Progress 09/01/22 to 08/31/23
Outputs
Target Audience:The audience of this project continues to be, but is not limited to, growers, farm advisors, extension specialists, and scientists with expertise in soil, crop, irrigation, and pest management, commodity groups, non-profit organizations, water districts, agricultural technology companies, and young scientists and students interested in agricultural data science careers. Audiences such as agricultural consultants, growers, and scientists were reached through conference presentations, scientific (technical and non-technical) publications, and internet media (e.g., https://ai4sa.ucr.edu/; https://twitter.com/ucr_ai4sa), and some press coverage. The education portion of the project aims to reach undergraduate students who major in environmental science, engineering, or related fields, and who are interested in learning more about digital agronomy as well as in gaining hands-on experience in the field. Students recruited have had majors in: statistics, mathematics, data science, environmental science, mathematics, and other. The education team utilizes campus communications to reach additional students and faculty, campus events to participate in, and creates or identifies networking opportunities to participate in. Project Co-PIs work with postdoctoral scholars funded by this grant. The extension portion of this project worked with American growers in the Southwest and regional extension personnel. Co-PI Bali and Co-PI Cahn conducted demonstrations for growers in California, demonstrating the technology, research, and resources like CropManage. Workshops reached growers, extension personnel, crop advisors, industry personnel, government officials, and journalists for agriculture-related publications. Changes/Problems:The project was set back by delays resulting from the COVID-19 pandemic during year one, which has resulted in an overall pushed-back timeline. The theft of research equipment during year three, shared by the PD Scudiero and Co-PD Anderson groups, resulted in fieldwork delays that will be rescheduled moving into year four. The postdoctoral scholar hiring process has also resulted in research delays. There have been challenges with the recruitment of undergraduate students to participate in the summer Digital Agriculture Fellowship program located at UC Riverside. Project members will determine the possibility of reallocating educational funds and restructuring the fellowship program to reflect this unforeseen difficulty. The Digital Agriculture Fellowship program continues to adjust based on feedback expressed by the participating students and faculty. Post-survey results showed a decrease in plans of pursuing a career in digital agriculture following completion of the program, with a reported increase in understanding of digital agriculture. The education team will focus on identifying additional qualitative factors that might contribute to this, beginning with the recruitment process of the students for the next Cohort, in collaboration with the evaluation/assessment team. Planned opportunities for hands-on research and networking will continue to be included, since these are areas reported to be of interest. The DAF also has had changes in the program itself due to delays in the student recruitment process; this resulted in Cohort II students concluding their fellowship experience with the summer research program, which was a change in the anticipated schedule. Extension project members will consider the feedback from the first two multi-state event survey evaluation/assessment reports when planning the next regional or multi-state events. Considerations include the possibility of adding smaller-scale farming topics, applied artificial intelligence topics, additional demonstrations, and specific topic areas depending on region. Additionally, to keep track of the stakeholders who are participating at multi-state outreach events, a streamlined method of collecting audience reach and audience participation information will need to be developed. What opportunities for training and professional development has the project provided?The education component of the project includes the Digital Agriculture Fellowship, which is intended to support the education and career paths of each participating undergraduate student. Digital Agriculture Fellowship Coordinator Noel Salunga organizes the DAF program in collaboration with the PD and Co-PD Connie Nugent, organizes the UC Riverside Research in Science and Engineering (RISE) summer program, and serves as the primary contact person for feedback from students, faculty, and the evaluation/assessment team. Organized activities during year three included three faculty/facility lab tours, Western Plant Health Networking night, College of Natural Agricultural Sciences Research and Engagement fair (table event with 250 peer guests); UC Santa Barbara Evaluation Presentation; Entrepreneurial Workshop; Industry Panel Workshop; poster presentations at the campus Precision Agriculture Workshop, Wonderful Company Talk (three students were selected to share their research with the CEO Wonderful Company Citrus), and the RISE summer hands-on research experience which includes 10 weeks of instructional activities and a final symposium. Postdoctoral scholars and student researchers supported by the project have conducted research, presented findings, attended workshops, attended meetings and conferences, co-authored papers, and have been encouraged to collaborate with other research groups on the project. Postdoctoral scholars have supported the Digital Agriculture Fellowship students and supported their work as well, acting as additional mentors. How have the results been disseminated to communities of interest?Results have been disseminated to communities of interest by using Twitter for updates regarding recently published papers and by participating in events/workshops. Multiple Co-PIs maintain updated personal web pages with a bibliography of their work. Through Twitter, affiliated websites (e.g., campus or department), and workshops the goal is to reach growers, academics, scientists, students, advisors, and technical personnel. Press coverage ideally supports the research reaching a local public audience. Student-led research has been disseminated through participation in the College of Natural and Agricultural Sciences RISE Symposium event, which was promoted using campus communications; about 300 guests attended, with about 111 of the guests being student peers. Additionally, the project's research groups participated in conferences to present their research findings, including attending the 2023 February California Plant and Soil Conference. The extension workshops hosted by Co-PI Bali utilize surveys developed by the UC Santa Barbara Evaluation and Assessment group led by Tarek Azzam and Natalie Jones. These surveys collect feedback from audience members to better determine what research topics would be most useful for the presentations to focus on. Project CropManage, Irrigation, and Nitrogen Management outreach activities include 28 workshops and presentations since 10/18/2022. These events took place regionally in California, as well as one event in Arizona and one in Colorado. 22 events were in person, while six events were conducted via Zoom. In total, over 700 participants have attended these events. Outreach Activities highlights by date include: 9/9/22 - Annual Alfalfa and Forage Field Day in Parlier, California - Co-PD Bali presented a talk called "Winter Flooding and Summer Deficit Irrigation of Alfalfa." 11/14/22 - World Alfalfa Congress in San Diego, California - Co-PD Bali moderated as well as presented the talk "Advances in Surface Irrigation Management in Alfalfa." 12/13/22 - Salinity Management in Pistachio in Kearney, California - Co-PD Bali presented "Effects of Soil Texture on Irrigating Pistachios in Saline Conditions" and Co-PI Cahn presented "Calculating Leaching Fractions and Requirements" along with a hands-on portion. 1/18/2023 - CropManage Hands-On Workshop - Co-PD Cahn organized a hands-on training workshop in Watsonville, California with 32 attendees. 2/15/2023 - CropManage Hands-On Workshop - Co-PD Cahn organized a hands-on training workshop in Modesto, California with 25 attendees. 2/21/23 - 2/24/23 - Visits to Agricultural Water Conservation Projects in Imperial Valley and other California portions of the Colorado River Basin - Co-PD Khaled Bali, Co-PD Charles Sanchez, and PD Elia Scudiero met with stakeholder groups, irrigation districts, and staffers from the offices of Senator Dianne Fienstein and Senator Mark Kelly, and the Bureau of Reclamation, DC. 2/23/23 - Using Planet Satellite Imagery in ArcGIS Workshop in Riverside, California - This UC Riverside campus event was organized by Co-PD Hoori Ajami, UCR Library, ESRI, and Planet. PD Elia Scudiero and Co-PD Ahmed Eldawy also supported the event. 13 guests attended including students and one faculty member. Participants received hands-on training with Planet imagery and a suite of ArcGIS products. 3/29/2023 - CropManage Hands-On Workshop - Co-PD Cahn organized a hands-on training workshop in San Martin, California with 33 attendees. 4/4/23 - Sustainable Agriculture with Artificial Intelligence Extension workshop in Maricopa, Arizona - This event was a project-organized event, led by Co-PD Khaled Bali and Co-PD Charles Sanchez. Speakers included multiple project Co-PDs, and this was the first multi-state workshop organized as part of Supporting Objective Three (SO3). 4/5/23 - Extension Public Seminar Sustainable Agriculture with Artificial Intelligence in Loma, Colorado - This event was a project-organized event, led by Co-PD Khlaed Bali and Co-PD Raj Khosla's group. Speakers included multiple project Co-PDs, and this was part of the multi-state workshops organized as part of SO3. 4/19/2023 - CropManage Hands-On Workshop - Co-PD Cahn organized a hands-on training workshop in Parlier, California with 30 attendees. 5/22/23 - Precision Agriculture Workshop in Riverside, California - This UC Riverside campus event was co-organized by PD Elia Scudiero, with the goal of promoting networking between campus precision agriculture researchers. Faculty, students, and industry partners participated. The event reached UC Riverside faculty, administrators, and students primarily in the College of Natural and Agricultural Sciences, School of Policy and Business, and the College of Engineering. There were about 60-70 guests throughout this all-day event. NEWS & PRESS 11/10/22 - "2022 FREP/WPH Nutrient Management Conference Recap" located on CA.gov's Fertilizer Research and Education Program blog - Emad Jahanzad provides a summary of the conference held in October, including a Kearney Ag. Research and Extension Center facility tour led by Co-PI Khaled Bali. 5/17/23 - "Efforts to Recharge California's Underground Aquifers Shows Mixed Results" located on npr.org - Nathan Rott transcribes an interview with Co-PD Khaled Bali, describing the effects of rainfall on aquifers. 