AN AGRO-ECOSYSTEM MODEL TO ACHIEVE AGRICULTURAL SUSTAINABILITY IN FOR THE DELMARVA

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: EVANS-ALLEN
• Reporting Frequency: Annual
• Accession No.: 1014941
• Project Start Date: Dec 29, 2017
• Project End Date: May 14, 2019
• Program Code: [(N/A)]- (N/A)

Recipient Organization
University of Maryland Eastern Shore
11868 College Backborne Road
Princess Anne,MD 21853

Performing Department
Agriculture

Non Technical Summary
The Delmarva Peninsula, home to some of the most productive agriculture in the nation, faces numerous environmental challenges that may jeopardize both its productivity and its economic viability. Climate change has the potential to impact food system productivity via a wide variety of mechanisms including direct effects on plants as a result of temperature, water, or CO2 changes and indirect effects by changing weed, insect, and pathogen pressure (Easton et al. 2014). Climate change can also directly impact animal productivity by restricting feed supplies, particularly for grazing systems (Walthall et al. 2013), and increasing potential for thermal stress which reduces feed intake and productivity (White et , al. 2014). Adaptation and mitigation strategies must be developed to feed the growing global population while coping with climate change and land availability. Delmarva agriculture is further constrained by the Chesapeake Bay Total Maximum Daily Load (TMDL) regulations which call for reductions in nitrogen, phosphorus and sediment loadings on the order of 30%, and which may considerably constrain production in order to comply. Our proposed research focuses on identification of promising management practices farmers can employ to adapt to climate change and improve water quality. Such adaptations might include fine-tuning practices or making major management changes, such as the timing of planting or harvest, double cropping, shifting to new crops/varieties, making greater use of stover as feed, increasing genetic selection pressure, or increasing the use of byproducts of human food processing.Farmers and other stakeholders need recommendations for how to adapt agricultural production systems to environmental and socio-economic change. Evaluating adaptation mechanisms requires tools to help us understand the consequences of natural and human-induced changes in environmental drivers and to help us anticipate human adaptations and predict responses to those adaptations. Eco-hydrological and agro-ecosystem models are common tools for these purposes. Human decisions about land-use and agricultural practices are made in the context of a changing climate and its other influences on ecosystems. Thus, anticipating change requires new approaches to examine how human responses to climate can amplify or dampen the impacts of climate and landuse change, and to evaluate effectiveness of policies aimed at mitigating impacts. We propose evaluating adaption mechanisms in the Manokin watershed in MD.The proposed work will 1) Develop an agro-ecosystem model comprised of linked crop, hydrology, livestock/poultry, and economic models for assessing the economic and environmental effects of adaptation practices; 2) Apply the agro-ecosystem model to Delmarva crop and livestock production in the Manokin watershed; and 3) Propose/identify adaptation strategies to increase resiliency of Delmarva agroecosystems while improving water quality.We will incorporate methods we are currently using in other watersheds (Easton et al. 2014; Bosch et al., 2016a) with the major adaptation being inclusion of livestock/poultry models and adjustment of the economic, hydrology, and crop models to account for livestock/poultry model outputs. Results from application of the agro-ecosystem model to Manokin watershed will demonstrate how adaptation practices and farm enterprise selection should be adjusted in response to the challenges of climate and socio-economic change. Identifying these strategies may help to improve resiliency of Delmarva agriculture and ensure that it remains competitive for the foreseeable future.

Animal Health Component

80%

Research Effort Categories

Basic

10%

Applied

80%

Developmental

10%

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
132	0320	2050	30%
601	0430	3010	30%
307	3299	3100	30%
112	0320	3100	10%

Knowledge Area
601 - Economics of Agricultural Production and Farm Management; 112 - Watershed Protection and Management; 132 - Weather and Climate; 307 - Animal Management Systems;

Subject Of Investigation
3299 - Poultry, general/other; 0320 - Watersheds; 0430 - Climate;

Field Of Science
2050 - Hydrology; 3010 - Economics; 3100 - Management;

