Calculation of photolysis rate coefficients from irradiance measurements to use in tropospheric and stratospheric models

GUSTAVO G. PALANCAR; RAFAEL P. FERNANDEZ; BEATRIZ M. TOSELLI

Salvador, Bahía

Congreso; 27a Reunião Anual da Sociedade Brasileira de Química y XXVI Congreso Latinoamericano de Química; 2004

Sociedade Brasileira de Química y Federación Latinoamericana de Asociaciones Químicas

Photodissociation reactions are the driving force of the atmospheric chemistry. These processes generate highly reactive species, which are involved in many mechanisms and are the responsible of the remotion of most trace gases in the troposphere. As a consequence of their importance for the chemical behavior of the atmosphere, the rates (J) at which the photolysis reactions occur must be included as input parameters in all photochemical models. Currently, the database of actinic flux measurements is very limited. On the contrary, a much more extensive database of global irradiance measurements exists. Although both radiometric quantities are closely related, they are not equal and the usage of irradiance instead of actinic flux to calculate J-values (e.g. O3 or NO2) leads to errors of approximately 35%. That is why a considerable effort has been devoted to find a robust method to convert irradiance to actinic flux. The instruments used in this work were manufactured by Yankee Environmental Systems, Inc. (YES). The YES UVB-1 pyranometer measures broadband global UV-B irradiance (280-315 nm) while the YES TSP-700 pyranometer measures broadband global total irradiance (300-3000 nm). Both instruments were mounted on a wide-open area in the university campus in Córdoba City, Argentina (31º 24´ S, 64º 11´ W, 454 meters above sea level). In this work we present the results of applying an approach for converting a set of broadband irradiance data to J-values. To show this approach we use the ozone (O3) and the formaldehyde (HCHO) photolysis frequencies at the surface level. The converted J-values are compared against the results of a discrete ordinate 8-stream radiative transfer model (TUV), which is used as a reference all through this work. However, considering that the 2-stream models and the broadband irradiance measurements are the most widely used tools throughout the world, an empirical relationship between them is also shown. The analysis is focused, on one hand, on the results of the daily variation of the J-values and, on the other hand, on their annual variation at midday, both at surface. The J-values for some specific reactions (N2O, HCFC, HNO3) are also calculated at different altitudes in the troposphere and stratosphere. Herein it is important to recognize that although the model permits calculations at high altitudes, some important considerations have to be taken into account regarding to the effect of the O2 absorption in the Schumann-Runge (SR) bands (175-205 nm). In these cases, a better resolution in the SR bands and in the wavelength grid of the model can lead to important differences in the photolysis frequencies without a significant increment in the CPU time. The J-values calculated at surface from irradiance measurements are used as input parameters of an atmospheric model to assess the air quality and the O3 formation in Córdoba City, Argentina. This study demonstrates that the potential exists to convert broadband global irradiance measurements into actinic flux and, finally, to J-values, using the structure of the TUV model under clear sky conditions. In this way, irradiance measurements can be used as the source of the J-values calculations at surface. This result is very important because there is a large data base of these measurements all over the world. However, to include experimental measurements in the model not only the knowledge of the direct and the diffuse fractions but also the irradiance values spectrally resolved are required. Although the uncertainties in converting from irradiance to actinic flux may be large, the data may still be of sufficient quality to investigate differences in photolysis frequencies between sites where differences in UV irradiances are large.