historicaltrendofhourlyextremerainfallinpeninsularmalaysia · 3 2719001 setor jps sikamat seremban...

ORIGINAL PAPER

A. H. Syafrina & M. D. Zalina & L. Juneng

Received: 9 May 2013 /Accepted: 27 March 2014 /Published online: 15 May 2014# The Author(s) 2014. This article is published with open access at Springerlink.com

Abstract Hourly rainfall data between the years 1975 and2010 across the Peninsular Malaysia were analyzed for trendsin hourly extreme rainfall events. The analyses were conduct-ed on rainfall occurrences during the northeast monsoon (No-vember–February) known as NEM, the southwest monsoon(May–August) known as SWM, and the two inter-monsoonseasons, i.e., March–April (MA) and September–October(SO). Several extreme rainfall indices were calculated at thestation level. The extreme rainfall events in Peninsular Ma-laysia showed an increasing trend between the years 1975 and2010. The trend analysis was conducted using linear regres-sion; no serial correlation was detected from the Durbin-Watson test. Ordinary kriging was used to determine thespatial patterns of trends in seasonal extremes. The totalamount of rainfall received during NEM is higher comparedto rainfall received during inter-monsoon seasons. However,intense rainfall is observed during the inter-monsoon seasonwith higher hourly total amount of rainfall. The eastern part ofpeninsular was most affected by stratiform rains, while con-vective rain contributes more precipitation to areas in thewestern part of the peninsular. From the distribution of spatialpattern of trend, the extreme frequency index (Freq >20) givessignificant contribution to the positive extreme rainfall trendduring the monsoon seasons. Meanwhile, both extreme

frequency and extreme intensity (24-Hr Max, Freq >95th,Tot >95th, Tot >99th, and Hr Max) indices give significantcontribution to the positive extreme rainfall trend during theinter-monsoon seasons. Most of the significant extreme indi-ces showed the positive sign of trends. However, a negativetrend of extreme rainfall was found in the northwest coast dueto the existence of Titiwangsa Range. The extreme intensity,extreme frequency, and extreme cumulative indices showedincreasing trends during the NEM and MA while extremeintensity and extreme frequency had similar trends duringthe SWM and SO throughout Peninsular Malaysia. Overall,the hourly extreme rainfall events in Peninsular Malaysiashowed an increasing trend between the year 1975 and 2010with notable increasing trends in short temporal rainfall duringinter-monsoon season. The result also proves that convectiverain during this period contributes higher intensity rains whichcan only be captured using short duration rainfall series.

1 Introduction

Recent increases in the frequency and intensity of extremerainfall events have raised concern that human activity mighthave resulted in an alteration of the climate system. It isbelieved that rise in both frequency and intensity of extremerainfall events are the major impacts of global warming (SenRoy 2009; Cheng et al. 2012). Intense rainfall occurrences inshort temporal scales or persistent rainfall over long period oftime often lead to massive floods resulting in hazardoussituations. Peninsular Malaysia experiences unpredictablerainfall events, which causes havoc, and fixing them costsmillions of Malaysian ringgit. The increase in massive floodcases, including flash flood and landslides in the last decade, isdue to the increase in rainfall intensities.

Several factors such as urban heat island (UHI) effect andlocal temperature changes contribute to the extreme rainfall

A. H. Syafrina (*) :M. D. ZalinaUTM Razak School of Engineering and Advanced Technology,Universiti Teknologi Malaysia, Jalan Semarak, 54100 KualaLumpur, Malaysiae-mail: [email protected]

M. D. Zalinae-mail: [email protected]

L. JunengSchool of Environment Natural Resources, Faculty of Science andTechnology, Universiti Kebangsaan Malaysia UKM, 43600 Bangi,Selangor, Malaysiae-mail: [email protected]

Theor Appl Climatol (2015) 120:259–285DOI 10.1007/s00704-014-1145-8

Historical trend of hourly extreme rainfall in Peninsular Malaysia

events. Large-scale climatic fluctuations, such as the positionof the Madden-Jullian oscillation (MJO) and the El-Ninosouthern oscillation (ENSO) phases, may also modulate thefrequency and amount of precipitation on a wet day (Cayanet al. 1999). For instance in Malaysia, the occurrence of thetwo most extreme ENSOs in 1982/1983 and 1997/1998 werereported to be the factor responsible for extreme climatechanges in Peninsular Malaysia (Zin et al. 2010). However,due to local factors too, extreme rainfall events may presentcontrasting behavior in different regions across Malaysia. Anumber of studies involving the determination of extremerainfall trend were done using daily rainfall data with differentintensity categories defining extreme rainfall events. For in-stance, Griffiths et al. (2003) investigated the relationshipbetween the South Pacific Convergence Zone and extremerainfall trend using indices which considered both extremefrequencies and intensities in the tropical South Pacific.Correlation between indices and the total rain was alsodetermined in this study. Schmidli and Frei (2005) foundincreasing trends in heavy precipitation during autumn andwinter in Switzerland over the period 1901–2000. In otherregion, Haylock and Nicholls (2000) found decreasing trendin extreme frequency and extreme intensity in southwestWestern Australia, while an increasing trend was found inextreme precipitation of eastern Australia.