Keywords

climate change

delmarva peninsula

agro-ecosystem model

manokin watershed

adaptation strategies

Goals / Objectives
The proposed work directly addresses three RFA focus areas: 1) Best management practices surrounding the environmental impacts of Delmarva agriculture; 2) Sociopolitical issues impacting the viability and profitability of Delmarva agriculture; and 3) Impacts of climate change and sea level rise on Delmarva agriculture, and could contribute significantly to evaluating alternative crop and animal agriculture enterprises. The proposed work will:1) Develop an agro-ecosystem model comprised of linked crop, hydrology, livestock/poultry, and economic models for assessing the economic and environmental impacts of climate change adaption and water quality improvement strategies on Delmarva agriculture.2) Apply the agro-ecosystem model in the Manokin watershed in the Delmarva peninsula.3) Propose/identify adaptation strategies to increase resiliency of Delmarva agroecosystems while simultaneously improving water quality.

Project Methods
Obj 1: Develop linked agroecosystem model: The major adaptation is including a livestock/poultry model, which has implications for crop, hydrology, and economic model development. The first step will be to decide on the types of management practices included within the livestock/poultry models, how the crop, hydrology, and economic models will be modified to accommodate these changes, and how data will be handled and shared among models. Adding the livestock/poultry model introduces complexity in that some livestock/poultry outputs such as manure production and utilization will be passed to the SWAT-VSA model and the resulting model outputs (predicted crop yields and nutrient loadings) will then be passed to the economic model. Other livestock/poultry outputs (such meat, or egg production) will be passed directly to the economic model, which will be maximizing net revenues subject to appropriate resource constraints including land type and quality, livestock facilities, crop rotations, animal and crop nutrient requirements, and limits on nutrient loadings to surface waters (Bosch et al., 2016a). Physical input parameters such as crop and livestock/poultry yields and estimated nutrient loadings will be provided by SWAT-VSA and livestock/poultry models. Per unit costs for crop and livestock/poultry enterprises will be based on enterprise budgets supplemented with expert opinion (Virginia Cooperative Extension, 2011; University of Maryland Extension, 2016). Delmarva crop and livestock prices will be obtained from NASS (https://www.nass.usda.gov/).Obj 2: The Manokin watershed will be our application region, because it represents the dominant soils (e.g., sandy loam), and production systems (e.g., cropland, poultry, dairy, forages, pasture) of the Delmarva and contains enough measured data to suitably evaluate our predictions. A headwater subbasin of the Manokin discharges to a gage (USGS 01486000 Manokin Branch near Princess Anne, MD) where several years of discharge data have been recorded daily, will be the focus for applying the agro-ecosystem model. We will represent current climate conditions with a 20-year time series from a Delmarva weather station covering recent weather patterns. Future climate data will be obtained using simulations from the Coupled Model Intercomparison Project 5 (CMIP) dataset.Livestock modules will be adapted to simulate animal agriculture in the Manokin sub-watershed. Livestock numbers, physical locations, and available management practice data will serve as inputs. Least-cost rations will be formulated for all livestock populations and used to define dietary inputs. The modeled nutrient outputs from animal agriculture will be outputted back to SWAT-VSA to serve as informed estimates of N, P and other nutrient excretions included in Delmarva-relevant field management practices for current and future time periods.Once verified, this baseline scenario will define the current conditions against which the effects of climate change will be evaluated. The model will then be forced with a series of scenarios to predict food system productivity, crop shifts, and economic and environmental consequences for two future time periods, 2041-2065 and 2075-2099. Using SWAT-VSA we will evaluate predicted crop yields and nutrient loadings results from at least three climate projection models for a 20-80 year future period in order to determine the sensitivity of our results to variations in climate predictions.Obj 3 We will generate a series of recommended adaptation strategies for Delmarva agriculture. The set of characteristics or practices that define at risk (or resilient) Delmarva agricultural systems will be identified from the results in Obj 2. The datasets generated by Obj 2 will help to identify characteristics of Delmarva agriculture that are at high (or low) economic or productive risk under climate change. After evaluating potential risk (or resilient) areas, a comprehensive set of potential landuse and management practices will be proposed. The range of outputs for each input scenario will be cataloged and analyzed to identify input scenarios with above average performance relative to environmental impact, profitability, and food production goals. The characteristics of these input scenarios will be used to identify optimal adaptations to climate stress.

