Source: The Regents of University of California submitted to
AGRICULTURAL SALINITY MANAGEMENT VIA AN INTEGRATION OF PROXIMAL AND REMOTE SENSING WITH BIG GEODATA MODELING
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
TERMINATED
Funding Source
Reporting Frequency
Annual
Accession No.
1019467
Grant No.
2019-67022-29696
Project No.
CALW-2018-07212
Proposal No.
2018-07212
Multistate No.
(N/A)
Program Code
A1521
Project Start Date
Jul 1, 2019
Project End Date
Jun 30, 2023
Grant Year
2019
Project Director
Scudiero, E.
Recipient Organization
The Regents of University of California
200 University Office Building
Riverside,CA 92521
Performing Department
Environmental Sciences
Non Technical Summary
For irrigation to be sustainable in arid and semi-arid regions, the salt balance of the soil root zone must be maintained by irrigating in excess of a crop's water requirements. The extra irrigation water leaches salts down into the subsurface and prevents harmful salinity build-up near the surface (0-1 m depth). However, the traditional practice of over-irrigating for salinity management is being scrutinized strongly due to increasing water scarcity. This project aims to combine proximal and remote sensing with new, multi-scale, high-resolution big geodata modeling to provide accurate information on irrigation requirements for optimal crop growth and salinity control. The area of interest in this project is the Central, Coachella, Imperial, Salinas, and San Jacinto Valleys of California. We will use available USDA-ARS and University of California historical (mid-1970s - present) soil salinity datasets, as well as new surveys, to calibrate a machine-learning salinity prediction model based on remote sensing and other geodata. Then, available spatial evapotranspiration datasets will be integrated with the salinity prediction to provide early season irrigation advice. Example field data include real-time spatial evapotranspiration estimates, remote-sensing canopy reflectance, historical soil datasets, and new field-scale salinity surveys with proximal sensing and micro-scale solute transport modeling. Salinity predictions and water management advice will be delivered through an interactive online geographical information systems decision support tool. Success of this project will significantly help mitigate the footprint of agriculture in water-scarce US and global farmland by increasing water use efficiency and maintaining long-term soil quality, while sustaining and improving crop yields. The main beneficiaries of this project are American farmers, irrigation consultants, and state and federal natural-resource scientists and managers.
Animal Health Component
0%
Research Effort Categories
Basic
40%
Applied
20%
Developmental
40%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
1030110202050%
1020210205050%
Goals / Objectives
The long-term goal of this project is to significantly help mitigate the footprint of agriculture in saline farmland of the U.S. and elsewhere by i) increasing water use efficiency, ii) increasing efficiency of irrigation as a tool for salinity hazard control; iii) maintaining long-term soil quality; and iv) sustaining and improving crop yields. This will be done by developing a decision support tool that will provide accurate information on irrigation requirements for optimal crop growth and salinity control at the sub-field scale for American farmers, irrigation consultants, and state and federal natural-resource scientists and managers. California farmland, including the Imperial, Coachella, San Jacinto, Salinas, and San Joaquin Valleys, will be used as study area in this project.The project comprises four supporting objectives (SO):SO 1) Generate up-to-date broad-scale spatiotemporal soil salinity predictions based on remote sensing and other geodata;SO 2) Develop ground-truth and remote sensing based information on root-zone soil salinity for root-zone imaging in micro-irrigated orchards;SO 3) Integrate spatial predictions of soil salinity with spatial crop evapotranspiration information to determine field and sub-field water requirements;SO 4) Deliver spatial water management information through an easily accessible online geographical information systems decision support tool.This project will investigate on the use of engineered devices, such as the use of electromagnetic induction and gamma-ray spectrometry to characterize soil property in the root-zone at the field-scale (supporting objectives 1 & 2). We will focus on the use of technology, such as: soil solute transport modeling software, big-data machine learning, and geostatistical and spatial statistical software to integrate soil sampling data with geospatial near-ground and remote sensing measurements (supporting objectives 1 & 2). This project proposed to integrate novel information technology for spatial salinity estimation with available real-time crop water requirement tools and datasets to improve water management over large regions (supporting objective 3). We will create an online geographical information systems decision support tool to improve agriculturally relevant plant (e.g., tree and herbaceous crops in both organic and conventional agriculture) and natural resource (i.e., soils and water) systems (supporting objective 4).By accomplishing all four supporting objectives, this project will address the following Agricultural Engineering Program Area Priorities:We will enable engineering, soil and plant sensing, big geodata computing and modeling, and online geographical information systems for better management of soil and water natural resources, and sustainable (economically, environmentally, and socially) plant production in salt-affected US farmlands;We will develop and evaluate precision technologies for soil salinity monitoring, measurement, and detection in agricultural systems;We will develop advanced computational methods and technologies for mining, management, visualization, modeling, and communication of big data that will enable a more effective understanding of agricultural soil salinization and of crop water requirement in salt-affected soils over very-broad areas;We will provide a tool to help improve water management and efficiency of water use in salt-affected US farmland. The developed tool will provide information on where, when, and how to irrigate, at the sub-field and whole-field scales.
Project Methods
Supporting objectives (SO) 1 To accomplish SO 1, a large ground-truth soil salinity dataset will be required to calibrate and evaluate the broad-scale spatiotemporal soil salinity prediction model. For this project, historical and current salinity data will be used. Since the mid-1970s, scientists at the US Salinity Laboratory, in collaboration with UC staff, have been surveying hundreds of fields across saline farmland in California. Most of these salinity surveys were carried out over rain-fed fallow land and flood- and/or sprinkler-irrigated annual crops. Historical salinity datasets were collected over different soil types and varying climate conditions across central and southern California. Additionally, new 100×100-m footprints at ten to fifteen fields cropped with annual crops will be surveyed.A soil salinity prediction model will be calibrated on the ground-truth data using remote sensing and other geodata. Remote sensing data will be obtained from the Landsat collection (1972 to present). Surface reflectance will be used to calculate the normalized vegetation index (NDVI). Two to seven years of NDVI data may be needed at each site to separate the effect of salinity on crops from those of other stressors. We will, therefore, calculate NDVI time series for all seven years preceding each salinity survey campaign, at each ground-truth pixel. Smoothed NDVI time-series will be calculated. Additional geodata covariates will include: crop, land-use, soil physical properties, elevation, and meteorological data.Scaling our previous work up to the state, national, world levels will be challenging due to the limited computing capabilities in the traditional methods. This proposed project will overcome this challenge by adapting the state-of-the-art big-data (machine learning) technology with remote sensing data. When using machine learning it is easy to overfit the data. Overfitted models provide highly inaccurate estimates when extrapolated (spatially or temporally). Evaluation of the model will be done through cross-validation. A 10-fold cross validation will be used. Additionally, a spatially independent leave-one-field-out cross-validation will be also performed. Even parsimonious models - with small cross-validation errors - are likely to provide estimates with greater uncertainty with increasing distance from calibration points. To allow model users to improve prediction accuracy at previously unsampled locations, we will include an updating algorithm to the model. The algorithm will allow users to feed new soil salinity data at any location. The new salinity data will be added to the historical salinity dataset and will be used to re-calibrate the model. The integration of new data to increase the model estimation accuracy will be tested by using field-scale salinity surveys done during this project. A limitation of the remote sensing approach to estimate root-zone soil salinity is that it only delivers estimates for average root-zone salinity and does not provide any measure of risk of soil salinization from salts that are below the root-zone of plants.SO 2a. All historical and new salinity surveys in SO 1 were/will not be carried out over orchards. Within SO 2a, average root-zone soil salinity will be estimated for 100×100-m footprints at ca. 30 micro-irrigated orchards for calibration of the remote sensing -based salinity prediction model. At each orchard, 6 locations will be selected according to the spatial variability of apparent electrical conductivity and gamma ray spectrometry sensor maps. At each of the 6 location, 3 soil cores will be obtained down to 1.2-1.5 m. The cores will be below the irrigation emitter (e.g., dripper) and, perpendicularly, 0.5, and 1 m away from the emitter. This sampling scheme should allow yielding a good estimate of root-zone soil salinity for each of the selected sampling locations. Subsequently, salinity data for the orchards will be imputed to the remote-sensing salinity prediction model.SO 2b. At two orchards, we will monitor seasonal changes of root-zone salinity patterns with sensors throughout Yr.2 and 3 at four locations per field, hourly. Additionally, we will carry out repeated apparent electrical conductivity (ECa) sensor measurements (e.g., one survey in fall of Yr. 2, one survey in spring of Yr. 3). At soil-water monitoring location, we will sample 3 cores down to 1.2-1.5 m. Bulk density will be estimated from the cores. Soil will be analyzed for salinity and for sand, silt, and clay percentage content in the laboratory. Additionally, saturated hydraulic conductivity will be estimated from undisturbed samples at multiple depths, below the emitter. Numerical hydrological modeling will be used to model the geometry and temporal variation of water and salinity in the root-zone. Simulated soil profiles will be used to evaluate ECa vertical inversion modeling. Inversion of ECa data allows estimating ECa continuously through the soil profile, as opposed to average profile (e.g., 0 to 0.75 m) provided by electromagnetic induction sensors. If a good match is found between inverted ECa profiles and hydrological simulations, we will test the extrapolation of numerical hydrological modeling outputs at every location with associated ECa sensor measurements.SO 3. Crops water requirements throughout the growing season can be determined using evapotranspiration (ET) data. Crop ET (ETc) can be estimated using a crop coefficient (Kc) to adjust reference ET (ETo). Reference ET can be calculated from meteorological data. Spatial estimations of ETo are available using spatial meteorological datasets. In California, daily spatial ETo with 2 km pixel size are available from the California Department of Water Resources' California Irrigation Management Information System (CIMIS) service called Spatial CIMIS. The Kc values represent crop phenological stage and are variety, location, and agronomic practice specific. Kc values are generally available for most crops and agricultural systems. During the growing season, Kc generally starts with small values and increases to values equal or larger than 1 when plants have the largest water demand. After maturity and during later phenological stages, Kc decreases. Spatial CIMIS ETo and Kc values for different crops will be used to estimate crop water requirements for non-saline conditions.We will integrate the remote-sensing soil salinity estimations (30×30-m resolution) with ET data in a transient-state model used to forecast crop-water productivity curves, including the effects of salinity, at the beginning of each growing season. The model will output expected potential yield according to initial soil conditions, and the amount and quality of irrigation water. The loading of salts leached below the root-zone will be also calculated. Mathematical models can simulate water flow and chemical transport processes in irrigated soils, and can account for dynamic factors such as irrigation scheduling and seasonally variable water quality and crop salt tolerance. Variably-saturated water flow will be simulated with the Richards equation, while solute movement will be simulated with an advection-dispersion type equation. The model will generate irrigation recommendations and crop-water production functions for a variety of different irrigation strategies.SO 4. In this objective, we aim at building an interactive web-based interface to access the proposed system. The goal is to reach out to scientists and users to demonstrate our work and allow the stakeholder community to use the proposed models. The web tool will give users access to up-to-date soil salinity map and beginning-of-season irrigation prescriptions for used selected areas of interest.

