ENHANCING THE CAPABILITIES OF SWAT+ TO SIMULATE RUNOFF AND POLLUTANT TRANSPORT IN AGRICULTURAL LANDSCAPES

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: HATCH
• Reporting Frequency: Annual
• Accession No.: 1012803
• Project Start Date: May 2, 2017
• Project End Date: Apr 17, 2022
• Program Code: [(N/A)]- (N/A)

Recipient Organization
TEXAS A&M UNIVERSITY
750 AGRONOMY RD STE 2701
COLLEGE STATION,TX 77843-0001

Performing Department
Temple-Blackland Res Cntr

Non Technical Summary
The management of watersheds is a very complex task because the requirements of both agriculture and the environment must be considered. Computer simulation models are frequently used to investigate the impact of management strategies due to their ability to quickly and efficiently compare different strategies. The Soil and Water Assessment Tool (SWAT) is a computer simulation model that is used all over the world to predict the movement of water, sediment, and agricultural chemicals from the landscape to water bodies and through the stream network of a watershed. SWAT can be used to assess the impact of changes in land use or climate on water resources and to estimate the effectiveness of conservation practices in agriculture. The model is constantly evolving in response to advancements in scientific knowledge and the needs of the user community. Recently, major modifications were made to the model that allow for a more realistic spatial representation of the connectivity of different elements within a watershed, e.g. upland areas, ponds, reservoirs, floodplains, and streams. The resulting new version of the model, SWAT+, will require extensive testing over the next few years to make sure it predicts fluxes of water, sediment, and agricultural chemicals as realistically as possible. Therefore, one of the objectives of this project is to apply SWAT+ in several watersheds in the United States and verify the model output using observed data. Also, the predictive capabilities of the model will be enhanced by dividing the watersheds into two landscape units, uplands and floodplains, to account for differences in their hydrologic characteristics, route water across the landscape, and improve the simulation of floodplain processes and stream-aquifer interactions. Furthermore, this project focuses on the development of national SWAT+ models that can be used to assess the effectiveness of conservation practices and provide daily updated information for stakeholders to optimize the scheduling of agricultural management operations. Finally, this project also aims at improving the simulation of flow and nutrient transport in tile-drained systems. While systems with sub-surface tile drainage networks experience increased crop growth and yield, tile drains also increase the connectivity of agricultural fields to surface ditches and streams, thus providing pathways for nutrient transport to nearby surface water bodies and leading to eutrophication and harmful algae blooms. Therefore, there is a fundamental need for a model that has the capability to investigate management strategies that can reduce nutrient loss in tile-drained systems. The main products of this project will be improved hydrologic modeling tools and information generated by applying these, which can be used by government agencies, watershed management teams, researchers, and stakeholders in the United States and worldwide to promote vigorous agriculture while at the same time preserving or restoring environmental health. The models can aid the development of more efficient management strategies and policies, thus leading to a decrease in money spent on inefficient use of resources or misplaced conservation practices.

Animal Health Component

15%

Research Effort Categories

Basic

15%

Applied

15%

Developmental

70%

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
112	0399	2050	100%

Knowledge Area
112 - Watershed Protection and Management;

Subject Of Investigation
0399 - Watersheds and river basins, general;

Field Of Science
2050 - Hydrology;

Keywords

hydrology

hydrologic modeling

watershed management

nutrient losses

conservation practices

sub-surface tile drains

soil and water assessment tool

Goals / Objectives
The overall goal of this project is to contribute to the ongoing development of SWAT+ (Soil and Water Assessment Tool) in response to advancements in scientific knowledge and the needs of the SWAT user community. This includes (1) testing SWAT+ extensively with regard to water quantity and quality (sediment, nutrients, and pesticides) in various watersheds, (2) the development and extensive testing of an approach to integrate landscape units in SWAT+, (3) contributing to the development of national SWAT+ models, and (4) developing an improved modeling framework for assessing nutrient transport in tile-drained watersheds.Testing SWAT+ in various watershedsVerification of plant growth, water balance, streamflow, sediment, nutrient, and pesticide modules in SWAT+.Identification of errors in the model subroutines and implementation of modifications to the SWAT+ code.Dissemination of SWAT+ to the user community.Integration of landscape units in SWAT+Identification of appropriate algorithms to describe the landscape and water body connectivity and implementation in SWAT+.Analysis of the impact of implementing landscape units in SWAT+ on model predictions.Development of national SWAT modelsCollection of required input data, e.g. topography, soils, land use and management, and climate.Delineation of upland and floodplain areas for all watersheds in the conterminous United States.Setup of SWAT+ models for all HUC8 (8-digit Hydrologic Unit Code) watersheds in the United States with HUC12 (12-digit Hydrologic Unit Code) watersheds as subbasins and calibration of the models against all available hard and soft data.Development of a website with county-level maps displaying model output that are updated daily and can help stakeholders make management decisions.Improved modeling framework for assessing nutrient transport in tile-drained watershedsDevelopment and testing of a process-based watershed model that explicitly simulates water flow and nutrient transport in subsurface tile drains.Improvement of our understanding of the hydrological and biogeochemical processes affecting the transport, storage, and transformation of nitrogen and phosphorus species in tile-drained systems.Assessment of viable conservation practices that maintain ecosystem services in the watershed.Dissemination of the developed modeling framework to the research community.

