UV radiation in Novi Sad (Serbia): UV Index monitoring and variability of high erythemal UV radiation doses

Abstract

The UV monitoring in Novi Sad has been in operation by broadband Yankee UVB-1 biometer at the campus of the University of Novi Sad (45.33° N, 19.85° E, 84 m above sea level) since 2003. Data of UV index are taken every 30 s, averaged over 10 min intervals, and automatically saved in the database. The time series were reconstructed using an improved reconstruction technique that is based on parametric numerical model NEOPLANTA calculations of erythemal radiation and the empirical relationship between the erythemal doses and sunshine duration. In this study, we presented the maximum daily UV index values over the period 2003-2018 and the variability of reconstructed high erythemal UV doses (hUVery) over the period 1971-2018. Additionally, the influence of low total ozone column (TOC), low cloud cover conditions, and high surface albedo on hUVery was analyzed on a seasonal basis. Analyzing the measurements we concluded that maximum daily values are almost the same in the period of sixteen years, in the summer months maximum values are about 9. Based on the 90th percentile of each month we assessed a set of 1691 days with hUVery (10.65 %). The fraction of hUVery days in the last two decades was considerably larger (12.29%) than in the period before (6.97%) and during (8.68%) TOC depletion. We observed a statistically significant increase in the annual number of days with hUVery of +6.26 days/decade (p<0.01). The increase was statistically significant in all seasons, except winter, while it was the most pronounced in summer (+2.44 days/decade, p<0.01). However, the increase was not uniform over the entire observation period. Over the period of TOC depletion (1971-1980), the number of days with hUVery decreased, followed by a steep significant increase over the period of TOC depletion (1981-1996) of +16.75 days per decade (p<0.05). After 1996 hUVery increased slowly (+3.19 days/decade), but the trend was not statistically significant. Considering the influence of each UV affecting factor separately, the analysis showed that low cloud cover had slightly more influence on the occurrence of hUVery days than TOC in all seasons except winter. Of the total number of hUVery days, 89.47% were recorded when the cloud cover was low, while 80.65% of hUVery were recorded when the TOC was low. However, the most frequent reason for hUVery is the combination of these two factors (75% of all hUVery days). In the winter season-high surface albedo influenced the appearance of 18.26% of hUVery days. It is important to emphasize that we analyzed the reconstructed erythemal doses, not the measured data. Reconstructed data (implying the quality of the reconstruction) provide valuable resources for insight into UV variability in the past although cannot replace measurements. The results indicate even if the TOC recovers in the 21st century as expected, the appearance of days with high erythemal doses will still be significantly affected by future changes in the cloud cover.