routine measurement of erythemally effective uv irradiance on inclined surfaces

Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94

www.elsevier.com/locate/jphotobiol

Routine measurement of erythemally effective UV irradianceon inclined surfaces

A. Oppenrieder a,b,*, P. Hoeppe a, P. Koepke b

a Institut und Poliklinik f€ur Arbeits- und Umweltmedizin der Ludwigs-Maximilians-Universit€at M€unchen, Ziemssenstr. 1, 80336 M€unchen, Germanyb Meteorologisches Institut der Ludwigs-Maximilians-Universit€at M€unchen, Theresienstr. 37, 80333 M€unchen, Germany

Received 30 May 2003; received in revised form 5 November 2003; accepted 11 November 2003

Available online 9 April 2004

Abstract

Measurements of erythemally weighted UV radiation are usually related to a horizontal surface. The radiation is weighted with

the sensitivity of the human skin, but the surface of the human body has only few horizontal surfaces. Therefore the UV radiation

on inclined surfaces has to be quantified to investigate UV effects on humans. To fulfill this task three fully automatic measuring

systems were built to measure the erythemally weighted UV radiation in 27 directions within 2 min. This system measures routinely

during the whole day and has now been in operation for nearly three years (in total 2000 measurement days) under any kind of

meteorological conditions. The measurements provide the informations needed for further investigations concerning the UV effects

on humans. The calibration of the erythemally weighting radiometers was performed in a way to provide reliable UV index

measurements for all directions. The results of four exemplary measurement days in summer and winter for clear sky and overcast

conditions are presented.

� 2004 Elsevier B.V. All rights reserved.

Keywords: Ultra violet; Exposure; Measurement; Long term; Erythemal; Inclined; Tilted; Orientation; Automatic

1. Introduction

The solar UV radiation is an environmental factor

with great influence on humans. Especially health risks

like sunburn and skin cancer have attracted the public

interest to UV investigations and have led to the foun-

dation of task groups under the umbrella of the World

Meteorological Organization (WMO) and the World

Health Organization (WHO) [1–3].In spite of the strong absorption of photons by ozone

in the solar UVB range (280–315 nm), a significant

fraction reaches the earth’s surface and has major bio-

logical effects. UVA photons (315–400 nm) are nearly

unaffected by the atmosphere’s ozone amount and a

greater number than in the UVB reach the ground.

Their energy is lower, because of the longer wavelength,

* Corresponding author. Tel.: +49-89-2180-4363; fax: +49-89-2180-

4381.

E-mail address: [email protected] (A. Oppenrieder).

1011-1344/$ - see front matter � 2004 Elsevier B.V. All rights reserved.

doi:10.1016/j.jphotobiol.2003.11.008

but they also have significant influence on biological andchemical processes. For effects on human skin neither

UVA nor UVB, but the total radiation weighted with

the sensitivity of the human skin is of importance. The

erythemally weighted UV irradiance in W/m2 multiplied

with 40 m2/W is defined as the UV index (UVI) [2]. An

international expert group of the WHO and WMO

recommends the UVI to inform the public about the risk

of UV caused skin damages [2].The internationally standardized UVI is defined to

characterize the effects of the UV irradiance on a hori-

zontal plane with the sensitivity of the human skin. But

generally the surfaces of biological bodies are not hori-

zontally oriented. For example the surface of the human

body can be approximated by a multitude of inclined

plane areas. To specify the effects of solar UV radiation

on the human body and specific parts of it the irradianceon arbitrarily orientated planes has to be characterized.

Only a few systematical UV measurements for in-

clined surfaces have been made by now and they are

mail to: [email protected]

86 A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94

confined to limited times and special locations. Blumt-

haler et al. [4] measured the solar UV irradiance on a

horizontal and a vertical surface in a high mountain

area, but the vertical surface was constantly facing the

south. As expected the horizontal UV irradiance washigher in summer and smaller in winter than the vertical.

At this measuring site roughly 3000 m above sea level

the irradiance ratio of the horizontal and vertical surface

depends to a great extent on the solar elevation and the

albedo.

Webb et al. [5] carried out measurements for vertical

surfaces facing east, south, west and north on a single day,

18 July 1995, at Iza~na station (2370 m above sea level) onTeneriffa, Spain. Except for the early and late hours of the

day the irradiance on the horizontal surface was greater

than the irradiance on the vertical surface, and the irra-

diance on the south facing surface was greater than the

irradiances on the surfaces facing the other directions.