Project Methods
Testing SWAT+ in various watershedsExtensive testing of SWAT+ will be necessary over the coming years to make sure that the changes made to the structure of code and input files have not modified the simulation of processes in the model. So far, three watersheds in the US have been selected, the Little River Experimental Watershed in Georgia, the Middle Bosque River Watershed in Texas, and the South Fork Wildcat Creek Watershed in Indiana. Additional test watersheds will be added as needed to evaluate the model performance at various scales and in as many different environmental settings as possible. Testing of the model will involve detailed analysis of model output on daily, monthly, yearly and average annual time steps. Model predictions of plant growth (biomass and leaf area index), evapotranspiration, runoff components in the landscape (surface runoff, lateral flow, tile flow, and groundwater flow), snowfall and snowmelt, streamflow, erosion, and transport of sediment and agricultural chemical loads across the landscape and through the stream network will be analyzed visually by plotting time series, flow duration curves, boxplots, and scatterplots and statistically using common model evaluation indices, e.g., Nash-Sutcliffe-Efficiency, coefficient of determination, and percent bias. All available hard and soft data will be used for comparison to simulated data. Whenever there are significant discrepancies between observed and simulated data or inconsistencies in the model output, the code will be analyzed to identify the cause and fix the problem. This is an iterative process that will continue over the entire duration of this project. New capabilities of the model that have been thoroughly tested will be described in scientific articles published in relevant journals to inform the scientific community. SWAT+ will be released and available for download as soon as the first round of testing is completed. A manual will be available to facilitate use of the model for new users and users that have used SWAT in the past. Also, technical support will be provided for users and workshops will be taught according to needs and opportunities. SWAT+ is expected to generate a lot of interest in the scientific community and with watershed planners and environmental agencies. The success of this part of my project will be indicated by the growing numbers of SWAT+ users and publication of SWAT+ studies over the coming years.Integration of landscape units in SWAT+This part of my project will involve an extensive literature research to find existing methods for identifying the dominant processes producing fluxes of water, sediment, and agricultural chemicals in watersheds. Subsequently, promising methods will be tested in the SWAT+ test watersheds mentioned above. Once the most appropriate method has been identified, it will undergo further testing and will be published in one or more scientific articles. The analysis of model output will be conducted using the methods described above in the section "Testing SWAT+ in various watersheds". The expected advantages of dividing the watershed into landscape units will be demonstrated by comparing results from model setups with and without landscape units. Development of national SWAT+ modelsThe development of national SWAT+ models at Blackland Research Center and Grassland, Soil and Water Laboratory in Temple includes the collection of input data, construction of the models, calibration and evaluation. The first step will be to compile all required input data (topography, soils, land use and management, climate) in large databases and to write code that will extract the data from those databases and use it to construct the datasets for all major river basins in the United States. For each watershed, two models will be set up, one without landscape units and one with two landscape units, uplands and floodplains. For the latter, floodplains will be delineated for the entire country using a method I have tested in the past. Once the models are constructed, they will be calibrated against all available hard and soft data. The analysis of model output will be conducted using the methods described above in the section "Testing SWAT+ in various watersheds". After successful calibration, the models will be used for running scenarios assessing the effectiveness of different conservation practices. Also, they will be run with daily updated weather predictions for the future, which can be obtained online. Output from the model will be available on a website that stakeholders can access to get up-to-date information and predictions into the future of variables that are relevant for scheduling agricultural management operations, e.g., soil moisture and plant growth.Improved modeling framework for assessing nutrient transport in tile-drained watershedsThe first step of this part of my project will be to incorporate tile drain flow and transport processes into an existing SWAT-MODFLOW-RT3D model. SWAT-MODFLOW is a physically-based, spatially-distributed coupled land surface / groundwater flow and solute transport model that incorporates the advantage of both base models: SWAT simulates land surface hydrological processes, whereas MODFLOW simulates three-dimensional groundwater flow in a heterogeneous aquifer system. RT3D simulates nitrate and phosphorus reactive transport in groundwater. Tile drain flow will be simulated using the Streamflow Routing (SFR) package of MODFLOW, and transport of nitrate and phosphorus within the tile drain flow will be simulated using the USGS finite difference modeling code OTIS (One-Dimensional Transport with Inflow and Storage). The tile drain simulations will be performed within the MODFLOW/RT3D routines, with output provided to SWAT. The new model will be tested in three watersheds in the United States, the Lower Arkansas River Valley in Colorado, the South Fork of the Iowa River Watershed in Iowa, and the Upper Cedar Creek Watershed in Indiana. The three test watersheds provide a contrast in climatic conditions and cultivation practices. Data for model testing are available for all three watersheds across multiple spatial scales: field scale, multi-field scale, and catchment scale. The analysis of model output will be conducted using the methods described above in the section "Testing SWAT+ in various watersheds". Model evaluation will focus on comparing nitrogen and phosphorus loading between the major transport pathways surface runoff, tile flow, and groundwater flow, investigating the impacts of varying weather patterns using multi-decadal historical weather data, and comparing results from the enhanced model with results from the original SWAT-MODFLOW-RT3D model, which is expected to highlight the need for tile drainage to be explicitly included in the model. Subsequently, viable conservation practices for nitrogen and phosphorus remediation in tile drain systems under wet and dry conditions will be assessed by simulating their effectiveness using the enhanced model. The assessment will be based on preservation or enhancement of agricultural productivity, total cost, vulnerability, equity, resilience, reliability, and chance of adoption. Also, the best economic policies to accompany the conservation practices will be investigated. Finally, the enhanced model will be disseminated to the research community via conference workshops, the development of a graphical user interface accompanied by a detailed user manual, and the development of decision-support tools for balancing ecosystem service maintenance with required crop production.

