A two‐leaf rectangular hyperbolic model for estimating GPP across vegetation types and climate conditions

Abstract There are mainly three types of gross primary production (GPP), including light use efficiency (LUE) model, rectangular hyperbolic model (RHM), and process‐based model (PBM). RHM is not widely used because its parameters, namely, quantum yield ( α ) and maximum photosynthetic rate (P m ), vary temporally with temperature and spatially with vegetation type under natural conditions. In the study, we present a temperature‐ and vegetation‐type‐adapted RHM by linking it to the Baldocchi's model to obtain the relationship between α ‐P m and V cmax,25 ‐temperature to overcome the shortcomings of traditional RHM. The modified RHM (MRHM) coupled with a two‐leaf upscaling strategy makes it possible to accurate and fast estimation of GPP at large scale. Twenty‐two CO 2 eddy‐covariance sites with different vegetation types, including evergreen needleleaf forest, deciduous broadleaf forest, grassland, and evergreen broadleaf forest, are used to evaluate the performance of MRHM for GPP estimation. The comparisons of the simulated GPP using MRHM with measured and Boreal Ecosystem Productivity Simulator‐simulated GPP demonstrate that the MRHM can simulate GPP as accurately as PBM and in the meantime with the advantage of simplicity as LUE model. These results show the promising potential of MRHM for accurately simulating GPP with relative high computational efficiency, providing an ideal alternative tool for large‐scale and long time series GPP simulations. Key Points A two‐leaf rectangular hyperbolic model (RHM) for GPP estimation was developed The two parameters of the modified RHM vary with temperature and vegetation type RHM has the potentials of simplicity as LUE model and accuracy as process model