Progress 07/01/19 to 06/30/23

Outputs
Target Audience:The target audience for this project encompasses American farmers, irrigation consultants, state and federal natural-resource scientists and managers, farm advisors, extension specialists, and scientists specializing in soil, crop, and irrigation sciences. Throughout the project's duration, multiple avenues were utilized to engage and inform these stakeholders. The project's outreach strategies included direct interactions with California growers during the first year, where the project investigators discussed project goals, gathered feedback, and obtained permissions for field data collection. The broader scientific community was reached through peer-reviewed publications (see "Products" section), providing a platform to share research findings and insights. Natural resource managers in both private and public sectors were engaged by the Project Director (PD) at a conference in Las Vegas in February 2020, organized by the Multi State Salinity Coalition. This event facilitated connections with stakeholders involved in technologies for desalination, saline water reuse, soil and water salinity control strategies, and water/energy efficiencies. During the second year, interactions continued with California growers, University of California Cooperative Extension personnel, and other stakeholders. These engagements included discussions about project goals, feedback solicitation, and permissions for field data collection, primarily conducted remotely due to the COVID-19 pandemic. The project's research findings were shared with the scientific community through peer-reviewed publications and presentations at various conferences, both within California and internationally. These presentations served as a vital channel for communicating the project's contributions and insights. Additionally, the PD sought advice from non-project scientists during the second year, including those who agreed to serve on the Advisory Committee. In the project's third year, the project investigators expanded their interactions to include California growers, University of California Cooperative Extension personnel, agricultural consultants/private companies, academics, research scientists, and students. These engagements included group and one-on-one interactions, both virtual and in-person, with discussions on project goals and the presentation of project discoveries. Feedback received from these stakeholders and advisors will inform project activities in the final year. Furthermore, the project's fundamental research accomplishments in big-geodata management were presented to the computer science community, with the project's funding contributing to research presented at scientific conferences through presentations and short papers. These contributions were pivotal in sharing knowledge about the use of big-geodata, including remote sensing time series and other spatial covariates, for mapping and monitoring soil salinity with stakeholders in California and internationally. Changes/Problems:The COVID-19 pandemic posed significant challenges and limitations to the complete success of the project in several ways. First and foremost, the pandemic forced a halt in regular project operations, including fieldwork, data collection, and in-person collaboration among team members and external partners. Travel restrictions, lockdowns, and health and safety concerns made it exceedingly difficult for project members to conduct essential research activities, including field visits to collect data and gather critical ground-truth measurements. The prolonged halting of these activities disrupted project timelines and hindered the completion of critical components of the research, such as the development of a finalized remote sensing model for multitemporal salinity mapping, a decision support tool for regional salinity control strategies, and the peer reviewed publication of project findings. Most project fingings were presented at scientific conferences, but not all of them were converted to peer-reviewed publications. These publications with be completed by the project personnel and collaborators in the future without the use of USDA-NIFA funds. What opportunities for training and professional development has the project provided?Over the course of the entire project, various training and professional development activities were diligently carried out to equip team members with the necessary skills and knowledge. These activities spanned multiple aspects of the project's goals and objectives. Training Activities: The project facilitated several training initiatives throughout its duration. Dr. Theodor Bughici, Dr. Mario Guevara (postdoctoral scholars) and University of California undergraduate students and technicians involved with the project received bi-weekly training sessions from the Project Director (PD) Dr. Scudiero and co-PD Dr. Todd Skaggs. These sessions covered the use of geophysical measurements and solute transport modeling to characterize salt-affected irrigated agricultural soils, with a focus on soil sensor operation, setup, and installation, as well as 1D and 2D simulations using the HYDRUS software. Co-PD Ahmed Eldawy provided one-on-one training to M.Sc. student Ms. Husna Sayedi on satellite data utilization for environmental information computation and web map interfacing for data visualization. Ph.D. student Ms. Samriddhi Singla received weekly mentoring sessions from Co-PD Eldawy on various topics, including big satellite data querying, pre and post-processing, analysis, and paper writing for peer-reviewed publications. Professional Development: The project team actively pursued opportunities for professional development. Dr. Bughici presented his work at the University of California - Riverside Postdoctoral Association's Postdoctoral Symposium, where he discussed data-driven irrigation scheduling methods. He also attended the "W4188: Soil, Water, and Environmental Physics to Sustain Agriculture and Natural Resources, 2020 Annual Meeting" in Las Vegas, NV, USA, to broaden his understanding of the field. Additionally, throughout the project, monthly seminars organized by the University of California Riverside and the USDA-ARS U.S. Salinity Laboratory were attended by Dr. Bughici, the PD, and co-PD Skaggs, featuring discussions on soil sciences, agricultural water management, and environmental sciences. Ms. Singla, a Ph.D. student, represented the project at the 27th International Conference on Advances in Geographic Information Systems (ACM SIGSPATIAL 2019), guided and mentored by Co-PD Eldawy on expanding the project's outreach within the geospatial community. Additionally, Ahmed Eldawy and graduate student attended the 30th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems (ACM SIGSPATIAL 2022) in Seattle, WA, to present the latest work and engage with the GIS community. The training and professional development efforts were ongoing and consistent throughout the entire project's duration, ensuring that team members continually enhanced their skills and knowledge, contributing significantly to the project's overall success. How have the results been disseminated to communities of interest?Throughout the entirety of the project, the dissemination of project results to the target audience was a comprehensive and ongoing process. These activities ensured that the project's findings reached a broad and relevant audience, as outlined below: Peer-Reviewed Publications: Project results were consistently shared through peer-reviewed publications. Notable contributions included research papers authored by Singla et al., Bughici et al. Corwin et al, Scudiero et al.; sreview papers by Corwin, and a book chapters co-authored by Scudiero, along with research articles co-authored by Skaggs. These publications were instrumental in sharing the project's research outcomes with the scientific community and beyond. Workshops and Meetings: The Project Director (PD) actively engaged in workshops and meetings where project results were presented to academic researchers, natural resource managers, water districts, and local policymakers. These interactions occurred in various settings, both physical and virtual. Notably, the PD participated in academic workshops and meetings where he discussed project results with faculty at the University of California in Davis, the Cooperative Extension Division of the University of California, and the Data Science Team of the University of California Riverside's Computer Science and Engineering Department. Advisory Board Engagement: In the second and third years of the project, the advisory board was expanded to include additional members who volunteered to serve as advisors. The engagement of the advisory board facilitated a more comprehensive sharing of project accomplishments and findings with key stakeholders. Conferences and Presentations: Project members actively participated in conferences, workshops, and meetings where they presented project results to a wide array of audiences, including growers, academic researchers, natural resource managers, water districts, and local policymakers. The dissemination efforts extended to international events, such as a presentation in Venice, Italy, where the PD discussed near-ground and remote sensing of soil salinity. Additionally, project members gave invited presentations at various conferences and symposiums, further solidifying the project's presence in the scientific community. Future Workshops: The project team is actively planning a workshop for the Food and Agriculture Organization of the United Nations (FAO) on the topic of "Remote sensing of soil salinity in farmed soils." This workshop, expected in 2024, will serve as a platform to share project insights and findings with a global audience. These comprehensive activities ensured the wide dissemination of project results to academic, industry, and policy communities of interest, fostering collaboration and sharing the project's valuable contributions to the fields of soil science, agriculture, and geospatial technology. