The Impact Of Soil Moisture And Saturated Vapor Pressure Deficit On Corn Yield In China And Its Predictio

As the most important food crop in China and even globally,understanding the effects of climate change-induced moisture stress and heat stress on maize yield is of great importance to ensure food supply.In this study,we investigated the effects of saturated water vapor pressure deficit(VPD)and soil moisture(SM)in the root zone on maize yield in China using current high-precision meteorological data and yield data and made predictions.Firstly,based on the maize yield data in the GDHY dataset and the meteorological data in the ERA5 reanalysis dataset,we analyzed the spatial and temporal variation of maize yield and meteorological factors such as soil moisture,and used a linear mixed-effects model to build a yield prediction model to select the most sensitive meteorological factors for maize yield and provide a research basis for subsequent studies.Then,the specific effects of soil moisture and VPD in the root zone on maize yield in historical periods in China were refined and analyzed,while a quadratic polynomial was constructed to analyze the future changes in maize yield in the context of climate change.Finally,multiple crop models based on the first phase of GGCMI(Global Gridded Crop Model Intercomparison)were compared with yield data from GDHY to analyze the model selection.Subsequently,based on the yield data from this model in Ag MIP(Agricultural Model Intercomparison and Improvement Project),the response of its maize yield to climate change was studied and analyzed in comparison with the predicted results based on soil moisture and VPD maize yield.The main findings are as follows:(1)It showed a solid upward trend in maize production during 1982-2016 in China,with each major producing region maintaining a homogeneous movement.The trend in maize production and growth was more pronounced in the northeast than in the other two production areas.The spatial distribution of regions with higher yield growth is largely consistent with those with higher maize yields.During the 2010-2016 maize growing season,there was relatively little variation in moisture conditions,which represent the need for crop growth and development,relative to the heat factor.The regions where VPD showed an upward trend were consistent with regions where temperatures increased significantly and were consistent with regions where soil moisture decreased.After comparison of yield prediction results based on a linear mixed effects model,the hydrothermal combination of VPD and soil moisture in the root zone had the highest correlation with maize yield.(2)The historical yield prediction model was validated to be highly reliable,and there was a trend of negative-positive-negative contribution of both VPD and root zone soil moisture to yield,but the negative impact on yield was relatively greater at higher soil moisture content.Yield loss can occur when there is an imbalance between atmospheric moisture supply and demand.The yield prediction results indicated that the yield loss would increase with increasing VPD,and using only a polynomial yield prediction model based on VPD for yield estimation would increase the yield loss estimation.Also,the optimal water balance conditions for maize in future periods tend to shrink relative to 2010-2016 as the emission scenario gradually increases.The response of maize yields will also become increasingly negative with increasing temperature increases.Also,yield loss increases with increasing emission scenarios at the same temperature increase.(3)The different crop models of GGCMI Phase I had slightly higher historical maize yields in irrigated conditions than in non-irrigated conditions.Meanwhile,the EPIC series models were more correlated with the reference data,while the TAMU models were relatively better matched among them.The spatial correlation between the EPIC-TAMU model and the reference data was higher for the non-irrigated condition when the model and meteorological data were the same.maize yields for the Ag MIP EPIC models were still greater for the irrigated condition than for the non-irrigated condition.maize yields for the EPIC crop model in future periods decreased gradually with increasing emission paths,with a greater trend of yield reduction for the non-irrigated condition.Maize yield loss increases with increasing temperature for all emission pathways and irrigated conditions;yield loss increases with increasing emission scenarios for the same temperature increase.Comparative analysis based on the predicted results of VPD-SM and Ag MIP models indicated that the former estimated yield response was higher than the response results of the latter for irrigated conditions and lower than the yield response for non-irrigated conditions,but their simulated results both showed a decreasing trend over time.