Progress 09/01/19 to 08/31/22

Outputs
Target Audience:The team has targeted several audiences to help educate and advocate for management of alfalfa to enhance its environmental sustainability as a crop, in crop rotation systems, and in related dairy production systems. Farmers, crop consultants, and dairy nutritionists are the primary audience in disseminating results, as farmers and their advisers ultimately make decisions on the farm regarding what crop rotations and production methods are used. Utilizing Extension outreach channels, general findings and potential management implication are being summarized for distribution in print and online formats. University of Minnesota Extension is serving as a primary distribution channel, utilizing the Extension website, Crop News Blogs, news releases, and listservs to distribute study results and recommendations. Results are also being summarized in print format to be distributed at field days and workshops held throughout the region targeting both crop and livestock producers. We are also creating fact sheets to disseminated to policymakers and others responsible for setting and reaching sustainability targets. Although farmers are responsible for changing production methods, it is equally important for public policy and sustainability goals to be in alignment with best management and recommended production methods. Outreach activities for this audience will primarily be through non-technical summaries and research reports summarizing project results. These same non-technical summaries along with infographics are being used to convey agriculture's efforts at increasing sustainability to the public. The final audience is academics and others interested in the technical details and analysis of this work. Changes/Problems:A challenge that has continued through the project is inability to find additional post-doctoral research help for the project. Hiring for the post-doc position in the grant proposal was made very difficult due to Covid-19 restrictions on USDA hiring and the overall lack of interested candidates. The candidate was envisioned to help both with project activities and, writing and disseminating results. An additional staffing challenge is the loss of our extension agronomist, Dr. Jarod Goplen, during the last few months of the project, who moved to the private sector. Dr. Craig Shaeffer- Professor (University of Minnesota, Dept. of Agronomy) and Extension Forage Specialist, who has brought his decades of alfalfa expertise to the project as an advisor is helping with getting the extension components finalized and disseminated. This reduced staffing has challenged the team to get results converted into written dissemination products as we close out the grant. What opportunities for training and professional development has the project provided?The project has supported a Ph.D. candidate. The student was tasked with GIS mapping of alfalfa production regions using the satellite based cropland data layer (CDL). How have the results been disseminated to communities of interest?While some results have already been disseminated to communities of interest (see target audiences), the information from the project is still being developed into products suitable for outreach. Initial presentations to researchers and alfalfa producers has begun. As more product/content is finalized, the project team will continue working with these groups and others to explain how alfalfa can be an important addition to cropping systems. Finalized documents will appear in research journals, trade journals, and on USDA-NAL and Univ. of Minn websites. Researchers from this project will also be using the information in current and future projects on alfalfa and perennial cropping systems. The major products from this project are or will be publications, both technical and non-technical that describe environmental and economic considerations of alfalfa production in the Midwest. The project team has decided on how to divide project results into content for a number of technical articles for peer-reviewed research journals. At this point, the research has been concluded and the papers are being written. The technical articles anticipated are: An alfalfa LCA paper with in-depth discussion of environmental impacts. A paper that examines LCA, economic, and soil science aspects of crop rotations that include alfalfa. A geospatial article that examines alfalfa production locations, crop rotations, and stand lengths. The soil impacts of alfalfa will be examined in a paper on carbon, organic matter, and nutrients. Non-Technical publications: The alfalfa producer survey results will be written up in an article for a forage trade journal. The production demographics and production regions will be discussed in trade journal. In addition to the technical articles, a larger report on the methods and data is being developed to accompany the LCA data that will be submitted to the USDA National Agriculture Library. It is expected that an initial technical report submission with the LCA modeling data will be updated as papers are published, and the project writings are refined. What do you plan to do during the next reporting period to accomplish the goals?Although this is the final report, we are continuing the process of outreach and dissemination. With the research completed, the project team is focused on getting the technical and non-technical written products ready for publication (see Dissemination section). The status of project writing varies from fairly complete to waiting to be started. The field specific nature of the models means that much of the technical writeup work is limited to the modelers on the team.