These findings agree with several studies done throughoutthe Asian region particularly in Malaysia. Suhaila et al. (2010)found increasing trends in both the total amount of rainfall andthe frequency of wet days during the northeast monsoon, whichgive rise to the increasing trend of rainfall intensity for the periodof 1971–2004. Similarly, increasing trends in the extreme inten-sity indices also were found by Zin et al. (2010) for the period of1975–2004. However, both studies have found a significantdecrease in the number of wet days during this period. Besidesextreme rainfall trends, extreme temperature and dry spell trendsare also being studied in some regions. Alexander and Arblaster(2009) found increasing trends in temperature extremes,particularly a significant increase in the number of warmnights and heat waves with much longer dry spellsinterspersed with periods of increased extreme precipitationover Australia. Studies in Peninsular Malaysia found that mostof the dry spell indices depicted downward trends during thenortheast monsoon over the years of 1975 to 2004.

Analyzing seasonal hourly trend of extreme rainfall eventswill give a better insight to extreme rainfall behavior com-pared to daily trends. Haylock and Nicholls (2000) argued thatanalysis of extreme rainfall, based on total daily rainfall isproblematic due to varying quality of the data. In 2009, SenRoy has proven that hourly data indicated more detailedresults, where the trend of maximum hourly rainfall in theeastern part of the Gangetic Plain and Uttaranchal showed anincreasing trend during winter, dry-summer, and wet-summermonsoon season, which contrasted an earlier study using daily

rainfall showing a decreasing trend of daily extreme rainfall inthese same areas (Sen Roy and Balling 2004). Similar studieson daily rainfall data was also conducted in Malaysia by Zinet al. (2010). Increasing trends in several extreme indices werereported. However, analyzing hourly rainfall could give abetter indication of the seasonal contribution to the annualextreme rainfall as mentioned earlier. Hence, this study pro-poses to (1) analyze the trends of extreme hourly rainfallevents across PeninsularMalaysia, and (2) describe the profileof hourly extreme rainfall in Peninsular Malaysia. The use ofhourly rainfall with a large number of indices and over a longhistorical period will produce a more accurate account of thespatial distribution and trend of extreme rainfall in Malaysia.

2 Data

Peninsular Malaysia is located between 1° and 6° N in thenorthern latitude and between 100° to 103° E longitude. Thesurface climate is influenced by the NEM between Novemberand February and by SWM between May and August. TheNEMmonsoon is usually associated with heavier rainfall withthe eastern and southern regions being the most affected areas.In between these two monsoons are the inter-monsoon sea-sons occurring inMarch–April (MA) and September–October(SO), which brings intense convective rainfall to the westerncoast of Peninsular Malaysia.

In this study, hourly rainfall data were sourced from theMalaysia Drainage and Irrigation Department (DID). Therainfall stations were selected based on two criteria: (1) ade-quacy of data and length of record and (2) even distribution ofrainfall stations across Peninsular Malaysia. Stations withmissing values, greater than 2 % of the total record hourswithin 1 January 1975 to 31 December 2010, were excluded.The average nearest neighbor (ANN) was used in ensuring thestations chosen are spread evenly over the Peninsular Malay-sia. The average nearest neighbor ratio is given as follows:

ANN ¼ D̄0

D̄E

ð1Þ

where D0 is the observed mean distance between each stationand their nearest neighbor,

D̄0 ¼Xi¼1

n din

ð2Þ

and DE is the expected mean distance for the stations given arandom pattern,

D̄E ¼ 0:5ffiffiffiffiffiffiffiffiffiffin.A

r ð3Þ

260 A.H. Syafrina et al.

di equals the distance between station i and its neareststation, n corresponds to the total number of features, and Ais the total study area. The zANN-score for the statistic iscalculated as follows:

zANN ¼ D̄0−D̄E

SEð4Þ

where

SE ¼ 0:26136ffiffiffiffiffiffiffiffiffiffiffiffin2.A

r ð5Þ

If the zANN-score is less than 1, the stations are clustered.Otherwise, the stations are spread evenly. The calculated zANN-score is 2, which falls into a category of dispersed distributionat 99 % level of significance. Using these criteria, 25 stations(Table 1) were selected for this study. The distribution ofstations is portrayed in Fig. 1.

3 Methodology

In this study, 8 seasonal extreme rainfall indices were exam-ined based on hourly rainfall data at the 25 selected stations.The extreme rainfall indices, chosen after Sen Roy (2009)considered both the intensity and frequency of extreme rain-fall events (Table 2). A rainfall amount of at least 0.1 mm ischosen as a threshold for hourly rainfall. In this study, percen-tiles were used rather than fixed threshold as it is a moreaccurate measure and suitable for a region like PeninsularMalaysia which has highly variable rainfall and high spatialrainfall intensity (Haylock and Nicholls 2000; Griffiths et al.2003). The 95th and 99th percentile was selected as thethreshold to represent extreme rainfall events. All precipita-tion events above 0.1 mm occurring throughout the entirestudy period were sorted in ascending order for each stationseparately, in order to determine the threshold value for the95th and 99th percentile. Next, the total amount and frequencyof events occurring each year above the 95th and 99th per-centile were calculated. Hours with rainfall exceeding 95thpercentile were referred to as very wet and hours with rainfallexceeding 99th percentile as extremely wet.

