Int J Biometeorol DOI 10.1007/s00484-015-1005-y

ORIGINAL PAPER

The angular distributions of ultraviolet spectral irradiance at different solar elevation angles under clear sky conditions Yan Liu 1,2 & LiWen Hu 1 & Fang Wang 1 & YanYan Gao 1 & Yang Zheng 1 & Yu Wang 1 & Yang Liu 1

Received: 10 October 2014 / Revised: 14 April 2015 / Accepted: 26 April 2015 # ISB 2015

Abstract To investigate the angular distributions of UVA, UVB, and effective UV for erythema and vitamin D (vitD) synthesis, the UV spectral irradiances were measured at ten inclined angles (from 0° to 90°) and seven azimuths (from 0° to 180°) at solar elevation angle (SEA) that ranged from 18.8° to 80° in Shanghai (31.22° N, 121.55° E) under clear sky and the albedo of ground was 0.1. The results demonstrated that in the mean azimuths and with the back to the sun, the UVA, UVB, and erythemally and vitD-weighted irradiances increased with the inclined angles and an increase in SEA. When facing toward the sun at 0°–60° inclined angles, the UVA first increased and then decreased with an increase in SEA; at other inclined angles, the UVA increased with SEA. At 0°–40° inclined angles, the UVB and erythemally and vitD-weighted irradiances first increased and then decreased with an increase in SEA, and their maximums were achieved at SEA 68.7°; at other inclined angles, the above three irradiances increased with an increase in SEA. The maximum UVA, UVB, and erythemally and vitD-weighted irradiances were achieved at an 80° inclined angle at SEA 80° (the highest in our measurements); the cumulative exposure of the half day achieved the maximum at a 60° inclined angle, but not on the

* Yang Liu [email protected] 1

Department of Environmental Health, School fo Public Health, China Medical University, Shenyang, China

2

Department of Biomedical Engineering, China Medical University, Shenyang, China

horizontal. This study provides support for the assessment of human skin sun exposure. Keywords Ultraviolet radiation Solar elevation angle (SEA) Inclined angles Erythemal Vitamin D

Introduction Solar ultraviolet radiation (UVR) is an important environmental factor, and it has dual effects on human health. Overexposure to the sun’s rays is responsible for the majority of skin cancer (de Gruijl 1999; Rigel 2008), as well as other negative health effects, such as sunburn, skin aging, immunosuppression, and some forms of eye cataracts (Gallagher and Lee 2006). However, moderate UVB triggers vitamin D (vitD) synthesis (Moan et al. 2008) and possibly contributes to protection against breast cancer, prostate cancer, and nonHodgkin’s lymphoma (Luscombe et al. 2001). The evaluation of human ultraviolet exposure depends not only on the intensity of the ambient UVR but also on the postures of humans. Humans partake in multiple activities outside, and the postures are random; therefore, the horizontal UVR cannot quantify the actual UVR exposure level of the human body. Based on these reasons, some researchers have investigated UVR using vertical angles because they could better account for the position of the eyes or some areas of the skin and subsequently compared these data with horizontal UVR (Webb et al. 1999; Sobolewski et al. 2008; Hu et al. 2010b). Other researchers have measured UVR on inclined planes with fixed angles (Esteve et al. 2006; Utrillas et al. 2009; Navntoft et al. 2012); Oppenrieder et al. (2004), (2005) measured erythemally weighted UV irradiance at azimuth intervals of 15° on 0° and 45° inclined planes by the use of a fully automatic measuring system, an angle scanning

