Regional cropland decline and associated climate feedbacks in the Mississippi delta

Abstract

The Yazoo–Mississippi Delta has undergone substantial land use and land cover transitions over the past century, with cropland expansion peaking in the mid-twentieth century followed by sustained decline and partial reforestation. These landscape changes modify surface properties and may influence regional hydroclimate through altered land–atmosphere coupling. Here, we integrate satellite observations (MODIS), historical reconstructions (LUH2), reanalysis products (ERA5), and high-resolution PRISM temperature data to quantify cropland trends and associated changes in vegetation structure, surface energy fluxes, boundary layer characteristics, and convective potential from 2001–2022, with historical context extending to 1850. To isolate local land-surface forcing from large-scale climate variability, we conduct idealized experiments using the Weather Research and Forecasting (WRF) model in which cropland surface characteristics are progressively replaced with deciduous broadleaf forest values under identical atmospheric forcing. Observations indicate an approximately 5% decline in cropland since 2001, accompanied by increases in woody vegetation. Reanalysis data suggest modest regional increases in latent heat and decrease in sensible heat, although spatial patterns are heterogeneous and influenced by broader southeastern U.S. climate trends. WRF simulations reveal a coherent and generally statistically significant responses in latent heat flux, sensible heat flux, Planetary boundary layer height, and near-surface air temperature during the growing season. Atmospheric stability metrics indicate a more favorable local thermodynamic environment for convection, whereas precipitation responses remain spatially variable and are not confined to the Delta. These results show that cropland decline and reforestation can alter surface energy partitioning and boundary layer structure in the Mississippi Delta, while precipitation remains strongly modulated by synoptic-scale forcing. Findings underscore the importance of representing agricultural land transitions in regional climate assessments of highly managed landscapes.