Location: Grassland Soil and Water Research Laboratory
2023 Annual Report
Objectives
Objective 1: In support of LTAR and CEAP, develop management strategies to maintain agronomic resilience through climate extremes that support natural resource conservation and agroecosystem sustainability in the Texas Gulf Coast region.
Sub-objective 1.A: Identify, develop, and evaluate strategies to improve precision agronomic management of croplands to incorporate precision conservation that will optimize agronomic, environmental and economic outcomes.
Sub-objective 1.B: Evaluate the role of landscape scale spatiotemporal soil, water and genetic variability within representative plant species to determine production potential through extreme climatic events in the Texas Gulf Coast region.
Sub-objective 1.C: Catalog, archive and synthesize observational research data across the LTAR network to facilitate collaborate research using advanced database and visualization technologies.
Objective 2. Enhance process-based model algorithms and structure using modern programming paradigms and new research findings from LTAR and CEAP.
Sub-objective 2.A: Improve the SWAT+ model using streamlined code, data structures and upgraded algorithms to better address U.S. and global environmental challenges.
Sub-objective 2.B: Improve ALMANAC predictive capacity using enhanced phenology algorithms and updated plant parameters derived from LTAR-Phenocam observational data on fraction leaf cover.
Objective 3: Develop tools using enhanced models and other new technologies to support improved agroecosystems management and policy formulation from the field to national scales.
Subobjective 3.A: Develop a trans-scale unified national modeling framework to support CEAP and LTAR.
Subobjective 3.B: Develop decision support tools to address unmet agricultural and environmental problems by synthesizing observational data and model predictions using emerging technologies (ML, AI, Drones).
Approach
The overall goal of this research is to improve agricultural and environmental sustainability by providing producers and policymakers with scientifically credible information to make good decisions. There is a strong need for improved data driven decision support tools which can predict the effects of both human activity and climate variability on agricultural production systems and the environment. These tools are needed and requested by producers, conservation/watershed planners, USDA leadership, State Agencies, Federal Agencies (NRCS, FPAC, EPA, NOAA, USGS), Non-Governmental Organizations (The Nature Conservancy, Environmental Defense Fund, Field to Market), and local stakeholder groups. The Grassland, Soil and Water Research Laboratory (GSWRL) is well positioned to address this need using complementary programs in field/ monitoring and hydrologic/water quality modeling.
There are three interlinked principal components of this project: 1) collection and synthesis of field data to aid in the evaluation of environmental and agro-economic impacts to support the development of more sustainable production strategies; 2) enhancement and testing of the Soil and Water Assessment Tool (SWAT) and Agricultural Land Management Alternative with Numerical Assessment Criteria (ALMANAC) model algorithms that represent field-, farm-, and watershed-scale processes using up-to-date scientific knowledge from Conservation Effects Assessment Program (CEAP), Long-Term Agro-ecosystem Research (LTAR), and other applied research programs; and 3) development of decision support tools using emerging technologies for conservation management, planning, and policy development at local, regional, and national scales.
Model enhancement is a foundational component of this project, models developed at GSWRL are critical components of local/regional, USDA, and legislative decision-making. These models are being used to assess USDA conservation policy in the second generation of the Office of Management and Budget (OMB) and congressionally mandated CEAP program. These models are widely used in Europe, Asia, Africa, and South America, and their enhancement has significant global impact.
Progress Report
Sub-objective 1.A. Precision ag technology data were used to develop information regarding sub-field profitability at the Riesel watersheds. Management was altered at four fields, two reduced inputs to 80% or 60% in unprofitable zones based on within-field yield potential and two fields eliminated production from unprofitable zones where yield potential has been consistently low. Fertility tests at the individual plant scale have continued. Results are variable but promising in terms of delivering fertilizer to plants in a more efficient manner to minimize environmental losses. An unmanned ground vehicle has been purchased to begin automation of fertilizer delivery.
Sub-objective 1.B. We have formulated plant parameters for an array of crops and grasses. Our primary focus is to accurately simulate various plant species and their respective production potentials under high and low rainfall. We have utilized the ALMANAC (Agricultural Land Management Alternative with Numerical Assessment Criteria) and SWAT (Soil and Water Assessment Tool) models to achieve these results, applying them to native grass systems, improved grass systems, and all regionally pertinent crops. This advances our ability to understand and predict regional agricultural outcomes under varying weather conditions.