The most extensive measurements on inclined sur-

faces have been made by Schauberger [6,7]. He used

measurements of erythemally weighted irradiance oninclined surfaces to calculate a correction in a model for

calculating the erythemally weighted irradiance on in-

clined surfaces. The model needs the erythemally

weighted irradiance on the horizontal surface, the al-

bedo and the inclination angle (angle between position

of the sun and the perpendicular of the surface) as input

parameters.

According to Weihs [8] the measurements of Schau-berger are limited to locations at low height above sea

level and in flat topography. Weihs developed a model

to calculate the erythemally weighted irradiance on in-

clined surfaces considering the topography and ground

albedo of the environment. Comparing his model with

the measured data of Schauberger [6,7] he found an

overestimation for vertical planes of more than 10% [8],

but another validation of his model with measured datawas not carried out.

In this paper results from a fully automatic measuring

system, Angle SCAnning RAdiometer for determination

of erythemally weighted irradiance on TIlted Surfaces

(ASCARATIS), are presented. This system was de-

signed and built to measure in all kinds of environ-

mental conditions, including extreme climatic conditions

of high altitudes, and to produce data sets that providereliable information about erythemally weighted UV

radiation on inclined surfaces [9].

Fig. 1. The measuring system ASCARATIS, movable and fixed

erythemally weighting broadband radiometer, at the site Hoher

Peissenberg.

2. Materials and methods

2.1. Angle scanning radiometer for determination of

erythemally weighted irradiance on tilted surfaces

The system ASCARATIS was developed and built

for continuous long term measurements in extreme

weather conditions by the Meteorological Institute and

the Institute and Outpatient Clinic for Occupational and

Environmental Medicine in M€unchen. Two erythemally

weighting broadband radiometers, one is movable and

the other permanently mounted horizontally, are si-multaneously measuring to allow the comparison of UV

irradiances on inclined planes and the horizontal plane.

The movable radiometer can be positioned in all direc-

tions by two stepping motors mounted perpendicularly.

The way of positioning can be easily programmed to

adjust the system to a given measuring task. In Fig. 1 the

measuring setup is shown for the site Hoher Peissenberg.

To scan 27 positions (see Table 1) in less than twominutes is a good compromise to get irradiance mea-

surements for different orientations, that are dense en-

ough to describe all possible tilt directions, on one hand

and to have constant sky conditions (clouds) while

scanning on the other hand. The conditions are classi-

fied as constant for UVI measurements, if the fixed ra-

diometer’s variations are less than 5% in the scanning

period of 2 min. The constant sky conditions allow thecomparison of the fixed and moving radiometer in the

period of one scan.

For the positions with number 1–12 (Table 1) the

movable radiometer is viewing towards the horizon

(elevation angle h is 0) and is moved in 30� steps from an

azimuth angel u of 15–345�. For the positions 13–24 h is

set to 45� and u changes again in 30� steps backwards

from 345� to 15�. In position 25 h is set to 90� and theradiometer is horizontally oriented looking upwards. In

this position, the movable radiometer is compared to the

fixed simultaneously measuring horizontal radiometer.

The comparability of the inclined and horizontal UVI

therefore is ensured every two minutes. After that, po-

sition 26 turns the radiometer for the albedo measure-

ment to the south and up side down, u ¼ 180� and

h ¼ �90�. Finally the actual position of the sun is

Table 1

Positions (#) of the movable erythemally weighting radiometer

# u (�) h (�)

1 15 0

2 45 0

3 75 0

4 105 0

5 135 0

6 165 0

7 195 0

8 225 0

9 255 0

10 285 0

11 315 0

12 345 0

13 345 45

14 315 45

15 285 45

16 255 45

17 225 45

18 195 45

19 165 45

20 135 45

21 105 45

22 75 45

23 45 45

24 15 45

25 15 90

26 180 )9027 SAA SEA

u is the azimuth angle and h the elevation angle for the orientation

of the radiometer’s view. In position 27 the view is directly orientated

to the sun. SAA stands for solar azimuth angle and SEA for solar

elevation angle.

A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94 87

calculated and the movable radiometer’s view is directed

to the sun (position 27). After that, the procedure starts

again with position 1. The scans are performed during

the whole day and in all weather conditions.

