ESTIMATE OF REFERENCE EVAPOTRANSPIRATION THROUGH CONTINUOUS PROBABILITY MODELLING
ABSTRACT This study aimed at testing the fit of continuous probability distributions to a daily reference evapotranspiration dataset (ET0) at a 75% probability level for designing of irrigation systems. Reference evapotranspiration was estimated by the Penman-Monteith method (FAO-56-PM) for eight locations, within the state of Espírito Santo (Brazil), where there are automatic gauge stations. The assessed probability distributions were beta, gamma, generalized extreme value (GEV), generalized logistic (GLO), generalized normal (GN), Gumbel (G), normal (N), Pearson type 3 (P3), Weibull (W), two- and three-parameter lognormal (LN2 and LN3). The fitting of the probability distributions to the ET0 daily dataset was checked by the Kolmogorov-Smirnov's test. Among the studied distributions, GN was the only one to fit the ET0 data for all studied months and locations. We should also infer that continuous probability models have a good fit to the studied ET0 dataset, enabling its estimation at 75% probability through a Generalized Normal distribution (GN). Therefore, it can be used for the sizing of irrigation systems according to a given degree of risk.
Main Authors: | , , , , |
---|---|
Format: | Digital revista |
Language: | English |
Published: |
Associação Brasileira de Engenharia Agrícola
2017
|
Online Access: | http://old.scielo.br/scielo.php?script=sci_arttext&pid=S0100-69162017000200257 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | ABSTRACT This study aimed at testing the fit of continuous probability distributions to a daily reference evapotranspiration dataset (ET0) at a 75% probability level for designing of irrigation systems. Reference evapotranspiration was estimated by the Penman-Monteith method (FAO-56-PM) for eight locations, within the state of Espírito Santo (Brazil), where there are automatic gauge stations. The assessed probability distributions were beta, gamma, generalized extreme value (GEV), generalized logistic (GLO), generalized normal (GN), Gumbel (G), normal (N), Pearson type 3 (P3), Weibull (W), two- and three-parameter lognormal (LN2 and LN3). The fitting of the probability distributions to the ET0 daily dataset was checked by the Kolmogorov-Smirnov's test. Among the studied distributions, GN was the only one to fit the ET0 data for all studied months and locations. We should also infer that continuous probability models have a good fit to the studied ET0 dataset, enabling its estimation at 75% probability through a Generalized Normal distribution (GN). Therefore, it can be used for the sizing of irrigation systems according to a given degree of risk. |
---|