Sub-objective 2.A. Several substantial enhancements were made to the SWAT+ model. A spatially detailed groundwater model was incorporated into the SWAT+ framework, which allows for improved streamflow and aquifer level prediction under irrigation pumping. At the same time, a more realistic simulation of nitrate and pesticide transport within aquifers was included. The model's capabilities to simulate carbon transformations was improved by developing soil and plant organic matter objects. The ongoing development is particularly focusing on the transportation of carbon in surface runoff, sediments, and soil percolate. In addition to these improvements, modules for managing rice paddy, including transplanting, puddling, and the application of fertilizers and pesticides, as well as irrigation management, were incorporated. An energy-based stream temperature algorithm was also developed and tested. Lastly, a comprehensive salt fate and transport model, covering atmospheric deposition, road salt application, plant stress, salt transport in rivers, reservoirs, and aquifers, and salt in irrigation water was developed, tested, and integrated into SWAT+.
Sub-objective 2.B. Significant progress was made in deriving plant parameters from PhenoCam observation. Results were published online at https://gmiller.tamu.edu/publication-archive/use-of-phenocam-measurements-and-image-analysis-to-inform-the-almanac-process-based-simulation-model/. We found that PhenoCam time series imagery can be used to improve leaf area development by adjusting parameter values to better match LAI (Leaf Area Index) derived values in new diverse environments. The results show how PhenoCam data can make a valuable contribution to validate process-based models, making these models much more realistic.
Subobjective 3.A. A draft version of the NAM (National Agroecosystems Model) was released to close collaborators for testing and use. Improvements were made to the water balance, reservoir release operations, irrigation sources, tile drains, stream connectivity, and crop yield calibration. Progress was also made on the development of nutrient and sediment loads for calibration by process domain.
Subobjective 3.B. To efficiently evaluate USDA conservation practices, a model called SWIFT (Speedy Watershed Integrated Forecasting Tool) was developed. Annual water, sediment, and nutrient landscape load predictions from NAM are utilized as input to inform SWIFT. SWIFT includes the identical routing structure of SWAT+ and has algorithms for river and reservoir transport. Although SWIFT only predicts annual loadings, the primary advantage is run time, cutting run times of large watersheds to a few seconds, facilitating its use as a decision support tool.
Remote sensing data collection at both the satellite and drone levels have also continued. In addition, in-situ readings (e.g. LAI, chlorophyll) are being collected. We have added an Oak Ridge Institute for Science and Education post-doc to this research which will expand our analytical possibilities in the use of machine learning and AI.
Accomplishments
1. Validating adoption of precision agriculture to optimize crop production, profitability and environmental outcomes. Over four years, ARS researchers in Temple, Texas, working in fields in Riesel, Texas, were intensively monitored, generating geospatial profitability maps from corn yields, input costs, and sales revenues. In 2022, management strategies were adjusted in four fields by reducing inputs in less profitable zones of two fields and converting unprofitable areas to grasslands in the other two. This change resulted in higher profitability compared to fields with unchanged management, under years with similar weather conditions. Furthermore, fields with altered management showed improved water quality compared to control fields. The study reaffirmed that precision agriculture can enhance crop productivity, profitability, and environmental stewardship.
2. Advancements in the SWAT+ (Soil and Water Assessment Tool) model. A spatially detailed groundwater model was integrated into the SWAT+ code, allowing for improved assessment of groundwater resources and the simulation of irrigation impacts on streamflow and aquifer levels. The model was also enhanced to provide a more realistic depiction of nitrate transport in aquifers. Testing was conducted in various locations with differing characteristics including the Mississippi Delta, Chesapeake Bay watershed, and Oregon. Additionally, development efforts focused on creating modules for rice paddy management and an energy-based stream temperature algorithm, as well as integrating a comprehensive salt fate and transport model into SWAT+. These efforts result in a more robust and versatile tool for environmental assessment and policy-making.
3. SWIFT: A speedy tool for evaluation of USDA conservation practices. To expedite the evaluation of USDA conservation practices, the SWIFT (Speedy Watershed Integrated Forecasting Tool) model was developed. SWIFT leverages annual water, sediment, and nutrient landscape loadings data from the enhanced SWAT+ model as inputs. Notably, SWIFT retains the same routing structure as SWAT+ and incorporates algorithms for river and reservoir transport. Despite its limitation to only predicting annual loadings, SWIFT's most significant advantage is its rapid execution, reducing the run times of large watersheds to mere seconds. This increased speed enables SWIFT to be the engine for future advanced decision support tools to inform national conservation policy.
4. Enhancing precision agriculture through machine learning and remote sensing techniques. Uptake of precision agriculture is rapidly expanding among agricultural stakeholders for the cost- and time-efficient benefits associated with identification of in-field biophysical and biochemical characteristics. As satellites and aerial vehicles obtain sensors with greater spectral sensitivities, these in-situ studies have potential to lead to large-scale fiscal savings. In addition, the increasing availability of machine learning/Artificial Intelligence (AI) presents opportunity to analyze these collected precision agriculture data through a novel and often improved lens. ARS researchers at Temple, Texas, through the implementation of new sensors and machine learning/AI methods alongside ongoing field-based research endeavors, precision agriculture methods are being analyzed in-depth and improved by this research. A superb example of this is in our recent publication entitled ‘Hyperspectral reflectance and machine learning to monitor legume biomass nitrogen accumulation’. Overall, the identification of best practices for precision agriculture can lead to more sustainable practices (e.g., nutrient application, irrigation management) and greater fiscal returns for United States agricultural producers.
