Location: Coastal Plain Soil, Water and Plant Conservation Research
2025 Annual Report
Objectives
1. Develop effective irrigation and water management techniques to improve water and nutrient use efficiency and increase water reuse for conservation.
1a. Improve site-specific/variable-rate irrigation management using decision support systems to improve water and nutrient use efficiency.
1b. Enhance multiscale prediction of water pathways under climate variability using Machine Learning (ML) with hydrological models.
1c. Evaluate the impact of advanced treatment technologies for livestock wastewater reuse.
2. Develop innovative cropping systems and rotations to improve water and nutrient use efficiency, profitability, climatic resiliency, and reduce environmental impacts.
2a. Quantify the impact of tillage and crop rotation interactions on optimizing water availability and crop productivity in rainfed agriculture with or without cover crops.
2b. Identify and develop novel cover and row crop systems that provide double cropping benefits, while improving soil and water conservation in the Southeastern United States.
2c. Evaluate available novel row and cover crop genetic resources for productivity and water-use in drought-prone soils.
2d. Evaluate how water availability and microbial population dynamics are influenced by soil management practices.
Approach
Water availability is essential to maintain and increase agricultural production to meet the new century’s growing food and fiber demands. Increasing demand for water for recreational, industrial, and ecosystem services is competing with agriculture for available water resources. Therefore, agriculture must be more efficient with its available water resources. The overall goal of this project is to improve water and nutrient management in humid regions. The research focuses on two main objectives. The first objective is to develop effective irrigation and water management techniques to improve water and nutrient use efficiency and increase water reuse. In this objective, we will evaluate and refine a decision support system for variable-rate irrigation management to improve water and nutrient use efficiency. Using hydrologic models and machine learning, we will improve the prediction of multiscale water and nutrient pathways under climatic variability. We will investigate the feasibility of reusing livestock wastewater for supplemental irrigation from improved treatment technologies. The second objective is to develop innovative cropping systems and rotations to improve water and nutrient use efficiency, profitability, climatic resiliency, and reduce environmental impacts. Much of the Southeastern Coastal Plain’s agriculture is in rainfed production. To address this, we will investigate and quantify the impact of tillage and novel crop rotations to optimize water availability and crop productivity and improve overall soil and water conservation. We will also investigate novel cover crops and their genetic resources to provide potential double-cropping benefits and improve soil and water conservation in the region’s drought-prone soils. Overall, this research will identify water and nutrient management practices that conserve water, sustain production, and enhance environmental quality. Conservation and protection of the nation’s water resources will ensure food and fiber production for current and future populations in an economically viable and environmentally sustainable manner.
Progress Report
In subobjective 1b, a validated SWAT model was employed to propose a spatial analysis method for retrieving SWAT reservoir parameters for water quality and quantity assessment for small watersheds. The reported method is a cost-effective alternative to traditional in situ hydrographic surveys and it is useful for addressing watersheds with small reservoirs. The procedure can be automated for assessing reservoirs through computer models or developing reservoir elevation–capacity–area curves.
In subobjective 1c, a newly developed learning-based filtering algorithm was applied to high-resolution aerial RGB and NIR ortho-imageries to retrieve the geolocations and areas of swine CAFOs lagoons across two agricultural watersheds in North Carolina. The new swine lagoons dataset is being integrated into SWATplus model to quantify the water reuse potential in animal-crop production systems. Different scenarios of waste treatment including conventional and advanced systems will be investigated for their impact on agricultural water balance and surface water quality at the watershed scale.
In subobjective 2a, the third year of the row crop-tillage-cover crop experiment was completed. Cotton and soybean yields along with biomass accumulation were collected and analyzed. Soil moisture content down to 24-inch depths were collected from each treatment. Water use efficiency was then calculated based on crop yields and observed rainfall over the growing season. Results from the second year of the trial were similar to preceding years, with crop yields and biomass accumulation not differing among the different tillage-cover crop treatment combinations. A manuscript is currently being prepared on soil health components measured in the trial. For the fourth year of the experiment, the 4-species cover crop mixture consisting of annual ryegrass, crimson clover, hairy vetch, and forage rapeseed was planted in fall 2024, followed by population counts and biomass measurements in spring 2025. Cotton and soybean were established in May 2025. Lysimeters were installed at 30- and 90-cm depths in each plot during the cover crop season and again during the row crop season. Soil leachates were collected from lysimeters following major rain events, and are currently awaiting laboratory analysis.