Every night the measured data are automatically

saved to a central server. At the server a data base im-

ports the data, calculates the UVI using the individualcalibration at the actual date for every radiometer and

displays the results of the measurements in graphs for

control.

2.2. Radiometers and calibration

The measurements of the erythemally weighted UV

irradiance are carried out by common erythemallyweighting broadband radiometers. Instruments were

chosen with relatively good cosine response and

broadband sensitivity representing erythemal sensitivity.

The characteristics of each instrument, however, differ

slightly in spectral and cosine response [1,3]. The indi-

vidual variation of the characteristics causes differences

in the output of the radiometers under the same atmo-

spheric conditions. A calibration is required to com-pensate these deviations.

Once a year the radiometers are calibrated by an

absolute standard at an independent laboratory. The

calibration is carried out considering the standards of

the WMO [1,3]. A matrix of calibration factors for every

instrument with values (in (W/m2)/V) depending on thesolar elevation angle and the total amount of ozone is

the result of this calibration. The position of the sun is

calculated by an astronomical algorithm [10] and TOMS

satellite data provide total amount of ozone. Depending

on the solar elevation and on ozone amount the values

of the matrix have to be linearly interpolated in two

dimensions for every measurement. The resulting abso-

lute calibration factor has to be multiplied with thevoltage output of the instrument and with the factor 40

m2/W to get the actual UVI [11].

For dates of measurements lying in the time between

two yearly absolute calibrations the two corresponding

calibration factors were linearly interpolated in time to

provide the correct calibration factor at the date of the

measurement.

The absolute calibration corrects the differing spectralsensitivities of the radiometers and therefore provides

comparable UVI measurements. This is essential for

broadband radiometers, because the spectral character

of the measured radiation is changing with the solar

elevation.

Also the not ideal angular sensitivity of the radiom-

eters is corrected by the calibration. Fig. 2 exemplarily

shows the typical angular sensitivity of one of our ra-diometers recorded at one of the calibrations. The de-

viation of the angular sensitivity from the cosine is lower

than 10% for incidence angles smaller than roughly 65�.For horizontally orientated radiometers the incidence

angle and the solar zenith angle are the same and the

errors due to the deviation of the angular sensitivity

from the cosine are simultaneously adjusted by the cal-

ibration factor depending on the solar elevation angle.The influence of differing angular sensitivity on the re-

cording of the direct solar radiation part of the global

UV irradiance is not very strong for horizontal radi-

ometers, because at zenith angles greater than 65� this

part is less than 10% of the total [12].

For inclined radiometers the incidence and solar

zenith angle are differing. Measurements at noon with

low solar zenith angles and incidence angles greaterthan 65� are possible. We could separate direct and

diffuse solar radiation with the standard setup of AS-

CARATIS and analyze the effects of the angular sen-

sitivity deviations on the recording of the direct solar

UV irradiance. A correction of these effects was carried

out. The difference of the UVI output of the inclined

radiometer before and after the correction never ex-

ceeds 8%. Fig. 3 exemplarily shows the consequence ofthe correction for an inclined radiometer in position 4,

azimuth angle u of 105� and elevation angle h of 0�(Table 1), on a clear day (14 June 2002) with maximal

0.70

0.75

0.80

0.85

0.90

0.95

1.00

1.05

1.10

1.15

1.20

0 10 20 30 40 50 60 70 80 90

IA [˚]

An

gu

lar

Sen

siti

vity

/ co

s (I

A)

Fig. 2. Ratio of the measured angular sensitivity of one of the erythemally weighting broadband radiometers to the ideal cosine. The incidence angle

(IA) is 0� for perpendicular incidence.

0.90

0.92

0.94

0.96

0.98

1.00

1.02

1.04

1.06

1.08

1.10

0 60 120 180 240 300 360

SAA [˚]

rati

o

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

UV

I

before/after before after

Fig. 3. UVI values of the inclined radiometer in position 4 (Table 1) corresponding to SAA (solar azimuth angle) before and after the correction of

the incorrect angular sensitivity effects.


solar elevation of 65� in M€unchen. If the behavior of

the angular sensitivity in Fig. 2 is compared with the

ratio in Fig. 3, it becomes evident that the correctioncompensates the error caused by the non-ideal angular

sensitivity. The deviations for high quality UVI mea-

surements with horizontal erythemally weighting

broadband radiometers are also within 8% [1]. There-

fore it is not essential to carry out this correction for

inclined radiometers, but it surely contributes to the

comparability of inclined and horizontal radiometers.

