Progress 09/01/23 to 08/31/24
Outputs
Target Audience:The DAWN project's primary aim is to reach farmers in the U.S. Corn Belt, as well as agricultural information providers such as extension educators and government agencies. Throughout the year our Extension personnel engaged with farmers both directly (via personal conversations) and indirectly (through speaking engagements, panel and breakout sessions, and poster presentations). Many of these interactions occurred in formal events such as Extension Field Days and Agricultural Conferences, while others involved leveraging existing relationships with farmer partners, particularly through the DAWN-partnered farming organization UnCommon Farms. We have also gained interest from agricultural service providers, including irrigation providers, insurance companies, and financial institutions. Overall we have reached approximately 1,500 potential DAWN users. In addition to engaging agricultural stakeholders, the DAWN project aims to reach a secondary audience of climate and agricultural scientists on topics including decision support science, stakeholder engagement, and advances in seasonal climate prediction. DAWN findings, outcomes, and lessons have been shared in a variety of academic and research contexts, including conference presentations, seminars, and academic papers. Finally, undergraduate students have been engaged in DAWN-related coursework, research projects, and the DAWN summer internship program. Graduate students and post-doctoral scholars have received professional development and research skills training. Changes/Problems:As described in our report last year, we have modified our scope in response to stakeholder feedback, focusing exclusively on S2S forecasting and providing local-scale rather than broad regional outcomes. Likewise, we have prioritized the integration of models that can operate well at these scales, and some of the initial suite of models may not be incorporated into the dashboard by the project end. DAWN was initially designed around a set of models chosen because of their proven scientific advances; however, our team strongly believes that decision support must be user-driven, not model driven, and thus this change reflects our stakeholder-centric focus. In particular, the economic component has been scaled back, since the initial model did not translate well to the new scale. We continue to work to find ways to integrate the most essential information and data into our dashboard, and part of that process involves remaining flexible and responsive to user feedback. We strongly believe that these changes will create a final product that is more tailored to stakeholder needs. What opportunities for training and professional development has the project provided?The fourth objective of the DAWN project is to provide transdisciplinary education for future leaders in agricultural water use and nutrient management. In addition to our undergraduate educational program described above, undergraduate students, graduate students, and postdocs all assisted in the development of DAWN's dashboard tools and coupled modeling system under the mentorship of senior researchers. Most of DAWN's web developers are University of Maryland and Colorado State University undergraduate and graduate students who, in addition to financial compensation for their time, receive hands-on experience and training from senior project personnel. Most of these students come from computer science and programming backgrounds, and working on DAWN provides exposure to applied climate and agricultural science as well as valuable experience working on real-world project development. Graduate students are also heavily involved throughout the DAWN research and extension efforts. They participate in regular team meetings, receive one-on-one mentorship from DAWN faculty, and are trained in process understanding, model development, extension outreach, tool evaluation, and interdisciplinary science. Many students working on research models have the unique opportunity to collaborate with extension personnel, encouraging them to cross research-to-operations divides and consider the real-world impact of their modeling research. Finally, DAWN extension personnel interacted with stakeholders and community members through a variety of field days, conferences, and one-on-one conversations. Besides educating stakeholders in potential uses of DAWN tools, these interactions provided exposure to the concept of seasonal forecasting using NOAA or CWRF predictions. How have the results been disseminated to communities of interest?Our primary community of interest is the agricultural community, which includes not only farmers but related industries (seed companies, insurance, etc.) and extension agents. The main objective of this project has always been dissemination of results to these communities, and the products we published this year, described above, are a unique means for agriculturalists to directly access climate information relevantto their particular farms. We engaged in significant outreach efforts associated with our Phase 2 launch, both to create awareness about the tools themselves, and to provide users with resources for exploring DAWN's capabilities. In particular, we updated existing and developed new help pages with Tango tutorials and video walkthroughs for the new tools so users can learn independently. These are all accessible through the DAWN website athttps://dawn.umd.edu/docs/48/Getting-Startedandhttps://dawn.umd.edu/docs/47/Tools. We also created a communications kit (including social media copy, email template, newsletter blurbs) to help team members promote the new tools and dashboard updates. We announced the Phase 2 launch to an external audience of stakeholders, including supporters and current and potential users, informing them about the new tools and updates and encouraging them to try DAWN. Aside from agricultural communities, we also work to disseminate our results to students and the larger research community. For students, we have both our summer internship program, in which they can directly engage with the project, and our global classroom class, where they learn about intersectional issues of water, agriculture, and food production. Finally, our project members attend workshops and conferences throughout the year, sharing DAWN information and knowledge with a variety of communities. This past year, major academic conferences included the American Geophysical Union and American Meteorological Society annual conferences, which were attended by a large number of DAWN members. Additionally, our Extension team attended multiple conferences and other academic activities, including the Agronomy, Crop and Soils Society Meeting, the Minnesota Water Resources Conference, the Midwest Crops Conference, Aligning Global Research Priorities & Investment Strategies for Sustainable Water Management, the International Precision Agriculture Conference, and Farming for the Future: Digital and Regenerative At Integration. At these events, the team reached more than 300 participants of various academic disciplines and backgrounds. Our primary community of interest is the agricultural community, which includes not only farmers but related industries (seed companies, insurance, etc.) and extension agents. The main objective of this project has always been dissemination of results to these communities, and the products we published this year, described above, are a unique means for agriculturalists to directly access climate information relevantto their particular farms. We engaged in significant outreach efforts associated with our Phase 2 launch, both to create awareness about the tools themselves, and to provide users with resources for exploring DAWN's capabilities. In particular, we updated existing and developed new help pages with Tango tutorials and video walkthroughs for the new tools so users can learn independently. These are all accessible through the DAWN website athttps://dawn.umd.edu/docs/48/Getting-Startedandhttps://dawn.umd.edu/docs/47/Tools. We also created a communications kit (including social media copy, email template, newsletter blurbs) to help team members promote the new tools and dashboard updates. We announced the Phase 2 launch to an external audience of stakeholders, including supporters and current and potential users, informing them about the new tools and updates and encouraging them to try DAWN. Aside from agricultural communities, we also work to disseminate our results to students and the larger research community. For students, we have both our summer internship program, in which they can directly engage with the project, and our global classroom class, where they learn about intersectional issues of water, agriculture, and food production. Finally, our project members attend workshops and conferences throughout the year, sharing DAWN information and knowledge with a variety of communities. This past year, major academic conferences included the American Geophysical Union and American Meteorological Society annual conferences, which were attended by a large number of DAWN members. Additionally, our Extension team attended multiple conferences and other academic activities, including the Agronomy, Crop and Soils Society Meeting, the Minnesota Water Resources Conference, the Midwest Crops Conference, Aligning Global Research Priorities & Investment Strategies for Sustainable Water Management, the International Precision Agriculture Conference, and Farming for the Future: Digital and Regenerative At Integration. At these events, the team reached more than 300 participants of various academic disciplines and backgrounds. What do you plan to do during the next reporting period to accomplish the goals?Our plans for next year prioritize outreach, feedback, and the continued development and refinement of our cohesive set of tools. We aim to publicly release Phase 3 of DAWN in January 2025 - this release will include a new DSSAT-based crop tool, as well as a significant expansion of the current Data Viewer to include historical hindcasts and skill evaluation. Following the launch, we will extensively engage with our agricultural community partners to solicit feedback on the tools' usefulness and usability; findings will be used to guide further refinement. Concurrently, we will be rolling out additional new capabilities, including changes to the Data Viewer based on our past year of user testing, and a new irrigation tool. To accomplish our first objective of proactively engaging agricultural producers in the collaborative design, testing, and implementation of the web-based decision support system, our Extension team will continue agriculturalists via webinars, demos, practitioner conferences, field days, and one-on-one trainings. We will debug and evaluate the upcoming Phase 3 and 4 products, including a stakeholder assessment of different communication strategies to support the correct interpretation of the hindcast product.Additionally, we will develop communications products for the Phase 3 rollout to help DAWN users understand how to use the new tools and to help with outreach at events. Further, we will share a quarterly external newsletter to engage with users about DAWN tools/features at key points during the year, and further develop the website's documentation. To better evaluate the DAWN tools, we will engage a diverse group of end-user stakeholder in a DAWN usability study using Q-methodology. Findings will clarify the diversity of perspective