Impacts
What was accomplished under these goals? Objective 1. Evaluate regional production practices Data collection for the cropping, economic, and soil science components of the project was conducted early in the project. A primary source of data was the USDA National Agricultural Statistic Service (NASS), using their annual and 5-year census of agriculture or specialty surveys like irrigation research. The project team also initiated a survey that was sent out to Midwestern forage producers to identify regional practices for alfalfa. An additional source of information was the FINBIN farm financial database, which collects Midwest producer data. LCA work used a commercial dataset with US LCI data from NREL and a major US Fluid Milk Production effort. GIS mapping of alfalfa production provided spatial data on the cultivation of alfalfa across the Midwest. This informed the development of archetypal sub-regions of alfalfa crop management based on management and productivity, which fed into the alfalfa modeling efforts. Initial spatial analysis calculations followed a revised methodology based on Matt Yoast et al (2014). Using geospatial software, CDL data from 2011 - 2019 were processed to model chronological sequence of crops grown in a given location. In addition to crop chronology, average stand lengths were calculated by state, county, field, and by sub-region. Table 1 Stand Length State Stand Length ND 4.2 SD 3.8 NE 4.1 MN 3.6 IA 3.3 WI 3.4 Objective 2. Develop scenarios to optimize alfalfa's benefits for cropping and dairy systems The project worked at three levels, with scenarios developed for each; alfalfa as a crop, alfalfa in rotations, and alfalfa in diets. Examples of these scenarios can be seen in regional analysis (table 1), crop rotations (table 2 below) and in the dairy diet tables (not shown). Table 1 Alfalfa Production Regions Region 1 - W. Nebraska. Region 2 - Iowa, S. Minnesota, SE South Dakota, W. Nebraska. Region 3 - Western Dakotas, Region 4 - NE Minnesota and E. North Dakota. Region 5 - Wisconsin, E. Minnesota, and NE Iowa. Region 6 - Minnesota Dairy Region-Average of two dairy regions (5 and 6) . Objective 3. Develop LCA, Soil Carbon/Nutrient Flux, and Economic Models LCA Modeling An LCA was conducted to assess the greenhouse gas, fossil energy, land use, and irrigation water intensity for alfalfa production in the Midwest. The Dairy Region base case greenhouse gas and fossil energy impact for haylage were 137 kg CO2 e. per ton DM and 985 MJ per ton DM at the point of harvest on-field. At the point of feeding stored forage, GHG emissions were 163 kg CO2 e. per ton DM and fossil energy use was 1244 MJ per ton DM in the base case scenario. Fertility inputs, soil processes, and cropping related fuels are the single largest consistent contributors. In the sub-regions analyzed, GHG emissions varied between 104 to 167 kg CO2 e. per ton DM. Irrigation and liming were large contributors in regional GHG variation. Yields also influenced regional variations in GHG emissions. Fossil energy consumption exhibited similar patters GHG impacts. However, soil related N2O and lime application were not important in the fossil energy results. Overall, alfalfa has a relatively low environmental footprint for GHG emissions and fossil energy use compared to other crops. Development of more tolerant varieties that perform well with low inputs or higher production may help to improve the footprint. An additional consideration in alfalfa's overall environmental footprint is factors such as biodiversity and soil health. Both producers and the larger food industry should understand these additional benefits in addition to LCA results as they consider whether alfalfa will fit into their production systems or management. Soil Carbon We used the Environmental Policy Integrated Climate (EPIC) model to quantify changes in soil organic carbon (SOC) in alfalfa production systems. We simulated a baseline "business-as-usual" rotation