Additionally to the complete absolute calibration the

radiometers of the different stations were compared

against each other in regular periods of three months toallow a compensation of non-linear temporal effects in

the characteristics of the radiometers. To compare the

radiometers they were mounted next to each other on a

horizontal plate and each output voltage was logged

simultaneously and therefore corresponds to the same

global UV irradiance. The data of the comparison were

also absolutely calibrated in the mentioned way and the


one hour average of the UVI at noon calculated for each

instrument. The hourly average offset of each instrument

in relation to the hourly average of all instruments is

corrected by a factor. This correction factor can reach

values of 5% and has to be linearly interpolated forthe measurement dates lying in between the times of

comparisons.

All mentioned calibrations and corrections are per-

formed to control and guarantee the quality of the

measurements and allow an optimal comparability of

the six radiometers of the three measuring systems. The

use of simultaneously measuring radiometers, moved

and fixed, in one system also provides an additionalcontrol of the comparability of the radiometers, when

the movable radiometer is oriented horizontally in

position 25 (Table 1).

2.3. Measuring sites and time schedule

Data sets to characterize the effects of UV irradiance

on humans have to cover the variability of all relevantparameters influencing the UV irradiance on inclined

surfaces. The relevant parameters for this are solar ele-

vation angle, amount and properties of clouds, turbid-

ity, surface albedo and height above sea level.

The measuring sites were chosen in a way, that the

measurements represent the UV conditions in central

Europe. The basic characteristics of the measuring sites

are presented in Table 2. The sites vary between urbanconditions in a small town with rural environment and

relatively warm climate (W€urzburg), rural conditions inan area with agriculture (Frankendorf) and another in a

higher elevation with pasture farming (Hoher Peissen-

berg). Urban conditions are found at the site in

M€unchen, where ASCARATIS was situated on the

gravel roof of the building of the Meteorological Insti-

tute. A station (Schneefernerhaus) near a high mountainskiing area (Zugspitze) completes the list of sites.

At the end of March 2003 the systems have been

measuring in total for nearly 2000 days and they provide

a reliable data base for the statistical analysis of the

typical UV environments at the measuring sites. They

therefore characterize the typical UV radiation on in-

clined surfaces for central Europe.

Table 2

Description of the measuring sites

Measuring site GLO

(�E)GLA

(�E)HASL

(m)

Type of location

W€urzburg 9.9 49.7 200 Urban

Frankendorf 12.0 48.3 450 Rural/agriculture

M€unchen 11.6 48.1 530 Urban

Hoher Peissenberg 11.0 47.8 1000 Rural/pasture

Zugspitze 11.0 47.4 2650 Mountainous

GLO stands for geographical longitude, GLA for geographical

latitude and HASL for height above sea level.

3. Results

In this paper exemplary data for all measurement are

presented, one clear sky day and one overcast day each

for summer and winter at the measuring site inM€unchen.

The Figs. 4–7 show the measured UVI as function of

the solar azimuth angle (SAA), which corresponds with

the time scale during the day. The figures are always

divided into three graphs. In the upper graph positions

25, 26 and 27 are shown (Table 1), i.e., the movable

radiometer is horizontally oriented, up side down for

albedo measurement and facing the sun. The middlegraph of the figures shows the positions 1–12, i.e., the

movable radiometer is viewing the horizon. The lower

graph of these figures shows the positions 13–24, i.e., the

radiometer’s elevation angle is 45�. For the middle and

lower graphs of the figure the lower envelope of the

curve cluster represents the diffuse irradiance incidence

on the vertical (middle graph), respectively, 45� inclinedplane (lower graph).

Fig. 4 represents the measured UVI on a typical clear

day in summer, 4 July 2002. The UVI on the horizontal

surface reaches a value of 8.0 on this day on the roof of

the Meteorological Institute in M€unchen. The measured

erythemally weighted albedo of the gravel roof is about

8.8% (0.7 UVI). The maximal UVI value of this day

amounts to 8.3 in position 27 facing the sun, 3.8% more

than in the horizontally oriented position 25.Looking at the middle graph of Fig. 4 the diffuse UV

irradiance seems to be nearly isotropic in the positions

with the same elevation angle h for the radiometers in-

tegrating over the half sphere. Therefore the difference

of the UVI values at times without direct sun influence is

small. The diffuse radiation in positions 13–24 reaches

more than twice the magnitude of the positions 1–12,

because in the latter case a larger fraction of the viewedhemisphere is formed by the ground from which less UV

radiation arrives due to reflection. The diffuse radiation

in positions 13–24 nearly reaches the same values as the

maximum of all UVI values (with direct solar radiation)

in positions 1–12.

