Research ArticleLong-Term Variations of Temperature and Precipitation inthe Megacity of Istanbul for the Development of AdaptationStrategies to Climate Change

Huumlseyin Toros1 Mohsen Abbasnia1 Mustafa Sagdic2 andMete Tayanccedil3

1Department of Meteorology Istanbul Technical University Maslak 34469 Istanbul Turkey2Department of Social Studies Faculty of Education Yıldız Technical University 34210 Istanbul Turkey3Department of Environmental Engineering Marmara University Goztepe Istanbul Turkey

Correspondence should be addressed to Mohsen Abbasnia abbasniaituedutr

Received 5 June 2017 Revised 21 September 2017 Accepted 17 October 2017 Published 19 November 2017

Academic Editor Julio Diaz

Copyright copy 2017 Huseyin Toros et alThis is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Istanbul as one of the four anchor megacities of Europe has shown a rise of 094∘C in average annual temperature over the longperiod of 1912ndash2016 under impacts of anthropogenic climate change A notable increase in temperatures has started after the 1940swhich is in parallel with the beginning of industrialization era in IstanbulThis warming is associated with an extensive populationgrowth and accompanied the decrease in vegetation cover Increasing in minimum series of temperature is more evident thanmaximum values and the rising rate of temperature values has been more pronounced during recent decades The first significantupward trend in precipitation series has periodically started in 1920s while there has been a stable trend from 2001 till todayThe daily average of rainfall amount increased with a mean value of 58mm during the total study period Rising rate of dailymaximum precipitation has been more evident in the last 3 decades which is shown by the increased frequency of heavy rainfallIn this regard both of the temperature and precipitation series had higher mean values (139∘C and 878mm) for the final period(1965ndash2016) compared to the mean values (136∘C and 799mm) belonging to the first period (1912ndash1964)

1 Introduction

Modernization and industrialization and their socioeco-nomic effects have caused an increase in rapid urbanizationall over the world The probability of climate variability andclimate change especially in megacities is the cause of greatconcern among scientists governments and lawmakersUncontrolled changes in demographic values land-use areavehicle and industrial product types and building featurescan play an important role in regional climate change andcan produce further problems in settlement areas due tomore occurrences of climatic disasters Previous researchersrepresent that since the beginning of the 20th century averageglobal temperature has increased about 06∘C [1] Accordingto the 5th assessment report (AR5) of the IntergovernmentalPanel onClimate Change (IPCC) global average temperaturehas shown a 085∘C increase over the period of 1800ndash2012 and

this trend of global warming is predicted to likely increaseduring the 21st century under the all Representative Concen-tration Pathways scenarios [2] Thus climate change can bea serious threat to sustainable development particularly inmegacities in terms of economic social and environmentaleffects From the beginning of the industrial revolution ageto the present the excessive use of fossil fuel for the sakeof economic growth has affected the delicate balance ofthe atmosphere by leading to an increase in greenhousegas emissions So it is accepted today that human lifestyleand the increasing demands on natural resources are themain causes of climate change Global climate change willhave diverse impacts on human health because the healthof the human population depends fundamentally on theconditions of social and natural environments [3] On theother hand urban heat island (UHI) is related to the fasturbanization and industrialization of the cities In this case a

HindawiAdvances in MeteorologyVolume 2017 Article ID 6519856 14 pageshttpsdoiorg10115520176519856

2 Advances in Meteorology

good understanding of rising trends in population and waterconsumption is necessary to prevent anywater shortage in thecoming decades

One of the consequences of climate change is the increas-ing frequency of extreme events [4 5] such as floodsdroughts heat waves associated with negative effects on theenvironment losing the life and property and disturbanceof human climate comfort For example many heat-relateddeaths occurred during the heat wave events in Chicago in1995 inWestern Europe in 2003 and in East Asia in 1994 [6ndash8] Also recent studies revealed significant worldwide warm-ing and a general increase in the frequency and persistence ofextreme temperature and precipitation events (eg [9ndash13])So far extensive researches have been conducted to detect theeffects of climate change on climatic variables at a regionalscale Meanwhile a comprehensive assessment of the globalwarming impacts on climatic parameters is necessary forany regions especially urban areas which are affected moreby negative effects of urbanization and anthropogenic onthe environment [14] In the study area some studies havebeen investigated to understand climate change impacts byfocusing on temperature and precipitation variability In thisconnection one of the conclusions is that these studies agreeon the urban development influence on the temperature andprecipitation values Tayanc et al [15] carried out a studyusing temperature and precipitation data for the period of1950ndash2004 to classify the weather stations into two groupsaccording to their populations under climate change overTurkeyThese results showed that a significant warming trendis observed in both urban and suburban weather stations inthe last decade and the variability of precipitation time seriesin the urban area is generally larger than rural area and urbanweather stations are experiencing more frequent and severedroughts and floods hazards Toros [16] found a significantwarming trend in both annual maximum and minimumtemperature parameters Increasing trends in warm periodsare comparatively stronger than cold periods especially in themaximum temperatures Toros [17] indicated that there is ageneral decreasing tendency in the total annual precipitationover more than 34 of weather stations of Turkey AlsoTurkes et al [18] indicated that all 3 air temperature seriesminimum maximum and average have increased after the1980s over Turkey In this case there is a negative trendin precipitation parameter along with the positive trend intemperature parameter in Turkey which may significantlyaffect water sustainability regionally There is no doubt thatmore water supply facilities will be needed due to a reductionin rainwater and increase in evaporation under warmerconditions as well as increasing rate in the population withmore water demands

In this regard the reports of IPCC demonstrate thateach of the last three decades has been warmer than all theprevious decades [19] Also rising inland temperatures arelarger than the sea temperatures mainly owing to the land-use characteristics and urban heat island The demographicstudies of United Nations [20] showed that the proportionof the world population in urban areas has increased froma mere 13 in 1900 to 54 in 2014 This is showing thatauthorities need to prepare and develop adaptation strategies

for the urban areas especially megacities Therefore thisstudy aimed to evaluate the climate variability in Istanbulone of the largest and most populated megacities in theworld under the light of anthropogenic effects and climatechange consequences the study is carried out based on 105years of recorded meteorological data of the weather stationKandilli In this work various statistical analyses are used todetect the long-term climatic changes in the temperature andprecipitation time seriesThus the results of this study relyingon the long-term meteorological data can be a referenceand guidance for researchers and decision-makers of urbanplanning climate change adaptation andmanagement owingto the strategic location of Istanbul

2 Study Area and Data

The megacity of Istanbul with the geographical coordinatesof 28∘5710158405310158401015840E and 41∘0110158400710158401015840N is the biggest city in Turkeyand is one of the most populous cities in the world Istanbulis geographically located in the northwest of Turkey and theweather station of Kandilli is in the middle of the Asian sideof this city (Figures 1(a) and 1(b)) This megacity is the heartof Turkey owing to the industry commercialization cultureand tourism This city covers a total area of approximately5400 km2 andhas a lot of hills with the highest point of 540mat theAydosHill [21] Owing to the effects of industrializationin Turkey significant environmental impacts have startedto be seen after the 1970s Today almost every kind ofenvironmental pollution can be found around the greaterIstanbul metropolitan area Istanbulrsquos climate is usually warmand dry in summer and cold and wet in winter Istanbul con-tains many hills and valleys topographically Thus Istanbulrsquostemperature and precipitation are changed from one side tothe other The average annual temperature is about 138∘Cwith a monthly maximum of 228∘C in August and amonthlyminimumof 53∘C in FebruaryThe total annual precipitationis 8375mm with a monthly maximum of 1284mm inDecember and a monthly minimum of 319mm in July Thepopulation of Istanbul is about 15million people according tothe results of population registration system in 2016 [22]Thetrend of population growth in Istanbul shows an increasingtendency from 2000 to 2017 (Figure 1(c)) Also it is possibleto see the different views of Istanbul for vegetation changes byusing the Normalized Difference Vegetation Index (NDVI)over the time (Figure 1(d))This variability can be the drivingforce of extrememeteorological events with damaging effectsunder climate change conditions

In this study we used climatic data of Kandilli stationwhich is located in a park and its neighboring site is an areawithout urban density This is an advantage for recording theweather data which has not been affected by urban activitiesand the results of the data analysis can show reliable climatevariability representing the region For the quality analysisof the data several methods are considered There is noinformation about the history of meteorological instrumentswhich have been used to record the data at the Kandillistation Some of earlier records of instruments which arepresented by Erinc [23] are neither complete enough to beused in the analysis nor fully available from any other source

Advances in Meteorology 3

(a)

Istanbul area

Turkey country

Other countries

Water resource

24∘ 0

㰀 0㰀㰀

E

26∘ 0

㰀 0㰀㰀

E

28∘ 0

㰀 0㰀㰀

E

30∘ 0

㰀 0㰀㰀

E

42∘0㰀0㰀㰀N

40∘0㰀0㰀㰀N

38∘0㰀0㰀㰀N

36∘0㰀0㰀㰀N

(b)

105

115

125

135

145

155

165

2000

2002

2004

2006

2008

2010

2012

2014

2016

2018

2020

2022

Ista

nbul

rsquos po

pula

tion

(mill

ions

)

YearsObserved PProjected P

(c)

1985 1995

2005 2015

(d)

Figure 1 Geographical location of Istanbul (a) and Turkey (b) population growth (c) and NDVI changes (d)

[24] Thus we decided to use a data period of 1912ndash2016 thatis larger than a century which has high-quality data and canprovide information about climate change Climate changeis a long-term continuous change (increase or decrease) toaverage weather conditions (eg average temperature) or therange of weather (eg more frequent and severe extremestorms) Both can also happen simultaneously Therefore inthe present study the meteorological data of Kandilli stationdue to the valuable longest period of recording the obser-vational data from 1912 to 2016 is used for a comprehensiveassessment of the climate fluctuations in Istanbul area inadvance to global climate change effects In this case the long-term daily minimum average and maximum temperatureand total annual precipitation time series are prepared frommeteorological service of Kandilli station which is situated atAsian part of Istanbul (see Figure 1(a))

3 Methodology

Different time series ofmeteorological data at the station scaleof Kandilli including daily temperature and precipitationare considered for analysis of the climate change effects inIstanbul city over the time First reliability and accuracy

of these data at the significance level of 95 (Table 1) arechecked via SPSS software for based quality tests such as Runtest of randomness Kolmogorov-Smirnov test of normalityand Levenersquos test of homogeneity In this case Run test isused to test whether the data time series are random or not[25] Also K-S test is applied to check whether the datatime series is normally distributed (or bell-shaped) with 0mean 1 standard deviation and a symmetric bell-shapedcurve [26] Finally Levenersquos test is used to assess if thegroups have equal variances or to test the assumption ofhomogeneity of variance by comparing climatic series of theKandilli station over the time and with its nearby stationcalled Sariyer (refer to Figure 1(a)) inside of Istanbul area[27] Thence the nonparametric Mann-Kendall (MK) trendtest and statistical regression analysis are used to discovervariability and trend in the data time series The result of MKtest shows any increasing or decreasing trends in the datatime series whereas a value of 119906(119905) higher than 196 showsa positive trend and a value of 119906(119905) lower than ndash196 showsa negative trend at the significance level of 95 [28 29]The Mann-Kendall test examines whether to accept the nullhypothesis H0 (shows nomonotonic trend) or the alternativehypothesis H1 (shows monotonic trend) As (1) shows the

4 Advances in Meteorology

MK test is applicable in cases when the data values 119883119894 of atime series can be assumed to obey the model

119883119894 = 119891 (119905119894) + 120576119894 (1)

where 119891(119905) is a continuous monotonic increasing or decreas-ing function of time the residuals 120576119894 can be assumed to befrom the same distribution with zero mean

Then an upward or downward trend is given by a positiveor negative value of 119885 For this purpose at first the varianceof 119878 is computed using the following equation

VAR (119878) =119899 (119899 minus 1) (2119899 minus 5) minus sum119898

119894=1119905 (119905 minus 1) (2119905 + 5)

18 (2)

where119898 is the number of tied groups 119905 is the number of datavalues in the 119894 group Then the values of 119878 and VAR(119878) areused to compute the MK test statistic of119885 as is present in (3)(119885 119886 119887 119888)

119885119886 =119878 minus 1

radicVAR (119878) if 119878 gt 0

119885119887 = 0 if 119878 = 0

119885119888 =119878 + 1

radicVAR (119878) if 119878 lt 0

(3)

Moreover remote sensing phenology studies use data gath-ered by satellite sensors which measure wavelengths of lightabsorbed and reflected by green plants Certain pigmentin plant leaves strongly absorbs wavelengths of visible light(red) The leaves themselves strongly reflect wavelengths ofnear-infrared light which is invisible to human eyes Thishas long been used to monitor the vegetation and changesin the vegetation of the entire Earth Although there areseveral vegetation indices one of the most widely used is theNormalized Difference Vegetation Index (NDVI) (see (4))

NDVI = (Band 4 minus Band 3)(Band 4 + Band 3)

(4)

where RED (Band 3) and NIR (Band 4) stand for the spectralreflectance measurements acquired in the visible (RED) andnear-infrared (NIR) regions respectively NDVI values rangeis from +01 to minus01 A zero means no vegetation and closeto +1 (08ndash09) indicates the highest possible density of greenleaves Areas of barren rock sand or snow usually showvery low NDVI values Sparse vegetation such as shrubs andgrasslands or senescing crops may result in moderate NDVIvalues High NDVI values correspond to dense vegetationsuch as that found in temperate and tropical forests or cropsat their peak growth stage [30] In this case the NDVIindex is extracted based on accessible remote sensing imagesof the Landsat-7 ETM+ during the years of 1985 to 2015(refer to Figure 1(d)) Finally statistical analysis of populationgrowth rates in Istanbul city is calculated during the yearsof 2000 to 2017 to better understand the trends in climaticparameters according to anthropogenic behaviors affected bygrowing urbanization in Istanbul area over the time (refer toFigure 1(c))

4 Results and Discussion

Trend analysis of the temperatures and precipitation seriesusing the method of the linear best-fit curve and Man-Kendall test are applied for subperiods throughout the wholeperiod of 1912ndash2016 In this case the first half of studiedperiod (1912ndash1964) is compared to second half of studiedperiod (1964ndash2016) for knowing the actual changes in cli-matic parameters Also since the globe on average was hotterin the early 1960s according to fifth assessment report ofIPCC in 2013 a comparing analysis between two subperiodsof 1912ndash1980 and 1981ndash2016 is performed by choosing theturning point in the year of 1980 In this case the resultsof basic quality tests for all studied periods showed that thedistribution of all-time series at the significance level of 95does not follow the randomness principle by use of Run testand the normality principle by using Kolmogorov-Smirnovtest for both of Kandilli station and its nearby station ofSariyer Then the results of Levenersquos test of homogeneityand ANOVA test for assessing the absolute homogeneityshowed that minimum temperature and precipitation timeseries of the Kandilli station are homogeneous by comparisonbetween two subperiodsrsquo data including 1st and 2nd parts ofthe entire period of 1912ndash2016 Also in the case of assessingthe relative homogeneity between two separate data groupsincluding the Kandilli station and its nearby station of Sariyershowed that the average temperature and precipitation timeseries are homogeneous for the same studied period of1965ndash2016 Moreover results of the homogeneity test showedno homogeneity in the other series whether between meteo-rological series at the Kandilli station and its nearby station ofSariyer in the same period or between two subperiods of theKandilli stationrsquos series (Table 1)These uneven results of basictests onto data time periods and subperiods can be affectedby urbanization along with global warming effects So thepopulation of Istanbul has almost increased by a factor of 17from 1927 to 2016 and the effects of population growth arechanged in the land use and land cover Also it is possible tosee the different views of Istanbul for vegetation changes byusing the Normalized Difference Vegetation Index (NDVI)over the time from 1985 to 2015 (see Figure 1(d)) In this caseadverse effect of urbanization is the decline in forest coverMany countries have witnessed severe deforestation due torapid urban growth On the global scale the world lost 3 ofits total forest area between 1990 and 2005 with an average02 decrease per year Urban areas in Istanbul have alsoexpanded about 879 from 1977 to 2007 while forest areasdeclined about 54 in the same period One of the mostsignificant results of the study was that total forest areas inIstanbul have almost increased about 03 between 2000 and2007 [31] The increase of forest areas as opposed to growthin population size may well be explained by the movementof population from rural areas to Istanbul city Thus it canprovide an opportunity to release human pressure from forestareas probably resulting in a positive development of forest[32 33] The same period has also witnessed an increase in arate of 31 times in the cityrsquos population during the last thirtyyears Thus Istanbulrsquos population is rapidly increasing andconsequently the residential settlement is widening This

Advances in Meteorology 5

Table1Statisticalou

tputso

falltests

ford

ataq

ualitycontrol

Cases

NPartestslowast

Testof

norm

alitylowast

Testof

homogeneityof

varia

ncelowast

Parameter

Station

Perio

dTo

tal

Valid

Missing

Std

Runs

test

Kolm

ogorov-Smirn

ovANOVA

test

Levenersquostest

119873(day)119873(

)119873(day)

119885119886

Sig

Mean

Testvalues

Statistic

Sig

119865Sig

Statistic

Sig

119879min

Kand

illi

1965ndash2016

18980

999

2366minus137

00

105

105

0065

00

367

00

2100

Sarıy

er18998

100

167minus136

00

109

109

0061

00

Kand

illi

1st

38353

999

2366minus170

00

104

107

067

00

686

00

183

017

2nd

119879avg

Kand

illi

1965ndash2016

18985

100

471minus138

00

139

139

007

00

031

057

108

029

Sarıy

er18998

100

171minus135

00

139

138

0069

00

Kand

illi

1st

38353

100

471minus169

00

138

141

007

00

332

00

509

002

2nd

119879max

Kand

illi

1965ndash2016

18990

999

1884minus130

00

186

186

007

00

249

00

204

00

Sarıy

er18998

100

178minus129

00

175

175

006

800

Kand

illi

1st

38335

100

1883minus161

00

183

188

006

00

894

00

272

00

2nd

Precip

Kand

illi

1965ndash2016

7691

389

12238

96minus13

00

612

612

026

00

104

031

135

024

Sarıy

er7855

35

11074

92minus10

00

627

63

00253

00

Kand

illi

1st

38353

371

24121

96minus164

00

61

25

026

00

700

0799

037

2nd

ldquoardquom

eans

basedon

them

ediansta

tistic

allylowastshow

sthe

significantlevelof

test

ldquo1strdquo

show

sfirstsectio

nof

thes

tudied

perio

d1912ndash196

4ldquo2nd

rdquosho

wssecon

dsectionof

studied

perio

d1965ndash2016

6 Advances in Meteorology

120

125

130

135

140

145

150

155

160Av

erag

e tem

pera

ture

(∘C)

1916

1920

1924

1928

1932

1936

1940

1944

1948

1952

1956

1960

1964

1968

1972

1976

1980

1984

1988

1992

1996

2000

2004

2008

2012

2016

1912

Years

Average 138∘CMaximum 153∘C in 1966Minimum 124∘C in 1920

R2 = 01644y = 0009x + 13315

R2 = 0075y = 00097x + 13592

R2 = 04619

y = 00514x + 132671912ndash2016 1951ndash2012 1984ndash2016

(a)

Sept

embe

r

Dec

embe

rN

ovem

ber

Febr

uary

Janu

ary

June July

Augu

st

Oct

ober

Mar

ch

May

April

Months

Average 137∘CMaximum 227 ∘C in July

Minimum 52∘C in January

00

50

100

150

200

250

Aver

age t

empe

ratu

re (∘

C)(b)

Figure 2 Trend analysis of average annual temperature (a) and average monthly temperature (b)

is the main reason why the forests as Istanbulrsquos life sourceare being destroyed Overall this suggests that progress hasbeen made in Istanbul not only in having sustainable urbangrowth but also in preserving restoring and even expandingforest areas especially after the year 2000 In the followingafter assessing the population growth and land use change inIstanbul the results obtained from main climatic parameterstrend are analyzed for previous decades compared to lastdecades

41 Temperature The analysis of average yearly maximumand minimum temperature has been carried out using thestatistical method of least squares and the MK test at thesignificance level of 5 is shown with a dashed line (plusmn2)in Figures 2ndash5 The results of linear trends and MK rankcorrelation test showed that there is an increase in the averagedaily temperature series for the whole period of 1912ndash2016 inIstanbul (Figure 2(a)) However this upward trend includesimportant differences While there is not any significantchange in the annual temperature time series from 1912to 1944 the upward trend became apparent between 1944and 1969 A cool period exists in the period of 1969ndash1993and after that a significant rising trend is pronounced inaverage annual temperature at the 95 confidence levelThere has been a rise of about 094∘C in the average dailytemperature over the last 100 years at Kandilli Temperaturehas increased 060∘C in the period of 1951ndash2016 and 170∘C inthe period of 1981ndash2016 In this case the global combined landand ocean surface temperatures show an increase of about

089∘C (069∘Cndash108∘C) over the period of 1901ndash2016 andabout 072∘C (049∘Cndash089∘C) over the period of 1951ndash2016according to fifth assessment report of IPCC in 2013

Nevertheless the data time series of Kandilli stationshows that the increase in temperature parameter after the1940s is in parallel with the beginning of industrialization erain Istanbul Of course regime changes in temperature timeseries on regional scales cannot be totally explained by thenature destruction and pollutant emissions in and aroundIstanbul alone These effects can also be considered as areflection of the general situation happening in the worldon a larger scale Temperature time series of the Kandillistation shows an annual average about 137∘C and absoluteminimum and maximum in 1920 with a value of 124∘C andin 1966 with a value of 153∘C respectively (Figure 2(a))Owing to the Mediterranean climate of the study areamonthly temperature time series is increased from FebruarytoAugust and is decreased fromAugust to February reachinga minimum of 52∘C in February and a maximum of 227∘Cin August (Figure 2(b))

A general tendency of a warming trend in the dailytemperature series is found for the whole studied periodof 1912ndash2016 by using the least squares regression analysisTherefore the trend analysis revealed that the daily averageof temperature has increased by a rate of 09 for the periodfrom 1912 to 2016 (Figures 3(b1) and 3(b2)) Also there is apositive trend about 149∘C in daily maximum temperatureseries during the whole studied period Furthermore theincreasing trends have notably occurred in two subperiods

Advances in Meteorology 7

Tmin Tavg Tmax

Tmax Tmin DTR

u(t)u(t)

u(t)

u㰀(t)u㰀(t)

u㰀(t)

u(t)u㰀(t)

u㰀(t)

u(t)

u(t)

R2 = 02642

y = 00149x minus 10794R2 = 0163

y = 0009x minus 3926

R2 = 0233

y = 00116x minus 12375

R2 = 00182y = 00032x + 16875

R2 = 00734

y = minus00934x + 20292R2 = 02513

y = 02039x minus 37934

(a1)

(a2)

(b1)

(b2)

(c1)

(c2)

(d1)

(d2)

(e1)

(e2)

(f1)

(f2)

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1912

1976

1984

1992

2000

2008

2016

1968

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

2016

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1912

1976

1984

1992

2000

2008

2016

1968

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

minus4minus3minus2minus1

01234567

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus6minus5minus4minus3minus2minus1

0123456

minus4minus3minus2minus1

0123

010203040506070

Num

ber o

f day

s (gt30∘ C)

0102030405060

65707580859095

100

Tem

pera

ture

(∘C)

minus3minus2minus1

0123456

minus3minus2minus1

012345

minus3minus2minus1

01234567

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

9

10

11

12

13Te

mpe

ratu

re (∘

C)

12

13

14

15

16

Tem

pera

ture

(∘C)

16

17

18

19

20

21

Tem

pera

ture

(∘C)

Num

ber o

f day

s (lt0∘ C)

Figure 3 Trend analysis of the temperatures time series using linear best-fit curve andMK test for daily minimum temperature (a1 a2) dailyaverage temperature (b1 b2) daily maximum temperature (c1 c2) number of days with daily maximum temperature gt 30∘C (d1 d2) numberof days with daily minimum temperature lt 0∘C (e1 e2) and daily temperature range (f1 f2)

between 1944 and 1969 and between 1991 and 2016 (Figures3(c1) and 3(c2)) There is a positive trend about 11∘C indaily minimum temperature series during the whole studiedperiod too while this positive trend value is smaller than thepositive trend value of maximum temperature The trend ofminimum temperature series is stable during the period of1912ndash1944 Then it shows a steady increase during 1942 to1989 which is most pronounced during 1944ndash1954 Finallythe rising trends in the minimum temperature series became

significant based onMK testrsquos result for the last period of 2003to 2016 (Figures 3(a1) and 3(a2)) In this case the minimumtemperature shows increasing change for all season wherethe most increasing change has happened in summer andspring season in all studied periods respectively (Table 2)Analysis of the number of days having daily maximumtemperature higher than 30∘C for the long time period of1912 to 2016 by focusing precisely on its subperiods showedthat there is almost no trend during 1912ndash1945 a significant

8 Advances in Meteorology

400

500

600

700

800

900

1000

1100

1200

1300

1912 1920 1928 1936 1944 1952 1960 1968 1976 1984 1992 2000 2008 2016

Tota

l pre

cipi

tatio

n (m

m)

Years

Average 8375 mmMaximum 12894 mm in 1981Minimum 4487 mm in 1921

(a)

0

20

40

60

80

100

120

140

Janu

ary

Febr

uary

Mar

chAp

rilM

ayJu

ne July

Augu

stSe

ptem

ber

Oct

ober

Nov

embe

rD

ecem

ber

Tota

l pre

cipi

tatio

n (m

m)

Months

Average 699 mmMaximum 1284 mm in December

Minimum 319 mm in July

(b)

Figure 4 Time series of total annual precipitation (a) and total monthly precipitation (b)

increasing trend during 1945ndash1969 a relative reduction trendduring 1969ndash1977 an increasing trend during 1977ndash1992and a more pronounced increasing trend during 1992ndash2007(Figures 3(d1) and 3(d2)) On the other hand analysis of thenumber of dayswith dailyminimum temperatures lower than0∘C shows a decreasing trend generally So there is a relativeincrease during 1915ndash1930 almost no trend during 1930ndash1954an increase during 1954ndash1963 a decrease during 1963ndash1985almost no trend during 1985ndash1995 and a decrease during1995ndash2011 by a detailed focus on subperiods of the time series(Figures 3(e1) and 3(e2)) When the daily temperature rangein the whole long period is analyzed an increment can beclearly observed The situation was stable in the period of1912ndash1950There is an increasing trend during 1950ndash1970 andan obvious decrement during 1970ndash1985 Also there is anincrease between 1985 and 2010 (Figures 3(f1) and 3(f2))

The comparison results between the periods of 1912ndash1964and 1965ndash2016 and the periods of 1912ndash1980 and 1981ndash2016showed that (Table 1) there is an increment in the dailyaverage temperature series about 04∘C for the periods of1965ndash2016 (14∘C) compared to previous period of 1912ndash1964(136∘C) and an increment about 05∘C for the periods of1981ndash2016 (141∘C) compared to previous period of 1912ndash1980(136∘C) respectively In addition the standard deviation isincreased by a coefficient between 01∘C and 02∘C in thewhole studied period The increment of the minimum valuesin the daily average temperature series is more evident thanthe maximum values by a mean coefficient of about 6 timesfor both of studied periodsTheminimumvalue of daily aver-age temperatures in the periods of 1912ndash1964 and 1965ndash2016revealed an increment of 06∘C and 08∘C respectively Themaximum value of daily average temperatures in the periodsof 1912ndash1964 and 1965ndash2016 showed a decrement of 02∘Cand 01∘C respectively When the monthly values of dailyaverage temperature are analyzed the highest increment in

the periods of 1912ndash1964 and 1965ndash2016 is found to be inJune at a rate of 1∘C The highest increment in the periodsof 1912ndash1980 and 1981ndash2016 appeared to be in June againwith a rate of 11∘C Also there has been temperature decre-ment in October November and December for all studiedperiods The highest decrement was in November with arate of minus05∘C in the periods of 1912ndash1964 and 1965ndash2016 aswell as minus04∘C in the periods of 1912ndash1980 and 1981ndash2016In this case the analysis of seasonal values of the dailyaverage temperatures showed that the highest increment inthe periods of 1912ndash1964 and 1965ndash2016 took place in thesummer season at a rate of 07∘C The highest increment inthe periods of 1912ndash1980 and 1981ndash2016 was found to be insummer again by a rate of 11∘C Also there was a temperaturedecrement in the autumn season which was about minus010∘Conly for the periods of 1912ndash1964 and 1965ndash2016 Analysisof the percentile thresholds of daily average temperaturesshowed that the temperature increment at the 5th percentileis 05∘C for the periods of 1912ndash1964 and 1965ndash2016 while thisincrement is 03∘C for the periods of 1912ndash1980 and 1981ndash2016Therefore it can be said that the rate of temperature risinghas increased further as much as the time is closer to the endyears of studied period Also these increment values indicatean increase of 02∘C and 03∘C at the 25th percentile 05∘Cand 08∘C at the 75th percentile and 06∘C and 11∘C at the95th percentile thresholds for thewhole studied subperiods of1912ndash1964 1965ndash2016 1912ndash1980 and 1981ndash2016 respectivelyThis situation shows that the increment of higher values inthe daily average temperatures is greater than the incrementof lower values and also this increment ismore evident for theperiods of 1912ndash1980 and 1981ndash2016 compared to the periodsof 1912ndash1964 and 1965ndash2016

Comparison of the daily maximum temperature seriesfor two periods of 1912ndash1964 and 1965ndash2016 revealed anincrement of 08∘C which can be divided into an average

Advances in Meteorology 9

1912

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

Annual total

Years

Annual maximum daily precip

Annual heavy precipitation STD

R2 = 00549

y = 12491x minus 16171

R2 = 00103

y = 0087x minus 1104

u(t)u(t)

u㰀(t)

u㰀(t)

R2 = 2E minus 08y = minus1E minus 05x + 92115

R2 = 00091

y = 00082x minus 99366

u(t)

u㰀(t)

u(t)

u㰀(t)

(a1)

(a2)

(b1)

(b2)

(c1)

(c2)

(d1)

(d2)

400500600700800900

1000110012001300

Am

ount

of p

reci

p (m

m)

Am

ount

of p

reci

p (m

m)

20

40

60

80

100

120

140

160

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

minus3

minus2

minus1

0

1

2

3

4

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus3

minus2

minus1

0

1

2

3

2016

1912

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1928

Years

2468

101214161820

Am

ount

of p

reci

p (m

m)

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

0

2

4

6

8

10

12

14

Num

ber o

f day

s (gt25

mm

)

minus3

minus2

minus1

0

1

2

3

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus3

minus2

minus1

0

1

2

3

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

2016

1912

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1928

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

Figure 5 Trend analysis of the precipitation time series using least squares linear regression and MK trend test for total annual precipitation(a1 a2) daily maximum precipitation (b1 b2) daily maximum precipitation greater than 25mm (c1 c2) and standard deviation in the dailyprecipitation series (d1 d2)

10 Advances in Meteorology

Table2Statisticalanalyses

outputso

ftem

perature

andprecipitatio

nparameters(SD

stand

arddeviation

Inc

increasedecdecrease)

Parameters

Minim

umtemperature

(∘ C)

Averagetem

perature

(∘ C)

Maxim

umtemperature

(∘ C)

Precipitatio

n(m

m)

Statistics

Perio

ds

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Incor

dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

Average

total

102

107

052

102

109

071

136

140

037

136

141

043

180

187

075

181

187

060

8113

8656

544

8267

8591

324

SD66

66

004

65

67

022

7172

007

7173

019

82

84

021

82

85

025

9391minus03

9193

02

Minim

umminus54minus42

116minus53minus39

138minus27minus21

062minus27minus19

074minus08minus01

076minus07

00

068

00

00

00

00

00

00

Maxim

um223

229

069

222

232

099

276

278

023

277

277

000

356

362

059

358

361

029

589

620

31

579

651

71Percentile5

minus04

02

061minus03

03

055

21

25

049

22

24

026

45

50

052

46

48

015

01

01minus01

01

01

00

Percentile2

547

52

045

48

52

038

7780

024

7778

01 0

114

119

054

115

116

011

07

05minus03

06

Percentile5

0106

108

019

105

106

007

140

141

013

140

138minus020

186

190

040

187

185minus024

26

21minus05

23

24

01

Percentile7

5159

165

066

158

164

059

199

204

054

199

201

022

250

261

105

252

255

034

7672minus04

7374

01

Percentile9

5196

203

074

195

202

074

236

242

065

236

239

032

296

307

110

297

300

027

229

226minus03

227

223minus05

gt30

Cggt80

mm

00

00

00

00

00

00

1626

1017

2810

00

00

00

gt25

Cggt40

mm

00

00

00

712

57

147

90107

1794

108

142

20

22

0lt0C

gmdash

2316minus6

2216minus6

76minus2

76minus1

21minus1

21minus1

mdashmdash

mdashmdash

mdashmdash

ltminus5C

gmdash

21minus1

21minus1

00

00

00

00

00

00

mdashmdash

mdashmdash

mdashmdash

Winter

33

36

032

33

38

048

59

62

031

60

62

020

9095

047

9195

035

1033

1096

63

1056

1066

09

Sprin

g74

87

131

7582

073

111

124

121

113

119

060

160

171

116

162

170

076

496

513

17515

511minus04

Summer

174

187

124

174

186

125

216

224

082

216

225

091

269

273

043

271

282

109

334

331minus03

330

368

38

Autumn

124

127

028

123

127

036

156

156

000

156

155minus001

199

197minus020

200

201

019

842

831minus10

854

919

65

Ann

ual

101

109

079

101

108

070

136

141

057

136

140

043

179

184

042

181

187

060

676

719

43

689

716

27

Advances in Meteorology 11

value of 18∘C for the period of 1912ndash1964 and an averagevalue of 187∘C for the period of 1965ndash2016 On the otherhand comparison of the daily maximum temperature seriesfor the periods of 1912ndash1980 and 1981ndash2016 has revealed anincrement of 06∘C with an average value between 182∘C and188∘C for both of these periods respectively In additionthe standard deviation values are increased between 02∘Cand 03∘C during the whole studied periodThe increment inthe minimum values of daily maximum temperature seriesis more evident than the maximum values The minimumvalues of daily maximum temperatures have exhibited anincrement of 08∘C for the period of 1965ndash2016 compared tothe previous period of 1912ndash1964 and also an increment of07∘C for the period of 1981ndash2016 compared to the previousperiod of 1912ndash1980 Also themaximumvalues of dailymaxi-mum temperature series exhibited amean increment of 05∘Cfor both studied periods Seasonal analysis of maximumtemperature series for both of studied section periods showedthat the highest rising temperature has been happened bya value of 12∘C in summer On the other hand the lowestincrement has happened in the autumn season The monthlyanalysis of daily maximum temperature series showed thatthe highest increment took place during the last monthsof the spring season and the first month of the summerseason in all of the studied periods When the percentiles ofdaily maximum temperatures are analyzed the temperatureincrement based on the 5th percentile threshold is 05∘C forthe periods of 1912ndash1964 and 1965ndash2016 while the incrementis 03∘C for the periods of 1912ndash1980 and 1981ndash2016 This canbe considered as an important sign of rising temperature overthe time In this case the value of percentile thresholds isincreased with extending the length of the time period withextending the length of the first section of studied period infavor of last years than the previous ones In this regard thesevalues are 05∘C and 04∘C at the 25th percentile 1∘C and11∘C at the 75th percentile and 11∘C and 12∘C at the 95thpercentile for both sections of studied periods respectivelyThis situation shows that the increment of higher values inthe dailymaximum temperatures is greater than lower values

The comparison analysis of daily minimum temperaturesbetween the periods of 1912ndash1964 and 1965ndash2016 and theperiods of 1912ndash1980 and 1981ndash2016 showed that there is ageneral increment of 05∘C during the first section periodswhich can be given as 102∘C and 107∘C for the individualperiods of 1912ndash1964 and 1965ndash2016 and also a generalincrement of 08∘C during the second section periods whichcan be provided as 102∘C and 11∘C for the individual periodsof 1912ndash1980 and 1981ndash2016 respectively In addition thestandard deviation values increased among these sectionperiods from 0∘C to 03∘C The increment in the minimumvalues of daily minimum temperature series is more evidentthan the maximum values The minimum values of dailyminimum temperature series showed an increment of 12∘Cfrom the period of 1912ndash1964 to 1965ndash2016 and also anincrement of 16∘C from the period of 1912ndash1980 to 1981ndash2016The maximum values of daily minimum temperature serieshave shown an increment of 07∘C for the section periodsof 1912ndash1964 and 1965ndash2016 and an increment of 12∘C forthe section periods of 1912ndash1980 and 1981ndash2016 Overall

the minimum values have had an average increment of14∘C while the maximum values have had an averageincrement of 1∘C during the last century which can showthe higher rate of upward trends in the temperature timeseries Monthly analysis of minimum temperature seriesshowed that the highest increment for the section studiedperiods of 1912ndash1964 and 1965ndash2016 has occurred by a valueof 1∘C in June while the highest increment for the sectionstudied periods of 1912ndash1980 and 1981ndash2016 has occurred bya value of 16∘C in August Meanwhile there is a temperaturedecrement in November in both of these periods Howeverthere is a clear decrement in the first half of these periodsin October and December whereas there is an incrementin the second half of these periods The seasonal analysisof the daily minimum temperature series showed that thehighest increment has happened in the summer season witha mean value of 07∘C for the section periods of 1912ndash1964and 1965ndash2016 and with a mean value of 13∘C for the sectionperiods of 1912ndash1980 and 1981ndash2016 respectively Then theincreasing rate during the summer season became moreevident during the recent decades Analysis of the percentilethresholds of daily minimum temperature series showed thatthe temperature increment at the 5th percentile is 06∘C forall studied time periods of 1912ndash1964 1965ndash2016 1912ndash1980and 1981ndash2016 Also these increment values indicate anincrease of 04∘C and 06∘C at the 25th percentile 07∘C and11∘C at the 75th percentile and 07∘C and 14∘C at the 95thpercentile thresholds for the whole studied subperiods of1912ndash1964 1965ndash2016 1912ndash1980 and 1981ndash2016 respectivelyThese rising rates in minimum temperature series are moreevident for the periods of 1912ndash1980 and 1981ndash2016 than theperiods of 1912ndash1980 and 1981ndash2016 Therefore it can be saidthat the rate of temperature rising has increased further aslong as the studied time period is closer to the last years

42 Precipitation Annual average precipitation in Istanbulis 838mm with a range of minimum value of 449mm in1921 and a maximum value of 1289mm in 1981 based on theobservatory data of Kandilli station during the whole studiedperiod from 1912 to 2016 (Figure 4(a)) Also monthly averageprecipitation is 699mm with a range of the minimum of326mm in July and the maximum of 1286mm in Decemberduring the whole studied time period (Figure 4(b))

Analysis of the trend in the annual average precipitationtime series by the methods of linear regression analysisand MK trend test has shown that periodically there arepartial increments and significant differences during the totalstudied period from 1912 to 2016 (Figures 5(a1) and 5(a2))But this increment in the precipitation time series is notas clear as the increment in the temperature time seriesHowever it is obvious in the precipitation time series thatthere is an increment between the years of 1917 and 1925a stable condition between the years of 1925 and 1954 anincrement between the years of 1954 and 1965 a decrementbetween the years of 1965 and 1974 an increment between theyears of 1974 and 2001 and again a no change situation from2001 till the end The trend analysis of the daily maximumprecipitation series showed an increment of 29mm for theperiods of 1912ndash1964 and 1965ndash2016 as well as an increment

12 Advances in Meteorology

of 93mm for the periods of 1912ndash1980 and 1981ndash2016 Thesevalues indicated that rainfall has increased at a rate of morethan three times over the last decades than the previous ones(Figures 5(b1) and 5(b2)) Also the number of days with dailyprecipitation greater than 25mm presented an incrementduring the period of 1912ndash2016 although this is not significantat the confidence level of 005 (Figures 5(c1) and 5(c2))The trend of standard deviation in the daily precipitationtime series showed a slowly increasing trend during the totalstudied period (Figures 5(d1) and 5(d2))

The comparison of the results of statistical analysisbetween the daily average rainfall amounts belonging tothe periods of 1912ndash1964 and 1965ndash2016 with those of theperiods of 1912ndash1980 and 1981ndash2016 revealed that there is anincrement of 78mm from the period of 1912ndash1964 to theperiod of 1965ndash2016 and an increment of 38mm from theperiod of 1912ndash1980 to the period of 1981ndash2016 In additionthe analysis of standard deviation exhibited a decrementfrom the period of 1912ndash1964 to the period of 1965ndash2016 andan increment from the period of 1912ndash1980 to the periodof 1981ndash2016 (Table 2) The analysis of monthly averageprecipitation time series showed that the highest incrementhas happened in October with 227mm for the sectionperiods of 1912ndash1964 and 1965ndash2016 and with 347mm forthe section periods of 1912ndash1980 and 1981ndash2016 On the otherhand the highest decrease took placewith a value ofminus04mmduring both May and July for the periods of 1912ndash1964 and1965ndash2016 and with a value of minus16mm in September for thesection periods of 1912ndash1980 and 1981ndash2016 The analysis ofseasonal average precipitation time series showed that thehighest increase occurred in autumn with a value of 72mmfor the section periods of 1912ndash1964 and 1965ndash2016 as wellas with a value of 64mm for the periods of 1912ndash1980 and1981ndash2016 respectively There is no remarkable decreasingchange seasonally except in winter season during the periodsof 1912ndash1980 and 1981ndash2016 Furthermore the percentilethresholds of daily average precipitation indicated that thereis an insignificant negative trend based on all percentilesand for all studied periods Overall the statistical analysisshowed that the total average precipitation of Istanbul hasincreased while this increasing trend is more pronounced inthe previous decades than the last 3 decades On the otherhand the increasing rate of daily maximum precipitation ismore evident during the last 3 decades than the previousdecades which can be proven by the increasing frequency ofheavy rainfall events in Istanbul

Generally the results of trend analysis of Kandilli stationduring the last 105 years of 1912ndash2016 showed that there isa warming significant trend in the precipitation time seriesby using both methods of linear regression analysis andMK trend test On the contrary previous climate studiesconducted over Turkey put forward that there has been adecreasing trend in annual precipitation time series duringthe recent decades regionally The results of a previouslyconducted study by using the daily precipitation and tem-perature data sets of Florya and Goztepe meteorologicalstations in Istanbul area between 1960 and 2013 showedthat most notably the precipitation during the warm periodshas decreased but the frequency of the intense rain has

increased and the majority of these episodes of intense raincoincided with the warm periods Other determinationswere the rise in the annual average temperature and theextension of the warm periods in a year This differentiationof the temperature features can lead to the aggravation of theevaporation and it can be effective for a longer period duringthe year [15] Thus it will make Istanbul be confronted withthe much more important problems of water managementand flood [34] Also the results of the current study for trendanalysis in the long period from 1912 to 2016 showed that themost striking spell is between the years of 1968 and 1998 dueto the existence of least number of rainfall events in IstanbulIt can be owing to industrialization along with the increasingair pollution as well as irregular urbanization in Istanbul areaIn this case severe droughts taking place during the yearsof 1988 1992 1993 and 2008 have threatened the reservoirswhich supplied fresh water of the city These years werecharacterized by not having enough rainfall events Theseyears are also characterized by more persistent high-pressuresystems and less occurrence of low-pressure systems in termsof number and strength As statements made by officialinstitutions the formation conditions for the atmosphericlayer of air pollution due to air pollutant emissions from fossilfuel combustion and industrial activities are more providedduring anticyclone or high-pressure system eventsThereforethis leads to warming up and generating an inversion layerin the boundary layer of atmosphere especially over city area[35] The inversion layer or urban heat island intensity isincreasing with the increasing city size andor populationa phenomenon that was also reported by others [36ndash38]Moreover these last climatic events also have affected somepolitical results beyond natural effects The most obviousexample of this is related to the local election of 1994 inIstanbul In this case extreme drought during the summers of1992 and 1993 has caused groundwater reservoirs to dry up inthe city discontinuance of water was experienced for severaldays or even weeks In those times the mayor of Istanbul cityhas lost the election of 1994 and he realized that this result wasdue to the peoplersquos reaction about the water shortages Manypolitical reports of that period also support this scientificview

5 Conclusions

Statistical analysis in temperature time series of the Kandillistation from 1912 to 2016 established that there is a notableincrease in temperature values after the 1940s which is inparallel with the beginning of industrialization era in Istan-bul There has been a rise about 094∘C in the daily averagetemperature series since the beginning of the last century Asignificant positive trend in the daily maximum temperatureseries is found about 156∘C Also there is a positive trendabout 087∘C in the daily minimum temperature series Onthe other hand analysis of the number of days with thedaily maximum temperature higher than 30∘C showed thatthere is an increasing trend Meanwhile analysis of thenumber of days with daily minimum temperatures lowerthan 0∘C showed a decreasing trend The increment in theminimum values of the daily minimum temperature series

Advances in Meteorology 13

is more evident than the maximum values of this series Inthis case these rising rates in the minimum temperatureseries are more evident for the section periods of 1912ndash1980and 1981ndash2016 than the section periods of 1912ndash1964 and1965ndash2016 This again shows that there is an increment inthe positive temperature trend from past to present decadesThe increment in the precipitation time series is not asclear as the increment in the temperature time series dueto periodic variability The trend analysis in the total annualprecipitation time series showed that the first significantupward trend has periodically been started from the 1920swhile there is a stable trend from 2001 till 2016 The dailyaverage of rainfall amounts has increased with a value of58mm during the period of 1912ndash2016 Also the analysisof heavy precipitation trend showed an increase of 61mmOverall the total average precipitation of Istanbul increasedwhile this increasing trend is more pronounced during theearly decades than the last 3 decades On the other handthe increasing rate of daily maximum precipitation is morepronounced in the last 3 decades than the previous decadesThen it was shown that the frequency of heavy rainfallat Istanbul has increased during the recent decades Thusthe precipitation changes in Istanbul have some differencescompared to the general tendency in precipitation trendthat was put forward by other studies as a decreasing trendover the whole of Turkey This result can be expressedas a positive effect of population overgrowth of Istanbulmegacity Comparison of the results in the first half of thestudy period (1912ndash1964) with the second half of the studyperiod (1965ndash2016) showed that both the average temperatureand average precipitation have higher values of 139∘C and878mm for the final phase compared to the values of 136∘Cand 799mm belonging to the initial phase Therefore it canbe stated that the megacity of Istanbul is directly affectedby the climate change and its consequences In this contextpotential risks of climate change in Istanbul megacity underhigher temperature conditions can be expressed as the rise inthe sea level increase in the rate of evapotranspiration andincrease in the frequency of heavy rainfall Also this city maynot be able to handle this uncontrolled population growthand its associated irreversible changes which is alreadypushing the natural limits by destroying the environmentTherefore the local governors of any megacity like Istanbulshould give more emphasis on the importance of sustainableurban development Thus it is urgent to prepare local andnational climate change strategies and action plans for themegacities

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

The authors are grateful to the Bogazici University theobservatory of Kandilli weather station and the Earthquake

Research Institute for providing the research data and tech-nical support The authors also gratefully acknowledge con-tributions of Assoc Professor Dr Yuksel Demirkaya Schoolof Social Sciences Marmara University This work has beensupported by Scientific and Technological Research Councilof Turkey (TUBITAK) under Grants 113R019 and 106Y258and by Marmara University (BAPKO) with projects FEN-E-120314-0066 FEN-C-YLP-090414-0102 FEN-L-250416-0180 and FEN-A-100413-0127

References

[1] D R Easterling B Horton P D Jones et al ldquoMaximum andminimum temperature trends for the globerdquo Science vol 277no 5324 pp 364ndash367 1997

[2] IPCC Climate Change 2014 IPCC Fifth Assessment Synthe-sis Report-Summary for Policymakers-an Assessment of Inter-Governmental Panel on Climate Change Cambridge UniversityPress Cambridge UK 2014

[3] J Carmin N Nadkarni and C Rhie Progress and Challenges inUrban Climate Adaptation Planning Results of a Global SurveyMIT Cambridge UK 2012

[4] A F Young ldquoUrban expansion and environmental risk in theSao Paulo Metropolitan Areardquo Climate Research vol 57 no 1pp 73ndash80 2013

