PREDICCIÓ OF RADIACIÓ SPECTRAL SOLAR UV MITJANÇANT OF DISPERSIÓ MULTIPLE MODELS. APLICACIÓ TO PREDICCIÓ DE L'INDEX UV A CATALUNYAAuthor:
CAMPMANY PONS ELIES.
Year:
2004.
University:
BARCELONA [
www.ub.es].
Place of defense: FACULTAT DE FÍSICA.
Place of preparation: UNIVERSITAT DE BARCELONA.
Summary: SUMMARY: The prediction of the UV Index is to identify the irradíancia biologically effective impact in a later time from knowledge of the factors that inteNienen in this: ozone, the zenith angle, the elevation of the site, the dispersion of air, the clouds and albedo. The influence of these factors is evident in the formulation of radiative transfer equations that solve numerically models spectral irradiance. This thesis has worked with 2 dispersion models Simplex (SMARTS2 and UVAGOA) who believe the atmosphere as a single layer and 2 models of multiple scattering (SBDART and STAR), which takes into account the no-homogeneidad vertical atmosphere and divided into an overlay layer. There has been a comparative study between them and has concluded that the models of multiple scattering get a better deal with differences between them of 2 to 3% for angles cenitales under 600.También are compared with measures instrumental and have found differences ranging between 2 and e115% in the central hours of the day. The measures instrumentals used both to compare with the departures of these models to determine its input parameters were: 4 years of írradiancia UV spectrum (285 to 400 nm every 0.5 nm) of a espectroradiómetro Bentham 300, 3 years of the UV Index measures a piranómetro ESR and 3 piranómetros Solar Light, 4 years of daily actions of a photometer Microtops and 1 year of a photometer measures CIMEL. For the determination of total column ozone has been used measures of espectroradiómetro Bentham and the photometer Microtops and have found differences of 9%. By comparison with measurements of remote sensing (TOMS and GOME) is reached RMSE of 4% for Microtops. As for the scattering due to aerosols were used as input parameters the optical thickness calculated from photometer CIMEL, while the mass of precipitable water, linked to the growth of aerosols hygroscopic, it has been determined from CIMEL light meters and Microtops and also from radiosonde. Differences obseNadas are around 12-15% among light meters and radiosonde. All these measures have been applied to model SBDART for predicting UV index in Catalonia during the years 2000-2002, using the assumption persistence of total column ozone in this work has been found to have an error of 6% summer days of clear skies. The results have been validated with piranómetro ESR and have obseNado differences around 14%. In the year 2003, he joined the prediction of ozone Deutscher Wetterdienst to 36 hours and for days of clear skies, the results improved significantly from a gap of 20% to 16%.