Progress 12/29/17 to 05/14/19

Outputs
Target Audience:Our target audience for this SEED grant included 1) extension educators, 2) industry service providers including fertilizer, feed, and seed dealers, and 3) policy makers. Our primary target audience was extension educators who were directly involved in the project. Changes/Problems:The model evaluation and scenario development required greater amounts of time and effort because of coding issues. However, the revisiting of the code provided the opportunity to improve model documentation. What opportunities for training and professional development has the project provided?AAEC Ph.D. Graduate Research Assistant, Yuelu Xu's work on a nearby watershed inPennsylvania contributes to this research. AAEC Ph.D. Graduate Research Assistant, Nasim Ebadi's work on a nearby watershed inPennsylvania contributes to this research. APSC Ph.D. Graduate Research Assistant, Douglas Liebe's work modeling livestock productionsystems contributes to this research. UMES undergraduates were employed and trained to work on GIS spatial analysis, watersampling and analysis, and data organization contributing to this research How have the results been disseminated to communities of interest?Darrell Bosch, Amy Collick, Robin White, and Zach Easton. 2017. An Agro-ecosystem Model toAchieve Agricultural Sustainability for the Delmarva (Part I). Lightning Presentation, 2017Delmarva Cooperative Seed Grant Program Summit. University of Maryland Eastern Shore.Princess Anne, MD. December 13. Yuelu Xu and Darrell J. Bosch. Costs of Nitrogen Loading Reductions by Macro-farm TargetingMethods. Paper presented at the Southern Agricultural Economics Association Annual Meeting,Birmingham, Alabama, February 3-6, 2019. Nasim Ebadi, Darrell J. Bosch, and Robin White. Trade-offs Among Increasing Farm NetReturns and Reducing Emissions of Nitrogen, Phosphorus, Greenhouse Gas Equivalents, andAmmonia in a Dairy Farm. Paper presented at the Southern Agricultural Economics AssociationAnnual Meeting, Birmingham, Alabama, February 3-6, 2019. Amy S. Collick, Darrell Bosch, Yuelu Xu, Moges B. Wagena, Doug M. Liebe, Rooin White, andZach M. Easton. An Agro-ecosystem Model to Achieve Agricultural Sustainability for theDelmarva (Part II). Presentation, 2019. Delmarva Cooperative Seed Grant Program Summit. University of Delaware. Newark, DE. March 7, 2019. Doug M. Liebe, Moges B. Wagena, Zach M. Easton, Robin R. White. 2019. Modeling BroilerProduction in the Chesapeake Bay Area: Effects of Management Decisions on N and P Outputs.Proceedings of Advances in Animal Biosciences. Annual Meeting of the Workshop onModelling Nutrient Digestion and Utilization in Farm Animals. September 14th to 17th, 2019.Itamambuca, Brazil. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Field monitoring data has been collected, compiled and organized to help improve modelingparameterization in the difficult terrain (very flat) of the Manokin. Using long-term data fromresearch (UMES Research and Teaching Farm) and commercial farms (some of whom we havecollaborated with for over five years), we are developing monitoring datasets that includegroundwater, ditch and stream monitoring (flow and water quality), soil characteristics, fieldmanagement, and subsurface imaging. A SWAT model initialization of the Manokin watershed has been developed using GIS spatiallayers: digital elevation models (DEMs), land use including field delineations, andtopographically derived soils. Farm management has been investigated in order to developcommon crop rotations for baseline (current) model runs. Scenarios can then be incorporatedinto this model initialization for future and climate change scenarios. A flexible framework for modeling livestock production was constructed. The framework isdesigned to allow real-time integration into the SWAT model for improved representation ofhow livestock feeding and management influences excretion of N and P. Detailed equations havebeen developed for poultry and dairy initializations to allow for subsequent simulation. A farm economic model was developed using GAMS (2019) to evaluate crop and poultryopportunities for DelMarva. The farm objective was assumed to be maximization of farm netreturns. The following agricultural enterprises were evaluated: corn grain, soybeans (full-seasonand double-cropped), wheat, and grass hay. Constraints include total land area in the watershedthat is suitable for crop production, 1,564 ha. Net revenues are estimated for each field modeled in SWAT where net revenue equals yieldtimes crop price minus costs. Costs include fertilizer cost plus machinery operating andownership costs, seed, agricultural pesticides, drying costs, crop insurance, and operating interest. Gross revenues, costs, and net revenues are summed for the farm. Prices, costs andyields are taken from USDA (U.S.D.A. Quickstats, 2019) and extension budgets for Maryland(corn grain, soybeans, double cropped soybeans and wheat) and Virginia (hay) (University ofMaryland Cooperative Extension, 2019; Virginia Cooperative Extension 2007, 2014). Cropyields and fertilizer applications are obtained from the SWAT model. We are continuing to test the model to determine its ability to differentiate livestock nutrientmanagement scenarios. Livestock nutrient management corresponds to low nitrogen feeding andphytase amendments to reduce phosphorus additions to the diet. Results from running thelivestock and SWAT models separately will be compared with results from running theintegrated SWAT-livestock model. In summary, the framework comprised of linked poultry, crop/hydrology, and economic models provides away to assess physical and economic implications of alternative strategies to reduce nutrientloadings from agriculture. The models can be run under alternative climate scenarios in order toassess the costs and effectiveness of such practices under climate change. Insights gained fromthese models can be used to assist Delmarva stakeholders in adapting to changing demandsimposed by evolving markets and environmental policies. Work is currently underway to use theframework to evaluate alternative poultry feeding strategies including phytase amendments andlow N feeding. These results will contribute to improved economic sustainability of Delmarvaagriculture.