6/12/23 - "Digital Ag. Fellowship Opens Doors to the Future" located on UC Riverside's online College of Natural and Agricultural Sciences blog - PD Elia Scudiero describes the Digital Ag. Fellowship opportunity and how UCR students can participate. Products (models, instruments, audio/visual, curricula, data, databases, collab): https://cropmanage.ucanr.edu/ CropManage is a free online crop management decision support tool managed by Co-PI Michael Cahn as part of the University of California Cooperative Extension. https://ai4sa.ucr.edu The AI4SA project-based website was created using the management system maintained by UC Riverside. Planet Labs, Inc. collaboration: AI4SA has collaborated with Planet Labs, Inc. to extend the use of the product license to all UCR personnel with a UCR email address through the year 2025. 77 total campus personnel have accessed the license to create a general user Planet account since year one, with 21 participants joining in year three. Of the 21 newly added participants in year three, three total are affiliated with the AI4SA project and four total are campus faculty members. What do you plan to do during the next reporting period to accomplish the goals?During year three, internal management adjusted to streamline communication chains and improve collaboration. Moving into year four, this process will continue in order to align with the increased communication needs required for completing supporting objectives. The extension groups will work towards continuing planning outreach workshops and events, with continued effort in establishing a multi-state network. Using the feedback from survey data and the evaluation/assessment reports, topics will be adjusted to match regional interests. Project members will increase specific stakeholder attendance tracking with a goal of streamlining this process for the project's needs. Co-PD Cahn will investigate interfacing OpenET with CropManage (ETo, ETz), improving user experience, adding a model to estimate N mineralization from organic amendments and fertilizers, and add leaching calculator for salinity management in collaboration with project members Co-PD Eldawy and Co-PD Vellidis' group will add new states as the data becomes available, they will increase the maximum area to get the soil statistics data, and they will use beta testers' feedback for bug fixes and improvements. The education portion of the project will continue as a fellowship program that supports each student's education and career paths. The education group will continue in the recruitment of eligible students to participate in the Digital Agriculture Fellowship program, with an updated emphasis on recruiting students with expressed interest in precision agriculture. The outcomes of this will be tracked with surveys in collaboration with the Evaluation and Assessment team. The program itself will continue to be adjusted based on student feedback described in the project's evaluation/assessment reports. Students will participate in planned tours, conferences, hands-on research, and networking events; students indicate via survey that networking with peers and faculty is an area of interest.
Impacts
What was accomplished under these goals?
In its third year, the project has demonstrated remarkable progress in the use of AI-driven agricultural management and precision farming. Breakthroughs in remotely sensed imagery and AI have enabled more precise irrigation and pest detection strategies, particularly in specialty crops. The implementation of advanced models like BAITSSS and the integration with novel data streams such as Hydrosat and Planet have significantly enhanced early season irrigation detection and soil moisture analysis. Field experiments have shown the efficacy of multispectral drones in disease and weed detection in crops like cantaloupe. Collaborative efforts have led to the improvement of soil class maps through Polaris Version 2, aiding in the more accurate prediction of surface temperatures. The project's extension efforts have substantially influenced water management in California, while educational initiatives under the Digital Agriculture Fellowship have fostered a new cadre of data-savvy agricultural professionals. Overall, Year 3 has consolidated the project's foundation in integrating AI and remote sensing with practical agricultural applications, paving the way for more efficient and sustainable farming practices in the U.S. Southwest. Research Research activities in Year 3 faced challenges, including equipment theft, but made notable progress in AI and remote sensing for agriculture. Co-PDs Anderson and French (SO1) continued satellite remote sensing work, focusing on the BAITSSS model for specialty crops and collaborating with Hydrosat for high-resolution imagery. This work enhanced early season irrigation detection using daily Planet data for vegetable crops, a significant advancement over traditional multispectral satellite sources. The team is processing evaporative fraction data for validating the evapotranspiration model and plans to integrate BAITSSS with improved data streams, including POLARIS and PLANET, and Hydrosat for irrigation assessment. Co-PD Chaney advanced Polaris to Version 2, improving soil class maps and surface temperature predictiveness. The team plans to merge Polaris v2 properties with in-situ soil data. Chaney works with PhD student Emma Xu on this aspect. The Khosla group's fieldwork in Colorado involved extensive data collection, including leaf tissue nitrogen and soil readings. They used Planet satellite data to estimate plant traits and applied different nitrogen treatments to enhance algorithm training. The group is also focusing on soil moisture data collection and plans to expand test sites. Co-PD Papalexakis's group tackled AI challenges in crop data layer prediction and self-supervised learning, grappling with issues like class imbalance and lack of deep learning datasets. They are investigating the application of the Segment Everything (SAM) model for crop region segmentation and preparing research papers in collaboration with Co-PI Eldawy's lab. Sanchez and French expanded a salt and water-tracking database, addressing nitrogen usage efficiency and the lack of variable rate application technology among stakeholders. Co-PD McGiffen's group experimented with multispectral drones for disease and weed detection in cantaloupe, achieving high accuracy. PD Scudiero and Co-PD Skaggs contributed to soil surveys and moisture sensing for precision input management. They investigated soil-plant relationships using Planet scoping and developed models for mapping available water capacity, using comprehensive soil moisture data from sensors. Overall, Year 3 saw significant advancements in integrating AI and remote sensing with practical agricultural practices, overcoming challenges and setting the stage for future innovations in sustainable farming. Extension In SO3, Co-PI Bali focused on groundwater management in California's Central Valley, working on reducing demand, enhancing irrigation efficiency, and promoting groundwater recharge. His work extended to deficit irrigation in the Lower Colorado River Basin and improving irrigation and nitrogen management in diverse regions. Bali also engaged in subsurface drip irrigation on alfalfa, incorporating new technologies and collaborating with growers to optimize irrigation and crop management. He continues collecting survey data to refine workshop content, with the surveys designed by UC Santa Barbara. Co-PI Cahn leads digital agriculture training for the CropManage tool, which tracks water, fertilizer, and soil samples, and supports weather-based irrigation scheduling. Efforts include integrating open ET and expanding service areas in the Western U.S. Recent updates to CropManage include incorporating the Arizona weather station network and enhancing algorithms for crop stress and deficit irrigation. Cahn's demonstrations and support in commercial fields have contributed to a 36% increase in CropManage usage. In SO4, Co-PD Eldawy's team, along with Co-PD Vellidis and developer José Andreis, are developing FutureFarmNow, a tailored recommendation app for growers. Hosted at UC Riverside, its backend supports data access via an API, and the iOS frontend is in beta testing. The app includes data on farmlands, regions, and soil products, allowing for detailed visualization and analysis. Education SO5: During year three, eight students participated in Cohort II of the Digital Agriculture Fellowship (DAF). Student majors included statistics, mathematics for teachers, data science, environmental science, mathematics, and other. Students conducted research with faculty mentors, participated in Digital Agronomy Club activities, networked, toured faculty labs, traveled to a state conference, participated in workshops, and kept in contact with DAF peers through monthly check-ins. Students recently either interned off campus or continued their research in the UC Riverside RISE summer program and symposium. One student interned at Aquaspy, one student interned at Farmsense, four students continued in the UCR RISE program, and one student transferred to another program. Five students have been recruited for Cohort III, with three more students planned to be recruited in September of 2023. Majors include: data science, bioengineering, and electrical engineering. Students from Cohort III have participated in the RISE summer program and symposium and will continue with their student-led research. DAF students are all registered as Digital Agronomy Club members, which is an affiliated chapter of the Students of Agronomy, Soils, and Environmental Sciences (SASES). Plans include sending DAF students to the annual SASES meeting. Three DAF scholars have been recognized for their work: V. Gajjewar was awarded the "2023 National Student Recognition Award" by ASA, CSSA, and SSSA; H. Dingilian was awarded third place at the 2023 California Plant and Soil Conference poster competition, and M. Nolasco was awarded the 2023 ASA, CSSA, and SSSA Golden Opportunity Scholar. The Evaluation and Assessment continued to develop surveys in collaboration with the education coordinator and project director to gather feedback from the DAF undergraduate students. The results are used to develop the program with the student's input and plan professional development activities moving forward.