Int J Biometeorol

radiometer. Casale et al. (2012) used south-facing polysulphone (PS) dosimetry to measure erythemal intensity on the seven inclined planes of which the angular intervals were 15°. The former study focused on the influence of azimuths, while the latter study focused on the influence of inclined angles under different weather conditions. At the same time, some researchers have placed detectors on manikins or a human body to measure the UVR on inclined planes. The inclined angles could represent eye position (Sasaki et al. 2011; Gao et al. 2012; Hu et al. 2013) or anatomical parts of a manikin body (Parisi et al. 2003; Wright et al. 2004; Hu et al. 2010a); thus, the measurements could reflect exposure levels of specific body areas. Downs and Parisi (2009) used miniaturized PS dosimeters to investigate the relation of human anatomical UV exposure and skin cancers and subsequently studied mean exposure fractions in relation to solar zenith angle (SZA) in different ambient climates (Downs and Parisi 2012). Other investigators have placed UV dosimeters on the bodies of cyclists (Kimlin et al. 2006), farmers (Schmalwieser et al. 2010), and individuals who performed various activities (Weihs et al. 2013) to reveal actual levels of UV exposure. According to the measurements of UVR, some mathematical models have been studied to calculate the UVR exposure at inclined angles (Bird 1986; Schauberger 1990; Serrano et al. 2012). Pope and Godar (2010) developed a model to convert horizontal UVR to a cylinder to simulate human exposure. Vernez et al. (2011) used a numeric model of 3D human to simulate an individual’s solar ultraviolet exposure and then calculated the anatomical exposure patterns of the skin (Vernez et al. 2012). Some researchers used human models to study the relation of body postures, solar elevation, and azimuth angles (Kubaha et al. 2004; Park and Tuller 2011). Seckmeyer et al. (2013) used a mathematical method to calculate vitD-weighted exposure via integration of the incident solar spectral irradiance over all relevant parts of the human body. Miyauchi et al. (2013) used a numeric model to estimate the solar time required for vitD synthesis in the human body and observed locations in Japan to verify the model. There are many academic researchers working on the UV spectra at inclined angles. Parisi and Kimlin (1999) compared horizontal and sun-normal spectral biologically effective ultraviolet irradiance (erythema, actinic, and deoxyribonucleic acid (DNA)) at solar elevation angle (SEA) from 26° to 55°. Schouten and Parisi (2011) measured the UV spectra at seven inclined angles in which the intervals were 15° in four directions (E, W, N, and S) at SEA from 32° to 42°. Baczynska et al. (2013) used the same measurements at the local solar noon when SEA was approximately 45°, and the results of the erythemally and vitD-weighted irradiances both closely followed the behavior of the UVB component of the spectra. We investigated the change of UV spectra at the inclined angles at different SEAs and azimuths. This current research measured UV spectra at ten inclined angles with 10° angle

intervals in seven directions that covered 180° azimuths and transferred the spectra to erythemally and vitD-weighted irradiances. We investigated the angular distributions of UVA (320–400 nm), UVB (300–320 nm), and erythemally and vitD-weighted irradiances at a wide range of SEAs, which included six times from 18.8° to 80°, and the results indicated different trends, especially at lower SEAs. The results could be combined to human exposure postures and supply a database of human ultraviolet exposure evaluations.

Materials and methods Measurement device The chosen solar UV sensor was a fiber optic spectrometer (AvaSpec-2048x14-2-USB2, Netherlands). In this study, the detector was placed on a tripod that was approximately 1 m from the ground. The tripod head could rotate horizontally and vertically. To avoid the legs of the tripod affecting the measurements, a shelf was fixed on the tripod head and the detector was placed in front of the shelf. The fiber optic spectrometer (AvaSpec-2048x14-2-USB2, Netherlands) with high-quantum efficiency and high-UV sensitivity was used to measure UV spectral irradiance. The spectrometer is based on the AvaBench-75 symmetrical CzernyTurner design with a 2048 pixel back-thinned CCD detector array, and the advantage of the CCD detector is the good UV sensitivity, combined with good S/N and dynamic range. The full width at half maximum (FWHM) resolution of the spectrometer is 2.0 nm, and the stray light is less than 0.1 %. The signal-to-noise ratio is 500 dB. The spectrometer has a diffraction grating with 1200 lines per millimeter and a slit size of 200 μm. A USB2 interface has ultrafast data sampling at 450 spectra per second and data transfer at 2.24 ms. The detector has a cosine corrector (CC-UV⁄VIS) with an active area of 3.9 mm and could accept light from a 180° solid angle. For absolute irradiance measurements, the spectrometer was configured and radiometrically calibrated over a range from 200 to 400 nm. The calibration was performed by the National Physical Laboratory (NPL) before the experiment (Gao et al. 2012). Measurement location The study was conducted in the Pudong District, Shanghai city (31.22° N, 121.55° E, 0 m a.s.l.), China. In the experiment, the experimental apparatus was located on the roof of a three-story building with an unobstructed view. The ground was concrete. The measurements were performed on sunny days with clear or only slightly cloudy skies. We conducted the study for two whole days on July 9 and 10, 2013 in Shanghai, so there are four groups of data corresponding to SEA. Through comparing and analysis of ambient irradiance, as is shown in Fig. 1, the irradiances in the morning of July 10 were best with SEAs, so we choose this part of data. We measured the surface reflectivity with the detector almost vertically downward, and the distance from

Fig. 1 The ambient irradiance changed with SEA, for in the morning on July 9 (a), in the afternoon on July 9 (b), in the morning on July 10 (c), and in the afternoon on July 10 (d)

UVR (µWcm-2)

Int J Biometeorol 6000 5000 4000 3000 2000 1000 0

UVR (µWcm-2)

(a)

6000 5000 4000 3000 2000 1000 0 0 10 20 30 40 50 60 70 80 90 SEA (˚)

10 20 30 40 50 60 70 80 90 SEA (˚)

6000 5000 4000 3000 2000 1000 0

6000 5000 4000 3000 2000 1000 0 (c)

0 (b)