amongst end-users and help researchers enhance the overall usability and outreach for climate decision support tools like DAWN, while embedding DAWN into a broader "theory of change," that better captures the entire decision environment of DAWN's end-users. This research will be submitted for publication and presented at scientific conferences, will support the training of a postdoc and an underrepresented undergraduate student, and improve the DAWN extension and communication activities and products. For the second objective of building effective decision support systems based on a coupled modeling system, we plan to release our Phase 3 tools in January 2025. These will include the DSSAT-based crop tool and the hindcast expansion of the Data Viewer. Following that, we will move towards Phase 4, which will include new dashboard tools such as an irrigation planner, and feedback-based revisions to the existing tools, particularly the Data Viewer. These include building out design changes that have already been planned, including incorporating confidence metrics for all dashboard tools, and designing a Scenario Builder that can be used across dashboard tools. Additionally, we will incorporate a secondary backup site to maintain DAWN services during issues such as power outages at our main hosting area. We will add more explicit documentation for the DAWN API, enabling use of DAWN information with other applications or servers. Our third objective is the production of reliable forecasts. We intend to standardize the CWRF-DSSAT seasonal climate-crop prediction and facilitate its integration into the DSSAT-based crop tool. We will simultaneously enhance CWRF-DSSAT prediction skills through refining physics representation, machine learning, superensemble optimization, and bias correction; this includes implementation of DSSAT multi-cultivar crop ensemble modeling, CWRF climate lateral boundary bias correction approach, and ensemble optimization techniques to improve overall DAWN dashboard forecast skills. To improve our overall presentation of forecast information to stakeholders, we will develop objective metrics to present the skills and uncertainties of the DAWN predictions in the way that can be well understood by agricultural stakeholders. We will also improve the DAWN forecast system initialization by constructing CWRF initial conditions using observational data assimilated CSSP long-term integrations. We will stabilize model codes to ensure CWRF and DSSAT runs smoothly on different computer systems with varying compilers for sustained DAWN dashboard service. We will also improve the data post-processing procedure and conduct a comprehensive analysis of the hindcast and forecast system (5-day updates; 9-month lead) outputs. Beyond CWRF and DSSAT, we will continue to update our core suite of models to move towards dashboard operationality. In particular, we will evaluate spatial (multi-site) prediction confidence/validation for BioCro soybean and corn growth models. We will also continue to update the irrigation models using feedback gathered during the process of creating the new irrigation tool. Finally, we will document the many changes that have been made in these models to make them more applicable for local S2S prediction, and publish in peer-reviewed journals. Finally, to accomplish the fourth objective of providing transdisciplinary education, DAWN team members will continue and improve upon many of the activities carried out in the past years. Education specialists will administer the fifth year of the DAWN summer internship program, and organize and administer professional development workshops for DAWN early-career scientists. DAWN's leadership will also communicate information about its objectives, products, and research findings through the publication of a special issue focused on moving S2S modeling from research to operations.
Impacts
What was accomplished under these goals?
DAWN aims to develop a dashboard of co-produced tools that address key stakeholder needs at decision-relevant scales, closing a long-standing gap between scientific advances in earth system modeling and the stakeholders who make agricultural decisions. Year 4 marked the Phase 2 public launch of the DAWN Dashboard, which included improvements to existing tools (Crop Progress, Growing Degree Days), and added a new data viewer that shows NOAA and CWRF predictions for temperature and precipitation. Additionally, we developed and implemented mechanisms for stakeholder engagement, communication, and co-production; improved the capabilities of the foundational models; and provided training and educational opportunities to students and early career scientists. In particular, the incorporation of CWRF was a significant accomplishment because it provides field-scale access to both global and downscaled temperature and precipitation outlooks, which are foundational inputs for all the DAWN seasonal-to-subseasonal (S2S) decision support tools. Objective 1: The DAWN extension team introduced farmers, certified crop advisors, agricultural extension agents, and on-farm advisors to the DAWN tools to support product use and to understand scientific and design opportunities to increase the usefulness of the tools for strategic agricultural decisions. These engagements reached over 1,000 agriculturalists. Leading up to the release of Phase 2, our extension team conducted think aloud exercises to observe how new users engaged with the materials. We identified which tool design aspects were easy to understand and useful for decisions, and which were difficult to navigate or understand. We synthesized the observations into internal recommendations to modify and improve the usefulness of DAWN's tools. Many were implemented within the Phase 2 release, while others are under consideration for future releases. Additionally, DAWN's extension effort supported a multi-month co-production process to evaluate the pre-public release of DAWN. This included internal and external engagement on a daily-to-weekly basis to identify bugs and make design recommendations to improve the Phase 2 products given the significant backend improvements and incorporation of CWRF. This ensured that the Phase 2 products would perform more reliably and provide agriculturalists in the Corn Belt region with a unique market product that supports strategic decision-making. Objective 2: DAWN's Phase 2 was publicly launched in June 2024. It included significant improvements to the Phase 1 tools, including significant UI upgrades to all existing tools, the conversion of tools to run on specific fields rather than being restricted to latitude/longitude, the addition of a new Drydown Calculator, and a field selection tool allowing users to upload shapefiles or manually select regions. Additionally, the CWRF climate model predictionswere made publicly available alongside NOAA's CFSv2 predictions. Temperature and precipitation predictions for the next sixth months are now available on the new Data Viewer tool, which places predictions in the context of historical patterns, allowing users to see whether temperature and rainfall are predicted to be average or below/above average. Several significant behind-the-scenes upgrades were also implemented. We overhauled the online content management system to facilitate improved communication and content generation. We evaluated and improved site accessibility. All web pages were updated to be mobile friendly, to better facilitate access. We updated our user feedback and comment management system to provide a better experience for users, allowing them to see the developers' response to their feedback. On the back end, we automated the data ingest and testing to streamline the forecast upload process and preempt potential errors before they reach users. Our Design Team, which consists of a mix of developers, and experts in modeling, extension, and visualization, guided the Phase 2 tool development to ensure that the highest scientific rigor was combined with the prioritization of user needs and the logistics of the online delivery system. The new data viewer is the result of this interdisciplinary effort. The design process is iterative, with feedback solicited both on preliminary mockups and then throughout the tool-building process. Finally, the Design and Development teams have begun the next phase of DAWN tools. The Phase 3 release, currently slated for January 2025, will add hindcasts to the Data Viewer, allowing users to see how well particular climate models have performed in their region in the past. Additionally, the DSSAT-based crop tool will allow them to explore the combined impact of climate and fertilization strategies on crop growth. Aside from this, additional development includes refining the Data Viewer based on feedback received following the phase 2 launch, and the back-end development of a SWAP-based irrigation scheduling tool. Objective 3: We continue to revise the underlying forecasts and models on which this process is based, improving the models' applicability for decisions, the post-processing of results, and the delivery of those results to the web dashboard system. This year was focused on optimizing the available human and computing resources for S2S prediction. We began producing the initial version of the operational CWRF for the Phase 2 launch, with forecasts running every five days with a seven-month lead time. We also ran coupled CWRF-DSSAT for forecasts and hindcasts, in preparation for the Phase 3 release of our new crop tool. The delivery of all forecasts to the web system has been automated and refined. Additionally, we evaluated the NOAA and CWRF climate predictive skill using hindcasts and assessed lateral forcing bias correction methods to further enhance CWRF downscaling capabilities. In addition to our climate modeling efforts, we continue to support and develop several supporting models still heading towards integration into the DAWN system. We extensively tested the performance of S2S forecasts used for real-time irrigation scheduling for corn fields in the continental US, using the new Illinois Irrigation Scheduling Tool. We also continued developing the BioCro model, which will supplement DSSAT for crop forecasting. We integrated a new soil-water module that more accurately predicts changes in soil water content. This is expected to improve spatial prediction (multi-site validations). We also developed and added a new process-based carbon (C) allocation that make it more robust to new environments, needing less re-training, making it more suitable for predictions over a range of sites/locations. We expect this to improve the ability of BioCro to be used for operational purposes. Objective 4: DAWN's ten-week summer internship program provided six undergraduate students from three institutions the opportunity to work closely with DAWN research, extension, and education personnel from June 3rd to August 9th 2024. During this time, the interns worked on individual projects of interest and attended a week longprofessional development a workshop. DAWN educators also taught an undergraduate course relevant to the science of DAWN during Fall 2023 and 2024. Additionally, experiential education is deeply rooted into all stages of the DAWN project; beyond our undergraduate classes and student internships, we employ student developers as programmers, and incorporate graduate students and postdoctoral fellows in every step of the process. These other activities are described in more detail in the next question.