Progress 05/02/17 to 04/17/22

Outputs
Target Audience:The tools that are being developed/improved by this project (SWAT and SWAT+, SWAT+ with groundwater module, SWAT+ with tile drainage routines) are used by researchers across the United States and worldwide. The project contributes to the development of a national SWAT+ model, which is used by the Conservation Effects Assessment Project to provide the data needed by the federal government to allocate funds for conservation practices in agriculture. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?WAT+ Introductory Workshop, 18-22 November 2019, Sydney, Australia, 16 participants SWAT+ Introductory Workshop, 28-29 November 2019, Christchurch, New Zealand, 30 participants SWAT+ Introductory Workshop, 01-04 March, 2021. Online. >30 participants How have the results been disseminated to communities of interest?Peer-reviewed journal articles. SWAT+ Session at the 2020 Virtual Geospatial Water Technology Conference of the American Water Resources Association SWAT+ workshops in 2019 and 2021 What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Computer simulation models are frequently used to investigate the impact of management strategies due to their ability to quickly and efficiently compare different strategies. Improved modeling tools are essential for the management of natural and agricultural resources. The Soil and Water Assessment Tool (SWAT) is a computer simulation model that is used all over the world to predict the movement of water, sediment, and agricultural chemicals from the landscape to water bodies and through the stream network of a watershed. The model is constantly evolving in response to advancements in scientific knowledge and the needs of the user community. This project contributes significantly to improving the model and testing new model components, thereby providing researchers, watershed managers, and stakeholders with a better tool for making more informed management decisions. Accomplishments in testing SWAT+ in various watersheds: Testing of SWAT+ has continued over the reporting period and will be ongoing for several years to come. During the entire period the SWAT+ model was applied to the entire Contiguous U.S. The simulation of hydrological processes in the model was improved and various bugs in the code were identified and fixed. Accomplishments in development of national SWAT models: The representation of stream network connectivity in the national SWAT+ setups was finalized. The models were calibrated against average annual water budgets and crop yields, and results were satisfactory in a large majority of watersheds across the US. Accomplishments in improved modeling framework for assessing nutrient transport in tile-drained watersheds: A new module that explicitly simulates water flow and nutrient transport in subsurface tile drains was integrated in SWAT+. The model was applied to the South Fork of the Iowa River Watershed in Iowa and after calibration it provided good simulations of water budgets and streamflow.

Publications

• Type: Other Status: Submitted Year Published: 2022 Citation: Natalja ?erkasova, Michael White, Jeffrey Arnold, Katrin Bieger, Peter Allen, Jungang Gao, Marilyn Gambone, Manyowa Meki, James Kiniry, Philip Gassman. Field Scale, National Scope SWAT+ Modeling of Corn and Soybean Yields for the Contiguous United States. Manuscript number: AGSY-D-22-00469

Progress 10/01/20 to 09/30/21

Outputs
Target Audience:The tools that are being developed and improved by this project (SWAT and SWAT+,are used by researchers across the United States, the EU and worldwide. The project contributes to the development of a US national SWAT+ model, which is used by the Conservation Effects Assessment Project (CEAP-Wildlife) to provide the data needed by the federal government to allocate funds for conservation practices in agriculture. The project has recently got a major internatiopnal ineterst, whenan EU funded project under the Horison 2020 programme (H2020 OPTAIN) has adopted SWAT+ as the main model for their studies. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Online SWAT+ Workshop: March 1-4, 2021 Using Google Earth Engine for Land Monitoring Applications: June 16, 23, 30,2021 How have the results been disseminated to communities of interest? Peer-reviewed journal articles - see list of publications Project meetings (internal) What do you plan to do during the next reporting period to accomplish the goals?Testing SWAT+ in various watersheds: Testing of SWAT+ will continue on the US National and a more focused - local scale. The work will continue on the selected UMRB sub-catchments. SWAT+ will be applied in 14 different EU countries by collaborators within the H2020 OPTAIN project. It is expected that in many cases a certain degree of consultancy or advise will be needed and provided. Integration of landscape units in SWAT+: Guidelines on how to properly define the landscape units within various catchments will be prepared and disseminated. Proper testing of different landscape setup methods will be performed and assessed. Development of national SWAT models: US National Scale model "hard" calibration will be continued, with more areas tested against a set of observations. Some identified model setup bugs will be addressed and corrected. Other national models (LT and LV) should be completed in the upcoming months by the partners. Improved modeling framework for assessing nutrient transport in tile-drained watersheds: The testing of the tile-drainage module will continue on a focused group of selected watersheds, most likely the ones that are situated in the UMRB. The results of these tests will be disseminated withing the project group.