0 10 20 30 40 50 60 70 80 90 SEA (˚)

the ground was 0.5 m. The mean albedo of UVR and UVB are 0.11 and 0.1, respectively. The site has a relatively unpolluted atmosphere, where the daily air quality index (AQI) was 68 (http://www.semc.com.cn/home/index.aspx). The atmosphere ozone level was 303 DU (National Aeronautics and Space Administration, NASA; http://ozoneaq.gsfc.nasa.gov/tools/ ozonemap/). UV irradiance measurements of inclined angles The direction of measurements was initiated facing toward the sun, and the inclined angles were changed from 0° to 90° with intervals of 10°. Because only the top of the detector received solar irradiance, the inclined angle was the angle between the normal receiving surface and the horizontal; thus, an inclined angle of 90° indicates the horizontal measurement (Fig. 2). The ten measurements represent the irradiations at different tilted angles, and they were conducted within 1.5 min; thus, they could be considered to be at the same SEA. After facing the sun measurements, the horizontal azimuth angle was changed by 30° in an anticlockwise manner, and measurements were subsequently conducted from 0° to 90° vertically. The Btrue^ position facing toward the sun was considered the reference angle and was considered to be 0° relative azimuth angle. Therefore, a point due opposite from the sun had a relative azimuth angle of 180°. The rotation increment of the angle was 30° from facing the sun to the opposite of the sun (Fig. 2). At any direction, the measurements of tilted angles from 0° to 90° were conducted; thus, one rotation had 70 measurements. The rotation could be measured within nearly 10 min, and the SEA may change a little. The measurements were conducted from 6:30 to 11:30 (China Standard Time (CST)), and the sampling interval was 1 h. The SEA correspondingly ranged from 18.8° to 80°. To compare the irradiance at inclined angles of different azimuths, the change of SEA in each rotation was corrected by the method of linear interpolation. From the first

(d)

0 10 20 30 40 50 60 70 80 90 SEA (˚)

measurement facing toward the sun to the last measurement opposite the sun, the time elapsed was 10 min; thus, the measurements behind the sun were pushed forward with a correspondingly changed SEA based on the data of two adjacent samplings by the use of linear interpolation. The same method was used in other directions, but the time consumed was only 1.5 min in the adjacent measurement. The UV irradiance for every inclined angle was calculated from the integration of the UVA and UVB bands. Overall, there were 70 groups of irradiance data per rotation. Calculation of biologically effective UV (UVBE) irradiance at inclined angles To analyze the biologically effective UV irradiance at different angles, the UVBE was calculated as follows:

90º 70º 80º

inclined angles

60º

Received surface was horizontal

50º 40º 30º 20º SEA

10º 0º

180º 30º

60º 120º 90º relative azimuth angles

150º

Fig. 2 For each measurement, the spectrometer detector, which was mounted on a tripod, was initially positioned facing toward the sun. The inclined angle between the detector and the horizontal surface ranged from 0° to 90° with intervals of 10°, and the detector collected data at every inclined angle. The tripod was rotated anticlockwise in increments of 30° from 0° to 180°, which was the relative azimuth angle between the spectrometer detector and the sun. At every relative azimuth angle, the spectrometer detector collected data ten times at ten different inclined angles using the previously described method. The receiver surface of the detector was on a plane that was perpendicular to the inclined angles

Int J Biometeorol

(a) 30

ðICNIRP 2004Þ

where H is the cumulative UV exposure, SUV is the UV irradiance, and t is the exposure time. In our study, t is 1 h, and SUV is the mean value of two measurements. We calculated the UVexposure of 1 h, and H is the cumulative UV exposure of a half day from 6:30 to 11:30 CST.

UV intensity(μWcm-2nm-1)

where S(λ) is the measured spectral irradiance, A(λ) is a particular biological process with an action spectrum, and Δλ is the bandwidth of nanometers of the intervals. In this study, the action spectra for erythema (Webb et al. 2011) and vitD synthesis (CIE 2006) were employed, which ranged from 300 to 400 nm. The respective action spectra were linearly interpolated between the 1 nm data points. The cumulative UV over time could be further calculated based on the following formula: X H¼ SU V t

0˚ Solar elevation angle 18.8e Time 6:30 CST

20˚ 40˚

25

60˚

374

312

90˚

20 15 10 5 0 300

310

320

330

340

350

360

370

380

390

400

wavelength(nm)

(b) 100

Solar elevation angle 43e Time 8:30 CST

90 80

UV intensity(μWcm-2nm-1)

U V BE ¼ ∑S ðλÞ AðλÞ Δλ

70 60 50 40 30 20 10 0 300

Results

310

320

330

340

350

360

370

380

390

400

wavelength(nm)

120

UV intensity(μWcm-2nm-1)