Publications
- Type:
Peer Reviewed Journal Articles
Status:
Published
Year Published:
2024
Citation:
Liang, Xin-Zhong, Drew Gower, Jennifer A. Kennedy, Melissa Kenney, Michael C Maddox, Michael Gerst, Guillermo Balboa, Talon Becker, Ximing Cai, Roger Elmore, Wei Gao, Yufeng He, Kang Liang, Shane Lotton, Leena Malayil, Megan L. Matthews, Alison M. Meadow, Christopher Neale, Greg Newman, Amy Rebecca Sapkota, Sanghoon Shin, Jonathan Straube, Chao Sun, You Wu, Yun Yang, and Xuesong Zhang, 2024: DAWN: Dashboard for Agricultural Water use and Nutrient management - A predictive decision support system to improve crop production in a changing climate. Bulletin of the American Meteorological Society, 105, E432-441, doi: 10.1175/BAMS-D-22-0221.1.
- Type:
Peer Reviewed Journal Articles
Status:
Published
Year Published:
2023
Citation:
" Zhang, Jingting, Hanqin Tian, Yongfa You, Xin-Zhong Liang, Zutao Ouyang, Naiqing Pan, and Shufen Pan, 2023: Balancing non-CO2 GHG emissions and soil carbon change in U.S. rice paddies: A retrospective meta-analysis and agricultural modeling study. AGU Advances, 5, e2023AV001052, doi: 10.1029/2023AV001052.
- Type:
Peer Reviewed Journal Articles
Status:
Published
Year Published:
2023
Citation:
He, Yufeng, Deepak Jaiswal, Stephen P. Long, Xin-Zhong Liang, and Megan L. Matthews, 2023: Biomass yield potential on U.S. marginal land and its contribution to reach net-zero emission. Global Change Biology-Bioenergy, 16, e13128, doi: 10.1111/gcbb.13128.
- Type:
Peer Reviewed Journal Articles
Status:
Published
Year Published:
2023
Citation:
Dangol, Sijal, Xuesong Zhang, Xin-Zhong Liang, and Elena Blanc-Betes, 2023: Advancing the SWAT model to simulate perennial bioenergy crops: A case study on switchgrass growth. Environmental Modelling and Software, 105834, doi: 10.1016/j.envsoft.2023.105834.
- Type:
Peer Reviewed Journal Articles
Status:
Published
Year Published:
2023
Citation:
" You, Yongfa, Hanqin Tian, Shufen Pan, Hao Shi, Chaoqun Lu, William Batchelor, Bo Cheng, Dafeng Hui, David Kicklighter, Xin-Zhong Liang, Xiaoyong Li, Jerry M Melillo, Naiqing Pan, Stephen Prior, and John Reilly, 2023: Net greenhouse gas balance in U.S. croplands: How can soils be a part of the climate solution? Global Change Biology, 30, e17109, doi: 10.1111/gcb.17109
- Type:
Peer Reviewed Journal Articles
Status:
Published
Year Published:
2023
Citation:
Clark, S. J.F. Wolfinger, M.A. Kenney, M.D. Gerst, H.A. Roop. (2023) Assessing stakeholder climate data needs for farm-level decision-making in the U.S. Corn Belt. Geoscience Communication. https://doi.org/10.5194/gc-6-27-2023
|
Progress 09/01/22 to 08/31/23
Outputs
Target Audience:The DAWN project's primary aim is to reach farmers in the U.S. Corn Belt, as well as agricultural information providers such as extension educators and government agencies. Throughout the year our Extension personnel engaged with farmers both directly (via personal conversations) and indirectly (through speaking engagements, panel and breakout sessions, and poster presentations). Many of these interactions occurred in formal events such as Extension Field Days and Agricultural Conferences, while others involved leveraging existing relationships with farmer partners, particularly through the DAWN-partnered farming organization UnCommon Farms. We have also gained interest from agricultural service providers, including irrigation providers, insurance companies, and financial institutions. In addition to engaging agricultural stakeholders, the DAWN project aims to reach a secondary audience of climate and agricultural scientists on topics including decision support science, stakeholder engagement, and advances in seasonal climate prediction. DAWN findings, outcomes, and lessons have been shared in a variety of academic and research contexts, including conference presentations, seminars, and academic papers. Finally, undergraduate students have been engaged in DAWN-related coursework and through a summer internship program, while graduate students and early career scientists have received professional development and subject-related training. Changes/Problems:A major decision made during year three was to focus exclusively on field-scale seasonal-to-subseasonal (S2S) forecasting, rather than including tools or products focused on weather-scale operations or large regional outcomes. This shift was made based largely on stakeholder and extension feedback, particularly the following considerations: 1) many more agricultural resources currently exist at the weather scale than at the S2S scale; 2) models focused at the weather scale typically requires data that many farmers may be unable or unwilling to input into a new tool system; 3) the prohibitive computational cost of running daily simulations. DAWN was initially designed around a set of models chosen because of their proven scientific advances; however, our team strongly believes that decision support must be user-driven, not model driven, and thus this change reflects our stakeholder-centric focus. As a result, the development of hydrological tools has been pushed back to year 5 as our modelers work to refocus from daily to subseasonal scales. Additionally, rather than providing large scale general equilibrium modeling, our economic modelers are creating smaller yield-focused tools that help to translate predicted yield values into financial outcomes. We strongly believe that these changes will create a final product that is more tailored to stakeholder needs. What opportunities for training and professional development has the project provided?The fourth objective of the DAWN project is to provide transdisciplinary education for future leaders in agricultural water use and nutrient management, as described above. Undergraduate students, graduate students, and postdocs all assisted in the development of DAWN's dashboard tools and coupled modeling system under the mentorship of senior researchers. The majority of DAWN's web developers are University of Maryland and Colorado State University undergraduate and graduate students who, in addition to financial compensation for their time, receive hands-on experience and training from senior project personnel. Most of these students come from computer science and programming backgrounds, and working on DAWN provides exposure to applied climate and agricultural science as well as valuable experience working on real-world project development. Graduate students are also heavily involved throughout the DAWN research and extension efforts. They participate in regular team meetings, receive one-on-one mentorship from DAWN faculty, and are trained in process understanding, model development, and interdisciplinary science. Many students working on research models have the unique opportunity to collaborate with extension personnel, encouraging them to cross research-to-operations divides and consider the real-world impact of their modeling research. DAWN's ten-week summer internship program provided eight undergraduate students from five different institutions the opportunity to work closely with DAWN research, extension, and education personnel from June to August 2023. During this time, the interns worked on individual projects of interest and attended weekly professional and writing workshops. DAWN educators also taught an undergraduate course relevant to the science of DAWN during spring 2023. Finally, DAWN extension personnel interacted with stakeholders and community members through a variety of field days, conferences, and one-on-one conversations. Besides educating stakeholders in potential uses of DAWN tools, these interactions provided exposure to the concept of seasonal forecasting using NOAA or CWRF predictions. How have the results been disseminated to communities of interest?Dissemination of scientific results and actionable information to agricultural stakeholders is one of DAWN's primary goals. The DAWN website went public in March 2023, making the first wave of predictive information publicly available. Since then, significant focus has been placed on engaging potential users and early adopters in the currently-available tools. DAWN team members increased awareness of the project and identified interested stakeholders via updates on the website, distribution of informational materials, and presence at farmer-centric workshops such as: the Huskers Harvest Days (the world's largest totally irrigated working farm show, September 2022); the UnCommon Farms Winter Conference (February 2023), Agronomy Peer Group (March 2023) and Regional Meeting (August 2023), and numerous one-on-one interactions from extension personnel. These final interactions have provided valuable opportunities for stakeholders to test DAWN tools while in conversation with extension experts, which provides two-way benefits: DAWN collects feedback and opinions, while users can explore climate and crop predictions, with some growers expressing the potential for adjusting their planting plans based on predictions of lowered growing degree days. Major DAWN announcements were supported through the project's media strategy and widely disseminated via DAWN's member organizations as well as team members' personal networks. DAWN members also shared the results of their project-related research through academic publications, seminars, presentations, and conferences. DAWN's Dashboard is intended as a mechanism to disseminate relevant and usable scientific information to decision makers. Following the Phase 1 release in which an initial set of tools was made publicly available, a Phase 2 release of an additional suite of tools is slated for early 2024. In addition to communicating with stakeholders and scientists about the project outcomes, the DAWN team is committed to sharing the methods and practices used in the development of this large-scale project, transparently discussing both successes and challenges. Besides presenting in academic forums, the project team has developed a research paper describing the first three years of DAWN development describing the project framework, progress, and challenges. The paper was submitted in 2023, and is currently under review by the Bulletin of the American Meteorological Society (BAMS). What do you plan to do during the next reporting period to accomplish the goals?To accomplish the first objective of proactively engaging agricultural producers in the collaborative design, testing, and implementation of the web-based decision support system, our internal extension experts will regularly test the functionality, usability, and accuracy of the existing and developing tools to help finetune and upgrade the products. They will continue to collaborate with modelers, programmers, and the DAWN design team to develop user stories that guide the transition from scientific models to user-oriented tools. Additionally, DAWN's extension team will continue its close engagement with stakeholders to introduce users to the dashboard, and educate them as to its