This is Fred's final draft 20.05.2001


Norsang Gelsor (1), Fred Sigernes (2), Yngvar Gjessing (3), Ladislav Kocbach (4)

(1) Department of Math. & Physics, University of Tibet, Tibet, China;

(2) University Courses on Svalbard, Longyearbyen, Norway;

(3)  Institute of Geophysics, University of Bergen, Norway;

(4)  Institute of Physics, University of Bergen, Norway

An experiment has been conducted in Lhasa, the capital of Tibet, at an altitude of 3650 m above sea level, to test the protective capability of T-shirts from solar UV radiation. The experiment was conducted on the 9'th of June, 2000, with clear sky conditions using 6 different types of T-shirts, which were selected randomly according to color, fabrics and condition. The global solar UV irradiance and the corresponding transmission properties of the T-shirts were obtained by two identical instruments. The effective erythemal dose rate ratios between covered and uncovered instruments were measured to be in the range from 0.4 up to 6 %, depending mainly on  fabrics and thickness of the shirts. Estimates of the corresponding daily Tibetan erythemal UV dose show that  3 of the shirts exceed the 1 MED dose level during summer. Thin cotton shirts have daily doses in the range 0.2 to 1.2 MED. The highest daily dose was found to be above 2.5 MED for the mixed fabricated shirt labeled 6% polyester and 94% cotton. Thick and closely woven T-shirts of 100 % cotton have the best protective capability with daily doses less than 0.35 MED.    

During the last centuries, man's actions may have disturbed the ozone balance in the middle atmosphere. The main reason for the ozone depletion in the stratosphere is believed to be connected to the increase in anthropogenic compounds such as chlorine or bromine (cf. Brasseur and Solomon, 1986). These constituents are believed to affect the chemical and photochemical processes connected to the natural production and loss of ozone. One important feature of the stratospheric ozone layer is that it has the ability to absorb UV radiation emitted by the Sun. This feature makes the ozone layer a molecular shield which protects the Earth against UV radiation. As the ozone layer gets thinner, UV radiation at the surface of the Earth increases. If the ozone content decreases by 10 % during the spring and summer, the annual UV dose increases by about 12 %. Human risks connected to a prolonged or repeated exposure to the UV are sunburns, skin cancers, premature skin aging and eye damage (Tevini, 1993). The most common effect is sunburn, which indicates and warns us that our skin has been over exposed to ultraviolet radiation (especially UVA, wavelengths from 315 - 400nm). The more serious scenario is skin cancer related to an extreme over exposure to UVB (290 - 315 nm). Protecting the human body from the sun's harmful UV radiation is becoming more and more necessary for people who live in high altitude and low ozone concentration regions. The most common and simplest way to protect from UV radiation is to wear clothes.

Regions like the Tibetan Plateau, the largest and highest plateau in the word (average altitude of about 4000 m a. s. l.) has a much higher erythemal UV dose rate at ground level compared to most other regions. According to previous measurements done by Ci Ren et al. (1997), Lhasa's dose rate can reach up to 500 mW/m2 in the mid summer at solar noon. Figure 1.shows the yearly variation of UV radiation from July 1996 to December 1997. In summer, the dose rates are above the half maximum (approximately 250 mW/m2 )  in a period of  5 hours centered around local noon.

As a support, the ozone climatology measurements by the Total Ozone Mapping Satellite (TOMS) show that a region with low ozone concentration appears over the Tibetan plateau in summer ( Zhou et al., 1994; Han, 1996). Wheter the low ozone concentration over Tibet is a natural dynamic effect caused by the high mountain range,  the increased release of anthropogenic CFC gases or by the sun itself, is not yet understood.

Figure 1. Diurnal and daily variation of the erythemal UV dose rates in Lhasa during the period from June, 1996 to December, 1997 (Ci Ren et al., 1999).

A new site for ground based monitoring of UV radiation has been established at the University of Tibet, Lhasa, 1999. The site has been supplied with both broadband filter instruments and an imaging spectrometer (Sigernes et al., 2000). A yearly calibration have been carried out on these instruments to secure the quality of the data and support the already existing instrument that was installed at the Meteorological Bureau in Lhasa, 1996 (cf. Ci Ren, 1998). The aim is to monitor the global UV irradiance at ground level and the stratospheric ozone content to seek out if there is any detectable climatic trend. However, as an interesting thing and somewhat urgent for people who live and visit Tibet, a 'T-shirt/UV experiment'  was designed to provide in situ information on the protective property of T-shirts when exposed to high doses of UV radiation.

Two identical multi-channel filter NILU-UV Irradiance Meters were used in this study. The experiment was conducted at the roof of the Institute for Math & Physics, University of Tibet, two hours after local noon (14:32-15:48 LT) on the 9'th of June, 2000. The sky condition was partly cloudy with clear sky in the zenith. The instruments are produced by the Norwegian Institute for Air Research (Dahlback, 1996). Each instrument has six filters, five in UV and one in the visible part of the spectrum. The center wavelengths are 305, 312, 320, 340 and 380 nm, representing  the spectral regions classified as UVB (290 - 315 nm) and UVA (315 - 400 nm) of the solar spectrum. The Full Widths at Half Maximum (FWHM) are close to 10 nm. The sixth filter covers the whole visible region from 400 to 700 nm.

