Previous: Introduction Next: Validation of the method for retrieving UV irradiance variations Up: Ext. Abst.

2. Data and method description

Following the RAF ideology described in [Booth and Madronich, 1994] which is based on the existence of power law dependence between the UVB irradiance and total ozone we calculated UVB relative changes due to ozone using the following equation:

where RAF=-1.1 for CIE erythemally weighted (EW) irradiance and RAF=-2.1 for Setlow weighted irradiance [Madronich et al., 1998].

To estimate the cloud effects we use the CQ_g parameter which can be evaluated using only visual cloud observations and precalculated dependence of UV transmittance on different cloud amount [Chubarova, 1998]. According to Chubarova and Nezval’ [2000] the resulting UV transmittance by clouds (CQ_g) can be written as:

where Pi(N,Nl) is the frequency of low level cloud amount (Nl) with different amounts of total cloudiness (N); Pi(10,Nl) is the frequency when total cloud amount is equal to 10, always corresponding to overcast conditions, but with different amount of low level clouds; CQ(Nl) is the UV transmittance by low cloudiness; CQ_up is the UV transmittance by upper level clouds under overcast conditions.

In our previous study [Chubarova and Nezval’, 2000] it was shown that speaking of year-to-year variations in UVB irradiance, the changes in total ozone and in CQ_g cloud parameter play the main role if compare with variations in aerosol and in cloud optical thickness. This fact enables us to calculate the changes in UVB due to cloudiness and ozone when there is no any radiation measurements and, hence, to reconstruct the historically changes in UVB irradiance. For calculating the CQ_g parameter we use 3 and 6 hourly cloud amount observations from 19 meteorological stations located in different regions of the former Soviet Union since the middle of the 30s which were available from cdiac.esd.ornl.gov/epubs/ndp/ndp048. We also use the observations from several ground ozone stations stored in WOUDC archive (www.tor.ec.gc.ca/woudc/data/). The longest period of ground ozone observation covers 1926-1998 (Arosa, Switserland, [Staehelin et al., 1998]).

We also used TOMS ozone and UV Reflectivity at 380 nm (R380) data, which were available from http://jwocky.gsfc.nasa.gov/ for the periods of 1979-1992 and 1997-1999. R380 was used to estimate the cloud-aerosol transmittance in the atmosphere according to the method proposed in [Eck et al., 1995]:

where Rs is the surface reflectivity. We consider Rs=0.03.