Efecto de la incertidumbre de las observaciones hidrometeorológicas en estudios hidrológicos (Zhurucay, Soldados y Mazar)

Abstract

In recent decades a number of hydrological models of different complexity have been developed to ensure the accuracy of hydrological studies. Yet very few hydrologists question the data quality used in the studies. Therefore, this thesis is aimed at determining the effect of hydrometeorological observations on hydrological studies. In particular, the study of how does the atmospheric pressure data recorded by automatic sensors, affects the water level measurement in small streams. It also examines the effect of sensors temperature records (T) and relative humidity records (RH) on estimating reference evapotranspiration (ETo) calculated with the Thornthwaite, Hargreaves and FAO Penman Monteith (PM) methods. The study was done with the sensors of hydrological and meteorological stations installed in the ecohydrological observatories of Zhurucay and Soldados (páramos ecosystems – 3200 to 4200 m a.s.l.), and Mazar (montane forest – 2600 to 3500 m a.s.l.), located in the southern Ecuadorian Andes. The results reveal that the atmospheric pressure estimation using linear regressions is efficient compared to the observed values, with Nash - Sutclife coefficients greater than 0.71, and that leads to calculated flow rate errors below 5%. Moreover, the discharge coefficients of weirs determined by gauging differ from theoretical values by up to 15%, which significantly affects the calculation of flow rates. Regarding the ETo estimated with the Penman Monteith method, the results reveal that the RH has a greater effect (65 %) in the calculation of the ETo than T (35 %). For a daily temperature error of -1.07 ° C, the effect is greater for the Hargreaves method (0.24 mm/day) than for PM (0.148 mm/day). With the same error of T (-1.07° C) in the calculation of ETo, the effect is smaller for the Thornthwaite method (0.832 mm/month) than for the Hargreaves (6.12 mm/month) and Penman Monteith (6.27 mm/month) methods. In general conclusion, it is important to consider the quality of the data with which we work, to obtain correct results and make informed decisions about the water resources management. It is suggested to ensure that the records are within precision established ranges before installing the sensors.