Figure 2. Experimental setup for the two NILU-UVs mounted at the roof of the Institute of Math & Physics, University of Tibet.

Teflon diffusors are used as front optics to the filters to obtain the global radiation.  Both instruments were calibrated prior to the experiment. The experimental setup is shown in figure 2.

The experimental procedure was first to check that both instruments measured the same intensities. Secondly, one of the instruments was covered by a T-shirt for about 10 minutes.  The transmission T was then calculated as the ratio between the intensity S of the covered and the intensity R of the uncovered instrument: 

where l is the center wavelength. The procedure was repeated for each T-shirt. Six T-shirts were selected randomly according to color, fabric, woven state, condition and number of volunteer owners.  Table 1.shows the specific characteristics of each T-shirt.
T-shirt  Qualitative description Fabrics Time [LT]
1 White; thin;  (new) 100% cotton 14:32-14:45
2 Light gray w/ dark pattern; thin 100% cotton 14:47-14:57
3 Black; thick; closely-woven 100% cotton 15:00-15:10
4 Light gray; normal thickness 94% cotton; 6% polyester 15:15-15:23
5 Brown; tightly-woven; thin 100% cotton;  15:27-15:35
6 Dark blue; thick; closely woven; (new) 100% cotton 15:38-15:48

Table 1. A list of the characteristics of the six T-shirts used in the experiment together with sample period. The shirts are in the Qualitative row labeled by color. White is thin and light colored cotton shirts. Light gray is thick dark closely woven cotton shirts, while dark gray represents the only shirt which is not marked 100% cotton (medium thickness). 

The biological effective dose rate of a NILU-UV covered by a T-shirt is calculated as a discrete sum

where Il is the measured intensity at center wavelength l. Note that the set of weights or dose coefficients al are calculated for each instrument by a procedure outlined by Dahlback (1996).


The aim of this study is to provide information on the transmission characteristics of a typical T-shirt in order to retrieve the corresponding effective UV dose that people are exposed to at the Tibetan Plateau.  The dose rates visualized in figure 1. are extremely high. According to the US National Oceanic and Atmospheric Administration (NOAA),  the UV Index  range from 0  to 15 or 16 (in the tropics at high elevations under clear skies). The UV index is divided into minimal (0 - 3), low (3 - 4), moderate (5 -6), high (7 - 9), and very high  ( > 10) levels. These levels are defined according to the time it takes to risk sun burns. At the Tibetan Plateau, the UV index is greater than 10 around solar noon from April to October, which indicate that the length of time to skin damage may be less than 10 minutes without protection. The maximum UV index in mid summer is measured to be as high as 20. This is to the authors knowledge the highest index levels measured in areas populated by humans.




CIE dose rate [mW/m2]

CIE dose rate w/ T-shirt

Dose rate ratio [%]

1 1.78 2.07 3.62 2.01 1.27  55.92




2 2.06 2.31 3.22 3.88 5.23  5.23




3 0.13 0.10 0.62 0.39 0.24   0.17




4 4.37 3.47 7.44 6.45 5.78  21.74




5 0.27 0.13 3.07 3.03 2.10  16.61




6 0.22 0.17 1.26 0.71 0.50   0.39




Table 2. Transmission coefficients for six T-shirts as a function of center wavelengths 305, 312, 320, 340 and 380 nm. Each with a bandwidth of approximately 10 nm.  The visible transmission factor covers the spectral region from 400 - 700 nm. The corresponding biological effective dose rates are calculated for each T-shirt.   The biological effective  ratio between the doses are also shown. White cells groups the thin and  light colored shirts, while light gray cells represents the thick closely woven dark shirts. Both groups are marked 100% cotton.  Dark gray represents shirt no. 4, which is light gray 94% cotton and 6% polyester (medium thickness). 

Table 2. shows the transmission factors calculated by equation (1) for each T-shirt used in the experiment. As expected, the transmission factors in the visible wavelength region seem to follow the color and thickness of the shirts. Thick dark colored T-shirts transmit less than thin light colored ones. The transmission in UV is in general much lower and more complex than in the visible. The state of the material becomes of equal importance. The thick dark colored and closely woven T-shirts 3 and 6 have a mean transmission of 0.3 and 0.6%, while the thin and light colored shirts 1, 2, and 5 are correspondingly up to approximately 10% more transparent. All of the shirts are labeled 100% cotton, except for T-shirt 4 which consist of 6% polyester and 94% cotton. This shirt stands out to have the overall highest transmission, especially in UVB.  Note that the transmission curve for all shirts increase with wavelength in UVB. In UVA, a local maximum transmission occurs at approximately 320 nm. T-shirt 2 is the only exception. It has an increasing transmission with a small local maximum shifted down to 312 nm. This effect is believed to be associated with the shirt's dark gray pattern. A combined effect of the transmission from light and dark gray areas of the shirt causes the local maximum at 320 nm to disappear.