Table 1 Name of stations withlongitude and latitude No. Station Name of station Long (°) Lat (°)

1 1737001 Sek. Men. Bukit Besar, Kota Tinggi Johor 103.72 1.76

2 2224038 Chin Chin (Tepi Jalan) Melaka 102.49 2.28

3 2719001 Setor JPS Sikamat Seremban 101.54 2.73

4 2815001 Pejabat JPS Sungai Mangga Selangor 101.54 2.82

5 2818110 SMK Bandar Tasik Kesuma, Semenyih Selangor 101.87 2.89

6 2913001 Pusat Kawalan P/S Telok Gong Selangor 101.39 2.93

7 3117070 JPS Ampang, Selangor 101.75 3.15

8 3118102 Sek. Keb. Kg. Lui Selangor 101.87 3.17

9 3314001 Rumah Pam JPS Jaya Setia Selangor 101.41 3.36

10 3411017 Stor JPS, Tg Karang Selangor 101.17 3.42

11 3516022 Loji Air Kuala Kubu Bharu Selangor 101.45 3.57

12 3613004 Ibu Bekalan Sg. Bernam Selangor 101.35 3.69

13 3710006 Rumah Pam JPS Bangunan Terap, Selangor 101.08 3.72

14 4908018 Pusat Kesihatan Kecil, Batu Kurau Perak 100.80 4.97

15 2831179 Kg. Kedaik, Pahang 103.18 2.88

16 3533102 Rumah Pam Pahang Tua, Pekan 103.35 3.56

17 3924072 Rumah Pam Paya Kangsar, Pahang 102.43 3.90

18 4234109 JPS Kemaman, Terengganu 103.42 4.23

19 4734079 Sek. Men. Sultan Omar Dungun, Terengganu 103.41 4.76

20 4819027 Gua Musang, Kelantan 101.99 4.87

21 4930038 Kg. Menerong, Terengganu 103.06 4.94

22 5331048 Setor JPS, Kuala Terengganu 103.13 5.31

23 5504035 Lahar Ikan Mati Kepala Batas, Penang 100.43 5.53

24 6207032 Ampang Pedu, Kedah 100.77 6.24

25 6401002 Padang Katong, Kangar 100.18 6.44

Historical trend of hourly extreme rainfall 261

The trends for each of these 32 indices (8 extremerainfall indices for 4 seasons) were calculated usingleast square linear regression analysis method at eachstation level. For each station, a matrix of 36 rows, 1for each year from 1975 to 2010, and 9 columnsrepresenting the year and each of the 8 variables wasdeveloped. Simple linear regression (LS) analysis withthe year of record serving as independent variable wasestablished to obtain the standardized regression coeffi-cient, representing the strength and sign of any trend,for each of the eight variables. The significance of thislinear fit was tested using t test at 5 % level of signif-icance. Another test to detect the trend within the timeseries is the nonparametric Mann-Kendall (MK) test.The Mann-Kendall test is based on the statistic S. Each

pair of observed values yi, yj (i> j) of the randomvariable is inspected to find out whether yi>yj or yi<yj. P isthe number of the former type of pairs, and M is thenumber of the latter type of pairs. Then S is defined asfollows:

S ¼ P−M ð6Þ

For n>10, the sampling distribution of S is as follows. Zfollows the standard normal distribution where

Z ¼S−1ð Þ

. ffiffiffiffiffiffiffiffiffiffiffiffiffiffiVar Sð Þ

pif S > 0;

0 if S ¼ 0;

S þ 1ð Þ. ffiffiffiffiffiffiffiffiffiffiffiffiffiffi

Var Sð Þp

if S < 0

8>><>>: ð7Þ

4908018

2831179

2224038

1737001

6401002

6207032

5504035

5331048

49300384819027

4734079

4234109

3924072

37100063613004 35331023516022

34110173314001

31181023117070

2913001 28181102815001 2719001

0 0.4 0.8 1.2 1.60.2Decimal Degrees

Legend

Rainfall stations

STRAITS OF M ALACCA

THAILAND

MALAYSIA SOUTHCHINA

SEA

Fig. 1 Distribution of stationsused in this study


There is a correction for ties when yi=yj. The null hypoth-esis of no trend is rejected when the computed Z value isgreater than zα in absolute value. These were done separatelyaccording to four different seasons of Peninsular Malaysia.After that, the field significance test for each extreme index istested using the regional average Mann-Kendall test (RAMK)as suggested by Yue et al. (2002). No correlation exist in therainfall data, hence the regional MK test without spatial andtemporal correlation was used in the study. The sum of MK

statistics of m independent sites is given by SSUM ¼ ∑k¼1

m

Sk .

The standardized RAMK statistic S is

Z̄ ¼S̄−1

� �. ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiVar S̄

� �rif S̄ > 0;

0 if S̄ ¼ 0;

S̄ þ 1� �. ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

Var S̄� �r

if S̄ < 0

8>>>>><>>>>>:

ð8Þ

where S¼ SSUM=m is normally distributed and can be

represented by S e N 1m ∑k¼1

m

μk ; ∑k¼1

m

σ 2k

� �. The sample

mean and variance of the RAMK statistic are E S� � ¼ 0 and

Var S ¼ 1

m2 ∑k¼1

m

Var Skð Þ . Next, ordinary kriging method

was employed to determine the spatial patterns of trends inseasonal extremes. Furthermore, the autocorrelation in the rainfalldata was also tested using the Durbin-Watson (DW) test where

H0:ρ=0 vs. H1:ρ>0. The test statistic is d ¼∑i¼2

n

ei−ei−1ð Þ2

∑i¼1

ne 2i

where ei ¼ yi−byi and yi andbyi are the observed and predictedvalues of the response variable for individual i. The DW-statistic for the data was found to be approximately 2, which

indicates that there is no autocorrelation in the rainfall data atthe 5 % level of significance.

4 Results and discussion

This section will discuss the spatial profile of extreme rainfallindices as well as the spatial trends of seasonal extremerainfall indices for the whole of Peninsular Malaysia.