The influence of the direct solar radiation can be seen

at times when the UVI value of the individual positions

deviates from the mentioned low envelope of the diffuseirradiance. The transition from not influenced to influ-

enced times and vice versa is more distinctive in positions

1–12, because the diffuse radiation part of the measured

UVI there is smaller than in the positions 13–24.

Comparing the inclined positions to the horizontal,

the irradiance of the 45� inclined surfaces can reach the

same maximal value, if they are oriented to the south.

The more the surfaces are rotated out of the southernorientation the smaller the maximal possible UVI in this

direction gets. The orientations facing the horizon reach

UVI values of about 4.4, 55% of the maximum of the

0.0

0.01.02.03.04.05.06.07.08.09.0

252627

0.00.51.01.52.02.53.03.54.04.55.0

UV

I

123456789101112

131415161718192021222324

0 60 120 180 240 300 360

SAA [˚]

0 60 120 180 240 300 360

SAA [˚]

0 60 120 180 240 300 360

SAA [˚]

UV

I

1.02.03.04.05.06.07.08.09.0

UV

I

Fig. 4. UVI values, average corresponding to 10� SAA (solar azimuth angle), in clear sky conditions recorded by the movable radiometer on 4 July

2001 in M€unchen (positions 1–27, see Table 1).


horizontal UVI value, and the time range in which this

maximum lies is extended.

Fig. 5 shows the measurements on a clear sky winterday with snow cover in M€unchen (15 December 2001). It

is obvious that the UVI values for all directions are

below 1.1 due to low sun. The UVI on the horizontal

surface, position 25, is roughly 0.87 and the albedo

measurement, position 26, is 0.61 (70.1% albedo) due to

the snow cover on the gravel roof. The maximal UVI

value is recorded again in position 27, i.e., radiometer

facing the sun, with an UVI value of 1.06, 21.8% morethan in the horizontal position 25. Compared to Fig. 4

the UVI ratio viewing the sun relative to the horizontal

orientation has increased from 3.8% to 21.8%.

The UVI values are relatively higher for tilted sur-

faces than for the horizontal surface. Two effects are

responsible for this behavior. First the solar azimuthangle does not get as big as in summer and therefore the

direct solar radiation is weighted less in the horizontal

due to the cosine sensitivity than in the tilted orienta-

tions. Second the diffuse part of the radiation reflected

from the ground has increased due to the snow cover.

The horizontal radiometer cannot see the ground and

therefore misses this reflected radiation.

These consequences can also be seen in the middleand lower graph of Fig. 5. The measurements of posi-

tion 1–12 and 13–24 are quite similar. The same maxi-

mal UVI values are reached and comparing positions

252627

123456789101112

131415161718192021222324

0 60 120 180 240 300 360

SAA [˚]

0.00

0.05

0.10

0.15

0.20

0.25

0.30

UV

I

0 60 120 180 240 300 360

SAA [˚]

0.00

0.05

0.10

0.15

0.20

0.25

0.30

UV

I

0 60 120 180 240 300 360

SAA [˚]

0.00

0.05

0.10

0.15

0.20

0.25

0.30

UV

I

Fig. 5. UVI values, average corresponding to 10� SAA (solar azimuth angle), in clear sky conditions recorded by the movable radiometer on 15

December 2001 in M€unchen (positions 1–27, see Table 1).


with the same azimuth angle u the progression over the

day is similar. Also the asymmetric feature to the 180�solar azimuth angle of the daily progression has nearlyvanished. The contribution of the direct solar radiation

has decreased significantly in winter.

Figs. 6 and 7 show measurements in overcast con-

ditions for summer, respectively, winter. In summer, 11

July 2001 (Fig. 6), the UVI appears to be maximal in

horizontal orientation, position 25. The variability of

the UVI in time is caused by the high variability of the

optical thickness of clouds. There is no direct sun inovercast conditions and the horizontally orientated

radiometer records the most diffuse radiation, because

it sees the greatest part of the overcast sky. The more

the radiometer is inclined the less diffuse skylight it sees

and the more radiation reflected from the ground is

detected. The amount of diffuse radiation reflectedfrom the ground is just a fraction of the amount of

diffuse skylight. Thus the irradiance and in conse-

quence the UVI is reduced with increased tilt angle.

