- Email: [email protected]
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 irradiance on inclined surfaces A. Oppenrieder a
a,b,*
, P. Hoeppe a, P. Koepke
b
Institut und Poliklinik f€ur Arbeits- und Umweltmedizin der Ludwigs-Maximilians-Universit€at M€unchen, Ziemssenstr. 1, 80336 M€unchen, Germany b 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. Ó 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 foundation 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 biological 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: +49-89-21804381. E-mail address: [email protected] (A. Oppenrieder). 1011-1344/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.jphotobiol.2003.11.008
but they also have significant influence on biological and chemical 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 horizontal plane with the sensitivity of the human skin. But generally the surfaces of biological bodies are not horizontally 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 irradiance on arbitrarily orientated planes has to be characterized. Only a few systematical UV measurements for inclined 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. Blumthaler 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 was higher 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) on Teneriffa, 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 irradiance on the south facing surface was greater than the irradiances on the surfaces facing the other directions. The most extensive measurements on inclined surfaces have been made by Schauberger [6,7]. He used measurements of erythemally weighted irradiance on inclined surfaces to calculate a correction in a model for calculating the erythemally weighted irradiance on inclined surfaces. The model needs the erythemally weighted irradiance on the horizontal surface, the albedo 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 Schauberger 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 inclined 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 data was 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 designed and built to measure in all kinds of environmental conditions, including extreme climatic conditions of high altitudes, and to produce data sets that provide reliable information about erythemally weighted UV radiation on inclined surfaces [9].
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 simultaneously measuring to allow the comparison of UV irradiances on inclined planes and the horizontal plane. The movable radiometer can be positioned in all directions 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 two minutes is a good compromise to get irradiance measurements for different orientations, that are dense enough 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 classified as constant for UVI measurements, if the fixed radiometer’s variations are less than 5% in the scanning period of 2 min. The constant sky conditions allow the comparison 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 the radiometer 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, position 26 turns the radiometer for the albedo measurement to the south and up side down, u ¼ 180° and h ¼ 90°. Finally the actual position of the sun is
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
Fig. 1. The measuring system ASCARATIS, movable and fixed erythemally weighting broadband radiometer, at the site Hoher Peissenberg.
A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94 Table 1 Positions (#) of the movable erythemally weighting radiometer #
u (°)
h (°)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
15 45 75 105 135 165 195 225 255 285 315 345 345 315 285 255 225 195 165 135 105 75 45 15 15 180 SAA
0 0 0 0 0 0 0 0 0 0 0 0 45 45 45 45 45 45 45 45 45 45 45 45 90 )90 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.
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 imports the data, calculates the UVI using the individual calibration 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 erythemally weighting 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 individual variation of the characteristics causes differences in the output of the radiometers under the same atmospheric conditions. A calibration is required to compensate these deviations.
87
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 the solar 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 absolute calibration factor has to be multiplied with the voltage 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 spectral sensitivities 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 radiometers is corrected by the calibration. Fig. 2 exemplarily shows the typical angular sensitivity of one of our radiometers recorded at one of the calibrations. The deviation 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 calibration factor depending on the solar elevation angle. The influence of differing angular sensitivity on the recording of the direct solar radiation part of the global UV irradiance is not very strong for horizontal radiometers, 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 greater than 65° are possible. We could separate direct and diffuse solar radiation with the standard setup of ASCARATIS and analyze the effects of the angular sensitivity 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 exceeds 8%. Fig. 3 exemplarily shows the consequence of the 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
88
A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94 1.20 1.15
Angular Sensitivity / cos (IA)
1.10 1.05 1.00 0.95 0.90 0.85 0.80 0.75 0.70 0
10
20
30
40
50
60
70
80
90
IA [˚]
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.