Publications

• Type: Conference Papers and Presentations Status: Accepted Year Published: 2019 Citation: Yuelu Xu and Darrell J. Bosch. Costs of Nitrogen Loading Reductions by Macro-farm Targeting Methods. Paper presented at the Southern Agricultural Economics Association Annual Meeting, Birmingham, Alabama, February 3-6, 2019.
• Type: Conference Papers and Presentations Status: Accepted Year Published: 2019 Citation: Nasim Ebadi, Darrell J. Bosch, and Robin White. Trade-offs Among Increasing Farm Net Returns and Reducing Emissions of Nitrogen, Phosphorus, Greenhouse Gas Equivalents, and Ammonia in a Dairy Farm. Paper presented at the Southern Agricultural Economics Association Annual Meeting, Birmingham, Alabama, February 3-6, 2019.
• Type: Conference Papers and Presentations Status: Accepted Year Published: 2019 Citation: Amy S. Collick, Darrell Bosch, Yuelu Xu, Moges B. Wagena, Doug M. Liebe, Rooin White, and Zach M. Easton. An Agro-ecosystem Model to Achieve Agricultural Sustainability for the Delmarva (Part II). Presentation, 2019. Delmarva Cooperative Seed Grant Program Summit.University of Delaware. Newark, DE. March 7, 2019.
• Type: Conference Papers and Presentations Status: Accepted Year Published: 2019 Citation: Doug M. Liebe, Moges B. Wagena, Zach M. Easton, Robin R. White. 2019. Modeling Broiler Production in the Chesapeake Bay Area: Effects of Management Decisions on N and P Outputs. Proceedings of Advances in Animal Biosciences. Annual Meeting of the Workshop on Modelling Nutrient Digestion and Utilization in Farm Animals. September 14th to 17th, 2019. Itamambuca, Brazil.
• Type: Conference Papers and Presentations Status: Accepted Year Published: 2018 Citation: Collick, A.S., D.R. Fuka, T.L. Veith, A. Buda, A. Allen, , R. Bryant, and Z.M. Easton. 2018. Employing fine resolution spatial information and extensive field research to evaluate best management practice (BMP) scenario evaluations across the Chesapeake Bay. 2018 Chesapeake Community Research & Modeling Symposium. Annapolis, MD. June 12  14, 2018.
• Type: Conference Papers and Presentations Status: Accepted Year Published: 2018 Citation: Liebe, D.M. and White, R.R. 2018. A dynamic modeling framework to compare culling strategies and their effect on N excretion in dairy herds. Proceedings of the 2018 Meeting of the Animal Science Modelling Group. Knoxville, TN. June 24th to June 27th, 2018.
• Type: Other Status: Accepted Year Published: 2019 Citation: USDA-NIFA Capacity Building Grant (CBG) submitted by Amy Collick (UMES), Zachary Easton (Virginia Tech), and Ray Bryant (USDA-ARS). UMES Drain Water Management Research Facility: Investigating drainage from high legacy phosphorus (P) soils and evaluating P-reduction innovations for poultry house system drainage. $499,952.
• Type: Conference Papers and Presentations Status: Accepted Year Published: 2017 Citation: Collick, A.S., Z.M. Easton, P.J.A. Kleinman, D. Auerbach, B. Buchanan,D.R. Fuka. 2017. Predicting Phosphorus Dynamics Across Physiographic Regions Using a Mixed Hortonian Non-Hortonian Hydrology Model. Annual Meeting of American Geophysical Union (AGU). New Orleans, LA. December 11  15, 2018.