potential uses, and collect feedback and information about user needs and desires. They will conduct demos and trainings with users, including one-on-ones, workshops during field days and conferences, and follow-up trainings to track DAWN use. They will reengage or develop new relationships with key stakeholders, with a particular Year 4 focus on corn growers, as well as supporting the upcoming Phase 2 release through the preparation of training and communication materials. For the second objective of building effective decision support systems based on a coupled modeling system, we plan to release a suite of new predictive tools in the upcoming Phase 2 release, currently scheduled for February 2024. These include CWRF-driven predictions, field-centric tools, the ability to upload shapefiles, a corn dry down calculator, a 3rd party API implementation, and an improved documentation and feedback mechanism. Following the release, our team will work with extension personnel and our early stakeholder adopters to identify and address any usability issues and add defined quality control measures to the current tools. Concurrently, the design team is working closely with the modelers, extension experts, and developers to design the next set of tools, which will include multi-scenario development and historical confidence metrics. Our third objective is the production of reliable forecasts. A major year 4 goal is the real-time production of 9-month CWRF-DSSAT climate-crop forecasts, updated every five days, which is currently scheduled for public release in February 2023. Following that, the focus of climate-crop forecasting will be on expanding the size of the forecast ensemble, improving bias correction and pre-/post-processing, and developing robust forecast reliability metrics that can be easily understood by end-users. Securing additional computational resources to carry out these objectives remains a challenge and goal. Additionally, the economic tool and hydrology models will continue to be development. The economic tool, which will translate crop yield predictions into financial terms, is planned for development and release during year four, while the hydrologic and irrigation models will continue development to ensure that they are localizable, calibrated, peer-revied, and technically ready to support tools in year 5. Finally, to accomplish the fourth objective of providing transdisciplinary education, DAWN team members will continue and improve upon many of the activities carried out in the past year. Education specialists will administer the fourth year of the DAWN summer internship program, and organize and administer professional development workshops for DAWN early-career scientists. Additionally, they will work with extension specialists to plan and implement at least one service or outreach event that will include students working on the DAWN project. DAWN's leadership will also communicate information about its objectives, products, and research findings through the publication of a Bulletin of the American Meteorological Society In-Box article about the project and the organization of a special issue devoted to themes related to DAWN in a journal yet to be chosen.
Impacts
What was accomplished under these goals?
DAWN aims to develop a dashboard of co-produced tools that address key stakeholder needs at decision-relevant scales, closing a long-standing gap between scientific advances in earth system modeling and the stakeholders who make agricultural decisions. Year 3 marked the Phase 1 public launch of the DAWN Dashboard. Additionally, we developed and implemented mechanisms for stakeholder engagement, communication, and co-production; improved the capabilities of the foundational models; and provided training and educational opportunities to students and early career scientists. Objective 1: Engage agricultural producers to collaboratively design, test and implement a dynamic evidence-based decision support system. Phase 1 of the DAWN dashboard was publicly launched in March 2023, including the Crop Progress and Growing Degree Day tools, both of which display predictive information for the upcoming season in comparison to historical trends and analog years. The launch was publicized through social media and DAWN partner institutions. The release particularly targeted early adopters interested in engaging with a beta-level product, and user feedback and responses have been continuously collected since the release and used to improve and develop the tools. DAWN continued pre-existing efforts to engage stakeholders and extension experts in providing feedback around the existing and planned tools, particularly through think-aloud interviews, demo-meetings and tool walkthroughs, and conference breakout sessions. These solicited valuable feedback both on direct design decisions, such as how users are interested in applying tools and ease-of-use, as well as broader questions of agricultural decision support, such as at what confidence thresholds farmers were likely to change their plans based on forecasts. In addition to guiding dashboard development, the ongoing conversations with stakeholders and extension experts led to evolution in DAWN priorities. DAWN was started with an initial set of advanced models chosen to capture key components of the earth's climate-agro-hydro-economic system. Stakeholder interactions revealed that some of the original plans for model-based tools required extensive site information and user input that would likely be prohibitive for adoption or use. Thus, project modeling and tool priorities were refocused away from weather-scale operations and towards seasonal-to-sub-seasonal planning. Objective 2: Build a coupled science modeling system to transfer holistic systems knowledge to effective decision support tools at stakeholder-relevant scales. DAWN web developers released the first public version of the dashboard with tools developed in year 2, while continuing to build out a suite of new tools on the internal development and staging sites. Several key changes were implemented on the backend to streamline future development and production. Based on lessons learned in the first two years, developers created a standardized input interface and refactored code for improved efficiency. Tool interfaces and usability were continuously refined based on user feedback. Additionally, the communications team built resources to accompany these public tools, including fact sheets, walk throughs, and user guides. Alongside the year 3 public releases, developers continued to build tools intended for DAWN's Phase 2 release in year 4. These include an improved online feedback system, a data viewer which will provide easy access to a wide variety of climate predictions, a dry down calculator, and a new field-based replacement for the location selector that allows users to upload shapefiles of their fields. Finally, while the Phase 1 release relied on NOAA Coupled Forecast System Model (CFSv2) predictions, the existing and upcoming tools were configured to accept the improved CWRF climate predictions, which will be rolled out publicly in the next release. Objective 3: Produce reliable forecasts to support agricultural and water managers' decisions. DAWN modelers are focused on both system development (improving model accuracy and relevance to stakeholder needs) and forecast operations (implementing workflows to produce updated forecasts to drive dashboard tools). In terms of system development, CWRF climate and DSSAT crop forecasts were improved through dynamic coupling feedback alongside advanced bias correction, machine learning, and ensemble optimization, enhancing the overall skill and confidence. Additionally, we developed high-resolution national maps of crop cultivation and crop-specific nutrient and irrigation application; these will provide baseline parameters for crop growth and yield forecasts. The bioenergy crop model BioCro was improved in predicting miscanthus, switchgrass and energycane growth/yield and validated against field measurements; this will be used to expand current and future tools to provide farmers with options in growing staple (corn, soybean) or bioenergy crops for balancing profitability and sustainability. The watershed model SWAT, which will drive water and farm management tools, was enhanced to provide improved simulation of nutrient loading, irrigation, and perennial grasses. The irrigation model SETMI was converted to open source for future integration into the dashboard, while the field water cycle model SWAP is being parameterized to work more broadly across the U.S. Corn Belt. Furthermore, based on extension feedback, our economic modelers created a simplified version of a financial assessment tool that will help translate the impacts of seasonal climate anomalies on crop yields into economic values. In terms of forecast operations, we developed a procedure to automate CWRF seasonal climate and DSSAT crop growth forecasts downscaled from NOAA operational climate forecasts at a 9-month lead time, updating every five days to produce a multiphysics ensemble prediction that will be used to drive all the dashboard tools. Generating high-resolution forecasts to support farm-level decision-making requires extensive computational resources, which are not funded by USDA and nor provided by in-house facilities. We have recieved supercomputing support from NSF- and DOE-funded high-performance computing centers, but more computing resources are needed to improve the predictive skills of DAWN's ensemble forecasts by running multiple realizations using different initial conditions and model physics configurations. Objective 4: Integrate research and extension into experiential, transdisciplinary education to teach, train and inspire future leaders in agricultural water use and nutrient management. The DAWN education team revised and taught an undergraduate global classroom focused on food, energy, water nexus-related topics: MIEH333: Every Drop Counts; Water, Food and Global Public Health (UMD). This course is an in-depth interdisciplinary study of public health issues related to water use for global food production. In Year 3, we expanded this global classroom and included a total of 51 students from UMD, Kathmandu University in Nepal, Mbeya University of Science and Technology in Tanzania and the Hebrew University of Jerusalem in Israel. The DAWN education team successfully administered the third year of the DAWN Undergraduate Summer Internship Program, including 12 interns (75% women; 7 universities and 1 private liberal arts college represented; 2 international students). During the 10-week internship, each intern participated in an orientation week; a research, extension or education project; professional development webinars; and final project presentations. In addition to the internship program, the DAWN education team also offered four professional development workshops targeted to DAWN early-career scientists.