Publications
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2023
Citation:
Dingilian et al. Using Drone Multispectral Imaging for Weed Discrimination: A Case Study from Riverside, CA. California Plant and Soil Conference. February 7-8, 2023. Fresno, CA. California Plant and Soil Conference. February 7-8, 2023. Fresno, CA
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2023
Citation:
Guillory et al. Machine Learning with Landsat Satellite Data for Crop Mapping in the Yuma Valley Region of the Lower Colorado River Basin . California Plant and Soil Conference. February 7-8, 2023. Fresno, CA. California Plant and Soil Conference. February 7-8, 2023. Fresno, CA
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2023
Citation:
Jimenez et al. Using Gamma-ray Surveys to Predict Soil Properties in Perennial Cropping Systems. California Plant and Soil Conference. February 7-8, 2023. Fresno, CA. California Plant and Soil Conference. February 7-8, 2023. Fresno, CA
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Dhungel, R., Anderson, R.G., French, A.N. et al. Assessing evapotranspiration in a lettuce crop with a two-source energy balance model. Irrig Sci 41, 183196 (2023). https://doi.org/10.1007/s00271-022-00814
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Dhungel, R., Anderson, R.G., French, A.N. et al. Remote sensing-based energy balance for lettuce in an arid environment: influence of management scenarios on irrigation and evapotranspiration modeling. Irrig Sci 41, 197214 (2023). https://doi.org/10.1007/s00271-023-00848-9
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Dhungel, R., Anderson, R.G., French, A.N. et al. Early season irrigation detection and evapotranspiration modeling of winter vegetables based on Planet satellite using water and energy balance algorithm in lower Colorado basin. Irrig Sci (2023). https://doi.org/10.1007/s00271-023-00874-7
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Xu, C., Torres-Rojas, L., Vergopolan, N., & Chaney, N. W. (2023). The benefits of using state-of-the-art digital soil properties maps to improve the modeling of soil moisture in land surface models. Water Resources Research, 59, e2022WR032336. https://doi.org/10.1029/2022WR032336
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Cahn, Michael, et al. Field evaluations of the cropmanage decision support tool for improving irrigation and nutrient use of cool season vegetables in California. Agricultural Water Management, vol. 287, 2023, p. 108401, https://doi.org/10.1016/j.agwat.2023.108401.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2022
Citation:
Xu, C., & Chaney, N. W. (2022) Towards Polaris 2.0: Revisiting the Prediction of Soil Classes [Abstract]. ASA, CSSA, SSSA International Annual Meeting, Baltimore, MD. https://scisoc.confex.com/scisoc/2022am/meetingapp.cgi/Paper/144328
- Type:
Conference Papers and Presentations
Status:
Awaiting Publication
Year Published:
2022
Citation:
Dhungel, R., Anderson, R. G., French, A. N., and Scudiero, E., High-resolution Planet Multispectral Images for Evapotranspiration Estimation of Lettuce Using a Two-Source Water and Energy Balance Model , AGU Fall Meeting 2022, held in Chicago, IL, 12-16 December 2022, id. H54C-03.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2022
Citation:
Scudiero, E., Singh, A., Huang, J., Ellegaard, P., Skaggs, T. H. (2022) Soil Moisture Estimation of Agricultural Fields Using Remote Sensing and Machine Learning [Abstract]. ASA, CSSA, SSSA International Annual Meeting, Baltimore, MD. https://scisoc.confex.com/scisoc/2022am/meetingapp.cgi/Paper/143481
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2022
Citation:
Scudiero, E., Singh, A., Mahajan, G., Chatziparaschis, D., Karydis, K., Houtz, D., & Skaggs, T. H. (2022) On-the-Go Microwave Radiometry for High-Resolution Soil Moisture Mapping in Micro-Irrigated Orchards [Abstract]. ASA, CSSA, SSSA International Annual Meeting, Baltimore, MD. https://scisoc.confex.com/scisoc/2022am/meetingapp.cgi/Paper/143485
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2022
Citation:
Scudiero, E., Burgess, W., Ahmed, S., Dingilian, H., Gajjewar, V., Ho, C., Lincoln, K., Morales, I., Urrutia, K., Guillory, J., Jimenez, N., Singh, S. (2022) The University of California, Riverside's Digital Agriculture and Agronomy Club [Abstract]. ASA, CSSA, SSSA International Annual Meeting, Baltimore, MD. https://scisoc.confex.com/scisoc/2022am/meetingapp.cgi/Paper/145470
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2022
Citation:
Kayad, A., Putman, A., Scudiero, E., Dingilian, H., & McGiffen, M. (2022) A Simple Weed Detection Technique through Drone Images for Transplanted Vegetables [Abstract]. ASA, CSSA, SSSA International Annual Meeting, Baltimore, MD. https://scisoc.confex.com/scisoc/2022am/meetingapp.cgi/Paper/143059
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2022
Citation:
Mahajan, G., Singh, A., Montazar, A., Corwin, D. L., & Scudiero, E. (2022) Investigating Soil-Plant Relationship in Salt-Affected Farmland Using Soil Apparent Electrical Conductivity -Directed Soil Sampling and Planetscope Time-Series Analysis. [Abstract]. ASA, CSSA, SSSA International Annual Meeting, Baltimore, MD. https://scisoc.confex.com/scisoc/2022am/meetingapp.cgi/Paper/142084
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2023
Citation:
Nolasco et al. Creating a Rapid GIS Workflow to Correct On-the-go Gamma Ray Soil Spectrometry Data According to Survey Speed. California Plant and Soil Conference. February 7-8, 2023. Fresno, CA
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2023
Citation:
Attride et al. Investigating the Relationship of Maize Yield with Sentinel 2 Time Series Data Over Hundreds of Fields in the Conterminous United States. California Plant and Soil Conference. February 7-8, 2023. Fresno, CA
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2023
Citation:
Banik et al. Investigating the Relationship of Maize Yield with Sentinel 2 Time Series Data Over Hundreds of Fields in the Conterminous United States. California Plant and Soil Conference. February 7-8, 2023. Fresno, CA. California Plant and Soil Conference. February 7-8, 2023. Fresno, CA
|
Progress 09/01/21 to 08/31/22
Outputs
Target Audience:The audience of this project includes, but is not limited to, growers, farm advisors, extension specialists, and scientists with expertise in soil, crop, irrigation, and pest management, commodity groups, non-profit organizations, water districts, agricultural technology companies, and young scientists and students interested in agricultural data science careers. Members of these target audiences were reached during the reporting period by the project teams via several outreach activities. Audiences such as agricultural consultants, growers, and scientists were reached through conference presentations, scientific (technical and non-technical) publications, and internet media (e.g., https://ai4sa.ucr.edu/; https://twitter.com/ucr_ai4sa). As part of the education portion of the project, seven undergraduate students continued to conduct student-led research with the support of UC Riverside-based faculty in the Digital Agriculture Fellowship's (DAF) first cohort. These students participated in the DAF Spring Symposium on May 18, 2022, which was a virtual event open to peers and faculty hosted at UC Riverside via Zoom. Students from DAF's first cohort founded a UC Riverside campus Digital Agriculture and Agronomy club that was accepted as a Students of Agronomy, Soils, and Environmental Sciences (SASES) chapter on March 23, 2022. Eight additional students were selected through an open application process at the UC Riverside campus to participate in the next DAF cohort, prior to the start of the 2022 Fall Quarter. These students are also members of the campus Digital Agriculture and Agronomy club and have plans to expand the reach of the campus club to additional students. The education portion of the project aims to reach undergraduate students who major in science, and who are interested in learning more about digital agronomy as well as in gaining hands-on experience in the field. Majors from the first and second cohorts include mathematics, computer science, engineering, computational mathematics, mathematics for teachers of secondary school, data science, statistics, and environmental science. Project Co-PIs work with postdoctoral scholars funded by this grant, with several scholars receiving job opportunities following working on the project. The extension portion of this project worked with American growers in the Southwest and regional extension personnel. Co-PI Bali and Co-PI Cahn