and management scenarios with different stand lengths and management (see Table 2). The SOC changes were multiplied by a characterization factor (CF) to calculate the climate benefits. SOC change was calculated for all management scenarios and compared the final reference state SOC stock (SOCRef) values in the business-as-usual scenario. The cumulative change in dairy region SOC stock ranged from -1.0 to +0.45 Mt CO2 ha-1 yr-1, with corresponding GWPs ranging from -46 to 1051 kg CO2-e yr-1 (Table 2). The greatest SOC loss was observed in Continuous Silage Corn and the greatest gain in the Heavy Alfalfa rotation. Increasing the proportion of alfalfa and decreasing the proportion of silage corn in the rotation further improved SOC accrual. When increasing the alfalfa stand length from 2 to 4 years, each additional year of alfalfa increased SOC storage by an average of 0.5 Mt CO2 ha-1 yr-1. Finally, when averaged over all rotations, the average change in SOC under alfalfa production was 0.36 ± 0.33 Mt CO2 ha-1 yr-1, corresponding to a net reduction in GWP of -5.7 kg CO2-e Ton-1 yr-1. Soybean, grain corn, and silage corn were net sources of emissions, contributing 5.3, 27.7, and 76.8 kg CO2-e Ton-1 yr-1, respectively. Table 2: Mean annual change in soil organic carbon (ΔSOC)in Mt CO2 ha-1 yr-1 and Global Warming Potential (GWP) in kg CO2-e yr-1 for 100-year EPIC simulations. Rotation ΔSOC GWP100 Continuous silage corn (C-C) -1.05 1051 Corn-soybean grain rotation (C-S) 0.10 -10 Light alfalfa (C-C-C-C-A-A) -0.55 550 Medium alfalfa (C-C-C-A-A-A) -0.04 42 Heavy alfalfa (C-C-A-A-A-A) 0.46 -46 Heavy grass mix (C-C-A-A-O-O) 0.01 -1 High protein rotation (C-S-Can-A-A-A) 0.08 -8 Alfalfa with oat companion (C-C-O/A-A-A-A) 0.39 -39 Alfalfa interseeding (C-C-C/A-A-A-A) 0.27 -27 Table 3: Mean annual change in SOC and GWP for 100-year simulations. Crop ΔSOC GWP100 Alfalfa 0.36 -5.7 Grain Corn 0.06 5.3 Silage Corn -0.17 27.7 Soybeans -0.74 76.8 Economic The overall objective of the economic analysis was to compare the profitability of alfalfa crop rotations (similar to those in Table 2) in the region. The analysis considered differences in corn and alfalfa costs and yields across years in a given rotation, and different seeding methods. The crop yields and prices were based generally on annual estimates by NASS. Two price scenarios are analyzed due to price increase in prices around the beginning of 2022. Scenario 1 is based on NASS averages over 2019-21. Scenario 2 is based on prices for January - April 2022. Three alfalfa seeding methods were considered: direct-seeded with herbicide, seeded with oat companion, and interseeded into corn. The expenses for growing the crops were mainly from FINBIN 2020 or 2021 crop enterprise reports and rented land, with machinery costs from previous work by Investigator Lazarus. Hybrid rye had the highest net return of the individual crops under the 2019-21 prices (other than mature stand alfalfa, which would need to be averaged with an establishment year for a fair comparison) while corn silage had the highest return under the 2022 prices. Net returns are negative for the alfalfa seeding years including the corn interseeded with alfalfa. The CRAAA rotation was most profitable of the rotations using the prices from 2019-21 while the CS rotation was most profitable based on the early 2022 prices. Under 2019-21 prices, the corn interseeded with alfalfa was more profitable than the other non-rye alfalfa rotations. This is mainly due to the fact that the first year of alfalfa had a mature yield rather than a lower yield as with the other alfalfa rotations. The CCCCAAA rotation is least profitable under the 2019-21 prices while the heavy grass rotation was least profitable with early 2022 prices Objective 4. Evaluate and disseminate results As can be seen from objective 3, we are well along with evaluating results. The outreach and dissemination activities and progress are more fully described below.