At a higher SEA (>60°), the spectral irradiance increased with an increase in the inclined angles; thus, the spectra changed in parallel at different inclined angles. However, at a lower SEA, there were crossing points in the curves of the spectra at different inclined angles. For example, at SEA 18.8°, the spectra at the inclined angles of 0° and 90° had two crossing points: They crossed at 312 and 374 nm. It indicated that if the wavelength was less than 312 nm, the perpendicular irradiance was higher compared with the horizontal irradiance; if the wavelength was greater than 312 nm and less than 374 nm, the horizontal irradiance was higher than the perpendicular; and if the wavelength was more than 374 nm, the horizontal irradiance was lower (see Fig. 3). The spectra at other inclined angles also crossed, which indicated that the wavelength had different distributions of intensity at different inclined angles. To demonstrate the change in UVB on inclined surfaces clearly, the wavelength irradiance was normalized by horizontal values (the inclined angle 90°) (see Fig. 4). The normalized values were the relative values of inclined surfaces and the horizontal. At a lower SEA, the normalized values of smaller wavelengths (300–320 nm) decreased sharply, and the values slowly increased with the wavelength. The relative maximum appeared at the shortest wavelength and the inclined angle nearest to SEA, such as at SEA 18.8°, whereas the maximum was 4.77 at 300 nm, 10° inclined angle. With an increase in

(c) Solar elevation angle 68.7e Time 10:30 CST

100 80 60 40 20 0 300

310

320

330

340

350

360

370

380

390

400

wavelength(nm)

(d) 120 UV intensity(μWcm-2nm-1)

The UV spectral irradiance of inclined angles when facing toward the sun

Solar elevation angle 80e Time 11:30 CST

100 80 60 40 20 0 300

310

320

330

340

350

360

370

380

390

400

wavelength(nm)

Fig. 3 The UV spectrums of different inclined angles when facing towards the sun, for SEA 18.8° (a), SEA 43° (b), SEA 68.7° (c), and SEA 80° (d)

Int J Biometeorol

5

0˚ 10˚ 20˚ 30˚ 40˚ 50˚ 60˚ 70˚ 80˚ 90˚

Solar elevation angle 18.8 Time 6:30 CST

4 3 2 1

2.5

1.5 1 0.5

0

1.2

300 310 320 330 340 350 360 370 380 390 400 wavelength(nm)

(b)

Solar elevation angle 68.7 Time 10:30 CST

1.2

1

normalized value

normalized value

0 300 310 320 330 340 350 360 370 380 390 400 wavelength(nm)

(a)

Solar elevation angle 43 Time 8:30 CST

2 normalized value

6

normalized value

Fig. 4 The irradiance at inclined angles normalized by horizontal when facing toward the sun, for SEA 18.8° (a), SEA 43° (b), SEA 68.7° (c), and SEA 80° (d)

0.8 0.6 0.4

Solar elevation angle 80 Time 11:30 CST

1 0.8 0.6 0.4 0.2

0.2

0 300 310 320 330 340 350 360 370 380 390 400

(d)

wavelength(nm)

change was only 10°, and the short and long wavelengths had the same inclined angle at which the maximum irradiance appeared. At SEA 68.7° and 80°, almost all wavelengths had the maximum value at the inclined angle, which was the same as the SEA. The angular distributions of wavelength were influenced not only by SEA but also by the background, especially at lower SEAs when the direct irradiance was less. The ground reflected more irradiance at lower inclined angles.

80 70 Solar elevation angle

60 50 40

14.9˚

30

18.8˚

20

29.4˚

10 300 307 314 321 328 335 342 349 356 363 370 377 384 391 398

0

(a)

wavelength (nm) Solar elevation angle

90 80 70 60 50 40 30 20 10 0

43˚ 55.6˚ 68.7˚ 80˚

301 308 315 322 329 336 343 350 357 364 371 378 385 392 399

SEA, the relative maximum decreased. At SEA 43°, the maximum was 2.25 at 300 nm, 40° inclined angle, and at SEA 68.7°, it was 1.14 at 303 nm, 70° inclined angle. When the SEA was greater than 60°, the normalized values on different inclined surfaces paralleled each other and did not substantially change with the wavelength. The figure clearly demonstrated that the change of shortwave irradiance on inclined surfaces was substantially more compared with the horizontal; thus, the horizontal values could not represent the inclined surfaces very well. At every SEA, each wavelength had different values on different inclined surfaces. The inclined angles when the maximum irradiance at each wavelength appeared were recorded and are summarized in Fig. 5. At a lower SEA (

The angular distributions of ultraviolet spectral irradiance at different solar elevation angles under clear sky conditions.

To investigate the angular distributions of UVA, UVB, and effective UV for erythema and vitamin D (vitD) synthesis, the UV spectral irradiances were m...
490KB Sizes 0 Downloads 7 Views