Impacts
What was accomplished under these goals? The significant part of the last years study was dedicated to testing the capabilities of SWAT+ to accurately represent the hydrological cycle and the plant growth simulations at the national scale. Testing SWAT+ in various watersheds: Testing of SWAT+ has continued over the reporting period and will be ongoing.As the model is developed further, so is the testing necessity of various model functions and modules. Now the testing will be focused on the Upper Mississippi River Basin (UMRB). Model sensitivity for streamflow-related parametershas already been tested for three specific HUC8 catchments(Upper Sangamon 07130006, LeSueur 07020011, North Raccoon 07100006). Most sensitive parameters have been identified. The study will now continue to test the water quality related parameters. Integration of landscape units in SWAT+: The integration of the landscape units has been fully completed and is now a part of the SWAT+ model concept. Further guide on the usage of landscape units is being developed in collaboration with EU collaborators:dr. Michael Strauch (UFZ, Germany), dr. Christoph Schürz (UFZ, Germany),dr. Miko?aj Piniewski (SGGW, Poland), and others. Development of national SWAT models: The US national model has been calibrated for water balance components, and crop yields (corn and soybeans). The model shows good agreement with observed or reported valuesafter the completed "soft" calibration. The "hard" calibration is ongoing, where we use the set of measured streamflow and water quality data of selected gages. The SWAT+ is used for National surface water qualitymodels of Lithuania and Latvia (Europe). We have aided and consulted the development team on the best course of action regarding the model application and calibration. The models are in development; hence no results are presented yet. Improved modeling framework for assessing nutrient transport in tile-drained watersheds: A new module that explicitly simulates water flow and nutrient transport in subsurface tile drains was integrated in SWAT+. Several inaccuracies (bugs) were detected during the testing phase of the model, and a in-depth debugging was undergone. The model is repeatedly tested, withpromisingresults.

Publications

• Type: Journal Articles Status: Published Year Published: 2021 Citation: Natalja ?erkasova, Georg Umgiesser, Ali Ert�rk, Modelling framework for flow, sediments and nutrient loads in a large transboundary river watershed: A climate change impact assessment of the Nemunas River watershed. Journal of Hydrology, Volume 598, 2021, 126422, ISSN 0022-1694, https://doi.org/10.1016/j.jhydrol.2021.126422.
• Type: Journal Articles Status: Submitted Year Published: 2021 Citation: Michael J. White, Jeffrey G. Arnold, Katrin Bieger, Jungang Goa, Natalja Cerkasova, Marilyn Gambone, Seonggyu Park, David Bosch, Haw Yen, Javier Osorio. Development of a Field Scale SWAT+ Modeling Framework for the Contiguous U.S. Journal of the American Water Resources Association.

Progress 10/01/19 to 09/30/20

Outputs
Target Audience:The tools that are being developed/improved by this project (SWAT and SWAT+, SWAT+ with groundwater module, SWAT+ with tile drainage routines) are used by researchers across the United States and worldwide. The project contributes to the development of a national SWAT+ model, whichis used by the Conservation Effects Assessment Project to provide the data needed by the federal government to allocate funds for conservation practices in agriculture. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?SWAT+ IntroductoryWorkshop, 18-22 November 2019, Sydney, Australia, 16 participants SWAT+IntroductoryWorkshop, 28-29 November 2019, Christchurch, New Zealand, 30 participants How have the results been disseminated to communities of interest?Peer-reviewed journal articles - see list of publications. SWAT+ Session at the 2020 Virtual Geospatial Water Technology Conference of the American Water Resources Association. What do you plan to do during the next reporting period to accomplish the goals? Testing SWAT+ in various watersheds: Continue testing of SWAT+ in the Little River Experimental Watershed in GA, the Middle Bosque Watershed in TX, the South Fork of the Iowa River Watershed in IA, the Kielstau Catchment in Germany, and the Weatherley Catchment in South Africa. Integration of landscape units in SWAT+: Work on the hydropedological approach to simulatingstreamflow and soil water contents with SWAT+ will continue. The paper on the integration of landscape units in SWAT+ will be finalized. Development of national SWAT models: Continued calibration of crop yields across the US. Calibration of simulated streamflow and nutrient loads against observed data from all available USGS gages. Improved modeling framework for assessing nutrient transport in tile-drained watersheds: Calibration of nutrient loads in the South Fork Watershed in Iowa to validate the simulation of nitrogen and phosphorus transport in tile drains. Construction of a SWAT+ model for two watersheds in Ohio and Indiana to use as additional test watersheds for the new tile drainage routines for SWAT+.

