Location: Cropping Systems and Water Quality Research
2024 Annual Report
Objectives
Objective 1: Optimize production systems for irrigated cotton, corn, soybean, and rice to improve crop water productivity under variable weather and soil conditions.
1A: Develop improved methods for determining the appropriate values of field capacity for use in irrigation scheduling.
1B: Develop a database of crop canopy sensing data for calculating crop coefficient in fields with uniform soil to serve as baseline for determining site-specific crop coefficients.
Objective 2: Evaluate and/or develop site-specific best management irrigation practices based on localized soil and environmental conditions to optimize crop production while minimizing water usage.
2A: Evaluate the potential for use of the ARSPivot program for variable-rate irrigation management in the sub-humid U.S. Mid-South.
2B: Document the spatial variability of crop water coefficient and other crop and soil properties in a field and how they interact to affect crop water productivity.
Approach
Our team will address impediments to the overall goal of improving performance, profitability, and sustainability of irrigated agriculture in humid and sub-humid climates. We will develop and refine tools to improve irrigation scheduling and develop improved methods for determining appropriate values for a specific soil’s field capacity, information which is essential for optimal water management. Building on our previous research and as part of a multi-location, multi-disciplinary team, we will investigate how best to achieve site-specific irrigation management through use of the ARSPivot computer program to manage mechanized irrigation systems, and observations of soil and crop variability within the field.
Progress Report
This project includes objectives from ARS scientists and collaboration of University of Missouri (MU) scientists through a Non-Assistance Cooperative Agreement.
Objective 1B: Collected canopy spectral reflectance and height data across several cotton varieties multiple times during the growing season. This data will be combined with data from previous years to develop a database of crop canopy sensing data for calculating crop coefficient in fields with uniform soil and management to serve as baseline for determining site-specific crop coefficients.
Through collaboration with the University of Missouri: (1) Maintained three real-time weather stations at research facilities in southeast Missouri with free web access to the information as part of the Missouri Mesonet statewide network of weather stations (mesonet.missouri.edu). (2) Prepared and planted row rice seeding rate trials utilizing four cultivars at five seeding rates across the top, middle, and bottom of four row rice fields across the Bootheel. A graduate student is collecting a more complete dataset this year than in past years. (3) A series of plots separated by berms have been created and cotton established with or without cover crop to compare the effects on water runoff and its quality. (4) Collaborated on the third year of a crop rotation study including corn, cotton, soybean and peanut to address how to better design cropping systems for the climate of the Upper Mississippi Delta.
Objective 2B: Collected a third year of canopy spectral reflectance and height data multiple times during the growing season with ground-based mobile sensors and a cooperator-operated unmanned aerial vehicle (UAV) in a cotton field. The goal was to observe differences between two cotton varieties in this field with highly variable soils, to better understand within-field variability in the cotton crop coefficient.
Through collaboration with the University of Missouri: High-resolution imagery data were collected using an unmanned aerial vehicle (UAV) system to generate digital maps for stand count of cotton at different seeding rates. Large spatial variations at the same seeding rate were related to soil texture differences. The goal is to use this information for prescribing variable-rate irrigation.
Accomplishments
1. Developed new methods of utilizing remote sensing for cotton yield prediction. Cotton yield prediction is important for farmers wanting to optimize irrigation management and make marketing decisions. Information on soil texture and weather conditions, supplemented with unmanned aerial vehicle (UAV) imagery, can provide an accurate estimate of yield. Combining images from multiple years can lead to even more accurate predictions. A field study was conducted by ARS scientists at Portageville and Columbia, Missouri, in collaboration with University of Missouri scientists with the goal to quantify cotton yield variation due to soil texture and weather conditions using UAV imagery and deep learning, an advanced machine learning technique. Soil apparent electrical conductivity, which farmers can readily collect with commercial implements, provided detailed information about the spatial variation of soil texture within fields, while UAV images provided information about the spatial and temporal variation of plant properties within fields. Seasonal weather and irrigation data were also used by the prediction model. Results showed that a model trained with data from two years could accurately predict cotton yield in a third year. This approach may be useful to researchers and farmers who are interested in predicting cotton yield at a high spatial resolution for marketing or management decisions.
2. Developed faster methods to accurately document cotton plant emergence and allow action targeted to problem areas. Quickly and accurately documenting cotton crop emergence can identify problem areas of the field and allow replanting if necessary. The small size of the newly emerged plants makes them difficult to identify with unmanned aerial vehicle (UAV) remote sensing and the time required for extensive data processing to accurately detect them often prevents remediation based on the measurements. ARS researchers at Portageville and Columbia, Missouri, and University of Missouri collaborators used a UAV to capture early-season images of irrigated cotton fields and then developed and refined methods to quickly process the images and provide emergence results. The method was more than 90% accurate in identifying the number of plant seedlings in the field and in recognizing weeds and other extraneous material to avoid including them in the seedling count. The near- real-time processing with this method was much faster than traditional image processing methods that may take days of computer time. This new methodology will allow cotton producers throughout the world to better manage their crops for more efficient production systems, ensuring a stable supply of food (cottonseed oil), feed (cottonseed meal), and fiber.
Review Publications
Nguyen, A., Thompson, A.L., Sudduth, K.A., Vories, E.D. 2023. Automatic management zone delineation for center pivot variable rate irrigation using field data. Journal of the ASABE. 66(6):1527-1545. https://doi.org/10.13031/ja.15528.
Tian, F., Ransom, C.J., Zhou, J., Wilson, B., Sudduth, K.A. 2024. Assessing the impact of soil and field conditions on cotton crop emergence using UAV-based imagery. Computers and Electronics in Agriculture. 218. Article 108738. https://doi.org/10.1016/j.compag.2024.108738
Feng, A., Zhou, J., Vories, E.D., Sudduth, K.A. 2023. Prediction of cotton yield based on soil texture, weather conditions and UAV imagery using deep learning. Precision Agriculture. 25:303-326. https://doi.org/10.1007/s11119-023-10069-x.