[5] P Tian X Mu J Liu J Hu and C Gu ldquoImpacts of ClimateVariability and Human Activities on the Changes of Runoff andSediment Load in a Catchment of the Loess Plateau ChinardquoAdvances inMeteorology vol 2016 Article ID 4724067 15 pages2016

[6] R S Kovats and K L Ebi ldquoHeatwaves and public health inEuroperdquo European Journal of Public Health vol 16 no 6 pp592ndash599 2006

[7] S Conti P Meli G Minelli et al ldquoEpidemiologic studyof mortality during the Summer 2003 heat wave in ItalyrdquoEnvironmental Research vol 98 no 3 pp 390ndash399 2005

[8] J Kysely and J Kim ldquoMortality during heat waves in SouthKorea 1991 to 2005 how exceptional was the 1994 heat waverdquoClimate Research vol 38 no 2 pp 105ndash116 2009

[9] B Yan Z Xia F Huang L Guo and X Zhang ldquoClimatechange detection and annual extreme temperature analysis ofthe amur river basinrdquoAdvances inMeteorology vol 2016 ArticleID 6268938 14 pages 2016

[10] E M Fischer and R Knutti ldquoAnthropogenic contribution toglobal occurrence of heavy-precipitation and high-temperatureextremesrdquo Nature Climate Change vol 5 no 6 pp 560ndash5642015

[11] X Zhang L Alexander G C Hegerl et al ldquoIndices for moni-toring changes in extremes based on daily temperature andprecipitation datardquo Climate Change vol 2 no 6 pp 851ndash8702011

[12] K H Schlunzen P Hoffmann G Rosenhagen and W RieckeldquoLong-term changes and regional differences in temperatureand precipitation in the metropolitan area of Hamburgrdquo Inter-national Journal of Climatology vol 30 no 8 pp 1121ndash11362010

[13] G Bartolini M Morabito A Crisci et al ldquoRecent trends inTuscany (Italy) summer temperature and indices of extremesrdquoInternational Journal of Climatology vol 28 no 13 pp 1751ndash1760 2008

14 Advances in Meteorology

[14] S C Sheridan and T J Dolney ldquoHeat mortality and levelof urbanization measuring vulnerability across Ohio USArdquoClimate Research vol 24 no 3 pp 255ndash265 2003

[15] M Tayanc U Im M Dogruel andM Karaca ldquoClimate changein Turkey for the last half centuryrdquo Climatic Change vol 94 no3-4 pp 483ndash502 2009

[16] H Toros ldquoSpatio-temporal variation of daily extreme tempera-tures over Turkeyrdquo International Journal of Climatology vol 32no 7 pp 1047ndash1055 2012

[17] H Toros ldquoSpatio-temporal precipitation change assessmentsover Turkeyrdquo International Journal of Climatology vol 32 no9 pp 1310ndash1325 2012

[18] M Turkes C Yozgatlıgil I Batmaz et al ldquoHas the climate beenchanging in Turkey Regional climate change signals based on acomparative statistical analysis of two consecutive time periods1950-1980 and 1981-2010rdquoClimate Research vol 70 no 1 pp 77ndash93 2016

[19] IPCC ldquoummary for Policymakers In Climate Change 2013rdquo inThe Physical Science Basis The contribution of Working Group Ito the Fifth Assessment Report of the Intergovernmental Panel onClimate Change Cambridge University Press Cambridge UK2013

[20] United Nations ldquoDepartment of Economic and Social AffairsPopulation Division 2006 World Urbanization Prospects The2005 Revisionrdquo Working Paper ESAPWP200 2011

[21] Y Demirkaya Sayılarla Istanbul ITO Istanbul Turkey 2011[22] TUIK ldquoAddress based population registration system results of

2014rdquo Turkish Statistical Institute (TUIK) 2017 httpraporytuikgovtr10-03-2015-184727-842632346446643191142126876html

[23] S Erinc Climatology and its Methods Alfa Basım YayımDagitim Istanbul Turkey 4th edition 1965

[24] O M Gokturk D Bozkurt O L Sen and M Karaca ldquoQualitycontrol and homogeneity of Turkish precipitation datardquoHydro-logical Processes vol 22 no 16 pp 3210ndash3218 2008

[25] C Ley and D Paindaveine ldquoRuns Testsrdquo in Encyclopedia ofEnvironmetrics 2012

[26] A Ghasemi and S Zahediasl ldquoNormality tests for statisticalanalysis a guide for non-statisticiansrdquo International Journal ofEndocrinology andMetabolism vol 10 no 2 pp 486ndash489 2012

[27] G V Glass ldquoTesting Homogeneity of Variancesrdquo AmericanEducational Research Journal vol 3 no 3 pp 187ndash190 1966

[28] H BMann ldquoNonparametric tests against trendrdquo Econometricavol 13 pp 245ndash259 1945

[29] M G Kendall Rank Correlation Method Charles GriffinLondon UK 4th edition 1975

[30] H Turoglu ldquoDetection of Changes on Temperature and Precip-itation Features in Istanbul (Turkey)rdquo Atmospheric and ClimateSciences vol 04 no 04 pp 549ndash562 2014

[31] A Karaburun A Demirci and I-S Suen ldquoImpacts of urbangrowth on forest cover in Istanbul (1987-2007)rdquo EnvironmentalModeling amp Assessment vol 166 no 1-4 pp 267ndash277 2010

[32] K K Karanth LM Curran and J D Reuning-Scherer ldquoVillagesize and forest disturbance in Bhadra Wildlife SanctuaryWestern Ghats Indiardquo Biological Conservation vol 128 no 2pp 147ndash157 2006

[33] G Cakir C Un E Z Baskent S Kose F Sivrikaya andS Keles ldquoEvaluating urbanization fragmentation and landuseland cover change pattern in Istanbul city Turkey from 1971to 2002rdquo Land Degradation amp Development vol 19 no 6 pp663ndash675 2008

[34] R B Myneni F G Hall P J Sellers and A L Marshak ldquoTheinterpretation of spectral vegetation indexesrdquo IEEE Transac-tions on Geoscience and Remote Sensing vol 33 no 2 pp 481ndash486 1995

[35] Y S Unal H Toros A Deniz and S Incecik ldquoInfluence ofmeteorological factors and emission sources on spatial and tem-poral variations of PM10 concentrations in Istanbul metropoli-tan areardquo Atmospheric Environment vol 45 no 31 pp 5504ndash5513 2011

[36] M KaracaM Tayanc andH Toros ldquoEffects of urbanization onclimate of Istanbul and Ankarardquo Atmospheric Environment vol29 no 23 pp 3411ndash3421 1995

[37] Y Ezber O L Sen T Kindap and M Karaca ldquoClimatic effectsof urbanization in Istanbul a statistical and modeling analysisrdquoInternational Journal of Climatology vol 27 no 5 pp 667ndash6792007

[38] H S Park ldquoFeatures of the heat island in seoul and its sur-rounding citiesrdquo Atmospheric Environment (1967) vol 20 no10 pp 1859ndash1866 1986

Advances in Meteorology 3

(a)

Istanbul area

Turkey country

Other countries

Water resource

24∘ 0

㰀 0㰀㰀

E

26∘ 0

㰀 0㰀㰀

E

28∘ 0

㰀 0㰀㰀

E

30∘ 0

㰀 0㰀㰀

E

42∘0㰀0㰀㰀N

40∘0㰀0㰀㰀N

38∘0㰀0㰀㰀N

36∘0㰀0㰀㰀N

(b)

105

115

125

135

145

155

165

2000

2002

2004

2006

2008

2010

2012

2014

2016

2018

2020

2022

Ista

nbul

rsquos po

pula

tion

(mill

ions

)

YearsObserved PProjected P

(c)

1985 1995

2005 2015

(d)

Figure 1 Geographical location of Istanbul (a) and Turkey (b) population growth (c) and NDVI changes (d)

[24] Thus we decided to use a data period of 1912ndash2016 thatis larger than a century which has high-quality data and canprovide information about climate change Climate changeis a long-term continuous change (increase or decrease) toaverage weather conditions (eg average temperature) or therange of weather (eg more frequent and severe extremestorms) Both can also happen simultaneously Therefore inthe present study the meteorological data of Kandilli stationdue to the valuable longest period of recording the obser-vational data from 1912 to 2016 is used for a comprehensiveassessment of the climate fluctuations in Istanbul area inadvance to global climate change effects In this case the long-term daily minimum average and maximum temperatureand total annual precipitation time series are prepared frommeteorological service of Kandilli station which is situated atAsian part of Istanbul (see Figure 1(a))

3 Methodology

Different time series ofmeteorological data at the station scaleof Kandilli including daily temperature and precipitationare considered for analysis of the climate change effects inIstanbul city over the time First reliability and accuracy

of these data at the significance level of 95 (Table 1) arechecked via SPSS software for based quality tests such as Runtest of randomness Kolmogorov-Smirnov test of normalityand Levenersquos test of homogeneity In this case Run test isused to test whether the data time series are random or not[25] Also K-S test is applied to check whether the datatime series is normally distributed (or bell-shaped) with 0mean 1 standard deviation and a symmetric bell-shapedcurve [26] Finally Levenersquos test is used to assess if thegroups have equal variances or to test the assumption ofhomogeneity of variance by comparing climatic series of theKandilli station over the time and with its nearby stationcalled Sariyer (refer to Figure 1(a)) inside of Istanbul area[27] Thence the nonparametric Mann-Kendall (MK) trendtest and statistical regression analysis are used to discovervariability and trend in the data time series The result of MKtest shows any increasing or decreasing trends in the datatime series whereas a value of 119906(119905) higher than 196 showsa positive trend and a value of 119906(119905) lower than ndash196 showsa negative trend at the significance level of 95 [28 29]The Mann-Kendall test examines whether to accept the nullhypothesis H0 (shows nomonotonic trend) or the alternativehypothesis H1 (shows monotonic trend) As (1) shows the

4 Advances in Meteorology

MK test is applicable in cases when the data values 119883119894 of atime series can be assumed to obey the model

119883119894 = 119891 (119905119894) + 120576119894 (1)

where 119891(119905) is a continuous monotonic increasing or decreas-ing function of time the residuals 120576119894 can be assumed to befrom the same distribution with zero mean

Then an upward or downward trend is given by a positiveor negative value of 119885 For this purpose at first the varianceof 119878 is computed using the following equation

VAR (119878) =119899 (119899 minus 1) (2119899 minus 5) minus sum119898

119894=1119905 (119905 minus 1) (2119905 + 5)

18 (2)

where119898 is the number of tied groups 119905 is the number of datavalues in the 119894 group Then the values of 119878 and VAR(119878) areused to compute the MK test statistic of119885 as is present in (3)(119885 119886 119887 119888)

119885119886 =119878 minus 1

radicVAR (119878) if 119878 gt 0

119885119887 = 0 if 119878 = 0

119885119888 =119878 + 1

radicVAR (119878) if 119878 lt 0

(3)

Moreover remote sensing phenology studies use data gath-ered by satellite sensors which measure wavelengths of lightabsorbed and reflected by green plants Certain pigmentin plant leaves strongly absorbs wavelengths of visible light(red) The leaves themselves strongly reflect wavelengths ofnear-infrared light which is invisible to human eyes Thishas long been used to monitor the vegetation and changesin the vegetation of the entire Earth Although there areseveral vegetation indices one of the most widely used is theNormalized Difference Vegetation Index (NDVI) (see (4))

NDVI = (Band 4 minus Band 3)(Band 4 + Band 3)

(4)

where RED (Band 3) and NIR (Band 4) stand for the spectralreflectance measurements acquired in the visible (RED) andnear-infrared (NIR) regions respectively NDVI values rangeis from +01 to minus01 A zero means no vegetation and closeto +1 (08ndash09) indicates the highest possible density of greenleaves Areas of barren rock sand or snow usually showvery low NDVI values Sparse vegetation such as shrubs andgrasslands or senescing crops may result in moderate NDVIvalues High NDVI values correspond to dense vegetationsuch as that found in temperate and tropical forests or cropsat their peak growth stage [30] In this case the NDVIindex is extracted based on accessible remote sensing imagesof the Landsat-7 ETM+ during the years of 1985 to 2015(refer to Figure 1(d)) Finally statistical analysis of populationgrowth rates in Istanbul city is calculated during the yearsof 2000 to 2017 to better understand the trends in climaticparameters according to anthropogenic behaviors affected bygrowing urbanization in Istanbul area over the time (refer toFigure 1(c))

4 Results and Discussion

Trend analysis of the temperatures and precipitation seriesusing the method of the linear best-fit curve and Man-Kendall test are applied for subperiods throughout the wholeperiod of 1912ndash2016 In this case the first half of studiedperiod (1912ndash1964) is compared to second half of studiedperiod (1964ndash2016) for knowing the actual changes in cli-matic parameters Also since the globe on average was hotterin the early 1960s according to fifth assessment report ofIPCC in 2013 a comparing analysis between two subperiodsof 1912ndash1980 and 1981ndash2016 is performed by choosing theturning point in the year of 1980 In this case the resultsof basic quality tests for all studied periods showed that thedistribution of all-time series at the significance level of 95does not follow the randomness principle by use of Run testand the normality principle by using Kolmogorov-Smirnovtest for both of Kandilli station and its nearby station ofSariyer Then the results of Levenersquos test of homogeneityand ANOVA test for assessing the absolute homogeneityshowed that minimum temperature and precipitation timeseries of the Kandilli station are homogeneous by comparisonbetween two subperiodsrsquo data including 1st and 2nd parts ofthe entire period of 1912ndash2016 Also in the case of assessingthe relative homogeneity between two separate data groupsincluding the Kandilli station and its nearby station of Sariyershowed that the average temperature and precipitation timeseries are homogeneous for the same studied period of1965ndash2016 Moreover results of the homogeneity test showedno homogeneity in the other series whether between meteo-rological series at the Kandilli station and its nearby station ofSariyer in the same period or between two subperiods of theKandilli stationrsquos series (Table 1)These uneven results of basictests onto data time periods and subperiods can be affectedby urbanization along with global warming effects So thepopulation of Istanbul has almost increased by a factor of 17from 1927 to 2016 and the effects of population growth arechanged in the land use and land cover Also it is possible tosee the different views of Istanbul for vegetation changes byusing the Normalized Difference Vegetation Index (NDVI)over the time from 1985 to 2015 (see Figure 1(d)) In this caseadverse effect of urbanization is the decline in forest coverMany countries have witnessed severe deforestation due torapid urban growth On the global scale the world lost 3 ofits total forest area between 1990 and 2005 with an average02 decrease per year Urban areas in Istanbul have alsoexpanded about 879 from 1977 to 2007 while forest areasdeclined about 54 in the same period One of the mostsignificant results of the study was that total forest areas inIstanbul have almost increased about 03 between 2000 and2007 [31] The increase of forest areas as opposed to growthin population size may well be explained by the movementof population from rural areas to Istanbul city Thus it canprovide an opportunity to release human pressure from forestareas probably resulting in a positive development of forest[32 33] The same period has also witnessed an increase in arate of 31 times in the cityrsquos population during the last thirtyyears Thus Istanbulrsquos population is rapidly increasing andconsequently the residential settlement is widening This

Advances in Meteorology 5

Table1Statisticalou

tputso

falltests

ford

ataq

ualitycontrol

Cases

NPartestslowast

Testof

norm

alitylowast

Testof

homogeneityof

varia

ncelowast

Parameter

Station

Perio

dTo

tal

Valid

Missing

Std

Runs

test

Kolm

ogorov-Smirn

ovANOVA

test

Levenersquostest

119873(day)119873(

)119873(day)

119885119886

Sig

Mean

Testvalues

Statistic

Sig

119865Sig

Statistic

Sig

119879min

Kand

illi

1965ndash2016

18980

999

2366minus137

00

105

105

0065

00

367

00

2100

Sarıy

er18998

100

167minus136

00

109

109

0061

00

Kand

illi

1st

38353

999

2366minus170

00

104

107

067

00

686

00

183

017

2nd

119879avg

Kand

illi

1965ndash2016

18985

100

471minus138

00

139

139

007

00

031

057

108

029

Sarıy

er18998

100

171minus135

00

139

138

0069

00

Kand

illi

1st

38353

100

471minus169

00

138

141

007

00

332

00

509

002

2nd

119879max

Kand

illi

1965ndash2016

18990

999

1884minus130

00

186

186

007

00

249

00

204

00

Sarıy

er18998

100

178minus129

00

175

175

006

800

Kand

illi

1st

38335

100

1883minus161

00

183

188

006

00

894

00

272

00

2nd

Precip

Kand

illi

1965ndash2016

7691

389

12238

96minus13

00

612

612

026

00

104

031

135

024

Sarıy

er7855

35

11074

92minus10

00

627

63

00253

00

Kand

illi

1st

38353

371

24121

96minus164

00

61

25

026

00

700

0799

037

2nd

ldquoardquom

eans

basedon

them

ediansta

tistic

allylowastshow

sthe

significantlevelof

test

ldquo1strdquo

show

sfirstsectio

nof

thes

tudied

perio

d1912ndash196

4ldquo2nd

rdquosho

wssecon

dsectionof

studied

perio

d1965ndash2016

6 Advances in Meteorology

120

125

130

135

140

145

150

155

160Av

erag

e tem

pera

ture

(∘C)

1916

1920

1924

1928

1932

1936

1940

1944

1948

1952

1956

1960

1964

1968

1972

1976

1980

1984

1988

1992

1996

2000

2004

2008

2012

2016

1912

Years

Average 138∘CMaximum 153∘C in 1966Minimum 124∘C in 1920

R2 = 01644y = 0009x + 13315

R2 = 0075y = 00097x + 13592

R2 = 04619

y = 00514x + 132671912ndash2016 1951ndash2012 1984ndash2016

(a)

Sept

embe

r

Dec

embe

rN

ovem

ber

Febr

uary

Janu

ary

June July

Augu

st

Oct

ober

Mar

ch

May

April

Months

Average 137∘CMaximum 227 ∘C in July

Minimum 52∘C in January

00

50

100

150

200

250

Aver

age t

empe

ratu

re (∘

C)(b)

Figure 2 Trend analysis of average annual temperature (a) and average monthly temperature (b)

is the main reason why the forests as Istanbulrsquos life sourceare being destroyed Overall this suggests that progress hasbeen made in Istanbul not only in having sustainable urbangrowth but also in preserving restoring and even expandingforest areas especially after the year 2000 In the followingafter assessing the population growth and land use change inIstanbul the results obtained from main climatic parameterstrend are analyzed for previous decades compared to lastdecades

41 Temperature The analysis of average yearly maximumand minimum temperature has been carried out using thestatistical method of least squares and the MK test at thesignificance level of 5 is shown with a dashed line (plusmn2)in Figures 2ndash5 The results of linear trends and MK rankcorrelation test showed that there is an increase in the averagedaily temperature series for the whole period of 1912ndash2016 inIstanbul (Figure 2(a)) However this upward trend includesimportant differences While there is not any significantchange in the annual temperature time series from 1912to 1944 the upward trend became apparent between 1944and 1969 A cool period exists in the period of 1969ndash1993and after that a significant rising trend is pronounced inaverage annual temperature at the 95 confidence levelThere has been a rise of about 094∘C in the average dailytemperature over the last 100 years at Kandilli Temperaturehas increased 060∘C in the period of 1951ndash2016 and 170∘C inthe period of 1981ndash2016 In this case the global combined landand ocean surface temperatures show an increase of about

089∘C (069∘Cndash108∘C) over the period of 1901ndash2016 andabout 072∘C (049∘Cndash089∘C) over the period of 1951ndash2016according to fifth assessment report of IPCC in 2013

Nevertheless the data time series of Kandilli stationshows that the increase in temperature parameter after the1940s is in parallel with the beginning of industrialization erain Istanbul Of course regime changes in temperature timeseries on regional scales cannot be totally explained by thenature destruction and pollutant emissions in and aroundIstanbul alone These effects can also be considered as areflection of the general situation happening in the worldon a larger scale Temperature time series of the Kandillistation shows an annual average about 137∘C and absoluteminimum and maximum in 1920 with a value of 124∘C andin 1966 with a value of 153∘C respectively (Figure 2(a))Owing to the Mediterranean climate of the study areamonthly temperature time series is increased from FebruarytoAugust and is decreased fromAugust to February reachinga minimum of 52∘C in February and a maximum of 227∘Cin August (Figure 2(b))

A general tendency of a warming trend in the dailytemperature series is found for the whole studied periodof 1912ndash2016 by using the least squares regression analysisTherefore the trend analysis revealed that the daily averageof temperature has increased by a rate of 09 for the periodfrom 1912 to 2016 (Figures 3(b1) and 3(b2)) Also there is apositive trend about 149∘C in daily maximum temperatureseries during the whole studied period Furthermore theincreasing trends have notably occurred in two subperiods

Advances in Meteorology 7

Tmin Tavg Tmax

Tmax Tmin DTR

u(t)u(t)

u(t)

u㰀(t)u㰀(t)

u㰀(t)

u(t)u㰀(t)

u㰀(t)

u(t)

u(t)

R2 = 02642

y = 00149x minus 10794R2 = 0163

y = 0009x minus 3926

R2 = 0233

y = 00116x minus 12375

R2 = 00182y = 00032x + 16875

R2 = 00734

y = minus00934x + 20292R2 = 02513

y = 02039x minus 37934

(a1)

(a2)

(b1)

(b2)

(c1)

(c2)

(d1)

(d2)

(e1)

(e2)

(f1)

(f2)

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1912

1976

1984

1992

2000

2008

2016

1968

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

2016

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1912

1976

1984

1992

2000

2008

2016

1968

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

minus4minus3minus2minus1

01234567

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus6minus5minus4minus3minus2minus1

0123456

minus4minus3minus2minus1

0123

010203040506070

Num

ber o

f day

s (gt30∘ C)

0102030405060

65707580859095

100

Tem

pera

ture

(∘C)

minus3minus2minus1

0123456

minus3minus2minus1

012345

minus3minus2minus1

01234567

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

9

10

11

12

13Te

mpe

ratu

re (∘

C)

12

13

14

15

16

Tem

pera

ture

(∘C)

16

17

18

19

20

21

Tem

pera

ture

(∘C)

Num

ber o

f day

s (lt0∘ C)

Figure 3 Trend analysis of the temperatures time series using linear best-fit curve andMK test for daily minimum temperature (a1 a2) dailyaverage temperature (b1 b2) daily maximum temperature (c1 c2) number of days with daily maximum temperature gt 30∘C (d1 d2) numberof days with daily minimum temperature lt 0∘C (e1 e2) and daily temperature range (f1 f2)

between 1944 and 1969 and between 1991 and 2016 (Figures3(c1) and 3(c2)) There is a positive trend about 11∘C indaily minimum temperature series during the whole studiedperiod too while this positive trend value is smaller than thepositive trend value of maximum temperature The trend ofminimum temperature series is stable during the period of1912ndash1944 Then it shows a steady increase during 1942 to1989 which is most pronounced during 1944ndash1954 Finallythe rising trends in the minimum temperature series became

significant based onMK testrsquos result for the last period of 2003to 2016 (Figures 3(a1) and 3(a2)) In this case the minimumtemperature shows increasing change for all season wherethe most increasing change has happened in summer andspring season in all studied periods respectively (Table 2)Analysis of the number of days having daily maximumtemperature higher than 30∘C for the long time period of1912 to 2016 by focusing precisely on its subperiods showedthat there is almost no trend during 1912ndash1945 a significant

8 Advances in Meteorology

400

500

600

700

800

900

1000

1100

1200

1300

1912 1920 1928 1936 1944 1952 1960 1968 1976 1984 1992 2000 2008 2016

Tota

l pre

cipi

tatio

n (m

m)

Years

Average 8375 mmMaximum 12894 mm in 1981Minimum 4487 mm in 1921

(a)

0

20

40

60

80

100

120

140

Janu

ary

Febr

uary

Mar

chAp

rilM

ayJu

ne July

Augu

stSe

ptem

ber

Oct

ober

Nov

embe

rD

ecem

ber

Tota

l pre

cipi

tatio

n (m

m)

Months

Average 699 mmMaximum 1284 mm in December

Minimum 319 mm in July

(b)

Figure 4 Time series of total annual precipitation (a) and total monthly precipitation (b)

increasing trend during 1945ndash1969 a relative reduction trendduring 1969ndash1977 an increasing trend during 1977ndash1992and a more pronounced increasing trend during 1992ndash2007(Figures 3(d1) and 3(d2)) On the other hand analysis of thenumber of dayswith dailyminimum temperatures lower than0∘C shows a decreasing trend generally So there is a relativeincrease during 1915ndash1930 almost no trend during 1930ndash1954an increase during 1954ndash1963 a decrease during 1963ndash1985almost no trend during 1985ndash1995 and a decrease during1995ndash2011 by a detailed focus on subperiods of the time series(Figures 3(e1) and 3(e2)) When the daily temperature rangein the whole long period is analyzed an increment can beclearly observed The situation was stable in the period of1912ndash1950There is an increasing trend during 1950ndash1970 andan obvious decrement during 1970ndash1985 Also there is anincrease between 1985 and 2010 (Figures 3(f1) and 3(f2))

The comparison results between the periods of 1912ndash1964and 1965ndash2016 and the periods of 1912ndash1980 and 1981ndash2016showed that (Table 1) there is an increment in the dailyaverage temperature series about 04∘C for the periods of1965ndash2016 (14∘C) compared to previous period of 1912ndash1964(136∘C) and an increment about 05∘C for the periods of1981ndash2016 (141∘C) compared to previous period of 1912ndash1980(136∘C) respectively In addition the standard deviation isincreased by a coefficient between 01∘C and 02∘C in thewhole studied period The increment of the minimum valuesin the daily average temperature series is more evident thanthe maximum values by a mean coefficient of about 6 timesfor both of studied periodsTheminimumvalue of daily aver-age temperatures in the periods of 1912ndash1964 and 1965ndash2016revealed an increment of 06∘C and 08∘C respectively Themaximum value of daily average temperatures in the periodsof 1912ndash1964 and 1965ndash2016 showed a decrement of 02∘Cand 01∘C respectively When the monthly values of dailyaverage temperature are analyzed the highest increment in

the periods of 1912ndash1964 and 1965ndash2016 is found to be inJune at a rate of 1∘C The highest increment in the periodsof 1912ndash1980 and 1981ndash2016 appeared to be in June againwith a rate of 11∘C Also there has been temperature decre-ment in October November and December for all studiedperiods The highest decrement was in November with arate of minus05∘C in the periods of 1912ndash1964 and 1965ndash2016 aswell as minus04∘C in the periods of 1912ndash1980 and 1981ndash2016In this case the analysis of seasonal values of the dailyaverage temperatures showed that the highest increment inthe periods of 1912ndash1964 and 1965ndash2016 took place in thesummer season at a rate of 07∘C The highest increment inthe periods of 1912ndash1980 and 1981ndash2016 was found to be insummer again by a rate of 11∘C Also there was a temperaturedecrement in the autumn season which was about minus010∘Conly for the periods of 1912ndash1964 and 1965ndash2016 Analysisof the percentile thresholds of daily average temperaturesshowed that the temperature increment at the 5th percentileis 05∘C for the periods of 1912ndash1964 and 1965ndash2016 while thisincrement is 03∘C for the periods of 1912ndash1980 and 1981ndash2016Therefore it can be said that the rate of temperature risinghas increased further as much as the time is closer to the endyears of studied period Also these increment values indicatean increase of 02∘C and 03∘C at the 25th percentile 05∘Cand 08∘C at the 75th percentile and 06∘C and 11∘C at the95th percentile thresholds for thewhole studied subperiods of1912ndash1964 1965ndash2016 1912ndash1980 and 1981ndash2016 respectivelyThis situation shows that the increment of higher values inthe daily average temperatures is greater than the incrementof lower values and also this increment ismore evident for theperiods of 1912ndash1980 and 1981ndash2016 compared to the periodsof 1912ndash1964 and 1965ndash2016

Comparison of the daily maximum temperature seriesfor two periods of 1912ndash1964 and 1965ndash2016 revealed anincrement of 08∘C which can be divided into an average

Advances in Meteorology 9

1912

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

Annual total

Years

Annual maximum daily precip

Annual heavy precipitation STD

R2 = 00549

y = 12491x minus 16171

R2 = 00103

y = 0087x minus 1104

u(t)u(t)

u㰀(t)

u㰀(t)

R2 = 2E minus 08y = minus1E minus 05x + 92115

R2 = 00091

y = 00082x minus 99366

u(t)

u㰀(t)

u(t)

u㰀(t)

(a1)

(a2)

(b1)

(b2)

(c1)

(c2)

(d1)

(d2)

400500600700800900

1000110012001300

Am

ount

of p

reci

p (m

m)

Am

ount

of p

reci

p (m

m)

20

40

60

80

100

120

140

160

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

minus3

minus2

minus1

0

1

2

3

4

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus3

minus2

minus1

0

1

2

3

2016

1912

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1928

Years

2468

101214161820

Am

ount

of p

reci

p (m

m)

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

0

2

4

6

8

10

12

14

Num

ber o

f day

s (gt25

mm

)

minus3

minus2

minus1

0

1

2

3

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus3

minus2

minus1

0

1

2

3

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

2016

1912

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1928

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

Figure 5 Trend analysis of the precipitation time series using least squares linear regression and MK trend test for total annual precipitation(a1 a2) daily maximum precipitation (b1 b2) daily maximum precipitation greater than 25mm (c1 c2) and standard deviation in the dailyprecipitation series (d1 d2)

10 Advances in Meteorology

Table2Statisticalanalyses

outputso

ftem

perature

andprecipitatio

nparameters(SD

stand

arddeviation

Inc

increasedecdecrease)

Parameters

Minim

umtemperature

(∘ C)

Averagetem

perature

(∘ C)

Maxim

umtemperature

(∘ C)

Precipitatio

n(m

m)

Statistics

Perio

ds

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Incor

dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

Average

total

102

107

052

102

109

071

136

140

037

136

141

043

180

187

075

181

187

060

8113

8656

544

8267

8591

324

SD66

66

004

65

67

022

7172

007

7173

019

82

84

021

82

85

025

9391minus03

9193

02

Minim

umminus54minus42

116minus53minus39

138minus27minus21

062minus27minus19

074minus08minus01

076minus07

00

068

00

00

00

00

00

00

Maxim

um223

229

069

222

232

099

276

278

023

277

277

000

356

362

059

358

361

029

589

620

31

579

651

71Percentile5

minus04

02

061minus03

03

055

21

25

049

22

24

026

45

50

052

46

48

015

01

01minus01

01

01

00

Percentile2

547

52

045

48

52

038

7780

024

7778

01 0

114

119

054

115

116

011

07

05minus03

06

Percentile5

0106

108

019

105

106

007

140

141

013

140

138minus020

186

190

040

187

185minus024

26

21minus05

23

24

01

Percentile7

5159

165

066

158

164

059

199

204

054

199

201

022

250

261

105

252

255

034

7672minus04

7374

01

Percentile9

5196

203

074

195

202

074

236

242

065

236

239

032

296

307

110

297

300

027

229

226minus03

227

223minus05

gt30

Cggt80

mm

00

00

00

00

00

00

1626

1017

2810

00

00

00

gt25

Cggt40

mm

00

00

00

712

57

147

90107

1794

108

142

20

22

0lt0C

gmdash

2316minus6

2216minus6

76minus2

76minus1

21minus1

21minus1

mdashmdash

mdashmdash

mdashmdash

ltminus5C

gmdash

21minus1

21minus1

00

00

00

00

00

00

mdashmdash

mdashmdash

mdashmdash

Winter

33

36

032

33

38

048

59

62

031

60

62

020

9095

047

9195

035

1033

1096

63

1056

1066

09

Sprin

g74

87

131

7582

073

111

124

121

113

119

060

160

171

116

162

170

076

496

513

17515

511minus04

Summer

174

187

124

174

186

125

216

224

082

216

225

091

269

273

043

271

282

109

334

331minus03

330

368

38

Autumn

124

127

028

123

127

036

156

156

000

156

155minus001

199

197minus020

200

201

019

842

831minus10

854

919

65

Ann

ual

101

109

079

101

108

070

136

141

057

136

140

043

179

184

042

181

187

060

676

719

43

689

716

27

Advances in Meteorology 11

value of 18∘C for the period of 1912ndash1964 and an averagevalue of 187∘C for the period of 1965ndash2016 On the otherhand comparison of the daily maximum temperature seriesfor the periods of 1912ndash1980 and 1981ndash2016 has revealed anincrement of 06∘C with an average value between 182∘C and188∘C for both of these periods respectively In additionthe standard deviation values are increased between 02∘Cand 03∘C during the whole studied periodThe increment inthe minimum values of daily maximum temperature seriesis more evident than the maximum values The minimumvalues of daily maximum temperatures have exhibited anincrement of 08∘C for the period of 1965ndash2016 compared tothe previous period of 1912ndash1964 and also an increment of07∘C for the period of 1981ndash2016 compared to the previousperiod of 1912ndash1980 Also themaximumvalues of dailymaxi-mum temperature series exhibited amean increment of 05∘Cfor both studied periods Seasonal analysis of maximumtemperature series for both of studied section periods showedthat the highest rising temperature has been happened bya value of 12∘C in summer On the other hand the lowestincrement has happened in the autumn season The monthlyanalysis of daily maximum temperature series showed thatthe highest increment took place during the last monthsof the spring season and the first month of the summerseason in all of the studied periods When the percentiles ofdaily maximum temperatures are analyzed the temperatureincrement based on the 5th percentile threshold is 05∘C forthe periods of 1912ndash1964 and 1965ndash2016 while the incrementis 03∘C for the periods of 1912ndash1980 and 1981ndash2016 This canbe considered as an important sign of rising temperature overthe time In this case the value of percentile thresholds isincreased with extending the length of the time period withextending the length of the first section of studied period infavor of last years than the previous ones In this regard thesevalues are 05∘C and 04∘C at the 25th percentile 1∘C and11∘C at the 75th percentile and 11∘C and 12∘C at the 95thpercentile for both sections of studied periods respectivelyThis situation shows that the increment of higher values inthe dailymaximum temperatures is greater than lower values

The comparison analysis of daily minimum temperaturesbetween the periods of 1912ndash1964 and 1965ndash2016 and theperiods of 1912ndash1980 and 1981ndash2016 showed that there is ageneral increment of 05∘C during the first section periodswhich can be given as 102∘C and 107∘C for the individualperiods of 1912ndash1964 and 1965ndash2016 and also a generalincrement of 08∘C during the second section periods whichcan be provided as 102∘C and 11∘C for the individual periodsof 1912ndash1980 and 1981ndash2016 respectively In addition thestandard deviation values increased among these sectionperiods from 0∘C to 03∘C The increment in the minimumvalues of daily minimum temperature series is more evidentthan the maximum values The minimum values of dailyminimum temperature series showed an increment of 12∘Cfrom the period of 1912ndash1964 to 1965ndash2016 and also anincrement of 16∘C from the period of 1912ndash1980 to 1981ndash2016The maximum values of daily minimum temperature serieshave shown an increment of 07∘C for the section periodsof 1912ndash1964 and 1965ndash2016 and an increment of 12∘C forthe section periods of 1912ndash1980 and 1981ndash2016 Overall

the minimum values have had an average increment of14∘C while the maximum values have had an averageincrement of 1∘C during the last century which can showthe higher rate of upward trends in the temperature timeseries Monthly analysis of minimum temperature seriesshowed that the highest increment for the section studiedperiods of 1912ndash1964 and 1965ndash2016 has occurred by a valueof 1∘C in June while the highest increment for the sectionstudied periods of 1912ndash1980 and 1981ndash2016 has occurred bya value of 16∘C in August Meanwhile there is a temperaturedecrement in November in both of these periods Howeverthere is a clear decrement in the first half of these periodsin October and December whereas there is an incrementin the second half of these periods The seasonal analysisof the daily minimum temperature series showed that thehighest increment has happened in the summer season witha mean value of 07∘C for the section periods of 1912ndash1964and 1965ndash2016 and with a mean value of 13∘C for the sectionperiods of 1912ndash1980 and 1981ndash2016 respectively Then theincreasing rate during the summer season became moreevident during the recent decades Analysis of the percentilethresholds of daily minimum temperature series showed thatthe temperature increment at the 5th percentile is 06∘C forall studied time periods of 1912ndash1964 1965ndash2016 1912ndash1980and 1981ndash2016 Also these increment values indicate anincrease of 04∘C and 06∘C at the 25th percentile 07∘C and11∘C at the 75th percentile and 07∘C and 14∘C at the 95thpercentile thresholds for the whole studied subperiods of1912ndash1964 1965ndash2016 1912ndash1980 and 1981ndash2016 respectivelyThese rising rates in minimum temperature series are moreevident for the periods of 1912ndash1980 and 1981ndash2016 than theperiods of 1912ndash1980 and 1981ndash2016 Therefore it can be saidthat the rate of temperature rising has increased further aslong as the studied time period is closer to the last years

42 Precipitation Annual average precipitation in Istanbulis 838mm with a range of minimum value of 449mm in1921 and a maximum value of 1289mm in 1981 based on theobservatory data of Kandilli station during the whole studiedperiod from 1912 to 2016 (Figure 4(a)) Also monthly averageprecipitation is 699mm with a range of the minimum of326mm in July and the maximum of 1286mm in Decemberduring the whole studied time period (Figure 4(b))

Analysis of the trend in the annual average precipitationtime series by the methods of linear regression analysisand MK trend test has shown that periodically there arepartial increments and significant differences during the totalstudied period from 1912 to 2016 (Figures 5(a1) and 5(a2))But this increment in the precipitation time series is notas clear as the increment in the temperature time seriesHowever it is obvious in the precipitation time series thatthere is an increment between the years of 1917 and 1925a stable condition between the years of 1925 and 1954 anincrement between the years of 1954 and 1965 a decrementbetween the years of 1965 and 1974 an increment between theyears of 1974 and 2001 and again a no change situation from2001 till the end The trend analysis of the daily maximumprecipitation series showed an increment of 29mm for theperiods of 1912ndash1964 and 1965ndash2016 as well as an increment

12 Advances in Meteorology

of 93mm for the periods of 1912ndash1980 and 1981ndash2016 Thesevalues indicated that rainfall has increased at a rate of morethan three times over the last decades than the previous ones(Figures 5(b1) and 5(b2)) Also the number of days with dailyprecipitation greater than 25mm presented an incrementduring the period of 1912ndash2016 although this is not significantat the confidence level of 005 (Figures 5(c1) and 5(c2))The trend of standard deviation in the daily precipitationtime series showed a slowly increasing trend during the totalstudied period (Figures 5(d1) and 5(d2))

The comparison of the results of statistical analysisbetween the daily average rainfall amounts belonging tothe periods of 1912ndash1964 and 1965ndash2016 with those of theperiods of 1912ndash1980 and 1981ndash2016 revealed that there is anincrement of 78mm from the period of 1912ndash1964 to theperiod of 1965ndash2016 and an increment of 38mm from theperiod of 1912ndash1980 to the period of 1981ndash2016 In additionthe analysis of standard deviation exhibited a decrementfrom the period of 1912ndash1964 to the period of 1965ndash2016 andan increment from the period of 1912ndash1980 to the periodof 1981ndash2016 (Table 2) The analysis of monthly averageprecipitation time series showed that the highest incrementhas happened in October with 227mm for the sectionperiods of 1912ndash1964 and 1965ndash2016 and with 347mm forthe section periods of 1912ndash1980 and 1981ndash2016 On the otherhand the highest decrease took placewith a value ofminus04mmduring both May and July for the periods of 1912ndash1964 and1965ndash2016 and with a value of minus16mm in September for thesection periods of 1912ndash1980 and 1981ndash2016 The analysis ofseasonal average precipitation time series showed that thehighest increase occurred in autumn with a value of 72mmfor the section periods of 1912ndash1964 and 1965ndash2016 as wellas with a value of 64mm for the periods of 1912ndash1980 and1981ndash2016 respectively There is no remarkable decreasingchange seasonally except in winter season during the periodsof 1912ndash1980 and 1981ndash2016 Furthermore the percentilethresholds of daily average precipitation indicated that thereis an insignificant negative trend based on all percentilesand for all studied periods Overall the statistical analysisshowed that the total average precipitation of Istanbul hasincreased while this increasing trend is more pronounced inthe previous decades than the last 3 decades On the otherhand the increasing rate of daily maximum precipitation ismore evident during the last 3 decades than the previousdecades which can be proven by the increasing frequency ofheavy rainfall events in Istanbul

Generally the results of trend analysis of Kandilli stationduring the last 105 years of 1912ndash2016 showed that there isa warming significant trend in the precipitation time seriesby using both methods of linear regression analysis andMK trend test On the contrary previous climate studiesconducted over Turkey put forward that there has been adecreasing trend in annual precipitation time series duringthe recent decades regionally The results of a previouslyconducted study by using the daily precipitation and tem-perature data sets of Florya and Goztepe meteorologicalstations in Istanbul area between 1960 and 2013 showedthat most notably the precipitation during the warm periodshas decreased but the frequency of the intense rain has

increased and the majority of these episodes of intense raincoincided with the warm periods Other determinationswere the rise in the annual average temperature and theextension of the warm periods in a year This differentiationof the temperature features can lead to the aggravation of theevaporation and it can be effective for a longer period duringthe year [15] Thus it will make Istanbul be confronted withthe much more important problems of water managementand flood [34] Also the results of the current study for trendanalysis in the long period from 1912 to 2016 showed that themost striking spell is between the years of 1968 and 1998 dueto the existence of least number of rainfall events in IstanbulIt can be owing to industrialization along with the increasingair pollution as well as irregular urbanization in Istanbul areaIn this case severe droughts taking place during the yearsof 1988 1992 1993 and 2008 have threatened the reservoirswhich supplied fresh water of the city These years werecharacterized by not having enough rainfall events Theseyears are also characterized by more persistent high-pressuresystems and less occurrence of low-pressure systems in termsof number and strength As statements made by officialinstitutions the formation conditions for the atmosphericlayer of air pollution due to air pollutant emissions from fossilfuel combustion and industrial activities are more providedduring anticyclone or high-pressure system eventsThereforethis leads to warming up and generating an inversion layerin the boundary layer of atmosphere especially over city area[35] The inversion layer or urban heat island intensity isincreasing with the increasing city size andor populationa phenomenon that was also reported by others [36ndash38]Moreover these last climatic events also have affected somepolitical results beyond natural effects The most obviousexample of this is related to the local election of 1994 inIstanbul In this case extreme drought during the summers of1992 and 1993 has caused groundwater reservoirs to dry up inthe city discontinuance of water was experienced for severaldays or even weeks In those times the mayor of Istanbul cityhas lost the election of 1994 and he realized that this result wasdue to the peoplersquos reaction about the water shortages Manypolitical reports of that period also support this scientificview

5 Conclusions

Statistical analysis in temperature time series of the Kandillistation from 1912 to 2016 established that there is a notableincrease in temperature values after the 1940s which is inparallel with the beginning of industrialization era in Istan-bul There has been a rise about 094∘C in the daily averagetemperature series since the beginning of the last century Asignificant positive trend in the daily maximum temperatureseries is found about 156∘C Also there is a positive trendabout 087∘C in the daily minimum temperature series Onthe other hand analysis of the number of days with thedaily maximum temperature higher than 30∘C showed thatthere is an increasing trend Meanwhile analysis of thenumber of days with daily minimum temperatures lowerthan 0∘C showed a decreasing trend The increment in theminimum values of the daily minimum temperature series

Advances in Meteorology 13

is more evident than the maximum values of this series Inthis case these rising rates in the minimum temperatureseries are more evident for the section periods of 1912ndash1980and 1981ndash2016 than the section periods of 1912ndash1964 and1965ndash2016 This again shows that there is an increment inthe positive temperature trend from past to present decadesThe increment in the precipitation time series is not asclear as the increment in the temperature time series dueto periodic variability The trend analysis in the total annualprecipitation time series showed that the first significantupward trend has periodically been started from the 1920swhile there is a stable trend from 2001 till 2016 The dailyaverage of rainfall amounts has increased with a value of58mm during the period of 1912ndash2016 Also the analysisof heavy precipitation trend showed an increase of 61mmOverall the total average precipitation of Istanbul increasedwhile this increasing trend is more pronounced during theearly decades than the last 3 decades On the other handthe increasing rate of daily maximum precipitation is morepronounced in the last 3 decades than the previous decadesThen it was shown that the frequency of heavy rainfallat Istanbul has increased during the recent decades Thusthe precipitation changes in Istanbul have some differencescompared to the general tendency in precipitation trendthat was put forward by other studies as a decreasing trendover the whole of Turkey This result can be expressedas a positive effect of population overgrowth of Istanbulmegacity Comparison of the results in the first half of thestudy period (1912ndash1964) with the second half of the studyperiod (1965ndash2016) showed that both the average temperatureand average precipitation have higher values of 139∘C and878mm for the final phase compared to the values of 136∘Cand 799mm belonging to the initial phase Therefore it canbe stated that the megacity of Istanbul is directly affectedby the climate change and its consequences In this contextpotential risks of climate change in Istanbul megacity underhigher temperature conditions can be expressed as the rise inthe sea level increase in the rate of evapotranspiration andincrease in the frequency of heavy rainfall Also this city maynot be able to handle this uncontrolled population growthand its associated irreversible changes which is alreadypushing the natural limits by destroying the environmentTherefore the local governors of any megacity like Istanbulshould give more emphasis on the importance of sustainableurban development Thus it is urgent to prepare local andnational climate change strategies and action plans for themegacities

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

The authors are grateful to the Bogazici University theobservatory of Kandilli weather station and the Earthquake

Research Institute for providing the research data and tech-nical support The authors also gratefully acknowledge con-tributions of Assoc Professor Dr Yuksel Demirkaya Schoolof Social Sciences Marmara University This work has beensupported by Scientific and Technological Research Councilof Turkey (TUBITAK) under Grants 113R019 and 106Y258and by Marmara University (BAPKO) with projects FEN-E-120314-0066 FEN-C-YLP-090414-0102 FEN-L-250416-0180 and FEN-A-100413-0127

References

[1] D R Easterling B Horton P D Jones et al ldquoMaximum andminimum temperature trends for the globerdquo Science vol 277no 5324 pp 364ndash367 1997

[2] IPCC Climate Change 2014 IPCC Fifth Assessment Synthe-sis Report-Summary for Policymakers-an Assessment of Inter-Governmental Panel on Climate Change Cambridge UniversityPress Cambridge UK 2014

[3] J Carmin N Nadkarni and C Rhie Progress and Challenges inUrban Climate Adaptation Planning Results of a Global SurveyMIT Cambridge UK 2012

[4] A F Young ldquoUrban expansion and environmental risk in theSao Paulo Metropolitan Areardquo Climate Research vol 57 no 1pp 73ndash80 2013

[5] P Tian X Mu J Liu J Hu and C Gu ldquoImpacts of ClimateVariability and Human Activities on the Changes of Runoff andSediment Load in a Catchment of the Loess Plateau ChinardquoAdvances inMeteorology vol 2016 Article ID 4724067 15 pages2016

[6] R S Kovats and K L Ebi ldquoHeatwaves and public health inEuroperdquo European Journal of Public Health vol 16 no 6 pp592ndash599 2006

[7] S Conti P Meli G Minelli et al ldquoEpidemiologic studyof mortality during the Summer 2003 heat wave in ItalyrdquoEnvironmental Research vol 98 no 3 pp 390ndash399 2005

[8] J Kysely and J Kim ldquoMortality during heat waves in SouthKorea 1991 to 2005 how exceptional was the 1994 heat waverdquoClimate Research vol 38 no 2 pp 105ndash116 2009

[9] B Yan Z Xia F Huang L Guo and X Zhang ldquoClimatechange detection and annual extreme temperature analysis ofthe amur river basinrdquoAdvances inMeteorology vol 2016 ArticleID 6268938 14 pages 2016

[10] E M Fischer and R Knutti ldquoAnthropogenic contribution toglobal occurrence of heavy-precipitation and high-temperatureextremesrdquo Nature Climate Change vol 5 no 6 pp 560ndash5642015

[11] X Zhang L Alexander G C Hegerl et al ldquoIndices for moni-toring changes in extremes based on daily temperature andprecipitation datardquo Climate Change vol 2 no 6 pp 851ndash8702011

[12] K H Schlunzen P Hoffmann G Rosenhagen and W RieckeldquoLong-term changes and regional differences in temperatureand precipitation in the metropolitan area of Hamburgrdquo Inter-national Journal of Climatology vol 30 no 8 pp 1121ndash11362010

