Improving the Gross Primary Productivity Estimate by Simulating the Maximum Carboxylation Rate of the Crop Using Machine Learning Algorithms

The current regional-scale process-based photosynthesis models use biome-specified values of maximum carboxylation rate at 25 degrees C (V-m25) in simulating ecosystem gross primary productivity (GPP). These models ignore the variations in V-m25 over time and space, resulting in substantial errors in regional estimates of cropland GPP. Thus, to resolve this problem, we used the ensemble Kalman filter (EnKF) to assimilate tower-based GPP from five maize flux sites into a process-based mode to obtain the "apparent" value of V-m25 and then modeled this parameter using machine learning (ML) algorithms. The results showed that V-m25 increased during the early growing season and then decreased after reaching a peak value in the middle of the growing season. The coefficient of determination (R-2) root mean square error (RMSE) for satellite-driven coupled photosynthesis and evapotranspiration simulator (SCOPES)-Crop with EnKF-derived varied V-m25 in simulating daily GPP across all site-days increased (decreased) by 0.17 (5.63 mu mol m(-2) s(-1)) on average compared to that for the model with fixed V-m25. We used four ML algorithms, namely artificial neural network, random forest, extreme gradient enhancement, and convolutional neural network (CNN), to model the V-m25 of maize. The CNN algorithm yielded the best results. The average of the R-2 (RMSE) values of simulated GPP using CNN-based V-m25 over the three flux sites is 0.93 (1.95 mu mol m(-2)s(-1)), higher (smaller) than that using fixed V-m25 This study implies that representing the seasonal variations in V-m25 can facilitate improved estimates of GPP and the ML methods are useful tools for modeling the variation in V-m25.