conducted demonstrations for growers in California, demonstrating the technology, research, and resources like CropManage. Workshops reached growers, extension personnel, crop advisors, industry personnel, and journalists for agriculture-related publications. Changes/Problems:The project was set back by delays resulting from the COVID-19 pandemic during year one. Lab work and fieldwork were delayed to remain in accordance with site-specific safety measures. The initial delays affected the planned progression of the education, research, and extension activities of the project into year two. Changes in staffing resulted in additional delays, particularly occurring in education and research. What opportunities for training and professional development has the project provided?The education component of the project includes the Digital Agriculture Fellowship, which is intended to support the education and career paths of each participating undergraduate student. Activities during year two included resume writing support, scientific presentations, graphical abstract creation, student-led research support, training in tasks like drone flying depending on their research focus, and internship application support. Postdoctoral scholars and student researchers supported by the project have conducted research, presented findings, attended workshops, attended meetings and conferences, co-authored papers, and have been encouraged to collaborate with other research groups on the project. Postdoctoral scholars have supported the Digital Agriculture Fellowship students and supported their work as well, acting as additional mentors. How have the results been disseminated to communities of interest?Results have been disseminated to communities of interest by using Twitter for updates regarding recently published papers and by participating in events/workshops. Multiple Co-PIs maintain updated personal web pages with a bibliography of their work. The local and campus press have covered Co-PI Eldawy's work twice. Through Twitter, affiliated websites (e.g., campus or department), and workshops the goal is to reach growers, academics, scientists, students, advisors, and technical personnel. Press coverage ideally supports the research reaching a local public audience. One of the undergraduate students participating in the Digital Agriculture Fellowship was also highlighted in a campus news article for their work in Co-PI Eldawy's research group. Student-led research has been disseminated to the campus community using the virtual DAF Spring Symposium event, which was promoted using a campus newsletter sent to undergraduate students. Additionally, the project's research groups participated in conferences to present their recent findings. The extension workshops hosted by Co-PI Bali utilize surveys developed by the UC Santa Barbara Evaluation and Assessment group led by Tarek Azzam and Natalie Jones. These surveys collect feedback from audience members to better determine what research topics would be most useful for the presentations to focus on. Project outreach activities include Outreach Activities by date: 1/26/22 - AgTechx Ed. Summit in San Bernardino, Ca -- PD Scudiero participated on a panel of educators and academics discussing education and agricultural workforce development. 1/28/22 - CropManage Training Workshop in Stanislaus County, Ca -- Co-PI Cahn provided in-person Crop Manage tool training. 2/10/22 - CropManage virtual Training Workshop -- Co-PI Cahn provided a virtual Crop Manage tool training. 2/17/22 - Virtual Machine Learning Lunch Seminar hosted by Vanderbilt University -- Co-PI Papalexakis discussed tensor decompositions in a virtual seminar hosted by Vanderbilt University. The event was free. 2/24/22 - Southwest Ag Summit - Co-PI Sanchez and Co-PI French presented on Water & Salt Balance. 3/10/22 - Hybrid Vegetable Crops and IPM Workshop in Imperial County -- Co-PI Putman participated in a virtual workshop and discussed downy mildew in vegetable crops. 5/18/22 - Digital Agriculture Fellowship Spring Symposium - This event showcased the research of the seven students participating in the fellowship and concluded the first cohort's participation in DAF. The virtual event was recorded and the YouTube video is currently located on the project's website. 5/20/22 - Hybrid Irrigation and Nutrient Mgmt./CropManage Workshop in Parlier, Ca -- Co-PD Cahn organized this hands-on extension event focusing on irrigation and nutrient management in the central valley, including a hands-on demonstration of how to use CropManage. Co-PD Bali and PD Scudiero moderated and presented at this event. 5/24/22 -- TerraRad Tech free demonstration/meet & greet at UC Riverside -- PD Scudiero co-organized a free demonstration event showcasing TerraRad Tech's drone technology. The event was promoted to the campus and surrounding community using the campus Eventbrite calendar. 6/22/22 - 6/23/22 -- Advances in Citrus Water Management Workshops in Coachella Valley, Ca and Escondido, Ca -- Co-PD Bali co-organized this extension workshop offering interactive presentations related to citrus growing water management. Co-PD Scudiero presented at this event. 7/7/22 - Advances in Pistachio Water Management Workshop in Tulare, Ca -- Co-PD Bali co-organized this extension workshop offering interactive presentations related to pistachio growing water management in the central valley. PD Scudiero presented at this event. Received ag. press coverage: "Pistachio Nutrition with Limited Water" by Cecilia Parsons at West Coast Nut. NEWS & PRESS 12/8/2022 - "Wildfire dataset could help firefighters save lives and property" located on news.ucr.edu -- The article by Holly Ober describes Co-PI Eldawy's open source dataset WildfireDB which includes millions of US wildfire data points. Co-PI Eldawy's research team includes Samriddhi Singla and Digital Agriculture Fellow Vinayak Gajjewar. 2/1/2022 -- "Researchers build a 1-stop shop for historical wildfire data" by Jo Kwon for Spectrum News 1 (local television news channel) -- Reporter Jo Kwon interviews Co-PI Eldawy to highlight WildfireDB. Products (models, instruments, audio/visual, curricula, data, databases, collab): https://cropmanage.ucanr.edu/ CropManage is a free online crop management decision support tool managed by Co-PI Michael Cahn as part of the University of California Cooperative Extension. https://ai4sa.ucr.edu The AI4SA project-based website was created using the management system maintained by UC Riverside. Planet Labs, Inc. collaboration: AI4SA has collaborated with Planet Labs, Inc. to extend the use of the product license to all UCR personnel with a UCR email address through the year 2025. 56 total campus personnel have accessed the license to create a general user Planet account since year one, with 17 participants joining in year two. Of the 17 added participants in year two, two are faculty members while the rest are not. One of the two faculty members is affiliated with the project. What do you plan to do during the next reporting period to accomplish the goals?The initial phase of the project established research activities at the participating labs. Collaboration across disciplines was a key component of research in the first two years. Collaboration across thrusts and task forces will be key components in year 3 and beyond. The project communication between groups will increase in frequency to encourage feedback and collaboration, supported by the academic coordinator. The extension will work towards establishing a foundation for a multi-state cooperative extension program by organizing collaborative and regionally focused workshops. The next step will be continued planning meetings. The Eldawy and Vellidis groups will continue to develop the iOS application, with plans for interface bug fixes, user input to determine new features, and creating the beta version. The education portion of the project will continue to develop a fellowship program that supports each student's education and career paths. During year two, there were challenges with conference participation, job turnover, payment schedules, cohort selection, and fellowship enrollment/participation at the sub-awarded institutions. Due to this, events and procedures have been and will continue to be planned earlier to allow for time to offset any unexpected challenges. An additional project member will be hired to support the UC Riverside-based education coordinator. Sub-awarded institutions will plan to double their number of recruited students. Survey development will be adjusted to allow for more frequent feedback to gauge the effectiveness of the program.
