**2.2 Instruments and measurments**

100 Solar Radiation

variations in UV radiation and the ratio of UV to global solar radiation (*R*s). Additional studies have addressed long-term trends in the variations of UV through reconstructions of past UV radiation based on ground-based and satellite data (Kaurola et al., 2000; Fioletov et al., 2001; Lindfors et al., 2007; Feister et al., 2008;Hu et al., 2010a). In the last few decades, there has been a progressive increase and great concern in the amount of UV reaching the Earth's surface as a consequence of the thinning of the stratospheric ozone. Despite its anthropogenic importance and impacts, concern about the amount of UV radiation reaching the Earth's surface has only recently been developed, primarily as a result of the thinning of ozone layer linked to the depletion of stratospheric ozone in the 1980s (Su et al., 2005). UV

The objective of this chapter, apart from showing seasonal variations of UV and UV*/R*<sup>s</sup> values in Beijing, based on the reconstruction method, is to develop a long-term data set of

Beijing, the capital of the People's Republic of China, is located at 39°56' N latitude and 116°20' E longitude. East, North and West of Beijing are surrounded by mountains. The climate of Beijing is an East Asia monsoon type, with cold and dry winters, and hot and humid summers. During winter, the Siberian air masses that move southward across the Mongolian Plateau are accompanied by cold and dry air. In summer, the air mass is hot owing to warm and humid monsoon winds from the southeast, bringing most of the annual

The East Asian Monsoon season is the dominating climate of Beijing. In spring (March, April, May), the content of the water vapor in the atmosphere is low and there is little rain in this region. The rainy season begins in June, ends in August and then comes in the dry season. The main rainfall is in July. In autumn (September, October, November), the sky condition is always clear; for Beijing, it is often controlled by the anticyclone. In winter (December, January, February), Siberian anticyclones frequently take place in the Beijing area. Rainless and cloudless conditions are the dominant sky conditions in the region. In this paper, spring and winter are called the dry season because there are few rainfall events in these months. . Summer and autumn are called the humid season, for most of the rainfall

The monsoon starts in July, and ends in October when the dry season begins. There are many active synoptic systems (depressions) in the humid period,. while stabilization system

In the humid season, the southern wind from the ocean prevails bringing abundant vapor in to the atmosphere, thus the vapor content of the atmosphere is high in this season. Water vapor markedly affects the long wavelength radiation by absorbing them, leaving the UV spectral portion and the short wavelength spectral radiation for possible scattering and reflection. . Consequently, the general decrease in the global radiation could cause the ratio

control prevails in the dry season and most days are clear in this period.

of UV to global radiation to increase as the water vapor increases.

radiation-measuring networks are extremely scarce, particularly in China.

UV radiation, and also study variation characteristics of UV in Beijing.

**2. Methods** 

precipitation in Beijing area.

occur in these two seasons.

**2.1 Site** 

A solar radiation observation system was set up on the top of the two-floor building (10 m). Rs, PAR, direct radiation, diffuse radiation, concentration and meteorology parameters(temperature, relative humidity, air pressure) measurements are being carried out in this observatory.. Rs is measured by using a Kipp&Zonen radiometer CM-21 (Delft, The Netherlands). Rs measurements have an estimated experimental error of 2-3%. UV radiation (290–400 nm) is measured using CUV3 radiometers (Kipp & Zonen, Delft, Netherlands) with an accuracy of 5%. All radiation values were recorded at 1-min intervals, and an hourly average value was obtained by integrating the 1-min values. Temperature and relative humidity were measured with HMP45D(Vaisala, Finland),the accuracy of temperature and relative humidity are 0.1 and 3% respectively. The solar radiation parameters observation system is completed with a data acquisition system (Vaisala M520, Finland).