Impacts
What was accomplished under these goals? Computer simulation models are frequently used to investigate the impact of management strategies due to their ability to quickly and efficiently compare different strategies. Improved modeling tools are essential for the management of natural and agricultural resources. The Soil and Water Assessment Tool (SWAT) is a computer simulation model that is used all over the world to predict the movement of water, sediment, and agricultural chemicals from the landscape to water bodies and through the stream network of a watershed. The model is constantly evolving in response to advancements in scientific knowledge and the needs of the user community. This project contributes significantly to improving the model and testing new model components, thereby providing researchers, watershed managers, and stakeholders with a better tool for making more informed management decisions. Testing SWAT+ in various watersheds: Testing of SWAT+ has continued over the reporting period and will be ongoing for several years to come. In the past year, testing has included theLittle River Experimental Watershed in Georgia, the Middle Bosque River Watershed in Texas, the South Fork of the Iowa River Watershed in Iowa,the Kielstau Catchment in Germany, and the Weatherley Watershed in South Africa. The simulation of hydrological processes in the model was improved and various bugs in the code were identified and fixed. Integration of landscape units in SWAT+: A new collaboration was started with Johan van Tol, a researcher from South Africa, to investigate the use of landscape units in SWAT+ to incorporatehydropedological information for simulating internal watershed processes and flow connectivity. Interpreting soil morphological properties and relating them to the hydrological processes which formed them can be used to determine dominant flowpaths, storage mechanisms and connectivity between different flowpaths. This allows for a more holistic understanding of the hydrological functioning of landscapes at different scales (i.e. catchments and /or hillslopes) and can be used in process based modelling, wetland protection and pollution studies. Development of national SWAT models: The representation of stream network connectivity in the national SWAT+ setups was finalized. The models were calibrated against average annual water budgets and results were satisfactory in a large majority of watersheds across the US. Calibration of crop yields across the nation has started and results arepromising for the major crops corn, soybeans and winter wheat. Improved modeling framework for assessing nutrient transport in tile-drained watersheds: A new modulethat explicitly simulates water flow and nutrient transport in subsurface tile drains was integrated in SWAT+. The model was applied to the South Fork of the Iowa River Watershed in Iowa and after calibration it provided good simulations of water budgets and streamflow. Two additional test watersheds located in Ohio and Indiana were selected and the collection of input data needed for SWAT+ model construction has started.

Publications

• Type: Journal Articles Status: Published Year Published: 2020 Citation: Jungang Gao, Michael J. White, Katrin Bieger, and Jeffrey G. Arnold, 2021. Design and development of a Python-based interface for processing massive data with the LOAD ESTimator (LOADEST). Environmental Modelling & Software 135, 104897. DOI: 10.1016/j.envsoft.2020.104897
• Type: Journal Articles Status: Published Year Published: 2020 Citation: Jeffrey G. Arnold, Michael J. White, Peter M. Allen, Philip W. Gassman, and Katrin Bieger, 2020. Conceptual Framework of Connectivity for a National Agroecosystem Model Based on Transport Processes and Management Practices. Journal of the American Water Resources Association (JAWRA). DOI: 10.1111/1752-1688.12890
• Type: Journal Articles Status: Published Year Published: 2020 Citation: Ryan T. Bailey, Katrin Bieger, Jeffrey G. Arnold, and David D. Bosch, 2020. A new physically-based spatially-distributed groundwater flow module for SWAT+. Hydrology 7, 75. DOI: 10.3390/hydrology7040075
• Type: Journal Articles Status: Published Year Published: 2020 Citation: Jungang Gao, Katrin Bieger, Michael J. White, and Jeffrey G. Arnold, 2020. Development and accuracy assessment of a HUC-12 based Real-Time climate database for hydrologic models in the US. Journal of Hydrology 586, 124817. DOI: 10.1016/j.jhydrol.2020.124817
• Type: Journal Articles Status: Published Year Published: 2020 Citation: Ryan T. Bailey, Seonggyu Park, Katrin Bieger, Jeffrey G. Arnold, and Peter M. Allen, 2020. Enhancing SWAT+ Simulation of Groundwater Flow and Groundwater-Surface Water Interactions using MODFLOW routines. Environmental Modelling & Software 126, 104660. DOI: 10.1016/j.envsoft.2020.104660

Progress 10/01/18 to 09/30/19

Outputs
Target Audience:The tools that are being developed or improved by this project (SWAT/SWAT+, SWAT-MODFLOW) areused by researchers across the United States and worldwide. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?A PhD student from the University of Natural Resources and Life Sciences in Vienna, Austria was trained to use SWAT+ during a 4-month visit in Temple, TX from August to December 2018. A two-day Introductory SWAT+ Workshop was conducted on July 15-16, 2019 during the 2019 International SWAT Conference in Vienna, Austria. The workshop was attended by 35 participants from all over the world. Custom training was provided to several SWAT+ users as part of user support activities. How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals?Testing of SWAT+ in several watersheds will continue. More test watersheds will be added. Testing will focus more specifically on sediment and nutrients. Additional SWAT+ workshops will be offered over the coming years. A paper about the integration of landscape units in SWAT+ is currently under preparation. Errors in the representation ofstream network connectivity in the national SWAT+ models will be fixed. The national models will be calibrated against average annual values of water balance components and observed streamflow. Calibration of the SWAT+ model for the South Fork of the Iowa River Watershed in Iowa will continue. The SWAT+ model will be linked with a new algorithmthat explicitly simulates water flow and nutrient transport in subsurface tile drains.

