Análise de sensibilidade dos métodos de estimativa da evapotranspiração de referência e da cultura cana-de-açúcar.

Abstract

The sensitivity analysis technique assesses the contribution of each input variable of the model on other variable. This technique was applied to data obtained from field experiment carried out in the Distillery Miriri, Capim, PB. Both errors and sensitivity analysis were applied to the data of reference evapotranspiration (ETo) obtained by the Penman-Monteith (FAO/56) approach and to energy balance equation based on Bowen ratio for estimating of crop evapotranspiration (ETc). Air temperature, wind speed, relative humidity and sunshine data from September 1, 2008 to September 30, 2009 were used to obtain ETo. The energy balance equation was applied to the following days: March 19; May 9, August 30 and September 01 during the 2009 year to obtain the evaporatranspiration in sugar cane at the study area. The sensitivity and error analysis were applied to Bowen ratio method in measurement intervals of 15, 30, 60 and 120 minutes. Also, the sensitivity analysis was applied to the ETo data on daily basis over a period of one year. The most sensitive variable to the Penman-Monteith (FAO/56) equation was the net radiation, with sensitivity coefficient ranging from 0.19 to 0.97 and the lowest sensitivity was in air temperature, with values ranging between -1.4 and -0.04. The result shows that net radiation is the most sensitive variable in the calculation of energy balance, while the soil heat flux provides the lowest contribution. The relative errors for all variables in energy balance equation increase significantly with increase in sampling interval. The most sensitive variable in reference evapotranspiration by Penmam-Monteith method is net radiation, followed by relative humidity, wind speed at 2m and air temperature.