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Substantial progress was made toward addressing the complex issues of optimizing water use and managing soil salinity in arid and semi-arid regions. The project's multidisciplinary approach demonstrated the potential to address critical agricultural and environmental challenges. The use of extensive historical datasets, advanced data science tools, and fieldwork allowed for the development of novel models and data science tools for predicting and mapping soil salinity across different scales. Despite facing challenges (such as delays caused by the COVID-19 pandemic), the project made significant contributions to the fields of soil science, agriculture, and geospatial technology, promising long-term impacts and paving the way for further advancements in sustainable agricultural practices and water resource management in arid regions. SO1 The approach involved compiling a rich dataset of historical soil salinity measurements across agricultural land in California. This dataset included measurements from over 150 time x location sites, with each site covering an average size of approximately 18 hectares. These measurements were obtained from historical field data collected by the project investigators and other research institutions over several decades. During the the project, additional data was collected from agricultural sites across California. A significant product of the project was the generation of a publicly available database (DOI: 10.15482/USDA.ADC/1527809) including these data. To enable state-wide, real-time processing of vast geodata volumes, Co-PD Eldawy developed an innovative methodology to facilitate rapid analysis of big data and its integration into the salinity assessment models being developed by the project team (See the project publications authored by Singla et al.). The soil salinity prediction models were further refined using machine learning techniques, specifically the random forest algorithm by project postdoc Guevara. This enabled the field scale prediction of soil salinity in areas where measurements were limited, provided the apparent electrical conductivity (ECa) exceeded a specific threshold. For soils with lower ECa measurements, traditional ground-truth measurements remained essential for accurate predictions. The integration of remote sensing time series data, including information from Landsat satellites, allowed the identification of variations in crop phenology, which were then correlated with ground-truth soil salinity measurements. This information was crucial for the development of a "beta-verion" of a machine learning models for predicting root-zone (0-1.2 meters) soil salinity. Additional environmental data, such as historical weather information and soil attributes, were incorporated to enhance the accuracy of the soil salinity predictions. This exploratory model was presented and discussed by Dr. Guevara a the 2021 and 2022 ASA, CSSA, SSSA International Annual Meetings. The model will be presented and discussed in peer-review publications which are under preparation. Although the findings of Dr. Guevara were novel, the project leadership concluded that the remote sensing modeling framework needed additional development for the soil salinity predictions to be sufficiently robust to possibly be used in decision making and policy development by growers, academics, natural resource managers and practitioners. Due to the COVID-19 pandemic, unfortunately, the work on the development of the multi-temporal remote sensing of soil salinity model were critically delayed. The postdoc hired for this task ended their contract and returned to their country of origin before the work on model was completed. SO2 The second objective focused on addressing the unique challenges of measuring and monitoring soil salinity in micro-irrigated orchards, where traditional methods often proved inadequate. To tackle this challenge, soil samples were collected from twelve micro-irrigated orchard sites. These samples were critical in understanding the patterns of salt accumulation, which depended on factors like soil type and irrigation management. One significant discovery during this phase was the identification of salt buildup occurring between 0.6 and 1.2 meters from the micro-irrigation emitters. This insight prompted the development of a mobile system for on-the-go measurements of soil apparent electrical conductivity (ECa) and gamma-ray spectrometry. This innovative approach allowed for high resolution monitoring soil properties, a key step in informed decision-making. The development process, experimental field data, and a detailed description of the mobile platform were documented in a manuscript authored by Scudiero et al (doi:j.still.2023.105899). Additionally, the project's findings led Co-PD Corwin to develop modified protocols for mapping soil salinity under micro-irrigation with electromagnetic geospatial sensing. The project identified that previous ECa-directed sampling protocols for mapping salinity under drip irrigation were inadequate. The new protocols recommend ground truthing ECa measurements over the entire rootzone for a single tree. The results have been disseminated to growers, academics, and practitioners. Additionally, Project Postdoc Bughici, played a crucial role in advancing the understanding of spatial and temporal variations in soil moisture and salinity within the soil profile. Bughici employed an ensemble approach to simulate likely scenarios for spatial and temporal changes in moisture and salinity patterns, accounting for variations in soil types, crop types, irrigation management strategies, and water quality using data from California Pistachio fields. The outputs generated by these two-dimensional models held significant promise in guiding on-the-go sensor surveys, estimating the timing of reliable sensor measurements of apparent electrical conductivity in areas with high salt accumulation following an irrigation event, guiding soil sampling to characterize short-scale soil salinity heterogeneity, and interpreting on-the-go sensor measurements to distinguish signals related to soil moisture from those associated with salinity. This comprehensive approach to understanding root-zone soil salinity in micro-irrigated orchards, as undertaken in the second objective, marked a significant step forward in addressing the challenges faced in these specialized agricultural settings. SO3 The third objective aimed to integrate information on soil salinity with data related to crop evapotranspiration. Bughici et al. (2022), and Helaila et al (2022 a,b) used numerical modeling to understand the impact of drought and changing water sources on water use and soil salinity of almond and pistachio orchards in Central California. Due to the COVID-19 pandemic, unfortunately, the work on the development of the regional salinity control model were critically delayed. The postdoc hired for this task ended their contract and returned to their country of origin before the work on the salinity leaching model was completed. SO4 The development of a web-based geographical information system (GIS) formed an integral part of the project, with the objective of providing spatial water management information in an accessible and user-friendly format. The first three years witnessed significant progress in this area, led by Co-PD Eldawy and a (former) graduate student, Dr. Singla. The project team developed preliminary versions of GIS interfaces that demonstrated the capability to display state-wide maps and vector data (point or polygon) in real-time. These interfaces allowed for queries and statistical modeling of remote sensing data and other environmental factors, including soil texture maps and spatial weather data. Because the regional scale salinity maps and the salinity leaching models were not completely developed, the GIS platform was not publicly published.