All pyranometers were calibrated by using the 'alternate method' (*Bruce*, 1996). During the process of calibration, we were required to take on-site measurements of global, diffuse, and direct (pyrheliometer) sensor voltages in clear and sunny conditions. The pyrheliometer was calibrated against a reference pyranometer, which had been calibrated against a standard pyrheliometer (PMO6), i.e. absolute irradiance radiometer (Switzerland). This absolute irradiance radiometer is periodically calibrated every five years at the World Radiation Center in Davos, Switzerland.

Manufacturers usually calibrate UV sensors using standard lamps with a known spectral irradiance. The calibration of the UV3 sensor was conducted with a standard light source in standard spectral irradiance that can be traceable to the National Bureau of Standards lamp. A spectroradiometer measures a standard lamp spectral irradiance, and then retrieves the spectral sensitivity under standard lamp conditions. By using the same method, we could deduce the spectral sensitivity under sunshine conditions (equation 2).

$$K\_f^D = \frac{V\_{D,\Lambda\lambda}}{\lambda} \Big/\_{E\_{D,\Lambda\lambda}} = \tau\_{\Lambda\lambda} \cdot S\_{\Lambda\lambda} \tag{1}$$

where *Kf <sup>D</sup>* is the spectral sensitivity of the spectroradiometer under standard lamp conditions, *VD,*△λ is the respond voltage in response to the standard lamp in △λ, and *ED,*△λ is the standard irradiance of the standard lamp.

$$K\_f^S = \begin{aligned} \prescript{}{}{V\_{S,\Lambda\dot{\lambda}}} \Big\langle \prescript{}{}{}{V\_{S,\Lambda\dot{\lambda}}} = \tau\_{\Lambda\dot{\lambda}} \cdot S\_{\Lambda\dot{\lambda}} \end{aligned} \tag{2}$$

where *Kf <sup>S</sup>* is the spectral sensitivity of the spectroradiometer under sunshine conditions, *VS,*△λ is the respond voltage in response to the standard lamp in △λ, and *ES,*△λ is the standard irradiance of solar radiation.τ*S,*△λ is the transmittance of the fiber optic extension cord.

In narrow wavebands, the spectral sensitivity *Kf <sup>D</sup>*is equal to *Kf <sup>S</sup>*, and thus the spectroradiometer and standard lamp can be used to calibrate the quantum sensor. The spectral sensitivity *Kf <sup>D</sup>*for each narrow waveband can be derived from the lamp spectral irradiance. Then, this spectroradiometer can be used to measure the sun irradiance and the integral sun spectral irradiance between 290–400nm to calculate UV. At the same time, the UV sensor to be calibrated measures the respond voltage. The difference between measured results of the quantum sensor calibrated by the spectroradiometer and the new quantum sensor is not more than 1%.

Quality control of the UV radiation measurements was based on two main principles: that observed UV radiation should be less than the extraterrestrial UV radiation at the same geographical location, and that the range of the ratio of UV radiation to *Rs* should be limited to values between 0.02 and 0.08. Elhadidy et al. (1990) noted that the UV radiation/*Rs* ratio can be as low as 0.02 on days experiencing high dust levels, while the UV radiation/*Rs* ratio at the top of atmosphere is 0.08 as derived from integral sun spectral irradiance (Geuymard, 2004). Therefore, the UV radiation/*Rs* ratio should be between 0.02 and 0.08. Values outside this range were flagged as questionable and excluded from the data set. About 1% of the measurements were eliminated based on these quality control processes. Quality control for *Rs* was similar to that for UV radiation, and the smallest acceptable value of the ratio of *Rs* to extraterrestrial global solar radiation was 0.03 (Geiger, et al., 2002).