Publications

• Type: Conference Papers and Presentations Status: Other Year Published: 2022 Citation: Tallaksen. J. 2022. Life Cycle Assessment Research and Resources at the University of Minnesota, USDA Life Cycle Assessment Workshop, June 27-28, 2022- FTallaksen. J. 2022. Life Cycle Assessment Research and Resources at the University of Minnesota, USDA Life Cycle Assessment Workshop, June 27-28, 2022- Ft. Collins, CO t. Collins, CO
• Type: Conference Papers and Presentations Status: Other Year Published: 2022 Citation: Joel Tallaksen, John Grabber, Josh Gamble, Jared Goplen, William Lazarus, Bradley Heins, Craig Sheaffer, and Robin Sehler. 2022. Improving Alfalfa-Based Livestock Forage Production Systems Using Life Cycle Analysis, North American Alfalfa Improvement Conference, June 8, 2022- Kalamazoo, MI
• Type: Other Status: Other Year Published: 2022 Citation: Joel Tallaksen, John Grabber, Josh Gamble, Jared Goplen, William Lazarus, Bradley Heins, Craig Sheaffer, and Robin Sehler. 2022. Improving Alfalfa-Based Livestock Forage Production Systems Using Life Cycle Analysis, Gardner Seminar Series: USDA-ARS-NCSCRL (Morris) and U of MN, West Central Research and Outreach Center, October 6, 2022 Morris, MN
• Type: Other Status: Other Year Published: 2022 Citation: John Grabber, Joel Tallaksen, Josh Gamble, Jared Goplen, William Lazarus, Bradley Heins, Craig Sheaffer, and Robin Sehler. 2022. Improving Alfalfa Livestock Production Systems Using Life Cycle Analysis. Midwest Forage Association- Symposium 2022, Feb 21-23, 2022, Chula Vista, WI and Online

Progress 09/01/20 to 08/31/21

Outputs
Target Audience:During this reporting period, we have described our project in a mass e-mail as part of outreach for an alfalfa production survey to farmers and consultants. Changes/Problems:The first object of identifying regional practices has been more difficult than imagined due to the many variations of alfalfa production across the Midwest. Depending on the climate, presence of dairy, soil types, stand length, cutting frequency, chemical inputs, and yields, the production inputs and outputs vary considerably. These differences are also compounded by which data is being analyzed, work at research farms vs typical producers (via survey). Therefore, we have taken much longer than expected in research and discussion on how to address the variations. The geographical distribution and background of the research team assembled for this project was an advantage in recognizing these variations. We had not initially intended to survey producers, but realized that they could provide critical information in understanding and accounting for these variations. Based on their feedback, we have included more scenarios to consider issues such as companion crops. An ongoing issue with the project has been the hiring restrictions due to COVID-19. Our initial plan of hiring a post-doc proved to be difficult due to USDA-ARS policy on hiring foreign nationals during the COVID-19. Ultimately, we were not able to hire a post-doc due to a lack of available candidates. We therefore have extended the graduate student's appointment and increased the percentage of staff time to complete the work we had intended for the post-doc. What opportunities for training and professional development has the project provided?The project is funding the work of a graduate student in the soil science modeling aspects of the work. The student is using geographical information systems (GIS) to identify alfalfa production across the Midwest and which other crops are being produced with alfalfa. This training will be used in the student's Masters degree program. 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?In the final year of the project, we will begin the process of combining the data to refine many of the initial models that we have in rough form and add additional information for sensitivity analysis. It is expected that during this time and at the conclusion of the project (end of the next reporting period) that multiple technical writings will be prepared or in preparation. Additionally, we will write articles for trade journal publications describing the work we are doing. A final step will be preparing the LCA data for publication in LCA databases. Specific dissemination targets are: Paper on alfalfa stand length and crop rotations Paper or papers that include soil modeling, LCA, and economics (likely 2) Publication in trade journals- survey data and the variety of production methods Producer focused publications on Extension or other University websites

Impacts
