Premium
Assessment of evapotranspiration models applied to a watershed of Canadian Prairies with mixed land‐uses
Author(s) -
Biftu Getu Fana,
Yew Gan Thian
Publication year - 2000
Publication title -
hydrological processes
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.222
H-Index - 161
eISSN - 1099-1085
pISSN - 0885-6087
DOI - 10.1002/(sici)1099-1085(200005)14:7<1305::aid-hyp44>3.0.co;2-e
Subject(s) - evapotranspiration , environmental science , potential evaporation , hydrology (agriculture) , water balance , streamflow , watershed , atmosphere (unit) , drainage basin , atmospheric sciences , meteorology , geology , geography , ecology , geotechnical engineering , cartography , machine learning , computer science , biology
Using meteorological data collected in the summer of 1996 and 1997, three evapotranspiration (ET) models, the Penman–Monteith (PM), the modified Penman for non‐saturated surface, and the two‐source models were applied to estimate hourly ET from different land‐use covers of the Paddle River Basin (area = 265 km 2 ). By assuming closed canopy conditions, the PM model could estimate the ET of coniferous forest and agricultural land because the contribution of soil evaporation under these land covers is not significant. For mixed forest and pasturelands that are partially vegetated, however, the amount of soil evaporation is significant and so PM underestimated the total ET. The modified Penman model mainly underestimated hourly ET during daytime when the atmosphere is unstable, but overestimated ET during early morning and late afternoon (stable atmosphere), particularly for pastureland. By re‐establishing a relationship for the relative evaporation, the ET simulated by the modified Penman model improved. Both PM and the modified Penman are assessed with respect to the comprehensive, two‐source model's simulated ET because the latter agrees favourably with the ET obtained from the basin water balance. The water balance ET is reliable because it is based on the streamflow, surface temperature, net radiation and soil moisture simulated by the semi‐distributed hydrological model, Semi‐distributed Physically based Hydrologic Model–Remote Sensing, DPHM‐RS (host to the two‐source model), all of which have been demonstrated to agree well with the observed data. Copyright © 2000 John Wiley & Sons, Ltd.