Variability of the UV-Index in Paramaribo, Surinam.

The ultimate intention for this research is to better understand the variability of the UV-index (UVI) in the Tropics and also to be able to improve the prediction of the UVI at Paramaribo (5.81°N, 55.21°W), Surinam. This is the research question: What is the variability of the UV-index and what is its cause in Paramaribo under cloudless circumstances and corrected for the total ozone column, SZA and Earth-Sun distance? The ozone profile and aerosols are known to have an effect on UVI, but have not been investigated for Paramaribo. Cloud effects on UVI also have not yet been studied in Paramaribo, but for this research it is too complex to look at the combined effects of clouds, ozone profile and aerosols. Therefore, we restrict ourselves to cloudless situations. The sub-research question is: To what extent is the empirical algorithm, used to predict the UVI in Paramaribo, valid for cloudless moments? If not, what is a typical deviation of the measurements from the algorithm? What is the cause of this deviation?

In Chapter 1 a short description is given of the effects of UV radiation for human health. Further, an explanation of UVI, our objectives and research questions. In Chapter 2 information is given about Paramaribo station and afterwards a more technical description of the UVI and its affecting parameters. These parameters are: surface altitude, surface albedo, solar zenith angle (SZA), Earth-Sun distance, clouds, total ozone column, ozone profile, stratospheric temperature, aerosols and trace gases. In Chapter 3 the selection of cloudless situations is described. This is done in three steps. The first selection is based on known effects of clouds on UVA. An upper and lower limit is chosen for the irradiation (W/m 2 /nm) at two wavelengths in the UVA. The second selection is based on comparison of the UV-spectrum with a reference spectrum. The third selection is a check of the remaining situations with pictures of a total Sky Imager (TSI) and the solar Radiation Station. Only few selected (totally cloudless) situations remained after selection 3. For 9 months with pictures of TSI and 11 SZA ranges of 1º, only 58 cloudless situations were found. Data of tropospheric ozone and aerosols were also very scarce. That is why we decided to use also the remaining situations after selection 2 (‘cloudy’ but ‘sunny’ situations) for this research. In Chapter 4 results are given of the variability of the UVI and its affecting parameters. First, the variability of the absolute (measured) UVI is considered. The measured UVI variability at Paramaribo is defined within SZA ranges of 1º. The variability of the UVI is much higher for the situations after selection 2 than after selection 3. The maximum variability for a SZA range after selection 2 is 4 points of UV-index. For selection 3, it is about 1 point of UV-index. This difference is due to effects of clouds after selection 2, which are eliminated in selection 3. Clouds can cause an increase as well as a decrease of the UVI.

Second, the variability of the relative UVI is considered. In this report this is defined as the measured UVI divided by the calculated UVI. The UVI is calculated using the empirical algorithm with the Earth-Sun distance, the total ozone column and SZA as input parameters. After selection 2 the relative UVI varies typically between 0.8 and 1.1; after selection 3 typically between 0.85 and 1.05. The smaller range after selection 3 is again due to eliminating the cloud effects in this selection. The expectation was that total ozone column and SZA are not related with the relative UVI. However, our results suggest that for cloudless situations the relative UVI increases with SZA. For the situations after selection 2 as well as selection 3 the average relative UVI is 0.96. This suggests a calibration error of the Brewer (the instrument measuring UV-radiation).

For quantifying the effects of the ozone profile on the UVI, the tropospheric ozone column is used as parameter. The tropospheric ozone column (0-5 km) at Paramaribo varies between 7 and 16 DU. There is no significant variability of the UVI caused by this variability of the tropospheric ozone column. An increase of the aerosol optical depth (AOD) with 0.4 can cause a significant variability in the UVI of typically 10%. The AOD at Paramaribo varies between 0 and 0.5. The AOD was typically higher in the dry season than in the wet season of 2003. Diffuse radiation at cloudless circumstances is a possible indicator for the AOD. In Chapter 5 a summary is given of the conclusions from this study. Also some recommendations are given for further research and improvement of the algorithm predicting the UVI.