[13] G Bartolini M Morabito A Crisci et al ldquoRecent trends inTuscany (Italy) summer temperature and indices of extremesrdquoInternational Journal of Climatology vol 28 no 13 pp 1751ndash1760 2008

14 Advances in Meteorology

[14] S C Sheridan and T J Dolney ldquoHeat mortality and levelof urbanization measuring vulnerability across Ohio USArdquoClimate Research vol 24 no 3 pp 255ndash265 2003

[15] M Tayanc U Im M Dogruel andM Karaca ldquoClimate changein Turkey for the last half centuryrdquo Climatic Change vol 94 no3-4 pp 483ndash502 2009

[16] H Toros ldquoSpatio-temporal variation of daily extreme tempera-tures over Turkeyrdquo International Journal of Climatology vol 32no 7 pp 1047ndash1055 2012

[17] H Toros ldquoSpatio-temporal precipitation change assessmentsover Turkeyrdquo International Journal of Climatology vol 32 no9 pp 1310ndash1325 2012

[18] M Turkes C Yozgatlıgil I Batmaz et al ldquoHas the climate beenchanging in Turkey Regional climate change signals based on acomparative statistical analysis of two consecutive time periods1950-1980 and 1981-2010rdquoClimate Research vol 70 no 1 pp 77ndash93 2016

[19] IPCC ldquoummary for Policymakers In Climate Change 2013rdquo inThe Physical Science Basis The contribution of Working Group Ito the Fifth Assessment Report of the Intergovernmental Panel onClimate Change Cambridge University Press Cambridge UK2013

[20] United Nations ldquoDepartment of Economic and Social AffairsPopulation Division 2006 World Urbanization Prospects The2005 Revisionrdquo Working Paper ESAPWP200 2011

[21] Y Demirkaya Sayılarla Istanbul ITO Istanbul Turkey 2011[22] TUIK ldquoAddress based population registration system results of

2014rdquo Turkish Statistical Institute (TUIK) 2017 httpraporytuikgovtr10-03-2015-184727-842632346446643191142126876html

[23] S Erinc Climatology and its Methods Alfa Basım YayımDagitim Istanbul Turkey 4th edition 1965

[24] O M Gokturk D Bozkurt O L Sen and M Karaca ldquoQualitycontrol and homogeneity of Turkish precipitation datardquoHydro-logical Processes vol 22 no 16 pp 3210ndash3218 2008

[25] C Ley and D Paindaveine ldquoRuns Testsrdquo in Encyclopedia ofEnvironmetrics 2012

[26] A Ghasemi and S Zahediasl ldquoNormality tests for statisticalanalysis a guide for non-statisticiansrdquo International Journal ofEndocrinology andMetabolism vol 10 no 2 pp 486ndash489 2012

[27] G V Glass ldquoTesting Homogeneity of Variancesrdquo AmericanEducational Research Journal vol 3 no 3 pp 187ndash190 1966

[28] H BMann ldquoNonparametric tests against trendrdquo Econometricavol 13 pp 245ndash259 1945

[29] M G Kendall Rank Correlation Method Charles GriffinLondon UK 4th edition 1975

[30] H Turoglu ldquoDetection of Changes on Temperature and Precip-itation Features in Istanbul (Turkey)rdquo Atmospheric and ClimateSciences vol 04 no 04 pp 549ndash562 2014

[31] A Karaburun A Demirci and I-S Suen ldquoImpacts of urbangrowth on forest cover in Istanbul (1987-2007)rdquo EnvironmentalModeling amp Assessment vol 166 no 1-4 pp 267ndash277 2010

[32] K K Karanth LM Curran and J D Reuning-Scherer ldquoVillagesize and forest disturbance in Bhadra Wildlife SanctuaryWestern Ghats Indiardquo Biological Conservation vol 128 no 2pp 147ndash157 2006

[33] G Cakir C Un E Z Baskent S Kose F Sivrikaya andS Keles ldquoEvaluating urbanization fragmentation and landuseland cover change pattern in Istanbul city Turkey from 1971to 2002rdquo Land Degradation amp Development vol 19 no 6 pp663ndash675 2008

[34] R B Myneni F G Hall P J Sellers and A L Marshak ldquoTheinterpretation of spectral vegetation indexesrdquo IEEE Transac-tions on Geoscience and Remote Sensing vol 33 no 2 pp 481ndash486 1995

[35] Y S Unal H Toros A Deniz and S Incecik ldquoInfluence ofmeteorological factors and emission sources on spatial and tem-poral variations of PM10 concentrations in Istanbul metropoli-tan areardquo Atmospheric Environment vol 45 no 31 pp 5504ndash5513 2011

[36] M KaracaM Tayanc andH Toros ldquoEffects of urbanization onclimate of Istanbul and Ankarardquo Atmospheric Environment vol29 no 23 pp 3411ndash3421 1995

[37] Y Ezber O L Sen T Kindap and M Karaca ldquoClimatic effectsof urbanization in Istanbul a statistical and modeling analysisrdquoInternational Journal of Climatology vol 27 no 5 pp 667ndash6792007

[38] H S Park ldquoFeatures of the heat island in seoul and its sur-rounding citiesrdquo Atmospheric Environment (1967) vol 20 no10 pp 1859ndash1866 1986

4 Advances in Meteorology

MK test is applicable in cases when the data values 119883119894 of atime series can be assumed to obey the model

119883119894 = 119891 (119905119894) + 120576119894 (1)

where 119891(119905) is a continuous monotonic increasing or decreas-ing function of time the residuals 120576119894 can be assumed to befrom the same distribution with zero mean

Then an upward or downward trend is given by a positiveor negative value of 119885 For this purpose at first the varianceof 119878 is computed using the following equation

VAR (119878) =119899 (119899 minus 1) (2119899 minus 5) minus sum119898

119894=1119905 (119905 minus 1) (2119905 + 5)

18 (2)

where119898 is the number of tied groups 119905 is the number of datavalues in the 119894 group Then the values of 119878 and VAR(119878) areused to compute the MK test statistic of119885 as is present in (3)(119885 119886 119887 119888)

119885119886 =119878 minus 1

radicVAR (119878) if 119878 gt 0

119885119887 = 0 if 119878 = 0

119885119888 =119878 + 1

radicVAR (119878) if 119878 lt 0

(3)

Moreover remote sensing phenology studies use data gath-ered by satellite sensors which measure wavelengths of lightabsorbed and reflected by green plants Certain pigmentin plant leaves strongly absorbs wavelengths of visible light(red) The leaves themselves strongly reflect wavelengths ofnear-infrared light which is invisible to human eyes Thishas long been used to monitor the vegetation and changesin the vegetation of the entire Earth Although there areseveral vegetation indices one of the most widely used is theNormalized Difference Vegetation Index (NDVI) (see (4))

NDVI = (Band 4 minus Band 3)(Band 4 + Band 3)

(4)

where RED (Band 3) and NIR (Band 4) stand for the spectralreflectance measurements acquired in the visible (RED) andnear-infrared (NIR) regions respectively NDVI values rangeis from +01 to minus01 A zero means no vegetation and closeto +1 (08ndash09) indicates the highest possible density of greenleaves Areas of barren rock sand or snow usually showvery low NDVI values Sparse vegetation such as shrubs andgrasslands or senescing crops may result in moderate NDVIvalues High NDVI values correspond to dense vegetationsuch as that found in temperate and tropical forests or cropsat their peak growth stage [30] In this case the NDVIindex is extracted based on accessible remote sensing imagesof the Landsat-7 ETM+ during the years of 1985 to 2015(refer to Figure 1(d)) Finally statistical analysis of populationgrowth rates in Istanbul city is calculated during the yearsof 2000 to 2017 to better understand the trends in climaticparameters according to anthropogenic behaviors affected bygrowing urbanization in Istanbul area over the time (refer toFigure 1(c))

4 Results and Discussion

Trend analysis of the temperatures and precipitation seriesusing the method of the linear best-fit curve and Man-Kendall test are applied for subperiods throughout the wholeperiod of 1912ndash2016 In this case the first half of studiedperiod (1912ndash1964) is compared to second half of studiedperiod (1964ndash2016) for knowing the actual changes in cli-matic parameters Also since the globe on average was hotterin the early 1960s according to fifth assessment report ofIPCC in 2013 a comparing analysis between two subperiodsof 1912ndash1980 and 1981ndash2016 is performed by choosing theturning point in the year of 1980 In this case the resultsof basic quality tests for all studied periods showed that thedistribution of all-time series at the significance level of 95does not follow the randomness principle by use of Run testand the normality principle by using Kolmogorov-Smirnovtest for both of Kandilli station and its nearby station ofSariyer Then the results of Levenersquos test of homogeneityand ANOVA test for assessing the absolute homogeneityshowed that minimum temperature and precipitation timeseries of the Kandilli station are homogeneous by comparisonbetween two subperiodsrsquo data including 1st and 2nd parts ofthe entire period of 1912ndash2016 Also in the case of assessingthe relative homogeneity between two separate data groupsincluding the Kandilli station and its nearby station of Sariyershowed that the average temperature and precipitation timeseries are homogeneous for the same studied period of1965ndash2016 Moreover results of the homogeneity test showedno homogeneity in the other series whether between meteo-rological series at the Kandilli station and its nearby station ofSariyer in the same period or between two subperiods of theKandilli stationrsquos series (Table 1)These uneven results of basictests onto data time periods and subperiods can be affectedby urbanization along with global warming effects So thepopulation of Istanbul has almost increased by a factor of 17from 1927 to 2016 and the effects of population growth arechanged in the land use and land cover Also it is possible tosee the different views of Istanbul for vegetation changes byusing the Normalized Difference Vegetation Index (NDVI)over the time from 1985 to 2015 (see Figure 1(d)) In this caseadverse effect of urbanization is the decline in forest coverMany countries have witnessed severe deforestation due torapid urban growth On the global scale the world lost 3 ofits total forest area between 1990 and 2005 with an average02 decrease per year Urban areas in Istanbul have alsoexpanded about 879 from 1977 to 2007 while forest areasdeclined about 54 in the same period One of the mostsignificant results of the study was that total forest areas inIstanbul have almost increased about 03 between 2000 and2007 [31] The increase of forest areas as opposed to growthin population size may well be explained by the movementof population from rural areas to Istanbul city Thus it canprovide an opportunity to release human pressure from forestareas probably resulting in a positive development of forest[32 33] The same period has also witnessed an increase in arate of 31 times in the cityrsquos population during the last thirtyyears Thus Istanbulrsquos population is rapidly increasing andconsequently the residential settlement is widening This

Advances in Meteorology 5

Table1Statisticalou

tputso

falltests

ford

ataq

ualitycontrol

Cases

NPartestslowast

Testof

norm

alitylowast

Testof

homogeneityof

varia

ncelowast

Parameter

Station

Perio

dTo

tal

Valid

Missing

Std

Runs

test

Kolm

ogorov-Smirn

ovANOVA

test

Levenersquostest

119873(day)119873(

)119873(day)

119885119886

Sig

Mean

Testvalues

Statistic

Sig

119865Sig

Statistic

Sig

119879min

Kand

illi

1965ndash2016

18980

999

2366minus137

00

105

105

0065

00

367

00

2100

Sarıy

er18998

100

167minus136

00

109

109

0061

00

Kand

illi

1st

38353

999

2366minus170

00

104

107

067

00

686

00

183

017

2nd

119879avg

Kand

illi

1965ndash2016

18985

100

471minus138

00

139

139

007

00

031

057

108

029

Sarıy

er18998

100

171minus135

00

139

138

0069

00

Kand

illi

1st

38353

100

471minus169

00

138

141

007

00

332

00

509

002

2nd

119879max

Kand

illi

1965ndash2016

18990

999

1884minus130

00

186

186

007

00

249

00

204

00

Sarıy

er18998

100

178minus129

00

175

175

006

800

Kand

illi

1st

38335

100

1883minus161

00

183

188

006

00

894

00

272

00

2nd

Precip

Kand

illi

1965ndash2016

7691

389

12238

96minus13

00

612

612

026

00

104

031

135

024

Sarıy

er7855

35

11074

92minus10

00

627

63

00253

00

Kand

illi

1st

38353

371

24121

96minus164

00

61

25

026

00

700

0799

037

2nd

ldquoardquom

eans

basedon

them

ediansta

tistic

allylowastshow

sthe

significantlevelof

test

ldquo1strdquo

show

sfirstsectio

nof

thes

tudied

perio

d1912ndash196

4ldquo2nd

rdquosho

wssecon

dsectionof

studied

perio

d1965ndash2016

6 Advances in Meteorology

120

125

130

135

140

145

150

155

160Av

erag

e tem

pera

ture

(∘C)

1916

1920

1924

1928

1932

1936

1940

1944

1948

1952

1956

1960

1964

1968

1972

1976

1980

1984

1988

1992

1996

2000

2004

2008

2012

2016

1912

Years

Average 138∘CMaximum 153∘C in 1966Minimum 124∘C in 1920

R2 = 01644y = 0009x + 13315

R2 = 0075y = 00097x + 13592

R2 = 04619

y = 00514x + 132671912ndash2016 1951ndash2012 1984ndash2016

(a)

Sept

embe

r

Dec

embe

rN

ovem

ber

Febr

uary

Janu

ary

June July

Augu

st

Oct

ober

Mar

ch

May

April

Months

Average 137∘CMaximum 227 ∘C in July

Minimum 52∘C in January

00

50

100

150

200

250

Aver

age t

empe

ratu

re (∘

C)(b)

Figure 2 Trend analysis of average annual temperature (a) and average monthly temperature (b)

is the main reason why the forests as Istanbulrsquos life sourceare being destroyed Overall this suggests that progress hasbeen made in Istanbul not only in having sustainable urbangrowth but also in preserving restoring and even expandingforest areas especially after the year 2000 In the followingafter assessing the population growth and land use change inIstanbul the results obtained from main climatic parameterstrend are analyzed for previous decades compared to lastdecades

41 Temperature The analysis of average yearly maximumand minimum temperature has been carried out using thestatistical method of least squares and the MK test at thesignificance level of 5 is shown with a dashed line (plusmn2)in Figures 2ndash5 The results of linear trends and MK rankcorrelation test showed that there is an increase in the averagedaily temperature series for the whole period of 1912ndash2016 inIstanbul (Figure 2(a)) However this upward trend includesimportant differences While there is not any significantchange in the annual temperature time series from 1912to 1944 the upward trend became apparent between 1944and 1969 A cool period exists in the period of 1969ndash1993and after that a significant rising trend is pronounced inaverage annual temperature at the 95 confidence levelThere has been a rise of about 094∘C in the average dailytemperature over the last 100 years at Kandilli Temperaturehas increased 060∘C in the period of 1951ndash2016 and 170∘C inthe period of 1981ndash2016 In this case the global combined landand ocean surface temperatures show an increase of about

089∘C (069∘Cndash108∘C) over the period of 1901ndash2016 andabout 072∘C (049∘Cndash089∘C) over the period of 1951ndash2016according to fifth assessment report of IPCC in 2013

Nevertheless the data time series of Kandilli stationshows that the increase in temperature parameter after the1940s is in parallel with the beginning of industrialization erain Istanbul Of course regime changes in temperature timeseries on regional scales cannot be totally explained by thenature destruction and pollutant emissions in and aroundIstanbul alone These effects can also be considered as areflection of the general situation happening in the worldon a larger scale Temperature time series of the Kandillistation shows an annual average about 137∘C and absoluteminimum and maximum in 1920 with a value of 124∘C andin 1966 with a value of 153∘C respectively (Figure 2(a))Owing to the Mediterranean climate of the study areamonthly temperature time series is increased from FebruarytoAugust and is decreased fromAugust to February reachinga minimum of 52∘C in February and a maximum of 227∘Cin August (Figure 2(b))

A general tendency of a warming trend in the dailytemperature series is found for the whole studied periodof 1912ndash2016 by using the least squares regression analysisTherefore the trend analysis revealed that the daily averageof temperature has increased by a rate of 09 for the periodfrom 1912 to 2016 (Figures 3(b1) and 3(b2)) Also there is apositive trend about 149∘C in daily maximum temperatureseries during the whole studied period Furthermore theincreasing trends have notably occurred in two subperiods

Advances in Meteorology 7

Tmin Tavg Tmax

Tmax Tmin DTR

u(t)u(t)

u(t)

u㰀(t)u㰀(t)

u㰀(t)

u(t)u㰀(t)

u㰀(t)

u(t)

u(t)

R2 = 02642

y = 00149x minus 10794R2 = 0163

y = 0009x minus 3926

R2 = 0233

y = 00116x minus 12375

R2 = 00182y = 00032x + 16875

R2 = 00734

y = minus00934x + 20292R2 = 02513

y = 02039x minus 37934

(a1)

(a2)

(b1)

(b2)

(c1)

(c2)

(d1)

(d2)

(e1)

(e2)

(f1)

(f2)

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1912

1976

1984

1992

2000

2008

2016

1968

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

2016

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1912

1976

1984

1992

2000

2008

2016

1968

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

minus4minus3minus2minus1

01234567

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus6minus5minus4minus3minus2minus1

0123456

minus4minus3minus2minus1

0123

010203040506070

Num

ber o

f day

s (gt30∘ C)

0102030405060

65707580859095

100

Tem

pera

ture

(∘C)

minus3minus2minus1

0123456

minus3minus2minus1

012345

minus3minus2minus1

01234567

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

9

10

11

12

13Te

mpe

ratu

re (∘

C)

12

13

14

15

16

Tem

pera

ture

(∘C)

16

17

18

19

20

21

Tem

pera

ture

(∘C)

Num

ber o

f day

s (lt0∘ C)

Figure 3 Trend analysis of the temperatures time series using linear best-fit curve andMK test for daily minimum temperature (a1 a2) dailyaverage temperature (b1 b2) daily maximum temperature (c1 c2) number of days with daily maximum temperature gt 30∘C (d1 d2) numberof days with daily minimum temperature lt 0∘C (e1 e2) and daily temperature range (f1 f2)

between 1944 and 1969 and between 1991 and 2016 (Figures3(c1) and 3(c2)) There is a positive trend about 11∘C indaily minimum temperature series during the whole studiedperiod too while this positive trend value is smaller than thepositive trend value of maximum temperature The trend ofminimum temperature series is stable during the period of1912ndash1944 Then it shows a steady increase during 1942 to1989 which is most pronounced during 1944ndash1954 Finallythe rising trends in the minimum temperature series became

significant based onMK testrsquos result for the last period of 2003to 2016 (Figures 3(a1) and 3(a2)) In this case the minimumtemperature shows increasing change for all season wherethe most increasing change has happened in summer andspring season in all studied periods respectively (Table 2)Analysis of the number of days having daily maximumtemperature higher than 30∘C for the long time period of1912 to 2016 by focusing precisely on its subperiods showedthat there is almost no trend during 1912ndash1945 a significant

8 Advances in Meteorology

400

500

600

700

800

900

1000

1100

1200

1300

1912 1920 1928 1936 1944 1952 1960 1968 1976 1984 1992 2000 2008 2016

Tota

l pre

cipi

tatio

n (m

m)

Years

Average 8375 mmMaximum 12894 mm in 1981Minimum 4487 mm in 1921

(a)

0

20

40

60

80

100

120

140

Janu

ary

Febr

uary

Mar

chAp

rilM

ayJu

ne July

Augu

stSe

ptem

ber

Oct

ober

Nov

embe

rD

ecem

ber

Tota

l pre

cipi

tatio

n (m

m)

Months

Average 699 mmMaximum 1284 mm in December

Minimum 319 mm in July

(b)

Figure 4 Time series of total annual precipitation (a) and total monthly precipitation (b)

increasing trend during 1945ndash1969 a relative reduction trendduring 1969ndash1977 an increasing trend during 1977ndash1992and a more pronounced increasing trend during 1992ndash2007(Figures 3(d1) and 3(d2)) On the other hand analysis of thenumber of dayswith dailyminimum temperatures lower than0∘C shows a decreasing trend generally So there is a relativeincrease during 1915ndash1930 almost no trend during 1930ndash1954an increase during 1954ndash1963 a decrease during 1963ndash1985almost no trend during 1985ndash1995 and a decrease during1995ndash2011 by a detailed focus on subperiods of the time series(Figures 3(e1) and 3(e2)) When the daily temperature rangein the whole long period is analyzed an increment can beclearly observed The situation was stable in the period of1912ndash1950There is an increasing trend during 1950ndash1970 andan obvious decrement during 1970ndash1985 Also there is anincrease between 1985 and 2010 (Figures 3(f1) and 3(f2))

The comparison results between the periods of 1912ndash1964and 1965ndash2016 and the periods of 1912ndash1980 and 1981ndash2016showed that (Table 1) there is an increment in the dailyaverage temperature series about 04∘C for the periods of1965ndash2016 (14∘C) compared to previous period of 1912ndash1964(136∘C) and an increment about 05∘C for the periods of1981ndash2016 (141∘C) compared to previous period of 1912ndash1980(136∘C) respectively In addition the standard deviation isincreased by a coefficient between 01∘C and 02∘C in thewhole studied period The increment of the minimum valuesin the daily average temperature series is more evident thanthe maximum values by a mean coefficient of about 6 timesfor both of studied periodsTheminimumvalue of daily aver-age temperatures in the periods of 1912ndash1964 and 1965ndash2016revealed an increment of 06∘C and 08∘C respectively Themaximum value of daily average temperatures in the periodsof 1912ndash1964 and 1965ndash2016 showed a decrement of 02∘Cand 01∘C respectively When the monthly values of dailyaverage temperature are analyzed the highest increment in

the periods of 1912ndash1964 and 1965ndash2016 is found to be inJune at a rate of 1∘C The highest increment in the periodsof 1912ndash1980 and 1981ndash2016 appeared to be in June againwith a rate of 11∘C Also there has been temperature decre-ment in October November and December for all studiedperiods The highest decrement was in November with arate of minus05∘C in the periods of 1912ndash1964 and 1965ndash2016 aswell as minus04∘C in the periods of 1912ndash1980 and 1981ndash2016In this case the analysis of seasonal values of the dailyaverage temperatures showed that the highest increment inthe periods of 1912ndash1964 and 1965ndash2016 took place in thesummer season at a rate of 07∘C The highest increment inthe periods of 1912ndash1980 and 1981ndash2016 was found to be insummer again by a rate of 11∘C Also there was a temperaturedecrement in the autumn season which was about minus010∘Conly for the periods of 1912ndash1964 and 1965ndash2016 Analysisof the percentile thresholds of daily average temperaturesshowed that the temperature increment at the 5th percentileis 05∘C for the periods of 1912ndash1964 and 1965ndash2016 while thisincrement is 03∘C for the periods of 1912ndash1980 and 1981ndash2016Therefore it can be said that the rate of temperature risinghas increased further as much as the time is closer to the endyears of studied period Also these increment values indicatean increase of 02∘C and 03∘C at the 25th percentile 05∘Cand 08∘C at the 75th percentile and 06∘C and 11∘C at the95th percentile thresholds for thewhole studied subperiods of1912ndash1964 1965ndash2016 1912ndash1980 and 1981ndash2016 respectivelyThis situation shows that the increment of higher values inthe daily average temperatures is greater than the incrementof lower values and also this increment ismore evident for theperiods of 1912ndash1980 and 1981ndash2016 compared to the periodsof 1912ndash1964 and 1965ndash2016

Comparison of the daily maximum temperature seriesfor two periods of 1912ndash1964 and 1965ndash2016 revealed anincrement of 08∘C which can be divided into an average

Advances in Meteorology 9

1912

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

Annual total

Years

Annual maximum daily precip

Annual heavy precipitation STD

R2 = 00549

y = 12491x minus 16171

R2 = 00103

y = 0087x minus 1104

u(t)u(t)

u㰀(t)

u㰀(t)

R2 = 2E minus 08y = minus1E minus 05x + 92115

R2 = 00091

y = 00082x minus 99366

u(t)

u㰀(t)

u(t)

u㰀(t)

(a1)

(a2)

(b1)

(b2)

(c1)

(c2)

(d1)

(d2)

400500600700800900

1000110012001300

Am

ount

of p

reci

p (m

m)

Am

ount

of p

reci

p (m

m)

20

40

60

80

100

120

140

160

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

minus3

minus2

minus1

0

1

2

3

4

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus3

minus2

minus1

0

1

2

3

2016

1912

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1928

Years

2468

101214161820

Am

ount

of p

reci

p (m

m)

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

0

2

4

6

8

10

12

14

Num

ber o

f day

s (gt25

mm

)

minus3

minus2

minus1

0

1

2

3

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus3

minus2

minus1

0

1

2

3

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

2016

1912

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1928

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

Figure 5 Trend analysis of the precipitation time series using least squares linear regression and MK trend test for total annual precipitation(a1 a2) daily maximum precipitation (b1 b2) daily maximum precipitation greater than 25mm (c1 c2) and standard deviation in the dailyprecipitation series (d1 d2)

10 Advances in Meteorology

Table2Statisticalanalyses

outputso

ftem

perature

andprecipitatio

nparameters(SD

stand

arddeviation

Inc

increasedecdecrease)

Parameters

Minim

umtemperature

(∘ C)

Averagetem

perature

(∘ C)

Maxim

umtemperature

(∘ C)

Precipitatio

n(m

m)

Statistics

Perio

ds

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Incor

dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

Average

total

102

107

052

102

109

071

136

140

037

136

141

043

180

187

075

181

187

060

8113

8656

544

8267

8591

324

SD66

66

004

65

67

022

7172

007

7173

019

82

84

021

82

85

025

9391minus03

9193

02

Minim

umminus54minus42

116minus53minus39

138minus27minus21

062minus27minus19

074minus08minus01

076minus07

00

068

00

00

00

00

00

00

Maxim

um223

229

069

222

232

099

276

278

023

277

277

000

356

362

059

358

361

029

589

620

31

579

651

71Percentile5

minus04

02

061minus03

03

055

21

25

049

22

24

026

45

50

052

46

48

015

01

01minus01

01

01

00

Percentile2

547

52

045

48

52

038

7780

024

7778

01 0

114

119

054

115

116

011

07

05minus03

06

Percentile5

0106

108

019

105

106

007

140

141

013

140

138minus020

186

190

040

187

185minus024

26

21minus05

23

24

01

Percentile7

5159

165

066

158

164

059

199

204

054

199

201

022

250

261

105

252

255

034

7672minus04

7374

01

Percentile9

5196

203

074

195

202

074

236

242

065

236

239

032

296

307

110

297

300

027

229

226minus03

227

223minus05

gt30

Cggt80

mm

00

00

00

00

00

00

1626

1017

2810

00

00

00

gt25

Cggt40

mm

00

00

00

712

57

147

90107

1794

108

142

20

22

0lt0C

gmdash

2316minus6

2216minus6

76minus2

76minus1

21minus1

21minus1

mdashmdash

mdashmdash

mdashmdash

ltminus5C

gmdash

21minus1

21minus1

00

00

00

00

00

00

mdashmdash

mdashmdash

mdashmdash

Winter

33

36

032

33

38

048

59

62

031

60

62

020

9095

047

9195

035

1033

1096

63

1056

1066

09

Sprin

g74

87

131

7582

073

111

124

121

113

119

060

160

171

116

162

170

076

496

513

17515

511minus04

Summer

174

187

124

174

186

125

216

224

082

216

225

091

269

273

043

271

282

109

334

331minus03

330

368

38

Autumn

124

127

028

123

127

036

156

156

000

156

155minus001

199

197minus020

200

201

019

842

831minus10

854

919

65

Ann

ual

101

109

079

101

108

070

136

141

057

136

140

043

179

184

042

181

187

060

676

719

43

689

716

27

Advances in Meteorology 11

value of 18∘C for the period of 1912ndash1964 and an averagevalue of 187∘C for the period of 1965ndash2016 On the otherhand comparison of the daily maximum temperature seriesfor the periods of 1912ndash1980 and 1981ndash2016 has revealed anincrement of 06∘C with an average value between 182∘C and188∘C for both of these periods respectively In additionthe standard deviation values are increased between 02∘Cand 03∘C during the whole studied periodThe increment inthe minimum values of daily maximum temperature seriesis more evident than the maximum values The minimumvalues of daily maximum temperatures have exhibited anincrement of 08∘C for the period of 1965ndash2016 compared tothe previous period of 1912ndash1964 and also an increment of07∘C for the period of 1981ndash2016 compared to the previousperiod of 1912ndash1980 Also themaximumvalues of dailymaxi-mum temperature series exhibited amean increment of 05∘Cfor both studied periods Seasonal analysis of maximumtemperature series for both of studied section periods showedthat the highest rising temperature has been happened bya value of 12∘C in summer On the other hand the lowestincrement has happened in the autumn season The monthlyanalysis of daily maximum temperature series showed thatthe highest increment took place during the last monthsof the spring season and the first month of the summerseason in all of the studied periods When the percentiles ofdaily maximum temperatures are analyzed the temperatureincrement based on the 5th percentile threshold is 05∘C forthe periods of 1912ndash1964 and 1965ndash2016 while the incrementis 03∘C for the periods of 1912ndash1980 and 1981ndash2016 This canbe considered as an important sign of rising temperature overthe time In this case the value of percentile thresholds isincreased with extending the length of the time period withextending the length of the first section of studied period infavor of last years than the previous ones In this regard thesevalues are 05∘C and 04∘C at the 25th percentile 1∘C and11∘C at the 75th percentile and 11∘C and 12∘C at the 95thpercentile for both sections of studied periods respectivelyThis situation shows that the increment of higher values inthe dailymaximum temperatures is greater than lower values

The comparison analysis of daily minimum temperaturesbetween the periods of 1912ndash1964 and 1965ndash2016 and theperiods of 1912ndash1980 and 1981ndash2016 showed that there is ageneral increment of 05∘C during the first section periodswhich can be given as 102∘C and 107∘C for the individualperiods of 1912ndash1964 and 1965ndash2016 and also a generalincrement of 08∘C during the second section periods whichcan be provided as 102∘C and 11∘C for the individual periodsof 1912ndash1980 and 1981ndash2016 respectively In addition thestandard deviation values increased among these sectionperiods from 0∘C to 03∘C The increment in the minimumvalues of daily minimum temperature series is more evidentthan the maximum values The minimum values of dailyminimum temperature series showed an increment of 12∘Cfrom the period of 1912ndash1964 to 1965ndash2016 and also anincrement of 16∘C from the period of 1912ndash1980 to 1981ndash2016The maximum values of daily minimum temperature serieshave shown an increment of 07∘C for the section periodsof 1912ndash1964 and 1965ndash2016 and an increment of 12∘C forthe section periods of 1912ndash1980 and 1981ndash2016 Overall

the minimum values have had an average increment of14∘C while the maximum values have had an averageincrement of 1∘C during the last century which can showthe higher rate of upward trends in the temperature timeseries Monthly analysis of minimum temperature seriesshowed that the highest increment for the section studiedperiods of 1912ndash1964 and 1965ndash2016 has occurred by a valueof 1∘C in June while the highest increment for the sectionstudied periods of 1912ndash1980 and 1981ndash2016 has occurred bya value of 16∘C in August Meanwhile there is a temperaturedecrement in November in both of these periods Howeverthere is a clear decrement in the first half of these periodsin October and December whereas there is an incrementin the second half of these periods The seasonal analysisof the daily minimum temperature series showed that thehighest increment has happened in the summer season witha mean value of 07∘C for the section periods of 1912ndash1964and 1965ndash2016 and with a mean value of 13∘C for the sectionperiods of 1912ndash1980 and 1981ndash2016 respectively Then theincreasing rate during the summer season became moreevident during the recent decades Analysis of the percentilethresholds of daily minimum temperature series showed thatthe temperature increment at the 5th percentile is 06∘C forall studied time periods of 1912ndash1964 1965ndash2016 1912ndash1980and 1981ndash2016 Also these increment values indicate anincrease of 04∘C and 06∘C at the 25th percentile 07∘C and11∘C at the 75th percentile and 07∘C and 14∘C at the 95thpercentile thresholds for the whole studied subperiods of1912ndash1964 1965ndash2016 1912ndash1980 and 1981ndash2016 respectivelyThese rising rates in minimum temperature series are moreevident for the periods of 1912ndash1980 and 1981ndash2016 than theperiods of 1912ndash1980 and 1981ndash2016 Therefore it can be saidthat the rate of temperature rising has increased further aslong as the studied time period is closer to the last years

42 Precipitation Annual average precipitation in Istanbulis 838mm with a range of minimum value of 449mm in1921 and a maximum value of 1289mm in 1981 based on theobservatory data of Kandilli station during the whole studiedperiod from 1912 to 2016 (Figure 4(a)) Also monthly averageprecipitation is 699mm with a range of the minimum of326mm in July and the maximum of 1286mm in Decemberduring the whole studied time period (Figure 4(b))

Analysis of the trend in the annual average precipitationtime series by the methods of linear regression analysisand MK trend test has shown that periodically there arepartial increments and significant differences during the totalstudied period from 1912 to 2016 (Figures 5(a1) and 5(a2))But this increment in the precipitation time series is notas clear as the increment in the temperature time seriesHowever it is obvious in the precipitation time series thatthere is an increment between the years of 1917 and 1925a stable condition between the years of 1925 and 1954 anincrement between the years of 1954 and 1965 a decrementbetween the years of 1965 and 1974 an increment between theyears of 1974 and 2001 and again a no change situation from2001 till the end The trend analysis of the daily maximumprecipitation series showed an increment of 29mm for theperiods of 1912ndash1964 and 1965ndash2016 as well as an increment

12 Advances in Meteorology

of 93mm for the periods of 1912ndash1980 and 1981ndash2016 Thesevalues indicated that rainfall has increased at a rate of morethan three times over the last decades than the previous ones(Figures 5(b1) and 5(b2)) Also the number of days with dailyprecipitation greater than 25mm presented an incrementduring the period of 1912ndash2016 although this is not significantat the confidence level of 005 (Figures 5(c1) and 5(c2))The trend of standard deviation in the daily precipitationtime series showed a slowly increasing trend during the totalstudied period (Figures 5(d1) and 5(d2))

The comparison of the results of statistical analysisbetween the daily average rainfall amounts belonging tothe periods of 1912ndash1964 and 1965ndash2016 with those of theperiods of 1912ndash1980 and 1981ndash2016 revealed that there is anincrement of 78mm from the period of 1912ndash1964 to theperiod of 1965ndash2016 and an increment of 38mm from theperiod of 1912ndash1980 to the period of 1981ndash2016 In additionthe analysis of standard deviation exhibited a decrementfrom the period of 1912ndash1964 to the period of 1965ndash2016 andan increment from the period of 1912ndash1980 to the periodof 1981ndash2016 (Table 2) The analysis of monthly averageprecipitation time series showed that the highest incrementhas happened in October with 227mm for the sectionperiods of 1912ndash1964 and 1965ndash2016 and with 347mm forthe section periods of 1912ndash1980 and 1981ndash2016 On the otherhand the highest decrease took placewith a value ofminus04mmduring both May and July for the periods of 1912ndash1964 and1965ndash2016 and with a value of minus16mm in September for thesection periods of 1912ndash1980 and 1981ndash2016 The analysis ofseasonal average precipitation time series showed that thehighest increase occurred in autumn with a value of 72mmfor the section periods of 1912ndash1964 and 1965ndash2016 as wellas with a value of 64mm for the periods of 1912ndash1980 and1981ndash2016 respectively There is no remarkable decreasingchange seasonally except in winter season during the periodsof 1912ndash1980 and 1981ndash2016 Furthermore the percentilethresholds of daily average precipitation indicated that thereis an insignificant negative trend based on all percentilesand for all studied periods Overall the statistical analysisshowed that the total average precipitation of Istanbul hasincreased while this increasing trend is more pronounced inthe previous decades than the last 3 decades On the otherhand the increasing rate of daily maximum precipitation ismore evident during the last 3 decades than the previousdecades which can be proven by the increasing frequency ofheavy rainfall events in Istanbul

Generally the results of trend analysis of Kandilli stationduring the last 105 years of 1912ndash2016 showed that there isa warming significant trend in the precipitation time seriesby using both methods of linear regression analysis andMK trend test On the contrary previous climate studiesconducted over Turkey put forward that there has been adecreasing trend in annual precipitation time series duringthe recent decades regionally The results of a previouslyconducted study by using the daily precipitation and tem-perature data sets of Florya and Goztepe meteorologicalstations in Istanbul area between 1960 and 2013 showedthat most notably the precipitation during the warm periodshas decreased but the frequency of the intense rain has

increased and the majority of these episodes of intense raincoincided with the warm periods Other determinationswere the rise in the annual average temperature and theextension of the warm periods in a year This differentiationof the temperature features can lead to the aggravation of theevaporation and it can be effective for a longer period duringthe year [15] Thus it will make Istanbul be confronted withthe much more important problems of water managementand flood [34] Also the results of the current study for trendanalysis in the long period from 1912 to 2016 showed that themost striking spell is between the years of 1968 and 1998 dueto the existence of least number of rainfall events in IstanbulIt can be owing to industrialization along with the increasingair pollution as well as irregular urbanization in Istanbul areaIn this case severe droughts taking place during the yearsof 1988 1992 1993 and 2008 have threatened the reservoirswhich supplied fresh water of the city These years werecharacterized by not having enough rainfall events Theseyears are also characterized by more persistent high-pressuresystems and less occurrence of low-pressure systems in termsof number and strength As statements made by officialinstitutions the formation conditions for the atmosphericlayer of air pollution due to air pollutant emissions from fossilfuel combustion and industrial activities are more providedduring anticyclone or high-pressure system eventsThereforethis leads to warming up and generating an inversion layerin the boundary layer of atmosphere especially over city area[35] The inversion layer or urban heat island intensity isincreasing with the increasing city size andor populationa phenomenon that was also reported by others [36ndash38]Moreover these last climatic events also have affected somepolitical results beyond natural effects The most obviousexample of this is related to the local election of 1994 inIstanbul In this case extreme drought during the summers of1992 and 1993 has caused groundwater reservoirs to dry up inthe city discontinuance of water was experienced for severaldays or even weeks In those times the mayor of Istanbul cityhas lost the election of 1994 and he realized that this result wasdue to the peoplersquos reaction about the water shortages Manypolitical reports of that period also support this scientificview

5 Conclusions

Statistical analysis in temperature time series of the Kandillistation from 1912 to 2016 established that there is a notableincrease in temperature values after the 1940s which is inparallel with the beginning of industrialization era in Istan-bul There has been a rise about 094∘C in the daily averagetemperature series since the beginning of the last century Asignificant positive trend in the daily maximum temperatureseries is found about 156∘C Also there is a positive trendabout 087∘C in the daily minimum temperature series Onthe other hand analysis of the number of days with thedaily maximum temperature higher than 30∘C showed thatthere is an increasing trend Meanwhile analysis of thenumber of days with daily minimum temperatures lowerthan 0∘C showed a decreasing trend The increment in theminimum values of the daily minimum temperature series

Advances in Meteorology 13

is more evident than the maximum values of this series Inthis case these rising rates in the minimum temperatureseries are more evident for the section periods of 1912ndash1980and 1981ndash2016 than the section periods of 1912ndash1964 and1965ndash2016 This again shows that there is an increment inthe positive temperature trend from past to present decadesThe increment in the precipitation time series is not asclear as the increment in the temperature time series dueto periodic variability The trend analysis in the total annualprecipitation time series showed that the first significantupward trend has periodically been started from the 1920swhile there is a stable trend from 2001 till 2016 The dailyaverage of rainfall amounts has increased with a value of58mm during the period of 1912ndash2016 Also the analysisof heavy precipitation trend showed an increase of 61mmOverall the total average precipitation of Istanbul increasedwhile this increasing trend is more pronounced during theearly decades than the last 3 decades On the other handthe increasing rate of daily maximum precipitation is morepronounced in the last 3 decades than the previous decadesThen it was shown that the frequency of heavy rainfallat Istanbul has increased during the recent decades Thusthe precipitation changes in Istanbul have some differencescompared to the general tendency in precipitation trendthat was put forward by other studies as a decreasing trendover the whole of Turkey This result can be expressedas a positive effect of population overgrowth of Istanbulmegacity Comparison of the results in the first half of thestudy period (1912ndash1964) with the second half of the studyperiod (1965ndash2016) showed that both the average temperatureand average precipitation have higher values of 139∘C and878mm for the final phase compared to the values of 136∘Cand 799mm belonging to the initial phase Therefore it canbe stated that the megacity of Istanbul is directly affectedby the climate change and its consequences In this contextpotential risks of climate change in Istanbul megacity underhigher temperature conditions can be expressed as the rise inthe sea level increase in the rate of evapotranspiration andincrease in the frequency of heavy rainfall Also this city maynot be able to handle this uncontrolled population growthand its associated irreversible changes which is alreadypushing the natural limits by destroying the environmentTherefore the local governors of any megacity like Istanbulshould give more emphasis on the importance of sustainableurban development Thus it is urgent to prepare local andnational climate change strategies and action plans for themegacities

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

The authors are grateful to the Bogazici University theobservatory of Kandilli weather station and the Earthquake

Research Institute for providing the research data and tech-nical support The authors also gratefully acknowledge con-tributions of Assoc Professor Dr Yuksel Demirkaya Schoolof Social Sciences Marmara University This work has beensupported by Scientific and Technological Research Councilof Turkey (TUBITAK) under Grants 113R019 and 106Y258and by Marmara University (BAPKO) with projects FEN-E-120314-0066 FEN-C-YLP-090414-0102 FEN-L-250416-0180 and FEN-A-100413-0127

References

[1] D R Easterling B Horton P D Jones et al ldquoMaximum andminimum temperature trends for the globerdquo Science vol 277no 5324 pp 364ndash367 1997

[2] IPCC Climate Change 2014 IPCC Fifth Assessment Synthe-sis Report-Summary for Policymakers-an Assessment of Inter-Governmental Panel on Climate Change Cambridge UniversityPress Cambridge UK 2014

[3] J Carmin N Nadkarni and C Rhie Progress and Challenges inUrban Climate Adaptation Planning Results of a Global SurveyMIT Cambridge UK 2012

[4] A F Young ldquoUrban expansion and environmental risk in theSao Paulo Metropolitan Areardquo Climate Research vol 57 no 1pp 73ndash80 2013

[5] P Tian X Mu J Liu J Hu and C Gu ldquoImpacts of ClimateVariability and Human Activities on the Changes of Runoff andSediment Load in a Catchment of the Loess Plateau ChinardquoAdvances inMeteorology vol 2016 Article ID 4724067 15 pages2016

[6] R S Kovats and K L Ebi ldquoHeatwaves and public health inEuroperdquo European Journal of Public Health vol 16 no 6 pp592ndash599 2006

[7] S Conti P Meli G Minelli et al ldquoEpidemiologic studyof mortality during the Summer 2003 heat wave in ItalyrdquoEnvironmental Research vol 98 no 3 pp 390ndash399 2005

[8] J Kysely and J Kim ldquoMortality during heat waves in SouthKorea 1991 to 2005 how exceptional was the 1994 heat waverdquoClimate Research vol 38 no 2 pp 105ndash116 2009

[9] B Yan Z Xia F Huang L Guo and X Zhang ldquoClimatechange detection and annual extreme temperature analysis ofthe amur river basinrdquoAdvances inMeteorology vol 2016 ArticleID 6268938 14 pages 2016

[10] E M Fischer and R Knutti ldquoAnthropogenic contribution toglobal occurrence of heavy-precipitation and high-temperatureextremesrdquo Nature Climate Change vol 5 no 6 pp 560ndash5642015

[11] X Zhang L Alexander G C Hegerl et al ldquoIndices for moni-toring changes in extremes based on daily temperature andprecipitation datardquo Climate Change vol 2 no 6 pp 851ndash8702011

[12] K H Schlunzen P Hoffmann G Rosenhagen and W RieckeldquoLong-term changes and regional differences in temperatureand precipitation in the metropolitan area of Hamburgrdquo Inter-national Journal of Climatology vol 30 no 8 pp 1121ndash11362010

[13] G Bartolini M Morabito A Crisci et al ldquoRecent trends inTuscany (Italy) summer temperature and indices of extremesrdquoInternational Journal of Climatology vol 28 no 13 pp 1751ndash1760 2008

14 Advances in Meteorology

[14] S C Sheridan and T J Dolney ldquoHeat mortality and levelof urbanization measuring vulnerability across Ohio USArdquoClimate Research vol 24 no 3 pp 255ndash265 2003

[15] M Tayanc U Im M Dogruel andM Karaca ldquoClimate changein Turkey for the last half centuryrdquo Climatic Change vol 94 no3-4 pp 483ndash502 2009

[16] H Toros ldquoSpatio-temporal variation of daily extreme tempera-tures over Turkeyrdquo International Journal of Climatology vol 32no 7 pp 1047ndash1055 2012

[17] H Toros ldquoSpatio-temporal precipitation change assessmentsover Turkeyrdquo International Journal of Climatology vol 32 no9 pp 1310ndash1325 2012

[18] M Turkes C Yozgatlıgil I Batmaz et al ldquoHas the climate beenchanging in Turkey Regional climate change signals based on acomparative statistical analysis of two consecutive time periods1950-1980 and 1981-2010rdquoClimate Research vol 70 no 1 pp 77ndash93 2016

[19] IPCC ldquoummary for Policymakers In Climate Change 2013rdquo inThe Physical Science Basis The contribution of Working Group Ito the Fifth Assessment Report of the Intergovernmental Panel onClimate Change Cambridge University Press Cambridge UK2013

[20] United Nations ldquoDepartment of Economic and Social AffairsPopulation Division 2006 World Urbanization Prospects The2005 Revisionrdquo Working Paper ESAPWP200 2011

[21] Y Demirkaya Sayılarla Istanbul ITO Istanbul Turkey 2011[22] TUIK ldquoAddress based population registration system results of

2014rdquo Turkish Statistical Institute (TUIK) 2017 httpraporytuikgovtr10-03-2015-184727-842632346446643191142126876html

[23] S Erinc Climatology and its Methods Alfa Basım YayımDagitim Istanbul Turkey 4th edition 1965

[24] O M Gokturk D Bozkurt O L Sen and M Karaca ldquoQualitycontrol and homogeneity of Turkish precipitation datardquoHydro-logical Processes vol 22 no 16 pp 3210ndash3218 2008

[25] C Ley and D Paindaveine ldquoRuns Testsrdquo in Encyclopedia ofEnvironmetrics 2012

[26] A Ghasemi and S Zahediasl ldquoNormality tests for statisticalanalysis a guide for non-statisticiansrdquo International Journal ofEndocrinology andMetabolism vol 10 no 2 pp 486ndash489 2012

[27] G V Glass ldquoTesting Homogeneity of Variancesrdquo AmericanEducational Research Journal vol 3 no 3 pp 187ndash190 1966

[28] H BMann ldquoNonparametric tests against trendrdquo Econometricavol 13 pp 245ndash259 1945

[29] M G Kendall Rank Correlation Method Charles GriffinLondon UK 4th edition 1975

[30] H Turoglu ldquoDetection of Changes on Temperature and Precip-itation Features in Istanbul (Turkey)rdquo Atmospheric and ClimateSciences vol 04 no 04 pp 549ndash562 2014

[31] A Karaburun A Demirci and I-S Suen ldquoImpacts of urbangrowth on forest cover in Istanbul (1987-2007)rdquo EnvironmentalModeling amp Assessment vol 166 no 1-4 pp 267ndash277 2010

[32] K K Karanth LM Curran and J D Reuning-Scherer ldquoVillagesize and forest disturbance in Bhadra Wildlife SanctuaryWestern Ghats Indiardquo Biological Conservation vol 128 no 2pp 147ndash157 2006

[33] G Cakir C Un E Z Baskent S Kose F Sivrikaya andS Keles ldquoEvaluating urbanization fragmentation and landuseland cover change pattern in Istanbul city Turkey from 1971to 2002rdquo Land Degradation amp Development vol 19 no 6 pp663ndash675 2008

[34] R B Myneni F G Hall P J Sellers and A L Marshak ldquoTheinterpretation of spectral vegetation indexesrdquo IEEE Transac-tions on Geoscience and Remote Sensing vol 33 no 2 pp 481ndash486 1995

[35] Y S Unal H Toros A Deniz and S Incecik ldquoInfluence ofmeteorological factors and emission sources on spatial and tem-poral variations of PM10 concentrations in Istanbul metropoli-tan areardquo Atmospheric Environment vol 45 no 31 pp 5504ndash5513 2011