Impacts
What was accomplished under these goals?
Precision agriculture (e.g., site-specific management) is not common along the lower Colorado River Basin and other areas of the U.S. Southwest. The project found that many agricultural areas in the U.S. Southwest are characterized by high spatiotemporal variability crop x environment x management relationships. Rapid and accurate assessment of such variability is key to improving agricultural management. High spatial and temporal resolution remote sensing, from aerial and spaceborne sensors, as well as proximal sensing, fixed or on-the-go, can be used with the aid of machine learning to characterize such heterogeneities in the US Southwest agricultural systems. Knowledge of these systems enables improved management: our precision fertigation trials are showing that water and nitrogen use efficiency can be remarkably increased in vegetable crops when spatial information on soil and crop is included in management decisions. This project's findings strongly suggest that precision agriculture is feasible in many areas in the US Southwest. This project is educating growers in CO, AZ, and CA on novel agricultural technologies. This project is training students, mostly undergraduates, to develop skills in the transdisciplinary field of agricultural data sciences. Research SO1: Co-PI Eldawy and Co-PI Papalexakis established a shared server hosted at UC Riverside. They are currently integrating raster and vector data at scale using a cluster of machines that focus on raster processing. The groups are testing Predict Crop Data Layer crop types in Landsat8 imagery of crop fields by classifying pixels using semantic segmentation. They are exploring real-time crop classification with Planet imagery and data from the Bureau of Land Management. Co-PI Anderson, Co-PI Sanchez, and Co-PI French tested the balance model (BAITSSS) in specialty vegetable crops (e.g., lettuce) in Lower Colorado River Basin and Salinas sites. Evapotranspiration (ET) modeled by BAITSSS generally agreed well with measured ET for both sparse and full canopy periods. Presently, Co-PI Anderson's USDA-ARS group is working to integrate BAITSSS into real-time processing streams with improved data streams from POLARIS and Planet. The group faced challenges with modeling vegetable crops due to the short growing seasons, variable phrenology, and hydrologic modeling. They used regional evaporative fraction energy balance (REFEB) to make spatially distributed ET observations in Yuma, Palo Verde/CRIT, and Salinas. They found that the rapid sampling design of REFEB compared to EC and road network enabled 40-60 field observations and that increased field numbers are likely for revisited regions. Co-PI Sanchez and Co-PI French have started to focus on water management for alfalfa and citrus, with challenges related to implementing logistically suitable methodologies for citrus that will give them data on a scale that is similar to other commodities in their evaluations. They evaluated variable rate nitrogen management for several economically important vegetable crops. Zone management provides economic and environmental improvements over whole field management; however, methodologies for defining zones need further development. They have compiled robust databases for water and salt management using the state-of-the-art methodology for the most important vegetable crops and some rotational crops in the low desert. Co-PI Khosla's group selected two sites in the Colorado River Basin and two sites in the Kansas River Valley to conduct soil sampling, spatial EC mapping, moisture availability, leaf-tissue sampling (i.e., corn, soy), UAS and satellite imagery, and grain yield. The group might select one additional testing location, and this site might be selected with Co-PI Chief. PD Scudiero and Co-PI Anderson led field-scale soil surveys and processing (40 fields since year one) in Yuma, CRIT, Salinas, UC Riverside, and Imperial Valley. PD Scudiero's group is investigating improving surface soil moisture estimation using backscatter, optical remote sensing data, and point measurements of in-situ soil moisture. Co-PI Bali has been utilizing sensors and technologies in the Low Desert to investigate ETa, applied water, soil moisture, ET, yield, drone, and water table under different irrigation and nutrient conditions. They are investigating automated surface irrigation to improve real-time decision-making. SO2: Co-PI Putman and Co-PI McGiffen's group tested weed detection using spatiotemporal analysis of drone images for transplanted vegetables in a UC Riverside experimental field. The results suggest that the approach could detect >99.8% of pepper plants and >90% of weeds. They are testing weed (species) detection by spectral reflectance, and they are developing a deep-learning method for weed detection from RGB images. Extension SO3: Co-PI Bali has collected survey information from growers to determine topics that would be most useful to present at future workshops. The surveys were developed by Natalie Jones and Tarek Azzam at UC Santa Barbara to help track the effectiveness of the meetings based on the audience's feedback, as well as to determine preferred or useful topics to present at workshops. Co-PI Bali co-organized three California-based water workshops with regional crop focuses. Co-PI Cahn continued organizing and hosting digital agriculture-focused training for the CropManage irrigation and nutrient management decision support tool (cropmanage.ucanr.edu). SO4: FutureFarmNow is planned to be a map-based prototype iOS application. It is currently being hosted at UC Riverside, with the backend being developed and will be maintained by Co-PI Eldawy's group. The front end is being developed by Co-PI Vellidis's group and Austin developer José Andreis. The backend is currently hosting two types of data: farmland and soil product data. Processes Co-PI Eldawy's group has developed include: visualizing downloads, querying one farmland, querying a specific region, running reports, and exporting as CSV. Currently, the run report is a work in progress. The front end's main screen is a map-based interface that can be zoomed in to load farmlands polygons. Farmlands polygons display a panel with available data from the project's Soil Statistics API. Farmlands data can be shown using desired parameters (e.g., soil depth) with other features being possible, depending on the stakeholders' needs. There is a limit of 20,000 acres currently. Education SO5: During year two, seven students participated in cohort one of the Digital Agriculture Fellowship. One student left the fellowship early to pursue a valuable internship. The students continued their student-led research, founded a UC Riverside campus Digital Agriculture and Agronomy Club, and presented their research at the Digital Agriculture Symposium. Three of the students were selected for paid internships with ESRI, Hortau, and AquaSpy during the summer. Club activities included career exploration, with external professionals providing information about their careers in digital agriculture. The club petitioned and was accepted as an affiliated chapter of the Students of Agronomy, Soils, and Environmental Sciences (SASES). Eight new students were selected for the next DAF cohort, and each student is also a member of the campus Digital Agriculture and Agronomy club. The Evaluation and Assessment team led by Tarek Azzam and Natalie Jones (UC Santa Barbara) develop surveys in collaboration with the education coordinator and project director to gather feedback from the Digital Agriculture Fellowship undergraduate students. The results are used to develop the program with the student's input and plan professional development activities moving forward.