before/after
before
after
1.10
4.5
1.08
4.0
1.06
3.5
1.04
3.0 2.5 UVI
ratio
1.02 1.00 2.0 0.98 1.5
0.96
1.0
0.94
0.5
0.92 0.90 0
60
120
180
240
300
0.0 360
SAA [˚]
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 correction compensates the error caused by the non-ideal angular sensitivity. The deviations for high quality UVI measurements with horizontal erythemally weighting broadband radiometers are also within 8% [1]. Therefore 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 to allow 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
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 for the measurement dates lying in between the times of comparisons. All mentioned calibrations and corrections are performed 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 additional control 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 relevant parameters influencing the UV irradiance on inclined surfaces. The relevant parameters for this are solar elevation angle, amount and properties of clouds, turbidity, 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 urban conditions in a small town with rural environment and relatively warm climate (W€ urzburg), rural conditions in an area with agriculture (Frankendorf) and another in a higher elevation with pasture farming (Hoher Peissenberg). Urban conditions are found at the site in M€ unchen, where ASCARATIS was situated on the gravel roof of the building of the Meteorological Institute. A station (Schneefernerhaus) near a high mountain skiing 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 inclined surfaces for central Europe. Table 2 Description of the measuring sites Measuring site W€ urzburg Frankendorf M€ unchen Hoher Peissenberg Zugspitze
GLO (°E) 9.9 12.0 11.6 11.0 11.0
GLA (°E)
HASL (m)
Type of location
49.7 48.3 48.1 47.8 47.4
200 450 530 1000 2650
Urban Rural/agriculture Urban Rural/pasture Mountainous
GLO stands for geographical longitude, GLA for geographical latitude and HASL for height above sea level.
89
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 in M€ 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 middle graph 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° inclined plane (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 integrating 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 viewed hemisphere 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 diffuse irradiance. The transition from not influenced to influenced 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 southern orientation 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
A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94
UVI
90
9.0 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0
25 26 27
0
60
120
180
240
300
360
UVI
SAA [˚] 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0
60
120
180
240
300
360
UVI
SAA [˚]
9.0 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 0
60
120
180 SAA [˚]
240
300
360
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
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 winter day 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% more than 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 surfaces than for the horizontal surface. Two effects are responsible for this behavior. First the solar azimuth angle 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 orientations. 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 middle and lower graph of Fig. 5. The measurements of position 1–12 and 13–24 are quite similar. The same maximal UVI values are reached and comparing positions
A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94
91
0.30 0.25
UVI
0.20
25 26 27
0.15 0.10 0.05 0.00 0
60
120
180
240
300
360
SAA [˚] 0.30 0.25
UVI
0.20 0.15 0.10 0.05 0.00 0
60
120
180
240
300
360
SAA [˚]
0.30 0.25
UVI
0.20 0.15 0.10 0.05 0.00 0
60
120
180 SAA [˚]
240
300
360
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
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 nearly vanished. The contribution of the direct solar radiation has decreased significantly in winter. Figs. 6 and 7 show measurements in overcast conditions 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 in overcast 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 reflected from the ground is just a fraction of the amount of diffuse skylight. Thus the irradiance and in consequence 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 measurements 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 sky summer conditions (Fig. 4). In winter overcast conditions, 17 December 2001 (Fig. 7), the UVI measurements for all orientations are
92
A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94
1.2 1.0
UVI
0.8
25 26 27
0.6 0.4 0.2 0.0 0
60
120
180
240
300
360
SAA [˚] 0.5
UVI
0.4 0.3 0.2 0.1 0.0 0
60
120
180
240
300
360
UVI
SAA [˚]
1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0
60
120
180 SAA [˚]
240
300
360
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
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 orientated radiometer, with a value of 0.38. The reasons are the same as in summer overcast conditions. The measurements in the albedo position 26 reach an UVI of 0.22 (57.9% albedo). Compared to the clear sky conditions on 15 December 2001 the albedo has decreased from 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), however, the albedo is high and causes the reflection of a significant fraction of the diffuse skylight. Yet, the skylight 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 radiation reflected 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-
A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94
93
0.30 0.25
UVI
0.20
25 26 27
0.15 0.10 0.05 0.00 0
60
120
180
240
300
360
SAA [˚] 0.25
UVI
0.20 0.15 0.10 0.05 0.00 0
60
120
180
240
300
360
SAA [˚]
0.30 0.25
UVI
0.20 0.15 0.10 0.05 0.00 0
60
120
180
240
SAA [˚]
300
360
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
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 erythemally 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 effects 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 measurement days are now recorded by the three ASCARATIS systems. The broadband radiometers are calibrated in the best possible way and therefore provide 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 information 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 for horizontal and inclined planes at the measuring locations 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].
