Location: Water Management and Systems Research2021 Annual Report
Objective 1: Quantify changes in agricultural production and fluxes of water and associated nutrients (N and P) and sediment from field to watershed scales over the next several decades at fine temporal resolutions in response to changes in water availability, land use, management practices, and climate. Sub-objective 1.1. Understand and quantify the effects of variable irrigation practices on crop production responses by assessing genotype x environment x limited-water management (GEM) interactions for different land use, management, and climate scenarios at field to watershed scales. Sub-objective 1.2. Improve estimates of water redistribution and storage by resolving spatial scale issues related to the measurement and simulation of soil moisture in cropland and grassland ecosystems at field to watershed scales. Simulate hydroecology within the SPRB and the Central Plains Experimental Range (CPER) Long-Term Agroecosystem Research (LTAR) site to extend experimental results to larger areas and different management scenarios. Sub-objective 1.3. Understand how the effectiveness of spatially distributed water conservation strategies and agricultural best management practices (BMPs) for nutrient and sediment control vary with landscape position, geographic/geologic characteristics of the field, farm, or watershed, and other factors. Objective 2: Assess key ecosystem services for projected water requirements and water quality targets in the South Platte River Basin, Colorado, at field to watershed scales in response to changes in water availability, land use, management, and climate. Objective 3: Develop and disseminate a web-based geospatial data management system as a repository of data, models, and tools for accelerating collaborative research and facilitating sustainable management of water, nutrients, and sediment.
Objectives 1 and 2 focus on enhancing scientific knowledge for incorporation into the Agricultural Ecosystems Services (AgES) distributed watershed model with subsequent testing and application of AgES. Objective 1 is divided into three sub-objectives integrated from smaller to larger scales, which focus on: (1.1) improved model components for plant modeling of GEM interactions, particularly for irrigated water management, (1.2) soil water modeling emphasizing spatial scaling of soil water and surface runoff in dryland cropping and rangeland systems, and (1.3) simulation of conservation effects over regional watersheds, primarily in Iowa where collaborators have been investigating and monitoring water quality impacts over decadal time scales. In Objective 2, the AgES model will be used to simulate a series of land use, management practice and climate scenarios for hydrologic and water quality ecosystem service indicators in eastern Colorado. Objective 3 involves development of a web-based Geospatial Portal for Scientific Research (GPSR) for technology transfer of geospatial information. GPSR will be used for dissemination of the results of the present project together with broader technology transfer by ARS and collaborators, such as experimental results generated from Long-Term Agricultural Research sites and Climate Hubs.
Objective 1. Sub-objective 1.1: ARS researchers at Fort Collins, Colorado, further developed the Unified Plant Growth Model (UPGM) to characterize response of crop phenology to water-stress across a range of genetic variation (grouped by variety maturity class) for 11 key crops: maize (corn), sorghum, soybean, dry (pinto) bean, winter wheat, spring wheat, winter barley, spring barley, sunflower, proso millet, and hay (foxtail) millet. Field data have been compiled to test interactions between genotype, environment, and management (GEM) for each crop at multiple sites in the Central Great Plains. Preliminary calibrations enhanced both UPGM and the soil-water module of the coupled UPGM/Agricultural Ecosystems Services (AgES) model. Objective 1. Sub-objective 1.3: The AgES watershed model (version 0.3.0) with field-level tile drainage simulated daily hydrology and nitrogen cycling and transport in the South Fork Iowa River. AgES parameters were calibrated to fit daily streamflow rates and nitrate concentrations at the outlet. In consultation with ARS researchers in Ames, Iowa and the Southfork Watershed Association, the recommended fertilizer and swine manure application rates were adjusted to match nitrate loads at the gauging station. The resulting simulations provided estimates of reduced inorganic fertilizer and increased manure application. Objective 1. Sub-objective 1.3: Two small watersheds in Brazil illustrate effects of an environmental services payment program. ARS researchers in Fort Collins, Colorado collaborated with a team in Brazil led by Embrapa Environment, where cereal, fiber, and oil crops occupies ~61 million hectares (ha), and about 90% of soybean is produced in no-tillage systems. The team evaluated the ecological and hydrological conditions of two small agricultural catchments in Itaí, São Paulo. The in-stream concentration of suspended solids was used to calculate the Index of Dissipation of Energy (IDE) as a rainfall index of protection from soil erosion. The resulting values of IDE indicate that agricultural conservation practices help control soil losses and maintain streamflow, which provided a quantitative