Review Publications
Kiniry, J.R., Williams, A.S., Reisner, L., Hatfield, J.L., Kim, S. 2023. Effects of two categorically differing emergent wetland plants on evapotranspiration. Agrosystems, Geosciences & Environment. 6(1). Article e20331. https://doi.org/10.1002/agg2.20331.
Meki, M.N., Osorio-Leyton, J., Steglich, E.M., Kiniry, J.R., Propato, M., Winchell, M., Rathjens, H., Angerer, J.P., Norfleet, L.M. 2023. Plant parameterization and APEXgraze model calibration and validation for U.S. land resource region H grazing lands. Agricultural Systems. 207. Article 103631. https://doi.org/10.1016/j.agsy.2023.103631.
White, M.J., Arnold, J.G., Bieger, K., Allen, P.M., Gao, J., Cerkasova, N., Gambone, M.A., Park, S., Bosch, D.D., Yen, H., Osorio, J.M. 2022. Development of a field scale SWAT+ modeling framework for the contiguous U.S. Journal of the American Water Resources Association. 58(6):1545-1560. https://doi.org/10.1111/1752-1688.13056.
Elias, E.H., Tsegaye, T.D., Hapeman, C.J., Mankin, K.R., Kleinman, P.J., Cosh, M.H., Peck, D.E., Coffin, A.W., Archer, D.W., Alfieri, J.G., Anderson, M.C., Baffaut, C., Baker, J.M., Bingner, R.L., Bjorneberg, D.L., Bryant, R.B., Gao, F.N., Gao, S., Heilman, P., Knipper, K.R., Kustas, W.P., Leytem, A.B., Locke, M.A., McCarty, G.W., McElrone, A.J., Moglen, G.E., Moriasi, D.N., O'Shaughnessy, S.A., Reba, M.L., Rice, P.J., Silber-Coats, N., Wang, D., White, M.J., Dobrowolski, J.P. 2023. A vision for integrated, collaborative solutions to critical water and food challenges. Journal of Soil and Water Conservation. 78(3):63A-68A. https://doi.org/10.2489/jswc.2023.1220A.
Baez-Gonzalez, A.D., Melgoza-Castillo, A., Royo-Marquez, M.H., Kiniry, J.R., Meki, M.N. 2022. Modeling the distribution of wild cotton Gossypium aridum in Mexico using flowering growing degree days and annual available soil water. Sustainability. 14(11). Article 6383. https://doi.org/10.3390/su14116383.
Meki, N.N., Osorio, J.M., Steglich, E.M., Kiniry, J.R. 2022. Drought-induced nitrogen and phosphorus carryover nutrients in corn/soybean rotations in the Upper Mississippi River Basin. Sustainability. 14(22). Article 15108. https://doi.org/10.3390/su142215108.
Liu, W., Zhou, Y., Dong, J., Zhang, G., Yang, T., You, N., Flynn, K.C., Wagle, P., Yang, H. 2023. Cooling effects of increased green fodder area on native grassland in the northeastern Tibetan Plateau. Environmental Research Letters. 18(6). Article 064006. https://doi.org/10.1088/1748-9326/acc9d3.
Knutson, D., Irgens, M.S., Flynn, K.C., Norvilitis, J., Bauer, L.M., Berkessel, J.B., Cascalheira, C., Cera, J.L., Choi, N., Cuccolo, K., Danielson, D.K., Dascano, K.N., Edlund, J.E., Fletcher, T., Flinn, R., Gosnell, C.L., Heermans, G., Horne, M., Howell, J.L., Hua, J., Ijebor, E.E., Jia, F., McGullivray, S., Ogba, K., Shane-Simpson, C., Staples, A., Ugwu, C.F., Wang, S.C., Yockey, A., Zheng, Z., Zlokovich, M.S. 2023. Associations between primary residence and mental health in global marginalized populations. Community Mental Health Journal. https://doi.org/10.1007/s10597-023-01088-z.
Adhikari, K., Smith, D.R., Hajda, C.B., Kharel, T.P. 2023. Within-field yield stability and gross margin variations across corn fields and implications for precision conservation. Precision Agriculture. 24(4):1401-1416. https://doi.org/10.1007/s11119-023-09995-7.
Flynn, K.C., Baath, G., Lee, T.O., Gowda, P.H., Northup, B.K. 2023. Hyperspectral reflectance and machine learning to monitor legume biomass and nitrogen accumulation. Computers and Electronics in Agriculture. 211. Article 107991. https://doi.org/10.1016/j.compag.2023.107991.
