Evaluación de imágenes de precipitación de un radar meteorológico a diferentes resoluciones espacios-temporales en la cuenca alta del Río Paute

Abstract

The demand for the use of water resources due to the growth of the different production areas, such as: agriculture, livestock, electric generation, industrial, etc. Have increased in recent decades. Therefore, it is essential to understand the characteristics that govern the hydrological cycle. Water resources have been studied with greater interest in the paramos of our country. Ecuador possesses unique characteristics in terms of its hydro-meteorological processes due to the effects of the intertropical convergence zone, orography and air masses from both the Pacific and the Amazon. As a support for the study of precipitation behavior, a meteorological radar called CAXX was implemented at the northern boundary of the Cajas National Park, the radar initially recorded reflectivity values (Z) in [dbZ], which are converted at the rate of rain in [mm/h] from a power law Z=A R^b where A and b are empirical coefficients. The present study aimed to evaluate the precipitation estimates of the CAXX radar images from four functions Z=A R^b: a) used by default in the equipment and b) three additional functions adapted with local data. By comparison with the data observed in pluviometric sensors at different temporal space resolutions. The method of comparison between radar images and ground-based rain gauge data was based on statistical errors: Abs-Bias, root mean square error and coefficient of determination. This study evaluated the accuracy of radar estimates at the temporal and spatial level and also generated a baseline on the evaluation of the images in their primary state, without additional calibrations