Publications
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2021
Citation:
Singla, S., Eldawy, A., Diao, T., Mukhopadhyay, A., & Scudiero, E. (2021, November). The Raptor Join Operator for Processing Big Raster+ Vector Data. In Proceedings of the 29th International Conference on Advances in Geographic Information Systems (pp. 324-335).
- Type:
Journal Articles
Status:
Published
Year Published:
2022
Citation:
8/26/22 Dhungel, R., Anderson, R. G., French, A. N., Saber, M., Sanchez, C. A., & Scudiero, E. (2022). Assessing evapotranspiration in a lettuce crop with a two-source energy balance model. Irrigation Science, 1-14.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2021
Citation:
Dhungel, R., Anderson, R., French, A., & Sccudiero, E. (2021, December). Improving satellite evapotranspiration for vegetable irrigation management in the Lower Colorado River Basin with novel satellite platforms, artificial intelligence. In AGU Fall Meeting Abstracts (Vol. 2021, pp. H31F-04).
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2021
Citation:
rench, A., Sanchez, C., Saber, M., Scott, A., Shields, J., & Wisniewski, E. (2021, December). Mapping Irrigation Crop Water Use with Sentinel 2-adjusted FAO-56 Growth Duration Values in Yuma, Arizona. In AGU Fall Meeting Abstracts (Vol. 2021, pp. H52E-02)
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2021
Citation:
Xu, C., Torres-Rojas, L., Vergopolan, N., & Chaney, N. (2021, December). Investigating the influence of vertical heterogeneity and Machine Learning derived soil properties in land surface modeling accuracy. In AGU Fall Meeting Abstracts (Vol. 2021, pp. H42B-04).
- Type:
Journal Articles
Status:
Published
Year Published:
2022
Citation:
Garg, A., A. Sapkota and A. Haghverdi. 2022. SAMZ-Desert: A Satellite-based agricultural management zoning tool for the desert agriculture region of southern California. Computers and Electronics in Agriculture 194: 106803. doi:https://doi.org/10.1016/j.compag.2022.106803.
|
Progress 09/01/20 to 08/31/21
Outputs
Target Audience:The audience of this project includes, but is not limited to, growers, farm advisors, extension specialists, and scientists with expertise in soil, crop, irrigation, and pest management, commodity groups, non-profit organizations, water districts, agricultural technology companies, and young scientists and students interested in agricultural data science careers. Members of these target audiences were reached during the report period by the project teams via several outreach activities. Meetings with agricultural technology (over 15 companies), commodity boards (Citrus Research Board of California, Almond Board of California), and advocacy groups (Western Growers) were carried out to present the project objectives and preliminary results as well as seeking collaboration and feedback on how to improve project impacts. Audiences such as agricultural consultants, growers, and scientists were reached through conference presentations, scientific (technical and non-technical) publications, and internet media (e.g., https://ai4sa.ucr.edu/; https://twitter.com/ucr_ai4sa). Students and postdoctoral scientists were also reached, as follows: 1. Co-PI Ahmed Eldawy: Implemented training of a graduate student; worked with an undergraduate student hired through the Digital Agriculture Fellowship on developing a web-based application for exploring and visualizing large-scale geospatial data. 2. Co-PI Vagelis Papalexakis: Worked with and trained master's degree student Krishna Kabi and doctoral student Rutuja Gurav; worked with Digital Agriculture Fellow Ingrid Morales on unsupervised tensor modeling. 3. PD Elia Scudiero: Oversaw (with Co-PD Nugent) the student-led research program on Digital Agriculture; trained a postdoctoral scholar and other support personnel with time for feedback during scheduled biweekly meetings. 4. Co-PI Nate Chaney: Working with a second-year Ph.D. student who is funded by the project; she will submit her first paper, with her second paper also in preparation. The papers are meant to lead to an improved POLARIS dataset. 5. Co-PI Raj Khosla: Worked with and trained two postdoctoral scholars. The extension portion of this project worked with American growers in the Southwest and local extension personnel. Co-PI Bali and Co-PI Cahn conducted demonstrations for growers in Holtville and Salinas, demonstrating the following technologies: Tule technologies (Eta), Aquaspy (soil moisture and EC), Watermark (soil moisture), Hobo (shallow water level sensors and EC sensors), drones (multispectral, infrared, crop geometry), flow meter (uncommonly utilized), irrigation systems (flood, surface drip, subsurface drip). The irrigation management practices demonstrated included optimizing drip establishment of lettuce, improving application uniformity of drip by decreasing the length of drip lines and using pressure regulation, and ET-based irrigation scheduling using CropManage. The education portion of this project reached specifically undergraduate students with science-based majors. A goal of this project's Digital Agriculture Fellowship is to expand to undergraduate audiences adjacent to the agricultural science community by showcasing opportunities for science-based work within agriculture; fellows selected for the first cohort are majoring in subjects including: mathematics, computer science, engineering, and computational mathematics. UC Riverside-based project co-PIs were paired with undergraduate students as part of the Digital Agriculture Fellowship's first cohort. Training activities covered topics within the scope of the faculty's expertise that aligned with the students' majors and career goals. During Year One of the project, students participating in the Fellowship completed the RISE Summer Symposium, where they shared their projects with 102 peers participating in six additional partner programs, as well as the faculty of UC Riverside. Students met in weekly group meetings with their peers and mentors to discuss their work, in preparation for the Symposium presentation. Changes/Problems:The COVID-19 pandemic resulted in the closure of multiple research labs to comply with state and county mandates. Labs reopened in accordance with campus safety measures, which were site-specific. Due to the closures of campuses as a result of the pandemic, hiring was slowed or completely halted. Onboarding new personnel took longer than anticipated as a result. Co-PI Raj Khosla changed campus sites during the first year of the project and transitioned from Colorado State University to Kansas State University. This transition resulted in needing additional time for onboarding and re-establishing personnel, including two hired postdocs. PD Elia Scudiero experienced field vehicle failure preventing aspects of fieldwork from being conducted for intermittent amounts of time until the vehicle was successfully repaired leading into project year two. Fieldwork was significantly affected by the campus closure as a result of the pandemic as well. The USDA ARS facilities in Riverside and Maricopa have been closed as a result of the COVID-19 pandemic, which delayed the first year of fieldwork. No students from the sub-awarded institutions were enrolled in the Digital Agricultural Fellowship in year 1. Attendance to the RISE program would have not been possible for these students. Hoping that the COVID-19 pandemic will be over in the near future, the sub awarded institutions plan to double their number of students in the summer of year 2 (through year 3) of the project. What opportunities for training and professional development has the project provided?Project members worked with five postdoctoral scholars, four graduate students, and eight Digital Agriculture Fellowship undergraduate students during year one. Postdoctoral scholars participated in project-related meetings, meetings with project faculty, and collaborated with campus-based faculty to conduct fieldwork. Graduate students also had opportunities to participate in project-related meetings and collaborated with their faculty supervisors and peers. The undergraduate students in the Digital Agriculture Fellowship were paired with