[36] M KaracaM Tayanc andH Toros ldquoEffects of urbanization onclimate of Istanbul and Ankarardquo Atmospheric Environment vol29 no 23 pp 3411ndash3421 1995

[37] Y Ezber O L Sen T Kindap and M Karaca ldquoClimatic effectsof urbanization in Istanbul a statistical and modeling analysisrdquoInternational Journal of Climatology vol 27 no 5 pp 667ndash6792007

[38] H S Park ldquoFeatures of the heat island in seoul and its sur-rounding citiesrdquo Atmospheric Environment (1967) vol 20 no10 pp 1859ndash1866 1986

Advances in Meteorology 5

Table1Statisticalou

tputso

falltests

ford

ataq

ualitycontrol

Cases

NPartestslowast

Testof

norm

alitylowast

Testof

homogeneityof

varia

ncelowast

Parameter

Station

Perio

dTo

tal

Valid

Missing

Std

Runs

test

Kolm

ogorov-Smirn

ovANOVA

test

Levenersquostest

119873(day)119873(

)119873(day)

119885119886

Sig

Mean

Testvalues

Statistic

Sig

119865Sig

Statistic

Sig

119879min

Kand

illi

1965ndash2016

18980

999

2366minus137

00

105

105

0065

00

367

00

2100

Sarıy

er18998

100

167minus136

00

109

109

0061

00

Kand

illi

1st

38353

999

2366minus170

00

104

107

067

00

686

00

183

017

2nd

119879avg

Kand

illi

1965ndash2016

18985

100

471minus138

00

139

139

007

00

031

057

108

029

Sarıy

er18998

100

171minus135

00

139

138

0069

00

Kand

illi

1st

38353

100

471minus169

00

138

141

007

00

332

00

509

002

2nd

119879max

Kand

illi

1965ndash2016

18990

999

1884minus130

00

186

186

007

00

249

00

204

00

Sarıy

er18998

100

178minus129

00

175

175

006

800

Kand

illi

1st

38335

100

1883minus161

00

183

188

006

00

894

00

272

00

2nd

Precip

Kand

illi

1965ndash2016

7691

389

12238

96minus13

00

612

612

026

00

104

031

135

024

Sarıy

er7855

35

11074

92minus10

00

627

63

00253

00

Kand

illi

1st

38353

371

24121

96minus164

00

61

25

026

00

700

0799

037

2nd

ldquoardquom

eans

basedon

them

ediansta

tistic

allylowastshow

sthe

significantlevelof

test

ldquo1strdquo

show

sfirstsectio

nof

thes

tudied

perio

d1912ndash196

4ldquo2nd

rdquosho

wssecon

dsectionof

studied

perio

d1965ndash2016

6 Advances in Meteorology

120

125

130

135

140

145

150

155

160Av

erag

e tem

pera

ture

(∘C)

1916

1920

1924

1928

1932

1936

1940

1944

1948

1952

1956

1960

1964

1968

1972

1976

1980

1984

1988

1992

1996

2000

2004

2008

2012

2016

1912

Years

Average 138∘CMaximum 153∘C in 1966Minimum 124∘C in 1920

R2 = 01644y = 0009x + 13315

R2 = 0075y = 00097x + 13592

R2 = 04619

y = 00514x + 132671912ndash2016 1951ndash2012 1984ndash2016

(a)

Sept

embe

r

Dec

embe

rN

ovem

ber

Febr

uary

Janu

ary

June July

Augu

st

Oct

ober

Mar

ch

May

April

Months

Average 137∘CMaximum 227 ∘C in July

Minimum 52∘C in January

00

50

100

150

200

250

Aver

age t

empe

ratu

re (∘

C)(b)

Figure 2 Trend analysis of average annual temperature (a) and average monthly temperature (b)

is the main reason why the forests as Istanbulrsquos life sourceare being destroyed Overall this suggests that progress hasbeen made in Istanbul not only in having sustainable urbangrowth but also in preserving restoring and even expandingforest areas especially after the year 2000 In the followingafter assessing the population growth and land use change inIstanbul the results obtained from main climatic parameterstrend are analyzed for previous decades compared to lastdecades

41 Temperature The analysis of average yearly maximumand minimum temperature has been carried out using thestatistical method of least squares and the MK test at thesignificance level of 5 is shown with a dashed line (plusmn2)in Figures 2ndash5 The results of linear trends and MK rankcorrelation test showed that there is an increase in the averagedaily temperature series for the whole period of 1912ndash2016 inIstanbul (Figure 2(a)) However this upward trend includesimportant differences While there is not any significantchange in the annual temperature time series from 1912to 1944 the upward trend became apparent between 1944and 1969 A cool period exists in the period of 1969ndash1993and after that a significant rising trend is pronounced inaverage annual temperature at the 95 confidence levelThere has been a rise of about 094∘C in the average dailytemperature over the last 100 years at Kandilli Temperaturehas increased 060∘C in the period of 1951ndash2016 and 170∘C inthe period of 1981ndash2016 In this case the global combined landand ocean surface temperatures show an increase of about

089∘C (069∘Cndash108∘C) over the period of 1901ndash2016 andabout 072∘C (049∘Cndash089∘C) over the period of 1951ndash2016according to fifth assessment report of IPCC in 2013

Nevertheless the data time series of Kandilli stationshows that the increase in temperature parameter after the1940s is in parallel with the beginning of industrialization erain Istanbul Of course regime changes in temperature timeseries on regional scales cannot be totally explained by thenature destruction and pollutant emissions in and aroundIstanbul alone These effects can also be considered as areflection of the general situation happening in the worldon a larger scale Temperature time series of the Kandillistation shows an annual average about 137∘C and absoluteminimum and maximum in 1920 with a value of 124∘C andin 1966 with a value of 153∘C respectively (Figure 2(a))Owing to the Mediterranean climate of the study areamonthly temperature time series is increased from FebruarytoAugust and is decreased fromAugust to February reachinga minimum of 52∘C in February and a maximum of 227∘Cin August (Figure 2(b))

A general tendency of a warming trend in the dailytemperature series is found for the whole studied periodof 1912ndash2016 by using the least squares regression analysisTherefore the trend analysis revealed that the daily averageof temperature has increased by a rate of 09 for the periodfrom 1912 to 2016 (Figures 3(b1) and 3(b2)) Also there is apositive trend about 149∘C in daily maximum temperatureseries during the whole studied period Furthermore theincreasing trends have notably occurred in two subperiods

Advances in Meteorology 7

Tmin Tavg Tmax

Tmax Tmin DTR

u(t)u(t)

u(t)

u㰀(t)u㰀(t)

u㰀(t)

u(t)u㰀(t)

u㰀(t)

u(t)

u(t)

R2 = 02642

y = 00149x minus 10794R2 = 0163

y = 0009x minus 3926

R2 = 0233

y = 00116x minus 12375

R2 = 00182y = 00032x + 16875

R2 = 00734

y = minus00934x + 20292R2 = 02513

y = 02039x minus 37934

(a1)

(a2)

(b1)

(b2)

(c1)

(c2)

(d1)

(d2)

(e1)

(e2)

(f1)

(f2)

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1912

1976

1984

1992

2000

2008

2016

1968

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

2016

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1912

1976

1984

1992

2000

2008

2016

1968

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

minus4minus3minus2minus1

01234567

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus6minus5minus4minus3minus2minus1

0123456

minus4minus3minus2minus1

0123

010203040506070

Num

ber o

f day

s (gt30∘ C)

0102030405060

65707580859095

100

Tem

pera

ture

(∘C)

minus3minus2minus1

0123456

minus3minus2minus1

012345

minus3minus2minus1

01234567

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

9

10

11

12

13Te

mpe

ratu

re (∘

C)

12

13

14

15

16

Tem

pera

ture

(∘C)

16

17

18

19

20

21

Tem

pera

ture

(∘C)

Num

ber o

f day

s (lt0∘ C)

Figure 3 Trend analysis of the temperatures time series using linear best-fit curve andMK test for daily minimum temperature (a1 a2) dailyaverage temperature (b1 b2) daily maximum temperature (c1 c2) number of days with daily maximum temperature gt 30∘C (d1 d2) numberof days with daily minimum temperature lt 0∘C (e1 e2) and daily temperature range (f1 f2)

between 1944 and 1969 and between 1991 and 2016 (Figures3(c1) and 3(c2)) There is a positive trend about 11∘C indaily minimum temperature series during the whole studiedperiod too while this positive trend value is smaller than thepositive trend value of maximum temperature The trend ofminimum temperature series is stable during the period of1912ndash1944 Then it shows a steady increase during 1942 to1989 which is most pronounced during 1944ndash1954 Finallythe rising trends in the minimum temperature series became

significant based onMK testrsquos result for the last period of 2003to 2016 (Figures 3(a1) and 3(a2)) In this case the minimumtemperature shows increasing change for all season wherethe most increasing change has happened in summer andspring season in all studied periods respectively (Table 2)Analysis of the number of days having daily maximumtemperature higher than 30∘C for the long time period of1912 to 2016 by focusing precisely on its subperiods showedthat there is almost no trend during 1912ndash1945 a significant

8 Advances in Meteorology

400

500

600

700

800

900

1000

1100

1200

1300

1912 1920 1928 1936 1944 1952 1960 1968 1976 1984 1992 2000 2008 2016

Tota

l pre

cipi

tatio

n (m

m)

Years

Average 8375 mmMaximum 12894 mm in 1981Minimum 4487 mm in 1921

(a)

0

20

40

60

80

100

120

140

Janu

ary

Febr

uary

Mar

chAp

rilM

ayJu

ne July

Augu

stSe

ptem

ber

Oct

ober

Nov

embe

rD

ecem

ber

Tota

l pre

cipi

tatio

n (m

m)

Months

Average 699 mmMaximum 1284 mm in December

Minimum 319 mm in July

(b)

Figure 4 Time series of total annual precipitation (a) and total monthly precipitation (b)

increasing trend during 1945ndash1969 a relative reduction trendduring 1969ndash1977 an increasing trend during 1977ndash1992and a more pronounced increasing trend during 1992ndash2007(Figures 3(d1) and 3(d2)) On the other hand analysis of thenumber of dayswith dailyminimum temperatures lower than0∘C shows a decreasing trend generally So there is a relativeincrease during 1915ndash1930 almost no trend during 1930ndash1954an increase during 1954ndash1963 a decrease during 1963ndash1985almost no trend during 1985ndash1995 and a decrease during1995ndash2011 by a detailed focus on subperiods of the time series(Figures 3(e1) and 3(e2)) When the daily temperature rangein the whole long period is analyzed an increment can beclearly observed The situation was stable in the period of1912ndash1950There is an increasing trend during 1950ndash1970 andan obvious decrement during 1970ndash1985 Also there is anincrease between 1985 and 2010 (Figures 3(f1) and 3(f2))

The comparison results between the periods of 1912ndash1964and 1965ndash2016 and the periods of 1912ndash1980 and 1981ndash2016showed that (Table 1) there is an increment in the dailyaverage temperature series about 04∘C for the periods of1965ndash2016 (14∘C) compared to previous period of 1912ndash1964(136∘C) and an increment about 05∘C for the periods of1981ndash2016 (141∘C) compared to previous period of 1912ndash1980(136∘C) respectively In addition the standard deviation isincreased by a coefficient between 01∘C and 02∘C in thewhole studied period The increment of the minimum valuesin the daily average temperature series is more evident thanthe maximum values by a mean coefficient of about 6 timesfor both of studied periodsTheminimumvalue of daily aver-age temperatures in the periods of 1912ndash1964 and 1965ndash2016revealed an increment of 06∘C and 08∘C respectively Themaximum value of daily average temperatures in the periodsof 1912ndash1964 and 1965ndash2016 showed a decrement of 02∘Cand 01∘C respectively When the monthly values of dailyaverage temperature are analyzed the highest increment in

the periods of 1912ndash1964 and 1965ndash2016 is found to be inJune at a rate of 1∘C The highest increment in the periodsof 1912ndash1980 and 1981ndash2016 appeared to be in June againwith a rate of 11∘C Also there has been temperature decre-ment in October November and December for all studiedperiods The highest decrement was in November with arate of minus05∘C in the periods of 1912ndash1964 and 1965ndash2016 aswell as minus04∘C in the periods of 1912ndash1980 and 1981ndash2016In this case the analysis of seasonal values of the dailyaverage temperatures showed that the highest increment inthe periods of 1912ndash1964 and 1965ndash2016 took place in thesummer season at a rate of 07∘C The highest increment inthe periods of 1912ndash1980 and 1981ndash2016 was found to be insummer again by a rate of 11∘C Also there was a temperaturedecrement in the autumn season which was about minus010∘Conly for the periods of 1912ndash1964 and 1965ndash2016 Analysisof the percentile thresholds of daily average temperaturesshowed that the temperature increment at the 5th percentileis 05∘C for the periods of 1912ndash1964 and 1965ndash2016 while thisincrement is 03∘C for the periods of 1912ndash1980 and 1981ndash2016Therefore it can be said that the rate of temperature risinghas increased further as much as the time is closer to the endyears of studied period Also these increment values indicatean increase of 02∘C and 03∘C at the 25th percentile 05∘Cand 08∘C at the 75th percentile and 06∘C and 11∘C at the95th percentile thresholds for thewhole studied subperiods of1912ndash1964 1965ndash2016 1912ndash1980 and 1981ndash2016 respectivelyThis situation shows that the increment of higher values inthe daily average temperatures is greater than the incrementof lower values and also this increment ismore evident for theperiods of 1912ndash1980 and 1981ndash2016 compared to the periodsof 1912ndash1964 and 1965ndash2016

Comparison of the daily maximum temperature seriesfor two periods of 1912ndash1964 and 1965ndash2016 revealed anincrement of 08∘C which can be divided into an average

Advances in Meteorology 9

1912

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

Annual total

Years

Annual maximum daily precip

Annual heavy precipitation STD

R2 = 00549

y = 12491x minus 16171

R2 = 00103

y = 0087x minus 1104

u(t)u(t)

u㰀(t)

u㰀(t)

R2 = 2E minus 08y = minus1E minus 05x + 92115

R2 = 00091

y = 00082x minus 99366

u(t)

u㰀(t)

u(t)

u㰀(t)

(a1)

(a2)

(b1)

(b2)

(c1)

(c2)

(d1)

(d2)

400500600700800900

1000110012001300

Am

ount

of p

reci

p (m

m)

Am

ount

of p

reci

p (m

m)

20

40

60

80

100

120

140

160

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

minus3

minus2

minus1

0

1

2

3

4

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus3

minus2

minus1

0

1

2

3

2016

1912

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1928

Years

2468

101214161820

Am

ount

of p

reci

p (m

m)

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

0

2

4

6

8

10

12

14

Num

ber o

f day

s (gt25

mm

)

minus3

minus2

minus1

0

1

2

3

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus3

minus2

minus1

0

1

2

3

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

2016

1912

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1928

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

Figure 5 Trend analysis of the precipitation time series using least squares linear regression and MK trend test for total annual precipitation(a1 a2) daily maximum precipitation (b1 b2) daily maximum precipitation greater than 25mm (c1 c2) and standard deviation in the dailyprecipitation series (d1 d2)

10 Advances in Meteorology

Table2Statisticalanalyses

outputso

ftem

perature

andprecipitatio

nparameters(SD

stand

arddeviation

Inc

increasedecdecrease)

Parameters

Minim

umtemperature

(∘ C)

Averagetem

perature

(∘ C)

Maxim

umtemperature

(∘ C)

Precipitatio

n(m

m)

Statistics

Perio

ds

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Incor

dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

Average

total

102

107

052

102

109

071

136

140

037

136

141

043

180

187

075

181

187

060

8113

8656

544

8267

8591

324

SD66

66

004

65

67

022

7172

007

7173

019

82

84

021

82

85

025

9391minus03

9193

02

Minim

umminus54minus42

116minus53minus39

138minus27minus21

062minus27minus19

074minus08minus01

076minus07

00

068

00

00

00

00

00

00

Maxim

um223

229

069

222

232

099

276

278

023

277

277

000

356

362

059

358

361

029

589

620

31

579

651

71Percentile5

minus04

02

061minus03

03

055

21

25

049

22

24

026

45

50

052

46

48

015

01

01minus01

01

01

00

Percentile2

547

52

045

48

52

038

7780

024

7778

01 0

114

119

054

115

116

011

07

05minus03

06

Percentile5

0106

108

019

105

106

007

140

141

013

140

138minus020

186

190

040

187

185minus024

26

21minus05

23

24

01

Percentile7

5159

165

066

158

164

059

199

204

054

199

201

022

250

261

105

252

255

034

7672minus04

7374

01

Percentile9

5196

203

074

195

202

074

236

242

065

236

239

032

296

307

110

297

300

027

229

226minus03

227

223minus05

gt30

Cggt80

mm

00

00

00

00

00

00

1626

1017

2810

00

00

00

gt25

Cggt40

mm

00

00

00

712

57

147

90107

1794

108

142

20

22

0lt0C

gmdash

2316minus6

2216minus6

76minus2

76minus1

21minus1

21minus1

mdashmdash

mdashmdash

mdashmdash

ltminus5C

gmdash

21minus1

21minus1

00

00

00

00

00

00

mdashmdash

mdashmdash

mdashmdash

Winter

33

36

032

33

38

048

59

62

031

60

62

020

9095

047

9195

035

1033

1096

63

1056

1066

09

Sprin

g74

87

131

7582

073

111

124

121

113

119

060

160

171

116

162

170

076

496

513

17515

511minus04

Summer

174

187

124

174

186

125

216

224

082

216

225

091

269

273

043

271

282

109

334

331minus03

330

368

38

Autumn

124

127

028

123

127

036

156

156

000

156

155minus001

199

197minus020

200

201

019

842

831minus10

854

919

65

Ann

ual

101

109

079

101

108

070

136

141

057

136

140

043

179

184

042

181

187

060

676

719

43

689

716

27

Advances in Meteorology 11

value of 18∘C for the period of 1912ndash1964 and an averagevalue of 187∘C for the period of 1965ndash2016 On the otherhand comparison of the daily maximum temperature seriesfor the periods of 1912ndash1980 and 1981ndash2016 has revealed anincrement of 06∘C with an average value between 182∘C and188∘C for both of these periods respectively In additionthe standard deviation values are increased between 02∘Cand 03∘C during the whole studied periodThe increment inthe minimum values of daily maximum temperature seriesis more evident than the maximum values The minimumvalues of daily maximum temperatures have exhibited anincrement of 08∘C for the period of 1965ndash2016 compared tothe previous period of 1912ndash1964 and also an increment of07∘C for the period of 1981ndash2016 compared to the previousperiod of 1912ndash1980 Also themaximumvalues of dailymaxi-mum temperature series exhibited amean increment of 05∘Cfor both studied periods Seasonal analysis of maximumtemperature series for both of studied section periods showedthat the highest rising temperature has been happened bya value of 12∘C in summer On the other hand the lowestincrement has happened in the autumn season The monthlyanalysis of daily maximum temperature series showed thatthe highest increment took place during the last monthsof the spring season and the first month of the summerseason in all of the studied periods When the percentiles ofdaily maximum temperatures are analyzed the temperatureincrement based on the 5th percentile threshold is 05∘C forthe periods of 1912ndash1964 and 1965ndash2016 while the incrementis 03∘C for the periods of 1912ndash1980 and 1981ndash2016 This canbe considered as an important sign of rising temperature overthe time In this case the value of percentile thresholds isincreased with extending the length of the time period withextending the length of the first section of studied period infavor of last years than the previous ones In this regard thesevalues are 05∘C and 04∘C at the 25th percentile 1∘C and11∘C at the 75th percentile and 11∘C and 12∘C at the 95thpercentile for both sections of studied periods respectivelyThis situation shows that the increment of higher values inthe dailymaximum temperatures is greater than lower values

The comparison analysis of daily minimum temperaturesbetween the periods of 1912ndash1964 and 1965ndash2016 and theperiods of 1912ndash1980 and 1981ndash2016 showed that there is ageneral increment of 05∘C during the first section periodswhich can be given as 102∘C and 107∘C for the individualperiods of 1912ndash1964 and 1965ndash2016 and also a generalincrement of 08∘C during the second section periods whichcan be provided as 102∘C and 11∘C for the individual periodsof 1912ndash1980 and 1981ndash2016 respectively In addition thestandard deviation values increased among these sectionperiods from 0∘C to 03∘C The increment in the minimumvalues of daily minimum temperature series is more evidentthan the maximum values The minimum values of dailyminimum temperature series showed an increment of 12∘Cfrom the period of 1912ndash1964 to 1965ndash2016 and also anincrement of 16∘C from the period of 1912ndash1980 to 1981ndash2016The maximum values of daily minimum temperature serieshave shown an increment of 07∘C for the section periodsof 1912ndash1964 and 1965ndash2016 and an increment of 12∘C forthe section periods of 1912ndash1980 and 1981ndash2016 Overall

the minimum values have had an average increment of14∘C while the maximum values have had an averageincrement of 1∘C during the last century which can showthe higher rate of upward trends in the temperature timeseries Monthly analysis of minimum temperature seriesshowed that the highest increment for the section studiedperiods of 1912ndash1964 and 1965ndash2016 has occurred by a valueof 1∘C in June while the highest increment for the sectionstudied periods of 1912ndash1980 and 1981ndash2016 has occurred bya value of 16∘C in August Meanwhile there is a temperaturedecrement in November in both of these periods Howeverthere is a clear decrement in the first half of these periodsin October and December whereas there is an incrementin the second half of these periods The seasonal analysisof the daily minimum temperature series showed that thehighest increment has happened in the summer season witha mean value of 07∘C for the section periods of 1912ndash1964and 1965ndash2016 and with a mean value of 13∘C for the sectionperiods of 1912ndash1980 and 1981ndash2016 respectively Then theincreasing rate during the summer season became moreevident during the recent decades Analysis of the percentilethresholds of daily minimum temperature series showed thatthe temperature increment at the 5th percentile is 06∘C forall studied time periods of 1912ndash1964 1965ndash2016 1912ndash1980and 1981ndash2016 Also these increment values indicate anincrease of 04∘C and 06∘C at the 25th percentile 07∘C and11∘C at the 75th percentile and 07∘C and 14∘C at the 95thpercentile thresholds for the whole studied subperiods of1912ndash1964 1965ndash2016 1912ndash1980 and 1981ndash2016 respectivelyThese rising rates in minimum temperature series are moreevident for the periods of 1912ndash1980 and 1981ndash2016 than theperiods of 1912ndash1980 and 1981ndash2016 Therefore it can be saidthat the rate of temperature rising has increased further aslong as the studied time period is closer to the last years

42 Precipitation Annual average precipitation in Istanbulis 838mm with a range of minimum value of 449mm in1921 and a maximum value of 1289mm in 1981 based on theobservatory data of Kandilli station during the whole studiedperiod from 1912 to 2016 (Figure 4(a)) Also monthly averageprecipitation is 699mm with a range of the minimum of326mm in July and the maximum of 1286mm in Decemberduring the whole studied time period (Figure 4(b))

Analysis of the trend in the annual average precipitationtime series by the methods of linear regression analysisand MK trend test has shown that periodically there arepartial increments and significant differences during the totalstudied period from 1912 to 2016 (Figures 5(a1) and 5(a2))But this increment in the precipitation time series is notas clear as the increment in the temperature time seriesHowever it is obvious in the precipitation time series thatthere is an increment between the years of 1917 and 1925a stable condition between the years of 1925 and 1954 anincrement between the years of 1954 and 1965 a decrementbetween the years of 1965 and 1974 an increment between theyears of 1974 and 2001 and again a no change situation from2001 till the end The trend analysis of the daily maximumprecipitation series showed an increment of 29mm for theperiods of 1912ndash1964 and 1965ndash2016 as well as an increment

12 Advances in Meteorology

of 93mm for the periods of 1912ndash1980 and 1981ndash2016 Thesevalues indicated that rainfall has increased at a rate of morethan three times over the last decades than the previous ones(Figures 5(b1) and 5(b2)) Also the number of days with dailyprecipitation greater than 25mm presented an incrementduring the period of 1912ndash2016 although this is not significantat the confidence level of 005 (Figures 5(c1) and 5(c2))The trend of standard deviation in the daily precipitationtime series showed a slowly increasing trend during the totalstudied period (Figures 5(d1) and 5(d2))

The comparison of the results of statistical analysisbetween the daily average rainfall amounts belonging tothe periods of 1912ndash1964 and 1965ndash2016 with those of theperiods of 1912ndash1980 and 1981ndash2016 revealed that there is anincrement of 78mm from the period of 1912ndash1964 to theperiod of 1965ndash2016 and an increment of 38mm from theperiod of 1912ndash1980 to the period of 1981ndash2016 In additionthe analysis of standard deviation exhibited a decrementfrom the period of 1912ndash1964 to the period of 1965ndash2016 andan increment from the period of 1912ndash1980 to the periodof 1981ndash2016 (Table 2) The analysis of monthly averageprecipitation time series showed that the highest incrementhas happened in October with 227mm for the sectionperiods of 1912ndash1964 and 1965ndash2016 and with 347mm forthe section periods of 1912ndash1980 and 1981ndash2016 On the otherhand the highest decrease took placewith a value ofminus04mmduring both May and July for the periods of 1912ndash1964 and1965ndash2016 and with a value of minus16mm in September for thesection periods of 1912ndash1980 and 1981ndash2016 The analysis ofseasonal average precipitation time series showed that thehighest increase occurred in autumn with a value of 72mmfor the section periods of 1912ndash1964 and 1965ndash2016 as wellas with a value of 64mm for the periods of 1912ndash1980 and1981ndash2016 respectively There is no remarkable decreasingchange seasonally except in winter season during the periodsof 1912ndash1980 and 1981ndash2016 Furthermore the percentilethresholds of daily average precipitation indicated that thereis an insignificant negative trend based on all percentilesand for all studied periods Overall the statistical analysisshowed that the total average precipitation of Istanbul hasincreased while this increasing trend is more pronounced inthe previous decades than the last 3 decades On the otherhand the increasing rate of daily maximum precipitation ismore evident during the last 3 decades than the previousdecades which can be proven by the increasing frequency ofheavy rainfall events in Istanbul

Generally the results of trend analysis of Kandilli stationduring the last 105 years of 1912ndash2016 showed that there isa warming significant trend in the precipitation time seriesby using both methods of linear regression analysis andMK trend test On the contrary previous climate studiesconducted over Turkey put forward that there has been adecreasing trend in annual precipitation time series duringthe recent decades regionally The results of a previouslyconducted study by using the daily precipitation and tem-perature data sets of Florya and Goztepe meteorologicalstations in Istanbul area between 1960 and 2013 showedthat most notably the precipitation during the warm periodshas decreased but the frequency of the intense rain has

increased and the majority of these episodes of intense raincoincided with the warm periods Other determinationswere the rise in the annual average temperature and theextension of the warm periods in a year This differentiationof the temperature features can lead to the aggravation of theevaporation and it can be effective for a longer period duringthe year [15] Thus it will make Istanbul be confronted withthe much more important problems of water managementand flood [34] Also the results of the current study for trendanalysis in the long period from 1912 to 2016 showed that themost striking spell is between the years of 1968 and 1998 dueto the existence of least number of rainfall events in IstanbulIt can be owing to industrialization along with the increasingair pollution as well as irregular urbanization in Istanbul areaIn this case severe droughts taking place during the yearsof 1988 1992 1993 and 2008 have threatened the reservoirswhich supplied fresh water of the city These years werecharacterized by not having enough rainfall events Theseyears are also characterized by more persistent high-pressuresystems and less occurrence of low-pressure systems in termsof number and strength As statements made by officialinstitutions the formation conditions for the atmosphericlayer of air pollution due to air pollutant emissions from fossilfuel combustion and industrial activities are more providedduring anticyclone or high-pressure system eventsThereforethis leads to warming up and generating an inversion layerin the boundary layer of atmosphere especially over city area[35] The inversion layer or urban heat island intensity isincreasing with the increasing city size andor populationa phenomenon that was also reported by others [36ndash38]Moreover these last climatic events also have affected somepolitical results beyond natural effects The most obviousexample of this is related to the local election of 1994 inIstanbul In this case extreme drought during the summers of1992 and 1993 has caused groundwater reservoirs to dry up inthe city discontinuance of water was experienced for severaldays or even weeks In those times the mayor of Istanbul cityhas lost the election of 1994 and he realized that this result wasdue to the peoplersquos reaction about the water shortages Manypolitical reports of that period also support this scientificview

5 Conclusions

Statistical analysis in temperature time series of the Kandillistation from 1912 to 2016 established that there is a notableincrease in temperature values after the 1940s which is inparallel with the beginning of industrialization era in Istan-bul There has been a rise about 094∘C in the daily averagetemperature series since the beginning of the last century Asignificant positive trend in the daily maximum temperatureseries is found about 156∘C Also there is a positive trendabout 087∘C in the daily minimum temperature series Onthe other hand analysis of the number of days with thedaily maximum temperature higher than 30∘C showed thatthere is an increasing trend Meanwhile analysis of thenumber of days with daily minimum temperatures lowerthan 0∘C showed a decreasing trend The increment in theminimum values of the daily minimum temperature series

Advances in Meteorology 13

is more evident than the maximum values of this series Inthis case these rising rates in the minimum temperatureseries are more evident for the section periods of 1912ndash1980and 1981ndash2016 than the section periods of 1912ndash1964 and1965ndash2016 This again shows that there is an increment inthe positive temperature trend from past to present decadesThe increment in the precipitation time series is not asclear as the increment in the temperature time series dueto periodic variability The trend analysis in the total annualprecipitation time series showed that the first significantupward trend has periodically been started from the 1920swhile there is a stable trend from 2001 till 2016 The dailyaverage of rainfall amounts has increased with a value of58mm during the period of 1912ndash2016 Also the analysisof heavy precipitation trend showed an increase of 61mmOverall the total average precipitation of Istanbul increasedwhile this increasing trend is more pronounced during theearly decades than the last 3 decades On the other handthe increasing rate of daily maximum precipitation is morepronounced in the last 3 decades than the previous decadesThen it was shown that the frequency of heavy rainfallat Istanbul has increased during the recent decades Thusthe precipitation changes in Istanbul have some differencescompared to the general tendency in precipitation trendthat was put forward by other studies as a decreasing trendover the whole of Turkey This result can be expressedas a positive effect of population overgrowth of Istanbulmegacity Comparison of the results in the first half of thestudy period (1912ndash1964) with the second half of the studyperiod (1965ndash2016) showed that both the average temperatureand average precipitation have higher values of 139∘C and878mm for the final phase compared to the values of 136∘Cand 799mm belonging to the initial phase Therefore it canbe stated that the megacity of Istanbul is directly affectedby the climate change and its consequences In this contextpotential risks of climate change in Istanbul megacity underhigher temperature conditions can be expressed as the rise inthe sea level increase in the rate of evapotranspiration andincrease in the frequency of heavy rainfall Also this city maynot be able to handle this uncontrolled population growthand its associated irreversible changes which is alreadypushing the natural limits by destroying the environmentTherefore the local governors of any megacity like Istanbulshould give more emphasis on the importance of sustainableurban development Thus it is urgent to prepare local andnational climate change strategies and action plans for themegacities

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

The authors are grateful to the Bogazici University theobservatory of Kandilli weather station and the Earthquake

Research Institute for providing the research data and tech-nical support The authors also gratefully acknowledge con-tributions of Assoc Professor Dr Yuksel Demirkaya Schoolof Social Sciences Marmara University This work has beensupported by Scientific and Technological Research Councilof Turkey (TUBITAK) under Grants 113R019 and 106Y258and by Marmara University (BAPKO) with projects FEN-E-120314-0066 FEN-C-YLP-090414-0102 FEN-L-250416-0180 and FEN-A-100413-0127

References

[1] D R Easterling B Horton P D Jones et al ldquoMaximum andminimum temperature trends for the globerdquo Science vol 277no 5324 pp 364ndash367 1997

[2] IPCC Climate Change 2014 IPCC Fifth Assessment Synthe-sis Report-Summary for Policymakers-an Assessment of Inter-Governmental Panel on Climate Change Cambridge UniversityPress Cambridge UK 2014

[3] J Carmin N Nadkarni and C Rhie Progress and Challenges inUrban Climate Adaptation Planning Results of a Global SurveyMIT Cambridge UK 2012

[4] A F Young ldquoUrban expansion and environmental risk in theSao Paulo Metropolitan Areardquo Climate Research vol 57 no 1pp 73ndash80 2013

[5] P Tian X Mu J Liu J Hu and C Gu ldquoImpacts of ClimateVariability and Human Activities on the Changes of Runoff andSediment Load in a Catchment of the Loess Plateau ChinardquoAdvances inMeteorology vol 2016 Article ID 4724067 15 pages2016

[6] R S Kovats and K L Ebi ldquoHeatwaves and public health inEuroperdquo European Journal of Public Health vol 16 no 6 pp592ndash599 2006

[7] S Conti P Meli G Minelli et al ldquoEpidemiologic studyof mortality during the Summer 2003 heat wave in ItalyrdquoEnvironmental Research vol 98 no 3 pp 390ndash399 2005

[8] J Kysely and J Kim ldquoMortality during heat waves in SouthKorea 1991 to 2005 how exceptional was the 1994 heat waverdquoClimate Research vol 38 no 2 pp 105ndash116 2009

[9] B Yan Z Xia F Huang L Guo and X Zhang ldquoClimatechange detection and annual extreme temperature analysis ofthe amur river basinrdquoAdvances inMeteorology vol 2016 ArticleID 6268938 14 pages 2016

[10] E M Fischer and R Knutti ldquoAnthropogenic contribution toglobal occurrence of heavy-precipitation and high-temperatureextremesrdquo Nature Climate Change vol 5 no 6 pp 560ndash5642015

[11] X Zhang L Alexander G C Hegerl et al ldquoIndices for moni-toring changes in extremes based on daily temperature andprecipitation datardquo Climate Change vol 2 no 6 pp 851ndash8702011

[12] K H Schlunzen P Hoffmann G Rosenhagen and W RieckeldquoLong-term changes and regional differences in temperatureand precipitation in the metropolitan area of Hamburgrdquo Inter-national Journal of Climatology vol 30 no 8 pp 1121ndash11362010

[13] G Bartolini M Morabito A Crisci et al ldquoRecent trends inTuscany (Italy) summer temperature and indices of extremesrdquoInternational Journal of Climatology vol 28 no 13 pp 1751ndash1760 2008

14 Advances in Meteorology

[14] S C Sheridan and T J Dolney ldquoHeat mortality and levelof urbanization measuring vulnerability across Ohio USArdquoClimate Research vol 24 no 3 pp 255ndash265 2003

[15] M Tayanc U Im M Dogruel andM Karaca ldquoClimate changein Turkey for the last half centuryrdquo Climatic Change vol 94 no3-4 pp 483ndash502 2009

[16] H Toros ldquoSpatio-temporal variation of daily extreme tempera-tures over Turkeyrdquo International Journal of Climatology vol 32no 7 pp 1047ndash1055 2012

[17] H Toros ldquoSpatio-temporal precipitation change assessmentsover Turkeyrdquo International Journal of Climatology vol 32 no9 pp 1310ndash1325 2012

[18] M Turkes C Yozgatlıgil I Batmaz et al ldquoHas the climate beenchanging in Turkey Regional climate change signals based on acomparative statistical analysis of two consecutive time periods1950-1980 and 1981-2010rdquoClimate Research vol 70 no 1 pp 77ndash93 2016

[19] IPCC ldquoummary for Policymakers In Climate Change 2013rdquo inThe Physical Science Basis The contribution of Working Group Ito the Fifth Assessment Report of the Intergovernmental Panel onClimate Change Cambridge University Press Cambridge UK2013

[20] United Nations ldquoDepartment of Economic and Social AffairsPopulation Division 2006 World Urbanization Prospects The2005 Revisionrdquo Working Paper ESAPWP200 2011

[21] Y Demirkaya Sayılarla Istanbul ITO Istanbul Turkey 2011[22] TUIK ldquoAddress based population registration system results of

2014rdquo Turkish Statistical Institute (TUIK) 2017 httpraporytuikgovtr10-03-2015-184727-842632346446643191142126876html

[23] S Erinc Climatology and its Methods Alfa Basım YayımDagitim Istanbul Turkey 4th edition 1965

[24] O M Gokturk D Bozkurt O L Sen and M Karaca ldquoQualitycontrol and homogeneity of Turkish precipitation datardquoHydro-logical Processes vol 22 no 16 pp 3210ndash3218 2008

[25] C Ley and D Paindaveine ldquoRuns Testsrdquo in Encyclopedia ofEnvironmetrics 2012

[26] A Ghasemi and S Zahediasl ldquoNormality tests for statisticalanalysis a guide for non-statisticiansrdquo International Journal ofEndocrinology andMetabolism vol 10 no 2 pp 486ndash489 2012

[27] G V Glass ldquoTesting Homogeneity of Variancesrdquo AmericanEducational Research Journal vol 3 no 3 pp 187ndash190 1966

[28] H BMann ldquoNonparametric tests against trendrdquo Econometricavol 13 pp 245ndash259 1945

[29] M G Kendall Rank Correlation Method Charles GriffinLondon UK 4th edition 1975

[30] H Turoglu ldquoDetection of Changes on Temperature and Precip-itation Features in Istanbul (Turkey)rdquo Atmospheric and ClimateSciences vol 04 no 04 pp 549ndash562 2014

[31] A Karaburun A Demirci and I-S Suen ldquoImpacts of urbangrowth on forest cover in Istanbul (1987-2007)rdquo EnvironmentalModeling amp Assessment vol 166 no 1-4 pp 267ndash277 2010

[32] K K Karanth LM Curran and J D Reuning-Scherer ldquoVillagesize and forest disturbance in Bhadra Wildlife SanctuaryWestern Ghats Indiardquo Biological Conservation vol 128 no 2pp 147ndash157 2006

[33] G Cakir C Un E Z Baskent S Kose F Sivrikaya andS Keles ldquoEvaluating urbanization fragmentation and landuseland cover change pattern in Istanbul city Turkey from 1971to 2002rdquo Land Degradation amp Development vol 19 no 6 pp663ndash675 2008

[34] R B Myneni F G Hall P J Sellers and A L Marshak ldquoTheinterpretation of spectral vegetation indexesrdquo IEEE Transac-tions on Geoscience and Remote Sensing vol 33 no 2 pp 481ndash486 1995

[35] Y S Unal H Toros A Deniz and S Incecik ldquoInfluence ofmeteorological factors and emission sources on spatial and tem-poral variations of PM10 concentrations in Istanbul metropoli-tan areardquo Atmospheric Environment vol 45 no 31 pp 5504ndash5513 2011

[36] M KaracaM Tayanc andH Toros ldquoEffects of urbanization onclimate of Istanbul and Ankarardquo Atmospheric Environment vol29 no 23 pp 3411ndash3421 1995

[37] Y Ezber O L Sen T Kindap and M Karaca ldquoClimatic effectsof urbanization in Istanbul a statistical and modeling analysisrdquoInternational Journal of Climatology vol 27 no 5 pp 667ndash6792007

[38] H S Park ldquoFeatures of the heat island in seoul and its sur-rounding citiesrdquo Atmospheric Environment (1967) vol 20 no10 pp 1859ndash1866 1986

6 Advances in Meteorology

120

125

130

135

140

145

150

155

160Av

erag

e tem

pera

ture

(∘C)

1916

1920

1924

1928

1932

1936

1940

1944

1948

1952

1956

1960

1964

1968

1972

1976

1980

1984

1988

1992

1996

2000

2004

2008

2012

2016

1912

Years

Average 138∘CMaximum 153∘C in 1966Minimum 124∘C in 1920

R2 = 01644y = 0009x + 13315

R2 = 0075y = 00097x + 13592

R2 = 04619

y = 00514x + 132671912ndash2016 1951ndash2012 1984ndash2016

(a)

Sept

embe

r

Dec

embe

rN

ovem

ber

Febr

uary

Janu

ary

June July

Augu

st

Oct

ober

Mar

ch

May

April

Months

Average 137∘CMaximum 227 ∘C in July

Minimum 52∘C in January

00

50

100

150

200

250

Aver

age t

empe

ratu

re (∘

C)(b)

Figure 2 Trend analysis of average annual temperature (a) and average monthly temperature (b)

is the main reason why the forests as Istanbulrsquos life sourceare being destroyed Overall this suggests that progress hasbeen made in Istanbul not only in having sustainable urbangrowth but also in preserving restoring and even expandingforest areas especially after the year 2000 In the followingafter assessing the population growth and land use change inIstanbul the results obtained from main climatic parameterstrend are analyzed for previous decades compared to lastdecades

41 Temperature The analysis of average yearly maximumand minimum temperature has been carried out using thestatistical method of least squares and the MK test at thesignificance level of 5 is shown with a dashed line (plusmn2)in Figures 2ndash5 The results of linear trends and MK rankcorrelation test showed that there is an increase in the averagedaily temperature series for the whole period of 1912ndash2016 inIstanbul (Figure 2(a)) However this upward trend includesimportant differences While there is not any significantchange in the annual temperature time series from 1912to 1944 the upward trend became apparent between 1944and 1969 A cool period exists in the period of 1969ndash1993and after that a significant rising trend is pronounced inaverage annual temperature at the 95 confidence levelThere has been a rise of about 094∘C in the average dailytemperature over the last 100 years at Kandilli Temperaturehas increased 060∘C in the period of 1951ndash2016 and 170∘C inthe period of 1981ndash2016 In this case the global combined landand ocean surface temperatures show an increase of about

089∘C (069∘Cndash108∘C) over the period of 1901ndash2016 andabout 072∘C (049∘Cndash089∘C) over the period of 1951ndash2016according to fifth assessment report of IPCC in 2013

Nevertheless the data time series of Kandilli stationshows that the increase in temperature parameter after the1940s is in parallel with the beginning of industrialization erain Istanbul Of course regime changes in temperature timeseries on regional scales cannot be totally explained by thenature destruction and pollutant emissions in and aroundIstanbul alone These effects can also be considered as areflection of the general situation happening in the worldon a larger scale Temperature time series of the Kandillistation shows an annual average about 137∘C and absoluteminimum and maximum in 1920 with a value of 124∘C andin 1966 with a value of 153∘C respectively (Figure 2(a))Owing to the Mediterranean climate of the study areamonthly temperature time series is increased from FebruarytoAugust and is decreased fromAugust to February reachinga minimum of 52∘C in February and a maximum of 227∘Cin August (Figure 2(b))

A general tendency of a warming trend in the dailytemperature series is found for the whole studied periodof 1912ndash2016 by using the least squares regression analysisTherefore the trend analysis revealed that the daily averageof temperature has increased by a rate of 09 for the periodfrom 1912 to 2016 (Figures 3(b1) and 3(b2)) Also there is apositive trend about 149∘C in daily maximum temperatureseries during the whole studied period Furthermore theincreasing trends have notably occurred in two subperiods

Advances in Meteorology 7

Tmin Tavg Tmax

Tmax Tmin DTR

u(t)u(t)

u(t)

u㰀(t)u㰀(t)

u㰀(t)

u(t)u㰀(t)

u㰀(t)

u(t)

u(t)

R2 = 02642

y = 00149x minus 10794R2 = 0163

y = 0009x minus 3926

R2 = 0233

y = 00116x minus 12375

R2 = 00182y = 00032x + 16875

R2 = 00734

y = minus00934x + 20292R2 = 02513

y = 02039x minus 37934

(a1)

(a2)

(b1)

(b2)

(c1)

(c2)

(d1)

(d2)

(e1)

(e2)

(f1)

(f2)

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1912

1976

1984

1992

2000

2008

2016

1968

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

2016

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1912

1976

1984

1992

2000

2008

2016

1968

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

minus4minus3minus2minus1

01234567

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus6minus5minus4minus3minus2minus1

0123456

minus4minus3minus2minus1

0123

010203040506070

Num

ber o

f day

s (gt30∘ C)

0102030405060

65707580859095

100

Tem

pera

ture

(∘C)

minus3minus2minus1

0123456

minus3minus2minus1

012345

minus3minus2minus1

01234567

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

9

10

11

12

13Te

mpe

ratu

re (∘

C)

12

13

14

15

16

Tem

pera

ture

(∘C)

16

17

18

19

20

21

Tem

pera

ture

(∘C)

Num

ber o

f day

s (lt0∘ C)

Figure 3 Trend analysis of the temperatures time series using linear best-fit curve andMK test for daily minimum temperature (a1 a2) dailyaverage temperature (b1 b2) daily maximum temperature (c1 c2) number of days with daily maximum temperature gt 30∘C (d1 d2) numberof days with daily minimum temperature lt 0∘C (e1 e2) and daily temperature range (f1 f2)

between 1944 and 1969 and between 1991 and 2016 (Figures3(c1) and 3(c2)) There is a positive trend about 11∘C indaily minimum temperature series during the whole studiedperiod too while this positive trend value is smaller than thepositive trend value of maximum temperature The trend ofminimum temperature series is stable during the period of1912ndash1944 Then it shows a steady increase during 1942 to1989 which is most pronounced during 1944ndash1954 Finallythe rising trends in the minimum temperature series became

significant based onMK testrsquos result for the last period of 2003to 2016 (Figures 3(a1) and 3(a2)) In this case the minimumtemperature shows increasing change for all season wherethe most increasing change has happened in summer andspring season in all studied periods respectively (Table 2)Analysis of the number of days having daily maximumtemperature higher than 30∘C for the long time period of1912 to 2016 by focusing precisely on its subperiods showedthat there is almost no trend during 1912ndash1945 a significant

8 Advances in Meteorology

400

500

600

700

800

900

1000

1100

1200

1300

1912 1920 1928 1936 1944 1952 1960 1968 1976 1984 1992 2000 2008 2016

Tota

l pre

cipi

tatio

n (m

m)

Years

Average 8375 mmMaximum 12894 mm in 1981Minimum 4487 mm in 1921

(a)

0

20

40

60

80

100

120

140

Janu

ary

Febr

uary

Mar

chAp

rilM

ayJu

ne July

Augu

stSe

ptem

ber

Oct

ober

Nov

embe

rD

ecem

ber

Tota

l pre

cipi

tatio

n (m

m)

Months

Average 699 mmMaximum 1284 mm in December

Minimum 319 mm in July

(b)

Figure 4 Time series of total annual precipitation (a) and total monthly precipitation (b)

increasing trend during 1945ndash1969 a relative reduction trendduring 1969ndash1977 an increasing trend during 1977ndash1992and a more pronounced increasing trend during 1992ndash2007(Figures 3(d1) and 3(d2)) On the other hand analysis of thenumber of dayswith dailyminimum temperatures lower than0∘C shows a decreasing trend generally So there is a relativeincrease during 1915ndash1930 almost no trend during 1930ndash1954an increase during 1954ndash1963 a decrease during 1963ndash1985almost no trend during 1985ndash1995 and a decrease during1995ndash2011 by a detailed focus on subperiods of the time series(Figures 3(e1) and 3(e2)) When the daily temperature rangein the whole long period is analyzed an increment can beclearly observed The situation was stable in the period of1912ndash1950There is an increasing trend during 1950ndash1970 andan obvious decrement during 1970ndash1985 Also there is anincrease between 1985 and 2010 (Figures 3(f1) and 3(f2))