Publications

  • Type: Book Chapters Status: Awaiting Publication Year Published: 2022 Citation: Zhuocheng Shang, and�Ahmed�Eldawy. "Object Delineation in Satellite Images", In SpatialGems, Volume 2 (IN PRESS)
  • Type: Journal Articles Status: Published Year Published: 2023 Citation: Elia Scudiero*, Dennis L. Corwin, Paul T. Markley, Alireza Pourreza, Tait Rounsaville, Theodor Bughici, and Todd H. Skaggs: A System for Concurrent On-the-go Soil Apparent Electrical Conductivity and Gamma-Ray Sensing in Micro-irrigated Orchards. Soil & Tillage Research. 2024. 235: 1058993


Progress 07/01/21 to 06/30/22

Outputs
Target Audience:The target audience of this project includes American farmers, irrigation consultants, state and federal natural-resource scientists and managers, farm advisors, extension specialists, and scientists in the USA and abroad with expertise in soil, crop, and irrigation sciences. During the project's third year (July 1st, 2021 through June 30th, 2022), the investigators reached out to California growers, University of California Cooperative Extension personnel, agricultural consultants/private companies, academics, research scientists, and students. These interactions were carried out via group and one-on-one interactions, virtual and in-person. We discussed the project goals, and with field demonstrations and formal presentations, we showed the project's discoveries and received advice on how to effectively transfer the knowledge developed by the project. The advice received from stakeholders and advisors will be incorporated into the project activities for the project's fourth and final year. The scientific community at large was reached via peer-reviewed publications (see "Products" section). Natural resource managers, agricultural consultants, and scientists were reached by the project investigators via talks organized in California and internationally (see "Accomplishments" section) The project's foundational-research accomplishments related to big-geodata management were presented to the computer science community. This project's funding partially contributed to the research presented at scientific conferences via presentations and short papers (see "Products" and "Accomplishments" section). These conference contributions contributed to transferring the knowledge on the use of big-geodata (remote sensing time series and other spatial covariates) to map and monitor soil salinity to stakeholders in California and internationally. Changes/Problems:Covid-19 slowed down the research activities of the project considerably. In Year 3, activities at UC Riverside and USDA -ARS (US Salinity Laboratory) resumed to pre-pandemic conditions. A one-year no-cost extension was requested by UC Riverside and approved by NIFA. In Year 4, the team will conclude the planned activities. We expect that all proposed project goals will be accomplished by the end of year 4. What opportunities for training and professional development has the project provided?The training and professional development provided through this project to students and early career scientist in Year 3, were analogous to these in year 1 and 2. They mostly consisted of one-on-one mentorship on weekly or bi-weekly basis and the opportunity to present scientific work at local, national, and international meetings. How have the results been disseminated to communities of interest?Project results were disseminated to the target audience through peer-reviewed publications(see "Products" section in this Progress Report). Project members participated in several conferences, workshops, and meetings where they presented project results to growers, academic researchers and natural resource managers, water districts, local policymakers. Additionally, the PD presented the project accomplishments to scientists that were not part of the project team, including faculty at the University of California in Davis and the Cooperative Extension Division of the University of California, the Data Science Team of the University of California Riverside's Computer Science and Engineering Department, and other academic researchers. Agronomy and Soil Science results dissemination: E. Scudiero and D.L. Corwin: "Mapping Soil Salinity in Micro-irrigated Pistachio Orchards" in Advances in Pistachio Water Management Workshop; July 7, 2022; International Agri-Center, Tulare, CA 93274. Organized by UC Cooperative Extension, UC Davis, California Pistachio Research Board, and the CDFA E. Scudiero "Near-ground and Remote Sensing of Soil Salinity in California Farmland" in "Monitoring Sea-water intrusion in coastal aquifers and Testing pilot projects for its mitigation" 22 June 2022; VENICE, ITALY. INVITED PRESENTATION E Scudiero: "Mapping and monitoring soil salinity across scales in California, USA". Machine Learning to Map and Monitor Soil Salinity. National Salinity Research Center of Iran, VIRTUAL, Dec 2021. INVITED PRESENTATION E Scudiero, A Eldawy, and R Anderson: "High-Resolution Remote Sensing to Characterize Agronomic Processes Across Scales". Management and Adaptation to Aridification in the Western United States. Fall Event of the Water Science and Technology Board, National Academy of Sciences. Washington DC, Virtual. Nov 2021. INVITED PRESENTATION E Scudiero, TH Skaggs, and DL Corwin: "Mapping root-zone agricultural soil salinity across scales in California, USA". Global Symposium on Salt-Affected Soils - Food and Agriculture Organization of the United Nations (FAO), Rome, Italy. Virtual. Oct 2021. INVITED PRESENTATION Guevara, M., Corwin, D. L., Todd-Brown, K. E., Rounsaville, T., Singh, A., Benes, S. E., Quinn, N., Skaggs, T. H., & Scudiero, E.: "Geospatial Measurements of Soil Electrical Conductivity and Saturation Percentage to Support Soil Salinity Assessments across California Irrigated Farmland". 2021 ASA, CSSA, SSSA International Annual Meeting, Salt Lake City, UT. NOV 2021. PRESENTATION Sharon E. Benes, S Singh, Elia Scudiero, U Gull, and DH. Putnam: "Characterization of Spatial and Temporal Variability in Soil Salinity in Relationship to Alfalfa (Medicago sativa L.) Productivity". 2021 ASA, CSSA, SSSA International Annual Meeting, Salt Lake City, UT. NOV 2021. PRESENTATION Sharon E. Benes, S Singh, U Gull, A Anderson, Elia Scudiero, RB Hutmacher, and DH. Putnam: "Characterization of Spatial and Temporal Variability in Soil Salinity in Relationship to Alfalfa (Medicago sativa L.) Productivity". Global Symposium on Salt-Affected Soils - Food and Agriculture Organization of the United Nations (FAO), Rome, Italy. Virtual. Oct 2021. POSTER Computer science result dissemination: Samriddhi Singla, Ahmed Eldawy, Tina Diao, Ayan Mukhopadhyay, and Elia Scudiero. 2021. The Raptor Join Operator for Processing Big Raster + Vector Data. In 29th International Conference on Advances in Geographic Information Systems (SIGSPATIAL '21), November 2-5, 2021, Beijing, China. ACM, New York, NY, USA, 12 pages. https://doi.org/10.1145/3474717.3483971 Singla, Samriddhi, Ayan Mukhopadhyay, Michael Wilbur, Tina Diao, Vinayak Gajjewar, Ahmed Eldawy, Mykel Kochenderfer, Ross Shachter, and Abhishek Dubey. "WildfireDB: An Open-Source Dataset Connecting Wildfire Spread with Relevant Determinants." In 35th Conference on Neural Information Processing Systems (NeurIPS 2021) Track on Datasets and Benchmarks. 2021. Ahmed Eldawy, Vagelis Hristidis, Saheli Ghosh, Majid Saeedan, Akil Sevim, A.B. Siddique, Samriddhi Singla, Ganesh Sivaram, Tin Vu, Yaming Zhang . 2021. Beast: Scalable Exploratory Analytics on Spatio-temporal Data. In Proceedings of the 30th ACM International Conference on Information and Knowledge Management (CIKM '21), November 1-5, 2021, Virtual Event, QLD, Australia. ACM, New York, NY, USA, 12 pages. https://doi.org/10.1145/3459637.3481897 What do you plan to do during the next reporting period to accomplish the goals?Any goals approved by USDA-NIFA that were not fully accomplished, mostly due to slow-downs related to the Covid-19 Pandemic, will be accomplished in year 4 following the agency-approved scope of work.