Publications
- Type:
Journal Articles
Status:
Published
Year Published:
2022
Citation:
Clark, S., Wolfinger, J. F., Kenney, M. A., Gerst, M. D., & Roop, H. A. (2022). Assessing stakeholder climate data needs for farm-level decision-making in the US Corn Belt. EGUsphere, 1-26.
- Type:
Journal Articles
Status:
Awaiting Publication
Year Published:
2023
Citation:
Jamal A., XM Cai, X. Qiao, L. Garcia, J. Wang, A. Amori, and H. Yang (2023). Real-time irrigation scheduling based on weather forecasts, field observations, and human-machine interactions, Wat. Resour. Res., in press.
- Type:
Journal Articles
Status:
Under Review
Year Published:
2023
Citation:
He Y, Jaiswal D, Long SP, Liang X-Z, Matthews ML, Biomass yield potential and yield stability on U.S. marginal land: implications on reaching net-zero emission. GCB-Bioenergy.
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Wu, Y., Meng, S., Liu, C., Gao, W., Liang, X.-Z. (2023). A bibliometric analysis of Research for Climate Impact on Agriculture, Frontiers in Sustainable Food Systems, Vol 7, https://doi.org/10.3389/fsufs.2023.1191305.
- Type:
Journal Articles
Status:
Submitted
Year Published:
2023
Citation:
Dangol, S., Zhang, X., Liang, X.-Z., and Blanc-Betes, E. Submitted to Environmental Modelling & Software. Dec 2023. Advancing the SWAT model to simulate perennial bioenergy crops: A case study on switchgrass growth.
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Sun, C., and X.-Z. Liang, 2023: Understanding and Reducing Warm and Dry Summer Biases in the Central United States: Improving Cumulus Parameterization. Journal of Climate, 36, 20352054, doi:10.1175/JCLI-D-22-0254.1.
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Kennedy, J., Hurtt, G. C., Liang, X. Z., Chini, L., & Ma, L. (2023). Changing cropland in changing climates: quantifying two decades of global cropland changes. Environmental Research Letters, 18(6), 064010.
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Liang, Kang, Xuesong Zhang, Xin-Zhong Liang, Virginia L. Jin, Girma Birru, Marty R. Schmer, G. Philip Robertson, Gregory W. McCarty, and Glenn E. Moglen, 2023: Simulating agroecosystem soil inorganic nitrogen dynamics under long-term management with an improved SWAT-C model. Science of the Total Environment, 879, 162906. doi: 10.1016/j.scitotenv.2023.162906
- Type:
Journal Articles
Status:
Under Review
Year Published:
2023
Citation:
Liang, Xin-Zhong; Drew Gower, Jennifer A. Kennedy, Melissa Kenney, Michael C Maddox, Michael Gerst, Guillermo Balboa, Talon Becker, Ximing Cai, Roger Elmore, Wei Gao, Yufeng He, Kang Liang, Shane Lotton, Leena Malayil, Megan L. Matthews, Alison M. Meadow, Christopher Neale, Greg Newman, Amy Rebecca Sapkota, Sanghoon Shin, Jonathan Straube, Chao Sun, You Wu, Yun Yang, and Xuesong Zhang, 2023: DAWN: Dashboard for Agricultural Water use and Nutrient management - A predictive decision support system to improve crop production in a changing climate. Bulletin of the American Meteorological Society (minor revision).
|
Progress 09/01/21 to 08/31/22
Outputs
Target Audience:The DAWN project aims to reach farmers and water managers in the U.S. Corn Belt as well as agricultural information providers such as extension educators and government agencies. This past year DAWN's extension team formed and convened a Stakeholder Working Group (SWG) composed of producers, extension agents, and scientists to advise it on dashboard tool development and outreach. Farmers belonging to FamilyFarms Group, a constituent part of DAWN, have also provided the project with information about the decisions they face and feedback on early versions of the dashboard. Additionally, members of the DAWN project have reached several secondary target audiences, including climate and agricultural scientists, through lectures and presentations. Lastly, they have also engaged undergraduate students through DAWN-related coursework and a summer internship program. Changes/Problems:The DAWN Project encountered a few major and several minor challenges in its second year. The largest problem was that the project member managing DAWN's server and website had to leave the project after several months of ill health. His departure delayed the release of the development and staging versions of the dashboard as well as new tool development. Another setback occurred when a significant weather event took down the network router for the site, which greatly reduced developer productivity for a month and delayed the release of the staging dashboard. Since then, DAWN's leadership designated a new website manager who adopted an agile management system that has helped streamline development work. DAWN's backend developers have also reorganized the system and eliminated unnecessary dependencies that would have automatically failed during future system outages. In addition to the delays in the deployment of DAWN's internal dashboard, developing robust communication among modelers, web developers, and extension specialists proved to be a challenge. DAWN's external reviewer and management team identified several key gaps slowing the development and release of tools, such as developers and extension specialists being unaware of model capabilities; modelers and web developers being unaware of stakeholder needs; and extension specialists and modelers being unaware of the development timeline for individual tools. DAWN's leadership addressed this problem by instituting the creation of user stories and scheduling regular demonstration sessions for each new tool. A user story is a detailed plan for the design and intended use of a specific tool developed collaboratively by modelers, web developers, and extension specialists. In this new framework, tool development can only begin once a user story has been written and shared among the three groups. Web developers have in turn begun scheduling demonstration sessions every two weeks for members of other groups to provide feedback about the latest tool updates. These changes have put DAWN on track to publicly release its dashboard in early February What opportunities for training and professional development has the project provided?The fourth objective of the DAWN project is to provide transdisciplinary education for future leaders in agricultural water use and nutrient management. In this respect, the past year was successful. Undergraduate students, graduate students, and postdocs all assisted in the development of DAWN's dashboard tools and coupled modeling system under the mentorship of senior researchers. The majority of DAWN's web developers are University of Maryland undergraduate students who, in addition to financial compensation for their time, receive hands-on experience and training from senior project personnel. DAWN's ten-week summer internship program provided eight undergraduate students from five different institutions the opportunity to work closely with DAWN research, extension, and education personnel from June to August 2022. During this time, the interns worked on individual projects of interest and attended weekly professional and writing workshops. DAWN educators also developed curricula for and taught two undergraduate courses relevant to the science of DAWN during spring and summer 2022. Lastly, DAWN personnel held a workshop with farmers to introduce them to data and modeling tools that can be used to support real-time irrigation scheduling. How have the results been disseminated to communities of interest?Dissemination of scientific results and actionable information to agricultural stakeholders is one of DAWN's primary goals. In its second year DAWN team members increased awareness of the project and identified interested stakeholders via updates on the website, distribution of informational materials, and presence at farmer-centric workshops such as the Daugherty Water for Food Global Institute Conference in October 2021, the FamilyFarms Winter Conference and Illinois Crop Management Conference in February 2022, and the University of Nebraska South Central Agricultural Laboratory Field Day in August 2022. DAWN's Extension team also held the inaugural meeting of its Stakeholder Working Group in May 2022. The Stakeholder Working Group serves as the first external audience for DAWN's decision support tools and has provided valuable feedback to aid in their development. DAWN members also shared the results of their project-related research through academic publications, seminars, presentations, and conferences, including convening an eLightning session on "Coupled Climate-Agrohydrosystem Modeling to Support Sustainable Agricultural Production" at the 2021 American Geophysical Union Fall Meeting. DAWN's Dashboard is intended as a mechanism to disseminate relevant and usable scientific information to decision makers. The tools developed this year are currently slated to be shared with a small group of test users in the coming months, and will be rolled out more publicly in early 2023. What do you plan to do during the next reporting period to accomplish the goals?To accomplish the first objective, DAWN's extension team will continue working with stakeholders to gather information and promote the project. First, team leaders will continue to meet regularly with members of the Stakeholder Working Group to better understand the needs of producers and solicit feedback on DAWN's decision support tools. Second, DAWN's communications manager will continue to create communications products, including fact sheets, brochures, and web copy, to increase the project's visibility, outreach, and impact. Third, team members will draft and publish standards of practice for the collection, storage, and use of DAWN user data. Fourth, DAWN extension specialists will release the first version of the public dashboard at the FamilyFarms Group Winter Conference in February 2023. Lastly, DAWN developers will build APIs for several of its tools so that they can be accessible through the FieldIQ software used by members of FamilyFarms Group. To accomplish the second objective, DAWN web developers and modelers will continue to build decision support tools and to integrate them into the web-based dashboard. Developers will replace the growing-degree-day-based algorithm driving the crop progress tool with the DSSATmodel, allowing it to forecast seasonal yields and take into account rainfall, nitrogen dynamics, and fertilizer application in addition to temperature. Developers will also add a corn drydown component to the crop progress tool, making it capable of forecasting the harvest date as well the maturity date. DAWN developers also plan to build and test a new data viewer tool that will display historical and predicted values for a range of environmental variables of interest to producers and to make