What was accomplished under these goals? Objective 1. Identifying regional practices Survey Based Regional Practice Data An online survey was distributed to alfalfa producers throughout the upper Midwest to gain a better understanding of alfalfa production practices in the region. A total of 132 responses to the survey were received. Farmers or farm managers made up 68% of the survey responses, with 26% being crop consultants. Those who completed the survey influenced over 570,000 total cropland acres, with 82,000 acres of alfalfa and 138,000 acres of corn silage being represented. The survey respondents represented over 370,000 head of dairy and beef cattle. The average alfalfa yield reported by survey respondents was 4.7 tons of alfalfa per acre for established stands. The average stand length of alfalfa for those who completed the survey was 4 years in production, not including the establishment year. The average number of times alfalfa was cut in the establishment year was 2 times, with 3.6 times per year in established stands. At seeding, nearly 60% of respondents reported using a companion crop. Oats was the most common companion crop reported. Alfalfa was reported as being seeded with at least some perennial grasses by 45% of the survey respondents. Orchardgrass, timothy, meadow fescue and tall fescue were the most common perennial grasses that were included. Over 77% of survey respondents utilize manure to fertilize alfalfa at some point during the life of the stand. Nearly half of the respondents utilize manure before establishment, and 37% use manure during the life of the stand. Rates used were reported at 8200 gallons per acre on average for liquid manure and 6 tons per acre when solid manure was used. Average stand length common crop rotations Across the Midwest The Soil Survey Geographic Database (SSURGO), Cropland Data Layer (CDL), and county and state boundary shapefiles were obtained. Using the software ArcGIS, the original 255 CDL crop category IDs were reclassified and condensed into 9 current crop categories. 9 years of CDL (2011 - 2019) were compiled into a single raster dataset with each raster containing a 9 digit crop ID code representing the chronological sequence of crops grown. The raster dataset was converted to polygons and then exported to Excel. Within Excel, the 9 digit rotations were used to identify alfalfa stands and calculate the stand length for each polygon, representing the field scale. Next, those figures were averaged within each state to obtain state average stand lengths. Additionally, the crop rotational codes were used to calculate crops grown one and two years previous to alfalfa stands and one and two years after alfalfa stands at the state level. Ongoing work will summarize this data by county for further analysis. Investigating production practices and yields from literature and databases A number of databases and university extension production guides for alfalfa are being used to guide modeling assumptions. USDA-NASS data provides county level data on production plantings and yields of alfalfa, as well as data for the location and scale of dairy operations. Economic data is being provided by USDA-AMS databases and the University of Minnesota Finbin (Farm Financial Data Set). Chemical, fertilizer, and other physical inputs are best calculated from survey data as there is no direct data source for exactly how much of these inputs are being used. Extension guidelines for alfalfa production inputs are fairly general and allow producers many options for production methods. In addition, the guidelines specify many inputs on an as needed basis. Objective 2. Develop Scenarios Over the last year, we have developed several scenarios that cover the dairy related production of alfalfa, crop rotations, and milk production. For individual crops, we are examining alfalfa hay, alfalfa haylage, corn grain, corn silage, soybeans, canola, grass hay (currently timothy), small grain (currently oats). These crops are being used in dairy rations and corresponding production scenarios to examine the complete system. The current variations of the complete dairy system are: Baseline: Corn Silage Continuous Cropping Alfalfa in Rotation (alfalfa baseline) Alternative: Corn Soybean Rotation (corn grain and silage) Alternative: Alfalfa Grass Mixture Alternative: Companion crop during establishment Alternative: Inter-seeding with corn We are also examining regional variations in alfalfa production for the areas with a dairy production presence. Though we have a rough map of these regions based on NASS production data, we are further refining the regions with the stand length data using principal component analysis. Objective 3 Modeling Economics A preliminary cost analysis was conducted for six dairy feed scenarios with different levels of alfalfa. The main focus of the cost analysis was on the operating costs to grow the alfalfa and other homegrown feeds required. The crops considered are corn grain, corn silage, alfalfa haylage, canola, and oats. The inputs included are seed, fertilizer, chemicals, and diesel fuel. The crop yields and input quantities in the economic analysis are taken from a spreadsheet LCA model developed by PI Tallaksen. Prices are from USDA Agricultural Market Service reports and the FINBIN database of 2020 farm business and financial management summaries for the six states. This preliminary analysis showed that the lowest-cost diet and cropping scenario is the baseline corn silage one while the highest-cost one is the heavy alfalfa one. The cost results are generally correlated with the energy requirements for the first four scenarios. The other two scenarios - alternative protein (canola) and