Impacts
What was accomplished under these goals? For SO1), year 3 accomplishments include the origination of a dataset published in Ag Data Commons which includes near-ground sensor measurements and soil salinity laboratory measurements. The data can be used to test and explore model relationships between ECe, SP, and ECa (EMv and EMh), as well as their spatial variability. In particular, the data may be useful for comparing and testing modeling approaches that account for both deterministic and random components of soil spatial variability at single-field and multi-field scales, and supporting high-resolution digital soil mapping studies across irrigated lands. The project team also carried out soil salinity surveys in Imperial County, Fresno County, and Monterey County. The data collected in these surveys (around 300 soil cores across over 25 fields) are being used as ground truth for the remote sensing of soil salinity models. For SO2), The scientific manuscripts by Corwin et al. (2022; "Modified ECa - ECe Protocols for Mapping Soil Salinity Under Micro-Irrigation". Agricultural Water Management. doi: 10.1016/j.agwat.2022.107640) and Bughici et al. (2022; "Ensemble HYDRUS-2D modeling to improve apparent electrical conductivity sensing of soil salinity under drip irrigation". Agricultural Water Management. doi: 10.1016/j.agwat.2022.107813) describe in detail how soil apparent electrical conductivity directed soil sampling can be used to create maps of soil salinity in micro irrigated orchards. The methodologies described in the two publications are being employed to generate ground-truth maps of soil salinity over 100 locations in Central and Southern California. In Year 2, a controlled experiment to characterize soil moisture and salinity spatiotemporal in a drip irrigated soil was carried out. The analysis of the data collected has begun in year 3 and will be completed in year 4. For SO3) No noteworthy accomplishments for this supporting objective in year 3. Year 4 research efforts will focus on completing SO3 activities. For 4), the Ahmed (co-PD) team continued the development of a web GIS that will make soil salinity maps and other project products available to growers and other stakeholders. In year 3, the team directed their efforts in addressing the stakeholders' and advisors' suggestions on how information should be delivered to users in the web GIS. In Year 4, the web GIS will be published and made publicly and freely available.

Publications

  • Type: Journal Articles Status: Published Year Published: 2022 Citation: Theodor Bughici, Todd H. Skaggs, Dennis L. Corwin, Elia Scudiero: Ensemble HYDRUS-2D modeling to improve apparent electrical conductivity sensing of soil salinity under drip irrigation. Agricultural Water Management. doi: 10.1016/j.agwat.2022.107813
  • Type: Journal Articles Status: Published Year Published: 2022 Citation: Dennis L. Corwin*, Daniele Zaccaria, Elia Scudiero: Modified ECa  ECe Protocols for Mapping Soil Salinity Under Micro-Irrigation. Agricultural Water Management. doi: 10.1016/j.agwat.2022.107640
  • Type: Conference Papers and Presentations Status: Published Year Published: 2021 Citation: Samriddhi Singla, Ahmed Eldawy, Tina Diao, Ayan Mukhopadhyay, and Elia Scudiero. 2021. The Raptor Join Operator for Processing Big Raster + Vector Data. In 29th International Conference on Advances in Geographic Information Systems (SIGSPATIAL 21), November 25, 2021, Beijing, China. ACM, New York, NY, USA, 12 pages. https://doi.org/10.1145/3474717.3483971
  • Type: Conference Papers and Presentations Status: Published Year Published: 2021 Citation: Singla, Samriddhi, Ayan Mukhopadhyay, Michael Wilbur, Tina Diao, Vinayak Gajjewar, Ahmed Eldawy, Mykel Kochenderfer, Ross Shachter, and Abhishek Dubey. "WildfireDB: An Open-Source Dataset Connecting Wildfire Spread with Relevant Determinants." In 35th Conference on Neural Information Processing Systems (NeurIPS 2021) Track on Datasets and Benchmarks. 2021.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2021 Citation: Ahmed Eldawy, Vagelis Hristidis, Saheli Ghosh, Majid Saeedan, Akil Sevim, A.B. Siddique, Samriddhi Singla, Ganesh Sivaram, Tin Vu, Yaming Zhang . 2021. Beast: Scalable Exploratory Analytics on Spatio-temporal Data. In Proceedings of the 30th ACM International Conference on Information and Knowledge Management (CIKM 21), November 15, 2021, Virtual Event, QLD, Australia. ACM, New York, NY, USA, 12 pages. https://doi.org/10.1145/3459637.3481897
  • Type: Book Chapters Status: Published Year Published: 2022 Citation: D. Corwin. "Soil Salinity" in Soil Constraints on Crop Production; EDs.:Neal Menzies, Ram Dalal, Yash Dang; Cambridge Scholars Publishing. Newcastle upon Tyne, United Kingdom. 2022


