Long-term variations of UV-B doses at three stations in northern Europe

Citation
J. Kaurola et al., Long-term variations of UV-B doses at three stations in northern Europe, J GEO RES-A, 105(D16), 2000, pp. 20813-20820
Citations number
17
Categorie Soggetti
Earth Sciences
Volume
105
Issue
D16
Year of publication
2000
Pages
20813 - 20820
Database
ISI
SICI code
Abstract
Recent analysis of the total ozone observations indicate a negative trend o f about 4%/decade in the Northern Hemisphere midlatitudes during the last t wo decades [WMO, 1999]. The effect of this decline on surface UV levels is of interest to a variety of applications. In this work the long-term variat ion of UV radiation at three stations located in northern Europe (Belsk, No rrkoping, and Jokioinen) has been studied using data from (1) ground-based observations, (2) surface UV doses determined using TOMS satellite measurem ents, and (3) reconstructed UV doses using observations of global radiation , total ozone, and radiative transfer modeling. For each station the estima tes of daily UV doses from various sources have been intercompared, and a t rend analysis has been performed to reveal long-term changes in the UV radi ation. Data sets, which start in the late 1970s or early 1980s, show a gene ral positive trend in annual doses of UV radiation. Some of these upward tr ends are statistically significant. For Belsk the increases are in the rang e of 5-15% per decade during spring and summer. The largest increases, abou t 20%/decade, has been observed in Norrkoping during spring. At Jokioinen t here has been a slight upward trend in UV throughout the year. The analysis of reconstructed Belsk data from 1966 onward shows that the positive trend since late 1970s was preceeded by a negative trend. The reason for such ch anges is probably not only related to the changes in the total ozone but al so to changes in aerosol content and cloudiness. The agreement of the UV se ries based on different data sources is good. This was studied using a subs et of data in which it was required that data from all possible sources wer e available. The different trend estimates were in very close agreement wit h each other. However, there were often differences in absolute values, whi ch is probably related to problems in calibration and limitations of the mo dels.