Impacts
What was accomplished under these goals? Computer simulation models are frequently used to investigate the impact of management strategies due to their ability to quickly and efficiently compare different strategies.Improved modeling tools are essential for the management of natural and agricultural resources.The Soil and Water Assessment Tool (SWAT) is a computer simulation model that is used all over the world to predict the movement of water, sediment, and agricultural chemicals from the landscape to water bodies and through the stream network of a watershed.The model is constantly evolving in response to advancements in scientific knowledge and the needs of the user community.This project contributes significantly to improving the model andtesting new model components, thereby providing researchers,watershed managers, and stakeholder with a better tool for making more informed management decisions. Testing SWAT+ in various watersheds: Testing of SWAT+ has continued over the reporting period and will be ongoing for several years to come. In the past year, testing has included theLittle River Experimental Watershed in Georgia, the Middle Bosque River Watershed in Texas, the South Fork of the Iowa River Watershed in Iowa, and the Kielstau Catchment in Germany. A revised version of SWAT+ was released to the user community in July 2019. Integration of Landscape Units in SWAT+: Landscape units are integrated in the latestversion of SWAT+. A paper analyzing the impacts of landscape units on the simulation of upland - floodplain - stream connectivity was published in the Journal of the American Water Resources Association in 2019. Development of national SWAT+ models:The construction of SWAT+ models for all HUC8 watersheds in the US was completed in 2019. However,there are still some erros in the representation of thestream network connectivity in the model. A website with county-level maps of model output is available under https://realtimeceap.brc.tamus.edu/. Improved modeling framework for assessing nutrient transport in tile-drained watersheds: A SWAT+ model for the South Fork of the Iowa River Watershed in Iowa was constructed, which will be used as a test watershed for a new process-based watershed model that explicitly simulates water flow and nutrient transport in subsurface tile drains. Calibration of the model started during the reporting period, but is still ongoing.

Publications

• Type: Journal Articles Status: Published Year Published: 2019 Citation: Katrin Bieger, Jeffrey G. Arnold, Hendrik Rathjens, Michael J. White, David D. Bosch, and Peter M. Allen, 2019. Representing the connectivity of upland areas to floodplains and streams in SWAT+. Journal of the American Water Resources Association (JAWRA) 55(3):578-590. DOI: 10.1111/1752-1688.12728
• Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Katrin Bieger, Jeffrey G. Arnold, Seonggyu Park, Ryan T. Bailey, and Peter M. Allen, 2019. Overview of SWAT+ options for simulating groundwater processes. 2019 International SWAT Conference, July 17-19, 2019, Vienna, Austria.

Progress 10/01/17 to 09/30/18

Outputs
Target Audience:The tools that are being developed or improved by this project (SWAT/SWAT+, SWAT-MODFLOW)are currently mostly used by researchers across the United States and worldwide. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?An introductory workshop about SWAT+ was offered at the 2018 iEMSs Conference from June 24 to 28, 2018 in Fort Collins, Colorado. How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Testing SWAT+ in various watersheds:The plant growth, water balance, and streamflow modules of SWAT+ were tested in the Little River Experimental Watershed in Georgia, the Middle Bosque River Watershed in Texas, and the Kielstau Catchment in Germany. Testing of the sediment and nutrient modules is ongoing.SWAT+ model debugging is an ongoing process that will continue for several years. SWAT+ has been released to the user community. Integration of landscape units in SWAT+: Landscape units are integrated in the recently released version of SWAT+. A paper analyzing the impact of landscape units on the simulation of upland - floodplain - stream connectivity was submitted to the Journal of the American Water Resources Association. Development of national SWAT models: The collection of required input data has been completed except for point source and wetland data. A floodplain map for the conterminous US is available. The setup of SWAT+ models for all HUC8 watersheds in the US is partly done. The definition of Hydrologic Response Units for simulating land phase processes is completed, but the definition of the stream network connectivity in the model is still ongoing. A first version of a website with county-level maps of model output is available. Improved modeling framework for assessing nutrient transport in tile-drained watersheds: A process-based watershed model that explicitly simulates water flow and nutrient transport in subsurface tile drains has been developed by including tile drain flow (Streamflow Routing (SFR) package of MODFLOW) and nutrient transport processes (OTIS - One-Dimensional Transport with Inflow and Storage) into an existing SWAT-MODFLOW-RT3D model. Model testing in several watersheds is ongoing.

Publications

• Type: Journal Articles Status: Published Year Published: 2017 Citation: Michael J. White, Katrin Bieger, Marilyn Gambone, Elizabeth Haney, Jeff Arnold, and Jungang Gao, 2017. Development of a Hydrologic Connectivity Dataset for SWAT Assessments in the US. Water 9, 892. DOI: 10.3390/w9110892
• Type: Journal Articles Status: Published Year Published: 2018 Citation: Jeffrey G. Arnold, Katrin Bieger, Michael J. White, Raghavan Srinivasan, John A. Dunbar, and Peter M. Allen, 2018. Use of Decision Tables to Simulate Management in SWAT+. Water 10, 713. DOI: 10.3390/w10060713
• Type: Conference Papers and Presentations Status: Other Year Published: 2018 Citation: Katrin Bieger, Jeffrey G. Arnold, Michael J. White, David D. Bosch, Peter M. Allen. Exploring the sensitivity of upland  floodplain  stream connectivity in SWAT+. 2018 International SWAT Conference, September 19-21, 2018, Brussels, Belgium.
• Type: Conference Papers and Presentations Status: Other Year Published: 2018 Citation: Katrin Bieger, Michael J. White, Jeffrey G. Arnold. Large-scale hydrologic assessment using SWAT. Integrated Assessment Models and the Social Costs of Water Pollution Conference. April 16-17, 2018, Ames, IA, USA.
• Type: Conference Papers and Presentations Status: Other Year Published: 2018 Citation: Katrin Bieger, Raghavan Srinivasan, Jeffrey G. Arnold. SWAT watershed modeling: Recent model development. Water and Livelihoods Initiative (WLI) Final Regional Workshop. March 26-27, 2018, Amman, Jordan.

