Evaluation of Irrigation Scheduling Strategies for Corn Production Using Crop Growth Modeling and Soil Moisture Dynamics in Southeast Alabama

Abstract

In the Southeastern United States, higher temperatures and more frequent flash droughts are increasing yield risk in irrigated corn, especially on sandy soils with low water-holding capacity. These conditions make it important to design irrigation schedules that protect yield while improving water use efficiency. This thesis evaluates irrigation scheduling strategies for corn in Southeast Alabama by combining crop growth modeling with soil moisture dynamics measurements. Chapter 2 focuses on growth-stage-based deficit irrigation (GSBD) using the CERES-Maize model in DSSAT for two contrasting soil types in Samson, Geneva County, Alabama. A field experiment with the corn hybrid Dynagro® 57VC65 under center-pivot irrigation provided data on phenology, leaf area index, biomass, soil volumetric water content, and grain yield for model calibration in 2022 and validation in 2019. Seven GSBD irrigation treatments were defined by adjusting soil water depletion thresholds for three main growth stages and simulated over 33 years of historical weather. Years were classified as wet or dry based on an abundant and well-distributed rainfall index. In the sand clay loam soil, a GSBD strategy with a 70% soil water depletion threshold during the vegetative–reproductive stage increased grain yield by up to 4.4% and water use efficiency by up to 4.7% in dry years, and increased grain yield and water use efficiency by 3.5% and 3.2%, respectively, in wet years. In contrast, the same strategy was less favorable in the sandy soil, indicating that effective deficit irrigation scheduling must account for crop growth stage, soil water-holding capacity, and seasonal rainfall conditions. Chapter 3 develops an algorithm to identify automatically field capacity (FC) and evaluate weather daily crop water use (CWU) can be estimated from in-situ soil moisture dynamics across different growth stages. Soil volumetric water content (SVWC) data from TDR sensors at 15, 30, and 60 cm were used to detect field capacity, smooth drying curves using a Savitzky–Golay filter and calculate CWU for selected drying periods in the 2022 (calibration) and 2025 (validation) seasons. The algorithm detected field capacity with small differences from soil core measurements and estimated daily crop water use in realistic ranges between about 3 and 7 mm day⁻¹ across drying cycles. HYDRUS-1D simulations showed good agreement with measured soil moisture, especially at 60 cm depth, and suggested that post–field-capacity decreases in soil moisture were mainly driven by crop water use, with limited deep drainage. Overall, this thesis shows that integrating crop simulation modeling with soil moisture–based CWU estimation can support more precise and site-specific irrigation scheduling for corn in humid subtropical environments.