The measurements for the albedo position, position 26,

show strong variations due to the very low UVI mea-

surements in this position. The signal-to-noise ratio is

too small for reliable results, but the order of size for

the albedo is about 9.0% and is similar to the clear skysummer conditions (Fig. 4).

In winter overcast conditions, 17 December 2001

(Fig. 7), the UVI measurements for all orientations are

0.0

0.2

0.4

0.6

0.8

1.0

1.2

252627

0.0

0.1

0.2

0.3

0.4

0.5

UV

I

123456789101112

0.00.10.20.30.40.50.60.70.80.91.0

UV

I

131415161718192021222324

0 60 120 180 240 300 360

0 60 120 180 240 300 360

0 60 120 180 240 300 360

SAA [˚]

SAA [˚]

SAA [˚]

UV

I

Fig. 6. UVI values, average corresponding to 10� SAA (solar azimuth angle), in overcast conditions recorded by the movable radiometer on 11 July

2001 in M€unchen (positions 1–27, see Table 1).


very low, and again the variability in time is high due to

the variations in the optical thickness of clouds. The

maximal UVI is measured for the horizontally orien-

tated radiometer, with a value of 0.38. The reasons are

the same as in summer overcast conditions. The mea-

surements in the albedo position 26 reach an UVI of

0.22 (57.9% albedo). Compared to the clear sky condi-

tions on 15 December 2001 the albedo has decreasedfrom 70.1% to 57.9%, which can be explained by the

change of the snow conditions during the two days.

Compared to summer conditions (Figs. 4 and 6), how-

ever, the albedo is high and causes the reflection of a

significant fraction of the diffuse skylight. Yet, the sky-

light still is greater than its reflected part, and thus the

irradiance on the radiometers tilted with 45� is still

higher than that on the vertical one.

Resuming and simplifying the results of the examples

the characteristics of the erythemally weighted and

over the hemisphere integrated UV irradiance are

caused by the environmentally dependent combination

of three components of UV radiation: the direct solar

radiation, the diffuse skylight and the diffuse radiationreflected from the ground. The measured UVI on the

inclined surfaces strongly depends on the individual

combination of the three components. In the examples

the maximal increase (21.8%) of the UVI due to the

inclination of surfaces is found in Winter, because of the

increased fraction of diffuse skylight and diffuse radia-

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0 60 120 180 240 300 360

0 60 120 180 240 300 360

0 60 120 180 240 300 360

SAA [˚]

SAA [˚]

SAA [˚]

UV

I 252627

0.00

0.05

0.10

0.15

0.20

0.25

UV

I

123456789101112

0.00

0.05

0.10

0.15

0.20

0.25

0.30

UV

I

131415161718192021222324

Fig. 7. UVI values, average corresponding to 10� SAA (solar azimuth angle), in overcast conditions recorded by the movable radiometer on 17

December 2001 in M€unchen (positions 1–27, see Table 1).


tion reflected from the ground relative to the direct solar

radiation. If the direct solar radiation is small or absent

the progression of the UVI on inclined surfaces over the

day gets symmetrical to 180� solar azimuth angle and

the influence of the elevation angle h increases compared

to the azimuth angle u.

4. Conclusion

To get information on the variability of the erythe-

mally weighted UV radiation on inclined surfaces a

reliable data set had to be created for a period of at

least one year under all occurring weather conditions

and different environments to characterize the UV ef-

fects on humans. Therefore three fully automatic

measurement systems ASCARATIS (Angle SCAnning

RAdiometer for determination of erythemally weighted

irradiance on TIlted Surfaces) were designed and built

to fulfill the measurement task. More than 2000 mea-surement days are now recorded by the three AS-

CARATIS systems. The broadband radiometers are

calibrated in the best possible way and therefore pro-

vide reliable and quality controlled UVI data. The data

are processed that far by an automatic data base to

allow further investigations with the actual UVI values

on inclined surfaces at the three measuring sites at the

same time.


The mentioned methods and exemplary results of the

measurements show the comprehensiveness of the in-

formation included in the already gained data set. This

data set has to be evaluated explicitly with regard to the

regional variability and climatology of UV radiation forhorizontal and inclined planes at the measuring loca-

tions to allow a better characterization of UV radiation

effecting the human body under real meteorological

conditions.

