Evapotranspiration estimation of Platycladus orientalis in Northern China based on various models

Platycladus orientalis is one of the most popular afforestation species and greening species for water management in arid and semi-arid regions of northern China. We applied various models to estimate and validate artificial P. orientalis forest evapotranspiration features with the goal of accurately estimating the water use of a P.orientalis plantation. The American Society of Civil Engineers Evapotranspiration–Penman–Monteith model(APM) and FAO56–Penman–Monteith model(FPM) are extensively applied for vegetation evapotranspiration estimation because their reliability has been validated by many scholars. The Priestley–Taylor model(PT) and Hargreaves model(HS) require only the daily maximum temperature,daily minimum temperature and solar radiation to estimate evapotranspiration and are thus widely applied to grasslands but not to forests. We used the Energy Balance Bowen Ratio(EBBR) system to validate the accuracy of the four models. The results indicated that:(1) Compared to the EBBR measurement annual value, APM was the most accurate, followed by FPM, and PT;(2) During the year, the accuracies of the four models varied. APM and FPM underestimated evapotranspiration during June, July and August, whereas PT and HS overestimated evapotranspiration during this period. In the rest of the year, the estimation accuracies were reversed;(3) An analysis of the possible reasons indicated that wind speed, air temperature and precipitation were the most important contributors.High temperatures were measured in June, July and August, which led to an overestimation by PT and HS because these two models only calculated the temperature and radiation without vegetation information. Underestimation also occurred when a low temperature was recorded. Though APM and FPM addressed both meteorological and vegetation factors, slight deviations still existed; and(4) The two models were modified based on EBBR-measured data. Relative humidity was introduced into PT, and parameter ‘‘A’’ in the HS estimation model was amend