Progress 07/01/20 to 06/30/21

Outputs
Target Audience:The target audience of this project includes: American farmers, irrigation consultants, state and federal natural-resource scientists and managers, farm advisors, extension specialists, and scientists in the USA and abroad with expertise in soil, crop, and irrigation sciences. During the project's second year, the investigators reached out to California growers and University of California Cooperative Extension personnel, via group and one-on-one interactions, discussing the project goals, asking for feedback, and asking (and receiving) permission to access their land for field data collection. These meetings were mostly carried out remotely. Due to the Covid-19 pandemic, a very limited amount of in-person meetings were conducted with local farmers. The scientific community at large was reached via peer-reviewed publications (see "Products" section). Natural resource managers, agricultural consultants, and scientists were reached by the PD via four invited talks organized in California and internationally: E Scudiero: "Mapping Agricultural Soil Salinity with Near-Ground and Remote Sensing". 2021 Soils and Crops Workshop. University of Saskatchewan, Canada. Virtual. Mar 2021. E Scudiero: "Spatial and Temporal Changes of Soil-Crop-Environment Relationships". University of California GIS WEEK, Virtual. Nov 2020. E Scudiero: "Understanding soil-plant-environment relationships using big geospatial data". UC Riverside 2020 Plants3D Retreat, Virtual. Nov 2020. E Scudiero: "Mapping and monitoring soil salinity across scales". International Salinity Webinar on Resilient Agriculture in Saline Environments Under Changing Climate. Jointly organized by the Central Soil Salinity Research Institute, Karnal, India and the International Center for Biosaline Agriculture, Dubai. Virtual. Nov 2020. These talks contributed to transfer the knowledge on the use of big-geodata (remote sensing time series and other spatial covariates) to map and monitor soil salinity to stakeholders in California and internationally. Such knowledge was developed through this project. During the second year, the PD sought advice about the project's research activities from non-project scientists, including, but not limited to, the scientists that agreed to serve in the Advisory Committee. Changes/Problems:Covid-19 slowed down the research activities of the project considerably. The project effort has been focused on carrying out work locally (e.g., close to Riverside, CA) rather than carrying out extensive state-wide field data collection campaigns. The project team hopes to carry out extensive state-wide field data collection in Year 3, as operational status for UC Riverside and USDA -ARS (US Salinity Laboratory) slowly go back to pre-pandemic conditions. Soil laboratory analyses are behind schedule as the laboratory facilties used for this project (at the USDA-ARS US Salinity Lab) could not be used for most of Year 2 because of Covid-19 facility closure and restrictions. Because of Covid-19, we decided to carry out the intensive monitoring of salinity and moisture dynamics in drip irrigated systems (Specific Objective 2) at the Agricultural Experimental Station of the University of California, Riverside, rather than in sites located in California's Central Valley, as originally planned. What opportunities for training and professional development has the project provided?1) Bi-weekly training for Dr. Theodor Bughici, the postdoctoral scholar hired by the project, and other personnel hired by Dr. Scudiero with the Project Director (PD) Dr. Scudiero, and co-PD Dr. Todd Skaggs, on the use of geophysical measurements and solute transport modeling to characterize salt-affected irrigated agricultural soils. More specifically training and guidance in: a) Soil sensors operation, setup, and installation to monitor soil water content and solute concentration; b) 1D and 2D simulations of water and solute transport modeling of flood irrigated row crops and drip-irrigated tree crops by the use of the HYDRUS software. 2) Co-PD Eldawy provided weekly one-on-one meetings with Ms. Samriddhi Singla, a Ph.D. student hired within the project, and other members of his lab, on topics of big satellite data querying, pre and post-processing, analysis, and visualization. Additionally, training on how to write research papers and other technical reports for peer-reviewed publishing was provided to Ms. Singla in these meetings. 4) Co-PD Eldawy included Ms. Singla in weekly group meetings in which Ms. Singla presented her work and got feedback. There was no Professional Developement activities for students and postdocs in Year 2, because of the limited availability of such events due to Covid-19 How have the results been disseminated to communities of interest?Project results were disseminated to the target audience through peer-reviewed publications, which included several research papers by Singla et al., one book chapter co-authored by Scudiero, and research articles co-authored by Skaggs (see "Products" section in this Progress Report). The PD participated in virtual workshops and meetings where he presented project results to academic researchers and natural resource managers, water districts, local policymakers. Additionally, the PD presented the project accomplishments to scientists that were not part of the project team, including faculty at the University of California in Davis and the Cooperative Extension Division of the University of California, the Data Science Team of the University of California Riverside's Computer Science and Engineering Department, and other academic researchers. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? For irrigation to be sustainable in arid and semi-arid regions, the salt balance of the soil root zone must be maintained by irrigating in excess of a crop's water requirements. The extra irrigation water leaches salts down into the subsurface and prevents harmful salinity build-up near the surface (0-1 m depth). However, the traditional practice of over-irrigating for salinity management is being scrutinized strongly due to increasing water scarcity. By developing novel and actionable engineering methods (including soil and plant sensing, big data computing, and modeling) that will enable precision technologies for real-time soil and plant monitoring and agronomic management, the first-year activities of this project are setting the foundations for optimized water use in salt-affected farmland in California, so that crop production is optimized and salinity accumulation in the soil profile is minimized. The research in this project is discovering novel and actionable means to (Objective 1) provide broad-scale and high-resolution inventories of soil salinity using remote sensing and data science tools; (Objective 2) measure and monitor salinity in micro irrigated orchards (for which there is no currently any reliable established method) using a combination of ground and remote measurements and data science tools; (Objective 3) use the generated salinity maps to model water requirements; and (Objective 4) create a web-based geographical information system to display the project products to stakeholders. For 1), the teams of Scudiero (PD), Corwin (co-PD), and Skaggs (Co-PD) have harmonized the ground-truth dataset for soil salinity and soil texture that was compiled in year 1. Machine learning (e.g., random forest) regression was evaluated to generate field-scale maps of soil salinity and texture based on field-scale sensor measurements of apparent electrical conductivity (ECa). When leveraging information from hundreds of historical soil salinity surveys, machine learning can be used to predict soil salinity, in soils with ECa greater than 1.5 dS/m, without additional ground truth measures with reasonable accuracy. In soils with lower ECa sensor measurements, ground truth soil cores and laboratory analyses are still required to obtain accurate field-scale soil salinity maps. The research team has led the use of remote sensing time series (e.g., Landsat satellites providing multispectral information at the 30x30 m resolution) and other geodata using machine learning. The remote sensing time series were used to identify relative differences (from the field to the whole-state scale) in crop phenology for the crops grown in California. Such phenology differences were compared to the available ground-truth soil salinity measurements to generate a machine learning prediction model for root-zone (0-1.2 m) soil salinity. Historical gridded weather data and ancillary soil information (e.g., soil texture maps from public repositories) were used to increase the accuracy of the soil salinity predictions. Next year, the team will further refine and evaluate (cross-validation and independent data) the remote-sensing-based soil salinity model. The teams of Scudiero (PD) and Corwin (co-PD) collected ground-truth data from agricultural sites in the Central Valley (2 fields), Imperial Valley (7 fields), San Jacinto Valley (3 fields), and the Riverside area (3 fields) in California. Laboratory analyses of the collected soil samples are still ongoing. Laboratory work was hindered by facilities closure due to the Covid-19 pandemic. In data acquisition and distribution, the Eldawy (Co-PD) finished the distributed loading of large-scale satellite data in Apache Spark, distributed computing engine. We also finished the distributed data loading and processing by introducing the Raptor Zonal Statistics (RZS) operator which was published in IEEE Big Data 2020. These components facilitated the building of the ground-truth data that will be used to calibrate the proposed data models. For 2), Dennis Corwin (co-PD) led the use of collected ground data from pistachio orchards to create ECa-directed soil salinity maps in drip-irrigated fields. The research was submitted (under review) to the Agricultural Water Management journal. Additionally, Scudiero's and Skaggs' teams carried out detailed geophysical (Electromagnetic induction, electrical resistivity tomography, and time-domain reflectometry) time-lapse measurements to characterize soil moisture and salinity spatiotemporal variability in a 0.2-acre drip-irrigated plot at the Agricultural Experimental Station of UC Riverside. The field measurements were collected during the spring and summer of 2021. In Year 3, Scudiero and Skaggs' teams will use the collected data to calibrate two-dimensional solute transport modeling to describe spatial and temporal variability of soil moisture and salinity in the soil profile with minimal laboratory soil data (e.g., without determining the water retention curve) For 3) Dr. Theodor Bughici, a postdoctoral scholar hired within this project, investigated the use of space weather data to model spatial and temporal variability of soil salinity at the field scale. This research is still being developed and has not produced accurate salinity estimations from the hydrologic modeling yet. This research will continue in Year 3. For 4), the Ahmed (co-PD) team is actively working on the development of the front-end web-based interface. A preliminary prototype was demonstrated in the International Conference on Very Large Data Bases (VLDB) 2019. The team has since worked on improving it to work with the models that we produce in this project.