The comparison results between the periods of 1912ndash1964and 1965ndash2016 and the periods of 1912ndash1980 and 1981ndash2016showed that (Table 1) there is an increment in the dailyaverage temperature series about 04∘C for the periods of1965ndash2016 (14∘C) compared to previous period of 1912ndash1964(136∘C) and an increment about 05∘C for the periods of1981ndash2016 (141∘C) compared to previous period of 1912ndash1980(136∘C) respectively In addition the standard deviation isincreased by a coefficient between 01∘C and 02∘C in thewhole studied period The increment of the minimum valuesin the daily average temperature series is more evident thanthe maximum values by a mean coefficient of about 6 timesfor both of studied periodsTheminimumvalue of daily aver-age temperatures in the periods of 1912ndash1964 and 1965ndash2016revealed an increment of 06∘C and 08∘C respectively Themaximum value of daily average temperatures in the periodsof 1912ndash1964 and 1965ndash2016 showed a decrement of 02∘Cand 01∘C respectively When the monthly values of dailyaverage temperature are analyzed the highest increment in

the periods of 1912ndash1964 and 1965ndash2016 is found to be inJune at a rate of 1∘C The highest increment in the periodsof 1912ndash1980 and 1981ndash2016 appeared to be in June againwith a rate of 11∘C Also there has been temperature decre-ment in October November and December for all studiedperiods The highest decrement was in November with arate of minus05∘C in the periods of 1912ndash1964 and 1965ndash2016 aswell as minus04∘C in the periods of 1912ndash1980 and 1981ndash2016In this case the analysis of seasonal values of the dailyaverage temperatures showed that the highest increment inthe periods of 1912ndash1964 and 1965ndash2016 took place in thesummer season at a rate of 07∘C The highest increment inthe periods of 1912ndash1980 and 1981ndash2016 was found to be insummer again by a rate of 11∘C Also there was a temperaturedecrement in the autumn season which was about minus010∘Conly for the periods of 1912ndash1964 and 1965ndash2016 Analysisof the percentile thresholds of daily average temperaturesshowed that the temperature increment at the 5th percentileis 05∘C for the periods of 1912ndash1964 and 1965ndash2016 while thisincrement is 03∘C for the periods of 1912ndash1980 and 1981ndash2016Therefore it can be said that the rate of temperature risinghas increased further as much as the time is closer to the endyears of studied period Also these increment values indicatean increase of 02∘C and 03∘C at the 25th percentile 05∘Cand 08∘C at the 75th percentile and 06∘C and 11∘C at the95th percentile thresholds for thewhole studied subperiods of1912ndash1964 1965ndash2016 1912ndash1980 and 1981ndash2016 respectivelyThis situation shows that the increment of higher values inthe daily average temperatures is greater than the incrementof lower values and also this increment ismore evident for theperiods of 1912ndash1980 and 1981ndash2016 compared to the periodsof 1912ndash1964 and 1965ndash2016

Comparison of the daily maximum temperature seriesfor two periods of 1912ndash1964 and 1965ndash2016 revealed anincrement of 08∘C which can be divided into an average

Advances in Meteorology 9

1912

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

Annual total

Years

Annual maximum daily precip

Annual heavy precipitation STD

R2 = 00549

y = 12491x minus 16171

R2 = 00103

y = 0087x minus 1104

u(t)u(t)

u㰀(t)

u㰀(t)

R2 = 2E minus 08y = minus1E minus 05x + 92115

R2 = 00091

y = 00082x minus 99366

u(t)

u㰀(t)

u(t)

u㰀(t)

(a1)

(a2)

(b1)

(b2)

(c1)

(c2)

(d1)

(d2)

400500600700800900

1000110012001300

Am

ount

of p

reci

p (m

m)

Am

ount

of p

reci

p (m

m)

20

40

60

80

100

120

140

160

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

minus3

minus2

minus1

0

1

2

3

4

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus3

minus2

minus1

0

1

2

3

2016

1912

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1928

Years

2468

101214161820

Am

ount

of p

reci

p (m

m)

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

0

2

4

6

8

10

12

14

Num

ber o

f day

s (gt25

mm

)

minus3

minus2

minus1

0

1

2

3

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus3

minus2

minus1

0

1

2

3

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

2016

1912

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1928

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

Figure 5 Trend analysis of the precipitation time series using least squares linear regression and MK trend test for total annual precipitation(a1 a2) daily maximum precipitation (b1 b2) daily maximum precipitation greater than 25mm (c1 c2) and standard deviation in the dailyprecipitation series (d1 d2)

10 Advances in Meteorology

Table2Statisticalanalyses

outputso

ftem

perature

andprecipitatio

nparameters(SD

stand

arddeviation

Inc

increasedecdecrease)

Parameters

Minim

umtemperature

(∘ C)

Averagetem

perature

(∘ C)

Maxim

umtemperature

(∘ C)

Precipitatio

n(m

m)

Statistics

Perio

ds

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Incor

dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

Average

total

102

107

052

102

109

071

136

140

037

136

141

043

180

187

075

181

187

060

8113

8656

544

8267

8591

324

SD66

66

004

65

67

022

7172

007

7173

019

82

84

021

82

85

025

9391minus03

9193

02

Minim

umminus54minus42

116minus53minus39

138minus27minus21

062minus27minus19

074minus08minus01

076minus07

00

068

00

00

00

00

00

00

Maxim

um223

229

069

222

232

099

276

278

023

277

277

000

356

362

059

358

361

029

589

620

31

579

651

71Percentile5

minus04

02

061minus03

03

055

21

25

049

22

24

026

45

50

052

46

48

015

01

01minus01

01

01

00

Percentile2

547

52

045

48

52

038

7780

024

7778

01 0

114

119

054

115

116

011

07

05minus03

06

Percentile5

0106

108

019

105

106

007

140

141

013

140

138minus020

186

190

040

187

185minus024

26

21minus05

23

24

01

Percentile7

5159

165

066

158

164

059

199

204

054

199

201

022

250

261

105

252

255

034

7672minus04

7374

01

Percentile9

5196

203

074

195

202

074

236

242

065

236

239

032

296

307

110

297

300

027

229

226minus03

227

223minus05

gt30

Cggt80

mm

00

00

00

00

00

00

1626

1017

2810

00

00

00

gt25

Cggt40

mm

00

00

00

712

57

147

90107

1794

108

142

20

22

0lt0C

gmdash

2316minus6

2216minus6

76minus2

76minus1

21minus1

21minus1

mdashmdash

mdashmdash

mdashmdash

ltminus5C

gmdash

21minus1

21minus1

00

00

00

00

00

00

mdashmdash

mdashmdash

mdashmdash

Winter

33

36

032

33

38

048

59

62

031

60

62

020

9095

047

9195

035

1033

1096

63

1056

1066

09

Sprin

g74

87

131

7582

073

111

124

121

113

119

060

160

171

116

162

170

076

496

513

17515

511minus04

Summer

174

187

124

174

186

125

216

224

082

216

225

091

269

273

043

271

282

109

334

331minus03

330

368

38

Autumn

124

127

028

123

127

036

156

156

000

156

155minus001

199

197minus020

200

201

019

842

831minus10

854

919

65

Ann

ual

101

109

079

101

108

070

136

141

057

136

140

043

179

184

042

181

187

060

676

719

43

689

716

27

Advances in Meteorology 11

value of 18∘C for the period of 1912ndash1964 and an averagevalue of 187∘C for the period of 1965ndash2016 On the otherhand comparison of the daily maximum temperature seriesfor the periods of 1912ndash1980 and 1981ndash2016 has revealed anincrement of 06∘C with an average value between 182∘C and188∘C for both of these periods respectively In additionthe standard deviation values are increased between 02∘Cand 03∘C during the whole studied periodThe increment inthe minimum values of daily maximum temperature seriesis more evident than the maximum values The minimumvalues of daily maximum temperatures have exhibited anincrement of 08∘C for the period of 1965ndash2016 compared tothe previous period of 1912ndash1964 and also an increment of07∘C for the period of 1981ndash2016 compared to the previousperiod of 1912ndash1980 Also themaximumvalues of dailymaxi-mum temperature series exhibited amean increment of 05∘Cfor both studied periods Seasonal analysis of maximumtemperature series for both of studied section periods showedthat the highest rising temperature has been happened bya value of 12∘C in summer On the other hand the lowestincrement has happened in the autumn season The monthlyanalysis of daily maximum temperature series showed thatthe highest increment took place during the last monthsof the spring season and the first month of the summerseason in all of the studied periods When the percentiles ofdaily maximum temperatures are analyzed the temperatureincrement based on the 5th percentile threshold is 05∘C forthe periods of 1912ndash1964 and 1965ndash2016 while the incrementis 03∘C for the periods of 1912ndash1980 and 1981ndash2016 This canbe considered as an important sign of rising temperature overthe time In this case the value of percentile thresholds isincreased with extending the length of the time period withextending the length of the first section of studied period infavor of last years than the previous ones In this regard thesevalues are 05∘C and 04∘C at the 25th percentile 1∘C and11∘C at the 75th percentile and 11∘C and 12∘C at the 95thpercentile for both sections of studied periods respectivelyThis situation shows that the increment of higher values inthe dailymaximum temperatures is greater than lower values

The comparison analysis of daily minimum temperaturesbetween the periods of 1912ndash1964 and 1965ndash2016 and theperiods of 1912ndash1980 and 1981ndash2016 showed that there is ageneral increment of 05∘C during the first section periodswhich can be given as 102∘C and 107∘C for the individualperiods of 1912ndash1964 and 1965ndash2016 and also a generalincrement of 08∘C during the second section periods whichcan be provided as 102∘C and 11∘C for the individual periodsof 1912ndash1980 and 1981ndash2016 respectively In addition thestandard deviation values increased among these sectionperiods from 0∘C to 03∘C The increment in the minimumvalues of daily minimum temperature series is more evidentthan the maximum values The minimum values of dailyminimum temperature series showed an increment of 12∘Cfrom the period of 1912ndash1964 to 1965ndash2016 and also anincrement of 16∘C from the period of 1912ndash1980 to 1981ndash2016The maximum values of daily minimum temperature serieshave shown an increment of 07∘C for the section periodsof 1912ndash1964 and 1965ndash2016 and an increment of 12∘C forthe section periods of 1912ndash1980 and 1981ndash2016 Overall

the minimum values have had an average increment of14∘C while the maximum values have had an averageincrement of 1∘C during the last century which can showthe higher rate of upward trends in the temperature timeseries Monthly analysis of minimum temperature seriesshowed that the highest increment for the section studiedperiods of 1912ndash1964 and 1965ndash2016 has occurred by a valueof 1∘C in June while the highest increment for the sectionstudied periods of 1912ndash1980 and 1981ndash2016 has occurred bya value of 16∘C in August Meanwhile there is a temperaturedecrement in November in both of these periods Howeverthere is a clear decrement in the first half of these periodsin October and December whereas there is an incrementin the second half of these periods The seasonal analysisof the daily minimum temperature series showed that thehighest increment has happened in the summer season witha mean value of 07∘C for the section periods of 1912ndash1964and 1965ndash2016 and with a mean value of 13∘C for the sectionperiods of 1912ndash1980 and 1981ndash2016 respectively Then theincreasing rate during the summer season became moreevident during the recent decades Analysis of the percentilethresholds of daily minimum temperature series showed thatthe temperature increment at the 5th percentile is 06∘C forall studied time periods of 1912ndash1964 1965ndash2016 1912ndash1980and 1981ndash2016 Also these increment values indicate anincrease of 04∘C and 06∘C at the 25th percentile 07∘C and11∘C at the 75th percentile and 07∘C and 14∘C at the 95thpercentile thresholds for the whole studied subperiods of1912ndash1964 1965ndash2016 1912ndash1980 and 1981ndash2016 respectivelyThese rising rates in minimum temperature series are moreevident for the periods of 1912ndash1980 and 1981ndash2016 than theperiods of 1912ndash1980 and 1981ndash2016 Therefore it can be saidthat the rate of temperature rising has increased further aslong as the studied time period is closer to the last years

42 Precipitation Annual average precipitation in Istanbulis 838mm with a range of minimum value of 449mm in1921 and a maximum value of 1289mm in 1981 based on theobservatory data of Kandilli station during the whole studiedperiod from 1912 to 2016 (Figure 4(a)) Also monthly averageprecipitation is 699mm with a range of the minimum of326mm in July and the maximum of 1286mm in Decemberduring the whole studied time period (Figure 4(b))

Analysis of the trend in the annual average precipitationtime series by the methods of linear regression analysisand MK trend test has shown that periodically there arepartial increments and significant differences during the totalstudied period from 1912 to 2016 (Figures 5(a1) and 5(a2))But this increment in the precipitation time series is notas clear as the increment in the temperature time seriesHowever it is obvious in the precipitation time series thatthere is an increment between the years of 1917 and 1925a stable condition between the years of 1925 and 1954 anincrement between the years of 1954 and 1965 a decrementbetween the years of 1965 and 1974 an increment between theyears of 1974 and 2001 and again a no change situation from2001 till the end The trend analysis of the daily maximumprecipitation series showed an increment of 29mm for theperiods of 1912ndash1964 and 1965ndash2016 as well as an increment

12 Advances in Meteorology

of 93mm for the periods of 1912ndash1980 and 1981ndash2016 Thesevalues indicated that rainfall has increased at a rate of morethan three times over the last decades than the previous ones(Figures 5(b1) and 5(b2)) Also the number of days with dailyprecipitation greater than 25mm presented an incrementduring the period of 1912ndash2016 although this is not significantat the confidence level of 005 (Figures 5(c1) and 5(c2))The trend of standard deviation in the daily precipitationtime series showed a slowly increasing trend during the totalstudied period (Figures 5(d1) and 5(d2))

The comparison of the results of statistical analysisbetween the daily average rainfall amounts belonging tothe periods of 1912ndash1964 and 1965ndash2016 with those of theperiods of 1912ndash1980 and 1981ndash2016 revealed that there is anincrement of 78mm from the period of 1912ndash1964 to theperiod of 1965ndash2016 and an increment of 38mm from theperiod of 1912ndash1980 to the period of 1981ndash2016 In additionthe analysis of standard deviation exhibited a decrementfrom the period of 1912ndash1964 to the period of 1965ndash2016 andan increment from the period of 1912ndash1980 to the periodof 1981ndash2016 (Table 2) The analysis of monthly averageprecipitation time series showed that the highest incrementhas happened in October with 227mm for the sectionperiods of 1912ndash1964 and 1965ndash2016 and with 347mm forthe section periods of 1912ndash1980 and 1981ndash2016 On the otherhand the highest decrease took placewith a value ofminus04mmduring both May and July for the periods of 1912ndash1964 and1965ndash2016 and with a value of minus16mm in September for thesection periods of 1912ndash1980 and 1981ndash2016 The analysis ofseasonal average precipitation time series showed that thehighest increase occurred in autumn with a value of 72mmfor the section periods of 1912ndash1964 and 1965ndash2016 as wellas with a value of 64mm for the periods of 1912ndash1980 and1981ndash2016 respectively There is no remarkable decreasingchange seasonally except in winter season during the periodsof 1912ndash1980 and 1981ndash2016 Furthermore the percentilethresholds of daily average precipitation indicated that thereis an insignificant negative trend based on all percentilesand for all studied periods Overall the statistical analysisshowed that the total average precipitation of Istanbul hasincreased while this increasing trend is more pronounced inthe previous decades than the last 3 decades On the otherhand the increasing rate of daily maximum precipitation ismore evident during the last 3 decades than the previousdecades which can be proven by the increasing frequency ofheavy rainfall events in Istanbul

Generally the results of trend analysis of Kandilli stationduring the last 105 years of 1912ndash2016 showed that there isa warming significant trend in the precipitation time seriesby using both methods of linear regression analysis andMK trend test On the contrary previous climate studiesconducted over Turkey put forward that there has been adecreasing trend in annual precipitation time series duringthe recent decades regionally The results of a previouslyconducted study by using the daily precipitation and tem-perature data sets of Florya and Goztepe meteorologicalstations in Istanbul area between 1960 and 2013 showedthat most notably the precipitation during the warm periodshas decreased but the frequency of the intense rain has

increased and the majority of these episodes of intense raincoincided with the warm periods Other determinationswere the rise in the annual average temperature and theextension of the warm periods in a year This differentiationof the temperature features can lead to the aggravation of theevaporation and it can be effective for a longer period duringthe year [15] Thus it will make Istanbul be confronted withthe much more important problems of water managementand flood [34] Also the results of the current study for trendanalysis in the long period from 1912 to 2016 showed that themost striking spell is between the years of 1968 and 1998 dueto the existence of least number of rainfall events in IstanbulIt can be owing to industrialization along with the increasingair pollution as well as irregular urbanization in Istanbul areaIn this case severe droughts taking place during the yearsof 1988 1992 1993 and 2008 have threatened the reservoirswhich supplied fresh water of the city These years werecharacterized by not having enough rainfall events Theseyears are also characterized by more persistent high-pressuresystems and less occurrence of low-pressure systems in termsof number and strength As statements made by officialinstitutions the formation conditions for the atmosphericlayer of air pollution due to air pollutant emissions from fossilfuel combustion and industrial activities are more providedduring anticyclone or high-pressure system eventsThereforethis leads to warming up and generating an inversion layerin the boundary layer of atmosphere especially over city area[35] The inversion layer or urban heat island intensity isincreasing with the increasing city size andor populationa phenomenon that was also reported by others [36ndash38]Moreover these last climatic events also have affected somepolitical results beyond natural effects The most obviousexample of this is related to the local election of 1994 inIstanbul In this case extreme drought during the summers of1992 and 1993 has caused groundwater reservoirs to dry up inthe city discontinuance of water was experienced for severaldays or even weeks In those times the mayor of Istanbul cityhas lost the election of 1994 and he realized that this result wasdue to the peoplersquos reaction about the water shortages Manypolitical reports of that period also support this scientificview

5 Conclusions

Statistical analysis in temperature time series of the Kandillistation from 1912 to 2016 established that there is a notableincrease in temperature values after the 1940s which is inparallel with the beginning of industrialization era in Istan-bul There has been a rise about 094∘C in the daily averagetemperature series since the beginning of the last century Asignificant positive trend in the daily maximum temperatureseries is found about 156∘C Also there is a positive trendabout 087∘C in the daily minimum temperature series Onthe other hand analysis of the number of days with thedaily maximum temperature higher than 30∘C showed thatthere is an increasing trend Meanwhile analysis of thenumber of days with daily minimum temperatures lowerthan 0∘C showed a decreasing trend The increment in theminimum values of the daily minimum temperature series

Advances in Meteorology 13

is more evident than the maximum values of this series Inthis case these rising rates in the minimum temperatureseries are more evident for the section periods of 1912ndash1980and 1981ndash2016 than the section periods of 1912ndash1964 and1965ndash2016 This again shows that there is an increment inthe positive temperature trend from past to present decadesThe increment in the precipitation time series is not asclear as the increment in the temperature time series dueto periodic variability The trend analysis in the total annualprecipitation time series showed that the first significantupward trend has periodically been started from the 1920swhile there is a stable trend from 2001 till 2016 The dailyaverage of rainfall amounts has increased with a value of58mm during the period of 1912ndash2016 Also the analysisof heavy precipitation trend showed an increase of 61mmOverall the total average precipitation of Istanbul increasedwhile this increasing trend is more pronounced during theearly decades than the last 3 decades On the other handthe increasing rate of daily maximum precipitation is morepronounced in the last 3 decades than the previous decadesThen it was shown that the frequency of heavy rainfallat Istanbul has increased during the recent decades Thusthe precipitation changes in Istanbul have some differencescompared to the general tendency in precipitation trendthat was put forward by other studies as a decreasing trendover the whole of Turkey This result can be expressedas a positive effect of population overgrowth of Istanbulmegacity Comparison of the results in the first half of thestudy period (1912ndash1964) with the second half of the studyperiod (1965ndash2016) showed that both the average temperatureand average precipitation have higher values of 139∘C and878mm for the final phase compared to the values of 136∘Cand 799mm belonging to the initial phase Therefore it canbe stated that the megacity of Istanbul is directly affectedby the climate change and its consequences In this contextpotential risks of climate change in Istanbul megacity underhigher temperature conditions can be expressed as the rise inthe sea level increase in the rate of evapotranspiration andincrease in the frequency of heavy rainfall Also this city maynot be able to handle this uncontrolled population growthand its associated irreversible changes which is alreadypushing the natural limits by destroying the environmentTherefore the local governors of any megacity like Istanbulshould give more emphasis on the importance of sustainableurban development Thus it is urgent to prepare local andnational climate change strategies and action plans for themegacities

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

The authors are grateful to the Bogazici University theobservatory of Kandilli weather station and the Earthquake

Research Institute for providing the research data and tech-nical support The authors also gratefully acknowledge con-tributions of Assoc Professor Dr Yuksel Demirkaya Schoolof Social Sciences Marmara University This work has beensupported by Scientific and Technological Research Councilof Turkey (TUBITAK) under Grants 113R019 and 106Y258and by Marmara University (BAPKO) with projects FEN-E-120314-0066 FEN-C-YLP-090414-0102 FEN-L-250416-0180 and FEN-A-100413-0127

References

[1] D R Easterling B Horton P D Jones et al ldquoMaximum andminimum temperature trends for the globerdquo Science vol 277no 5324 pp 364ndash367 1997

[2] IPCC Climate Change 2014 IPCC Fifth Assessment Synthe-sis Report-Summary for Policymakers-an Assessment of Inter-Governmental Panel on Climate Change Cambridge UniversityPress Cambridge UK 2014

[3] J Carmin N Nadkarni and C Rhie Progress and Challenges inUrban Climate Adaptation Planning Results of a Global SurveyMIT Cambridge UK 2012

[4] A F Young ldquoUrban expansion and environmental risk in theSao Paulo Metropolitan Areardquo Climate Research vol 57 no 1pp 73ndash80 2013

[5] P Tian X Mu J Liu J Hu and C Gu ldquoImpacts of ClimateVariability and Human Activities on the Changes of Runoff andSediment Load in a Catchment of the Loess Plateau ChinardquoAdvances inMeteorology vol 2016 Article ID 4724067 15 pages2016

[6] R S Kovats and K L Ebi ldquoHeatwaves and public health inEuroperdquo European Journal of Public Health vol 16 no 6 pp592ndash599 2006

[7] S Conti P Meli G Minelli et al ldquoEpidemiologic studyof mortality during the Summer 2003 heat wave in ItalyrdquoEnvironmental Research vol 98 no 3 pp 390ndash399 2005

[8] J Kysely and J Kim ldquoMortality during heat waves in SouthKorea 1991 to 2005 how exceptional was the 1994 heat waverdquoClimate Research vol 38 no 2 pp 105ndash116 2009

[9] B Yan Z Xia F Huang L Guo and X Zhang ldquoClimatechange detection and annual extreme temperature analysis ofthe amur river basinrdquoAdvances inMeteorology vol 2016 ArticleID 6268938 14 pages 2016

[10] E M Fischer and R Knutti ldquoAnthropogenic contribution toglobal occurrence of heavy-precipitation and high-temperatureextremesrdquo Nature Climate Change vol 5 no 6 pp 560ndash5642015

[11] X Zhang L Alexander G C Hegerl et al ldquoIndices for moni-toring changes in extremes based on daily temperature andprecipitation datardquo Climate Change vol 2 no 6 pp 851ndash8702011

[12] K H Schlunzen P Hoffmann G Rosenhagen and W RieckeldquoLong-term changes and regional differences in temperatureand precipitation in the metropolitan area of Hamburgrdquo Inter-national Journal of Climatology vol 30 no 8 pp 1121ndash11362010

[13] G Bartolini M Morabito A Crisci et al ldquoRecent trends inTuscany (Italy) summer temperature and indices of extremesrdquoInternational Journal of Climatology vol 28 no 13 pp 1751ndash1760 2008

14 Advances in Meteorology

[14] S C Sheridan and T J Dolney ldquoHeat mortality and levelof urbanization measuring vulnerability across Ohio USArdquoClimate Research vol 24 no 3 pp 255ndash265 2003

[15] M Tayanc U Im M Dogruel andM Karaca ldquoClimate changein Turkey for the last half centuryrdquo Climatic Change vol 94 no3-4 pp 483ndash502 2009

[16] H Toros ldquoSpatio-temporal variation of daily extreme tempera-tures over Turkeyrdquo International Journal of Climatology vol 32no 7 pp 1047ndash1055 2012

[17] H Toros ldquoSpatio-temporal precipitation change assessmentsover Turkeyrdquo International Journal of Climatology vol 32 no9 pp 1310ndash1325 2012

[18] M Turkes C Yozgatlıgil I Batmaz et al ldquoHas the climate beenchanging in Turkey Regional climate change signals based on acomparative statistical analysis of two consecutive time periods1950-1980 and 1981-2010rdquoClimate Research vol 70 no 1 pp 77ndash93 2016

[19] IPCC ldquoummary for Policymakers In Climate Change 2013rdquo inThe Physical Science Basis The contribution of Working Group Ito the Fifth Assessment Report of the Intergovernmental Panel onClimate Change Cambridge University Press Cambridge UK2013

[20] United Nations ldquoDepartment of Economic and Social AffairsPopulation Division 2006 World Urbanization Prospects The2005 Revisionrdquo Working Paper ESAPWP200 2011

[21] Y Demirkaya Sayılarla Istanbul ITO Istanbul Turkey 2011[22] TUIK ldquoAddress based population registration system results of

2014rdquo Turkish Statistical Institute (TUIK) 2017 httpraporytuikgovtr10-03-2015-184727-842632346446643191142126876html

[23] S Erinc Climatology and its Methods Alfa Basım YayımDagitim Istanbul Turkey 4th edition 1965

[24] O M Gokturk D Bozkurt O L Sen and M Karaca ldquoQualitycontrol and homogeneity of Turkish precipitation datardquoHydro-logical Processes vol 22 no 16 pp 3210ndash3218 2008

[25] C Ley and D Paindaveine ldquoRuns Testsrdquo in Encyclopedia ofEnvironmetrics 2012

[26] A Ghasemi and S Zahediasl ldquoNormality tests for statisticalanalysis a guide for non-statisticiansrdquo International Journal ofEndocrinology andMetabolism vol 10 no 2 pp 486ndash489 2012

[27] G V Glass ldquoTesting Homogeneity of Variancesrdquo AmericanEducational Research Journal vol 3 no 3 pp 187ndash190 1966

[28] H BMann ldquoNonparametric tests against trendrdquo Econometricavol 13 pp 245ndash259 1945

[29] M G Kendall Rank Correlation Method Charles GriffinLondon UK 4th edition 1975

[30] H Turoglu ldquoDetection of Changes on Temperature and Precip-itation Features in Istanbul (Turkey)rdquo Atmospheric and ClimateSciences vol 04 no 04 pp 549ndash562 2014

[31] A Karaburun A Demirci and I-S Suen ldquoImpacts of urbangrowth on forest cover in Istanbul (1987-2007)rdquo EnvironmentalModeling amp Assessment vol 166 no 1-4 pp 267ndash277 2010

[32] K K Karanth LM Curran and J D Reuning-Scherer ldquoVillagesize and forest disturbance in Bhadra Wildlife SanctuaryWestern Ghats Indiardquo Biological Conservation vol 128 no 2pp 147ndash157 2006

[33] G Cakir C Un E Z Baskent S Kose F Sivrikaya andS Keles ldquoEvaluating urbanization fragmentation and landuseland cover change pattern in Istanbul city Turkey from 1971to 2002rdquo Land Degradation amp Development vol 19 no 6 pp663ndash675 2008

[34] R B Myneni F G Hall P J Sellers and A L Marshak ldquoTheinterpretation of spectral vegetation indexesrdquo IEEE Transac-tions on Geoscience and Remote Sensing vol 33 no 2 pp 481ndash486 1995

[35] Y S Unal H Toros A Deniz and S Incecik ldquoInfluence ofmeteorological factors and emission sources on spatial and tem-poral variations of PM10 concentrations in Istanbul metropoli-tan areardquo Atmospheric Environment vol 45 no 31 pp 5504ndash5513 2011

[36] M KaracaM Tayanc andH Toros ldquoEffects of urbanization onclimate of Istanbul and Ankarardquo Atmospheric Environment vol29 no 23 pp 3411ndash3421 1995

[37] Y Ezber O L Sen T Kindap and M Karaca ldquoClimatic effectsof urbanization in Istanbul a statistical and modeling analysisrdquoInternational Journal of Climatology vol 27 no 5 pp 667ndash6792007

[38] H S Park ldquoFeatures of the heat island in seoul and its sur-rounding citiesrdquo Atmospheric Environment (1967) vol 20 no10 pp 1859ndash1866 1986

Advances in Meteorology 7

Tmin Tavg Tmax

Tmax Tmin DTR

u(t)u(t)

u(t)

u㰀(t)u㰀(t)

u㰀(t)

u(t)u㰀(t)

u㰀(t)

u(t)

u(t)

R2 = 02642

y = 00149x minus 10794R2 = 0163

y = 0009x minus 3926

R2 = 0233

y = 00116x minus 12375

R2 = 00182y = 00032x + 16875

R2 = 00734

y = minus00934x + 20292R2 = 02513

y = 02039x minus 37934

(a1)

(a2)

(b1)

(b2)

(c1)

(c2)

(d1)

(d2)

(e1)

(e2)

(f1)

(f2)

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1912

1976

1984

1992

2000

2008

2016

1968

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

2016

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1912

1976

1984

1992

2000

2008

2016

1968

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

minus4minus3minus2minus1

01234567

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus6minus5minus4minus3minus2minus1

0123456

minus4minus3minus2minus1

0123

010203040506070

Num

ber o

f day

s (gt30∘ C)

0102030405060

65707580859095

100

Tem

pera

ture

(∘C)

minus3minus2minus1

0123456

minus3minus2minus1

012345

minus3minus2minus1

01234567

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

9

10

11

12

13Te

mpe

ratu

re (∘

C)

12

13

14

15

16

Tem

pera

ture

(∘C)

16

17

18

19

20

21

Tem

pera

ture

(∘C)

Num

ber o

f day

s (lt0∘ C)

Figure 3 Trend analysis of the temperatures time series using linear best-fit curve andMK test for daily minimum temperature (a1 a2) dailyaverage temperature (b1 b2) daily maximum temperature (c1 c2) number of days with daily maximum temperature gt 30∘C (d1 d2) numberof days with daily minimum temperature lt 0∘C (e1 e2) and daily temperature range (f1 f2)

between 1944 and 1969 and between 1991 and 2016 (Figures3(c1) and 3(c2)) There is a positive trend about 11∘C indaily minimum temperature series during the whole studiedperiod too while this positive trend value is smaller than thepositive trend value of maximum temperature The trend ofminimum temperature series is stable during the period of1912ndash1944 Then it shows a steady increase during 1942 to1989 which is most pronounced during 1944ndash1954 Finallythe rising trends in the minimum temperature series became

significant based onMK testrsquos result for the last period of 2003to 2016 (Figures 3(a1) and 3(a2)) In this case the minimumtemperature shows increasing change for all season wherethe most increasing change has happened in summer andspring season in all studied periods respectively (Table 2)Analysis of the number of days having daily maximumtemperature higher than 30∘C for the long time period of1912 to 2016 by focusing precisely on its subperiods showedthat there is almost no trend during 1912ndash1945 a significant

8 Advances in Meteorology

400

500

600

700

800

900

1000

1100

1200

1300

1912 1920 1928 1936 1944 1952 1960 1968 1976 1984 1992 2000 2008 2016

Tota

l pre

cipi

tatio

n (m

m)

Years

Average 8375 mmMaximum 12894 mm in 1981Minimum 4487 mm in 1921

(a)

0

20

40

60

80

100

120

140

Janu

ary

Febr

uary

Mar

chAp

rilM

ayJu

ne July

Augu

stSe

ptem

ber

Oct

ober

Nov

embe

rD

ecem

ber

Tota

l pre

cipi

tatio

n (m

m)

Months

Average 699 mmMaximum 1284 mm in December

Minimum 319 mm in July

(b)

Figure 4 Time series of total annual precipitation (a) and total monthly precipitation (b)

increasing trend during 1945ndash1969 a relative reduction trendduring 1969ndash1977 an increasing trend during 1977ndash1992and a more pronounced increasing trend during 1992ndash2007(Figures 3(d1) and 3(d2)) On the other hand analysis of thenumber of dayswith dailyminimum temperatures lower than0∘C shows a decreasing trend generally So there is a relativeincrease during 1915ndash1930 almost no trend during 1930ndash1954an increase during 1954ndash1963 a decrease during 1963ndash1985almost no trend during 1985ndash1995 and a decrease during1995ndash2011 by a detailed focus on subperiods of the time series(Figures 3(e1) and 3(e2)) When the daily temperature rangein the whole long period is analyzed an increment can beclearly observed The situation was stable in the period of1912ndash1950There is an increasing trend during 1950ndash1970 andan obvious decrement during 1970ndash1985 Also there is anincrease between 1985 and 2010 (Figures 3(f1) and 3(f2))

The comparison results between the periods of 1912ndash1964and 1965ndash2016 and the periods of 1912ndash1980 and 1981ndash2016showed that (Table 1) there is an increment in the dailyaverage temperature series about 04∘C for the periods of1965ndash2016 (14∘C) compared to previous period of 1912ndash1964(136∘C) and an increment about 05∘C for the periods of1981ndash2016 (141∘C) compared to previous period of 1912ndash1980(136∘C) respectively In addition the standard deviation isincreased by a coefficient between 01∘C and 02∘C in thewhole studied period The increment of the minimum valuesin the daily average temperature series is more evident thanthe maximum values by a mean coefficient of about 6 timesfor both of studied periodsTheminimumvalue of daily aver-age temperatures in the periods of 1912ndash1964 and 1965ndash2016revealed an increment of 06∘C and 08∘C respectively Themaximum value of daily average temperatures in the periodsof 1912ndash1964 and 1965ndash2016 showed a decrement of 02∘Cand 01∘C respectively When the monthly values of dailyaverage temperature are analyzed the highest increment in

the periods of 1912ndash1964 and 1965ndash2016 is found to be inJune at a rate of 1∘C The highest increment in the periodsof 1912ndash1980 and 1981ndash2016 appeared to be in June againwith a rate of 11∘C Also there has been temperature decre-ment in October November and December for all studiedperiods The highest decrement was in November with arate of minus05∘C in the periods of 1912ndash1964 and 1965ndash2016 aswell as minus04∘C in the periods of 1912ndash1980 and 1981ndash2016In this case the analysis of seasonal values of the dailyaverage temperatures showed that the highest increment inthe periods of 1912ndash1964 and 1965ndash2016 took place in thesummer season at a rate of 07∘C The highest increment inthe periods of 1912ndash1980 and 1981ndash2016 was found to be insummer again by a rate of 11∘C Also there was a temperaturedecrement in the autumn season which was about minus010∘Conly for the periods of 1912ndash1964 and 1965ndash2016 Analysisof the percentile thresholds of daily average temperaturesshowed that the temperature increment at the 5th percentileis 05∘C for the periods of 1912ndash1964 and 1965ndash2016 while thisincrement is 03∘C for the periods of 1912ndash1980 and 1981ndash2016Therefore it can be said that the rate of temperature risinghas increased further as much as the time is closer to the endyears of studied period Also these increment values indicatean increase of 02∘C and 03∘C at the 25th percentile 05∘Cand 08∘C at the 75th percentile and 06∘C and 11∘C at the95th percentile thresholds for thewhole studied subperiods of1912ndash1964 1965ndash2016 1912ndash1980 and 1981ndash2016 respectivelyThis situation shows that the increment of higher values inthe daily average temperatures is greater than the incrementof lower values and also this increment ismore evident for theperiods of 1912ndash1980 and 1981ndash2016 compared to the periodsof 1912ndash1964 and 1965ndash2016

Comparison of the daily maximum temperature seriesfor two periods of 1912ndash1964 and 1965ndash2016 revealed anincrement of 08∘C which can be divided into an average

Advances in Meteorology 9

1912

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

Annual total

Years

Annual maximum daily precip

Annual heavy precipitation STD

R2 = 00549

y = 12491x minus 16171

R2 = 00103

y = 0087x minus 1104

u(t)u(t)

u㰀(t)

u㰀(t)

R2 = 2E minus 08y = minus1E minus 05x + 92115

R2 = 00091

y = 00082x minus 99366

u(t)

u㰀(t)

u(t)

u㰀(t)

(a1)

(a2)

(b1)

(b2)

(c1)

(c2)

(d1)

(d2)

400500600700800900

1000110012001300

Am

ount

of p

reci

p (m

m)

Am

ount

of p

reci

p (m

m)

20

40

60

80

100

120

140

160

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

minus3

minus2

minus1

0

1

2

3

4

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus3

minus2

minus1

0

1

2

3

2016

1912

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1928

Years

2468

101214161820

Am

ount

of p

reci

p (m

m)

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

0

2

4

6

8

10

12

14

Num

ber o

f day

s (gt25

mm

)

minus3

minus2

minus1

0

1

2

3

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus3

minus2

minus1

0

1

2

3

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

2016

1912

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1928

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

Figure 5 Trend analysis of the precipitation time series using least squares linear regression and MK trend test for total annual precipitation(a1 a2) daily maximum precipitation (b1 b2) daily maximum precipitation greater than 25mm (c1 c2) and standard deviation in the dailyprecipitation series (d1 d2)

10 Advances in Meteorology

Table2Statisticalanalyses

outputso

ftem

perature

andprecipitatio

nparameters(SD

stand

arddeviation

Inc

increasedecdecrease)

Parameters

Minim

umtemperature

(∘ C)

Averagetem

perature

(∘ C)

Maxim

umtemperature

(∘ C)

Precipitatio

n(m

m)

Statistics

Perio

ds

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Incor

dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

Average

total

102

107

052

102

109

071

136

140

037

136

141

043

180

187

075

181

187

060

8113

8656

544

8267

8591

324

SD66

66

004

65

67

022

7172

007

7173

019

82

84

021

82

85

025

9391minus03

9193

02

Minim

umminus54minus42

116minus53minus39

138minus27minus21

062minus27minus19

074minus08minus01

076minus07

00

068

00

00

00

00

00

00

Maxim

um223

229

069

222

232

099

276

278

023

277

277

000

356

362

059

358

361

029

589

620

31

579

651

71Percentile5

minus04

02

061minus03

03

055

21

25

049

22

24

026

45

50

052

46

48

015

01

01minus01

01

01

00

Percentile2

547

52

045

48

52

038

7780

024

7778

01 0

114

119

054

115

116

011

07

05minus03

06

Percentile5

0106

108

019

105

106

007

140

141

013

140

138minus020

186

190

040

187

185minus024

26

21minus05

23

24

01

Percentile7

5159

165

066

158

164

059

199

204

054

199

201

022

250

261

105

252

255

034

7672minus04

7374

01

Percentile9

5196

203

074

195

202

074

236

242

065

236

239

032

296

307

110

297

300

027

229

226minus03

227

223minus05

gt30

Cggt80

mm

00

00

00

00

00

00

1626

1017

2810

00

00

00

gt25

Cggt40

mm

00

00

00

712

57

147

90107

1794

108

142

20

22

0lt0C

gmdash

2316minus6

2216minus6

76minus2

76minus1

21minus1

21minus1

mdashmdash

mdashmdash

mdashmdash

ltminus5C

gmdash

21minus1

21minus1

00

00

00

00

00

00

mdashmdash

mdashmdash

mdashmdash

Winter

33

36

032

33

38

048

59

62

031

60

62

020

9095

047

9195

035

1033

1096

63

1056

1066

09

Sprin

g74

87

131

7582

073

111

124

121

113

119

060

160

171

116

162

170

076

496

513

17515

511minus04

Summer

174

187

124

174

186

125

216

224

082

216

225

091

269

273

043

271

282

109

334

331minus03

330

368

38

Autumn

124

127

028

123

127

036

156

156

000

156

155minus001

199

197minus020

200

201

019

842

831minus10

854

919

65

Ann

ual

101

109

079

101

108

070

136

141

057

136

140

043

179

184

042

181

187

060

676

719

43

689

716

27

Advances in Meteorology 11

value of 18∘C for the period of 1912ndash1964 and an averagevalue of 187∘C for the period of 1965ndash2016 On the otherhand comparison of the daily maximum temperature seriesfor the periods of 1912ndash1980 and 1981ndash2016 has revealed anincrement of 06∘C with an average value between 182∘C and188∘C for both of these periods respectively In additionthe standard deviation values are increased between 02∘Cand 03∘C during the whole studied periodThe increment inthe minimum values of daily maximum temperature seriesis more evident than the maximum values The minimumvalues of daily maximum temperatures have exhibited anincrement of 08∘C for the period of 1965ndash2016 compared tothe previous period of 1912ndash1964 and also an increment of07∘C for the period of 1981ndash2016 compared to the previousperiod of 1912ndash1980 Also themaximumvalues of dailymaxi-mum temperature series exhibited amean increment of 05∘Cfor both studied periods Seasonal analysis of maximumtemperature series for both of studied section periods showedthat the highest rising temperature has been happened bya value of 12∘C in summer On the other hand the lowestincrement has happened in the autumn season The monthlyanalysis of daily maximum temperature series showed thatthe highest increment took place during the last monthsof the spring season and the first month of the summerseason in all of the studied periods When the percentiles ofdaily maximum temperatures are analyzed the temperatureincrement based on the 5th percentile threshold is 05∘C forthe periods of 1912ndash1964 and 1965ndash2016 while the incrementis 03∘C for the periods of 1912ndash1980 and 1981ndash2016 This canbe considered as an important sign of rising temperature overthe time In this case the value of percentile thresholds isincreased with extending the length of the time period withextending the length of the first section of studied period infavor of last years than the previous ones In this regard thesevalues are 05∘C and 04∘C at the 25th percentile 1∘C and11∘C at the 75th percentile and 11∘C and 12∘C at the 95thpercentile for both sections of studied periods respectivelyThis situation shows that the increment of higher values inthe dailymaximum temperatures is greater than lower values

The comparison analysis of daily minimum temperaturesbetween the periods of 1912ndash1964 and 1965ndash2016 and theperiods of 1912ndash1980 and 1981ndash2016 showed that there is ageneral increment of 05∘C during the first section periodswhich can be given as 102∘C and 107∘C for the individualperiods of 1912ndash1964 and 1965ndash2016 and also a generalincrement of 08∘C during the second section periods whichcan be provided as 102∘C and 11∘C for the individual periodsof 1912ndash1980 and 1981ndash2016 respectively In addition thestandard deviation values increased among these sectionperiods from 0∘C to 03∘C The increment in the minimumvalues of daily minimum temperature series is more evidentthan the maximum values The minimum values of dailyminimum temperature series showed an increment of 12∘Cfrom the period of 1912ndash1964 to 1965ndash2016 and also anincrement of 16∘C from the period of 1912ndash1980 to 1981ndash2016The maximum values of daily minimum temperature serieshave shown an increment of 07∘C for the section periodsof 1912ndash1964 and 1965ndash2016 and an increment of 12∘C forthe section periods of 1912ndash1980 and 1981ndash2016 Overall

the minimum values have had an average increment of14∘C while the maximum values have had an averageincrement of 1∘C during the last century which can showthe higher rate of upward trends in the temperature timeseries Monthly analysis of minimum temperature seriesshowed that the highest increment for the section studiedperiods of 1912ndash1964 and 1965ndash2016 has occurred by a valueof 1∘C in June while the highest increment for the sectionstudied periods of 1912ndash1980 and 1981ndash2016 has occurred bya value of 16∘C in August Meanwhile there is a temperaturedecrement in November in both of these periods Howeverthere is a clear decrement in the first half of these periodsin October and December whereas there is an incrementin the second half of these periods The seasonal analysisof the daily minimum temperature series showed that thehighest increment has happened in the summer season witha mean value of 07∘C for the section periods of 1912ndash1964and 1965ndash2016 and with a mean value of 13∘C for the sectionperiods of 1912ndash1980 and 1981ndash2016 respectively Then theincreasing rate during the summer season became moreevident during the recent decades Analysis of the percentilethresholds of daily minimum temperature series showed thatthe temperature increment at the 5th percentile is 06∘C forall studied time periods of 1912ndash1964 1965ndash2016 1912ndash1980and 1981ndash2016 Also these increment values indicate anincrease of 04∘C and 06∘C at the 25th percentile 07∘C and11∘C at the 75th percentile and 07∘C and 14∘C at the 95thpercentile thresholds for the whole studied subperiods of1912ndash1964 1965ndash2016 1912ndash1980 and 1981ndash2016 respectivelyThese rising rates in minimum temperature series are moreevident for the periods of 1912ndash1980 and 1981ndash2016 than theperiods of 1912ndash1980 and 1981ndash2016 Therefore it can be saidthat the rate of temperature rising has increased further aslong as the studied time period is closer to the last years

42 Precipitation Annual average precipitation in Istanbulis 838mm with a range of minimum value of 449mm in1921 and a maximum value of 1289mm in 1981 based on theobservatory data of Kandilli station during the whole studiedperiod from 1912 to 2016 (Figure 4(a)) Also monthly averageprecipitation is 699mm with a range of the minimum of326mm in July and the maximum of 1286mm in Decemberduring the whole studied time period (Figure 4(b))

Analysis of the trend in the annual average precipitationtime series by the methods of linear regression analysisand MK trend test has shown that periodically there arepartial increments and significant differences during the totalstudied period from 1912 to 2016 (Figures 5(a1) and 5(a2))But this increment in the precipitation time series is notas clear as the increment in the temperature time seriesHowever it is obvious in the precipitation time series thatthere is an increment between the years of 1917 and 1925a stable condition between the years of 1925 and 1954 anincrement between the years of 1954 and 1965 a decrementbetween the years of 1965 and 1974 an increment between theyears of 1974 and 2001 and again a no change situation from2001 till the end The trend analysis of the daily maximumprecipitation series showed an increment of 29mm for theperiods of 1912ndash1964 and 1965ndash2016 as well as an increment

12 Advances in Meteorology

of 93mm for the periods of 1912ndash1980 and 1981ndash2016 Thesevalues indicated that rainfall has increased at a rate of morethan three times over the last decades than the previous ones(Figures 5(b1) and 5(b2)) Also the number of days with dailyprecipitation greater than 25mm presented an incrementduring the period of 1912ndash2016 although this is not significantat the confidence level of 005 (Figures 5(c1) and 5(c2))The trend of standard deviation in the daily precipitationtime series showed a slowly increasing trend during the totalstudied period (Figures 5(d1) and 5(d2))

The comparison of the results of statistical analysisbetween the daily average rainfall amounts belonging tothe periods of 1912ndash1964 and 1965ndash2016 with those of theperiods of 1912ndash1980 and 1981ndash2016 revealed that there is anincrement of 78mm from the period of 1912ndash1964 to theperiod of 1965ndash2016 and an increment of 38mm from theperiod of 1912ndash1980 to the period of 1981ndash2016 In additionthe analysis of standard deviation exhibited a decrementfrom the period of 1912ndash1964 to the period of 1965ndash2016 andan increment from the period of 1912ndash1980 to the periodof 1981ndash2016 (Table 2) The analysis of monthly averageprecipitation time series showed that the highest incrementhas happened in October with 227mm for the sectionperiods of 1912ndash1964 and 1965ndash2016 and with 347mm forthe section periods of 1912ndash1980 and 1981ndash2016 On the otherhand the highest decrease took placewith a value ofminus04mmduring both May and July for the periods of 1912ndash1964 and1965ndash2016 and with a value of minus16mm in September for thesection periods of 1912ndash1980 and 1981ndash2016 The analysis ofseasonal average precipitation time series showed that thehighest increase occurred in autumn with a value of 72mmfor the section periods of 1912ndash1964 and 1965ndash2016 as wellas with a value of 64mm for the periods of 1912ndash1980 and1981ndash2016 respectively There is no remarkable decreasingchange seasonally except in winter season during the periodsof 1912ndash1980 and 1981ndash2016 Furthermore the percentilethresholds of daily average precipitation indicated that thereis an insignificant negative trend based on all percentilesand for all studied periods Overall the statistical analysisshowed that the total average precipitation of Istanbul hasincreased while this increasing trend is more pronounced inthe previous decades than the last 3 decades On the otherhand the increasing rate of daily maximum precipitation ismore evident during the last 3 decades than the previousdecades which can be proven by the increasing frequency ofheavy rainfall events in Istanbul