metric for the benefits of government payments for conservation practices that may now be applied to other locations in Brazil and internationally. Objective 2: Following publication and presentations of AgES simulations of historical water flow and nitrate transport in the Big Dry Creek Watershed, Colorado, ARS and university researchers at Colorado State University responded to requests and comments from the Big Dry Watershed Association to simulate potential effects of best management practices (irrigation, fertilization, and tillage) on nitrate loads in the watershed. This ongoing work led to the current collaboration with Prairie View University, Texas (an historically black college). Researchers at Fort Collins, Colorado have developed and begun calibration of AgES simulations for burned and unburned watersheds in the upper montane zone in the northern Colorado Front Range to improve forecast ability in high elevation source watershed areas following wildfire. These sites include collaboration with Colorado State University (under a National Science Foundation rapid research grant) and Federal scientists (U.S. Army Research Lab) at Bighorn Creek (Estes Park, Colorado; Big Thompson River basin) an unburned analog to Dry Creek (Rustic, Colorado; Cache la Poudre River basin) which burned in the 2020 Cameron Peak fire – the largest fire to have burned in Colorado. Initial site work has included the installation of stream discharge gauges and weather stations to measure weather, precipitation, and soil moisture variables for the AgES model. Researchers have also begun developing an AgES simulation for the Fourmile burn area in the Boulder Creek watershed near Boulder, Colorado. This project is a collaboration with United States Geological Survey (USGS) scientists to assess hillslope impacts of fire and burn severity on post-fire soil moisture dynamics and vegetation recovery and impacts on site water balance. These interagency efforts address high-priority national research and have shaped objectives of the next project plan. Researchers in Fort Collins, Colorado have collected stream and snow samples for analysis of microplastics to assess atmospheric deposition versus streamflow contributions to mountain watersheds and their implications for water quality management. This project is a collaboration with local, regional, and federal stakeholders (e.g., City of Greeley, Colorado Department of Public Health and Environment, United States Forest Service). Initial data collection of water quality samples along an elevation gradient in the Cache la Poudre River basin, and of atmospheric deposition of microplastics during precipitation. Initial results indicate pervasive presence of microplastics in samples and suggest an important need to quantify spatial variation, sources, and impacts of microplastics on freshwater systems. This research is on the leading edge of an emerging environmental issue of local to state to national and international importance. In support of AgES applications, ARS researchers and collaborators at Colorado State University are developing and deploying web-based tools and services for processing spatial data and generating required model inputs. The Catchment areas delineation (Cadel) tool analyzes the watershed topology using a Digital Elevation Model and other spatial layers (such as soil and land use) to partition a watershed into interconnected Hydrological Response Units and generate essential input files for AgES. Model calibration (fitting simulated output to measured data by adjusting model parameters) is generally very challenging, time consuming, and computationally expensive. Existing methods used for AgES calibration are sequential (running one set of parameter values at a time) and can take weeks to months to find optimal values. ARS researchers in Fort Collins, Colorado are collaborating with Colorado State University and the University of Washington to develop new calibration methods that run multiple parameter sets in parallel to reduce this timeframe. A completely new approach combines concepts to implement Particle Swarm Optimization (PSO) with stepwise calibration of parameters grouped by biophysical processes. This project attracted $5000 in cloud computing credits awarded by Google. A new stepwise or multi-group PSO calibration service is being developed and tested using Google Cloud Services to run tens of AgES simulations in parallel. Integrated cloud services deliver results of complex process models using machine learning. New technology bridges boundaries between biophysical process-based models and deployment of rapid online solutions required by service delivery organizations, such as the Natural Resources Conservation Service (NRCS). ARS researchers and collaborators from Colorado State University and University of Trento, Italy, provided a computational framework to help streamline model setup, reduce runtime, and improve model infrastructure efficiency. Herein, an ensemble of artificial neural networks or surrogate models, captures the intrinsic knowledge of a process-based model. This automated process is validated and secured using blockchain technology. The proof-of-concept, wherein peak surface runoff provided by the curve number model is emulated with artificial intelligence, is prepared for adoption by the NRCS and other federal agencies and private corporations.
