Ground-Based Ultraviolet-Radiation Measurements during Springtime in the Southern Hemisphere

We report the first measurements obtained by a network of UV-B detectors established through Argentina and Chile, at locations covering latitudes extending from 53° S to 18° S. Evidence that UV-B increases are detected at these latitudes during the Austral spring 1993 is presented.

Introduction. -The destruction of the stratospheric ozone layer is generally considered one of the most serious environmental problems. It has been recently published that during October 1993 the ozone hole has been the deepest ever recorded[ll. Reduction of the stratospheric ozone layer was unambiguously detected about two decades ago in the Antarctic continent [2,3]. Since then it has been systematically monitored by different means (satellite, balloon soundings and ground station observations . One of the most serious consequences of the ozone content reduction is the increase of biologically effective ultraviolet doses, particularly the so-called ultraviolet-B radiation (280-320 nm) received at the Earth surface, with multiple possible hazards for living species [71. Despite these consequences, ground stations to check the W Sun radiation are not very numerous, particularly outside the circumpolar area in the Southern Hemisphere. With the aim of studying whether the ozone depletion over the Antarctic area has extended further, and whether the possibility that W-radiation increase may begin to affect inhabited regions, a network of W -B detectors has been established through Argentina and Chile, at locations covering latitudes extending from 53" S to 18" S ( fig. 1). Here we report the first measurements obtained by this network, providing evidence that W -B increases are detected at these latitudes during the last Austral spring.
Expem'mental results and discussion. -The devices built for this purpose have been designed to be of low cost and easy to manipulate, in order to implement an extended W network. The device consists of two EGG detectors (type DF) incorporating blue silicon photodiodes (spectral response 250-900 nm) and integrated interferencial filters (centred at 313 nm and 300 nm, whith-10 nm bandwidth). Two quartz-based telescopic assemblies and an equatorial mounting (ALSTAR) allow Sun tracking for direct solar-radiation measurement. The relative response of the devices has been carefully calibrated at the Centro de Investigaciones Opticas (CIOp), La Plata, Argentine.
Due to the substantial difference (a factor of six) in the absorption cross-section at the selected wavelengths (313 and 300nm)[8], changes in the irradiance ratio at these wavelengths clearly indicate a decrease in the total ozone amount. Although the irradiance ratio between these two close wavelengths corrects several effects, such as light scattering by clouds [9], all data presented in this work have been taken in clear days, with high horizontal visibility. Figure 2a) shows the variation of the total ozone during September 1993 (days 245 to 274 of the year), recorded in Punta Arenas (lat. = 52" S), Chile, with a Brewer spectrophotometer, calibrated in Dobson units. It can be observed that a remarkable depletion in the total ozone between the days 250 to 260 was detected, followed by narrow deeps occurring on the days 270 and 276. The overall ozone reduction registered between the days 248-270 was about 50% of the predepletion values.  .2a)).
These results indicate that during the time when ozone depletion in the Antarctica takes place, an increase in UV-B radiation reaching the Earth surface affected the American continent at latitudes about parallel 50" S. Figure 2b) also shows the R values recorded in Bariloche (lat. = 41" S) (squares), and Tan& (lat. = 37" S) (triangles), Argentina. These data reveal that, although much more attenuated than in Rio Grande, the enhancement of W -B radiation which occurs around the days 255, 270 and 276 was also detected at these latitudes. A search for W -B increase at higher (i.e. equatoward) latitudes was also possible with the devices installed in San Luis (lat. = 33" S), Mercedes (lat. = 34" S) and Valparaiso Oat. = 33" S). These results are presented in fig. 3 in an expanded scale.
Although due to atmospheric conditions the period of observation is limited to a reduced number of days (255-280 days) the combined data indicate, surprisingly, that an UV-B increase has also happened at these high latitudes around the day 271. A similar effect on days 270 and 276 was also found in Antofagasta (lat. = 23" S), Chile. Figure 4 shows the dependence of the percentage decrease in the irradiance ratio at 313nm and 300nm (R) with the latitude (A). It can be observed that there is a marked attenuation of the decrease at high latitudes, although it may still be detected at latitudes as meridional as 23" S. The inset shows a quantification of this dependence, which closely follows a relationship aR = A/cos4 A. Solid lines represent the least-squares fitting of the experimental data to that dependence.
Finally, we may mention that no variation in the R values was detected in the Arica station (Chile), located at a more meridional latitude (lat. = 18" S).
These results, obtained with the first eight operative devices of the network, show that an W -B radiation increase, very likely connected with an ozone layer depletion, was observed in the Southern continent during the beginning of the 1993 spring.
It is accepted that ozone depletion in the atmosphere is caused by a series of W-induced photochemical and catalytic reactions involving halogens as well as hydrogen and nitrogen compounds (HO, and NO,). The injection of these components, due to the polar-vortex destruction, is generally associated with the recurrent ozone depletion during the Austral spring.
In relation to the W increase at the Earth surface presented in fig.2 and 3, the appearance of episodes well localized in time during the days 270 and 276 is apparent. These episodes are of short (few days) duration and appear practically coincident at latitudes as different as in Rio Grande (53"s) and Antofagasta (23"s). Therefore, they seem to be inconsistent with a diffusion process of the .ozone hole. from the Antarctica, and other possibilities must be considered.
It is generally accepted that high-intensity ( X ) solar flares are capable of depositing large amounts of energy in a short time, involving the production of high concentrations of nitrogen oxide and affecting the ozone concentration at different atmospheric altitudes [lo-121,  although the connection of this lower intensity (M> events is not as clear as with the high-energy (x> events, which are occasionally associated with a measurable solar-proton event at the Earth surface (Ground-Level Solar-Proton Event). Therefore, although it cannot be unambiguously stated, the coincidence in time suggests a possible connection between solar flares and the W-radiation increase. We want to remark also that the ozone depletion here reported on days 270 and 276 of 1993 in Punta Arenas amounts to a 50% of its predepletion value. This is higher than the values previously reported (23% [ll] and = 5% [12]) in connection with solar-flare events.
The establishment of a more extensive network, devoted to permanent W-radiation monitoring, covering even higher (equatorward) latitudes than those shown here, would be very convenient to ascertain the origin of the observed W-radiation increase. * * *