Progress 12/29/17 to 09/30/18

Outputs
Target Audience:We proposed a target audience of our work to involve1) extension educators, 2) industry service providers including fertilizer, feed, and seed dealers, and3) policy makers. Our primary target audience currently has been extension educators who are directly involved in the project. The outreach of this project depends on the effective linkage of the agroecosystem-livestock-economic model and employment of the model. Much of our effort has been to finalize the model to then run scenarios and recommend adaptation options for the other audience. Changes/Problems:The model integration has required more time than anticipated due to challenges in combining model programming code. What opportunities for training and professional development has the project provided?Virginia Tech AAEC Ph.D. Graduate Research Assistant, Yuelu Xu's work on a nearby watershed in Pennsylvania contributes to this research. Virginia Tech APSC Ph.D. Graduate Research Assistant, Douglas Liebe's work modeling livestock production systems contributes to this research. UMES School of Agriculture and Natural Resource Graduate Research Assistant,Jhamyllia Rice"s workon GIS land use delineation and analysis contributes to the Manokin Watershed modeling being conducted for this research. How have the results been disseminated to communities of interest?While the model continues to be developed and corroboratedand scenarios run, dissemination will be limited to those directly involved in the work. The last several months of the project will be committed to dissemination of the model information and scenario results. The first year results were summarized for the Delmarva SEED grant annual meeting. Darrell Bosch, Amy Collick, Robin White, and Zach Easton. 2017. An Agro-ecosystem Model to Achieve Agricultural Sustainability for the Delmarva. Lightning Presentation, 2017 Delmarva Cooperative Seed Grant Program Summit. University of Maryland Eastern Shore. Princess Anne, MD. December 13. What do you plan to do during the next reporting period to accomplish the goals?The model integration will continue and scenarios will be run using historical climate and forecasted climate. The results will demonstrate how adaptation practices and farm enterprise selection may be adjusted in response to the challenges of climate and socioeconomic change. Furthermore, we will continue to identify a suite of adaptive agricultural practices and strategies to increase resiliency of Delmarvaagroecosystems. March 7, 2019 marks another annual Delmarva SEED meeting at University of Delaware in Newark, DE. A full presentation instead of a lightening presentation. An Agro-ecosystem Model to Achieve Agricultural Sustainability for the Delmarva Bosch, D., M.B. Wagena, D. Liebe, R. White, and Z.M. Easton, Virginia Tech, Blacksburg, VA Collick, A.S.*, University of Maryland Eastern Shore, Princess Anne, MD 21853 *Author presenting at annual Delmarva Seed Grant meeting Results from the application of the agro-ecosystem model to Manokin watershed demonstrate how adaptation practices and farm enterprise selection should be adjusted in response to the challenges of climate and socio-economic change. We will continue to identify adaptation strategies to increase resiliency of Delmarva agroecosystems and ensure that it remains competitive while improving water quality into the future. Dissemination of the suite of adaptive agricultural strategies will be disseminated to our target audience (extension, farm planning and agribusiness, andpolicy makers) described in the project initiation.