In subobjective 2b, the second year of the perennial groundcover crop field-scale trial was completed. This field-scale trial is approximately 5x larger than the pilot-scale study that was conducted during the first and second years of the project. In addition to perennial red and white clovers, the perennial grass, tall fescue, was also included as a new species. Soil moisture sensors collected data from all treatments during the growing season, and it was found that the presence of perennial cover crops helped retain soil moisture in the upper 6-inches of the soil profile, but caused soils to be drier at the intermediate 7 – 12-inch depth. Unlike in past years, extreme drought conditions were prevalent during most of the growing season, and drought symptoms were observed in cotton for all treatments. Based on data from the completed pilot-scale trial, a peer-reviewed journal article was published in Agronomy Journal on soil water availability, perennial groundcover crop water use, and cotton drought stress data. Data from the field-scale trial was also presented at the American Society of Agronomy/Crop Science Society of America Annual Meetings, the American Forage and Grassland Council annual meeting, and at the Beltwide Cotton Conferences.
In subobjective 2c, the fourth year of the drought tolerant crop trial was completed. The second rotation of drought tolerant vs. non-drought tolerant cotton and soybean were harvested and compared. Results were similar to those from the first cycle of this rotation in FY 23, with drought tolerant soybeans having greater yield while producing fewer beans per pod, but more pods per plant, and drought tolerant cotton using less soil water. A peer review journal article is currently being drafted.
In subobjective 2d, plots were re-established and microbial inoculant was applied. Non-drought portions of the experimental field were irrigated, soil samples were collected and processed, and crop yields were determined. Preliminary analysis indicates that, over the course of the study, microbial inoculants were not successful in improving yield outcomes in drought-like conditions, potentially due to lack of microbial establishment in the soil. A peer review journal article is currently being drafted to report these results.
Accomplishments
1. Using perennial cover crops to improve soil water availability for cotton production. Cotton production in the southeastern United States is exposed to intermittent drought stress that can reduce lint yields. New cover crop management strategies may offer solutions by reducing evaporative water losses in the topsoil, altering where plants use water first, and improving cotton water use efficiency. ARS scientists in Florence, South Carolina examined soil subsurface moisture content at 6, 12, and 18-inch depths when growing cotton with perennial red and white clover groundcover crops compared to conventional annual cover crop and fallow systems. Results indicated that cotton rooting depths were shallower under the perennial clover system due to greater water availability in the upper layers of soil. Additionally, water use efficiency was similar in cotton grown alongside perennial clover and following terminated annual cover crops or fallow. In addition to maximizing limited water resources use, incorporating perennial groundcover clovers allows cotton producers to gain the benefits of weed suppression and soil fertility with reduced input costs.
2. Machine learning tools for early-season crop yield prediction using drought. Rainfed agriculture is dominant across the hot, humid southeastern United States (US). However, over-reliance on precipitation exposes farmers to drought-induced yield instability. Under the rainfed southeastern US agricultural system, prediction tools are needed to make reliable projections of crop production. Scientists at ARS Florence, South Carolina and Clemson University developed supervised machine learning tools using K-nearest neighbors and random forests models for predicting corn, cotton, peanut, and soybean yield based on early season drought conditions. Although the models performances varied depending on the crop, model, and spatial scale, the results showed reasonable yield predictions. In addition, the machine learning tools consistently predicted corn, cotton, peanut, and soybean yield. Hence, these models can be used for making early-season crop yield projections or anticipating yield anomalies across the southeastern US.
Review Publications
Sohoulande Djebou, D.C. 2025. Spatial analysis to retrieve SWAT model reservoir parameters for water quality and quantity assessment. Water. 17(6):834. https://doi.org/10.3390/w17060834.
Billman, E.D., Stone, K.C., Paye, W.S. 2025. Perennial goundcover crop effects on soil moisture retention and water use efficiency in cotton cropping systems. Agronomy Journal. 117(3). https://doi.org/10.1002/agj2.70094.
Khedun, P., Sohoulande Djebou, D.C. 2025. Probabilistic estimates of drought-induced yield loss in the Southeastern United States. Agricultural Systems. 229. https://doi.org/10.1016/j.agsy.2025.104418.
Sohoulande Djebou, D.C., Ma, L., Qi, Z., Szogi, A., Stone, K.C., Harmel, R.D., Martin, J.H., Birru, G.A., Sima, M. 2024. Agronomic and environmental effects of forage-cutting schedule and nitrogen fertilization for bermudagrass (Cynodon Dactylon, L.). Agriculture, Ecosystems and Environment. 378. Article 109318. https://doi.org/10.1016/j.agee.2024.109318.