Publications

  • Type: Journal Articles Status: Published Year Published: 2020 Citation: 4. 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
  • Type: Conference Papers and Presentations Status: Published Year Published: 2021 Citation: 5. Tina Diao, Samriddhi Singla, Ayan Mukhopadhyay, Ahmed Eldawy, Ross Shachter, and MykelKochenderfer. Uncertainty Aware Wildfire Management, In Proceedings of the AI for Social Good - AAAI Fall Symposium 2020, November, 2020.
  • Type: Journal Articles Status: Published Year Published: 2021 Citation: Helalia, S.A., Anderson, R.G., Skaggs, T.H., Jenerette, G.D., Wang, D., `im?nek, J., 2021. Impact of Drought and Changing Water Sources on Water Use and Soil Salinity of Almond and Pistachio Orchards: 1. Observations. Soil Systems 5, 50.
  • Type: Journal Articles Status: Published Year Published: 2021 Citation: Helalia, S.A., Anderson, R.G., Skaggs, T.H., `im?nek, J., 2021. Impact of Drought and Changing Water Sources on Water Use and Soil Salinity of Almond and Pistachio Orchards: 2. Modeling. Soil Systems 5, 58.
  • Type: Book Chapters Status: Published Year Published: 2021 Citation: Hopmans, J.W., Qureshi, A.S., Kisekka, I., Munns, R., Grattan, S.R., Rengasamy, P., Ben-Gal, A., Assouline, S., Javaux, M., Minhas, P.S., Raats, P.A.C., Skaggs, T.H., Wang, G., De Jong van Lier, Q., Jiao, H., Lavado, R.S., Lazarovitch, N., Li, B., Taleisnik, E., 2021. Chapter One - Critical knowledge gaps and research priorities in global soil salinity, In: Sparks, D.L. (Ed.), Advances in Agronomy. Academic Press, pp. 1-191. doi: https://doi.org/10.1016/bs.agron.2021.03.001
  • Type: Journal Articles Status: Published Year Published: 2021 Citation: 1. Samriddhi Singla, Ahmed Eldawy, Tina Diao, Ayan Mukhopadhyay, and Elia Scudiero.Experimental Study of Big Raster and Vector Database Systems, In International Conference on Data Engineering, 2021. DOI>10.1109/ICDE51399.2021.00231
  • Type: Conference Papers and Presentations Status: Published Year Published: 2020 Citation: 2. Samriddhi Singla, Tina Diao, Ayan Mukhopadhyay, Ahmed Eldawy, Ross Shachter, and Mykel Kochenderfer. WildfireDB: A Spatio-Temporal Dataset Combining Wildfire Occurrence with Relevant Covariates, In AI for Earth Sciences Workshop at NeurIPS, 2020
  • Type: Conference Papers and Presentations Status: Published Year Published: 2020 Citation: 3. Samriddhi Singla, and Ahmed Eldawy. Flexible Computation of Multidimensional Histograms, In The 2nd ACM SIGSPATIAL International Workshop on Spatial Gems (SpatialGems 2020), November, 2020.
  • Type: Book Chapters Status: Accepted Year Published: 2021 Citation: 6. James D. Oster, Nigel W.T. Quinn, Aaron L.M. Daigh, Elia Scudiero. Agricultural subsurface drainage water: an unconventional source of water for Irrigation. In: Unconventional Water Resources, Editor: Qadir, M, Smakhtin, V., Koo-Oshima, S., Edeltruaud, E. Springer. 2021.
  • Type: Journal Articles Status: Under Review Year Published: 2022 Citation: Corwin, D.L., Scudiero, E., Zaccaria, D., Modified EC a  EC e Protocols for Micro Irrigation Soil Salinity Patterns. Under Review. Agricultural Water Management
  • Type: Journal Articles Status: Under Review Year Published: 2022 Citation: Elia Scudiero, Dennis L. Corwin, Paul T. Markley, Alireza Pourreza, Tait Rounsaville, and Todd H. Skaggs. A platform for on-the-go soil sensing in micro-irrigated orchards. Computers and Electronics in Agriculture. UNDER REVIEW