4.1 Spatial profile of extreme rainfall indices

As for the extreme intensity and extreme cumulative indices, itis interesting to note that the spatial pattern of rainfall activitiesover the inter-monsoon season, notably MA, shows a consis-tent pattern of higher indices in the western region (Figs. 2, 3,4, 5, and 6). The Hr-Max recorded a higher value in thisseason compared to other seasons with amount of rainfallreaching up to 49.93 mm. This is in agreement to the generalassumption that convective rain occurring during the inter-monsoon period has higher intensities compared to stratiformrain which occur during the monsoon season since the max-imum rainfall depicts higher value in a relatively short time(i.e., hourly) and such results cannot be captured using dailydata as portrayed by the results of Zin et al. (2010), in whichthe highest extreme intensities occurred during NEM season.The spatial pattern of the 5 Hr-Max and 24 Hr-Max followsthe same pattern. However, the Hr-Max presents a differentpattern with higher hourly maximums in the western regionindicating that the NEM brings stratiform rain which are longduration heavy rains. The western regions are affected by theNEM but to a lesser degree. Hence immense floods usuallyoccur in the eastern regions during the NEM. The northernpart of the peninsular receives little rain during the NEM dueto the existence of Titiwangsa Range (Banjaran Titiwangsa)which blocks the region from receiving the rain brought on the

Table 2 Rainfall indices with definitions and units

Definitions Index name Unit

Extreme cumulative rainfall

1. Largest seasonal total amount of rainfall recorded over 24 consecutive hours for each year (24) 24-Hr Max mm/24-h

Extreme intensity

2. Annual seasonal hourly maximum amount of rainfall—the seasonal hourly maximum amountsof rainfall were determined separately for each year.

Hr Max mm/hr

3. Largest seasonal total amount of rainfall recorded over five consecutive hours for each year 5-Hr Max mm/5-h

4. Total amount of rainfall exceeding 95th percentile for each year Tot>99th mm/hr

5. Total amount of rainfall exceeding 99th percentile for each year Tot>95th mm/hr

Extreme frequency

6. Frequency of seasonal rainfall events greater than 20 mm within 1 h for each year Freq>20 hr

7. Frequency of rainfall exceeding 95th percentile within 1 h for each year Freq>95th hr

8. Frequency of rainfall exceeding 99th percentile within 1 h for each year Freq>99th hr


Legend

16.02 - 20.620.6 - 23.823.8 - 28.3928.39 - 34.9534.95 - 44.36 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

21.49 - 27.5927.59 - 33.1733.17 - 38.2638.26 - 43.8343.83 - 49.93 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

32.5 - 36.8336.83 - 39.7239.72 - 41.6541.65 - 44.5444.54 - 48.87 0 0.4 0.8 1.2 1.60.2

Decimal Degrees


Legend

19.47 - 31.4331.43 - 36.336.3 - 38.2738.27 - 43.1443.14 - 55.1

a b

c d

Fig. 2 Spatial distribution of HrMax index for a northeast monsoon season (NEM), b southwest monsoon season (SWM), c inter-monsoon season (MA),and d inter-monsoon season (SO) over 36 years (1975–2010)


Legend

51.07 - 58.2358.23 - 61.0261.02 - 68.1968.19 - 86.6286.62 - 134.01 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

43.18 - 53.7153.71 - 61.7661.76 - 67.9167.91 - 72.672.6 - 76.18 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

16.02 - 37.7937.79 - 51.0651.06 - 59.1659.16 - 64.0964.09 - 72.19 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

7.08 - 37.8537.85 - 54.9954.99 - 64.5564.55 - 69.8769.87 - 79.43 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

a b

c d

Fig. 3 Spatial distribution of 5-Hr Max index for a northeast monsoon season (NEM), b southwest monsoon season (SWM), c inter-monsoon season(MA), and d inter-monsoon season (SO) over 36 years (1975–2010)


Legend

36.13 - 66.3166.31 - 86.4486.44 - 116.62116.62 - 161.85161.85 - 229.66 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

46.15 - 61.561.5 - 69.6769.67 - 74.0374.03 - 82.2182.21 - 97.55 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

30.07 - 50.850.8 - 64.0264.02 - 72.4572.45 - 77.8377.83 - 86.26 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

6.94 - 41.0241.02 - 63.7563.75 - 78.9278.92 - 89.0489.04 - 104.21 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

a b

c d

Fig. 4 Spatial distribution of 24-Hr Max index for a northeast monsoon season (NEM), b southwest monsoon season (SWM), c inter-monsoon season(MA), and d inter-monsoon season (SO) over 36 years (1975–2010)


Legend

101.5 - 154.6154.6 - 221.09221.09 - 304.35304.35 - 408.61408.61 - 491.87 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

123.01 - 157.77157.77 - 182.03182.03 - 216.78216.78 - 266.59266.59 - 337.97 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

55.99 - 84.1484.14 - 102.09102.09 - 130.25130.25 - 174.39174.39 - 243.59 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

100.95 - 117.9117.9 - 126.54126.54 - 143.49143.49 - 176.77176.77 - 242.1 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

a b

c d

Fig. 5 Spatial distribution of total amount for Tot >95th index, for a northeast monsoon season (NEM), b southwest monsoon season (SWM), c inter-monsoon season (MA), and d inter-monsoon season (SO) over 36 years (1975–2010)


Legend

33.74 - 53.52

53.52 - 65.05

65.05 - 84.83

84.83 - 118.76

118.76 - 176.98 0 0.4 0.8 1.2 1.60.2Decimal Degrees

Legend

43.01 - 53.6853.68 - 60.260.2 - 70.8870.88 - 88.3988.39 - 117.11 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

13.04 - 33.1633.16 - 45.4345.43 - 52.9152.91 - 65.1865.18 - 85.31 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

