On the Estimation of Global Plant Water Requirement

Abstract. Water supply is the most critical constraint for vegetation growth and food security. The amount of water demand by plant growth is usually estimated by plant water requirement which unfortunately cannot be directly measured at any large scale in field conditions. Different estimation methods have been proposed in the past seven decades for estimating plant water requirements using the concept of reference evapotranspiration (ET0) methods or potential evapotranspiration (PET) methods. In addition, using PET or ET0 to estimate actual evapotranspiration (ETa) is a critical approach in hydrological and climate models. However, different PET or ET0 models provide diverse results for irrigation water requirement (IWR) that in turn may result in a huge waste of irrigation water. Here, we assess the suitability of six common methods for estimating PET at 170 eddy covariance flux sites and propose a practical approach for estimating the IWR using a physically consistent model STEMMUS-SCOPE.

Notably, the Priestley-Taylor and LSA_SAF method excels in providing reasonable approximations of daily PET. Consequently, in scenarios where net radiation data and ground heat flux are accessible, the Priestley-Taypor method emerges as the recommended choice. The LSA_SAF method is the better one when only net radiation data is available. Alternatively, in cases where only global radiation data is available, the Makkink and Hargreaves methods serve as viable substitutes. Although the FAO56 Penman-Monteith method is much better than the original Penman-Monteith method when wind speed and air humidity data are at hand, its suitability falls short of the preferred status. This study contributes to understanding and quantifying the applicability of different methods in estimating PET and IWR, based on input data availability and physical considerations.