improvements to the field selector, weather widget, and dashboard landing page. DAWN modelers will continue to test irrigation scheduling tools in the field while developers construct appropriate user interfaces for them. Modelers will also introduce a new version of the SWAT model into the coupled modeling system that will allow it to predict nutrient use and availability. Moreover, modelers will optimize the coupled CWRF-DSSAT system to further enhance the skill of the dashboard tools in predicting corn and soybean yields. In support of the third objective, DAWN team members will improve the coupled climate-agricultural modeling system so that it can provide real-time, skillful forecasts to drive the decision support tools and data viewer. First, modelers will continue to improve the coupling between CWRF, DAWN's regional climate model, and the agrohydrosystem models (DSSAT, BiCro, SWAT) that simulate crop growth and hydrological processes. Second, modelers will develop and implement a cost-effective approach for generating superensemble forecasts (combining multiple model initialization and physics configurations) that can enhance overall predictive skills while remaining computationally affordable. Third, DAWN scientists will increase the resolution of the CWRF grid over the US Corn Belt to improve the accuracy of its predictions in that region. Fourth, modelers will begin weekly production of 16-day and 9-month climate forecasts and couple this output to existing decision support tools to provide users with the most up-to-date information. Lastly, modelers will systematically assess CWRF downscaling skills in predicting weather conditions and climate anomalies, which are critical to all DAWN decision support tools, and incorporate this skill information into dashboard tools to show information confidence and uncertainty. For all these modeling efforts, the team will write more proposals to secure the necessary supercomputing resources. To accomplish its fourth objective, DAWN team members will continue and improve upon many of the activities carried out in the past year. Education specialists will administer the third year of the DAWN summer internship program, teach one undergraduate course that is in line with DAWN focus areas, and organize and administer professional development workshops for DAWN early-career scientists. Additionally, they will work with extension specialists to plan and implement at least one service or outreach event that will include students working on the DAWN project. DAWN's leadership will also communicate information about its objectives, products, and research findings through the publication of a Bulletin of the American Meteorological Society In-Box article about the project and the organization of a special issue devoted to themes related to DAWN in a journal yet to be chosen
Impacts
What was accomplished under these goals?
The DAWN project's fundamental goal is to close a long-standing gap between scientific advances in earth system modeling and the community of stakeholders who make agricultural decisions. DAWN's Dashboard is intended to consist of a collection of tools co-produced by scientists, extension experts, and web developers to address key stakeholder needs at decision-relevant scales. To that end, our primary Year 2 accomplishments include: developing and implementing mechanisms for stakeholder engagement, communication, and co-production; creating the underlying web infrastructure for the Dashboard and developing processes for the creation of tools centered around user stories; improving the capabilities and better quantifying the uncertainty of the models around which the Dashboard is being built; and providing training and educational opportunities to students and early career scientists. Objective 1 DAWN's leadership formally established a stakeholder working group that includes seven stakeholders within the Corn Belt region, representing farmers, agronomists, and government officials. The stakeholder working group will be the first audience for and provide feedback and guidance on DAWN decision support tools. In another effort to shape the decision support tools, DAWN extension experts participated in think-aloud interviews on the initial DAWN growing degree day and crop progress tool to (i) to capture extension team members' user experience, (ii) elicit design feedback and (iii) test interview protocols that will be modified for end user engagement. The interviews identified tool features that were difficult to navigate or not intuitive given how farmers wish to use this information in decision-making processes, as well as additional features that would improve effectiveness. To better understand stakeholder climate information needs and priorities in the U.S. Corn Belt, DAWN extension specialists additionally conducted an analysis of 50 documents. This analysis found that stakeholders are primarily concerned with practical and tactical decision-making, including from whom they receive their information, the application of information to agricultural, water, and risk management, and desired economic outcomes, and are less concerned with social and environmental issues, and long-term climate resilience. DAWN extension specialists also reviewed the literature on decision calendars for agriculture and water resources to provide recommendations for tool development, features, and key decisions for the DAWN dashboard. The review found that many key decisions in a cropping year are made during the preceding fall and winter, demonstrating the need for tools that can leverage seasonal-scale climate forecasts. Objective 2 DAWN web developers built and released two internal versions of the DAWN dashboard: a development version for visualizing changes and a staging version to share updates with other members of the team or selected stakeholders. DAWN's developers also rebuilt the site's backend architecture to improve its reliability, made improvements to the existing DAWN growing degree day tool, and built a new crop progress tool, a weather widget, and a field selector, all of which will be part of the first public version of the dashboard. To facilitate co-production from the model-development stage, DAWN leadership instituted a new software design approach - user stories - to clearly identify users, features, and decision-making use cases before physical tool development and to centerthe needs of agriculturalists. Additionally, several improvements were made to the underlying physical models to improve key capabilities such as crop development and nutrient management. DAWN modelers at the University of Illinois Urbana Champaign added a multilayer bucket model incorporating downflow, saturation flow, upflow, runoff, and tile drainage to the BioCro crop growth model, which will be incorporated into DAWN's dashboard. Another team of modelers at the same institution developed a user interface and data assimilation algorithm for RTIST, an irrigation scheduling model that will also be part of the dashboard, and introduced it to Nebraska farmers at a workshop where they solicited feedback about its usefulness and functionality. Lastly, a modeling team based at the University of Maryland evaluated the SWAT-C model for simulating long-term soil inorganic nitrogen at multiple US Midwest agricultural sites with various management practices. They anticipate that the SWAT-C model can serve as an effective tool to assess the availability of soil inorganic nitrogen for plant growth in different cropping systems, thereby providing useful information to support decision-making related to fertilizer application and water quality management. Objective 3 DAWN climate modelers completed ensemble seasonal hindcasts for spring and summer over the period 2012 to 2021 and produced seasonal climate forecasts for spring and summer 2022 and 2-week weather forecasts for the 2019 growing season. The modelers also ran coupled CWRF-DSSAT historical simulations of corn yield and tested CWRF driven DSSAT model configurations for soybean growth. Two journal articles with the results of these simulations have been submitted, with additional manuscripts in preparation. We identified April 1st as the optimal initialization date to maximize prediction skill for the growing season, and demonstrated that CWRF downscaling significantly enhances the NOAA Climate Forecast System forecasts. A more systematic skill assessment is under way, which represents a first step towards the inclusion of forecast reliability and uncertainty estimates in the DAWN dashboard. Generating high-resolution forecasts to support decision-making requires extensive computational resources, which are not funded by USDA and nor provided by in-house facilities. The team has written several successful proposals for supercomputing support from NSF- and DOE-funded computing centers, but more computing resources are needed to further improve the predictive skills of DAWN's ensemble forecasts by running multiple realizations using different initial conditions and model physics configurations. Objective 4 DAWN education specialists revised and taught two undergraduate courses focused on food, energy, water nexus-related topics: 1) Crop Sciences 113; Environment, Agriculture, and Society (University of Illinois Urbana-Champaign (UIUC); and 2) MIEH333: Every Drop Counts; Water, Food and Global Public Health (University of Maryland College Park (UMD)). The UIUC course demonstrates the importance of agriculture and the environment to our society and delves into how each of the three influence each other. The UMD course is an in-depth interdisciplinary study of the public health issues related to water use for global food production. In Year 2, we transitioned the course into a "global classroom" course in collaboration with Kathmandu University in Nepal. A total of 35 students (16 undergraduate students from University of Maryland and 19 from Kathmandu University, Nepal) participated. In addition to coursework, DAWN education specialists successfully administered the second year of the DAWN summer internship program. The 8 interns represented 5 different universities and community colleges (including 1 international university) and 66.67% of the interns were women. Throughout the 10-week internship, each intern participated in a research, extension or education project; weekly professional development workshops; weekly journal clubs/seminars; and a professional writing workshop held over 5 sessions that focused on how to write a technical manuscript. Finally, DAWN representatives presented information about the project at the University of Maryland "Maryland Day" where tens of thousands of Marylanders visit the campus to learn about UMD research and educational activities.