Impacts
What was accomplished under these goals? Field monitoring data has been collected, compiled and organized to help improve modeling parameterization in the difficult terrain (very flat) of the Manokin. Using long-term data from research (UMES Research and Teaching Farm) and commercial farms (some of whom we have collaborated with for over five years), we are developing monitoring datasets that include groundwater, ditch and stream monitoring (flow and water quality), soil characteristics, field management, and subsurface imaging. A SWAT model initialization of the Manokin watershed has been developed using GIS spatial layers: digital elevation models (DEMs), land use including field delineations, and topographically derived soils. Farm management has been investigated in order to develop common crop rotations for baseline (current) model runs. Scenarios can then be incorporated into this model initialization for future and climate change scenarios. A flexible framework for modeling livestock production was constructed. The framework is designed to allow real-time integration into the SWAT model for improved representation of how livestock feeding and management influences excretion of N and P. Detailed equations have been developed for poultry and dairy initializations to allow for subsequent simulation. A farm economic model was developed using linear programming in GAMS (2017) to evaluate crop and poultry opportunities for DelMarva. The farm objective was assumed to be maximization of farm net returns. The following agricultural enterprises were evaluated: corn grain (irrigated and unirrigated), soybeans (full-season and double-cropped), wheat, switchgrass for biofuel, alfalfa hay, tomatoes, and broiler production. Constraints include total land area in the watershed that is suitable for crop production, 3,700 ha. To reflect marketing constraints, tomato production is limited to no more than 10 percent of land area. The watershed has capacity for 30 broiler houses. Prices, costs and yields are taken from USDA (Quickstats, 2017) and extension budgets for Maryland (corn grain, soybeans, and double cropped soybeans and wheat) and Virginia (alfalfa and tomato) (University of Maryland Cooperative Extension, 2017; Virginia Cooperative Extension 2007, 2014). Broiler production is 242,000 birds/house/year, which is contracted with a poultry integrator. Gross revenue per house is $70,674 not including revenues from broiler litter sales. Variable costs per house equal $16,820, and net returns to operator labor, management, and capital equal $53,854/year (Rhodes et al., 2011). Total farm net returns to land, risk, and management are $11,313,000 or an average of $3,058/ha ($1,238/ac). Included enterprises are broilers (30 houses), unirrigated corn grain (1,657 ha), wheat (1,622 ha), soybean (1,622 ha), alfalfa (50 ha), and tomato (370 ha). Major factors behind the high return are broiler and tomato production. When the tomato enterprise was eliminated, total returns declined to $5,240,000 ($1,416/ha, $573/ac). When broilers were eliminated as well, total returns declined to $2,934,000 ($792/ha, $321/ac).

Publications

• Type: Conference Papers and Presentations Status: Accepted Year Published: 2018 Citation: Collick, A.S., D.R. Fuka, T.L. Veith, A. Buda, A. Allen, , R. Bryant, and Z.M. Easton. 2018. Employing fine resolution spatial information and extensive field research to evaluate best management practice (BMP) scenario evaluations across the Chesapeake Bay. 2018 Chesapeake Community Research & Modeling Symposium. Annapolis, MD. June 12  14, 2018.
• Type: Conference Papers and Presentations Status: Accepted Year Published: 2017 Citation: Collick, A.S., Z.M. Easton, P.J.A. Kleinman, D. Auerbach, B. Buchanan, D.R. Fuka. 2017. Predicting Phosphorus Dynamics Across Physiographic Regions Using a Mixed Hortonian Non-Hortonian Hydrology Model. Annual Meeting of American Geophysical Union (AGU). New Orleans, LA. December 11  15, 2018.
• Type: Conference Papers and Presentations Status: Published Year Published: 2018 Citation: Liebe, D.M. and White, R.R. 2018. A dynamic modeling framework to compare culling strategies and their effect on N excretion in dairy herds. Proceedings of the 2018 Meeting of the Animal Science Modelling Group. Knoxville, TN. June 24th to June 27th, 2018.
• Type: Other Status: Submitted Year Published: 2018 Citation: USDA-NIFA Capacity Building Grant (CBG) submitted by Amy Collick (UMES), Zachary Easton (Virginia Tech), and Ray Bryant (USDA-ARS). UMES Drain Water Management Research Facility: Investigating drainage from high legacy phosphorus (P) soils and evaluating P-reduction innovations for poultry house system drainage. $499,952.