faculty mentors and participated in weekly update meetings. The Fellows also worked with peers within the Fellowship and with other students in programs participating in the 10-week RISE Summer Symposium. How have the results been disseminated to communities of interest?The extension portion of the project has reached out to growers directly and virtually through workshop events, such as the two-day Online Citrus Water Management Workshop held on September 9-10, 2021. Research-based virtual events have been advertised using social media channels, Twitter, as well as University of California Natural and Agricultural Resources event pages. During year one, the project's Twitter account was created and will continue to be utilized in updating the target audience with upcoming events and, moving forward, project-related highlights. The project (also called "AI4SA") has a landing page website using the content management system offered by the University of California, Riverside; this site offers information regarding the project as well as instructions on how to access the Planet Labs, Inc license. Distribution of the website address has so far been through email correspondence, the June 9, 2021 issue of the Inside UCR Weekly e-newsletter, and listed on campus websites and the project-based Twitter. Results of the undergraduate students' work have been disseminated to the UC Riverside campus community through the RISE Summer Symposium event. PRoject knowledge and results were dissiminated to the commiunities of interest through the project products and outreach activities, which include: University of California GIS Week Conference, November 17-19, 2020 PD Elia Scudiero and Co-PI Ahmed Eldawy participated in a three-day conference open to anyone interested in learning more from university-based experts about real-world applications of GIS. Hosted by the UC geospatial community, the event was virtual and free of charge. Understanding Ecohydrologic Processes of Agricultural Ecosystems from Headwaters to Groundwaters During Droughts, June 18, 2021 Co-PI Hoori Ajami led a virtual discussion on the complexities behind understanding water and ecohydrological processes during droughts. This free event was hosted by UC ANR and was open to all interested in learning more about this topic. The event was promoted on Twitter. RISE Summer Symposium Event, August 25, 2021 The virtual Research in Science and Engineering (RISE) Symposium event was hosted by the UC Riverside's STEM Pathway Program and completed the 10-week summer research component of the Digital Agriculture Fellowship. The Fellows presented their findings from the research they conducted with faculty mentors to peers, staff, and other university personnel. Six additional partner programs participated in this event with the Digital Agriculture Fellowship students. Advances in Citrus Water Management Workshop, September 9-10, 2021 This event took place virtually and was co-organized by Co-PI Khaled Bali, with PD Elia Scudiero serving as a moderator at the event. Participant survey information was taken prior to the event to determine audiences reached. The event was advertised using the UC Agriculture and Natural Resources website, the project's Twitter, and on the California Citrus Mutual website. Drought and Climate Change Impacts to Rural and Urban Water Users, October 5, 2021 Co-PI Kurt Schwabe participated in a free virtual panel discussion event hosted by UC ANR Cooperative Extension Specialists in Agricultural and Resource Economics (ARE). This discussion was open to all interested in learning more about drought and its effects on urban and rural water users and California communities. Geospatial/GIS Faculty Panel, October 14, 2021 Co-PI Hoori Ajami participated in a free virtual UC Riverside faculty panel that aimed to provide a space for college students interested in getting involved with GIS to ask questions and hear from experts about how different fields utilize GIS. This event was advertised on the UC Library website, the UC LIbrary Twitter as well as the project's Twitter page. Trainings: CropManage Hands on Virtual Seminar, May 4th, 2021 Led by Co-Pi Michael Cahn, this free training event took place virtually and provided an opportunity for growers to learn about how to use CropManage. Participants worked with the hosts in real time to practice using CropManage after setting up a free account to access the software. Products (models, instruments, audio/visual, curricula, data, databases, collab): https://cropmanage.ucanr.edu/ CropManage is a free online crop management decision support tool managed by Co-PI Michael Cahn as part of University of California Cooperative Extension. https://ai4sa.ucr.edu The AI4SA project-based website was created using the management system maintained by UC Riverside. Planet Labs, Inc. collaboration: AI4SA has collaborated with Planet Labs, Inc. to extend the use of the product license to all UCR personnel with a UCR email address through the year 2025. This license information is presented on the project's website, and was promoted on Twitter as well as through the UCR Weekly e-newsletter. 42 campus personnel have accessed the license to create a general user account, with 32 participants being non-faculty staff or students during year one. What do you plan to do during the next reporting period to accomplish the goals?1. Co-PI Ahmed Eldawy's team plans to purchase and set up the server, host the data, and have it accessible to project members. They plan to expand their analysis technique to integrate raster and vector data with machine learning and AI techniques. 2. Co-PI Vagelis Papalexakis's team plans to identify and work on richer/better/cleaner data in collaboration with co-PI Eldawy's lab. They plan to continue working as part of the Digital Agriculture Fellowship, as well as investigate hybrid models. 3. PD Scudiero's team plans to continue mentoring Digital Agriculture Fellows; they also plan to continue fieldwork and collaboration between the research groups within the project. Specific to communications, as fieldwork continues, the social media channel and website will be updated with highlights from project members' work. 4. Co-PI Putman and Co-PI McGiffen plan to begin controlled experiments in year two. 5. Co-PI Bali and Co-PI Cahn will improve CropManage by adding the ability to use AZMT weather data for the Yuma region, improving the root development model, improving the leaching fraction calculator estimation, and improving the leaching requirement recommendation model. Moving forward, additional demonstrations will include new sensors that will be installed by Co-PI Anderson. They will also hold meetings with Co-PI Sanchez and Co-PI Khosla to plan for joint events. 6. Co-PI Chaney's team will finish developing the new DSM algorithm and submit the related paper. They will also implement the DSM paper for soil property prediction over the project's study domain once the field data has been collected, analyzed, and processed. 7. Co-PI Raj Khosla's team will collect data from all identified sites beginning in the spring, working with PD Scudiero's team. They will install the sensors they received for Colorado sites in Spring; the sensors include Ag Arable and Acclima TDR soil moisture sensors. 7 ARS Riverside and Maricopa will incorporate higher resolution soil property data for better irrigation parameterization in BAITSSS, they will run crop water use models (BAITSSS) with high-resolution satellite and meteorological inputs, and they will conduct fieldwork validation with PD Scudiero's team. 8. The Digital Agriculture Fellowship's first cohort will continue into year two, with additional responsibilities added, including mentorship for the next cohort. Applications will be open in spring, and the second cohort will be selected to participate in the Fellowship. It is the goal to increase the second Fellowship cohort with opportunities for more in-person activities on their campuses. 9. Co-PI Charles Sanchez's team plans on adding an intern to their team in the summer of 2022. They will continue their second season of fieldwork.
Impacts
What was accomplished under these goals?