25.86 - 34.0834.08 - 44.3744.37 - 57.2557.25 - 73.3873.38 - 93.58 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

a b

c d

Fig. 6 Spatial distribution of total amount for Tot >99th index for a northeast monsoon season (NEM), b southwest monsoon season (SWM), c inter-monsoon season (MA), and d inter-monsoon season (SO) over 36 years (1975–2010)


Legend

1.86 - 3.793.79 - 5.345.34 - 7.277.27 - 9.699.69 - 12.72 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

2.67 - 3.653.65 - 4.364.36 - 5.345.34 - 6.696.69 - 8.53 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

0.81 - 1.661.66 - 2.162.16 - 3.023.02 - 4.484.48 - 7 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

2.11 - 2.652.65 - 3.013.01 - 3.563.56 - 4.374.37 - 5.58 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

a b

c d

Fig. 7 Spatial distribution of Freq >20 index for a northeast monsoon season (NEM), b southwest monsoon season (SWM), c inter-monsoon season(MA), and d inter-monsoon season (SO) over 36 years (1975–2010)


Legend

4.44 - 6.236.23 - 7.137.13 - 8.918.91 - 12.4112.41 - 19.28 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

4.56 - 5.845.84 - 6.656.65 - 7.937.93 - 9.949.94 - 13.08 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

2.69 - 3.363.36 - 4.194.19 - 5.235.23 - 6.546.54 - 7.58 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

3.94 - 4.484.48 - 4.714.71 - 5.255.25 - 6.516.51 - 9.47 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

a b

c d

Fig. 8 Spatial distribution of total amount for Freq >95th index for a northeast monsoon season (NEM), b southwest monsoon season (SWM), c inter-monsoon season (MA), and d inter-monsoon season (SO) over 36 years (1975–2010)


Legend

0.83 - 1.191.19 - 1.381.38 - 1.741.74 - 2.412.41 - 3.67 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

0.94 - 1.241.24 - 1.461.46 - 1.751.75 - 2.142.14 - 2.64 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

0.33 - 0.780.78 - 1.011.01 - 1.131.13 - 1.361.36 - 1.81 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

0.56 - 0.790.79 - 0.930.93 - 1.161.16 - 1.541.54 - 2.17 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

a b

c d

Fig. 9 Spatial distribution of total amount for Freq >99th index for a northeast monsoon season (NEM), b southwest monsoon season (SWM), c inter-monsoon season (MA), and d inter-monsoon season (SO) over 36 years (1975–2010)


Legend

-0.45 - -0.18-0.18 - 0.010.01 - 0.140.14 - 0.230.23 - 0.37 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.32 - -0.17-0.17 - -0.02-0.02 - 0.140.14 - 0.310.31 - 0.48 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.42 - -0.12-0.12 - 0.040.04 - 0.130.13 - 0.290.29 - 0.6 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.24 - -0.14-0.14 - -0.03-0.03 - 0.10.1 - 0.260.26 - 0.46 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

a b

c d

Fig. 10 Spatial pattern of trends for Hr Max for a northeast monsoon season (NEM), b southwest monsoon season (SWM), c inter-monsoon season(MA), and d inter-monsoon season (SO) over 36 years (1975–2010)


Legend

-0.35 - -0.15-0.15 - 0.030.03 - 0.170.17 - 0.290.29 - 0.43 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.38 - -0.18-0.18 - 00 - 0.170.17 - 0.330.33 - 0.5 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.51 - -0.14-0.14 - 00 - 0.050.05 - 0.190.19 - 0.57 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.32 - -0.19-0.19 - -0.08-0.08 - 0.050.05 - 0.20.2 - 0.38 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

a b

c d

Fig. 11 Spatial pattern of trends for 5-Hr Max for a northeast monsoon season (NEM), b southwest monsoon season (SWM), c inter-monsoon season(MA), and d inter-monsoon season (SO) over 36 years (1975–2010)


Legend

-0.48 - -0.26-0.26 - -0.04-0.04 - 0.20.2 - 0.450.45 - 0.7 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.58 - -0.33-0.33 - -0.11-0.11 - 0.080.08 - 0.290.29 - 0.54 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.47 - -0.22-0.22 - -0.02-0.02 - 0.140.14 - 0.260.26 - 0.42 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.28 - -0.17-0.17 - -0.08-0.08 - 0.030.03 - 0.160.16 - 0.32 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

a b

c d

Fig. 12 Spatial pattern of trends for 24-Hr Max for a northeast monsoon season (NEM), b southwest monsoon season (SWM), c inter-monsoon season(MA), and d inter-monsoon season (SO) over 36 years (1975–2010)


Legend

-0.27 - -0.07-0.07 - 0.110.11 - 0.280.28 - 0.430.43 - 0.57 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.52 - -0.19-0.19 - 0.050.05 - 0.230.23 - 0.360.36 - 0.54 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.4 - -0.13-0.13 - 0.030.03 - 0.130.13 - 0.290.29 - 0.56 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.33 - -0.2-0.2 - -0.08-0.08 - 0.060.06 - 0.20.2 - 0.34 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

a b

dc

Fig. 13 Spatial pattern of trends for Tot >95th for a northeast monsoon season (NEM), b southwest monsoon season (SWM), c inter-monsoon season(MA), and d inter-monsoon season (SO) over 36 years (1975–2010)