The biological effective dose rates during the experiment dropped from 237 to 112 mW/m2. This effect is clearly consistent with Figure 1 both according to the UV dose level and the time of year. During calibration of the instruments the dose rate was as high as 420 mW/m2. The corresponding dose rates for the T-shirts in Table 2 were 10.4, 10.6, 1.3, 23.1, 5.2 and 2.5 mW/m2, respectively. Table 2 also shows the dose rate ratio between covered and uncovered instrument. This ratio represents the effective biological transmission rate of the shirts. For simplicity we assume this transmission rate to be constant as a function of solar zenith angle in order to retrieve an estimate of the daily doses wearing T-shirts. The monthly mean daily erythemal UV doses of figure 1 was calculated by Ci Ren et al. (1999) to vary between 2.2 to 7.6 kJ/m2 A maximum daily dose of 9.18 kJ/m2 was obtained in July 1996. Table 3 shows the corresponding daily doses for our T-shirts.  


UV dose 

T-shirt #1

T-shirt #2

T-shirt #3

T-shirt #4

T-shirt #5

T-shirt #6






















Table 3. The daily UV dose using six T-shirts as a function of typical daily erythemal UV doses measured in Lhasa, Tibet. The cells are colored according to color, thickness and fabrics of the T-shirts.

The doses are given in units of MED (Minimal Erythemal Dose). 1 MED  equals 210 J/m2 and is defined as the minimum UV dose level for perceptible reddening of previously unexposed human skin. Note that human individuals are not equally sensitive to UV radiation due to different pigmentation / ethnic origin. Human skin is often classified into four main groups according to the skin’s ability to protect itself  (Fitzpatrick, 1975; Sayre et al., 1981). The value of 1 MED for skin types I, II, III and IV are approximately 200, 250, 350 and 450 J/m2 , respectively. The 1 MED value for Tibetans are to the authors not know. Nevertheless, according to the International Radiation Protection Association (1991), the daily doses of the ultraviolet radiation should not exceed the limit of 1 MED.  From Table 3. it is clear that thick dark and closely woven T-shirts (100% cotton) protect human skin from solar UV even under extreme dose levels. The light colored and  thin cotton shirts do also protect us, but not in the extreme cases. T-shirt # 4 stands out to have the worst protective characteristics with daily doses above 2.5 MED. Compared to color and thickness of the 100% cotton shirts, the result must be due to the mixture of polyester and cotton.   

From the above in-situ experiment we recommend people that visit Tibet during summer, as a minimum  wear protective thick closely woven T-shirts labeled 100% cotton, especially individuals of skin type I and II. Thin or even medium thick T-shirts made of polyester and cotton should not be used. Further studies are need to retrieve the Minimal Erythemal Dose levels for the local population. A larger ensemble and a qualitative description of  the shirts should also be included to improve future T-shirt experiments in Lhasa.


Brasseur, G., and Solomon, S., Aeronomy of the Middle Atmosphere, 2nd ed. (D. Reidel Publishing Company, Dordrecht, Holland, 1986), pp. 1 - 7.

Ci Ren, P.B, Sigernes, F. and Gjessing, Y., Ground-based measurements of solar UV-radiation in Tibet: preliminary results, Geophys. Res. Lett., 24, 1359-1362, 1997.

Ci Ren, P.B, Gjessing, Y., and Sigernes, F., Measurements of solar ultra violet radiation on the Tibetian Plateau and comparisons with discrete ordinate method simulations, J. Atm. Terr. Phys., 61, 425, 1999.

Ci Ren, P.B., Solar Ultraviolet Radiation On the Tibetan Plateau: Measurements and Modeling, Ph.D. Thesis, Geophysical Institute, University of Bergen, Norway, 1998.

Dahlback, A., Measurements of biological effective UV doses, total ozone abundance and cloud effect with multi-channel moderate bandwidth filter instruments, Applied Optics,  Vol. 35, No. 33, 6514-6521, 1996.

Fitzpatric, T.B., Soleil et peau., J Med Esthet, Vol. 2, :33-34, 1975.

Han, Z., Seasonal variation and trends of TOMS ozone over Tibet, Geophys. Res. Lett., Vol. 23, 1029 - 1032, 1996.

Sayre, R.M., Desrochers, D.L., Wilson, C.J., and Marlowe, E., Skin type, minimal erythema dose (MED), and sunlight acclimatization,  J Am Acad Dermatol, Vol. 5,  439-443, 1981.

Sigernes, F., Lorentzen, D.A., Heia, K., and Svenøe, T., A multi-purpose spectral imager, Applied Optics, Vol. 39, No. 18, 3143-3153, 2000.

Tevini, M. (ed.), UVB radiation and ozone depletion: effects on humans, animal, plants, microorganism, and materials, Lewis Publishers, Boca Raton, 1993.

Zou, X., and Chao, L., Ozone valley over Tibetan Plateau, Acta Meteorologica Sinca, 8(4), 505 - 506, 1994.

International Radiation Protection Association / International Non-Ionizing Radiation Committee 1991a, Health issues of ultraviolet: A sun beds used for cosmetic purposes, Health Phys 61 (2), 285-288, 1991.