[3]
[4]
[5]
[6]
[7]
Acknowledgements The study is part of the Bavarian Research Network (Bayerischer Forschungsverbund: Erh€ ohte UV-Strahlung 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 financial and technical support at the measuring site Zugspitze.
[8]
[9]
[10] [11]
[12]
References [1] K. Leszcynski, K. Jokela, L. Ylianttila, R. Visuri, M. Blumthaler, Report of the WMO/STUK Intercomparison of ErythemallyWeighted Solar UV Radiometers, Report 112, World Meteorological 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. 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. M. Blumthaler, W. Ambach, R. Ellinger, UV-Bestrahlung von horizontalen und vertikalen Fl€achen im Hochgebirge, Sonderdruck aus Wetter und Leben 48 (1996) 25–31. A.R. Webb, P. Weihs, M. Blumthaler, Spectral UV irradiance on vertical surfaces: a case study, Photochem. Photobiol. 69 (4) (1999) 464–470. G. Schauberger, Model for the global irradiance of the solar biologically-effective ultraviolet-radiation on inclined surfaces, Photochem. Photobiol. 52 (5) (1990) 1029–1032. G. Schauberger, Anisotropic model for the diffuse biologicallyeffective irradiance of solar UV-radiation on inclined surfaces, Theor. Appl. Climatol. 46 (1992) 45–51. P. Weihs, Influence of ground reflectivity and topography on erythemal UV-radiation on inclined surfaces, Int. J. Biometeorol. 46 (2002) 95–104. 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. O. Montenbruck, Astronomie mit dem Personal Computer, second ed., Springer Verlag, Berlin, Deutschland, 1965. A. Oppenrieder, P. Hoeppe, P. Koepke, J. Reuder, J. Schween, J. Schreder, Simplified calibration for broadband solar UV measurements, Photochem. Photobiol. 78 (6) (2003) 603– 606. 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 A. Oppenrieder a
a,b,*
, P. Hoeppe a, P. Koepke
b
Institut und Poliklinik f€ur Arbeits- und Umweltmedizin der Ludwigs-Maximilians-Universit€at M€unchen, Ziemssenstr. 1, 80336 M€unchen, Germany b 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. Ó 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 foundation 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 biological 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: +49-89-21804381. E-mail address: [email protected] (A. Oppenrieder). 1011-1344/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.jphotobiol.2003.11.008
but they also have significant influence on biological and chemical 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 horizontal plane with the sensitivity of the human skin. But generally the surfaces of biological bodies are not horizontally 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 irradiance on arbitrarily orientated planes has to be characterized. Only a few systematical UV measurements for inclined 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. Blumthaler 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 was higher 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) on Teneriffa, 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 irradiance on the south facing surface was greater than the irradiances on the surfaces facing the other directions. The most extensive measurements on inclined surfaces have been made by Schauberger [6,7]. He used measurements of erythemally weighted irradiance on inclined surfaces to calculate a correction in a model for calculating the erythemally weighted irradiance on inclined surfaces. The model needs the erythemally weighted irradiance on the horizontal surface, the albedo 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 Schauberger 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 inclined 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 data was 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 designed and built to measure in all kinds of environmental conditions, including extreme climatic conditions of high altitudes, and to produce data sets that provide reliable information about erythemally weighted UV radiation on inclined surfaces [9].
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 simultaneously measuring to allow the comparison of UV irradiances on inclined planes and the horizontal plane. The movable radiometer can be positioned in all directions 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 two minutes is a good compromise to get irradiance measurements for different orientations, that are dense enough 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 classified as constant for UVI measurements, if the fixed radiometer’s variations are less than 5% in the scanning period of 2 min. The constant sky conditions allow the comparison 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 the radiometer 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, position 26 turns the radiometer for the albedo measurement to the south and up side down, u ¼ 180° and h ¼ 90°. Finally the actual position of the sun is
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
Fig. 1. The measuring system ASCARATIS, movable and fixed erythemally weighting broadband radiometer, at the site Hoher Peissenberg.
