Behavior of the Hargreaves and Samani model in different meteorological conditions
Name: RAFAEL ESTEVES DOHLER
Publication date: 06/07/2016
Advisor:
Name |
Role |
|---|---|
| ROBERTO AVELINO CECÍLIO | Co-advisor * |
| SIDNEY SARA ZANETTI | Advisor * |
Examining board:
| Name |
Role |
|---|---|
| ALEXANDRE CÂNDIDO XAVIER | External Examiner * |
| JOSÉ EDUARDO MACEDO PEZZOPANE | Internal Examiner * |
| ROBERTO AVELINO CECÍLIO | Internal Examiner * |
| SIDNEY SARA ZANETTI | Advisor * |
Summary: The rational use of the water has become increasingly important in recent years due to poor distribution of rainfall and increased demand for water, such as in agricultural and forestry production. Evapotranspiration is an important variable of the hydrological cycle and one of the main components of the water balance in the soil. The use of simplified equations is a potential alternative to estimate the reference evapotranspiration when weather data are limited. The objective of this study was to apply and test different methods to estimate the reference evapotranspiration (ET0) for the Espírito Santo State (Brazil), from limited weather data using the method of Hargreaves and Samani, adopting the Penman-Monteith FAO-56 as a reference. Calibrated the ET0 by Hargreaves and Samani equation using linear regression, and adjusted to the coefficient of Hargreaves and Samani (CH) by the methods of Vanderlinden et al. (2004) and Martí et al. (2015). adjustments were performed by linear regression, considering all weather stations in this study (fit general), different types of weather (fit by climate), the dry and wet seasons of the year (fit dry period and fit rainy period), classes temperature range (fit by classes), and the type of climate combined with temperature range of classes (fit by climate and classes). Also it is estimated for comparison purposes, by methods ET0 Hargreaves and Samani (HS) and Penman Monteith with limited weather data (PML). In general, the mean absolute error (MAE) of the HS methods PML Vanderlinden et al. (2004) Martí et al. (2015), fit general, fit by classes, fit by climate, fit by period, fit by climate and classes, were 0.68, 1.46, 0.81, 0.77, 0.53, 0.51, 0.51, 0.51, 0.49 mm day-1, respectively. For the dry and rainy period separately, the errors (MAE) were 0.41 and 0.61 mm day-1, respectively. The fit by classes of temperature range provided better estimates of ET0 in drier days in which they need to better estimates for irrigation management in agriculture. The PML method had the worst performance among the tested methods, it is not recommended to estimate evapotranspiration in the state. The adjustments by linear regression obtained outperformed CH settings in which improved estimates of ET0 up to 30%. With limited meteorological data, the fit general method is regarded as the most recommended among tested methods, due to its simplicity of application. To estimate the ET0 between the months of April and September in the state, it is recommended the dry period fit method.