Progress 05/02/17 to 09/30/17

Outputs
Target Audience:Preliminary results of the project were presented to members of government agencies, academic research groups, and consulting firms at various national and international conferences during the reporting period. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest?1. Peer-reviewed articles in the "Journal of the American Water Resources Association" and "Water". 2. Oral presentations at the 2017 AWRA Spring Specialty Conference on Aquatic SystemConnectivity, the 2017 International SWAT Conference, and the 2017 AWRA Annual Conference. What do you plan to do during the next reporting period to accomplish the goals?Testing SWAT+ in various watersheds Verification of plant growth, water balance, streamflow, sediment, nutrient, and pesticide modules in SWAT+. Testing of SWAT+ will continue and include additional watersheds as well as water quality variables (sediment, nutrients, and pesticides). Identification of errors in the model subroutines and implementation of modifications to the SWAT+ code. Continued testing of the model is expected to reveal additional errors, which will be addressed as needed. Dissemination of SWAT+ to the user community. SWAT+ is expected to be officially released at the 2018 International SWAT Conference in Brussels, Belgium. Integration of landscape units in SWAT+ Identification of appropriate algorithms to describe the landscape and water body connectivity and implementation in SWAT+. The algorithms currently implemented in SWAT+ for describing the landscape and water body connectivity will be refined and improved. Analysis of the impact of implementing landscape units in SWAT+ on model predictions. The comparison of SWAT+ models with and without landscape units will continue and include additional watersheds. Development of national SWAT models Collection of required input data, e.g. topography, soils, land use and management, and climate. Input data will be refined if necessary. Delineation of upland and floodplain areas for all watersheds in the conterminous United States. For now, the national SWAT+ models will use a simpler setup without landscape units, so the delienation of upland and floodplains is not a top priority at this point. However, a collaboration with researchers from Colorado State University has been started to improve existing tools for floodplain delineation with the ultimate goal to approach and complete this task. Setup of SWAT+ models for all HUC8 (8-digit Hydrologic Unit Code) watersheds in the United States with HUC12 (12-digit Hydrologic Unit Code) watersheds as subbasins and calibration of the models against all available hard and soft data. Once debugging of the firstHUC8 (8-digit Hydrologic Unit Code) watershed SWAT+ models has been completed, models will be constructed for the remaining watersheds. Development of a website with county-level maps displaying model output that are updated daily and can help stakeholders make management decisions. Development of the website will continue as model output that is supposed to be displayed on the website becomes available. Improved modeling framework for assessing nutrient transport in tile-drained watersheds Development and testing of a process-based watershed model that explicitly simulates water flow and nutrient transport in subsurface tile drains. Development of the linked SWAT-MODFLOW model will continue and the models for simulating flow and nutrient transport in tile drains will be included in the modeling framework. Improvement of our understanding of the hydrological and biogeochemical processes affecting the transport, storage, and transformation of nitrogen and phosphorus species in tile-drained systems. Output from the enhanced modeling framework is expected to become available by the end of the next reporting period and to improve our understanding of hydrological and biogeochemical processes in tile-drained systems. Assessment of viable conservation practices that maintain ecosystem services in the watershed. This task will be approached after the end of the next reporting period. Dissemination of the developed modeling framework to the research community. This task will be approached after the end of the next reporting period.