Legend

-0.29 - -0.04-0.04 - 0.10.1 - 0.190.19 - 0.330.33 - 0.58 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.49 - -0.14-0.14 - 0.030.03 - 0.120.12 - 0.30.3 - 0.65 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.24 - 0.040.04 - 0.160.16 - 0.220.22 - 0.340.34 - 0.62 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.29 - -0.09-0.09 - 0.050.05 - 0.140.14 - 0.280.28 - 0.49 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

a b

dc

Fig. 14 Spatial pattern of trends for Tot >99th for a northeast monsoon season (NEM), b southwest monsoon season (SWM), c inter-monsoon season(MA), and d inter-monsoon season (SO) over 36 years (1975–2010)


Legend

-0.25 - -0.07-0.07 - 0.10.1 - 0.250.25 - 0.390.39 - 0.55 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.42 - -0.21-0.21 - -0.03-0.03 - 0.150.15 - 0.30.3 - 0.44 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.33 - -0.06-0.06 - 0.080.08 - 0.150.15 - 0.290.29 - 0.56 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.4 - -0.18-0.18 - -0.06-0.06 - 0.020.02 - 0.150.15 - 0.37 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

a b

dc

Fig. 15 Spatial pattern of trends for Freq >20 for a northeast monsoon season (NEM), b southwest monsoon season (SWM), c inter-monsoon season(MA), and d inter-monsoon season (SO) over 36 years (1975–2010)


Legend

-0.26 - -0.06-0.06 - 0.080.08 - 0.180.18 - 0.320.32 - 0.52 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.51 - -0.25-0.25 - -0.04-0.04 - 0.150.15 - 0.30.3 - 0.48 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.45 - -0.14-0.14 - 0.090.09 - 0.250.25 - 0.370.37 - 0.54 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.32 - -0.21-0.21 - -0.12-0.12 - -0.01-0.01 - 0.140.14 - 0.34 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

a b

dc

Fig. 16 Spatial pattern of trends for Freq >95th for a northeast monsoon season (NEM), b southwest monsoon season (SWM), c inter-monsoon season(MA), and d inter-monsoon season (SO) over 36 years (1975–2010)


Legend

-0.34 - -0.07-0.07 - 0.130.13 - 0.280.28 - 0.40.4 - 0.55 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.5 - -0.15-0.15 - 0.020.02 - 0.110.11 - 0.280.28 - 0.63 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.24 - 0.030.03 - 0.150.15 - 0.220.22 - 0.350.35 - 0.61 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

Legend

-0.27 - -0.08-0.08 - 0.060.06 - 0.180.18 - 0.330.33 - 0.52 0 0.4 0.8 1.2 1.60.2

Decimal Degrees

a b

dc

Fig. 17 Spatial pattern of trends for Freq >99th for a northeast monsoon season (NEM), b southwest monsoon season (SWM), c inter-monsoon season(MA), and d inter-monsoon season (SO) over 36 years (1975–2010)


north easterly winds. By comparing Hr Max and 24-Hr Maxindices, it is shown that the western region experiences shortduration extreme rainfall (Hr Max) during the NEM and MAbut the eastern including some areas in the northwesternregion experience long duration extreme rainfall (24-HrMax) during the same seasons. Such results explain the oc-currences of flash floods which frequently happen in urbanareas (i.e., west coast) as well as massive floods in nonurbanareas (i.e., east and northwest coasts).

It is also shown that the northern region experiences bothshort and long period of extreme rainfall during the SWM andSO. As for the Tot >95th and Tot >99th, NEM recorded thehighest amount of rainfall that exceeds the 95th and 99thpercentile throughout the 36 years followed by SWM andthe inter-monsoon seasons. Monsoon seasons accumulatehigher total amount of rainfall than inter-monsoon seasonsdue to monsoon seasons comprising of 4 months while inter-

monsoon seasons comprising of 3 months. Similarly, theeastern region receives highest rainfall during the NEM whilethe western region receives highest rainfall during the MA.Furthermore, the northern region receives highest rainfallduring the remaining seasons, SWM and SO. In relation tothat, as the numbers of days for the four seasons are not thesame, the extreme frequency indices will be compared be-tween monsoon and inter-monsoon seasons (Figs. 7, 8, and 9).During the monsoon seasons, the extreme rainfall eventsfrequently occurred during the NEM where the eastern regionrecorded the highest total amount of rainfall for the 95th and99th percentile with amount of rainfall reaching up to 491.87and 176.98 mm, respectively. During the inter-monsoon sea-sons, the extreme rainfall events frequently occurred duringthe SO where the northern region recorded the highest totalamount of rainfall for the 95th and 99th percentile withamount of rainfall reaching up to 242.1 and 93.58 mm,

Table 3 Significant hourly extreme indices for each station during NEM

Station Hr Max 5-Hr Max 24-Hr Max Freq> 20 Freq> 95th Tot> 95th Freq> 99th Tot> 99th

+ - + - + - + - + - + - + - + -

1737001 *

2224038 * * * * *

2719001 *

2815001

2818110

2913001

3117070 * * * *

3118102 * * * *

3314001

3411017

3516022

3613004 * * * * * *

3710006 * * * * * * *

4908018 *

2831179 * * * * * * * *

3533102 *

3924072

4234109 * * * * *

4734079

4819027

4930038

5331048

5504035 * * * * * * *

6207032

6401002 “+” represents positive trend, “–” represents negative trend where darkened box represents significant trend for LS test, and “*” represents positivesignificant trend for MK test


respectively. As for the Freq >20, the frequency of rainfallevents greater than 20 mm within 1 h is found to be higherduring the NEM and MA for the each monsoon and inter-monsoon seasons.