Generally the results of trend analysis of Kandilli stationduring the last 105 years of 1912ndash2016 showed that there isa warming significant trend in the precipitation time seriesby using both methods of linear regression analysis andMK trend test On the contrary previous climate studiesconducted over Turkey put forward that there has been adecreasing trend in annual precipitation time series duringthe recent decades regionally The results of a previouslyconducted study by using the daily precipitation and tem-perature data sets of Florya and Goztepe meteorologicalstations in Istanbul area between 1960 and 2013 showedthat most notably the precipitation during the warm periodshas decreased but the frequency of the intense rain has

increased and the majority of these episodes of intense raincoincided with the warm periods Other determinationswere the rise in the annual average temperature and theextension of the warm periods in a year This differentiationof the temperature features can lead to the aggravation of theevaporation and it can be effective for a longer period duringthe year [15] Thus it will make Istanbul be confronted withthe much more important problems of water managementand flood [34] Also the results of the current study for trendanalysis in the long period from 1912 to 2016 showed that themost striking spell is between the years of 1968 and 1998 dueto the existence of least number of rainfall events in IstanbulIt can be owing to industrialization along with the increasingair pollution as well as irregular urbanization in Istanbul areaIn this case severe droughts taking place during the yearsof 1988 1992 1993 and 2008 have threatened the reservoirswhich supplied fresh water of the city These years werecharacterized by not having enough rainfall events Theseyears are also characterized by more persistent high-pressuresystems and less occurrence of low-pressure systems in termsof number and strength As statements made by officialinstitutions the formation conditions for the atmosphericlayer of air pollution due to air pollutant emissions from fossilfuel combustion and industrial activities are more providedduring anticyclone or high-pressure system eventsThereforethis leads to warming up and generating an inversion layerin the boundary layer of atmosphere especially over city area[35] The inversion layer or urban heat island intensity isincreasing with the increasing city size andor populationa phenomenon that was also reported by others [36ndash38]Moreover these last climatic events also have affected somepolitical results beyond natural effects The most obviousexample of this is related to the local election of 1994 inIstanbul In this case extreme drought during the summers of1992 and 1993 has caused groundwater reservoirs to dry up inthe city discontinuance of water was experienced for severaldays or even weeks In those times the mayor of Istanbul cityhas lost the election of 1994 and he realized that this result wasdue to the peoplersquos reaction about the water shortages Manypolitical reports of that period also support this scientificview

5 Conclusions

Statistical analysis in temperature time series of the Kandillistation from 1912 to 2016 established that there is a notableincrease in temperature values after the 1940s which is inparallel with the beginning of industrialization era in Istan-bul There has been a rise about 094∘C in the daily averagetemperature series since the beginning of the last century Asignificant positive trend in the daily maximum temperatureseries is found about 156∘C Also there is a positive trendabout 087∘C in the daily minimum temperature series Onthe other hand analysis of the number of days with thedaily maximum temperature higher than 30∘C showed thatthere is an increasing trend Meanwhile analysis of thenumber of days with daily minimum temperatures lowerthan 0∘C showed a decreasing trend The increment in theminimum values of the daily minimum temperature series

Advances in Meteorology 13

is more evident than the maximum values of this series Inthis case these rising rates in the minimum temperatureseries are more evident for the section periods of 1912ndash1980and 1981ndash2016 than the section periods of 1912ndash1964 and1965ndash2016 This again shows that there is an increment inthe positive temperature trend from past to present decadesThe increment in the precipitation time series is not asclear as the increment in the temperature time series dueto periodic variability The trend analysis in the total annualprecipitation time series showed that the first significantupward trend has periodically been started from the 1920swhile there is a stable trend from 2001 till 2016 The dailyaverage of rainfall amounts has increased with a value of58mm during the period of 1912ndash2016 Also the analysisof heavy precipitation trend showed an increase of 61mmOverall the total average precipitation of Istanbul increasedwhile this increasing trend is more pronounced during theearly decades than the last 3 decades On the other handthe increasing rate of daily maximum precipitation is morepronounced in the last 3 decades than the previous decadesThen it was shown that the frequency of heavy rainfallat Istanbul has increased during the recent decades Thusthe precipitation changes in Istanbul have some differencescompared to the general tendency in precipitation trendthat was put forward by other studies as a decreasing trendover the whole of Turkey This result can be expressedas a positive effect of population overgrowth of Istanbulmegacity Comparison of the results in the first half of thestudy period (1912ndash1964) with the second half of the studyperiod (1965ndash2016) showed that both the average temperatureand average precipitation have higher values of 139∘C and878mm for the final phase compared to the values of 136∘Cand 799mm belonging to the initial phase Therefore it canbe stated that the megacity of Istanbul is directly affectedby the climate change and its consequences In this contextpotential risks of climate change in Istanbul megacity underhigher temperature conditions can be expressed as the rise inthe sea level increase in the rate of evapotranspiration andincrease in the frequency of heavy rainfall Also this city maynot be able to handle this uncontrolled population growthand its associated irreversible changes which is alreadypushing the natural limits by destroying the environmentTherefore the local governors of any megacity like Istanbulshould give more emphasis on the importance of sustainableurban development Thus it is urgent to prepare local andnational climate change strategies and action plans for themegacities

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

The authors are grateful to the Bogazici University theobservatory of Kandilli weather station and the Earthquake

Research Institute for providing the research data and tech-nical support The authors also gratefully acknowledge con-tributions of Assoc Professor Dr Yuksel Demirkaya Schoolof Social Sciences Marmara University This work has beensupported by Scientific and Technological Research Councilof Turkey (TUBITAK) under Grants 113R019 and 106Y258and by Marmara University (BAPKO) with projects FEN-E-120314-0066 FEN-C-YLP-090414-0102 FEN-L-250416-0180 and FEN-A-100413-0127

References

[1] D R Easterling B Horton P D Jones et al ldquoMaximum andminimum temperature trends for the globerdquo Science vol 277no 5324 pp 364ndash367 1997

[2] IPCC Climate Change 2014 IPCC Fifth Assessment Synthe-sis Report-Summary for Policymakers-an Assessment of Inter-Governmental Panel on Climate Change Cambridge UniversityPress Cambridge UK 2014

[3] J Carmin N Nadkarni and C Rhie Progress and Challenges inUrban Climate Adaptation Planning Results of a Global SurveyMIT Cambridge UK 2012

[4] A F Young ldquoUrban expansion and environmental risk in theSao Paulo Metropolitan Areardquo Climate Research vol 57 no 1pp 73ndash80 2013

[5] P Tian X Mu J Liu J Hu and C Gu ldquoImpacts of ClimateVariability and Human Activities on the Changes of Runoff andSediment Load in a Catchment of the Loess Plateau ChinardquoAdvances inMeteorology vol 2016 Article ID 4724067 15 pages2016

[6] R S Kovats and K L Ebi ldquoHeatwaves and public health inEuroperdquo European Journal of Public Health vol 16 no 6 pp592ndash599 2006

[7] S Conti P Meli G Minelli et al ldquoEpidemiologic studyof mortality during the Summer 2003 heat wave in ItalyrdquoEnvironmental Research vol 98 no 3 pp 390ndash399 2005

[8] J Kysely and J Kim ldquoMortality during heat waves in SouthKorea 1991 to 2005 how exceptional was the 1994 heat waverdquoClimate Research vol 38 no 2 pp 105ndash116 2009

[9] B Yan Z Xia F Huang L Guo and X Zhang ldquoClimatechange detection and annual extreme temperature analysis ofthe amur river basinrdquoAdvances inMeteorology vol 2016 ArticleID 6268938 14 pages 2016

[10] E M Fischer and R Knutti ldquoAnthropogenic contribution toglobal occurrence of heavy-precipitation and high-temperatureextremesrdquo Nature Climate Change vol 5 no 6 pp 560ndash5642015

[11] X Zhang L Alexander G C Hegerl et al ldquoIndices for moni-toring changes in extremes based on daily temperature andprecipitation datardquo Climate Change vol 2 no 6 pp 851ndash8702011

[12] K H Schlunzen P Hoffmann G Rosenhagen and W RieckeldquoLong-term changes and regional differences in temperatureand precipitation in the metropolitan area of Hamburgrdquo Inter-national Journal of Climatology vol 30 no 8 pp 1121ndash11362010

[13] G Bartolini M Morabito A Crisci et al ldquoRecent trends inTuscany (Italy) summer temperature and indices of extremesrdquoInternational Journal of Climatology vol 28 no 13 pp 1751ndash1760 2008

14 Advances in Meteorology

[14] S C Sheridan and T J Dolney ldquoHeat mortality and levelof urbanization measuring vulnerability across Ohio USArdquoClimate Research vol 24 no 3 pp 255ndash265 2003

[15] M Tayanc U Im M Dogruel andM Karaca ldquoClimate changein Turkey for the last half centuryrdquo Climatic Change vol 94 no3-4 pp 483ndash502 2009

[16] H Toros ldquoSpatio-temporal variation of daily extreme tempera-tures over Turkeyrdquo International Journal of Climatology vol 32no 7 pp 1047ndash1055 2012

[17] H Toros ldquoSpatio-temporal precipitation change assessmentsover Turkeyrdquo International Journal of Climatology vol 32 no9 pp 1310ndash1325 2012

[18] M Turkes C Yozgatlıgil I Batmaz et al ldquoHas the climate beenchanging in Turkey Regional climate change signals based on acomparative statistical analysis of two consecutive time periods1950-1980 and 1981-2010rdquoClimate Research vol 70 no 1 pp 77ndash93 2016

[19] IPCC ldquoummary for Policymakers In Climate Change 2013rdquo inThe Physical Science Basis The contribution of Working Group Ito the Fifth Assessment Report of the Intergovernmental Panel onClimate Change Cambridge University Press Cambridge UK2013

[20] United Nations ldquoDepartment of Economic and Social AffairsPopulation Division 2006 World Urbanization Prospects The2005 Revisionrdquo Working Paper ESAPWP200 2011

[21] Y Demirkaya Sayılarla Istanbul ITO Istanbul Turkey 2011[22] TUIK ldquoAddress based population registration system results of

2014rdquo Turkish Statistical Institute (TUIK) 2017 httpraporytuikgovtr10-03-2015-184727-842632346446643191142126876html

[23] S Erinc Climatology and its Methods Alfa Basım YayımDagitim Istanbul Turkey 4th edition 1965

[24] O M Gokturk D Bozkurt O L Sen and M Karaca ldquoQualitycontrol and homogeneity of Turkish precipitation datardquoHydro-logical Processes vol 22 no 16 pp 3210ndash3218 2008

[25] C Ley and D Paindaveine ldquoRuns Testsrdquo in Encyclopedia ofEnvironmetrics 2012

[26] A Ghasemi and S Zahediasl ldquoNormality tests for statisticalanalysis a guide for non-statisticiansrdquo International Journal ofEndocrinology andMetabolism vol 10 no 2 pp 486ndash489 2012

[27] G V Glass ldquoTesting Homogeneity of Variancesrdquo AmericanEducational Research Journal vol 3 no 3 pp 187ndash190 1966

[28] H BMann ldquoNonparametric tests against trendrdquo Econometricavol 13 pp 245ndash259 1945

[29] M G Kendall Rank Correlation Method Charles GriffinLondon UK 4th edition 1975

[30] H Turoglu ldquoDetection of Changes on Temperature and Precip-itation Features in Istanbul (Turkey)rdquo Atmospheric and ClimateSciences vol 04 no 04 pp 549ndash562 2014

[31] A Karaburun A Demirci and I-S Suen ldquoImpacts of urbangrowth on forest cover in Istanbul (1987-2007)rdquo EnvironmentalModeling amp Assessment vol 166 no 1-4 pp 267ndash277 2010

[32] K K Karanth LM Curran and J D Reuning-Scherer ldquoVillagesize and forest disturbance in Bhadra Wildlife SanctuaryWestern Ghats Indiardquo Biological Conservation vol 128 no 2pp 147ndash157 2006

[33] G Cakir C Un E Z Baskent S Kose F Sivrikaya andS Keles ldquoEvaluating urbanization fragmentation and landuseland cover change pattern in Istanbul city Turkey from 1971to 2002rdquo Land Degradation amp Development vol 19 no 6 pp663ndash675 2008

[34] R B Myneni F G Hall P J Sellers and A L Marshak ldquoTheinterpretation of spectral vegetation indexesrdquo IEEE Transac-tions on Geoscience and Remote Sensing vol 33 no 2 pp 481ndash486 1995

[35] Y S Unal H Toros A Deniz and S Incecik ldquoInfluence ofmeteorological factors and emission sources on spatial and tem-poral variations of PM10 concentrations in Istanbul metropoli-tan areardquo Atmospheric Environment vol 45 no 31 pp 5504ndash5513 2011

[36] M KaracaM Tayanc andH Toros ldquoEffects of urbanization onclimate of Istanbul and Ankarardquo Atmospheric Environment vol29 no 23 pp 3411ndash3421 1995

[37] Y Ezber O L Sen T Kindap and M Karaca ldquoClimatic effectsof urbanization in Istanbul a statistical and modeling analysisrdquoInternational Journal of Climatology vol 27 no 5 pp 667ndash6792007

[38] H S Park ldquoFeatures of the heat island in seoul and its sur-rounding citiesrdquo Atmospheric Environment (1967) vol 20 no10 pp 1859ndash1866 1986

8 Advances in Meteorology

400

500

600

700

800

900

1000

1100

1200

1300

1912 1920 1928 1936 1944 1952 1960 1968 1976 1984 1992 2000 2008 2016

Tota

l pre

cipi

tatio

n (m

m)

Years

Average 8375 mmMaximum 12894 mm in 1981Minimum 4487 mm in 1921

(a)

0

20

40

60

80

100

120

140

Janu

ary

Febr

uary

Mar

chAp

rilM

ayJu

ne July

Augu

stSe

ptem

ber

Oct

ober

Nov

embe

rD

ecem

ber

Tota

l pre

cipi

tatio

n (m

m)

Months

Average 699 mmMaximum 1284 mm in December

Minimum 319 mm in July

(b)

Figure 4 Time series of total annual precipitation (a) and total monthly precipitation (b)

increasing trend during 1945ndash1969 a relative reduction trendduring 1969ndash1977 an increasing trend during 1977ndash1992and a more pronounced increasing trend during 1992ndash2007(Figures 3(d1) and 3(d2)) On the other hand analysis of thenumber of dayswith dailyminimum temperatures lower than0∘C shows a decreasing trend generally So there is a relativeincrease during 1915ndash1930 almost no trend during 1930ndash1954an increase during 1954ndash1963 a decrease during 1963ndash1985almost no trend during 1985ndash1995 and a decrease during1995ndash2011 by a detailed focus on subperiods of the time series(Figures 3(e1) and 3(e2)) When the daily temperature rangein the whole long period is analyzed an increment can beclearly observed The situation was stable in the period of1912ndash1950There is an increasing trend during 1950ndash1970 andan obvious decrement during 1970ndash1985 Also there is anincrease between 1985 and 2010 (Figures 3(f1) and 3(f2))

The comparison results between the periods of 1912ndash1964and 1965ndash2016 and the periods of 1912ndash1980 and 1981ndash2016showed that (Table 1) there is an increment in the dailyaverage temperature series about 04∘C for the periods of1965ndash2016 (14∘C) compared to previous period of 1912ndash1964(136∘C) and an increment about 05∘C for the periods of1981ndash2016 (141∘C) compared to previous period of 1912ndash1980(136∘C) respectively In addition the standard deviation isincreased by a coefficient between 01∘C and 02∘C in thewhole studied period The increment of the minimum valuesin the daily average temperature series is more evident thanthe maximum values by a mean coefficient of about 6 timesfor both of studied periodsTheminimumvalue of daily aver-age temperatures in the periods of 1912ndash1964 and 1965ndash2016revealed an increment of 06∘C and 08∘C respectively Themaximum value of daily average temperatures in the periodsof 1912ndash1964 and 1965ndash2016 showed a decrement of 02∘Cand 01∘C respectively When the monthly values of dailyaverage temperature are analyzed the highest increment in

the periods of 1912ndash1964 and 1965ndash2016 is found to be inJune at a rate of 1∘C The highest increment in the periodsof 1912ndash1980 and 1981ndash2016 appeared to be in June againwith a rate of 11∘C Also there has been temperature decre-ment in October November and December for all studiedperiods The highest decrement was in November with arate of minus05∘C in the periods of 1912ndash1964 and 1965ndash2016 aswell as minus04∘C in the periods of 1912ndash1980 and 1981ndash2016In this case the analysis of seasonal values of the dailyaverage temperatures showed that the highest increment inthe periods of 1912ndash1964 and 1965ndash2016 took place in thesummer season at a rate of 07∘C The highest increment inthe periods of 1912ndash1980 and 1981ndash2016 was found to be insummer again by a rate of 11∘C Also there was a temperaturedecrement in the autumn season which was about minus010∘Conly for the periods of 1912ndash1964 and 1965ndash2016 Analysisof the percentile thresholds of daily average temperaturesshowed that the temperature increment at the 5th percentileis 05∘C for the periods of 1912ndash1964 and 1965ndash2016 while thisincrement is 03∘C for the periods of 1912ndash1980 and 1981ndash2016Therefore it can be said that the rate of temperature risinghas increased further as much as the time is closer to the endyears of studied period Also these increment values indicatean increase of 02∘C and 03∘C at the 25th percentile 05∘Cand 08∘C at the 75th percentile and 06∘C and 11∘C at the95th percentile thresholds for thewhole studied subperiods of1912ndash1964 1965ndash2016 1912ndash1980 and 1981ndash2016 respectivelyThis situation shows that the increment of higher values inthe daily average temperatures is greater than the incrementof lower values and also this increment ismore evident for theperiods of 1912ndash1980 and 1981ndash2016 compared to the periodsof 1912ndash1964 and 1965ndash2016

Comparison of the daily maximum temperature seriesfor two periods of 1912ndash1964 and 1965ndash2016 revealed anincrement of 08∘C which can be divided into an average

Advances in Meteorology 9

1912

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

Annual total

Years

Annual maximum daily precip

Annual heavy precipitation STD

R2 = 00549

y = 12491x minus 16171

R2 = 00103

y = 0087x minus 1104

u(t)u(t)

u㰀(t)

u㰀(t)

R2 = 2E minus 08y = minus1E minus 05x + 92115

R2 = 00091

y = 00082x minus 99366

u(t)

u㰀(t)

u(t)

u㰀(t)

(a1)

(a2)

(b1)

(b2)

(c1)

(c2)

(d1)

(d2)

400500600700800900

1000110012001300

Am

ount

of p

reci

p (m

m)

Am

ount

of p

reci

p (m

m)

20

40

60

80

100

120

140

160

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

minus3

minus2

minus1

0

1

2

3

4

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus3

minus2

minus1

0

1

2

3

2016

1912

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1928

Years

2468

101214161820

Am

ount

of p

reci

p (m

m)

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

0

2

4

6

8

10

12

14

Num

ber o

f day

s (gt25

mm

)

minus3

minus2

minus1

0

1

2

3

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus3

minus2

minus1

0

1

2

3

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

2016

1912

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1928

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

Figure 5 Trend analysis of the precipitation time series using least squares linear regression and MK trend test for total annual precipitation(a1 a2) daily maximum precipitation (b1 b2) daily maximum precipitation greater than 25mm (c1 c2) and standard deviation in the dailyprecipitation series (d1 d2)

10 Advances in Meteorology

Table2Statisticalanalyses

outputso

ftem

perature

andprecipitatio

nparameters(SD

stand

arddeviation

Inc

increasedecdecrease)

Parameters

Minim

umtemperature

(∘ C)

Averagetem

perature

(∘ C)

Maxim

umtemperature

(∘ C)

Precipitatio

n(m

m)

Statistics

Perio

ds

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Incor

dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

Average

total

102

107

052

102

109

071

136

140

037

136

141

043

180

187

075

181

187

060

8113

8656

544

8267

8591

324

SD66

66

004

65

67

022

7172

007

7173

019

82

84

021

82

85

025

9391minus03

9193

02

Minim

umminus54minus42

116minus53minus39

138minus27minus21

062minus27minus19

074minus08minus01

076minus07

00

068

00

00

00

00

00

00

Maxim

um223

229

069

222

232

099

276

278

023

277

277

000

356

362

059

358

361

029

589

620

31

579

651

71Percentile5

minus04

02

061minus03

03

055

21

25

049

22

24

026

45

50

052

46

48

015

01

01minus01

01

01

00

Percentile2

547

52

045

48

52

038

7780

024

7778

01 0

114

119

054

115

116

011

07

05minus03

06

Percentile5

0106

108

019

105

106

007

140

141

013

140

138minus020

186

190

040

187

185minus024

26

21minus05

23

24

01

Percentile7

5159

165

066

158

164

059

199

204

054

199

201

022

250

261

105

252

255

034

7672minus04

7374

01

Percentile9

5196

203

074

195

202

074

236

242

065

236

239

032

296

307

110

297

300

027

229

226minus03

227

223minus05

gt30

Cggt80

mm

00

00

00

00

00

00

1626

1017

2810

00

00

00

gt25

Cggt40

mm

00

00

00

712

57

147

90107

1794

108

142

20

22

0lt0C

gmdash

2316minus6

2216minus6

76minus2

76minus1

21minus1

21minus1

mdashmdash

mdashmdash

mdashmdash

ltminus5C

gmdash

21minus1

21minus1

00

00

00

00

00

00

mdashmdash

mdashmdash

mdashmdash

Winter

33

36

032

33

38

048

59

62

031

60

62

020

9095

047

9195

035

1033

1096

63

1056

1066

09

Sprin

g74

87

131

7582

073

111

124

121

113

119

060

160

171

116

162

170

076

496

513

17515

511minus04

Summer

174

187

124

174

186

125

216

224

082

216

225

091

269

273

043

271

282

109

334

331minus03

330

368

38

Autumn

124

127

028

123

127

036

156

156

000

156

155minus001

199

197minus020

200

201

019

842

831minus10

854

919

65

Ann

ual

101

109

079

101

108

070

136

141

057

136

140

043

179

184

042

181

187

060

676

719

43

689

716

27

Advances in Meteorology 11

value of 18∘C for the period of 1912ndash1964 and an averagevalue of 187∘C for the period of 1965ndash2016 On the otherhand comparison of the daily maximum temperature seriesfor the periods of 1912ndash1980 and 1981ndash2016 has revealed anincrement of 06∘C with an average value between 182∘C and188∘C for both of these periods respectively In additionthe standard deviation values are increased between 02∘Cand 03∘C during the whole studied periodThe increment inthe minimum values of daily maximum temperature seriesis more evident than the maximum values The minimumvalues of daily maximum temperatures have exhibited anincrement of 08∘C for the period of 1965ndash2016 compared tothe previous period of 1912ndash1964 and also an increment of07∘C for the period of 1981ndash2016 compared to the previousperiod of 1912ndash1980 Also themaximumvalues of dailymaxi-mum temperature series exhibited amean increment of 05∘Cfor both studied periods Seasonal analysis of maximumtemperature series for both of studied section periods showedthat the highest rising temperature has been happened bya value of 12∘C in summer On the other hand the lowestincrement has happened in the autumn season The monthlyanalysis of daily maximum temperature series showed thatthe highest increment took place during the last monthsof the spring season and the first month of the summerseason in all of the studied periods When the percentiles ofdaily maximum temperatures are analyzed the temperatureincrement based on the 5th percentile threshold is 05∘C forthe periods of 1912ndash1964 and 1965ndash2016 while the incrementis 03∘C for the periods of 1912ndash1980 and 1981ndash2016 This canbe considered as an important sign of rising temperature overthe time In this case the value of percentile thresholds isincreased with extending the length of the time period withextending the length of the first section of studied period infavor of last years than the previous ones In this regard thesevalues are 05∘C and 04∘C at the 25th percentile 1∘C and11∘C at the 75th percentile and 11∘C and 12∘C at the 95thpercentile for both sections of studied periods respectivelyThis situation shows that the increment of higher values inthe dailymaximum temperatures is greater than lower values

The comparison analysis of daily minimum temperaturesbetween the periods of 1912ndash1964 and 1965ndash2016 and theperiods of 1912ndash1980 and 1981ndash2016 showed that there is ageneral increment of 05∘C during the first section periodswhich can be given as 102∘C and 107∘C for the individualperiods of 1912ndash1964 and 1965ndash2016 and also a generalincrement of 08∘C during the second section periods whichcan be provided as 102∘C and 11∘C for the individual periodsof 1912ndash1980 and 1981ndash2016 respectively In addition thestandard deviation values increased among these sectionperiods from 0∘C to 03∘C The increment in the minimumvalues of daily minimum temperature series is more evidentthan the maximum values The minimum values of dailyminimum temperature series showed an increment of 12∘Cfrom the period of 1912ndash1964 to 1965ndash2016 and also anincrement of 16∘C from the period of 1912ndash1980 to 1981ndash2016The maximum values of daily minimum temperature serieshave shown an increment of 07∘C for the section periodsof 1912ndash1964 and 1965ndash2016 and an increment of 12∘C forthe section periods of 1912ndash1980 and 1981ndash2016 Overall

the minimum values have had an average increment of14∘C while the maximum values have had an averageincrement of 1∘C during the last century which can showthe higher rate of upward trends in the temperature timeseries Monthly analysis of minimum temperature seriesshowed that the highest increment for the section studiedperiods of 1912ndash1964 and 1965ndash2016 has occurred by a valueof 1∘C in June while the highest increment for the sectionstudied periods of 1912ndash1980 and 1981ndash2016 has occurred bya value of 16∘C in August Meanwhile there is a temperaturedecrement in November in both of these periods Howeverthere is a clear decrement in the first half of these periodsin October and December whereas there is an incrementin the second half of these periods The seasonal analysisof the daily minimum temperature series showed that thehighest increment has happened in the summer season witha mean value of 07∘C for the section periods of 1912ndash1964and 1965ndash2016 and with a mean value of 13∘C for the sectionperiods of 1912ndash1980 and 1981ndash2016 respectively Then theincreasing rate during the summer season became moreevident during the recent decades Analysis of the percentilethresholds of daily minimum temperature series showed thatthe temperature increment at the 5th percentile is 06∘C forall studied time periods of 1912ndash1964 1965ndash2016 1912ndash1980and 1981ndash2016 Also these increment values indicate anincrease of 04∘C and 06∘C at the 25th percentile 07∘C and11∘C at the 75th percentile and 07∘C and 14∘C at the 95thpercentile thresholds for the whole studied subperiods of1912ndash1964 1965ndash2016 1912ndash1980 and 1981ndash2016 respectivelyThese rising rates in minimum temperature series are moreevident for the periods of 1912ndash1980 and 1981ndash2016 than theperiods of 1912ndash1980 and 1981ndash2016 Therefore it can be saidthat the rate of temperature rising has increased further aslong as the studied time period is closer to the last years

42 Precipitation Annual average precipitation in Istanbulis 838mm with a range of minimum value of 449mm in1921 and a maximum value of 1289mm in 1981 based on theobservatory data of Kandilli station during the whole studiedperiod from 1912 to 2016 (Figure 4(a)) Also monthly averageprecipitation is 699mm with a range of the minimum of326mm in July and the maximum of 1286mm in Decemberduring the whole studied time period (Figure 4(b))

Analysis of the trend in the annual average precipitationtime series by the methods of linear regression analysisand MK trend test has shown that periodically there arepartial increments and significant differences during the totalstudied period from 1912 to 2016 (Figures 5(a1) and 5(a2))But this increment in the precipitation time series is notas clear as the increment in the temperature time seriesHowever it is obvious in the precipitation time series thatthere is an increment between the years of 1917 and 1925a stable condition between the years of 1925 and 1954 anincrement between the years of 1954 and 1965 a decrementbetween the years of 1965 and 1974 an increment between theyears of 1974 and 2001 and again a no change situation from2001 till the end The trend analysis of the daily maximumprecipitation series showed an increment of 29mm for theperiods of 1912ndash1964 and 1965ndash2016 as well as an increment

12 Advances in Meteorology

of 93mm for the periods of 1912ndash1980 and 1981ndash2016 Thesevalues indicated that rainfall has increased at a rate of morethan three times over the last decades than the previous ones(Figures 5(b1) and 5(b2)) Also the number of days with dailyprecipitation greater than 25mm presented an incrementduring the period of 1912ndash2016 although this is not significantat the confidence level of 005 (Figures 5(c1) and 5(c2))The trend of standard deviation in the daily precipitationtime series showed a slowly increasing trend during the totalstudied period (Figures 5(d1) and 5(d2))

The comparison of the results of statistical analysisbetween the daily average rainfall amounts belonging tothe periods of 1912ndash1964 and 1965ndash2016 with those of theperiods of 1912ndash1980 and 1981ndash2016 revealed that there is anincrement of 78mm from the period of 1912ndash1964 to theperiod of 1965ndash2016 and an increment of 38mm from theperiod of 1912ndash1980 to the period of 1981ndash2016 In additionthe analysis of standard deviation exhibited a decrementfrom the period of 1912ndash1964 to the period of 1965ndash2016 andan increment from the period of 1912ndash1980 to the periodof 1981ndash2016 (Table 2) The analysis of monthly averageprecipitation time series showed that the highest incrementhas happened in October with 227mm for the sectionperiods of 1912ndash1964 and 1965ndash2016 and with 347mm forthe section periods of 1912ndash1980 and 1981ndash2016 On the otherhand the highest decrease took placewith a value ofminus04mmduring both May and July for the periods of 1912ndash1964 and1965ndash2016 and with a value of minus16mm in September for thesection periods of 1912ndash1980 and 1981ndash2016 The analysis ofseasonal average precipitation time series showed that thehighest increase occurred in autumn with a value of 72mmfor the section periods of 1912ndash1964 and 1965ndash2016 as wellas with a value of 64mm for the periods of 1912ndash1980 and1981ndash2016 respectively There is no remarkable decreasingchange seasonally except in winter season during the periodsof 1912ndash1980 and 1981ndash2016 Furthermore the percentilethresholds of daily average precipitation indicated that thereis an insignificant negative trend based on all percentilesand for all studied periods Overall the statistical analysisshowed that the total average precipitation of Istanbul hasincreased while this increasing trend is more pronounced inthe previous decades than the last 3 decades On the otherhand the increasing rate of daily maximum precipitation ismore evident during the last 3 decades than the previousdecades which can be proven by the increasing frequency ofheavy rainfall events in Istanbul

Generally the results of trend analysis of Kandilli stationduring the last 105 years of 1912ndash2016 showed that there isa warming significant trend in the precipitation time seriesby using both methods of linear regression analysis andMK trend test On the contrary previous climate studiesconducted over Turkey put forward that there has been adecreasing trend in annual precipitation time series duringthe recent decades regionally The results of a previouslyconducted study by using the daily precipitation and tem-perature data sets of Florya and Goztepe meteorologicalstations in Istanbul area between 1960 and 2013 showedthat most notably the precipitation during the warm periodshas decreased but the frequency of the intense rain has

increased and the majority of these episodes of intense raincoincided with the warm periods Other determinationswere the rise in the annual average temperature and theextension of the warm periods in a year This differentiationof the temperature features can lead to the aggravation of theevaporation and it can be effective for a longer period duringthe year [15] Thus it will make Istanbul be confronted withthe much more important problems of water managementand flood [34] Also the results of the current study for trendanalysis in the long period from 1912 to 2016 showed that themost striking spell is between the years of 1968 and 1998 dueto the existence of least number of rainfall events in IstanbulIt can be owing to industrialization along with the increasingair pollution as well as irregular urbanization in Istanbul areaIn this case severe droughts taking place during the yearsof 1988 1992 1993 and 2008 have threatened the reservoirswhich supplied fresh water of the city These years werecharacterized by not having enough rainfall events Theseyears are also characterized by more persistent high-pressuresystems and less occurrence of low-pressure systems in termsof number and strength As statements made by officialinstitutions the formation conditions for the atmosphericlayer of air pollution due to air pollutant emissions from fossilfuel combustion and industrial activities are more providedduring anticyclone or high-pressure system eventsThereforethis leads to warming up and generating an inversion layerin the boundary layer of atmosphere especially over city area[35] The inversion layer or urban heat island intensity isincreasing with the increasing city size andor populationa phenomenon that was also reported by others [36ndash38]Moreover these last climatic events also have affected somepolitical results beyond natural effects The most obviousexample of this is related to the local election of 1994 inIstanbul In this case extreme drought during the summers of1992 and 1993 has caused groundwater reservoirs to dry up inthe city discontinuance of water was experienced for severaldays or even weeks In those times the mayor of Istanbul cityhas lost the election of 1994 and he realized that this result wasdue to the peoplersquos reaction about the water shortages Manypolitical reports of that period also support this scientificview

5 Conclusions

Statistical analysis in temperature time series of the Kandillistation from 1912 to 2016 established that there is a notableincrease in temperature values after the 1940s which is inparallel with the beginning of industrialization era in Istan-bul There has been a rise about 094∘C in the daily averagetemperature series since the beginning of the last century Asignificant positive trend in the daily maximum temperatureseries is found about 156∘C Also there is a positive trendabout 087∘C in the daily minimum temperature series Onthe other hand analysis of the number of days with thedaily maximum temperature higher than 30∘C showed thatthere is an increasing trend Meanwhile analysis of thenumber of days with daily minimum temperatures lowerthan 0∘C showed a decreasing trend The increment in theminimum values of the daily minimum temperature series

Advances in Meteorology 13

is more evident than the maximum values of this series Inthis case these rising rates in the minimum temperatureseries are more evident for the section periods of 1912ndash1980and 1981ndash2016 than the section periods of 1912ndash1964 and1965ndash2016 This again shows that there is an increment inthe positive temperature trend from past to present decadesThe increment in the precipitation time series is not asclear as the increment in the temperature time series dueto periodic variability The trend analysis in the total annualprecipitation time series showed that the first significantupward trend has periodically been started from the 1920swhile there is a stable trend from 2001 till 2016 The dailyaverage of rainfall amounts has increased with a value of58mm during the period of 1912ndash2016 Also the analysisof heavy precipitation trend showed an increase of 61mmOverall the total average precipitation of Istanbul increasedwhile this increasing trend is more pronounced during theearly decades than the last 3 decades On the other handthe increasing rate of daily maximum precipitation is morepronounced in the last 3 decades than the previous decadesThen it was shown that the frequency of heavy rainfallat Istanbul has increased during the recent decades Thusthe precipitation changes in Istanbul have some differencescompared to the general tendency in precipitation trendthat was put forward by other studies as a decreasing trendover the whole of Turkey This result can be expressedas a positive effect of population overgrowth of Istanbulmegacity Comparison of the results in the first half of thestudy period (1912ndash1964) with the second half of the studyperiod (1965ndash2016) showed that both the average temperatureand average precipitation have higher values of 139∘C and878mm for the final phase compared to the values of 136∘Cand 799mm belonging to the initial phase Therefore it canbe stated that the megacity of Istanbul is directly affectedby the climate change and its consequences In this contextpotential risks of climate change in Istanbul megacity underhigher temperature conditions can be expressed as the rise inthe sea level increase in the rate of evapotranspiration andincrease in the frequency of heavy rainfall Also this city maynot be able to handle this uncontrolled population growthand its associated irreversible changes which is alreadypushing the natural limits by destroying the environmentTherefore the local governors of any megacity like Istanbulshould give more emphasis on the importance of sustainableurban development Thus it is urgent to prepare local andnational climate change strategies and action plans for themegacities

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

The authors are grateful to the Bogazici University theobservatory of Kandilli weather station and the Earthquake

Research Institute for providing the research data and tech-nical support The authors also gratefully acknowledge con-tributions of Assoc Professor Dr Yuksel Demirkaya Schoolof Social Sciences Marmara University This work has beensupported by Scientific and Technological Research Councilof Turkey (TUBITAK) under Grants 113R019 and 106Y258and by Marmara University (BAPKO) with projects FEN-E-120314-0066 FEN-C-YLP-090414-0102 FEN-L-250416-0180 and FEN-A-100413-0127

References

[1] D R Easterling B Horton P D Jones et al ldquoMaximum andminimum temperature trends for the globerdquo Science vol 277no 5324 pp 364ndash367 1997

[2] IPCC Climate Change 2014 IPCC Fifth Assessment Synthe-sis Report-Summary for Policymakers-an Assessment of Inter-Governmental Panel on Climate Change Cambridge UniversityPress Cambridge UK 2014

[3] J Carmin N Nadkarni and C Rhie Progress and Challenges inUrban Climate Adaptation Planning Results of a Global SurveyMIT Cambridge UK 2012

[4] A F Young ldquoUrban expansion and environmental risk in theSao Paulo Metropolitan Areardquo Climate Research vol 57 no 1pp 73ndash80 2013

[5] P Tian X Mu J Liu J Hu and C Gu ldquoImpacts of ClimateVariability and Human Activities on the Changes of Runoff andSediment Load in a Catchment of the Loess Plateau ChinardquoAdvances inMeteorology vol 2016 Article ID 4724067 15 pages2016

[6] R S Kovats and K L Ebi ldquoHeatwaves and public health inEuroperdquo European Journal of Public Health vol 16 no 6 pp592ndash599 2006

[7] S Conti P Meli G Minelli et al ldquoEpidemiologic studyof mortality during the Summer 2003 heat wave in ItalyrdquoEnvironmental Research vol 98 no 3 pp 390ndash399 2005

[8] J Kysely and J Kim ldquoMortality during heat waves in SouthKorea 1991 to 2005 how exceptional was the 1994 heat waverdquoClimate Research vol 38 no 2 pp 105ndash116 2009

[9] B Yan Z Xia F Huang L Guo and X Zhang ldquoClimatechange detection and annual extreme temperature analysis ofthe amur river basinrdquoAdvances inMeteorology vol 2016 ArticleID 6268938 14 pages 2016

[10] E M Fischer and R Knutti ldquoAnthropogenic contribution toglobal occurrence of heavy-precipitation and high-temperatureextremesrdquo Nature Climate Change vol 5 no 6 pp 560ndash5642015

[11] X Zhang L Alexander G C Hegerl et al ldquoIndices for moni-toring changes in extremes based on daily temperature andprecipitation datardquo Climate Change vol 2 no 6 pp 851ndash8702011

[12] K H Schlunzen P Hoffmann G Rosenhagen and W RieckeldquoLong-term changes and regional differences in temperatureand precipitation in the metropolitan area of Hamburgrdquo Inter-national Journal of Climatology vol 30 no 8 pp 1121ndash11362010

[13] G Bartolini M Morabito A Crisci et al ldquoRecent trends inTuscany (Italy) summer temperature and indices of extremesrdquoInternational Journal of Climatology vol 28 no 13 pp 1751ndash1760 2008

14 Advances in Meteorology

[14] S C Sheridan and T J Dolney ldquoHeat mortality and levelof urbanization measuring vulnerability across Ohio USArdquoClimate Research vol 24 no 3 pp 255ndash265 2003

[15] M Tayanc U Im M Dogruel andM Karaca ldquoClimate changein Turkey for the last half centuryrdquo Climatic Change vol 94 no3-4 pp 483ndash502 2009

[16] H Toros ldquoSpatio-temporal variation of daily extreme tempera-tures over Turkeyrdquo International Journal of Climatology vol 32no 7 pp 1047ndash1055 2012

[17] H Toros ldquoSpatio-temporal precipitation change assessmentsover Turkeyrdquo International Journal of Climatology vol 32 no9 pp 1310ndash1325 2012

[18] M Turkes C Yozgatlıgil I Batmaz et al ldquoHas the climate beenchanging in Turkey Regional climate change signals based on acomparative statistical analysis of two consecutive time periods1950-1980 and 1981-2010rdquoClimate Research vol 70 no 1 pp 77ndash93 2016

[19] IPCC ldquoummary for Policymakers In Climate Change 2013rdquo inThe Physical Science Basis The contribution of Working Group Ito the Fifth Assessment Report of the Intergovernmental Panel onClimate Change Cambridge University Press Cambridge UK2013

[20] United Nations ldquoDepartment of Economic and Social AffairsPopulation Division 2006 World Urbanization Prospects The2005 Revisionrdquo Working Paper ESAPWP200 2011

[21] Y Demirkaya Sayılarla Istanbul ITO Istanbul Turkey 2011[22] TUIK ldquoAddress based population registration system results of

2014rdquo Turkish Statistical Institute (TUIK) 2017 httpraporytuikgovtr10-03-2015-184727-842632346446643191142126876html

[23] S Erinc Climatology and its Methods Alfa Basım YayımDagitim Istanbul Turkey 4th edition 1965

[24] O M Gokturk D Bozkurt O L Sen and M Karaca ldquoQualitycontrol and homogeneity of Turkish precipitation datardquoHydro-logical Processes vol 22 no 16 pp 3210ndash3218 2008

[25] C Ley and D Paindaveine ldquoRuns Testsrdquo in Encyclopedia ofEnvironmetrics 2012

[26] A Ghasemi and S Zahediasl ldquoNormality tests for statisticalanalysis a guide for non-statisticiansrdquo International Journal ofEndocrinology andMetabolism vol 10 no 2 pp 486ndash489 2012

[27] G V Glass ldquoTesting Homogeneity of Variancesrdquo AmericanEducational Research Journal vol 3 no 3 pp 187ndash190 1966

[28] H BMann ldquoNonparametric tests against trendrdquo Econometricavol 13 pp 245ndash259 1945

[29] M G Kendall Rank Correlation Method Charles GriffinLondon UK 4th edition 1975

[30] H Turoglu ldquoDetection of Changes on Temperature and Precip-itation Features in Istanbul (Turkey)rdquo Atmospheric and ClimateSciences vol 04 no 04 pp 549ndash562 2014

[31] A Karaburun A Demirci and I-S Suen ldquoImpacts of urbangrowth on forest cover in Istanbul (1987-2007)rdquo EnvironmentalModeling amp Assessment vol 166 no 1-4 pp 267ndash277 2010

[32] K K Karanth LM Curran and J D Reuning-Scherer ldquoVillagesize and forest disturbance in Bhadra Wildlife SanctuaryWestern Ghats Indiardquo Biological Conservation vol 128 no 2pp 147ndash157 2006

[33] G Cakir C Un E Z Baskent S Kose F Sivrikaya andS Keles ldquoEvaluating urbanization fragmentation and landuseland cover change pattern in Istanbul city Turkey from 1971to 2002rdquo Land Degradation amp Development vol 19 no 6 pp663ndash675 2008

[34] R B Myneni F G Hall P J Sellers and A L Marshak ldquoTheinterpretation of spectral vegetation indexesrdquo IEEE Transac-tions on Geoscience and Remote Sensing vol 33 no 2 pp 481ndash486 1995

[35] Y S Unal H Toros A Deniz and S Incecik ldquoInfluence ofmeteorological factors and emission sources on spatial and tem-poral variations of PM10 concentrations in Istanbul metropoli-tan areardquo Atmospheric Environment vol 45 no 31 pp 5504ndash5513 2011

[36] M KaracaM Tayanc andH Toros ldquoEffects of urbanization onclimate of Istanbul and Ankarardquo Atmospheric Environment vol29 no 23 pp 3411ndash3421 1995

[37] Y Ezber O L Sen T Kindap and M Karaca ldquoClimatic effectsof urbanization in Istanbul a statistical and modeling analysisrdquoInternational Journal of Climatology vol 27 no 5 pp 667ndash6792007

[38] H S Park ldquoFeatures of the heat island in seoul and its sur-rounding citiesrdquo Atmospheric Environment (1967) vol 20 no10 pp 1859ndash1866 1986

Advances in Meteorology 9

1912

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

Annual total

Years

Annual maximum daily precip

Annual heavy precipitation STD

R2 = 00549

y = 12491x minus 16171

R2 = 00103

y = 0087x minus 1104

u(t)u(t)

u㰀(t)

u㰀(t)

R2 = 2E minus 08y = minus1E minus 05x + 92115

R2 = 00091

y = 00082x minus 99366

u(t)

u㰀(t)

u(t)

u㰀(t)

(a1)

(a2)

(b1)

(b2)

(c1)

(c2)

(d1)

(d2)

400500600700800900

1000110012001300

Am

ount

of p

reci

p (m

m)

Am

ount

of p

reci

p (m

m)

20

40

60

80

100

120

140

160

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

minus3

minus2

minus1

0

1

2

3

4

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus3

minus2

minus1

0

1

2

3

2016

1912

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1928

Years

2468

101214161820

Am

ount

of p

reci

p (m

m)

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

0

2

4

6

8

10

12

14

Num

ber o

f day

s (gt25

mm

)

minus3

minus2

minus1

0

1

2

3

Man

n-Ke

ndal

l sta

tistic

s

Man

n-Ke

ndal

l sta

tistic

s

minus3

minus2

minus1

0

1

2

3

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

2016

1912

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

1920

1928

Years

1920

1928

1936

1944

1952

1960

1968

1976

1984

1992

2000

2008

2016

1912

Years

Figure 5 Trend analysis of the precipitation time series using least squares linear regression and MK trend test for total annual precipitation(a1 a2) daily maximum precipitation (b1 b2) daily maximum precipitation greater than 25mm (c1 c2) and standard deviation in the dailyprecipitation series (d1 d2)

10 Advances in Meteorology

Table2Statisticalanalyses

outputso

ftem

perature

andprecipitatio

nparameters(SD

stand

arddeviation

Inc

increasedecdecrease)

Parameters

Minim

umtemperature

(∘ C)

Averagetem

perature

(∘ C)

Maxim

umtemperature

(∘ C)

Precipitatio

n(m

m)

Statistics

Perio

ds

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Incor

dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

Average

total

102

107

052

102

109

071

136

140

037

136

141

043

180

187

075

181

187

060

8113

8656

544

8267

8591

324

SD66

66

004

65

67

022

7172

007

7173

019

82

84

021

82

85

025

9391minus03

9193

02

Minim

umminus54minus42

116minus53minus39

138minus27minus21

062minus27minus19

074minus08minus01

076minus07

00

068

00

00

00

00

00

00

Maxim

um223

229

069

222

232

099

276

278

023

277

277

000

356

362

059

358

361

029

589

620

31

579

651

71Percentile5

minus04

02

061minus03

03

055

21

25

049

22

24

026

45

50

052

46

48

015

01

01minus01

01

01

00

Percentile2

547

52

045

48

52

038

7780

024

7778

01 0

114

119

054

115

116

011

07

05minus03

06

Percentile5

0106

108

019

105

106

007

140

141

013

140

138minus020

186

190

040

187

185minus024

26

21minus05

23

24

01

Percentile7

5159

165

066

158

164

059

199

204

054

199

201

022

250

261

105

252

255

034

7672minus04

7374

01

Percentile9

5196

203

074

195

202

074

236

242

065

236

239

032

296

307

110

297

300

027

229

226minus03

227

223minus05

gt30

Cggt80

mm

00

00

00

00

00

00

1626

1017

2810

00

00

00

gt25

Cggt40

mm

00

00

00

712

57

147

90107

1794

108

142

20

22

0lt0C

gmdash

2316minus6

2216minus6

76minus2

76minus1

21minus1

21minus1

mdashmdash

mdashmdash

mdashmdash

ltminus5C

gmdash

21minus1

21minus1

00

00

00

00

00

00

mdashmdash

mdashmdash

mdashmdash

Winter

33

36

032

33

38

048

59

62

031

60

62

020

9095

047

9195

035

1033

1096

63

1056

1066

09

Sprin

g74

87

131

7582

073

111

124

121

113

119

060

160

171

116

162

170

076

496

513

17515

511minus04

Summer

174

187

124

174

186

125

216

224

082

216

225

091

269

273

043

271

282

109

334

331minus03

330

368

38

Autumn

124

127

028

123

127

036

156

156

000

156

155minus001

199

197minus020

200

201

019

842

831minus10

854

919

65

Ann

ual

101

109

079

101

108

070

136

141

057

136

140

043

179

184

042

181

187

060

676

719

43

689

716

27

Advances in Meteorology 11

value of 18∘C for the period of 1912ndash1964 and an averagevalue of 187∘C for the period of 1965ndash2016 On the otherhand comparison of the daily maximum temperature seriesfor the periods of 1912ndash1980 and 1981ndash2016 has revealed anincrement of 06∘C with an average value between 182∘C and188∘C for both of these periods respectively In additionthe standard deviation values are increased between 02∘Cand 03∘C during the whole studied periodThe increment inthe minimum values of daily maximum temperature seriesis more evident than the maximum values The minimumvalues of daily maximum temperatures have exhibited anincrement of 08∘C for the period of 1965ndash2016 compared tothe previous period of 1912ndash1964 and also an increment of07∘C for the period of 1981ndash2016 compared to the previousperiod of 1912ndash1980 Also themaximumvalues of dailymaxi-mum temperature series exhibited amean increment of 05∘Cfor both studied periods Seasonal analysis of maximumtemperature series for both of studied section periods showedthat the highest rising temperature has been happened bya value of 12∘C in summer On the other hand the lowestincrement has happened in the autumn season The monthlyanalysis of daily maximum temperature series showed thatthe highest increment took place during the last monthsof the spring season and the first month of the summerseason in all of the studied periods When the percentiles ofdaily maximum temperatures are analyzed the temperatureincrement based on the 5th percentile threshold is 05∘C forthe periods of 1912ndash1964 and 1965ndash2016 while the incrementis 03∘C for the periods of 1912ndash1980 and 1981ndash2016 This canbe considered as an important sign of rising temperature overthe time In this case the value of percentile thresholds isincreased with extending the length of the time period withextending the length of the first section of studied period infavor of last years than the previous ones In this regard thesevalues are 05∘C and 04∘C at the 25th percentile 1∘C and11∘C at the 75th percentile and 11∘C and 12∘C at the 95thpercentile for both sections of studied periods respectivelyThis situation shows that the increment of higher values inthe dailymaximum temperatures is greater than lower values

