Assimilating remote sensing-based VPM GPP into the WOFOST model for improving regional winter wheat yield estimation

Authors: ZHUO, WEN - Chinese Agricultural University; HUANG, JIANXI - Chinese Agricultural University; XIAO, XIANGMING - University Of Oklahoma; HUANG, HAI - Chinese Agricultural University; BAJGAIN, RAJEN - University Of Oklahoma; WU, XIAOCUI - University Of Oklahoma; GAO, XINRAN - Chinese Agricultural University; WANG, JIE - University Of Oklahoma; SU, WEI - Chinese Agricultural University; ZHANG, XIAODONG - Chinese Agricultural University; Wagle, Pradeep

Submitted to: European Journal of Agronomy
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/24/2022
Publication Date: 6/3/2022
Citation: Zhuo, W., Huang, J., Xiao, X., Huang, H., Bajgain, R., Wu, X., Gao, X., Wang, J., Su, W., Zhang, X., Wagle, P. 2022. Assimilating remote sensing-based VPM GPP into the WOFOST model for improving regional winter wheat yield estimation. European Journal of Agronomy. 139. Article 126556. https://doi.org/10.1016/j.eja.2022.126556.
DOI: https://doi.org/10.1016/j.eja.2022.126556

Interpretive Summary: Crop growth models can simulate the growth status and grain yields of crops. Gross primary production (GPP, total gain of carbon via photosynthesis) can be directly linked to crop yield. There is limited research on how to combine carbon fluxes such as GPP within crop data-model assimilation (CDMA) research. The CDMA framework is proposed to assimilate GPP estimates from the satellite-based vegetation photosynthesis model (VPM) into the WOrld FOod STudies (WOFOST) model. Results showed that WOFOST-simulated GPP agreed well with the eddy covariance (EC) tower-derived GPP. Assimilating GPP into WOFOST model improved site-scale estimates of GPP and regional-scale yield estimates of winter wheat. This study is the first to show the potential of GPP assimilation for regional-scale winter wheat yield estimation within a CDMA framework. This study improves our understanding of the importance of carbon flux processes of crops in crop yield estimation.

Technical Abstract: Crop growth models are powerful tools for predicting crop growth and yield. Gross primary production (GPP) is a major photosynthetic flux that is directly linked to crop yield. However, the potential for forecasting crop yield at the regional scale by incorporating GPP data with crop models has not been fully explored. In this study, a crop data-model assimilation (CDMA) framework is proposed that assimilates GPP estimates from the satellite-based vegetation photosynthesis model (VPM) into the WOrld FOod STudies (WOFOST) model using the ensemble Kalman filter (EnKF) algorithm to estimate winter wheat GPP and yield. Results showed that the WOFOST simulated GPP agreed with the GPP derived from eddy flux tower (R2 = 0.74 and 0.47 in 2015 and 2016, respectively). Assimilating GPPVPM into the WOFOST model improved site-scale GPP estimation (R2 = 0.85 and 0.63 in 2015 and 2016, respectively), and also improved regional-scale winter wheat yield estimates (R2 = 0.31 and 0.28; RMSE= 510 and 563 kg/ha in 2015 and 2016, respectively). Moreover, the accuracy of the simulated county-level winter wheat production using EnKF (R2 = 0.81 and 0.77; RMSE = 25 and 42 kton in 2015 and 2016, respectively) was higher than open loop simulated production (R2 = 0.71 and 0.69; RMSE = 29 and 47 kton in 2015 and 2016, respectively). Our study demonstrates the potential of predicting GPP for winter wheat yield estimation at the regional scale using the CDMA framework and improves our understanding of the importance of carbon flux process of crops in crop yield estimation.