Effects of Root Distribution and Root Water Compensation on Simulated Water Use in Maize Influenced by Shallow Groundwater

CORE IDEAS: Groundwater, root distribution, and compensation impact transpiration and NPP. Maize is less reliant on the compensated root water uptake mechanism as roots extend deeper. The impact of drought on corn lessens as roots become deeper. Maize increase its carbon fixation during dry years due to the compensated root water uptake. Shallow‐rooted corn benefits more from groundwater subsidy than deeper rooted corn. We investigated the potential impacts of shallow groundwater, root length density (RLD) distribution, and root water compensation on transpiration and net primary productivity (NPP). An agroecosystem model (AgroIBIS‐VSF) that is capable of simulating variably saturated water flow was driven with hourly weather observations in southern Wisconsin over 27 yr for various RLD distributions across a continuum of groundwater depth. The results indicated that the strength of the relationship between groundwater depth and water use in the critical water table depth zone is controlled by the root structure and root water uptake (RWU) strategy. In this zone, transpiration is progressively more sensitive to the groundwater level as roots become shallower. The impact of drought on corn (Zea mays L.) lessens and corn becomes less reliant on compensated RWU capabilities as roots extend deeper. Simulations indicated that the use of the compensated RWU approach results in NPP increases of 38.1 (3.81%), 30.8 (2.74%), and 6.4 (0.55%) g C m⁻² yr⁻¹ during the driest years (i.e., when growing season precipitation is below the 30th percentile of the long‐term observations) for shallow, intermediate, and deep RLDs, respectively. Moreover, shallow groundwater supported RWU, and corn with a shallow RLD benefited the most from shallow groundwater, with an increase in annual transpiration of 230 mm. Our findings underscore the importance of incorporating compensatory RWU and selecting an appropriate and representative RLD for contrasting vegetation types in ecosystem models to simulate a more realistic plant response to variable climate and groundwater depth conditions.