small grain (oats) diet, were less costly than the heavy alfalfa and heavy grass diets despite requiring more energy. Soil Carbon and Nutrient Flux Modeling Soil biogeochemical process models including DAYCENT/CENTURY, ArcAPEX, and APEX were evaluated to determine their suitability to accomplish the objectives of this project. The APEX model was chosen because of its flexibility in modelling generalized cropping system scenarios across regional scales and its well-documented accuracy in modelling both soil carbon and water quality responses to agricultural management. Preliminary baseline cropping scenarios were developed in alignment with the six dairy feed scenarios described above in Objective 2: Scenarios. The baseline cropping scenarios developed are: 1. Continuous silage corn (Unless noted otherwise, all corn is for silage) 2. Corn for grain - soybean rotation 3. Heavy alfalfa rotation (corn-corn-alfalfa-alfalfa-alfalfa-alfalfa) 4. Light alfalfa rotation (corn-corn-corn-corn-alfalfa-alfalfa) 5. High protein rotation (corn-corn/winter canola/soybean-alfalfa-alfalfa-alfalfa) 6. Heavy grass mixture (corn-corn-alfalfa-alfalfa-orchard grass-orchard grass) These baseline scenarios were modelled for an example farm near Morris, MN for which we have extensive crop, soil carbon, and water quality data for model validation. Simulations were run for 100 years of continuous management. This time horizon was chosen to align with the 100-year global warming potential for greenhouse gases used by IPCC. Model calibration work is ongoing as we examine unique rotations in specific regions. Life cycle assessment A basic life cycle assessment model for alfalfa has been created in SimaPro LCA software. This model is set up as a prototype for all variations of alfalfa production we plan to analyze. Preliminary models for other common crops had previously been developed and will serve as the basis for those crops in this project. As final practices and yields for the variations we are modeling are decided upon, they can be quickly put into the model for analysis. Objective 4- Evaluate and disseminate At this stage, it is early in the modeling to thoroughly dissect and evaluate the findings. Once this evaluation has occurred, we will begin disseminating project results.

Publications

Progress 09/01/19 to 08/31/20

Outputs
Target Audience: Nothing Reported Changes/Problems:The timeline for this project has changed substantially from what was written into the proposal. Though the initial award date was 9/1/19, the extended review and final approval of the award was not completed until later in February 2020. Therefore, we are effectively 3 months into the research. The changes to standard operations in the face of the Covid-19 situation has also slowed us somewhat as we have been onboarding graduate students and beginning the project from remote locations. As we continue the project, we are ramping up activities to make up some time lost in these delays. Bringing graduate students on board will be a great help in data collection and will allow us to proceed to more quickly in our modeling and scenario development objectives. What opportunities for training and professional development has the project provided? Nothing Reported 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?Note: Text for this question is also referenced in the progress report question above. Objective 1. Evaluate regional production practices As we continue this project, we will begin to refine production maps with information from regional research and extension publications, BMPs, and published yield data covering alfalfa and other rotational crops. Some of the key inputs for modeling include factors such as fertilization, irrigation, cutting frequency, alfalfa varieties and harvest methods. Objective 2. Develop scenarios to optimize alfalfa's benefits for cropping and dairy systems We will be using the information to develop modeling scenarios for A) building alfalfa production scenarios, B) building rotational cropping scenarios, and C) building forage (alfalfa) based dairy production system scenarios. The goal in looking at each of these areas is to identify innovative, easy to integrate, and financially viable techniques to improve the environmental sustainability in the alfalfa-to-dairy process. This work will be a collaborative effort that brings together the agronomy, dairy, and modeling experts. Objective 3. Develop LCA, Soil Carbon/Nutrient Flux, and Economic Models Soil Carbon and Nutrient Flux Modeling A preliminary parameterization of the ArcAPEX model is underway to examine changes in SOC for a small watershed in West Central Minnesota with respect to forage cropping practices. ArcAPEX supports the simulation of a variety of soil processes across a diverse field or at the small-catchment scale. This approach could be useful for a case-study approach to modelling alfalfa scenarios across the region. Alternatively, we are exploring approaches that can integrate process models such as APEX or CENTURY with geospatial data at broader spatial scales to develop regional spatialized SOC assessments. (Caddeo et