Changes in surface solar UV irradiances and total ozone during the solar eclipse of August 11, 1999

During the solar eclipse of August 11, 1999, intensive measurements of UV s olar irradiance and total ozone were performed at a number of observatories located near the path of the Moon's shadow. At the Laboratory of Atmospher ic Physics (LAP) of the Aristotle University of Thessaloniki, Greece, globa l and direct spectra of UV solar irradiances (285-365 nm) were recorded wit h a double monochromator, and erythemal irradiances were measured with broa dband pyranometers. in addition, higher- frequency measurements of global a nd direct irradiances at six UV wavelengths were performed with a single Br ewer spectrophotometer. Total ozone measurements were also performed with D obson and BI ewer spechophotometers at Hradec Kralove (Czech Republic), Isp ra (Italy), Sestola (Italy), Hohenpeissenberg (Germany), Bucharest (Romania ), Arosa (Switzerland), and Thessaloniki (Greece). From the spectral UV mea surements the limb darkening effect of the solar disk was tentatively quant ified from differences of measured solar spectral irradiances at the peak o f the eclipse (near to limb conditions) and before the eclipse. Two blackbo dy curves were fit to the preeclipse and peak eclipse spectra, which have s hown a difference in effective temperatures of about 165 degreesK between t he limb and the whole of the solar disk. The limb darkening effect is large r at the shorter UV wavelengths. The ratio of the diffuse to direct solar i rradiances during the eclipse shows that the diffuse component is reduced m uch less compared to the decline of the direct solar irradiance at the shor ter wavelengths. Moreover, a 20-min oscillation of erythemal UV-B solar irr adiance was observed before and after the time of the eclipse maximum under clear skies, indicating a possible 20-min fluctuation in total ozone, pres umably caused by the eclipse-induced gravity waves. This work also shows th at routine total ozone measurements with a Brewer or a Dobson spectrophotom eter should be used with caution during a solar eclipse. This is because th e diffuse light increases by more than 30% with respect to the direct solar radiation, increasing more at the shorter wavelength side of the UV spectr um. This plausible mechanism introduces an artificial decrease in total ozo ne during solar eclipse of more than 30 Dobson units (DU), which is confirm ed by all Brewer and Dobson measurements. Changes in total ozone cited earl ier in the refereed literature have not been confirmed in the present study .