routine measurement of erythemally effective uv irradiance on inclined surfaces
TRANSCRIPT
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
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.
88 A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94
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
A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94 89
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).
90 A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94
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).
A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94 91
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).
92 A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94
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).
A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94 93
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.
94 A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94
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).