Publications
- Type:
Conference Papers and Presentations
Status:
Other
Year Published:
2022
Citation:
Liang, X.-Z.: Improving Agricultural Water Use and Nutrient Management to Sustain Food and Energy Crops Production in the Corn Belt. Invited seminar at the NASA Goddard Space Flight Center, June 23, 2022.
- Type:
Conference Papers and Presentations
Status:
Other
Year Published:
2022
Citation:
Liang, X.-Z.: Regional Earth System Modeling to Support Sustainable Agricultural Production. Invited seminar at the School of Integrative Plant Science, Cornell University, Ithaca, NY, September 1, 2022.
- Type:
Conference Papers and Presentations
Status:
Other
Year Published:
2022
Citation:
Liang, K., X. Zhang, and X.-Z. Liang: Simulating Long-Term Soil Inorganic Nitrogen As Affected By Agricultural Management Practices in the US Midwest. ASA, CSSA, SSSA International Annual Meeting, Baltimore, MD, November 8, 2022.
- Type:
Journal Articles
Status:
Awaiting Publication
Year Published:
2022
Citation:
Sun, C., and X.-Z. Liang, 2022: Understanding and Reducing Warm and Dry Summer Biases in the Central United States: Analytical Modeling to Identify the Mechanisms for CMIP Ensemble Error Spread. Journal of Climate (in press).
- Type:
Journal Articles
Status:
Under Review
Year Published:
2022
Citation:
Clark, S., J.F. Wolfinger, M.A. Kenney, M.D. Gerst, and H.A. Roop, 2022: Assessing stakeholder climate data needs for farm-level decision-making in the U.S. Corn Belt. Geoscience Communication (in revision).
- Type:
Journal Articles
Status:
Under Review
Year Published:
2022
Citation:
Jamal, A., X. Cai, X. Qiao, L. Garcia, J. Wang, A. Amori, H. Yang, 2022: Real-time irrigation scheduling based on weather forecasts, field observations, and human-machine interactions. Water Resources Research (under review).
- Type:
Conference Papers and Presentations
Status:
Other
Year Published:
2022
Citation:
Liang, X.-Z.: Improving Agricultural Water Use and Nutrient Management to Sustain Food and Energy Crops Production in the Corn Belt. USDA NIFA Sustainable Agricultural Systems Project Directors Meeting, Kansas City, MO, April 18-20, 2022.
- Type:
Journal Articles
Status:
Under Review
Year Published:
2022
Citation:
Sun, C., and X.-Z. Liang, 2022: Understanding and Reducing Warm and Dry Summer Biases in the Central United States: Improving Cumulus Parameterization. Journal of Climate (revision under review).
|
Progress 09/01/20 to 08/31/21
Outputs
Target Audience:The DAWN project aims to reach farmers and water managers in the U.S. Corn Belt as well as agricultural information providers, such as extension educators and government agencies. Outreach to these groups will expand over the course of the project; the early years are focused on the development of a small working group of stakeholders, while rollout of the products to a wider audience will be a primary aim in later years. DAWN outreach efforts have been somewhat hindered by the COVID-19 pandemic, but the working group is close to deployment, and efforts to engage a wider audience are increasingly taking place as in-person agricultural events resume. Additionally, members of the DAWN project have reached several secondary target audiences, including climate and agricultural scientists, through lectures and presentations. They have also engaged undergraduate students through DAWN-related coursework and a summer internship program. Changes/Problems:The COVID-19 pandemic has presented significant challenges for the DAWN project in its first year. The home institutions of many DAWN CoPDs have severely restricted in-person meetings and instituted temporary hiring freezes. As a result of these policies, DAWN extension and education specialists were unable to hold meetings with stakeholders and other outreach events, while some CoPDs faced delays in hiring researchers to develop the coupled modeling system and decision-support tools. Additionally, DAWN team members were unable to assemble in-person for an annual workshop and instead conducted one virtually through Zoom. Such setbacks, however significant, have not changed the DAWN project's overall goals or approach and, as conditions improve, will be redressed in its second year. In addition, the project leadership made a few changes to the DAWN project in response to several smaller problems. First, the CoPDs determined that the project would benefit from a communications specialist, a position which was not included in the original budget. The budget has since been modified to hire a half-time specialist starting in the second year, reducing the support available for one graduate student. Second, to develop the project dashboard, the project director chose to hire a group of computer science students and a part-time IT specialist in place of a full-time software engineer. Third, a few scientists departed from DAWN for new positions, leaving temporary gaps that were filled after slight delays caused by the pandemic. Fourth, shipping delays and personnel absences pushed back the assembly of the project's data server and the launch of the project website by about six months. Lastly, the project director applied for and received allocations for supercomputing resources that are necessary for the development of DAWN's modeling system but that were not available locally or through USDA. What opportunities for training and professional development has the project provided?The fourth objective of the DAWN project is to provide transdisciplinary education for future leaders in agricultural water use and nutrient management. In this respect, the past year was successful. Undergraduate students, graduate students, and postdocs all assisted in the development of DAWN's dashboard tools and coupled modeling system under the mentorship of senior researchers. DAWN's eight-week summer internship program provided nine undergraduate students from six different institutions the opportunity to work closely with DAWN research, extension, and education personnel from June to August 2021. During this time, the interns worked on individual projects of interest and attended weekly professional and writing workshops. DAWN team members also co-organized a week-long professional development workshop for graduate students and early career professionals in August 2021. Lastly, DAWN educators developed curricula for and taught two undergraduate courses relevant to the science of DAWN during spring and summer 2021. How have the results been disseminated to communities of interest?Dissemination of scientific results and actionable information to agricultural stakeholders is DAWN's primary goal. Year one of the project was devoted to creating the decision-support tools that will be made available to farmers and water managers in the future. Simultaneously, DAWN team members increased awareness of the project and identified interested stakeholders via the development of the website, creation of informational materials, and presence at farmer-centric workshops such as the FamilyFarms LLC Summer Conference in July 2021. The DAWN leadership team has also explored new applications for its tools through collaboration with other national projects. These efforts include meeting with the USDA Honeybee and Pollinator Research team, presenting a lightning talk at the NSF-sponsored Convergence Accelerator Conference for Food Security in Extreme Environments and Food Deserts, and visiting USDA-ARS Long-Term Agricultural Research (LTAR) sites in the Choptank River Watershed. Finally, DAWN is ready to launch the Stakeholder Working Group in January 2022, which will provide active guidance and feedback on tool development and ease of use. What do you plan to do during the next reporting period to accomplish the goals?To accomplish the first objective, the DAWN team has identified a number of tasks related to stakeholder outreach and technological development. First, DAWN's extension team has identified a group of farmers, extension specialists, and managers with expertise and potential interest in collaborating on DAWN development. The team is also planning to integrate DAWN into existing extension programs and tools and to seek advice on DAWN activities and products from a formal stakeholder working group. Second, DAWN's new communications manager will refine and expand the communications strategy, and create communications products, including fact sheets, brochures, and web copy, to increase the project's visibility, outreach, and impact. Third, team members will draft and publish standards of practice for the collection, storage, and use of DAWN user data. Fourth, DAWN developers will release the first version of the public dashboard along with crop progress prediction and irrigation scheduling tools. Lastly, DAWN extension outreach specialists will develop agricultural decision calendars that they can use to schedule meetings with farmers and extension agents to demonstrate DAWN's tools. To accomplish the second objective, DAWN modelers will continue to develop the next generation of tools that will be integrated into its dashboard. The irrigation scheduling tools that have already been built will be tested in the field while DAWN web developers construct appropriate user interfaces for them. The modelers will also expand the number of crops whose yields can be predicted accurately and will couple the existing crop models with economic simulations that can forecast agricultural prices. Collaboration between the modeling groups will be enhanced (e.g., coupling the crop yield prediction model with the irrigation scheduling tool to explore optimized scheduling; using the crop area projected from the economic model to estimate irrigation water requirements). Lastly, DAWN personnel will collaborate with the ARS-Hydrology and Remote Sensing Laboratory to simulate nutrient flows in the Choptank River Watershed. These simulations will assist modelers in the future development of a nutrient-availability forecasting tool. In support of the third objective, DAWN team members will improve the coupled climate-agricultural modeling system so that it can provide real-time forecasts to drive the dashboard tools. First, modelers will continue to compare seasonal climate and crop yield hindcasts to historical records in order to develop better bias-correction algorithms. Second, team members will construct machine learning models of crop yields to improve prediction accuracy earlier in the growing season. Lastly, modelers will begin weekly production of 9-month climate forecasts and couple this output to existing crop progress and irrigation scheduling tools. To accomplish its fourth objective, DAWN team members will continue and improve upon many of the activities carried out in the past year. Education specialists will administer the second year of the DAWN summer internship program, teach two undergraduate courses that are in line with DAWN focus areas, and organize and administer professional development workshops for DAWN early-career scientists. Additionally, they will work with extension specialists to plan and implement at least one service/outreach event at each DAWN-participating institution. Finally, education specialists will explore how components of UMD's graduate MIEH 690 course can be scaled to other DAWN institutions.