The comparison analysis of daily minimum temperaturesbetween the periods of 1912ndash1964 and 1965ndash2016 and theperiods of 1912ndash1980 and 1981ndash2016 showed that there is ageneral increment of 05∘C during the first section periodswhich can be given as 102∘C and 107∘C for the individualperiods of 1912ndash1964 and 1965ndash2016 and also a generalincrement of 08∘C during the second section periods whichcan be provided as 102∘C and 11∘C for the individual periodsof 1912ndash1980 and 1981ndash2016 respectively In addition thestandard deviation values increased among these sectionperiods from 0∘C to 03∘C The increment in the minimumvalues of daily minimum temperature series is more evidentthan the maximum values The minimum values of dailyminimum temperature series showed an increment of 12∘Cfrom the period of 1912ndash1964 to 1965ndash2016 and also anincrement of 16∘C from the period of 1912ndash1980 to 1981ndash2016The maximum values of daily minimum temperature serieshave shown an increment of 07∘C for the section periodsof 1912ndash1964 and 1965ndash2016 and an increment of 12∘C forthe section periods of 1912ndash1980 and 1981ndash2016 Overall

the minimum values have had an average increment of14∘C while the maximum values have had an averageincrement of 1∘C during the last century which can showthe higher rate of upward trends in the temperature timeseries Monthly analysis of minimum temperature seriesshowed that the highest increment for the section studiedperiods of 1912ndash1964 and 1965ndash2016 has occurred by a valueof 1∘C in June while the highest increment for the sectionstudied periods of 1912ndash1980 and 1981ndash2016 has occurred bya value of 16∘C in August Meanwhile there is a temperaturedecrement in November in both of these periods Howeverthere is a clear decrement in the first half of these periodsin October and December whereas there is an incrementin the second half of these periods The seasonal analysisof the daily minimum temperature series showed that thehighest increment has happened in the summer season witha mean value of 07∘C for the section periods of 1912ndash1964and 1965ndash2016 and with a mean value of 13∘C for the sectionperiods of 1912ndash1980 and 1981ndash2016 respectively Then theincreasing rate during the summer season became moreevident during the recent decades Analysis of the percentilethresholds of daily minimum temperature series showed thatthe temperature increment at the 5th percentile is 06∘C forall studied time periods of 1912ndash1964 1965ndash2016 1912ndash1980and 1981ndash2016 Also these increment values indicate anincrease of 04∘C and 06∘C at the 25th percentile 07∘C and11∘C at the 75th percentile and 07∘C and 14∘C at the 95thpercentile thresholds for the whole studied subperiods of1912ndash1964 1965ndash2016 1912ndash1980 and 1981ndash2016 respectivelyThese rising rates in minimum temperature series are moreevident for the periods of 1912ndash1980 and 1981ndash2016 than theperiods of 1912ndash1980 and 1981ndash2016 Therefore it can be saidthat the rate of temperature rising has increased further aslong as the studied time period is closer to the last years

42 Precipitation Annual average precipitation in Istanbulis 838mm with a range of minimum value of 449mm in1921 and a maximum value of 1289mm in 1981 based on theobservatory data of Kandilli station during the whole studiedperiod from 1912 to 2016 (Figure 4(a)) Also monthly averageprecipitation is 699mm with a range of the minimum of326mm in July and the maximum of 1286mm in Decemberduring the whole studied time period (Figure 4(b))

Analysis of the trend in the annual average precipitationtime series by the methods of linear regression analysisand MK trend test has shown that periodically there arepartial increments and significant differences during the totalstudied period from 1912 to 2016 (Figures 5(a1) and 5(a2))But this increment in the precipitation time series is notas clear as the increment in the temperature time seriesHowever it is obvious in the precipitation time series thatthere is an increment between the years of 1917 and 1925a stable condition between the years of 1925 and 1954 anincrement between the years of 1954 and 1965 a decrementbetween the years of 1965 and 1974 an increment between theyears of 1974 and 2001 and again a no change situation from2001 till the end The trend analysis of the daily maximumprecipitation series showed an increment of 29mm for theperiods of 1912ndash1964 and 1965ndash2016 as well as an increment

12 Advances in Meteorology

of 93mm for the periods of 1912ndash1980 and 1981ndash2016 Thesevalues indicated that rainfall has increased at a rate of morethan three times over the last decades than the previous ones(Figures 5(b1) and 5(b2)) Also the number of days with dailyprecipitation greater than 25mm presented an incrementduring the period of 1912ndash2016 although this is not significantat the confidence level of 005 (Figures 5(c1) and 5(c2))The trend of standard deviation in the daily precipitationtime series showed a slowly increasing trend during the totalstudied period (Figures 5(d1) and 5(d2))

The comparison of the results of statistical analysisbetween the daily average rainfall amounts belonging tothe periods of 1912ndash1964 and 1965ndash2016 with those of theperiods of 1912ndash1980 and 1981ndash2016 revealed that there is anincrement of 78mm from the period of 1912ndash1964 to theperiod of 1965ndash2016 and an increment of 38mm from theperiod of 1912ndash1980 to the period of 1981ndash2016 In additionthe analysis of standard deviation exhibited a decrementfrom the period of 1912ndash1964 to the period of 1965ndash2016 andan increment from the period of 1912ndash1980 to the periodof 1981ndash2016 (Table 2) The analysis of monthly averageprecipitation time series showed that the highest incrementhas happened in October with 227mm for the sectionperiods of 1912ndash1964 and 1965ndash2016 and with 347mm forthe section periods of 1912ndash1980 and 1981ndash2016 On the otherhand the highest decrease took placewith a value ofminus04mmduring both May and July for the periods of 1912ndash1964 and1965ndash2016 and with a value of minus16mm in September for thesection periods of 1912ndash1980 and 1981ndash2016 The analysis ofseasonal average precipitation time series showed that thehighest increase occurred in autumn with a value of 72mmfor the section periods of 1912ndash1964 and 1965ndash2016 as wellas with a value of 64mm for the periods of 1912ndash1980 and1981ndash2016 respectively There is no remarkable decreasingchange seasonally except in winter season during the periodsof 1912ndash1980 and 1981ndash2016 Furthermore the percentilethresholds of daily average precipitation indicated that thereis an insignificant negative trend based on all percentilesand for all studied periods Overall the statistical analysisshowed that the total average precipitation of Istanbul hasincreased while this increasing trend is more pronounced inthe previous decades than the last 3 decades On the otherhand the increasing rate of daily maximum precipitation ismore evident during the last 3 decades than the previousdecades which can be proven by the increasing frequency ofheavy rainfall events in Istanbul

Generally the results of trend analysis of Kandilli stationduring the last 105 years of 1912ndash2016 showed that there isa warming significant trend in the precipitation time seriesby using both methods of linear regression analysis andMK trend test On the contrary previous climate studiesconducted over Turkey put forward that there has been adecreasing trend in annual precipitation time series duringthe recent decades regionally The results of a previouslyconducted study by using the daily precipitation and tem-perature data sets of Florya and Goztepe meteorologicalstations in Istanbul area between 1960 and 2013 showedthat most notably the precipitation during the warm periodshas decreased but the frequency of the intense rain has

increased and the majority of these episodes of intense raincoincided with the warm periods Other determinationswere the rise in the annual average temperature and theextension of the warm periods in a year This differentiationof the temperature features can lead to the aggravation of theevaporation and it can be effective for a longer period duringthe year [15] Thus it will make Istanbul be confronted withthe much more important problems of water managementand flood [34] Also the results of the current study for trendanalysis in the long period from 1912 to 2016 showed that themost striking spell is between the years of 1968 and 1998 dueto the existence of least number of rainfall events in IstanbulIt can be owing to industrialization along with the increasingair pollution as well as irregular urbanization in Istanbul areaIn this case severe droughts taking place during the yearsof 1988 1992 1993 and 2008 have threatened the reservoirswhich supplied fresh water of the city These years werecharacterized by not having enough rainfall events Theseyears are also characterized by more persistent high-pressuresystems and less occurrence of low-pressure systems in termsof number and strength As statements made by officialinstitutions the formation conditions for the atmosphericlayer of air pollution due to air pollutant emissions from fossilfuel combustion and industrial activities are more providedduring anticyclone or high-pressure system eventsThereforethis leads to warming up and generating an inversion layerin the boundary layer of atmosphere especially over city area[35] The inversion layer or urban heat island intensity isincreasing with the increasing city size andor populationa phenomenon that was also reported by others [36ndash38]Moreover these last climatic events also have affected somepolitical results beyond natural effects The most obviousexample of this is related to the local election of 1994 inIstanbul In this case extreme drought during the summers of1992 and 1993 has caused groundwater reservoirs to dry up inthe city discontinuance of water was experienced for severaldays or even weeks In those times the mayor of Istanbul cityhas lost the election of 1994 and he realized that this result wasdue to the peoplersquos reaction about the water shortages Manypolitical reports of that period also support this scientificview

5 Conclusions

Statistical analysis in temperature time series of the Kandillistation from 1912 to 2016 established that there is a notableincrease in temperature values after the 1940s which is inparallel with the beginning of industrialization era in Istan-bul There has been a rise about 094∘C in the daily averagetemperature series since the beginning of the last century Asignificant positive trend in the daily maximum temperatureseries is found about 156∘C Also there is a positive trendabout 087∘C in the daily minimum temperature series Onthe other hand analysis of the number of days with thedaily maximum temperature higher than 30∘C showed thatthere is an increasing trend Meanwhile analysis of thenumber of days with daily minimum temperatures lowerthan 0∘C showed a decreasing trend The increment in theminimum values of the daily minimum temperature series

Advances in Meteorology 13

is more evident than the maximum values of this series Inthis case these rising rates in the minimum temperatureseries are more evident for the section periods of 1912ndash1980and 1981ndash2016 than the section periods of 1912ndash1964 and1965ndash2016 This again shows that there is an increment inthe positive temperature trend from past to present decadesThe increment in the precipitation time series is not asclear as the increment in the temperature time series dueto periodic variability The trend analysis in the total annualprecipitation time series showed that the first significantupward trend has periodically been started from the 1920swhile there is a stable trend from 2001 till 2016 The dailyaverage of rainfall amounts has increased with a value of58mm during the period of 1912ndash2016 Also the analysisof heavy precipitation trend showed an increase of 61mmOverall the total average precipitation of Istanbul increasedwhile this increasing trend is more pronounced during theearly decades than the last 3 decades On the other handthe increasing rate of daily maximum precipitation is morepronounced in the last 3 decades than the previous decadesThen it was shown that the frequency of heavy rainfallat Istanbul has increased during the recent decades Thusthe precipitation changes in Istanbul have some differencescompared to the general tendency in precipitation trendthat was put forward by other studies as a decreasing trendover the whole of Turkey This result can be expressedas a positive effect of population overgrowth of Istanbulmegacity Comparison of the results in the first half of thestudy period (1912ndash1964) with the second half of the studyperiod (1965ndash2016) showed that both the average temperatureand average precipitation have higher values of 139∘C and878mm for the final phase compared to the values of 136∘Cand 799mm belonging to the initial phase Therefore it canbe stated that the megacity of Istanbul is directly affectedby the climate change and its consequences In this contextpotential risks of climate change in Istanbul megacity underhigher temperature conditions can be expressed as the rise inthe sea level increase in the rate of evapotranspiration andincrease in the frequency of heavy rainfall Also this city maynot be able to handle this uncontrolled population growthand its associated irreversible changes which is alreadypushing the natural limits by destroying the environmentTherefore the local governors of any megacity like Istanbulshould give more emphasis on the importance of sustainableurban development Thus it is urgent to prepare local andnational climate change strategies and action plans for themegacities

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

The authors are grateful to the Bogazici University theobservatory of Kandilli weather station and the Earthquake

Research Institute for providing the research data and tech-nical support The authors also gratefully acknowledge con-tributions of Assoc Professor Dr Yuksel Demirkaya Schoolof Social Sciences Marmara University This work has beensupported by Scientific and Technological Research Councilof Turkey (TUBITAK) under Grants 113R019 and 106Y258and by Marmara University (BAPKO) with projects FEN-E-120314-0066 FEN-C-YLP-090414-0102 FEN-L-250416-0180 and FEN-A-100413-0127

References

[1] D R Easterling B Horton P D Jones et al ldquoMaximum andminimum temperature trends for the globerdquo Science vol 277no 5324 pp 364ndash367 1997

[2] IPCC Climate Change 2014 IPCC Fifth Assessment Synthe-sis Report-Summary for Policymakers-an Assessment of Inter-Governmental Panel on Climate Change Cambridge UniversityPress Cambridge UK 2014

[3] J Carmin N Nadkarni and C Rhie Progress and Challenges inUrban Climate Adaptation Planning Results of a Global SurveyMIT Cambridge UK 2012

[4] A F Young ldquoUrban expansion and environmental risk in theSao Paulo Metropolitan Areardquo Climate Research vol 57 no 1pp 73ndash80 2013

[5] P Tian X Mu J Liu J Hu and C Gu ldquoImpacts of ClimateVariability and Human Activities on the Changes of Runoff andSediment Load in a Catchment of the Loess Plateau ChinardquoAdvances inMeteorology vol 2016 Article ID 4724067 15 pages2016

[6] R S Kovats and K L Ebi ldquoHeatwaves and public health inEuroperdquo European Journal of Public Health vol 16 no 6 pp592ndash599 2006

[7] S Conti P Meli G Minelli et al ldquoEpidemiologic studyof mortality during the Summer 2003 heat wave in ItalyrdquoEnvironmental Research vol 98 no 3 pp 390ndash399 2005

[8] J Kysely and J Kim ldquoMortality during heat waves in SouthKorea 1991 to 2005 how exceptional was the 1994 heat waverdquoClimate Research vol 38 no 2 pp 105ndash116 2009

[9] B Yan Z Xia F Huang L Guo and X Zhang ldquoClimatechange detection and annual extreme temperature analysis ofthe amur river basinrdquoAdvances inMeteorology vol 2016 ArticleID 6268938 14 pages 2016

[10] E M Fischer and R Knutti ldquoAnthropogenic contribution toglobal occurrence of heavy-precipitation and high-temperatureextremesrdquo Nature Climate Change vol 5 no 6 pp 560ndash5642015

[11] X Zhang L Alexander G C Hegerl et al ldquoIndices for moni-toring changes in extremes based on daily temperature andprecipitation datardquo Climate Change vol 2 no 6 pp 851ndash8702011

[12] K H Schlunzen P Hoffmann G Rosenhagen and W RieckeldquoLong-term changes and regional differences in temperatureand precipitation in the metropolitan area of Hamburgrdquo Inter-national Journal of Climatology vol 30 no 8 pp 1121ndash11362010

[13] G Bartolini M Morabito A Crisci et al ldquoRecent trends inTuscany (Italy) summer temperature and indices of extremesrdquoInternational Journal of Climatology vol 28 no 13 pp 1751ndash1760 2008

14 Advances in Meteorology

[14] S C Sheridan and T J Dolney ldquoHeat mortality and levelof urbanization measuring vulnerability across Ohio USArdquoClimate Research vol 24 no 3 pp 255ndash265 2003

[15] M Tayanc U Im M Dogruel andM Karaca ldquoClimate changein Turkey for the last half centuryrdquo Climatic Change vol 94 no3-4 pp 483ndash502 2009

[16] H Toros ldquoSpatio-temporal variation of daily extreme tempera-tures over Turkeyrdquo International Journal of Climatology vol 32no 7 pp 1047ndash1055 2012

[17] H Toros ldquoSpatio-temporal precipitation change assessmentsover Turkeyrdquo International Journal of Climatology vol 32 no9 pp 1310ndash1325 2012

[18] M Turkes C Yozgatlıgil I Batmaz et al ldquoHas the climate beenchanging in Turkey Regional climate change signals based on acomparative statistical analysis of two consecutive time periods1950-1980 and 1981-2010rdquoClimate Research vol 70 no 1 pp 77ndash93 2016

[19] IPCC ldquoummary for Policymakers In Climate Change 2013rdquo inThe Physical Science Basis The contribution of Working Group Ito the Fifth Assessment Report of the Intergovernmental Panel onClimate Change Cambridge University Press Cambridge UK2013

[20] United Nations ldquoDepartment of Economic and Social AffairsPopulation Division 2006 World Urbanization Prospects The2005 Revisionrdquo Working Paper ESAPWP200 2011

[21] Y Demirkaya Sayılarla Istanbul ITO Istanbul Turkey 2011[22] TUIK ldquoAddress based population registration system results of

2014rdquo Turkish Statistical Institute (TUIK) 2017 httpraporytuikgovtr10-03-2015-184727-842632346446643191142126876html

[23] S Erinc Climatology and its Methods Alfa Basım YayımDagitim Istanbul Turkey 4th edition 1965

[24] O M Gokturk D Bozkurt O L Sen and M Karaca ldquoQualitycontrol and homogeneity of Turkish precipitation datardquoHydro-logical Processes vol 22 no 16 pp 3210ndash3218 2008

[25] C Ley and D Paindaveine ldquoRuns Testsrdquo in Encyclopedia ofEnvironmetrics 2012

[26] A Ghasemi and S Zahediasl ldquoNormality tests for statisticalanalysis a guide for non-statisticiansrdquo International Journal ofEndocrinology andMetabolism vol 10 no 2 pp 486ndash489 2012

[27] G V Glass ldquoTesting Homogeneity of Variancesrdquo AmericanEducational Research Journal vol 3 no 3 pp 187ndash190 1966

[28] H BMann ldquoNonparametric tests against trendrdquo Econometricavol 13 pp 245ndash259 1945

[29] M G Kendall Rank Correlation Method Charles GriffinLondon UK 4th edition 1975

[30] H Turoglu ldquoDetection of Changes on Temperature and Precip-itation Features in Istanbul (Turkey)rdquo Atmospheric and ClimateSciences vol 04 no 04 pp 549ndash562 2014

[31] A Karaburun A Demirci and I-S Suen ldquoImpacts of urbangrowth on forest cover in Istanbul (1987-2007)rdquo EnvironmentalModeling amp Assessment vol 166 no 1-4 pp 267ndash277 2010

[32] K K Karanth LM Curran and J D Reuning-Scherer ldquoVillagesize and forest disturbance in Bhadra Wildlife SanctuaryWestern Ghats Indiardquo Biological Conservation vol 128 no 2pp 147ndash157 2006

[33] G Cakir C Un E Z Baskent S Kose F Sivrikaya andS Keles ldquoEvaluating urbanization fragmentation and landuseland cover change pattern in Istanbul city Turkey from 1971to 2002rdquo Land Degradation amp Development vol 19 no 6 pp663ndash675 2008

[34] R B Myneni F G Hall P J Sellers and A L Marshak ldquoTheinterpretation of spectral vegetation indexesrdquo IEEE Transac-tions on Geoscience and Remote Sensing vol 33 no 2 pp 481ndash486 1995

[35] Y S Unal H Toros A Deniz and S Incecik ldquoInfluence ofmeteorological factors and emission sources on spatial and tem-poral variations of PM10 concentrations in Istanbul metropoli-tan areardquo Atmospheric Environment vol 45 no 31 pp 5504ndash5513 2011

[36] M KaracaM Tayanc andH Toros ldquoEffects of urbanization onclimate of Istanbul and Ankarardquo Atmospheric Environment vol29 no 23 pp 3411ndash3421 1995

[37] Y Ezber O L Sen T Kindap and M Karaca ldquoClimatic effectsof urbanization in Istanbul a statistical and modeling analysisrdquoInternational Journal of Climatology vol 27 no 5 pp 667ndash6792007

[38] H S Park ldquoFeatures of the heat island in seoul and its sur-rounding citiesrdquo Atmospheric Environment (1967) vol 20 no10 pp 1859ndash1866 1986

10 Advances in Meteorology

Table2Statisticalanalyses

outputso

ftem

perature

andprecipitatio

nparameters(SD

stand

arddeviation

Inc

increasedecdecrease)

Parameters

Minim

umtemperature

(∘ C)

Averagetem

perature

(∘ C)

Maxim

umtemperature

(∘ C)

Precipitatio

n(m

m)

Statistics

Perio

ds

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Incor

dec

1912ndash

1980

1981ndash

2016

Incor

dec

1912ndash

1964

1965ndash

2016

Inc or dec

1912ndash

1980

1981ndash

2016

Inc or dec

Average

total

102

107

052

102

109

071

136

140

037

136

141

043

180

187

075

181

187

060

8113

8656

544

8267

8591

324

SD66

66

004

65

67

022

7172

007

7173

019

82

84

021

82

85

025

9391minus03

9193

02

Minim

umminus54minus42

116minus53minus39

138minus27minus21

062minus27minus19

074minus08minus01

076minus07

00

068

00

00

00

00

00

00

Maxim

um223

229

069

222

232

099

276

278

023

277

277

000

356

362

059

358

361

029

589

620

31

579

651

71Percentile5

minus04

02

061minus03

03

055

21

25

049

22

24

026

45

50

052

46

48

015

01

01minus01

01

01

00

Percentile2

547

52

045

48

52

038

7780

024

7778

01 0

114

119

054

115

116

011

07

05minus03

06

Percentile5

0106

108

019

105

106

007

140

141

013

140

138minus020

186

190

040

187

185minus024

26

21minus05

23

24

01

Percentile7

5159

165

066

158

164

059

199

204

054

199

201

022

250

261

105

252

255

034

7672minus04

7374

01

Percentile9

5196

203

074

195

202

074

236

242

065

236

239

032

296

307

110

297

300

027

229

226minus03

227

223minus05

gt30

Cggt80

mm

00

00

00

00

00

00

1626

1017

2810

00

00

00

gt25

Cggt40

mm

00

00

00

712

57

147

90107

1794

108

142

20

22

0lt0C

gmdash

2316minus6

2216minus6

76minus2

76minus1

21minus1

21minus1

mdashmdash

mdashmdash

mdashmdash

ltminus5C

gmdash

21minus1

21minus1

00

00

00

00

00

00

mdashmdash

mdashmdash

mdashmdash

Winter

33

36

032

33

38

048

59

62

031

60

62

020

9095

047

9195

035

1033

1096

63

1056

1066

09

Sprin

g74

87

131

7582

073

111

124

121

113

119

060

160

171

116

162

170

076

496

513

17515

511minus04

Summer

174

187

124

174

186

125

216

224

082

216

225

091

269

273

043

271

282

109

334

331minus03

330

368

38

Autumn

124

127

028

123

127

036

156

156

000

156

155minus001

199

197minus020

200

201

019

842

831minus10

854

919

65

Ann

ual

101

109

079

101

108

070

136

141

057

136

140

043

179

184

042

181

187

060

676

719

43

689

716

27

Advances in Meteorology 11

value of 18∘C for the period of 1912ndash1964 and an averagevalue of 187∘C for the period of 1965ndash2016 On the otherhand comparison of the daily maximum temperature seriesfor the periods of 1912ndash1980 and 1981ndash2016 has revealed anincrement of 06∘C with an average value between 182∘C and188∘C for both of these periods respectively In additionthe standard deviation values are increased between 02∘Cand 03∘C during the whole studied periodThe increment inthe minimum values of daily maximum temperature seriesis more evident than the maximum values The minimumvalues of daily maximum temperatures have exhibited anincrement of 08∘C for the period of 1965ndash2016 compared tothe previous period of 1912ndash1964 and also an increment of07∘C for the period of 1981ndash2016 compared to the previousperiod of 1912ndash1980 Also themaximumvalues of dailymaxi-mum temperature series exhibited amean increment of 05∘Cfor both studied periods Seasonal analysis of maximumtemperature series for both of studied section periods showedthat the highest rising temperature has been happened bya value of 12∘C in summer On the other hand the lowestincrement has happened in the autumn season The monthlyanalysis of daily maximum temperature series showed thatthe highest increment took place during the last monthsof the spring season and the first month of the summerseason in all of the studied periods When the percentiles ofdaily maximum temperatures are analyzed the temperatureincrement based on the 5th percentile threshold is 05∘C forthe periods of 1912ndash1964 and 1965ndash2016 while the incrementis 03∘C for the periods of 1912ndash1980 and 1981ndash2016 This canbe considered as an important sign of rising temperature overthe time In this case the value of percentile thresholds isincreased with extending the length of the time period withextending the length of the first section of studied period infavor of last years than the previous ones In this regard thesevalues are 05∘C and 04∘C at the 25th percentile 1∘C and11∘C at the 75th percentile and 11∘C and 12∘C at the 95thpercentile for both sections of studied periods respectivelyThis situation shows that the increment of higher values inthe dailymaximum temperatures is greater than lower values

The comparison analysis of daily minimum temperaturesbetween the periods of 1912ndash1964 and 1965ndash2016 and theperiods of 1912ndash1980 and 1981ndash2016 showed that there is ageneral increment of 05∘C during the first section periodswhich can be given as 102∘C and 107∘C for the individualperiods of 1912ndash1964 and 1965ndash2016 and also a generalincrement of 08∘C during the second section periods whichcan be provided as 102∘C and 11∘C for the individual periodsof 1912ndash1980 and 1981ndash2016 respectively In addition thestandard deviation values increased among these sectionperiods from 0∘C to 03∘C The increment in the minimumvalues of daily minimum temperature series is more evidentthan the maximum values The minimum values of dailyminimum temperature series showed an increment of 12∘Cfrom the period of 1912ndash1964 to 1965ndash2016 and also anincrement of 16∘C from the period of 1912ndash1980 to 1981ndash2016The maximum values of daily minimum temperature serieshave shown an increment of 07∘C for the section periodsof 1912ndash1964 and 1965ndash2016 and an increment of 12∘C forthe section periods of 1912ndash1980 and 1981ndash2016 Overall

the minimum values have had an average increment of14∘C while the maximum values have had an averageincrement of 1∘C during the last century which can showthe higher rate of upward trends in the temperature timeseries Monthly analysis of minimum temperature seriesshowed that the highest increment for the section studiedperiods of 1912ndash1964 and 1965ndash2016 has occurred by a valueof 1∘C in June while the highest increment for the sectionstudied periods of 1912ndash1980 and 1981ndash2016 has occurred bya value of 16∘C in August Meanwhile there is a temperaturedecrement in November in both of these periods Howeverthere is a clear decrement in the first half of these periodsin October and December whereas there is an incrementin the second half of these periods The seasonal analysisof the daily minimum temperature series showed that thehighest increment has happened in the summer season witha mean value of 07∘C for the section periods of 1912ndash1964and 1965ndash2016 and with a mean value of 13∘C for the sectionperiods of 1912ndash1980 and 1981ndash2016 respectively Then theincreasing rate during the summer season became moreevident during the recent decades Analysis of the percentilethresholds of daily minimum temperature series showed thatthe temperature increment at the 5th percentile is 06∘C forall studied time periods of 1912ndash1964 1965ndash2016 1912ndash1980and 1981ndash2016 Also these increment values indicate anincrease of 04∘C and 06∘C at the 25th percentile 07∘C and11∘C at the 75th percentile and 07∘C and 14∘C at the 95thpercentile thresholds for the whole studied subperiods of1912ndash1964 1965ndash2016 1912ndash1980 and 1981ndash2016 respectivelyThese rising rates in minimum temperature series are moreevident for the periods of 1912ndash1980 and 1981ndash2016 than theperiods of 1912ndash1980 and 1981ndash2016 Therefore it can be saidthat the rate of temperature rising has increased further aslong as the studied time period is closer to the last years

42 Precipitation Annual average precipitation in Istanbulis 838mm with a range of minimum value of 449mm in1921 and a maximum value of 1289mm in 1981 based on theobservatory data of Kandilli station during the whole studiedperiod from 1912 to 2016 (Figure 4(a)) Also monthly averageprecipitation is 699mm with a range of the minimum of326mm in July and the maximum of 1286mm in Decemberduring the whole studied time period (Figure 4(b))

Analysis of the trend in the annual average precipitationtime series by the methods of linear regression analysisand MK trend test has shown that periodically there arepartial increments and significant differences during the totalstudied period from 1912 to 2016 (Figures 5(a1) and 5(a2))But this increment in the precipitation time series is notas clear as the increment in the temperature time seriesHowever it is obvious in the precipitation time series thatthere is an increment between the years of 1917 and 1925a stable condition between the years of 1925 and 1954 anincrement between the years of 1954 and 1965 a decrementbetween the years of 1965 and 1974 an increment between theyears of 1974 and 2001 and again a no change situation from2001 till the end The trend analysis of the daily maximumprecipitation series showed an increment of 29mm for theperiods of 1912ndash1964 and 1965ndash2016 as well as an increment

12 Advances in Meteorology

of 93mm for the periods of 1912ndash1980 and 1981ndash2016 Thesevalues indicated that rainfall has increased at a rate of morethan three times over the last decades than the previous ones(Figures 5(b1) and 5(b2)) Also the number of days with dailyprecipitation greater than 25mm presented an incrementduring the period of 1912ndash2016 although this is not significantat the confidence level of 005 (Figures 5(c1) and 5(c2))The trend of standard deviation in the daily precipitationtime series showed a slowly increasing trend during the totalstudied period (Figures 5(d1) and 5(d2))

The comparison of the results of statistical analysisbetween the daily average rainfall amounts belonging tothe periods of 1912ndash1964 and 1965ndash2016 with those of theperiods of 1912ndash1980 and 1981ndash2016 revealed that there is anincrement of 78mm from the period of 1912ndash1964 to theperiod of 1965ndash2016 and an increment of 38mm from theperiod of 1912ndash1980 to the period of 1981ndash2016 In additionthe analysis of standard deviation exhibited a decrementfrom the period of 1912ndash1964 to the period of 1965ndash2016 andan increment from the period of 1912ndash1980 to the periodof 1981ndash2016 (Table 2) The analysis of monthly averageprecipitation time series showed that the highest incrementhas happened in October with 227mm for the sectionperiods of 1912ndash1964 and 1965ndash2016 and with 347mm forthe section periods of 1912ndash1980 and 1981ndash2016 On the otherhand the highest decrease took placewith a value ofminus04mmduring both May and July for the periods of 1912ndash1964 and1965ndash2016 and with a value of minus16mm in September for thesection periods of 1912ndash1980 and 1981ndash2016 The analysis ofseasonal average precipitation time series showed that thehighest increase occurred in autumn with a value of 72mmfor the section periods of 1912ndash1964 and 1965ndash2016 as wellas with a value of 64mm for the periods of 1912ndash1980 and1981ndash2016 respectively There is no remarkable decreasingchange seasonally except in winter season during the periodsof 1912ndash1980 and 1981ndash2016 Furthermore the percentilethresholds of daily average precipitation indicated that thereis an insignificant negative trend based on all percentilesand for all studied periods Overall the statistical analysisshowed that the total average precipitation of Istanbul hasincreased while this increasing trend is more pronounced inthe previous decades than the last 3 decades On the otherhand the increasing rate of daily maximum precipitation ismore evident during the last 3 decades than the previousdecades which can be proven by the increasing frequency ofheavy rainfall events in Istanbul

Generally the results of trend analysis of Kandilli stationduring the last 105 years of 1912ndash2016 showed that there isa warming significant trend in the precipitation time seriesby using both methods of linear regression analysis andMK trend test On the contrary previous climate studiesconducted over Turkey put forward that there has been adecreasing trend in annual precipitation time series duringthe recent decades regionally The results of a previouslyconducted study by using the daily precipitation and tem-perature data sets of Florya and Goztepe meteorologicalstations in Istanbul area between 1960 and 2013 showedthat most notably the precipitation during the warm periodshas decreased but the frequency of the intense rain has

increased and the majority of these episodes of intense raincoincided with the warm periods Other determinationswere the rise in the annual average temperature and theextension of the warm periods in a year This differentiationof the temperature features can lead to the aggravation of theevaporation and it can be effective for a longer period duringthe year [15] Thus it will make Istanbul be confronted withthe much more important problems of water managementand flood [34] Also the results of the current study for trendanalysis in the long period from 1912 to 2016 showed that themost striking spell is between the years of 1968 and 1998 dueto the existence of least number of rainfall events in IstanbulIt can be owing to industrialization along with the increasingair pollution as well as irregular urbanization in Istanbul areaIn this case severe droughts taking place during the yearsof 1988 1992 1993 and 2008 have threatened the reservoirswhich supplied fresh water of the city These years werecharacterized by not having enough rainfall events Theseyears are also characterized by more persistent high-pressuresystems and less occurrence of low-pressure systems in termsof number and strength As statements made by officialinstitutions the formation conditions for the atmosphericlayer of air pollution due to air pollutant emissions from fossilfuel combustion and industrial activities are more providedduring anticyclone or high-pressure system eventsThereforethis leads to warming up and generating an inversion layerin the boundary layer of atmosphere especially over city area[35] The inversion layer or urban heat island intensity isincreasing with the increasing city size andor populationa phenomenon that was also reported by others [36ndash38]Moreover these last climatic events also have affected somepolitical results beyond natural effects The most obviousexample of this is related to the local election of 1994 inIstanbul In this case extreme drought during the summers of1992 and 1993 has caused groundwater reservoirs to dry up inthe city discontinuance of water was experienced for severaldays or even weeks In those times the mayor of Istanbul cityhas lost the election of 1994 and he realized that this result wasdue to the peoplersquos reaction about the water shortages Manypolitical reports of that period also support this scientificview

5 Conclusions

Statistical analysis in temperature time series of the Kandillistation from 1912 to 2016 established that there is a notableincrease in temperature values after the 1940s which is inparallel with the beginning of industrialization era in Istan-bul There has been a rise about 094∘C in the daily averagetemperature series since the beginning of the last century Asignificant positive trend in the daily maximum temperatureseries is found about 156∘C Also there is a positive trendabout 087∘C in the daily minimum temperature series Onthe other hand analysis of the number of days with thedaily maximum temperature higher than 30∘C showed thatthere is an increasing trend Meanwhile analysis of thenumber of days with daily minimum temperatures lowerthan 0∘C showed a decreasing trend The increment in theminimum values of the daily minimum temperature series

Advances in Meteorology 13

is more evident than the maximum values of this series Inthis case these rising rates in the minimum temperatureseries are more evident for the section periods of 1912ndash1980and 1981ndash2016 than the section periods of 1912ndash1964 and1965ndash2016 This again shows that there is an increment inthe positive temperature trend from past to present decadesThe increment in the precipitation time series is not asclear as the increment in the temperature time series dueto periodic variability The trend analysis in the total annualprecipitation time series showed that the first significantupward trend has periodically been started from the 1920swhile there is a stable trend from 2001 till 2016 The dailyaverage of rainfall amounts has increased with a value of58mm during the period of 1912ndash2016 Also the analysisof heavy precipitation trend showed an increase of 61mmOverall the total average precipitation of Istanbul increasedwhile this increasing trend is more pronounced during theearly decades than the last 3 decades On the other handthe increasing rate of daily maximum precipitation is morepronounced in the last 3 decades than the previous decadesThen it was shown that the frequency of heavy rainfallat Istanbul has increased during the recent decades Thusthe precipitation changes in Istanbul have some differencescompared to the general tendency in precipitation trendthat was put forward by other studies as a decreasing trendover the whole of Turkey This result can be expressedas a positive effect of population overgrowth of Istanbulmegacity Comparison of the results in the first half of thestudy period (1912ndash1964) with the second half of the studyperiod (1965ndash2016) showed that both the average temperatureand average precipitation have higher values of 139∘C and878mm for the final phase compared to the values of 136∘Cand 799mm belonging to the initial phase Therefore it canbe stated that the megacity of Istanbul is directly affectedby the climate change and its consequences In this contextpotential risks of climate change in Istanbul megacity underhigher temperature conditions can be expressed as the rise inthe sea level increase in the rate of evapotranspiration andincrease in the frequency of heavy rainfall Also this city maynot be able to handle this uncontrolled population growthand its associated irreversible changes which is alreadypushing the natural limits by destroying the environmentTherefore the local governors of any megacity like Istanbulshould give more emphasis on the importance of sustainableurban development Thus it is urgent to prepare local andnational climate change strategies and action plans for themegacities

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

The authors are grateful to the Bogazici University theobservatory of Kandilli weather station and the Earthquake

Research Institute for providing the research data and tech-nical support The authors also gratefully acknowledge con-tributions of Assoc Professor Dr Yuksel Demirkaya Schoolof Social Sciences Marmara University This work has beensupported by Scientific and Technological Research Councilof Turkey (TUBITAK) under Grants 113R019 and 106Y258and by Marmara University (BAPKO) with projects FEN-E-120314-0066 FEN-C-YLP-090414-0102 FEN-L-250416-0180 and FEN-A-100413-0127

References

[1] D R Easterling B Horton P D Jones et al ldquoMaximum andminimum temperature trends for the globerdquo Science vol 277no 5324 pp 364ndash367 1997

[2] IPCC Climate Change 2014 IPCC Fifth Assessment Synthe-sis Report-Summary for Policymakers-an Assessment of Inter-Governmental Panel on Climate Change Cambridge UniversityPress Cambridge UK 2014

[3] J Carmin N Nadkarni and C Rhie Progress and Challenges inUrban Climate Adaptation Planning Results of a Global SurveyMIT Cambridge UK 2012

[4] A F Young ldquoUrban expansion and environmental risk in theSao Paulo Metropolitan Areardquo Climate Research vol 57 no 1pp 73ndash80 2013

[5] P Tian X Mu J Liu J Hu and C Gu ldquoImpacts of ClimateVariability and Human Activities on the Changes of Runoff andSediment Load in a Catchment of the Loess Plateau ChinardquoAdvances inMeteorology vol 2016 Article ID 4724067 15 pages2016

[6] R S Kovats and K L Ebi ldquoHeatwaves and public health inEuroperdquo European Journal of Public Health vol 16 no 6 pp592ndash599 2006

[7] S Conti P Meli G Minelli et al ldquoEpidemiologic studyof mortality during the Summer 2003 heat wave in ItalyrdquoEnvironmental Research vol 98 no 3 pp 390ndash399 2005

[8] J Kysely and J Kim ldquoMortality during heat waves in SouthKorea 1991 to 2005 how exceptional was the 1994 heat waverdquoClimate Research vol 38 no 2 pp 105ndash116 2009

[9] B Yan Z Xia F Huang L Guo and X Zhang ldquoClimatechange detection and annual extreme temperature analysis ofthe amur river basinrdquoAdvances inMeteorology vol 2016 ArticleID 6268938 14 pages 2016

[10] E M Fischer and R Knutti ldquoAnthropogenic contribution toglobal occurrence of heavy-precipitation and high-temperatureextremesrdquo Nature Climate Change vol 5 no 6 pp 560ndash5642015

[11] X Zhang L Alexander G C Hegerl et al ldquoIndices for moni-toring changes in extremes based on daily temperature andprecipitation datardquo Climate Change vol 2 no 6 pp 851ndash8702011

[12] K H Schlunzen P Hoffmann G Rosenhagen and W RieckeldquoLong-term changes and regional differences in temperatureand precipitation in the metropolitan area of Hamburgrdquo Inter-national Journal of Climatology vol 30 no 8 pp 1121ndash11362010

[13] G Bartolini M Morabito A Crisci et al ldquoRecent trends inTuscany (Italy) summer temperature and indices of extremesrdquoInternational Journal of Climatology vol 28 no 13 pp 1751ndash1760 2008

14 Advances in Meteorology

[14] S C Sheridan and T J Dolney ldquoHeat mortality and levelof urbanization measuring vulnerability across Ohio USArdquoClimate Research vol 24 no 3 pp 255ndash265 2003

[15] M Tayanc U Im M Dogruel andM Karaca ldquoClimate changein Turkey for the last half centuryrdquo Climatic Change vol 94 no3-4 pp 483ndash502 2009

[16] H Toros ldquoSpatio-temporal variation of daily extreme tempera-tures over Turkeyrdquo International Journal of Climatology vol 32no 7 pp 1047ndash1055 2012

[17] H Toros ldquoSpatio-temporal precipitation change assessmentsover Turkeyrdquo International Journal of Climatology vol 32 no9 pp 1310ndash1325 2012

[18] M Turkes C Yozgatlıgil I Batmaz et al ldquoHas the climate beenchanging in Turkey Regional climate change signals based on acomparative statistical analysis of two consecutive time periods1950-1980 and 1981-2010rdquoClimate Research vol 70 no 1 pp 77ndash93 2016

[19] IPCC ldquoummary for Policymakers In Climate Change 2013rdquo inThe Physical Science Basis The contribution of Working Group Ito the Fifth Assessment Report of the Intergovernmental Panel onClimate Change Cambridge University Press Cambridge UK2013

[20] United Nations ldquoDepartment of Economic and Social AffairsPopulation Division 2006 World Urbanization Prospects The2005 Revisionrdquo Working Paper ESAPWP200 2011

[21] Y Demirkaya Sayılarla Istanbul ITO Istanbul Turkey 2011[22] TUIK ldquoAddress based population registration system results of

2014rdquo Turkish Statistical Institute (TUIK) 2017 httpraporytuikgovtr10-03-2015-184727-842632346446643191142126876html

[23] S Erinc Climatology and its Methods Alfa Basım YayımDagitim Istanbul Turkey 4th edition 1965

[24] O M Gokturk D Bozkurt O L Sen and M Karaca ldquoQualitycontrol and homogeneity of Turkish precipitation datardquoHydro-logical Processes vol 22 no 16 pp 3210ndash3218 2008

[25] C Ley and D Paindaveine ldquoRuns Testsrdquo in Encyclopedia ofEnvironmetrics 2012

[26] A Ghasemi and S Zahediasl ldquoNormality tests for statisticalanalysis a guide for non-statisticiansrdquo International Journal ofEndocrinology andMetabolism vol 10 no 2 pp 486ndash489 2012

[27] G V Glass ldquoTesting Homogeneity of Variancesrdquo AmericanEducational Research Journal vol 3 no 3 pp 187ndash190 1966

[28] H BMann ldquoNonparametric tests against trendrdquo Econometricavol 13 pp 245ndash259 1945

[29] M G Kendall Rank Correlation Method Charles GriffinLondon UK 4th edition 1975

[30] H Turoglu ldquoDetection of Changes on Temperature and Precip-itation Features in Istanbul (Turkey)rdquo Atmospheric and ClimateSciences vol 04 no 04 pp 549ndash562 2014

[31] A Karaburun A Demirci and I-S Suen ldquoImpacts of urbangrowth on forest cover in Istanbul (1987-2007)rdquo EnvironmentalModeling amp Assessment vol 166 no 1-4 pp 267ndash277 2010

[32] K K Karanth LM Curran and J D Reuning-Scherer ldquoVillagesize and forest disturbance in Bhadra Wildlife SanctuaryWestern Ghats Indiardquo Biological Conservation vol 128 no 2pp 147ndash157 2006

[33] G Cakir C Un E Z Baskent S Kose F Sivrikaya andS Keles ldquoEvaluating urbanization fragmentation and landuseland cover change pattern in Istanbul city Turkey from 1971to 2002rdquo Land Degradation amp Development vol 19 no 6 pp663ndash675 2008

[34] R B Myneni F G Hall P J Sellers and A L Marshak ldquoTheinterpretation of spectral vegetation indexesrdquo IEEE Transac-tions on Geoscience and Remote Sensing vol 33 no 2 pp 481ndash486 1995

[35] Y S Unal H Toros A Deniz and S Incecik ldquoInfluence ofmeteorological factors and emission sources on spatial and tem-poral variations of PM10 concentrations in Istanbul metropoli-tan areardquo Atmospheric Environment vol 45 no 31 pp 5504ndash5513 2011

[36] M KaracaM Tayanc andH Toros ldquoEffects of urbanization onclimate of Istanbul and Ankarardquo Atmospheric Environment vol29 no 23 pp 3411ndash3421 1995

[37] Y Ezber O L Sen T Kindap and M Karaca ldquoClimatic effectsof urbanization in Istanbul a statistical and modeling analysisrdquoInternational Journal of Climatology vol 27 no 5 pp 667ndash6792007

[38] H S Park ldquoFeatures of the heat island in seoul and its sur-rounding citiesrdquo Atmospheric Environment (1967) vol 20 no10 pp 1859ndash1866 1986

Advances in Meteorology 11

value of 18∘C for the period of 1912ndash1964 and an averagevalue of 187∘C for the period of 1965ndash2016 On the otherhand comparison of the daily maximum temperature seriesfor the periods of 1912ndash1980 and 1981ndash2016 has revealed anincrement of 06∘C with an average value between 182∘C and188∘C for both of these periods respectively In additionthe standard deviation values are increased between 02∘Cand 03∘C during the whole studied periodThe increment inthe minimum values of daily maximum temperature seriesis more evident than the maximum values The minimumvalues of daily maximum temperatures have exhibited anincrement of 08∘C for the period of 1965ndash2016 compared tothe previous period of 1912ndash1964 and also an increment of07∘C for the period of 1981ndash2016 compared to the previousperiod of 1912ndash1980 Also themaximumvalues of dailymaxi-mum temperature series exhibited amean increment of 05∘Cfor both studied periods Seasonal analysis of maximumtemperature series for both of studied section periods showedthat the highest rising temperature has been happened bya value of 12∘C in summer On the other hand the lowestincrement has happened in the autumn season The monthlyanalysis of daily maximum temperature series showed thatthe highest increment took place during the last monthsof the spring season and the first month of the summerseason in all of the studied periods When the percentiles ofdaily maximum temperatures are analyzed the temperatureincrement based on the 5th percentile threshold is 05∘C forthe periods of 1912ndash1964 and 1965ndash2016 while the incrementis 03∘C for the periods of 1912ndash1980 and 1981ndash2016 This canbe considered as an important sign of rising temperature overthe time In this case the value of percentile thresholds isincreased with extending the length of the time period withextending the length of the first section of studied period infavor of last years than the previous ones In this regard thesevalues are 05∘C and 04∘C at the 25th percentile 1∘C and11∘C at the 75th percentile and 11∘C and 12∘C at the 95thpercentile for both sections of studied periods respectivelyThis situation shows that the increment of higher values inthe dailymaximum temperatures is greater than lower values

The comparison analysis of daily minimum temperaturesbetween the periods of 1912ndash1964 and 1965ndash2016 and theperiods of 1912ndash1980 and 1981ndash2016 showed that there is ageneral increment of 05∘C during the first section periodswhich can be given as 102∘C and 107∘C for the individualperiods of 1912ndash1964 and 1965ndash2016 and also a generalincrement of 08∘C during the second section periods whichcan be provided as 102∘C and 11∘C for the individual periodsof 1912ndash1980 and 1981ndash2016 respectively In addition thestandard deviation values increased among these sectionperiods from 0∘C to 03∘C The increment in the minimumvalues of daily minimum temperature series is more evidentthan the maximum values The minimum values of dailyminimum temperature series showed an increment of 12∘Cfrom the period of 1912ndash1964 to 1965ndash2016 and also anincrement of 16∘C from the period of 1912ndash1980 to 1981ndash2016The maximum values of daily minimum temperature serieshave shown an increment of 07∘C for the section periodsof 1912ndash1964 and 1965ndash2016 and an increment of 12∘C forthe section periods of 1912ndash1980 and 1981ndash2016 Overall

the minimum values have had an average increment of14∘C while the maximum values have had an averageincrement of 1∘C during the last century which can showthe higher rate of upward trends in the temperature timeseries Monthly analysis of minimum temperature seriesshowed that the highest increment for the section studiedperiods of 1912ndash1964 and 1965ndash2016 has occurred by a valueof 1∘C in June while the highest increment for the sectionstudied periods of 1912ndash1980 and 1981ndash2016 has occurred bya value of 16∘C in August Meanwhile there is a temperaturedecrement in November in both of these periods Howeverthere is a clear decrement in the first half of these periodsin October and December whereas there is an incrementin the second half of these periods The seasonal analysisof the daily minimum temperature series showed that thehighest increment has happened in the summer season witha mean value of 07∘C for the section periods of 1912ndash1964and 1965ndash2016 and with a mean value of 13∘C for the sectionperiods of 1912ndash1980 and 1981ndash2016 respectively Then theincreasing rate during the summer season became moreevident during the recent decades Analysis of the percentilethresholds of daily minimum temperature series showed thatthe temperature increment at the 5th percentile is 06∘C forall studied time periods of 1912ndash1964 1965ndash2016 1912ndash1980and 1981ndash2016 Also these increment values indicate anincrease of 04∘C and 06∘C at the 25th percentile 07∘C and11∘C at the 75th percentile and 07∘C and 14∘C at the 95thpercentile thresholds for the whole studied subperiods of1912ndash1964 1965ndash2016 1912ndash1980 and 1981ndash2016 respectivelyThese rising rates in minimum temperature series are moreevident for the periods of 1912ndash1980 and 1981ndash2016 than theperiods of 1912ndash1980 and 1981ndash2016 Therefore it can be saidthat the rate of temperature rising has increased further aslong as the studied time period is closer to the last years

