Modelling evapotranspiration using the modified Penman-Monteith equation and MODIS data over the Albany Thicket in South Africa

Abstract

Evapotranspiration (ET) is one of the least understood components of the water cycle, particularly in data scarce areas. In a context of climate change, evaluating water vapour fluxes of a particular area is crucial to help understand dynamics in water balance. In data scarce areas, ET modelling becomes vital. The study modelled ET using the Penman-Monteith-Leuning (PML) equation forced by Moderate Resolution Imaging Spectroradiometer (MODIS) leaf area index (LAI) and MODIS albedo with ancillary meteorological data from an automatic weather station. The study area is located on the Albany Thicket (AT) biome of South Africa and the dominant plant species is Portulacaria afra. The biggest challenge to the implementation of the PML is the parameterisation of surface and stomatal conductance. We tested the use of volumetric soil water content ( ), precipitation and equilibrium evaporation ratio ( ℎ ) and soil drying after precipitation ( ) approaches to account for the fraction ( ) of evaporation from the soil. ET from the model was validated using an eddy covariance system (EC). Post processing of eddy covariance data was implement using EddyPro software. The method performed better with a root mean square observations standard deviation ratio (RSR) of 0.97. The results suggest that modelling ET over the AT vegetation is delicate owing to strong vegetation phenological control of the ET process. The convergent evolution of the vegetation has resulted in high plant available water than the model can detect. It is vital to quantify plant available water in order to improve ET modelling in thicket vegetation.