Impacts
What was accomplished under these goals?
DAWN aims to transform the ability of agriculturalists and water resource managers to make timely and sustainable decisions that will reduce water use and nutrient pollution while maintaining profitable and productive operations. In the first year of the project, DAWN extension specialists identified and reached out to stakeholders, who will provide guidance and feedback on its products as part of a stakeholder working group. Likewise, DAWN modelers built and tested the climate modeling system that will produce its forecasts. They also designed several tools that will use these forecasts to aid in agricultural decision making. DAWN's activities also benefit future scientists and policy makers in the fields of climate, water resources, and agriculture. For example, DAWN education specialists developed and taught two courses in Spring and Summer 2021 that were based on the project's themes and carried out an internship program for students interested in the food-energy-water nexus. Additionally, DAWN's modeling system and information-delivery architecture will be made permanently available to the research community and USDA. To this end, DAWN web developers built and launched its project website and began to construct the dashboard that will eventually allow public access to its decision-support tools. Objective 1: Proactively engage agricultural producers and water managers to collaboratively design, test and implement a dynamic evidence-based decision support system and a web-based information delivery system for effective use by diverse stakeholder communities. DAWN's extension team has engaged their existing constituency groups to share the launch of the DAWN project, introduce them to DAWN, and to start setting up structures to integrate DAWN into existing extension programming and to solicit input on initial DAWN priorities and tools. The extension team has also identified an invited group of farmers, extension specialists, and managers who will form the inaugural DAWN Stakeholder Working Group. Simultaneously, DAWN web developers built and launched DAWN's website on a 500TB server cluster purchased specifically to support the project. These developers are continuously updating the website and are currently using publicly-available auditing tools to fix formatting issues that may limit the site's accessibility to visually-impaired audiences. The website will serve to introduce the DAWN project to stakeholders, share relevant news, and provide background information on its decision-support tools. The site will also give users access to DAWN's decision-support dashboard through a password-protected login page. Although the dashboard is currently unavailable to the public, DAWN developers have built a site-based development environment that is accessible to members of the project. This environment allows members of the DAWN team to demonstrate and test tool functionality in anticipation of a public release. Objective 2: Build a coupled science modeling system not only to represent interactions among climate, agriculture, land/water use, and economic/ environmental impacts, but also to transfer this holistic systems knowledge to effective decision support tools at stakeholder-relevant scales. DAWN modelers began development of several decision-support tools that will be integrated into the project dashboard. A tool for forecasting seasonal growing degree days and predicting crop maturity dates was built and tested. The tool uses historical and forecasted temperature data so that farmers can compare the current growing season to those in the past as well as the expected maturity dates for multiple cultivars of corn and soybean. Currently, the tool is being reviewed by DAWN extension specialists, beginning the iterative process through which it will be updated to improve usability and responsiveness to stakeholder needs. In addition, two tools that will support irrigation decision-making were also built and are undergoing field testing. One tool uses remote sensing products to estimate crop stress while the other simulates soil moisture evolution at the field scale. Other models predicting water quality, nutrient discharge, crop yields, and crop prices are in development, with completion projected in 2022.The modeling teams have been interacting with the extension and education teams for modeling development requirements from the user's perspective. Objective 3: Produce reliable forecasts to support agricultural and water managers' real-time and mid-range decisions as they navigate resource use, manage risks, and develop strategies to optimize crop productivity, profitability, and environmental quality. First, DAWN scientists have collected and processed the disparate data sources necessary for the initialization, forcing, and evaluation of DAWN forecasts, which include NOAA climate observations and operational forecasts, USDA crop production and management data, and NASA ecosystem and water resources remote-sensing products. Second, the core component of the DAWN prediction system, CWRF, has been improved to significantly reduce the long-standing warm-and-dry biases in the central U.S., thus improving forecasts for the Corn Belt. Third, DAWN climate modelers have started to produce seasonal hindcasts at a 15-km resolution using the coupled climate-agricultural modeling system. These forecasts are currently being evaluated against historical data to calibrate and reduce bias in the system before beginning forecast production for DAWN decision support. Preliminary comparisons indicated that CWRF downscaling significantly improved NOAA summer forecasts at a lead time of two months. Fourth, the project director of DAWN, Dr. Xin-Zhong Liang, was awarded a supercomputing allocation of 3.2 million SUs and 130 Tb storage valued at $22,609 through the Extreme Science and Engineering Discovery Environment, an NSF-funded institution. Dr. Liang also won an allocation of 1.5 million SUs and 1PB storage from the Center for Western Weather and Water Extremes. These resources will all be used to further develop DAWN's coupled modeling system. Objective 4: Integrate research and extension into experiential, transdisciplinary education to teach, train and inspire future leaders in agricultural water use and nutrient management. DAWN education specialists developed two undergraduate courses focused on topics related to the food-energy-water nexus: 1) Crop Sciences 113; Environment, Agriculture, and Society (University of Illinois Urbana-Champaign (UIUC); and 2) MD Inst. For Applied Env. Health 333: Every Drop Counts; Water, Food, and Global Public Health (University of Maryland College Park (UMD)). The UIUC course (61 students in Year 1) demonstrates the importance of agriculture and the environment to our society and delves into how each of the three influence one another. The UMD course (4 students in Year 1) is an in-depth interdisciplinary study of the public health issues related to water use for global food production. DAWN team members also began determining how the graduate MIEH 690 course could be scaled to other DAWN institutions. In addition to developing coursework, DAWN education specialists successfully launched the DAWN summer internship program. Thirty applications were received and 9 were awarded DAWN internships. During the 8-week program each intern participated in a research, extension, or education project; weekly professional development workshops; weekly journal clubs/seminars; and a 5-session professional writing workshop that focused on how to write a technical manuscript. Examples of internship projects include: 1) stakeholder scientific needs assessments; 2) analyzing satellite imagery and evapotranspiration data in support of the Decision Support System team; and 3) developing modules that use observations and climate model output to predict environmental and agricultural outcomes.
Publications
- Type:
Conference Papers and Presentations
Status:
Other
Year Published:
2020
Citation:
Liang, X.-Z: Food-Energy-Water Nexus: Food-Energy-Water Nexus: Observational and Modeling Data Analysis. Invited lecture (1.5 hours) at Maryland Institute for Applied Environmental Health for the select fellows of the NSF Research Traineeship Program ⿿ UMD Global STEWARDS: Project-Based Data Practicum at the Nexus of Food, Energy, and Water Systems (FEWS), University of Maryland, Maryland, October 21, 2020.
- Type:
Conference Papers and Presentations
Status:
Other
Year Published:
2020
Citation:
Liang, X.-Z: Combining Model Predictions with Satellite Data to Improve Agricultural Water Use and Nutrient Management. Invited talk at the Smart Farming and Natural Resource Management Enabled by Remotely Sensed Big Data, the AGU (American Geophysical Union) Fall Meeting (online), December 13-17, 2020.
- Type:
Conference Papers and Presentations
Status:
Other
Year Published:
2021
Citation:
Liang, X.-Z: Food-Energy-Water Nexus: Food-Energy-Water Nexus: Earth Modeling System and Decision Support. Invited lecture (1.5 hours) at Maryland Institute for Applied Environmental Health for the select fellows of the NSF Research Traineeship Program ⿿ UMD Global STEWARDS: Seminal Findings, and Research and Policy in Progress at the Food, Energy, Water Nexus, University of Maryland, Maryland, February 4.
- Type:
Conference Papers and Presentations
Status:
Other
Year Published:
2021
Citation:
Liang, X.-Z, Y. He, D. Jaiswal, C. Sun, S.P. Long: Growing perennial biomass crops on marginal land improves agricultural productivity and regional climate. Lightning talk at the NSF-sponsored Convergence Accelerator Conference for Food Security in Extreme Environments and Food Deserts, University of Texas at El Paso, May 19-21, 2021.
|
|