A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94 Table 1 Positions (#) of the movable erythemally weighting radiometer #
u (°)
h (°)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
15 45 75 105 135 165 195 225 255 285 315 345 345 315 285 255 225 195 165 135 105 75 45 15 15 180 SAA
0 0 0 0 0 0 0 0 0 0 0 0 45 45 45 45 45 45 45 45 45 45 45 45 90 )90 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.
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 imports the data, calculates the UVI using the individual calibration 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 erythemally weighting 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 individual variation of the characteristics causes differences in the output of the radiometers under the same atmospheric conditions. A calibration is required to compensate these deviations.
87
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 the solar 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 absolute calibration factor has to be multiplied with the voltage 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 spectral sensitivities 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 radiometers is corrected by the calibration. Fig. 2 exemplarily shows the typical angular sensitivity of one of our radiometers recorded at one of the calibrations. The deviation 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 calibration factor depending on the solar elevation angle. The influence of differing angular sensitivity on the recording of the direct solar radiation part of the global UV irradiance is not very strong for horizontal radiometers, 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 greater than 65° are possible. We could separate direct and diffuse solar radiation with the standard setup of ASCARATIS and analyze the effects of the angular sensitivity 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 exceeds 8%. Fig. 3 exemplarily shows the consequence of the 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
88
A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94 1.20 1.15
Angular Sensitivity / cos (IA)
1.10 1.05 1.00 0.95 0.90 0.85 0.80 0.75 0.70 0
10
20
30
40
50
60
70
80
90
IA [˚]
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.
before/after
before
after
1.10
4.5
1.08
4.0
1.06
3.5
1.04
3.0 2.5 UVI
ratio
1.02 1.00 2.0 0.98 1.5
0.96
1.0
0.94
0.5
0.92 0.90 0
60
120
180
240
300
0.0 360
SAA [˚]
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 correction compensates the error caused by the non-ideal angular sensitivity. The deviations for high quality UVI measurements with horizontal erythemally weighting broadband radiometers are also within 8% [1]. Therefore 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 to allow 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
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 for the measurement dates lying in between the times of comparisons. All mentioned calibrations and corrections are performed 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 additional control 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 relevant parameters influencing the UV irradiance on inclined surfaces. The relevant parameters for this are solar elevation angle, amount and properties of clouds, turbidity, 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 urban conditions in a small town with rural environment and relatively warm climate (W€ urzburg), rural conditions in an area with agriculture (Frankendorf) and another in a higher elevation with pasture farming (Hoher Peissenberg). Urban conditions are found at the site in M€ unchen, where ASCARATIS was situated on the gravel roof of the building of the Meteorological Institute. A station (Schneefernerhaus) near a high mountain skiing 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 inclined surfaces for central Europe. Table 2 Description of the measuring sites Measuring site W€ urzburg Frankendorf M€ unchen Hoher Peissenberg Zugspitze
GLO (°E) 9.9 12.0 11.6 11.0 11.0
GLA (°E)
HASL (m)
Type of location
49.7 48.3 48.1 47.8 47.4
200 450 530 1000 2650
Urban Rural/agriculture Urban Rural/pasture Mountainous
GLO stands for geographical longitude, GLA for geographical latitude and HASL for height above sea level.
