Application of a priori primary production model to estimate water vapor flux

Accurate estimation of terrestrial evapotranspiration (ET) is a basic request in researches about water cycle and the energy exchange at land-atmosphere interface. Modelling ET with water and carbon coupling theory has been proven to be a robust and effective strategy. However, there still remaining an assumption needs demonstration: if site-calibrated parameters in empirical models are universally accurate and could be generalized in different regions or future scenario. In this research, we present a prototype coupling ET and carbon assimilation based on a first-principle primary production model with only two parameters calibrated with independent datasets. The water vapor diffuses through leaf stomata, which is regulated by the ratio of leaf-internal to external CO₂ (χ) and could be estimated by environmental factors using our universal model. We validated the prototype with three steps. In the first step, we prove that diffusion process is the key linkage of water cycle and carbon assimilation at canopy scale. In the second step, comparation was carried over different vegetation types between predicted gross primary production (GPP) and tower-based observation, where results displays good agreement was found. Thirdly, we use the predicted χ and GPP to estimate canopy ET. Due to the strict description of physical and physiological process, our ET model is free of further consideration about the variation of parameters, thus could be ideally used in non-site region or future scenario. Sensitivity analysis results show that GPP would increase following the rising of CO₂ concentration, but exhibit a parabolic trend when faced with rising air temperature. On the other hand, simulated ET exhibits nearly linear trend against warmer environment, but nearly no obvious relation with rising CO₂ concentration.