Assessment of Agricultural Feedbacks in Noah-MP-Crop Land Surface Model on Regional Crop-Yield Simulations

Croplands significantly contribute to land-atmosphere coupling processes and can notably influence land surface heat, moisture, and momentum exchanges, alter mesoscale boundary convergence/convection, and ultimately modify local and regional weather and climate via the bio-geophysical characteristics such as albedo, soil moisture, surface roughness, canopy height, and leaf area index. Simultaneously from a reverse perspective, the environmental factors and changing climate can impact crop productivity and threaten food security at a regional and global scale. Nearly 20% of continental United States (and ~13% of the global land) are cover with croplands and further cropland expansion to produce more food is untenable due to lack of fertile land. In this situation, land management and agricultural practices will intensify to alleviate the negative impacts of climate variability on crop productivity. Crop models have been employed as the primary tool to assess the trade-off effect between climate and croplands, generate high spatiotemporal resolution regional agro-climatic related products, and evaluate adaptive strategies to increase crop productivity and mitigate the risk inherent in the food security. with the advancement of land surface models (LSM), the representation of croplands was incorporated in these models. Despite considerable signs of progress in LSM models, the majority of the models neglected growth characteristics and agricultural practices (e.g., planting dates, water absorption pattern, irrigation, fertilization) and assumed single cultivar for the whole research region. This resulted in large uncertainties in regional crop modeling simulations, particularly in simulating crop yield under projected changing climate. In response to these needs, dynamic corn (Zea mays) and soybean (Glycine max) growth simulations were introduced into Noah-MP (named as the Noah-MP-Crop model). The primary evaluations of the model showed promising enhancements in capturing spatiotemporal heterogeneity of crop leaf area index (LAI), simulated yield, surface energy fluxes, and the overall ability of the coupled version with WRF model in simulating the land-atmosphere interactions and mesoscale convection. Nevertheless, additional development is still required for several parts of the model as the previous simulations signified that the model was highly sensitive to cultivar information and agricultural practices. Additionally, preliminary results of testing the Noah-MP-Crop exhibited a high bias for latent heat flux (ET) under stress conditions. Accordingly, we postulate that by implementing more detailed information regarding agricultural practices, as well as considering another pathway during a water-stress condition by incorporating a canopy conductance-transpiration function, the model performance will be improved. The new updates will be applied to the Naoh-MP-Crop model and it will be tested over agricultural-dominated areas of U.S. Midwest, using datasets from field experiments. Implementing a more-detailed crop representation in the weather models not only foster our understanding of the role of agricultural feedbacks in the regional weather but it can also positively affect the numerical weather prediction of high impact events especially over the crop dominated landscapes. Therefore, we can expect a more accurate outcome to specify the impact of future agricultural productivity and regional climate.