89
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 in M€ 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 middle graph 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° inclined plane (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 integrating 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 viewed hemisphere 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 diffuse irradiance. The transition from not influenced to influenced 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 southern orientation 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
A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94
UVI
90
9.0 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0
25 26 27
0
60
120
180
240
300
360
UVI
SAA [˚] 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0
60
120
180
240
300
360
UVI
SAA [˚]
9.0 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 0
60
120
180 SAA [˚]
240
300
360
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
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 winter day 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% more than 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 surfaces than for the horizontal surface. Two effects are responsible for this behavior. First the solar azimuth angle 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 orientations. 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 middle and lower graph of Fig. 5. The measurements of position 1–12 and 13–24 are quite similar. The same maximal UVI values are reached and comparing positions
A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94
91
0.30 0.25
UVI
0.20
25 26 27
0.15 0.10 0.05 0.00 0
60
120
180
240
300
360
SAA [˚] 0.30 0.25
UVI
0.20 0.15 0.10 0.05 0.00 0
60
120
180
240
300
360
SAA [˚]
0.30 0.25
UVI
0.20 0.15 0.10 0.05 0.00 0
60
120
180 SAA [˚]
240
300
360
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
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 nearly vanished. The contribution of the direct solar radiation has decreased significantly in winter. Figs. 6 and 7 show measurements in overcast conditions 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 in overcast 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 reflected from the ground is just a fraction of the amount of diffuse skylight. Thus the irradiance and in consequence 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 measurements 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 sky summer conditions (Fig. 4). In winter overcast conditions, 17 December 2001 (Fig. 7), the UVI measurements for all orientations are
92
A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94
1.2 1.0
UVI
0.8
25 26 27
0.6 0.4 0.2 0.0 0
60
120
180
240
300
360
SAA [˚] 0.5
UVI
0.4 0.3 0.2 0.1 0.0 0
60
120
180
240
300
360
UVI
SAA [˚]
1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0
60
120
180 SAA [˚]
240
300
360
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
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 orientated radiometer, with a value of 0.38. The reasons are the same as in summer overcast conditions. The measurements in the albedo position 26 reach an UVI of 0.22 (57.9% albedo). Compared to the clear sky conditions on 15 December 2001 the albedo has decreased from 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), however, the albedo is high and causes the reflection of a significant fraction of the diffuse skylight. Yet, the skylight 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 radiation reflected 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-
A. Oppenrieder et al. / Journal of Photochemistry and Photobiology B: Biology 74 (2004) 85–94
93
0.30 0.25
UVI
0.20
25 26 27
0.15 0.10 0.05 0.00 0
60
120
180
240
300
360
SAA [˚] 0.25
UVI
0.20 0.15 0.10 0.05 0.00 0
60
120
180
240
300
360
SAA [˚]
0.30 0.25
UVI
0.20 0.15 0.10 0.05 0.00 0
60
120
180
240
SAA [˚]
300
360
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
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 erythemally 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 effects 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 measurement days are now recorded by the three ASCARATIS systems. The broadband radiometers are calibrated in the best possible way and therefore provide 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 information 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 for horizontal and inclined planes at the measuring locations 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].
[3]
[4]
[5]
[6]
[7]
Acknowledgements The study is part of the Bavarian Research Network (Bayerischer Forschungsverbund: Erh€ ohte UV-Strahlung 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 financial and technical support at the measuring site Zugspitze.
[8]
[9]
[10] [11]
[12]
References [1] K. Leszcynski, K. Jokela, L. Ylianttila, R. Visuri, M. Blumthaler, Report of the WMO/STUK Intercomparison of ErythemallyWeighted Solar UV Radiometers, Report 112, World Meteorological 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. 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. M. Blumthaler, W. Ambach, R. Ellinger, UV-Bestrahlung von horizontalen und vertikalen Fl€achen im Hochgebirge, Sonderdruck aus Wetter und Leben 48 (1996) 25–31. A.R. Webb, P. Weihs, M. Blumthaler, Spectral UV irradiance on vertical surfaces: a case study, Photochem. Photobiol. 69 (4) (1999) 464–470. G. Schauberger, Model for the global irradiance of the solar biologically-effective ultraviolet-radiation on inclined surfaces, Photochem. Photobiol. 52 (5) (1990) 1029–1032. G. Schauberger, Anisotropic model for the diffuse biologicallyeffective irradiance of solar UV-radiation on inclined surfaces, Theor. Appl. Climatol. 46 (1992) 45–51. P. Weihs, Influence of ground reflectivity and topography on erythemal UV-radiation on inclined surfaces, Int. J. Biometeorol. 46 (2002) 95–104. 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. O. Montenbruck, Astronomie mit dem Personal Computer, second ed., Springer Verlag, Berlin, Deutschland, 1965. A. Oppenrieder, P. Hoeppe, P. Koepke, J. Reuder, J. Schween, J. Schreder, Simplified calibration for broadband solar UV measurements, Photochem. Photobiol. 78 (6) (2003) 603– 606. 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).