4.2 Trend analysis of extreme rainfall indices

Figures 10, 11, 12, 13, and 14 illustrate the spatial pattern oftrends for all extreme intensity indices for each season.The values represent the coefficient of trends of whichpositive sign indicates upward trend, whereas negativesign indicates downward trend. All the extreme intensityindices show similar pattern of trend where almost allregions show increasing trends during all seasons. Thetrends are more pronounced towards the southern partof peninsular. Negative trends are mostly in the northernregion of the peninsular. Similarly, the extreme

frequency indices (Figs. 15, 16, and 17) reveal upwardtrends in almost all regions during all seasons. Suchtrends are also more significant towards the southernpart of peninsular during these seasons. The highestvalue of positive trend coefficient for Hr Max is ob-served for the MA (highest value of 0.6) while NEMhas the lowest. In contrast, NEM has the highest valueof positive coefficients of Freq >20, Freq >95th, andTot >95th while SO has the lowest. Besides, SWM hasthe highest value of positive coefficients of Freq >99thand Tot >99th whereas SO has the lowest. As for the 5-HrMax and 24-Hr Max, the highest value of positive coefficientsare shown during MA and the NEM, respectively, of whichboth indices has the lowest positive value of coefficientsduring SO. Again by comparing Hr Max and 24-Hr Max,the short duration of extreme rainfall depicts an upward trendduring MAwhile the long duration of extreme rainfall depicts

Table 4 Significant hourly extreme indices for each station during SWM


+ - + - + - + - + - + - + - + -

1737001

2224038 *

2719001 * * * * * * * *

2815001 * *

2818110

2913001

3117070 * * *

3118102

3314001

3411017

3516022 * * * * *

3613004 *

3710006

4908018

2831179 * * * * * * * *

3533102 * * * *

3924072 *

4234109

4734079

4819027

4930038

5331048

5504035 * * * * * *

6207032



the same trend during NEM over the years 1975–2010. Mostof the extreme indices (total of six out of eight indices) hashigher value of positive coefficients during the NEM and hasthe lowest during SO. Thus, it can be concluded that thepositive trend of extreme rainfall over the years 1975–2010is stronger during the NEM. This shows that the NEM bringsincreasing amount of rainfall compared to other seasons,hence the main contributing factor to the rainfall in Malaysiais the northeasterly winds over the South China Sea which isrelated to cold surges. The extreme rainfall in the west coast isalso caused by strong pulses of west wind associated with theMJO, as it passes through the maritime continent (Salahuddinand Curtis 2011). In addition, six indices have shown highervalue of positive coefficients during the monsoon seasons.This also proves that the positive trend of extreme rainfall isstronger during the monsoon seasons compared to the inter-monsoon seasons between the years 1975 and 2010.

4.3 The significant contribution of extreme indicesto Malaysia’s extreme rainfall

Tables 3, 4, 5, and 6 reveal the significant hourly extremeindices of each rainfall station using LS and MK test. Ingeneral, a large number of significant changes in extremeindices are found during the monsoon seasons that are theNEM and SWM, and this supports the findings discussed inSection 4.2. The analysis from MK test showed that all thesignificant extreme coefficients revealed upward trends. How-ever, the results obtained between LS and MK tests are slight-ly different due to different approaches employed by bothmethods in analyzing the trends (Zin et al. 2010). Figure 18depicts the significant trends of extreme rainfall indices for allseasons where the largest circle/triangle represents the largestcontribution of extreme rainfall indices to the extreme rainfalltrends. It shows that positive trends are more obvious during

Table 5 Significant hourly extreme indices for each station during MA


+ - + - + - + - + - + - + - + -

1737001

2224038

2719001 * * * * * * *

2815001

2818110

2913001

3117070

3118102 * * * * * * * *

3314001

3411017

3516022

3613004 * * *

3710006

4908018

2831179 * * * *

3533102

3924072

4234109

4734079 * * * * * * * *

4819027

4930038

5331048

5504035 * * * * * * *

6207032



the NEM while negative trends are more obvious during theSWM. The western region as well as the middle part of thepeninsular experience more positive trends during the NEM,SWM, and MA while the northern region experiences morenegative trends during the same seasons. Eighty percent of thesignificant extreme rainfall trends come from the monsoonseasons, i.e., NEM and SWM. In fact, 33 % relates to theNEM and SWM, while the rest 22 and 12% are related toMAand SO, respectively. In general, 65 % of the indices show apositive trend of which 14 % are significant while the rest35 % of the indices showed a negative trend of which only4 % are significant. The field significance test showed that allextreme intensity, extreme cumulative, and extreme frequencyindices revealed positive trends during the NEM and MAthroughout Peninsular Malaysia over the years 1975–2010.However, only extreme intensity and extreme frequency re-vealed the same trends during the SWM and SO. In particular,

Hr Max, Tot >95th, Tot >99th, Freq >99th, and 5-Hr Maxshowed upward trends during the SWM while other indicesremain unchanged. Meanwhile, Hr Max, Tot >99th, and Freq>99th depicted the similar trends during the SO while otherindices remain unchanged.