al., 2019) provide one such approach whereby the CENTURY model was integrated with land cover, soil, and climate layers via Microsoft Visual Basic© and R scripts for obtaining spatialized results of model runs. A graduate student with expertise in spatial analysis and spatial model development has recently been hired to work on this effort. LCA Modeling Going forward, we will continue to incorporate data from our initial analysis of alfalfa, cropping, and dairy systems into developing the alfalfa LCA framework. It will initially consist of the regions already identified, but will be expanded as more refined data and data for alternative scenarios is finalized. As this work proceeds and more data become available, the LCA methodology will be refined to accommodate issues such as material import/export and herd size. Another critical piece that will be added to the LCA is the results from soil modeling. Soil carbon and other nutrients are important components when considering environmental issues. These will be examined to look at how the changes in agronomics and crop rotations impact soil emissions and, ultimately, greenhouse gases emissions. Economic modeling We have had discussions about approaches and data sources for the economic modeling. Such economic modeling normally begins with enterprise budgets for scenarios considering the relevant crops, resources, operating inputs, yields, and prices. Most of that information is the same as will be required for the LCA modeling except for the prices. The most current prices available will be used. Many of the prices, such as farmgate sale prices for crops, are publicly available from sources such as the USDA National Agricultural Statistics Service or the Agricultural Marketing Service. We will need to conduct our own informal surveys in order to quantify typical prices for other items such as for seed and chemicals. Another key question concerns alfalfa stand length and crop rotation. The 2014 Agronomy Journal article by Yost et al. and the 2006 report by Burnett are two papers that provide suggestions for developing scenarios that include these considerations. Data from the Wisconsin Alfalfa Yield and Persistence (WAYAP) program suggests that alfalfa yields in Wisconsin do not decline in years 2 and 3 after establishment as quickly as Barnett assumed. Objective 4. Evaluation and dissemination goals The scenario evaluation and dissemination work for the project will start after the modeling work is further along. Scenarios will be developed based on the environmental impacts and economic results of the modeling. This will be a primary focus of the final months of the study as we finalize models and identify optimal production strategies.

Impacts
What was accomplished under these goals? Objective 1. Evaluate regional production practices The project team has begun the effort to identify key alfalfa production areas in the six-state region by using data from the USDA-Nass 2017 agricultural survey. We are using this rough data to identify factors influencing where and why alfalfa is being produced in certain regions. As part of this effort, we are examining the co-location of dairy facilities and herd numbers with their feed sources. Another component of regional production that we are interested in is the agronomic methods and inputs producers are employing. This data is being combined with other county-level crop production, land use, and environmental data to identify regions with similar alfalfa production systems and feed uses for LCA and other analysis. As we continue this project, we will begin to refine these maps with regional research and extension publications, BMPs, and published yield data covering alfalfa and other rotational crops. Some of the key inputs for modeling include factors such as fertilization, irrigation, cutting frequency, alfalfa varieties and harvest methods. Objective 2. Develop scenarios to optimize alfalfa's benefits for cropping and dairy systems Although some scenario development discussions have started, this objective requires more detailed information from objective 1 for completion. We will be using the information to develop modeling scenarios for A) building alfalfa production scenarios, B) building rotational cropping scenarios, and C) building forage (alfalfa) based dairy production system scenarios. The goal in looking at each of these areas is to identify innovative, easy to integrate, and financially viable techniques to improve the environmental sustainability in the alfalfa-to-dairy process. Objective 3. Develop LCA, Soil Carbon/Nutrient Flux, and Economic Models Soil Carbon and Nutrient Flux Modeling We have begun discussions about how to model soil carbon and nutrient flux in a way that best represents the spatial heterogeneity of alfalfa production acreage and alfalfa production practices across the 6-state target region. This effort is being developed in parallel to the evaluation of current production practices for Objective 1. Work has begun on an analysis of the USDA-NASS Cropland Data Layer to evaluate the spatial distribution of alfalfa acreage across the region and also to