The measured irradiances were already used for a

comparison with modelled ones [13] and will be used for

characterizing the UV effects on humans by combining

these data with a surface model of the human body [14].

Acknowledgements

The study is part of the Bavarian Research Network

(Bayerischer Forschungsverbund: Erh€ohte UV-Strah-

lung in Bayern – Folgen und Maßnahmen) and funded

by the Bavarian State Ministry for Science, Research

and Arts. We thank Dipl.-Ing. Meinhard Seefeldner,Dipl.-Phys. Dieter Rabus, Dr. Georg Praml and Dr.

Jochen Reuder for their valuable contributions to the

design and construction of the ASCARATIS measuring

system. We also thank the Environmental Research

Station Schneefernerhaus (Dr. Gerhard Enders) for fi-

nancial and technical support at the measuring site

Zugspitze.

References

[1] K. Leszcynski, K. Jokela, L. Ylianttila, R. Visuri, M. Blumthaler,

Report of the WMO/STUK Intercomparison of Erythemally-

Weighted Solar UV Radiometers, Report 112, World Meteoro-

logical Organization, Helsinki, Finland, 1995.

[2] WMO, Report of the WMO–WHO Meeting of Experts on

Standardization of UV Indices and Their Dissemination to the

Public, Report 127, World Meteorological Organization, Les

Diablerets, Switzerland, 1997.

[3] A. Bais, C. Topaloglou, S. Kazadtzis, M. Blumthaler, J. Schreder,

A. Schmalwieser, D. Henriques, M. Janouch, Report of the LAP/

COST/WMO Intercomparison of Erythemal Radiometers, Report

141, World Meteorological Organization, Thessaloniki, Greece,

1999.

[4] M. Blumthaler, W. Ambach, R. Ellinger, UV-Bestrahlung von

horizontalen und vertikalen Fl€achen im Hochgebirge, Sonderd-

ruck aus Wetter und Leben 48 (1996) 25–31.

[5] A.R. Webb, P. Weihs, M. Blumthaler, Spectral UV irradiance on

vertical surfaces: a case study, Photochem. Photobiol. 69 (4)

(1999) 464–470.

[6] G. Schauberger, Model for the global irradiance of the solar

biologically-effective ultraviolet-radiation on inclined surfaces,

Photochem. Photobiol. 52 (5) (1990) 1029–1032.

[7] G. Schauberger, Anisotropic model for the diffuse biologically-

effective irradiance of solar UV-radiation on inclined surfaces,

Theor. Appl. Climatol. 46 (1992) 45–51.

[8] P. Weihs, Influence of ground reflectivity and topography on

erythemal UV-radiation on inclined surfaces, Int. J. Biometeorol.

46 (2002) 95–104.

[9] P. Hoeppe, A. Oppenrieder, P. Koepke, J. Reuder, M. Seefeldner,

D. Nowak, Measurements of UV-irradiation on inclined surfaces

for exposure assessments of the human body, in: Proceedings of

the 15th Conference on Biometeorology and Aerobiology joint

with 16th International Congress on Biometeorology, Kansas

City, American Meteorological Society, Boston, USA, 2000, pp.

66–67.

[10] O. Montenbruck, Astronomie mit dem Personal Computer,

second ed., Springer Verlag, Berlin, Deutschland, 1965.

[11] A. Oppenrieder, P. Hoeppe, P. Koepke, J. Reuder, J. Schween, J.

Schreder, Simplified calibration for broadband solar UV mea-

surements, Photochem. Photobiol. 78 (6) (2003) 603–

606.

[12] P. Koepke, UV-Strahlung an der Oberfl€ache, in: R. Guderian

(Ed.), Handbuch der Umweltver€anderungen und €Okotoxikologie,

Atmosph€are, vol. 1B, Springer Verlag, Berlin, Germany, 2000, pp.

297–331.

[13] M. Mech, P. Koepke, Multifunctional model for UV irradiance

on arbitrarily oriented surfaces, Theor. Appl. Climatol. (2004) in

press.

[14] P. Hoeppe, A. Oppenrieder, C. Erianto, P. Koepko, J. Reuder, M.

Seefeldner, D. Nowak, Visualization of UV-exposure of the

human body based on data from a scanning UV-measuring

system, Int. J. Biometeorol. (in press).

routine measurement of erythemally effective uv irradiance on inclined surfaces

Documents