Impact statement During year one of the project, the project members established the foundation to reach each of the project objectives. This included hiring necessary support personnel, creating an accessible database that will meet the requirements of the project datasets, establishing contracts with relevant technology companies, beginning testing and data collection, outlining immediate goals for web interfaces, beginning the Digital Agriculture Fellowship, and establishing collaboration amongst project members. The work completed in year one of the project offered opportunities for training and collaboration with the project's target audiences, including growers and undergraduate students. Research AIM-AI Team (UCR, UC ANR, CSU, Duke, UA, UGA, USDA-ARS) EPD-AI Team (UCR) SO1, SO2: The data science team led by Co-PI Ahmed Eldawy and Co-PI Vagelis Papalexakis set up the data management framework that will assist with all incoming data from the research teams. This process included gaining access to Planet Labs, inc. API and building scripts to download the desired data. The team also began identifying the data products and spatial and temporal domains that define how the data will be downloaded and maintained. A survey was sent out to project members to determine and prioritize features of a geodatabase server that will host the project data, with accessibility in mind. Additionally, the data team took the initial step in developing a distributed query processing engine for analyzing raster and vector data, a technique that is planned to be expanded by the team moving forward. The Papalexakis team specifically is working on deep image segmentation for crop classification with goals of modeling improvements (e.g., tensor regression layer), transferring learning from one crop to another, and submitting a paper to a top-tier AI/ML/DS venue. The AIM-AI Team enrolled personnel during year one, including postdoctoral personnel and a project manager. As a result of COVID-19 safety mandates being lifted in compliance with campus safety protocols, the teams began fieldwork data collection. The Scudiero team has specifically hired a field technician, a postdoc, and a project manager, with hiring being slower than expected due to the COVID-19 pandemic. Additionally, PD Scudiero ensured contracts with Arable and Planet Labs, inc., conducted soil surveys at twelve Southern Californian sites (seven in Imperial Valley, three in San Jacinto Valley, and two at UCR-AES Riverside), conducted testing of Planet Skysat at 50 cm resolution for weed detection in date palm, and collaborated with co-PI Dr. Charles Sanchez and other non-project collaborators to begin harmonizing ground-truth ECe and SP dataset for the US Southwest region. Co-PI Chaney's team (Duke) is improving previous work on continental-scale digital soil mapping. Their current watershed algorithms are more accurate compared to previous algorithms. The research team overall has been working in collaboration with USDA-ARS Riverside and Maricopa. During year one, they onboarded Research Associate Ramesh Dhungel and technician Alexandra Schmale for project activities. They conducted preliminary remote sensing modeling with studies underway in Yuma Valley and a publication expected this fall. Initial fieldwork began locally in Southern California at Scott Brothers Dairy. The USDA ARS team worked with Co-PI Eldawy and Co-PI Papalexakis to optimize crop water use models for high-resolution remote sensing. Co-PI Raj Khosla's team identified four testing sites, with three sites in Colorado and one site in Topeka, Kansas. Data has been collected on two sites, including grid soil samples, electric conductivity, UAV imagery (MicaSense - RGB, RE, NIR), and yield monitoring. Co-PI Charles A. Sanchez's team conducted fieldwork focusing on irrigation, water use, and salt balance for broccoli, cauliflower, spinach, spring mix, and cotton, with initiated field sites for celery. One season of a crop begins in late summer and ends the following summer; the first season, 2020-2021, was completed with their second season beginning at the end of summer in 2021 (2021-2022). The team is evaluating variable rate nitrogen fertilizer for initiated celery and leaf lettuce field sites. Lab work and satellite imagery processing have remained on schedule. The Putman and McGiffen team have hired a postdoc, and they expect controlled experiments to begin soon at UCR-AES. The members of the EPD-AI team purchased equipment and began training using Planet Labs, Inc. imagery. Extension SO3: The Bali and Cahn team have conducted irrigation management demonstrations at five fields in the Salinas Valley. Co-PI Bali conducted irrigation research in the Imperial Valley with olive trees. Co-PI Cahn has continued to update and utilize CropManage as part of demonstrations, the online irrigation and nitrogen management decision support tool (cropmanage.ucanr.edu). Additionally, Co-PI Dr. Bali and Co-PI Cahn led a two-day Online Citrus Water Management Workshop and distributed a follow-up survey. The Citrus Workshop included about 50 participants, with a large amount of survey participation. The Extension portion of the project has worked to develop and distribute surveys to determine areas of interest from their target audience, US Southwest growers. They worked with Natalie Jones and Dr. Tarek Azzam at the Evaluation Center of the University of California Santa Barbara to develop a commodity board 60-minute survey. Education SO5: The project worked in collaboration with UC Riverside's STEM Pathway Program to develop the Digital Agriculture Fellowship (DAF) component of the project. The Fellowship began during year one, with eight undergraduate students selected from UC Riverside. Students selected met qualifications for the Fellowship, including GPA requirements and science majors, and were selected from a group of applicants. The Fellowship began with pairing the selected students with UC Riverside-based project members who specialize in research areas that align with the student's goals as well as interests. The first of the Fellowship activities included participation in the Research in Science & Engineering (RISE) 10-week summer research program. Students presented their research at the end of the summer research program to an audience of peers and campus faculty and staff. The Evaluation Center of the University of California Santa Barbara interviewed the DAF students before and after theie experience with RISE. Developing research skills, conveying findings, and providing insight towards writer research papers and tools to present research were the main themes that emerged when trainees were asked about how the Summer RISE program contributed to their academic pursuits. On average, the students' understanding of digital agriculture increased from pre test to post test. Trainees said that the Summer RISE program provided them with the opportunity to develop a clearer goal of their career trajectory, how they could apply what they learned to the real world, and the confidence to pursue higher education, such as: Exposure to different technologies used in their professional environment, Understanding and communicating technical concepts, Providing insight towards how to conduct and present research. The RISE program overall motivated trainees to pursue graduate school, one student said: "Because of this summer program, I was able to decide to pursue a Ph.D. and conduct research. Having lab experience and strong professional connections with my faculty mentor will also aid me in my future career."
Publications
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2021
Citation:
Merrick Campbell, Keran Ye, Elia Scudiero, and Konstantinos Karydis. A Portable Agricultural Robot for Continuous Apparent Soil Electrical Conductivity Measurements to Improve Irrigation Practices, In 2021 IEEE 17th International Conference on Automation Science and Engineering (CASE), 2021. [Full-length paper]
- Type:
Journal Articles
Status:
Published
Year Published:
2021
Citation:
Elia Scudiero*, Ray G. Anderson, Ramesh Dhungel, Sonia Rios, Robert R. Krueger: Can Artificial Intelligence Enhance the Profit and Environmental Sustainability of Agriculture? . In: Progressive Crop Consultant. 2021, 4:4-9 Online version: http://progressivecrop.com/2021/07/can-artificial-intelligence-enhance-the-profit-and-environmental-sustainability-of-agriculture/
- Type:
Journal Articles
Status:
Published
Year Published:
2020
Citation:
Samriddhi Singla and Ahmed Eldawy. Raptor Zonal Statistics: Fully Distributed Zonal Statistics of Big Raster + Vector Data, In Proceedings of the 2020 IEEE International Conference on Big Data (IEEE BigData 2020), December 2020. DOI>10.1109/BigData50022.2020.9377907
- Type:
Websites
Status:
Published
Year Published:
2021
Citation:
https://ai4sa.ucr.edu
The AI4SA project-based website was created using the management system maintained by UC Riverside.
- Type:
Websites
Status:
Published
Year Published:
2021
Citation:
https://twitter.com/ucr_ai4sa
Twitter account used for project outreach
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2021
Citation:
E Scudiero: Challenges and Opportunities for Artificial Intelligence and Automation in the US Southwest . 2021 California Plant and Soil Conference. California Chapter of the American Society of Agronomy, Virtual. Feb 2021.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2021
Citation:
E Scudiero: Spatial and Temporal Changes of Soil-Crop-Environment Relationships . University of California GIS WEEK, Virtual. Nov 2020.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2021
Citation:
E Scudiero: Understanding soil-plant-environment relationships using big geospatial data . 2020 Plants3D Retreat, Virtual. Nov 2020.
|