| dc.description.abstract | The radiation balance (Rn) represents the main energy source used by physical, chemical, biological and meteorological processes at the surface, and its estimation is essential to analyze soil-vegetation-atmosphere interactions. Due to the scarcity of reliable radiation balance data and the accurate estimation of shortwave radiation data, it is extremely important to obtain appropriate relationships between these variables. Thus, the main objective of this study is to determine the simple linear regression coefficients between downwelling shortwave radiation and the radiation balance, to estimate the radiation balance on the northern part of the Brazilian semi-arid region, considering seasonal variability and influence of cloud conditions. For that, it was used data of solar radiation, albedo and radiation balance, registered every 30 minutes, during 2014 and 2015, obtained from a net radiometer, from the Seridó Ecological Station (ESEC-Seridó). Simple linear regression coefficients were generated, and these measurements were analyzed seasonally and considering the influence of the clouds through transmittance. The calibration performance analysis was performed from two independent data sets of 30 minutes collected during an one year period, obtained in two different station areas belonging to the National Institute of the Semiarid (INSA) in Campina Grande-PB, and made using the Coefficient of Determination (R²), Mean Bias Error (MBE), Root Mean Square Error (RMSE), Mean Percentage Error (MPE) and Pearson Correlation Coefficient (r). Seasonal and global calibrations presented satisfactory and similar results among them and, when applied to an area of preserved caatinga, were better adjusted. The influence of albedo provided a clear improvement in model estimation, besides a little significant improvement for models of atmospheric transmittance. From these results, it was concluded that the calibrations provided reliable levels of accuracy, allowing the application of these to remote sensing data for Rn estimation at a regional scale, being indicated for areas of preserved caatinga and degraded caatinga, and considering the inclusion of the albedo to obtain more efficient results. In addition, these calibrations can serve as an alternative method in flow studies in caatinga environments. | |