42 Precipitation Annual average precipitation in Istanbulis 838mm with a range of minimum value of 449mm in1921 and a maximum value of 1289mm in 1981 based on theobservatory data of Kandilli station during the whole studiedperiod from 1912 to 2016 (Figure 4(a)) Also monthly averageprecipitation is 699mm with a range of the minimum of326mm in July and the maximum of 1286mm in Decemberduring the whole studied time period (Figure 4(b))

Analysis of the trend in the annual average precipitationtime series by the methods of linear regression analysisand MK trend test has shown that periodically there arepartial increments and significant differences during the totalstudied period from 1912 to 2016 (Figures 5(a1) and 5(a2))But this increment in the precipitation time series is notas clear as the increment in the temperature time seriesHowever it is obvious in the precipitation time series thatthere is an increment between the years of 1917 and 1925a stable condition between the years of 1925 and 1954 anincrement between the years of 1954 and 1965 a decrementbetween the years of 1965 and 1974 an increment between theyears of 1974 and 2001 and again a no change situation from2001 till the end The trend analysis of the daily maximumprecipitation series showed an increment of 29mm for theperiods of 1912ndash1964 and 1965ndash2016 as well as an increment

12 Advances in Meteorology

of 93mm for the periods of 1912ndash1980 and 1981ndash2016 Thesevalues indicated that rainfall has increased at a rate of morethan three times over the last decades than the previous ones(Figures 5(b1) and 5(b2)) Also the number of days with dailyprecipitation greater than 25mm presented an incrementduring the period of 1912ndash2016 although this is not significantat the confidence level of 005 (Figures 5(c1) and 5(c2))The trend of standard deviation in the daily precipitationtime series showed a slowly increasing trend during the totalstudied period (Figures 5(d1) and 5(d2))

The comparison of the results of statistical analysisbetween the daily average rainfall amounts belonging tothe periods of 1912ndash1964 and 1965ndash2016 with those of theperiods of 1912ndash1980 and 1981ndash2016 revealed that there is anincrement of 78mm from the period of 1912ndash1964 to theperiod of 1965ndash2016 and an increment of 38mm from theperiod of 1912ndash1980 to the period of 1981ndash2016 In additionthe analysis of standard deviation exhibited a decrementfrom the period of 1912ndash1964 to the period of 1965ndash2016 andan increment from the period of 1912ndash1980 to the periodof 1981ndash2016 (Table 2) The analysis of monthly averageprecipitation time series showed that the highest incrementhas happened in October with 227mm for the sectionperiods of 1912ndash1964 and 1965ndash2016 and with 347mm forthe section periods of 1912ndash1980 and 1981ndash2016 On the otherhand the highest decrease took placewith a value ofminus04mmduring both May and July for the periods of 1912ndash1964 and1965ndash2016 and with a value of minus16mm in September for thesection periods of 1912ndash1980 and 1981ndash2016 The analysis ofseasonal average precipitation time series showed that thehighest increase occurred in autumn with a value of 72mmfor the section periods of 1912ndash1964 and 1965ndash2016 as wellas with a value of 64mm for the periods of 1912ndash1980 and1981ndash2016 respectively There is no remarkable decreasingchange seasonally except in winter season during the periodsof 1912ndash1980 and 1981ndash2016 Furthermore the percentilethresholds of daily average precipitation indicated that thereis an insignificant negative trend based on all percentilesand for all studied periods Overall the statistical analysisshowed that the total average precipitation of Istanbul hasincreased while this increasing trend is more pronounced inthe previous decades than the last 3 decades On the otherhand the increasing rate of daily maximum precipitation ismore evident during the last 3 decades than the previousdecades which can be proven by the increasing frequency ofheavy rainfall events in Istanbul

Generally the results of trend analysis of Kandilli stationduring the last 105 years of 1912ndash2016 showed that there isa warming significant trend in the precipitation time seriesby using both methods of linear regression analysis andMK trend test On the contrary previous climate studiesconducted over Turkey put forward that there has been adecreasing trend in annual precipitation time series duringthe recent decades regionally The results of a previouslyconducted study by using the daily precipitation and tem-perature data sets of Florya and Goztepe meteorologicalstations in Istanbul area between 1960 and 2013 showedthat most notably the precipitation during the warm periodshas decreased but the frequency of the intense rain has

increased and the majority of these episodes of intense raincoincided with the warm periods Other determinationswere the rise in the annual average temperature and theextension of the warm periods in a year This differentiationof the temperature features can lead to the aggravation of theevaporation and it can be effective for a longer period duringthe year [15] Thus it will make Istanbul be confronted withthe much more important problems of water managementand flood [34] Also the results of the current study for trendanalysis in the long period from 1912 to 2016 showed that themost striking spell is between the years of 1968 and 1998 dueto the existence of least number of rainfall events in IstanbulIt can be owing to industrialization along with the increasingair pollution as well as irregular urbanization in Istanbul areaIn this case severe droughts taking place during the yearsof 1988 1992 1993 and 2008 have threatened the reservoirswhich supplied fresh water of the city These years werecharacterized by not having enough rainfall events Theseyears are also characterized by more persistent high-pressuresystems and less occurrence of low-pressure systems in termsof number and strength As statements made by officialinstitutions the formation conditions for the atmosphericlayer of air pollution due to air pollutant emissions from fossilfuel combustion and industrial activities are more providedduring anticyclone or high-pressure system eventsThereforethis leads to warming up and generating an inversion layerin the boundary layer of atmosphere especially over city area[35] The inversion layer or urban heat island intensity isincreasing with the increasing city size andor populationa phenomenon that was also reported by others [36ndash38]Moreover these last climatic events also have affected somepolitical results beyond natural effects The most obviousexample of this is related to the local election of 1994 inIstanbul In this case extreme drought during the summers of1992 and 1993 has caused groundwater reservoirs to dry up inthe city discontinuance of water was experienced for severaldays or even weeks In those times the mayor of Istanbul cityhas lost the election of 1994 and he realized that this result wasdue to the peoplersquos reaction about the water shortages Manypolitical reports of that period also support this scientificview

5 Conclusions

Statistical analysis in temperature time series of the Kandillistation from 1912 to 2016 established that there is a notableincrease in temperature values after the 1940s which is inparallel with the beginning of industrialization era in Istan-bul There has been a rise about 094∘C in the daily averagetemperature series since the beginning of the last century Asignificant positive trend in the daily maximum temperatureseries is found about 156∘C Also there is a positive trendabout 087∘C in the daily minimum temperature series Onthe other hand analysis of the number of days with thedaily maximum temperature higher than 30∘C showed thatthere is an increasing trend Meanwhile analysis of thenumber of days with daily minimum temperatures lowerthan 0∘C showed a decreasing trend The increment in theminimum values of the daily minimum temperature series

Advances in Meteorology 13

is more evident than the maximum values of this series Inthis case these rising rates in the minimum temperatureseries are more evident for the section periods of 1912ndash1980and 1981ndash2016 than the section periods of 1912ndash1964 and1965ndash2016 This again shows that there is an increment inthe positive temperature trend from past to present decadesThe increment in the precipitation time series is not asclear as the increment in the temperature time series dueto periodic variability The trend analysis in the total annualprecipitation time series showed that the first significantupward trend has periodically been started from the 1920swhile there is a stable trend from 2001 till 2016 The dailyaverage of rainfall amounts has increased with a value of58mm during the period of 1912ndash2016 Also the analysisof heavy precipitation trend showed an increase of 61mmOverall the total average precipitation of Istanbul increasedwhile this increasing trend is more pronounced during theearly decades than the last 3 decades On the other handthe increasing rate of daily maximum precipitation is morepronounced in the last 3 decades than the previous decadesThen it was shown that the frequency of heavy rainfallat Istanbul has increased during the recent decades Thusthe precipitation changes in Istanbul have some differencescompared to the general tendency in precipitation trendthat was put forward by other studies as a decreasing trendover the whole of Turkey This result can be expressedas a positive effect of population overgrowth of Istanbulmegacity Comparison of the results in the first half of thestudy period (1912ndash1964) with the second half of the studyperiod (1965ndash2016) showed that both the average temperatureand average precipitation have higher values of 139∘C and878mm for the final phase compared to the values of 136∘Cand 799mm belonging to the initial phase Therefore it canbe stated that the megacity of Istanbul is directly affectedby the climate change and its consequences In this contextpotential risks of climate change in Istanbul megacity underhigher temperature conditions can be expressed as the rise inthe sea level increase in the rate of evapotranspiration andincrease in the frequency of heavy rainfall Also this city maynot be able to handle this uncontrolled population growthand its associated irreversible changes which is alreadypushing the natural limits by destroying the environmentTherefore the local governors of any megacity like Istanbulshould give more emphasis on the importance of sustainableurban development Thus it is urgent to prepare local andnational climate change strategies and action plans for themegacities

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

The authors are grateful to the Bogazici University theobservatory of Kandilli weather station and the Earthquake

Research Institute for providing the research data and tech-nical support The authors also gratefully acknowledge con-tributions of Assoc Professor Dr Yuksel Demirkaya Schoolof Social Sciences Marmara University This work has beensupported by Scientific and Technological Research Councilof Turkey (TUBITAK) under Grants 113R019 and 106Y258and by Marmara University (BAPKO) with projects FEN-E-120314-0066 FEN-C-YLP-090414-0102 FEN-L-250416-0180 and FEN-A-100413-0127

References

[1] D R Easterling B Horton P D Jones et al ldquoMaximum andminimum temperature trends for the globerdquo Science vol 277no 5324 pp 364ndash367 1997

[2] IPCC Climate Change 2014 IPCC Fifth Assessment Synthe-sis Report-Summary for Policymakers-an Assessment of Inter-Governmental Panel on Climate Change Cambridge UniversityPress Cambridge UK 2014

[3] J Carmin N Nadkarni and C Rhie Progress and Challenges inUrban Climate Adaptation Planning Results of a Global SurveyMIT Cambridge UK 2012

[4] A F Young ldquoUrban expansion and environmental risk in theSao Paulo Metropolitan Areardquo Climate Research vol 57 no 1pp 73ndash80 2013

[5] P Tian X Mu J Liu J Hu and C Gu ldquoImpacts of ClimateVariability and Human Activities on the Changes of Runoff andSediment Load in a Catchment of the Loess Plateau ChinardquoAdvances inMeteorology vol 2016 Article ID 4724067 15 pages2016

[6] R S Kovats and K L Ebi ldquoHeatwaves and public health inEuroperdquo European Journal of Public Health vol 16 no 6 pp592ndash599 2006

[7] S Conti P Meli G Minelli et al ldquoEpidemiologic studyof mortality during the Summer 2003 heat wave in ItalyrdquoEnvironmental Research vol 98 no 3 pp 390ndash399 2005

[8] J Kysely and J Kim ldquoMortality during heat waves in SouthKorea 1991 to 2005 how exceptional was the 1994 heat waverdquoClimate Research vol 38 no 2 pp 105ndash116 2009

[9] B Yan Z Xia F Huang L Guo and X Zhang ldquoClimatechange detection and annual extreme temperature analysis ofthe amur river basinrdquoAdvances inMeteorology vol 2016 ArticleID 6268938 14 pages 2016

[10] E M Fischer and R Knutti ldquoAnthropogenic contribution toglobal occurrence of heavy-precipitation and high-temperatureextremesrdquo Nature Climate Change vol 5 no 6 pp 560ndash5642015

[11] X Zhang L Alexander G C Hegerl et al ldquoIndices for moni-toring changes in extremes based on daily temperature andprecipitation datardquo Climate Change vol 2 no 6 pp 851ndash8702011

[12] K H Schlunzen P Hoffmann G Rosenhagen and W RieckeldquoLong-term changes and regional differences in temperatureand precipitation in the metropolitan area of Hamburgrdquo Inter-national Journal of Climatology vol 30 no 8 pp 1121ndash11362010

[13] G Bartolini M Morabito A Crisci et al ldquoRecent trends inTuscany (Italy) summer temperature and indices of extremesrdquoInternational Journal of Climatology vol 28 no 13 pp 1751ndash1760 2008

14 Advances in Meteorology

[14] S C Sheridan and T J Dolney ldquoHeat mortality and levelof urbanization measuring vulnerability across Ohio USArdquoClimate Research vol 24 no 3 pp 255ndash265 2003

[15] M Tayanc U Im M Dogruel andM Karaca ldquoClimate changein Turkey for the last half centuryrdquo Climatic Change vol 94 no3-4 pp 483ndash502 2009

[16] H Toros ldquoSpatio-temporal variation of daily extreme tempera-tures over Turkeyrdquo International Journal of Climatology vol 32no 7 pp 1047ndash1055 2012

[17] H Toros ldquoSpatio-temporal precipitation change assessmentsover Turkeyrdquo International Journal of Climatology vol 32 no9 pp 1310ndash1325 2012

[18] M Turkes C Yozgatlıgil I Batmaz et al ldquoHas the climate beenchanging in Turkey Regional climate change signals based on acomparative statistical analysis of two consecutive time periods1950-1980 and 1981-2010rdquoClimate Research vol 70 no 1 pp 77ndash93 2016

[19] IPCC ldquoummary for Policymakers In Climate Change 2013rdquo inThe Physical Science Basis The contribution of Working Group Ito the Fifth Assessment Report of the Intergovernmental Panel onClimate Change Cambridge University Press Cambridge UK2013

[20] United Nations ldquoDepartment of Economic and Social AffairsPopulation Division 2006 World Urbanization Prospects The2005 Revisionrdquo Working Paper ESAPWP200 2011

[21] Y Demirkaya Sayılarla Istanbul ITO Istanbul Turkey 2011[22] TUIK ldquoAddress based population registration system results of

2014rdquo Turkish Statistical Institute (TUIK) 2017 httpraporytuikgovtr10-03-2015-184727-842632346446643191142126876html

[23] S Erinc Climatology and its Methods Alfa Basım YayımDagitim Istanbul Turkey 4th edition 1965

[24] O M Gokturk D Bozkurt O L Sen and M Karaca ldquoQualitycontrol and homogeneity of Turkish precipitation datardquoHydro-logical Processes vol 22 no 16 pp 3210ndash3218 2008

[25] C Ley and D Paindaveine ldquoRuns Testsrdquo in Encyclopedia ofEnvironmetrics 2012

[26] A Ghasemi and S Zahediasl ldquoNormality tests for statisticalanalysis a guide for non-statisticiansrdquo International Journal ofEndocrinology andMetabolism vol 10 no 2 pp 486ndash489 2012

[27] G V Glass ldquoTesting Homogeneity of Variancesrdquo AmericanEducational Research Journal vol 3 no 3 pp 187ndash190 1966

[28] H BMann ldquoNonparametric tests against trendrdquo Econometricavol 13 pp 245ndash259 1945

[29] M G Kendall Rank Correlation Method Charles GriffinLondon UK 4th edition 1975

[30] H Turoglu ldquoDetection of Changes on Temperature and Precip-itation Features in Istanbul (Turkey)rdquo Atmospheric and ClimateSciences vol 04 no 04 pp 549ndash562 2014

[31] A Karaburun A Demirci and I-S Suen ldquoImpacts of urbangrowth on forest cover in Istanbul (1987-2007)rdquo EnvironmentalModeling amp Assessment vol 166 no 1-4 pp 267ndash277 2010

[32] K K Karanth LM Curran and J D Reuning-Scherer ldquoVillagesize and forest disturbance in Bhadra Wildlife SanctuaryWestern Ghats Indiardquo Biological Conservation vol 128 no 2pp 147ndash157 2006

[33] G Cakir C Un E Z Baskent S Kose F Sivrikaya andS Keles ldquoEvaluating urbanization fragmentation and landuseland cover change pattern in Istanbul city Turkey from 1971to 2002rdquo Land Degradation amp Development vol 19 no 6 pp663ndash675 2008

[34] R B Myneni F G Hall P J Sellers and A L Marshak ldquoTheinterpretation of spectral vegetation indexesrdquo IEEE Transac-tions on Geoscience and Remote Sensing vol 33 no 2 pp 481ndash486 1995

[35] Y S Unal H Toros A Deniz and S Incecik ldquoInfluence ofmeteorological factors and emission sources on spatial and tem-poral variations of PM10 concentrations in Istanbul metropoli-tan areardquo Atmospheric Environment vol 45 no 31 pp 5504ndash5513 2011

[36] M KaracaM Tayanc andH Toros ldquoEffects of urbanization onclimate of Istanbul and Ankarardquo Atmospheric Environment vol29 no 23 pp 3411ndash3421 1995

[37] Y Ezber O L Sen T Kindap and M Karaca ldquoClimatic effectsof urbanization in Istanbul a statistical and modeling analysisrdquoInternational Journal of Climatology vol 27 no 5 pp 667ndash6792007

[38] H S Park ldquoFeatures of the heat island in seoul and its sur-rounding citiesrdquo Atmospheric Environment (1967) vol 20 no10 pp 1859ndash1866 1986

12 Advances in Meteorology

of 93mm for the periods of 1912ndash1980 and 1981ndash2016 Thesevalues indicated that rainfall has increased at a rate of morethan three times over the last decades than the previous ones(Figures 5(b1) and 5(b2)) Also the number of days with dailyprecipitation greater than 25mm presented an incrementduring the period of 1912ndash2016 although this is not significantat the confidence level of 005 (Figures 5(c1) and 5(c2))The trend of standard deviation in the daily precipitationtime series showed a slowly increasing trend during the totalstudied period (Figures 5(d1) and 5(d2))

The comparison of the results of statistical analysisbetween the daily average rainfall amounts belonging tothe periods of 1912ndash1964 and 1965ndash2016 with those of theperiods of 1912ndash1980 and 1981ndash2016 revealed that there is anincrement of 78mm from the period of 1912ndash1964 to theperiod of 1965ndash2016 and an increment of 38mm from theperiod of 1912ndash1980 to the period of 1981ndash2016 In additionthe analysis of standard deviation exhibited a decrementfrom the period of 1912ndash1964 to the period of 1965ndash2016 andan increment from the period of 1912ndash1980 to the periodof 1981ndash2016 (Table 2) The analysis of monthly averageprecipitation time series showed that the highest incrementhas happened in October with 227mm for the sectionperiods of 1912ndash1964 and 1965ndash2016 and with 347mm forthe section periods of 1912ndash1980 and 1981ndash2016 On the otherhand the highest decrease took placewith a value ofminus04mmduring both May and July for the periods of 1912ndash1964 and1965ndash2016 and with a value of minus16mm in September for thesection periods of 1912ndash1980 and 1981ndash2016 The analysis ofseasonal average precipitation time series showed that thehighest increase occurred in autumn with a value of 72mmfor the section periods of 1912ndash1964 and 1965ndash2016 as wellas with a value of 64mm for the periods of 1912ndash1980 and1981ndash2016 respectively There is no remarkable decreasingchange seasonally except in winter season during the periodsof 1912ndash1980 and 1981ndash2016 Furthermore the percentilethresholds of daily average precipitation indicated that thereis an insignificant negative trend based on all percentilesand for all studied periods Overall the statistical analysisshowed that the total average precipitation of Istanbul hasincreased while this increasing trend is more pronounced inthe previous decades than the last 3 decades On the otherhand the increasing rate of daily maximum precipitation ismore evident during the last 3 decades than the previousdecades which can be proven by the increasing frequency ofheavy rainfall events in Istanbul

Generally the results of trend analysis of Kandilli stationduring the last 105 years of 1912ndash2016 showed that there isa warming significant trend in the precipitation time seriesby using both methods of linear regression analysis andMK trend test On the contrary previous climate studiesconducted over Turkey put forward that there has been adecreasing trend in annual precipitation time series duringthe recent decades regionally The results of a previouslyconducted study by using the daily precipitation and tem-perature data sets of Florya and Goztepe meteorologicalstations in Istanbul area between 1960 and 2013 showedthat most notably the precipitation during the warm periodshas decreased but the frequency of the intense rain has

increased and the majority of these episodes of intense raincoincided with the warm periods Other determinationswere the rise in the annual average temperature and theextension of the warm periods in a year This differentiationof the temperature features can lead to the aggravation of theevaporation and it can be effective for a longer period duringthe year [15] Thus it will make Istanbul be confronted withthe much more important problems of water managementand flood [34] Also the results of the current study for trendanalysis in the long period from 1912 to 2016 showed that themost striking spell is between the years of 1968 and 1998 dueto the existence of least number of rainfall events in IstanbulIt can be owing to industrialization along with the increasingair pollution as well as irregular urbanization in Istanbul areaIn this case severe droughts taking place during the yearsof 1988 1992 1993 and 2008 have threatened the reservoirswhich supplied fresh water of the city These years werecharacterized by not having enough rainfall events Theseyears are also characterized by more persistent high-pressuresystems and less occurrence of low-pressure systems in termsof number and strength As statements made by officialinstitutions the formation conditions for the atmosphericlayer of air pollution due to air pollutant emissions from fossilfuel combustion and industrial activities are more providedduring anticyclone or high-pressure system eventsThereforethis leads to warming up and generating an inversion layerin the boundary layer of atmosphere especially over city area[35] The inversion layer or urban heat island intensity isincreasing with the increasing city size andor populationa phenomenon that was also reported by others [36ndash38]Moreover these last climatic events also have affected somepolitical results beyond natural effects The most obviousexample of this is related to the local election of 1994 inIstanbul In this case extreme drought during the summers of1992 and 1993 has caused groundwater reservoirs to dry up inthe city discontinuance of water was experienced for severaldays or even weeks In those times the mayor of Istanbul cityhas lost the election of 1994 and he realized that this result wasdue to the peoplersquos reaction about the water shortages Manypolitical reports of that period also support this scientificview

5 Conclusions

Statistical analysis in temperature time series of the Kandillistation from 1912 to 2016 established that there is a notableincrease in temperature values after the 1940s which is inparallel with the beginning of industrialization era in Istan-bul There has been a rise about 094∘C in the daily averagetemperature series since the beginning of the last century Asignificant positive trend in the daily maximum temperatureseries is found about 156∘C Also there is a positive trendabout 087∘C in the daily minimum temperature series Onthe other hand analysis of the number of days with thedaily maximum temperature higher than 30∘C showed thatthere is an increasing trend Meanwhile analysis of thenumber of days with daily minimum temperatures lowerthan 0∘C showed a decreasing trend The increment in theminimum values of the daily minimum temperature series

Advances in Meteorology 13

is more evident than the maximum values of this series Inthis case these rising rates in the minimum temperatureseries are more evident for the section periods of 1912ndash1980and 1981ndash2016 than the section periods of 1912ndash1964 and1965ndash2016 This again shows that there is an increment inthe positive temperature trend from past to present decadesThe increment in the precipitation time series is not asclear as the increment in the temperature time series dueto periodic variability The trend analysis in the total annualprecipitation time series showed that the first significantupward trend has periodically been started from the 1920swhile there is a stable trend from 2001 till 2016 The dailyaverage of rainfall amounts has increased with a value of58mm during the period of 1912ndash2016 Also the analysisof heavy precipitation trend showed an increase of 61mmOverall the total average precipitation of Istanbul increasedwhile this increasing trend is more pronounced during theearly decades than the last 3 decades On the other handthe increasing rate of daily maximum precipitation is morepronounced in the last 3 decades than the previous decadesThen it was shown that the frequency of heavy rainfallat Istanbul has increased during the recent decades Thusthe precipitation changes in Istanbul have some differencescompared to the general tendency in precipitation trendthat was put forward by other studies as a decreasing trendover the whole of Turkey This result can be expressedas a positive effect of population overgrowth of Istanbulmegacity Comparison of the results in the first half of thestudy period (1912ndash1964) with the second half of the studyperiod (1965ndash2016) showed that both the average temperatureand average precipitation have higher values of 139∘C and878mm for the final phase compared to the values of 136∘Cand 799mm belonging to the initial phase Therefore it canbe stated that the megacity of Istanbul is directly affectedby the climate change and its consequences In this contextpotential risks of climate change in Istanbul megacity underhigher temperature conditions can be expressed as the rise inthe sea level increase in the rate of evapotranspiration andincrease in the frequency of heavy rainfall Also this city maynot be able to handle this uncontrolled population growthand its associated irreversible changes which is alreadypushing the natural limits by destroying the environmentTherefore the local governors of any megacity like Istanbulshould give more emphasis on the importance of sustainableurban development Thus it is urgent to prepare local andnational climate change strategies and action plans for themegacities

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

The authors are grateful to the Bogazici University theobservatory of Kandilli weather station and the Earthquake

Research Institute for providing the research data and tech-nical support The authors also gratefully acknowledge con-tributions of Assoc Professor Dr Yuksel Demirkaya Schoolof Social Sciences Marmara University This work has beensupported by Scientific and Technological Research Councilof Turkey (TUBITAK) under Grants 113R019 and 106Y258and by Marmara University (BAPKO) with projects FEN-E-120314-0066 FEN-C-YLP-090414-0102 FEN-L-250416-0180 and FEN-A-100413-0127

References

[1] D R Easterling B Horton P D Jones et al ldquoMaximum andminimum temperature trends for the globerdquo Science vol 277no 5324 pp 364ndash367 1997

[2] IPCC Climate Change 2014 IPCC Fifth Assessment Synthe-sis Report-Summary for Policymakers-an Assessment of Inter-Governmental Panel on Climate Change Cambridge UniversityPress Cambridge UK 2014

[3] J Carmin N Nadkarni and C Rhie Progress and Challenges inUrban Climate Adaptation Planning Results of a Global SurveyMIT Cambridge UK 2012

[4] A F Young ldquoUrban expansion and environmental risk in theSao Paulo Metropolitan Areardquo Climate Research vol 57 no 1pp 73ndash80 2013

[5] P Tian X Mu J Liu J Hu and C Gu ldquoImpacts of ClimateVariability and Human Activities on the Changes of Runoff andSediment Load in a Catchment of the Loess Plateau ChinardquoAdvances inMeteorology vol 2016 Article ID 4724067 15 pages2016

[6] R S Kovats and K L Ebi ldquoHeatwaves and public health inEuroperdquo European Journal of Public Health vol 16 no 6 pp592ndash599 2006

[7] S Conti P Meli G Minelli et al ldquoEpidemiologic studyof mortality during the Summer 2003 heat wave in ItalyrdquoEnvironmental Research vol 98 no 3 pp 390ndash399 2005

[8] J Kysely and J Kim ldquoMortality during heat waves in SouthKorea 1991 to 2005 how exceptional was the 1994 heat waverdquoClimate Research vol 38 no 2 pp 105ndash116 2009

[9] B Yan Z Xia F Huang L Guo and X Zhang ldquoClimatechange detection and annual extreme temperature analysis ofthe amur river basinrdquoAdvances inMeteorology vol 2016 ArticleID 6268938 14 pages 2016

[10] E M Fischer and R Knutti ldquoAnthropogenic contribution toglobal occurrence of heavy-precipitation and high-temperatureextremesrdquo Nature Climate Change vol 5 no 6 pp 560ndash5642015

[11] X Zhang L Alexander G C Hegerl et al ldquoIndices for moni-toring changes in extremes based on daily temperature andprecipitation datardquo Climate Change vol 2 no 6 pp 851ndash8702011

[12] K H Schlunzen P Hoffmann G Rosenhagen and W RieckeldquoLong-term changes and regional differences in temperatureand precipitation in the metropolitan area of Hamburgrdquo Inter-national Journal of Climatology vol 30 no 8 pp 1121ndash11362010

[13] G Bartolini M Morabito A Crisci et al ldquoRecent trends inTuscany (Italy) summer temperature and indices of extremesrdquoInternational Journal of Climatology vol 28 no 13 pp 1751ndash1760 2008

14 Advances in Meteorology

[14] S C Sheridan and T J Dolney ldquoHeat mortality and levelof urbanization measuring vulnerability across Ohio USArdquoClimate Research vol 24 no 3 pp 255ndash265 2003

[15] M Tayanc U Im M Dogruel andM Karaca ldquoClimate changein Turkey for the last half centuryrdquo Climatic Change vol 94 no3-4 pp 483ndash502 2009

[16] H Toros ldquoSpatio-temporal variation of daily extreme tempera-tures over Turkeyrdquo International Journal of Climatology vol 32no 7 pp 1047ndash1055 2012

[17] H Toros ldquoSpatio-temporal precipitation change assessmentsover Turkeyrdquo International Journal of Climatology vol 32 no9 pp 1310ndash1325 2012

[18] M Turkes C Yozgatlıgil I Batmaz et al ldquoHas the climate beenchanging in Turkey Regional climate change signals based on acomparative statistical analysis of two consecutive time periods1950-1980 and 1981-2010rdquoClimate Research vol 70 no 1 pp 77ndash93 2016

[19] IPCC ldquoummary for Policymakers In Climate Change 2013rdquo inThe Physical Science Basis The contribution of Working Group Ito the Fifth Assessment Report of the Intergovernmental Panel onClimate Change Cambridge University Press Cambridge UK2013

[20] United Nations ldquoDepartment of Economic and Social AffairsPopulation Division 2006 World Urbanization Prospects The2005 Revisionrdquo Working Paper ESAPWP200 2011

[21] Y Demirkaya Sayılarla Istanbul ITO Istanbul Turkey 2011[22] TUIK ldquoAddress based population registration system results of

2014rdquo Turkish Statistical Institute (TUIK) 2017 httpraporytuikgovtr10-03-2015-184727-842632346446643191142126876html

[23] S Erinc Climatology and its Methods Alfa Basım YayımDagitim Istanbul Turkey 4th edition 1965

[24] O M Gokturk D Bozkurt O L Sen and M Karaca ldquoQualitycontrol and homogeneity of Turkish precipitation datardquoHydro-logical Processes vol 22 no 16 pp 3210ndash3218 2008

[25] C Ley and D Paindaveine ldquoRuns Testsrdquo in Encyclopedia ofEnvironmetrics 2012

[26] A Ghasemi and S Zahediasl ldquoNormality tests for statisticalanalysis a guide for non-statisticiansrdquo International Journal ofEndocrinology andMetabolism vol 10 no 2 pp 486ndash489 2012

[27] G V Glass ldquoTesting Homogeneity of Variancesrdquo AmericanEducational Research Journal vol 3 no 3 pp 187ndash190 1966

[28] H BMann ldquoNonparametric tests against trendrdquo Econometricavol 13 pp 245ndash259 1945

[29] M G Kendall Rank Correlation Method Charles GriffinLondon UK 4th edition 1975

[30] H Turoglu ldquoDetection of Changes on Temperature and Precip-itation Features in Istanbul (Turkey)rdquo Atmospheric and ClimateSciences vol 04 no 04 pp 549ndash562 2014

[31] A Karaburun A Demirci and I-S Suen ldquoImpacts of urbangrowth on forest cover in Istanbul (1987-2007)rdquo EnvironmentalModeling amp Assessment vol 166 no 1-4 pp 267ndash277 2010

[32] K K Karanth LM Curran and J D Reuning-Scherer ldquoVillagesize and forest disturbance in Bhadra Wildlife SanctuaryWestern Ghats Indiardquo Biological Conservation vol 128 no 2pp 147ndash157 2006

[33] G Cakir C Un E Z Baskent S Kose F Sivrikaya andS Keles ldquoEvaluating urbanization fragmentation and landuseland cover change pattern in Istanbul city Turkey from 1971to 2002rdquo Land Degradation amp Development vol 19 no 6 pp663ndash675 2008

[34] R B Myneni F G Hall P J Sellers and A L Marshak ldquoTheinterpretation of spectral vegetation indexesrdquo IEEE Transac-tions on Geoscience and Remote Sensing vol 33 no 2 pp 481ndash486 1995

[35] Y S Unal H Toros A Deniz and S Incecik ldquoInfluence ofmeteorological factors and emission sources on spatial and tem-poral variations of PM10 concentrations in Istanbul metropoli-tan areardquo Atmospheric Environment vol 45 no 31 pp 5504ndash5513 2011

[36] M KaracaM Tayanc andH Toros ldquoEffects of urbanization onclimate of Istanbul and Ankarardquo Atmospheric Environment vol29 no 23 pp 3411ndash3421 1995

[37] Y Ezber O L Sen T Kindap and M Karaca ldquoClimatic effectsof urbanization in Istanbul a statistical and modeling analysisrdquoInternational Journal of Climatology vol 27 no 5 pp 667ndash6792007

[38] H S Park ldquoFeatures of the heat island in seoul and its sur-rounding citiesrdquo Atmospheric Environment (1967) vol 20 no10 pp 1859ndash1866 1986

Advances in Meteorology 13

is more evident than the maximum values of this series Inthis case these rising rates in the minimum temperatureseries are more evident for the section periods of 1912ndash1980and 1981ndash2016 than the section periods of 1912ndash1964 and1965ndash2016 This again shows that there is an increment inthe positive temperature trend from past to present decadesThe increment in the precipitation time series is not asclear as the increment in the temperature time series dueto periodic variability The trend analysis in the total annualprecipitation time series showed that the first significantupward trend has periodically been started from the 1920swhile there is a stable trend from 2001 till 2016 The dailyaverage of rainfall amounts has increased with a value of58mm during the period of 1912ndash2016 Also the analysisof heavy precipitation trend showed an increase of 61mmOverall the total average precipitation of Istanbul increasedwhile this increasing trend is more pronounced during theearly decades than the last 3 decades On the other handthe increasing rate of daily maximum precipitation is morepronounced in the last 3 decades than the previous decadesThen it was shown that the frequency of heavy rainfallat Istanbul has increased during the recent decades Thusthe precipitation changes in Istanbul have some differencescompared to the general tendency in precipitation trendthat was put forward by other studies as a decreasing trendover the whole of Turkey This result can be expressedas a positive effect of population overgrowth of Istanbulmegacity Comparison of the results in the first half of thestudy period (1912ndash1964) with the second half of the studyperiod (1965ndash2016) showed that both the average temperatureand average precipitation have higher values of 139∘C and878mm for the final phase compared to the values of 136∘Cand 799mm belonging to the initial phase Therefore it canbe stated that the megacity of Istanbul is directly affectedby the climate change and its consequences In this contextpotential risks of climate change in Istanbul megacity underhigher temperature conditions can be expressed as the rise inthe sea level increase in the rate of evapotranspiration andincrease in the frequency of heavy rainfall Also this city maynot be able to handle this uncontrolled population growthand its associated irreversible changes which is alreadypushing the natural limits by destroying the environmentTherefore the local governors of any megacity like Istanbulshould give more emphasis on the importance of sustainableurban development Thus it is urgent to prepare local andnational climate change strategies and action plans for themegacities

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

The authors are grateful to the Bogazici University theobservatory of Kandilli weather station and the Earthquake

Research Institute for providing the research data and tech-nical support The authors also gratefully acknowledge con-tributions of Assoc Professor Dr Yuksel Demirkaya Schoolof Social Sciences Marmara University This work has beensupported by Scientific and Technological Research Councilof Turkey (TUBITAK) under Grants 113R019 and 106Y258and by Marmara University (BAPKO) with projects FEN-E-120314-0066 FEN-C-YLP-090414-0102 FEN-L-250416-0180 and FEN-A-100413-0127

References

[1] D R Easterling B Horton P D Jones et al ldquoMaximum andminimum temperature trends for the globerdquo Science vol 277no 5324 pp 364ndash367 1997

[2] IPCC Climate Change 2014 IPCC Fifth Assessment Synthe-sis Report-Summary for Policymakers-an Assessment of Inter-Governmental Panel on Climate Change Cambridge UniversityPress Cambridge UK 2014

[3] J Carmin N Nadkarni and C Rhie Progress and Challenges inUrban Climate Adaptation Planning Results of a Global SurveyMIT Cambridge UK 2012

[4] A F Young ldquoUrban expansion and environmental risk in theSao Paulo Metropolitan Areardquo Climate Research vol 57 no 1pp 73ndash80 2013

[5] P Tian X Mu J Liu J Hu and C Gu ldquoImpacts of ClimateVariability and Human Activities on the Changes of Runoff andSediment Load in a Catchment of the Loess Plateau ChinardquoAdvances inMeteorology vol 2016 Article ID 4724067 15 pages2016

[6] R S Kovats and K L Ebi ldquoHeatwaves and public health inEuroperdquo European Journal of Public Health vol 16 no 6 pp592ndash599 2006

[7] S Conti P Meli G Minelli et al ldquoEpidemiologic studyof mortality during the Summer 2003 heat wave in ItalyrdquoEnvironmental Research vol 98 no 3 pp 390ndash399 2005

[8] J Kysely and J Kim ldquoMortality during heat waves in SouthKorea 1991 to 2005 how exceptional was the 1994 heat waverdquoClimate Research vol 38 no 2 pp 105ndash116 2009

[9] B Yan Z Xia F Huang L Guo and X Zhang ldquoClimatechange detection and annual extreme temperature analysis ofthe amur river basinrdquoAdvances inMeteorology vol 2016 ArticleID 6268938 14 pages 2016

[10] E M Fischer and R Knutti ldquoAnthropogenic contribution toglobal occurrence of heavy-precipitation and high-temperatureextremesrdquo Nature Climate Change vol 5 no 6 pp 560ndash5642015

[11] X Zhang L Alexander G C Hegerl et al ldquoIndices for moni-toring changes in extremes based on daily temperature andprecipitation datardquo Climate Change vol 2 no 6 pp 851ndash8702011

[12] K H Schlunzen P Hoffmann G Rosenhagen and W RieckeldquoLong-term changes and regional differences in temperatureand precipitation in the metropolitan area of Hamburgrdquo Inter-national Journal of Climatology vol 30 no 8 pp 1121ndash11362010

[13] G Bartolini M Morabito A Crisci et al ldquoRecent trends inTuscany (Italy) summer temperature and indices of extremesrdquoInternational Journal of Climatology vol 28 no 13 pp 1751ndash1760 2008

14 Advances in Meteorology

[14] S C Sheridan and T J Dolney ldquoHeat mortality and levelof urbanization measuring vulnerability across Ohio USArdquoClimate Research vol 24 no 3 pp 255ndash265 2003

[15] M Tayanc U Im M Dogruel andM Karaca ldquoClimate changein Turkey for the last half centuryrdquo Climatic Change vol 94 no3-4 pp 483ndash502 2009

[16] H Toros ldquoSpatio-temporal variation of daily extreme tempera-tures over Turkeyrdquo International Journal of Climatology vol 32no 7 pp 1047ndash1055 2012

[17] H Toros ldquoSpatio-temporal precipitation change assessmentsover Turkeyrdquo International Journal of Climatology vol 32 no9 pp 1310ndash1325 2012

[18] M Turkes C Yozgatlıgil I Batmaz et al ldquoHas the climate beenchanging in Turkey Regional climate change signals based on acomparative statistical analysis of two consecutive time periods1950-1980 and 1981-2010rdquoClimate Research vol 70 no 1 pp 77ndash93 2016

[19] IPCC ldquoummary for Policymakers In Climate Change 2013rdquo inThe Physical Science Basis The contribution of Working Group Ito the Fifth Assessment Report of the Intergovernmental Panel onClimate Change Cambridge University Press Cambridge UK2013

[20] United Nations ldquoDepartment of Economic and Social AffairsPopulation Division 2006 World Urbanization Prospects The2005 Revisionrdquo Working Paper ESAPWP200 2011

[21] Y Demirkaya Sayılarla Istanbul ITO Istanbul Turkey 2011[22] TUIK ldquoAddress based population registration system results of

2014rdquo Turkish Statistical Institute (TUIK) 2017 httpraporytuikgovtr10-03-2015-184727-842632346446643191142126876html

[23] S Erinc Climatology and its Methods Alfa Basım YayımDagitim Istanbul Turkey 4th edition 1965

[24] O M Gokturk D Bozkurt O L Sen and M Karaca ldquoQualitycontrol and homogeneity of Turkish precipitation datardquoHydro-logical Processes vol 22 no 16 pp 3210ndash3218 2008

[25] C Ley and D Paindaveine ldquoRuns Testsrdquo in Encyclopedia ofEnvironmetrics 2012

[26] A Ghasemi and S Zahediasl ldquoNormality tests for statisticalanalysis a guide for non-statisticiansrdquo International Journal ofEndocrinology andMetabolism vol 10 no 2 pp 486ndash489 2012

[27] G V Glass ldquoTesting Homogeneity of Variancesrdquo AmericanEducational Research Journal vol 3 no 3 pp 187ndash190 1966

[28] H BMann ldquoNonparametric tests against trendrdquo Econometricavol 13 pp 245ndash259 1945

[29] M G Kendall Rank Correlation Method Charles GriffinLondon UK 4th edition 1975

[30] H Turoglu ldquoDetection of Changes on Temperature and Precip-itation Features in Istanbul (Turkey)rdquo Atmospheric and ClimateSciences vol 04 no 04 pp 549ndash562 2014

[31] A Karaburun A Demirci and I-S Suen ldquoImpacts of urbangrowth on forest cover in Istanbul (1987-2007)rdquo EnvironmentalModeling amp Assessment vol 166 no 1-4 pp 267ndash277 2010

[32] K K Karanth LM Curran and J D Reuning-Scherer ldquoVillagesize and forest disturbance in Bhadra Wildlife SanctuaryWestern Ghats Indiardquo Biological Conservation vol 128 no 2pp 147ndash157 2006

[33] G Cakir C Un E Z Baskent S Kose F Sivrikaya andS Keles ldquoEvaluating urbanization fragmentation and landuseland cover change pattern in Istanbul city Turkey from 1971to 2002rdquo Land Degradation amp Development vol 19 no 6 pp663ndash675 2008

[34] R B Myneni F G Hall P J Sellers and A L Marshak ldquoTheinterpretation of spectral vegetation indexesrdquo IEEE Transac-tions on Geoscience and Remote Sensing vol 33 no 2 pp 481ndash486 1995

[35] Y S Unal H Toros A Deniz and S Incecik ldquoInfluence ofmeteorological factors and emission sources on spatial and tem-poral variations of PM10 concentrations in Istanbul metropoli-tan areardquo Atmospheric Environment vol 45 no 31 pp 5504ndash5513 2011

[36] M KaracaM Tayanc andH Toros ldquoEffects of urbanization onclimate of Istanbul and Ankarardquo Atmospheric Environment vol29 no 23 pp 3411ndash3421 1995

[37] Y Ezber O L Sen T Kindap and M Karaca ldquoClimatic effectsof urbanization in Istanbul a statistical and modeling analysisrdquoInternational Journal of Climatology vol 27 no 5 pp 667ndash6792007

[38] H S Park ldquoFeatures of the heat island in seoul and its sur-rounding citiesrdquo Atmospheric Environment (1967) vol 20 no10 pp 1859ndash1866 1986

14 Advances in Meteorology

[14] S C Sheridan and T J Dolney ldquoHeat mortality and levelof urbanization measuring vulnerability across Ohio USArdquoClimate Research vol 24 no 3 pp 255ndash265 2003

[15] M Tayanc U Im M Dogruel andM Karaca ldquoClimate changein Turkey for the last half centuryrdquo Climatic Change vol 94 no3-4 pp 483ndash502 2009

[16] H Toros ldquoSpatio-temporal variation of daily extreme tempera-tures over Turkeyrdquo International Journal of Climatology vol 32no 7 pp 1047ndash1055 2012

[17] H Toros ldquoSpatio-temporal precipitation change assessmentsover Turkeyrdquo International Journal of Climatology vol 32 no9 pp 1310ndash1325 2012

[18] M Turkes C Yozgatlıgil I Batmaz et al ldquoHas the climate beenchanging in Turkey Regional climate change signals based on acomparative statistical analysis of two consecutive time periods1950-1980 and 1981-2010rdquoClimate Research vol 70 no 1 pp 77ndash93 2016

[19] IPCC ldquoummary for Policymakers In Climate Change 2013rdquo inThe Physical Science Basis The contribution of Working Group Ito the Fifth Assessment Report of the Intergovernmental Panel onClimate Change Cambridge University Press Cambridge UK2013

[20] United Nations ldquoDepartment of Economic and Social AffairsPopulation Division 2006 World Urbanization Prospects The2005 Revisionrdquo Working Paper ESAPWP200 2011

[21] Y Demirkaya Sayılarla Istanbul ITO Istanbul Turkey 2011[22] TUIK ldquoAddress based population registration system results of

2014rdquo Turkish Statistical Institute (TUIK) 2017 httpraporytuikgovtr10-03-2015-184727-842632346446643191142126876html

[23] S Erinc Climatology and its Methods Alfa Basım YayımDagitim Istanbul Turkey 4th edition 1965

[24] O M Gokturk D Bozkurt O L Sen and M Karaca ldquoQualitycontrol and homogeneity of Turkish precipitation datardquoHydro-logical Processes vol 22 no 16 pp 3210ndash3218 2008

[25] C Ley and D Paindaveine ldquoRuns Testsrdquo in Encyclopedia ofEnvironmetrics 2012

[26] A Ghasemi and S Zahediasl ldquoNormality tests for statisticalanalysis a guide for non-statisticiansrdquo International Journal ofEndocrinology andMetabolism vol 10 no 2 pp 486ndash489 2012

[27] G V Glass ldquoTesting Homogeneity of Variancesrdquo AmericanEducational Research Journal vol 3 no 3 pp 187ndash190 1966

[28] H BMann ldquoNonparametric tests against trendrdquo Econometricavol 13 pp 245ndash259 1945

[29] M G Kendall Rank Correlation Method Charles GriffinLondon UK 4th edition 1975

[30] H Turoglu ldquoDetection of Changes on Temperature and Precip-itation Features in Istanbul (Turkey)rdquo Atmospheric and ClimateSciences vol 04 no 04 pp 549ndash562 2014

[31] A Karaburun A Demirci and I-S Suen ldquoImpacts of urbangrowth on forest cover in Istanbul (1987-2007)rdquo EnvironmentalModeling amp Assessment vol 166 no 1-4 pp 267ndash277 2010

[32] K K Karanth LM Curran and J D Reuning-Scherer ldquoVillagesize and forest disturbance in Bhadra Wildlife SanctuaryWestern Ghats Indiardquo Biological Conservation vol 128 no 2pp 147ndash157 2006

[33] G Cakir C Un E Z Baskent S Kose F Sivrikaya andS Keles ldquoEvaluating urbanization fragmentation and landuseland cover change pattern in Istanbul city Turkey from 1971to 2002rdquo Land Degradation amp Development vol 19 no 6 pp663ndash675 2008

[34] R B Myneni F G Hall P J Sellers and A L Marshak ldquoTheinterpretation of spectral vegetation indexesrdquo IEEE Transac-tions on Geoscience and Remote Sensing vol 33 no 2 pp 481ndash486 1995

[35] Y S Unal H Toros A Deniz and S Incecik ldquoInfluence ofmeteorological factors and emission sources on spatial and tem-poral variations of PM10 concentrations in Istanbul metropoli-tan areardquo Atmospheric Environment vol 45 no 31 pp 5504ndash5513 2011

[36] M KaracaM Tayanc andH Toros ldquoEffects of urbanization onclimate of Istanbul and Ankarardquo Atmospheric Environment vol29 no 23 pp 3411ndash3421 1995

[37] Y Ezber O L Sen T Kindap and M Karaca ldquoClimatic effectsof urbanization in Istanbul a statistical and modeling analysisrdquoInternational Journal of Climatology vol 27 no 5 pp 667ndash6792007

[38] H S Park ldquoFeatures of the heat island in seoul and its sur-rounding citiesrdquo Atmospheric Environment (1967) vol 20 no10 pp 1859ndash1866 1986

Submit your manuscripts athttpswwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ClimatologyJournal of

EcologyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

EarthquakesJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mining

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 201

International Journal of

OceanographyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of Computational Environmental SciencesHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal ofPetroleum Engineering

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

GeochemistryHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Atmospheric SciencesInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

OceanographyHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MineralogyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MeteorologyAdvances in

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Paleontology JournalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Geological ResearchJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Geology Advances in