5 Conclusion

Analysis of extreme intensity indices shows higher values inthe eastern region during the NEM while the western regiondepicts higher indices during MA. For the northern region,high indices were observed during the SWM and SO. How-ever, short extreme intense rainfall (Hr Max) was observedmainly at stations located on the west coast where the hourlymaximum was recorded during the inter-monsoon seasonwhile in contrast, extreme cumulative rainfall (24-Hr Max)

Table 6 Significant hourly extreme indices for each station during SO


+ - + - + - + - + - + - + - + -

1737001

2224038

2719001 * * * *

2815001 * *

2818110 * * * * * * *

2913001

3117070 * * * * *

3118102

3314001

3411017

3516022 * * *

3613004

3710006

4908018

2831179 * * * * * * *

3533102

3924072

4234109

4734079

4819027

4930038

5331048

5504035 * * * * *

6207032



5504035

3613004

2831179

37100063924072

3117070

2224038

49080184819027


3118102

3533102

4234109

2815001

4734079

6207032

2719001 2831179

3516022

3117070

3533102

5504035

3924072

4908018

4734079

6401002

1737001


3613004

2815001

2913001

6207032

5504035

3118102

4734079

2831179

3533102

4908018


2815001

6207032

3924072

2831179

3516022

3117070

5504035


2719001

28181102913001

3314001

5331048

a b

c d

Fig. 18 Significant trends of extreme rainfall indices during a northeastmonsoon season (NEM), b southwest monsoon season (SWM), c inter-monsoon season (MA), and d inter-monsoon season (SO) (larger circle/

triangle indicates larger number of significant extreme rainfall indiceswhere “circle” represents significant positive trend, “triangle” representssignificant negative trend)


was observed at stations located on the eastern region wherethe highest value was recorded during the monsoon season.This shows that hourly data gives a better indication of theseasonal contribution to the annual extreme rainfall comparedto daily data. It gives a better profile of the extreme indiceswith regards to short and long duration rainfalls.

From the spatial pattern of trend analysis, stations locatedon the western and southern stretch which comprises of most-ly urban areas depict an increase in number of very wet andextremely wet hours. This explains the increase in the fre-quency of flash floods occurrences in these regions. Suchanalyses also conclude that the 95th percentile threshold isconsistent with the 99th percentile threshold. However, north-west region experiences decrease in number of very wet andextremely wet hours. Thus, negative trend of extreme rainfallwas found in the northwest region and can be considered asdrier area. In addition to that, the extreme frequency index(Freq >20) gives more significant contribution to the positiveextreme rainfall trend during the monsoon seasons. Mean-while, both extreme frequency and extreme intensity (24-HrMax, Freq >95th, Tot >95th, Tot >99th, and Hr Max) indicesgive more significant contribution to the positive extremerainfall trend during the inter-monsoon seasons. It is alsofound that the highest number of very wet hours and extreme-ly wet hours occur during the NEM which by large areproduced by stratiform rain.

In summary, the field significance test proved that most ofthe significant extreme indices showed the positive sign oftrends throughout Peninsular Malaysia mainly during theNEM and MA. Specifically, the extreme intensity, extremefrequency, and extreme cumulative indices showed increasingtrends during the NEM and MAwhile extreme intensity andextreme frequency had similar trends during the SWM andSO. Overall, the hourly extreme rainfall events in PeninsularMalaysia showed an increasing trend between the years 1975and 2010 with notable increasing trends in short temporalrainfall was observed during inter-monsoon season. Suchresult also shows that convective rain during this periodcontributes higher intensity rains which can only be capturedusing short duration rainfall series.

Acknowledgments This work was supported by the Drainage andIrrigation Department in providing the hourly rainfall data. This researchwas funded by the Ministry of Education Fundamental Research Grant(FRGS vote 4 F120), sponsorship from the International Islamic Universityof Malaysia and Ministry of Higher Education, Malaysia.

Open Access This article is distributed under the terms of the CreativeCommons Attribution License which permits any use, distribution, andreproduction in any medium, provided the original author(s) and thesource are credited.

References

Alexander LV, Arblaster JM (2009) Assessing trends in observed andmodelled climate extremes over Australia in relation to future pro-jections. Int J Climatol 29(3):417–435

Cayan DR, Redmond KT, Riddle LG (1999) ENSO and hydrologicextremes in the western United States. J Clim 12(12):3516

Cheng CS, Auld H, Li Q, Li G (2012) Possible impacts of climate changeon extreme weather events at local scale in south-central Canada.Clim Chang 112:963–979

Griffiths GM, SalingerMJ, Leleu I (2003) Trends in extreme daily rainfallacross the South Pacific and relationship to the South PacificConvergence Zone. Int J Climatol 23(8):847–869

Haylock M, Nicholls N (2000) Trends in extreme rainfall indices for anupdated high quality data set for Australia, 1910–1998. Int JClimatol 20(13):1533–1541

Salahuddin A, Curtis S (2011) Climate extremes in Malaysia and theequatorial South China Sea. Global Planet Chang 78(3–4):83–91

Schmidli J, Frei C (2005) Trends of heavy precipitation and wet and dryspells in Switzerland during the 20th century. Int J Climatol 25(6):753–771

Sen Roy S (2009) A spatial analysis of extreme hourly precipitationpatterns in India. Int J Climatol 29(3):345–355

Sen Roy S, Balling RC (2004) Trends in extreme daily precipitationindices in India. Int J Climatol 24(4):457–466

Suhaila J, Deni SM, Zin WZW, Jemain AA (2010) Trends inPeninsular Malaysia rainfall data during the Southwest Monsoonand Northeast Monsoon seasons: 1975–2004. Sains Malays 39(4):533–542

Yue S, Chun, Wang Y (2002) Regional streamflow trend detection withconsideration of both temporal and spatial correlation. Int J Climatol22:933–946

Zin WZW, Jamaludin S, Deni SM, Jemain AA (2010) Recent changes inextreme rainfall events in Peninsular Malaysia: 1971–2005. TheorAppl Climatol 99(3–4):303–314


historicaltrendofhourlyextremerainfallinpeninsularmalaysia · 3 2719001 setor jps sikamat seremban...

Documents