1. Guidelines for modeling water quality address international concerns of safe drinking water. In this synthesis, ARS researchers in Fort Collins, Colorado, collaborated with an international team (including the Australian National University, Utah State University, University of Napoli, University of Hohenheim, Helmholtz Center for Environmental Research, AgResearch New Zealand, and other U.S. federal agencies) to assess present research and anticipate future development needs in watershed water quality modeling. The work addresses improvement in the representation of freshwater systems pertaining to water quality, including environmental interfaces, in-stream water quality, soil health and land management, and (peri-) urban areas. Identification of the main challenges science, infrastructure, and practices led to three recommendations for improving watershed water quality modeling. These include building stronger collaborations between experimentalists and modelers, bridging gaps between modelers and stakeholders, and cultivating and applying procedural knowledge to better govern and support water quality modeling processes within organizations. The recommended approaches are being used and cited internationally by university researchers and government organizations solving water quality issues in receiving water bodies and drinking water.
2. Study of a highly managed peri-urban watershed reveals significant nitrate loading. In response to water quality concerns expressed by the City and County of Broomfield, City of Northglenn, and City of Westminster, ARS researchers in Fort Collins, Colorado, and Colorado State University compared spatial patterns of water flow and nitrate transport simulated using the fully distributed Agricultural Ecosystems Services (AgES) model and the semi-distributed Soil and Water Assessment Tool (SWAT) model. Patterns of water movement across the peri-urban watershed simulated with the two models were drastically different. AgES produced more realistic estimates of watershed-outlet streamflow, irrigation return flows, and in-stream water quality, indicating the value of AgES in water resource simulation in the semi-arid and arid western U.S. The Big Dry Creek Watershed Association and the Colorado State Department of Public Health and Environment are using these results to enhance water quality monitoring efforts and to explore agricultural conservation practices as offsets for municipal wastewater loads.
3. Gully erosion is an often forgotten but critical mechanism of soil erosion. Experts estimate that 10 to 94% of cropland erosion comes from ephemeral gullies, which are small channels eroded by concentrated overland flow. Model results are often used to guide implementation of soil erosion control practices, but because soil erosion models typically do not simulate ephemeral gully erosion accurately, or at all, ephemeral gullies can be overlooked in soil conservation efforts. An ARS researcher in Fort Collins, Colorado, collaborated with researchers at the University of Guelph - Canada and Kansas State University to summarize the current status of ephemeral gully modeling, describe shortcomings of existing models, and provide a framework for future research and development. This research has already provided foundational information for further ephemeral gully modeling work by stakeholders and other research teams internationally.
Moeser, C.D., Douglas-Mankin, K.R. 2021. Simulating hydrologic effects of wildfire on a small sub-alpine watershed in New Mexico, U.S. International Journal of Wildland Fire. 64(1):137-150. https://doi.org/10.13031/trans.13938.
Sheshukov, A.Y., Gao, J., Douglas-Mankin, K.R., Yen, H. 2021. Intercomparison of bias-correction data sources and their influence on watershed-specific downscaling climate projections. Transactions of the ASABE. 64(1):203-220. https://doi.org/10.13031/trans.14061.
Douglas-Mankin, K.R., Roy, S.K., Sheshukov, A.Y., Biswas, A., Gharabaghi, B., Binns, A., Rudra, R., Shrestha, N.K., Daggupati, P. 2020. A comprehensive review of ephemeral gully erosion models. Catena. 195. Article e104901. https://doi.org/10.1016/j.catena.2020.104901.
Douglas-Mankin, K.R., Helmers, M.J., Harmel, R.D. 2021. Review of filter strip performance and function for improving water quality from agricultural lands. Transactions of the ASABE. 64(2):659-674. https://doi.org/10.13031/trans.14169.
Valiya-Veettil, A., Green, T.R., Kipka, H., Arabi, M., Lighthart, N.P., Douglas-Mankin, K.R., Clary, J. 2021. Fully distributed versus semi-distributed process simulation of a highly managed watershed with mixed land use and irrigation return flow. Environmental Modelling & Software. 140. Article e105000. https://doi.org/10.1016/j.envsoft.2021.105000.
Fu, B., Horsburgh, J.S., Jakeman, A., Gualtieri, C., Arnold, T., Marshall, L., Green, T.R., Quinn, N.W., Volk, M., Hunt, R.J., Vezzaro, L., Croke, B., Jakeman, J., Snow, V., Rashleigh, B. 2020. Modeling water quality in freshwater systems: From here to the next generation. Water Resources Research. 56(11). Article e2020WR027721. https://doi.org/10.1029/2020WR027721.