update the alfalfa stand length and crop rotation analysis of (Yost et al., 2014) to include the years 2013 to 2019. This analysis will provide spatially explicit information on the length of alfalfa rotations and the most commonly used rotational crops across the region. This data will then be summarized by the identified LCA regions and overlaid with climate information and SSURGO soils data to determine how to best represent the variability in alfalfa production, soils, and climate for the soil carbon and nutrient flux modelling effort. This work will inform decision making for scenario development for Objective 2. We have also begun exploring various soil models for this effort. A preliminary parameterization of the ArcAPEX model is underway to examine changes in SOC for a small watershed in West Central Minnesota with respect to forage cropping practices. ArcAPEX supports the simulation of a variety of soil processes across a diverse field or at the small-catchment scale. This approach could be useful for a case-study approach to modelling alfalfa scenarios across the region. Alternatively, we are exploring approaches that can integrate process models such as APEX or CENTURY with geospatial data at broader spatial scales to develop regional spatialized SOC assessments. (Caddeo et al., 2019) provide one such approach whereby the CENTURY model was integrated with land cover, soil, and climate layers via Microsoft Visual Basic© and R scripts for obtaining spatialized results of model runs. A graduate student with expertise in spatial analysis and spatial model development has recently been hired to work on this effort. LCA Modeling Using the data on alfalfa production from objective 1 above and discussions with team members, LCA modeling efforts are first focusing on the distinct differences in production across the upper Midwest and how that can be best modeled. The source data is for these early regional production maps is fairly course county and state level information, so the data for each region will be updated further as the project continues. The initial LCA efforts have centered on the regions that are already producing substantial amounts of alfalfa, those that have a strong dairy presence, and those that have the potential for dairy-centered alfalfa production. Key questions in LCA methodology are being discussed, such as the functional units for analysis at each stage (i.e. alfalfa impacts per raw ton vs per unit of net energy of lactation adjusted ton) and the scale of dairy system being modeled in each region (i.e. 300 cow vs 8000 cow herds). Similarly, we have identified the regional transportation of alfalfa as well as transportation of alfalfa outside the six-state area as considerations in modeling the dairy systems. In combination with early data from extension publications and other sources, we are considering how to approach these issues and the availability of data in enhancing LCA resolution with different methods. Initial LCA work has been set up in OpenLCA software for output of data in the project. The cumulative energy demand (CED) and global warming potential (GWP) methods were imported and updated to function using NREL data and other US based cropping databases. Another critical piece that will be added to the LCA is the results from soil modeling. Soil carbon and other nutrients are important components when considering environmental issues. These will be examined to look at how the changes in agronomics and crop rotations impact soil emissions and, ultimately, greenhouse gases emissions. Economic modeling We have had discussions about approaches and data sources for the economic modeling. Such economic modeling normally begins with enterprise budgets for scenarios considering the relevant crops, resources, operating inputs, yields, and prices. Most of that information is the same as will be required for the LCA modeling except for the prices. The most current prices available will be used. Many of the prices, such as farmgate sale prices for crops, are publicly available from sources such as the USDA National Agricultural Statistics Service or the Agricultural Marketing Service. We will need to conduct our own informal surveys in order to quantify typical prices for other items such as for seed and chemicals. Another key question concerns alfalfa stand length and crop rotation. The 2014 Agronomy Journal article by Yost et al. and the 2006 report by Burnett are two papers that provide suggestions for developing scenarios that include these considerations. Data from the Wisconsin Alfalfa Yield and Persistence (WAYAP) program suggests that alfalfa yields in Wisconsin do not decline in years 2 and 3 after establishment as quickly as Barnett assumed. Objective 4. Evaluation and dissemination goals The scenario evaluation and dissemination work for the project will start after the modeling work is further along. Scenarios will be developed based on the environmental impacts and economic results of the modeling. This will be a primary focus of the final months of the study as we finalize models and identify optimal production strategies.

Publications