Progress 07/01/19 to 06/30/20

Outputs
Target Audience:The target audience of this project includes: American farmers, irrigation consultants, state and federal natural-resource scientists and managers, farm advisors, extension specialists, and scientists with expertise on soil, crop, and irrigation sciences. During the project first year, the lead investigators reached out to California growers, via one-on-one interactions, discussing about the project goals and preliminary results, asking for feedback, and by asking (and receiving) permission to access their land for field data collection. The scientific community at large was reached via peer-reviewed publications (see "Products" section). Natural resource manager, both in the private and public sectors, were reached by the PD at a conference in Las Vegas in February 2020, organized by the Multi State Salinity Coalition. At this meeting, the PD reached out to stakeholders involved in technologies for desalination and reuse of saline water, soil and water salinity control strategies, water/energy efficiencies, and related public policies that assist communities across California and the U.S. southwest. During the first year, the PD reached out to an external advisory board. The advisory board was limited to two people with expertise in project management, experimental setup, and outreach to growers. In the second and third year, the board will be expanded to all members identified during proposal preparation and project initiation. Changes/Problems:Field-data collection (i.e., the soil salinity surveys) has been halted since March 2020 because of COVID-19 pandemic. Following the directions of the CA Governor, the University of California Riverside (UCR) and the USDA-ARS U.S. Salinity Laboratory were closed to avoid spread of COVID-19. Field-data collection will be resumed in years 2 and 3 of the project, as soon as COVID-19 travel restriction for this type of field work will be lifted by UCR. It is too early to assess the impact of this problem on the project success. It is likely that field-data i) collected by the project before the outbreak of COVID-19 and ii) obtained by harmonizing historical datasets may be sufficient to calibrate the proposed regional scale models for soil salinity estimations in agronomic crops (Objective 1). Additional field data is needed to fully accomplish Objective 2; however, most of field activities related to Objective 2 are planned for project years 2 and 3. What opportunities for training and professional development has the project provided?Training activities 1) Bi-weekly training for Dr. Theodor Bughici, the postdoctoral scholar hired by the project, with the Project Director (PD) Dr. Scudiero, and co-PD Dr. Todd Skaggs, on the use of geophysical measurements and solute transport modeling to characterize salt-affected irrigated agricultural soils. More specifically training and guidance in: a) Soil sensors operation, setup and installation in order to monitor soil water content and solute concentration; b) 1D and 2D simulations of water and solute transport modeling of flood irrigated row crops and drip irrigated tree crops by the use of the HYDRUS software. 2) The co-PD Ahmed Eldawy provided one-to-one training to a M.Sc. student (Ms. Husna Sayedi) on the query, management, and employment of satellite data to compute country-level environmental information, e.g., surface temperature. Additionally, the student was trained on using web map interfacing to visualize the generated spatial information into user-friendly web interfaces. 3) Co-PD Eldawy provided weekly one-on-one meetings with Ms. Samriddhi Singla, a Ph.D. student hired within the project, on topics of big satellite data querying, pre and post processing, analysis, and visualization. Additionally, training on how to write research papers and other technical reports for peer-reviewed publishing was provided to Ms. Singla in these meetings. 4) Co-PD Eldawy included Ms. Singla in weekly group meetings in which Ms. Singla presented her work and got feedback. Professional development 1) On November 2019, Dr. Bughici gave an oral presentation on Postdoctoral Symposium 2019 Held by the University of California - Riverside Postdoctoral Association. In the presentation Dr. Bughici presented on data-driven irrigation scheduling methods. 1) In January 2020, Dr. Bughici attended the "W4188: Soil, Water, and Environmental Physics to Sustain Agriculture and Natural Resources, 2020 Annual Meeting" in Las Vegas, NV, USA, accompanied by the PD. 2) Throughout the fall of 2019-2020, monthly, Dr. Bughici attended the monthly seminars organized by the University of California Riverside, Department of Environmental Sciences, accompanied by the PD and co-PD Skaggs. The seminar series featured talks from invited speakers on topics related to soil sciences, agricultural water management, hydrology, and environmental sciences. 4) Throughout the fall of 2019-2020, Dr. Bughici attended the monthly seminars organized by the USDA-ARS U.S. Salinity Laboratory, accompanied by the PD and co-PD Skaggs. The seminar series featured talks from Salinity Laboratory's researchers and staff on topics related to soil salinity and its effects on plant growth, agriculture management and environment. 5) Ms. Singla attended the 27th International Conference on Advances in Geographic Information Systems (also known as ACM SIGSPATIAL 2019) in November 2019, Chicago, IL, USA. At the conference, co-PD Eldawy guided and mentored Ms. Singla on how to outreach to other research groups in the geospatial community to expand impact of the project's research outputs and products. How have the results been disseminated to communities of interest?Project results were disseminated to target audience through peer-reviewed publications, which included a research paper by Singla et al. and two review papers by Corwin and Corwin and Scudiero (see "Products" section in this Progress Report). The PD participated in workshops and meetings where he presented project results to academic researchers (at the OneWater Workshop in El Paso, TX, in November 2019) and to natural resource managers, water districts, local policy makers, and saline water management specialists from private and public sectors (at the Multi-State Salinity Coalition's meeting in Las Vegas, NV, in February 2020). Additionally, the PD presented the project's annual accomplishments to external advisors of the project, including Dr. Daniele Zaccaria from University of California in Davis and Dr. Ray Anderson form the USDA-ARS US Salinity Laboratory in Riverside. During the second and third year of the project, the advisory board will be expanded to include other members that have already volunteered to serve as advisors for this project, including growers and personnel form California's Department of Water Resources. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? For irrigation to be sustainable in arid and semi-arid regions, the salt balance of the soil root zone must be maintained by irrigating in excess of a crop's water requirements. The extra irrigation water leaches salts down into the subsurface and prevents harmful salinity build-up near the surface (0-1 m depth). However, the traditional practice of over-irrigating for salinity management is being scrutinized strongly due to increasing water scarcity. By developing novel and actionable engineering methods (including soil and plant sensing, big data computing and modeling) that will enable precision technologies for real-time soil and plant monitoring and agronomic management, the first-year activities of this project are setting the foundations for optimized water use in salt affected farmland in California, so that crop production is optimized and salinity accumulation in the soil profile is minimized. The research in this project is discovering novel and actionable means to (Objective 1) provide broad-scale and high resolution inventories of soil salinity using remote sensing and data science tools; (Objective 2) measure and monitor salinity in micro irrigated orchards (for which there is no currently any reliable established method) using a combination of ground and remote measurements and data science tools; (Objective 3) use the generated salinity maps to model water requirements; and (Objective 4) create a web-based geographical information system to display the project products to stakeholders. For 1), investigators Scudiero (PD), Corwin (co-PD), and Bali (co-PD) have been compiling ground datasets of soil salinity measurements across agricultural land in California. The dataset includes over 150 time x location sites with average size of about 18 hectares from historical (before the project start, since the mid-1980s) field-data collected by the project investigators and other scientists from the University of California, California State University, and USDA-Agricultural Research Service. In addition, during the first year of the project, data from agricultural sites in the Central Valley (17 fields), Blythe area (9 fields), and Coachella Valley (3 fields) were collected. These soil salinity measurements are being used to develop remote-sensing based state-wide maps of soil salinity for California farmland. Co-PD Eldawy developed query and analysis methods that enable state-wide real time (under an hour) use of big-data from satellites (e.g., 800 billion pixels) and other geodata. The developed methodology was tested on plant (e.g., vegetation indices) and environmental parameters (e.g., temperature) over the state of California. The processed data is being used in the salinity assessment models that the project team is developing. For 2), soil samples at twelve micro-irrigated orchard sites were collected by co-PD Corwin and his team. The field data indicates that, depending on the soil type and irrigation management, most of salts build up between 0.6 to 1.2-m away from the micro-irrigation emitters. Following these findings, Scudiero, Corwin, and Skaggs developed a mobile system to measure on-the-go soil apparent electrical conductivity and gamma ray spectrometry in orchards. Development notes, experimental field data examples collected at a 1-acre micro irrigated orange orchard the University of California Riverside Experimental Agricultural Station, and a detailed description of the platform were included in a manuscript that is currently being revised prior to submission to the journal Computer and Electronics in Agriculture (Scudiero et al. "A novel platform for on-the-go soil sensing in micro-irrigated orchards"). Additionally, under the guidance of Skaggs and Scudiero, Dr. Theodor Bughici, a postdoctoral scholar hired within this project, used two-dimensional solute transport modeling to describe spatial and temporal variability of soil moisture and salinity in the soil profile. Bughici formulated this modeling framework based on the field data (soil and crop water use on micro-irrigated pistachio) collected by co-PD Corwin and his collaborators. Bughici then used an ensemble approach to model likely scenarios for spatial and temporal changes of moisture and salinity patterns on different soil types, crop types, irrigation management strategies, water quality. In project years 2 and 3, field data will be acquired to validate and further improve the developed models. The outputs from these two-dimensional will be used to i) guide on-the-go sensor surveys by providing information on where, in relation to the micro-irrigation emitters, most of the salt is expected to accumulate in the soil profile; ii) estimate how soon after an irrigation event, the areas with high salt accumulation are going to have sufficient moisture content (i.e., at or close to field capacity) to allow reliable sensor measurements of apparent electrical conductivity ; iii) guide soil sampling in order to characterize short-scale heterogeneity of soil salinity; and iv) interpret on-the-go sensor measurements of apparent electrical conductivity, in order to distinguish signal from soil moisture from that of salinity. For 3), Theodor Bughici, under the supervision of PD Scudiero and co-PD Skaggs, carried out literature review on available models to describe broad scale water use, hydrology, and crop-production. In project years 2 and 3, the project investigators will use the products from objectives 1 and 2, and from publicly available data (e.g., weather and evapotranspiration estimations) to generate a decision support tool for irrigation in salt-affected soils in California. For 4), co-PD Eldawy and his mentee, Ms. Samriddhi Singla, a graduate student hired within this project, have developed preliminary versions of geographical information systems interfaces that are able to show stat-wide maps and vector (point or polygon) information, following real-time query and statistical modeling of remote sensing data and other environmental covariates (soil texture maps, spatial weather data). In project years 2 and 3, co-PD Eldawy and his team will include the project products from objectives 1-3 in a publicly available web tool.

Publications

  • Type: Journal Articles Status: Published Year Published: 2019 Citation: Singla, S., Eldawy, A., Alghamdi, R., Mokbel, M.F., 2019. Raptor: large scale analysis of big raster and vector data. Proceedings of the VLDB Endowment 12(12), 1950-1953.
  • Type: Book Chapters Status: Published Year Published: 2019 Citation: Corwin, D.L., Scudiero, E., 2019. Chapter One - Review of soil salinity assessment for agriculture across multiple scales using proximal and/or remote sensors. In: D.L. Sparks (Ed.), Advances in Agronomy. Academic Press, pp. 1-130. doi: doi:10.1016/bs.agron.2019.07.001.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2020 Citation: E Scudiero: Agricultural Salinity Management via an Integration of Proximal and Remote Sensing with Big Geodata Modeling. 2020 MSSC Annual Salinity Summit. Las Vegas, NV, USA. Feb 2020
  • Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: E Scudiero: Broad-scale Agricultural Soil Salinity Assessment. One Water Workshop. UT El Paso, TX, USA. Nov 2019
  • Type: Conference Papers and Presentations Status: Published Year Published: 2020 Citation: E Scudiero: Digital Agronomy. W4188: Soil, Water, and Environmental Physics to Sustain Agriculture and Natural Resources, 2020 Annual Meeting, Las Vegas, NV, USA. Jan 2020.
  • Type: Journal Articles Status: Published Year Published: 2020 Citation: Corwin, D.L., Climate change impacts on soil salinity in agricultural areas. European Journal of Soil Science n/a(n/a). doi: 10.1111/ejss.13010.