The extraterrestrial *UV*0 can be derived from equation 1 expressed as follow:

$$dU\_0 = I\_{scav} (\frac{12}{\pi \rho^2}) \int\_{\alpha\_l}^{\alpha\_2} \sin \alpha \qquad d \text{ or } \tag{3}$$

where α is the solar declination, ω is the hour angle and ρ is correction factor of the Earth's orbit.

$$I\_{scw} = 78 \text{ W m}^2 \text{\textdegree}^2 \text{\textdegree}^2 \tag{4}$$

*Iscuv* has been obtained from the spectral values given by Frohilich and Wherli (Lenobe,1993). *Ks* is the ratio of *Rs* to extraterrestrial broadband solar radiation (*H*0) given as follow:

$$K\_s = \bigvee\_{H\_0}^{R\_s} \!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\/,\tag{5}$$

is the respond voltage in response to the standard lamp in

,

spectroradiometer and standard lamp can be used to calibrate the quantum sensor. The

irradiance. Then, this spectroradiometer can be used to measure the sun irradiance and the integral sun spectral irradiance between 290–400nm to calculate UV. At the same time, the UV sensor to be calibrated measures the respond voltage. The difference between measured results of the quantum sensor calibrated by the spectroradiometer and the new quantum

Quality control of the UV radiation measurements was based on two main principles: that observed UV radiation should be less than the extraterrestrial UV radiation at the same geographical location, and that the range of the ratio of UV radiation to *Rs* should be limited to values between 0.02 and 0.08. Elhadidy et al. (1990) noted that the UV radiation/*Rs* ratio can be as low as 0.02 on days experiencing high dust levels, while the UV radiation/*Rs* ratio at the top of atmosphere is 0.08 as derived from integral sun spectral irradiance (Geuymard, 2004). Therefore, the UV radiation/*Rs* ratio should be between 0.02 and 0.08. Values outside this range were flagged as questionable and excluded from the data set. About 1% of the measurements were eliminated based on these quality control processes. Quality control for *Rs* was similar to that for UV radiation, and the smallest acceptable value of the ratio of *Rs* to

2

 , (3)

is the hour angle and ρ is correction factor of the Earth's

78 *scuv I* W m-2, (4)

*<sup>R</sup> <sup>K</sup> <sup>H</sup>* , (5)

*Iscuv* has been obtained from the spectral values given by Frohilich and Wherli (Lenobe,1993).

*<sup>s</sup> <sup>s</sup>*

0

*Ks* is the ratio of *Rs* to extraterrestrial broadband solar radiation (*H*0) given as follow:

,

△λ

*S S <sup>f</sup> <sup>S</sup> <sup>V</sup> <sup>K</sup> <sup>S</sup> <sup>E</sup>* 

is the respond voltage in response to the standard lamp in

*<sup>D</sup>* is the spectral sensitivity of the spectroradiometer under standard lamp

*<sup>S</sup>* is the spectral sensitivity of the spectroradiometer under sunshine conditions,

*<sup>D</sup>*for each narrow waveband can be derived from the lamp spectral

(2)

*<sup>D</sup>*is equal to *Kf*

△λ

is the transmittance of the fiber optic extension

△λ

, and *ES,*

, and *ED,*

△λ

*<sup>S</sup>*, and thus the

is the

△λ

where *Kf*

where *Kf*

*VS,*△λ

cord.

spectral sensitivity *Kf*

sensor is not more than 1%.

where α is the solar declination,

orbit.

conditions, *VD,*

△λ

is the standard irradiance of the standard lamp.

standard irradiance of solar radiation.τ*S,*

In narrow wavebands, the spectral sensitivity *Kf*

extraterrestrial global solar radiation was 0.03 (Geiger, et al., 2002).

ω

The extraterrestrial *UV*0 can be derived from equation 1 expressed as follow:

<sup>1</sup> <sup>0</sup> <sup>2</sup> <sup>12</sup> ( ) sin *UV Iscuv <sup>d</sup>* 

All of the calibration works were done at the beginning and the ending of the data collection. At the start and end of data collection, all measurement sensors were cross evaluated in Beijing. The maximum deviation of CM-11 sensors averaged 1.5% (the average deviation was 1.2%), and the maximum deviation of CUV3 sensors averaged 2.42% (the average deviation was 1.8%).