Impacts
What was accomplished under these goals? The management of watersheds is a very complex task anddue to their ability to quickly and efficiently compare different management strategies computer simulation models are frequently used to investigate their impact. The Soil and Water Assessment Tool (SWAT) is a computer simulation model that is used all over the world to predict the movement of water, sediment, and agricultural chemicals from the landscape to water bodies and through the stream network of a watershed. SWAT can be used to assess the impact of changes in land use or climate on water resources and to estimate the effectiveness of conservation practices in agriculture. The model is constantly evolving in response to advancements in scientific knowledge and the needs of the user community. Recently, major modifications were made to the model that allow for a more realistic spatial representation of the connectivity of different elements within a watershed, e.g. upland areas, ponds, reservoirs, floodplains, and streams. The resulting new version of the model, SWAT+,isbeing tested and verified in great detail in the context of this project. First applications of the model in the Little River Experimental Watershed in Georgia have demonstrated that the enhanced capabilities of SWAT+ allow it to provide more robust predictions of hydrologic and water quality variables. This is of great importance, as the improved hydrologic modeling tools and information generated by applying thesecan be used by government agencies, watershed management teams, researchers, and stakeholders inthe United States and worldwide to promote vigorous agriculture while at the same time preserving or restoring environmental health. SWAT can aid the development of more efficient management strategies and policies, thus leading to a decrease in money spent on inefficient use of resources (e.g., water, energy, seeds, fertilizer, and pesticides) or misplaced conservation practices. The overall goal of this project is to contribute to the ongoing development of SWAT+ (Soil and Water Assessment Tool) in response to advancements in scientific knowledge and the needs of the SWAT user community. This includes (1) testing SWAT+ extensively with regard to water quantity and quality (sediment, nutrients, and pesticides) in various watersheds, (2) the development and extensive testing of an approach to integrate landscape units in SWAT+, (3) contributing to the development of national SWAT+ models, and (4) developing an improved modeling framework for assessing nutrient transport in tile-drained watersheds. Testing SWAT+ in various watersheds Verification of plant growth, water balance, streamflow, sediment, nutrient, and pesticide modules in SWAT+. The SWAT+ simulations of water balance and streamflow have been analyzed and verified in great detail in the Little River Experimental Watershed in Georgia. Identification of errors in the model subroutines and implementation of modifications to the SWAT+ code. Through carefully analyzing SWAT+ model output for the Little River Experimental Watershed in Georgia, several errors and shortcomings in the model subroutines have been identified and modifications of the SWAT+ code have been made accordingly. Dissemination of SWAT+ to the user community. Nothing to report. Integration of landscape units in SWAT+ Identification of appropriate algorithms to describe the landscape and water body connectivity and implementation in SWAT+. A first, simple approach to representing landscape and water body connectivity has been implementedin SWAT+ and testing of this model setup has started in the Little River Experimental Watershed in Georgia. Analysis of the impact of implementing landscape units in SWAT+ on model predictions. Two different SWAT+ models have been set up for the Little River Experimental Watershed, one without and one with landscape units. Comparison of output from both models is currently in progress. Development of national SWAT models Collection of required input data, e.g. topography, soils, land use and management, and climate. The collection of the input data required for construction the national SWAT+ models has been completed. Delineation of upland and floodplain areas for all watersheds in the conterminous United States. Nothing to report. Setup of SWAT+ models for all HUC8 (8-digit Hydrologic Unit Code) watersheds in the United States with HUC12 (12-digit Hydrologic Unit Code) watersheds as subbasins and calibration of the models against all available hard and soft data. SWAT+ models have been set up for selectedHUC8 (8-digit Hydrologic Unit Code) watersheds and debugging is in progress. Development of a website with county-level maps displaying model output that are updated daily and can help stakeholders make management decisions. A first, simple version of the website has been developed. Work is still in progress. Improved modeling framework for assessing nutrient transport in tile-drained watersheds Development and testing of a process-based watershed model that explicitly simulates water flow and nutrient transport in subsurface tile drains. SWAT+ has been linked to MODFLOW and testing of the linked models is currently in progress. Improvement of our understanding of the hydrological and biogeochemical processes affecting the transport, storage, and transformation of nitrogen and phosphorus species in tile-drained systems. Nothing to report. Assessment of viable conservation practices that maintain ecosystem services in the watershed. Nothing to report. Dissemination of the developed modeling framework to the research community. Nothing to report.

Publications

• Type: Journal Articles Status: Published Year Published: 2017 Citation: Pfannerstill, M., Bieger, K., Guse, B., Bosch, D.D., Fohrer, N., and Arnold, J.G. (2017). How to Constrain Multi-Objective Calibrations of the SWAT Model Using Water Balance Components, Journal of the American Water Resources Association (JAWRA), 53(3): 532546.
• Type: Journal Articles Status: Published Year Published: 2017 Citation: Her, Y., Jeong, J., Bieger, K., Rathjens, H., Arnold, J., and Srinivasan, R. (2017). Implications of Conceptual Channel Representation on SWAT Streamflow and Sediment Modeling, Journal of the American Water Resources Association (JAWRA), 53(4): 725747.
• Type: Journal Articles Status: Published Year Published: 2017 Citation: White, M., Haglund, L., Gloe, M., Bieger, K., Namphong, B., Gambone, M., Hardy, E., Gao, J., Yen, H., and Arnold, J. (2017). Distribution of Selected Soil and Water Conservation Practices in the U.S. as Identified with Google Earth, Journal of the American Water Resources Association (JAWRA), 53(5): 12291240.
• Type: Journal Articles Status: Published Year Published: 2017 Citation: White, M.J., Bieger, K., Gambone, M., Haney, E., Arnold, J., and Gao, J. (2017). Development of a Hydrologic Connectivity Dataset for SWAT Assessments in the US, Water, 9, 892: 10 pages.
• Type: Journal Articles Status: Published Year Published: 2017 Citation: Merwade, V., Baffaut, C., Bieger, K., Boithias, L., Rathjens, H. (2017). Featured Series Conclusion: SWAT Applications for Emerging Hydrologic and Water Quality Challenges, Journal of the American Water Resources Association (JAWRA), 53(6): 13901392.
• Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Bieger, K., White, M.J., and Arnold, J.G. (2017). Representing landscape and aquatic system connectivity in a national watershed model, 2017 AWRA Spring Specialty Conference Aquatic System Connectivity, May 1-3, 2017, Snowbird, UT, USA.
• Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Bieger, K., White, M.J., and Arnold, J.G. (2017). Watershed Configuration in SWAT+, 2017 International SWAT Conference, June 28-30, 2017, Warsaw, Poland.
• Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Bieger, K., Bailey, R.T., Park, S., and Arnold, J.G. (2017). Development of an improved modeling framework for simulating nutrient transport in tile-drained watersheds, 2017 AWRA Annual Conference, November 6-9, 2017, Portland, OR, USA.