Estimación de la radiación solar y neta a partir de la temperatura en ecosistemas de páramo y comparación con sensores a nivel del suelo

Abstract

Radiation is a meteorological parameter related to the processes of evapotranspiration, photosynthesis, and climate change. In order to understand the influence of radiation on an ecosystem, is necessary a continuing monitoring using specialized equipment, or using mathematic models to calculate it. Study probes the relation between observed and estimated radiation, and the influence of cloudiness over the solar radiation. It was developed using information of three meteorological stations, situated in Quinoas watershed. Objectives were: (i) Estimating hourly solar radiation, (ii) Estimating hourly net radiation. Both of them were calculated using equations described in FAO-56. Results were compared with the data from pyranometers and net radiometers respectively, using statistical equations: root mean square error (RMSE), mean bias error (MBE), and coefficient of determination (R2). (iii) Relating the cloudiness percentage, determined by Landsat ETM+(B1) satellite images, with the solar radiation. Results showed a poor estimation of the solar radiation using Hargreaves-Samani model (HS). RMSE% errors were higher than 64.5%, MBE% errors were higher than -31.8% and the mean R2 was 0.52. Net radiation calculated from values of solar radiation observed in pyranometers was acceptable. RMSE% were lower than 41.9%, MBE% were lower than 23.9%, and the mean R2 was 0.96. There was a moderate to high inverse relation (R2=0.73) between cloudiness percentage and the effective fraction of solar radiation, for the area that contains the pyranometers (72.26 Km2). This study has evaluated the reliability/plausibility of the HS model, and the importance of relating the solar radiation and cloudiness in a páramo ecosystem.