Funded Research Projects
Satellite and UAS Imagery Use to Implement Timely Irrigation Strategies
Jose Chávez, Colorado State University; Daran Rudnick, University of Nebraska Lincoln; Juan Enciso, Texas A&M University; Florence Cassel, Fresno State University
Industry partner: Lindsay Corporation
Abstract: Unmanned Aerial Systems (UASs) and satellites have been recognized as excellent platforms to provide near real time feedback of temporal and spatial conditions found in agricultural fields throughout the growing season. UASs have also allowed selecting the best management practices that optimize the management of soil and water resources. The main objective of this study is to assess the accuracy of using UASs and satellites (multispectral imagery) to determine crop actual water use and soil water deficit. This study will be conducted at two research facilities in Colorado, one research station in NE, at a commercial orchard and in Weslaco, TX, one research site in Garden City, KS; and at commercial farms in Fresno, CA. Remote sensing (RS) imagery (0.05 – 5 m pixel size) will be acquired over research fields to determine crop evapotranspiration (ET) and soil water deficits on a daily basis. Eddy covariance and soil water sensors/probes will be used to assess remote sensing platform and crop ET model (e.g., reflectance crop coefficient, energy balance) weekly, monthly, and seasonal accuracy
Field Testing of Precision Ag Irrigation Data Exchange Standards
Stephen Smith, Colorado State University
Industry partner: AgGateway
Abstract: The proposed project addresses a fundamental issue of multiple and conflicting formats used by different manufacturers, governmental agencies and academic organizations who are involved in data-driven irrigation solutions. These differences are a key barrier to farmers adopting precision irrigation technologies. Big data solutions rely upon the capture and analysis of multiple sets of data from various pieces of equipment and databases to support analysis, planning, operations and reporting of results. The proposed project is to field test a data exchange standard for agricultural irrigation developed by a team of industry and academic experts working with AgGateway. The Precision Ag Irrigation Language (PAIL) standard represents weather, soil moisture and other relevant data, currently stored in a variety of Original Equipment Manufacturer (OEM) formats, in an industry-wide format that can be used by irrigation data analysis and prescription programs. The standard has been approved by the American Society of Agricultural and Biological Engineers (ASABE) and the American National Standards Institute (ANSI). However, it has not been field tested for use with precision drip irrigation solutions. Members of the PAIL team will train university personnel who are already engaged in, or planning, an irrigation planning and operations project on the PAIL standard and how to apply it. Participants will document how they are using the PAIL standard, issues that occur, and resolutions or recommendations for those issues. These results will be used to enhance the support and documentation of the standard in order to accelerate market adoption. The standard will be open for use by those developing big data solutions, where the standard can provide a fundamental base of compatibility.
Multi-Environment Vertical Agricultural Technologies: Innovative Irrigation And Monitoring Solutions With Machine Learning Integration
Tim Jacobsen and Shawn Ashkan, Fresno State University
Industry partners: Vertical Irrigation LLC and Trestle Optics
Abstract: Vertical indoor farming is capturing a larger market share of food production, yet problems in this nascent field remain. Indoor farming benefits from reduced water consumption by employing closed system techniques such as hydroponics and aquaponics, however, optimizations can be further improved with systems automation and waste reduction. Reducing operation expenses is achievable by deploying passive and reusable sensors which acquire actionable information that can then be used for machine learning optimizations. Our state-of-the-art opto-electronic and photonic sensors monitor environmental conditions, and, when paired with linear regression-based analysis techniques can correlate the signal outputs to product yield and operation efficiencies. The goal of this proposal is to study the efficacy of this “big-data” approach to agriculture by comparing profit margins of vertical farming systems with and without the use of our custom remote monitoring system over multiple grow cycles. We expect to see a simplified work flow by automating plant and systems monitoring, which reduces labor cost, and we expect a more uniform and high-volume product yield, both of which boost profit margins.
Next Generation Soil Moisture Technology for Improving Irrigation Management in Turfgrass and Landscape
Cathie Lavis, Dale Bremer, Jack Fry and Jared Hoyle, Kansas State University; Benjamin Wherley and Ambika Chandra, Texas A&M AgriLife Research; Jay Ham, Tony Koski and Yaling Qian, Colorado State University
Industry partner: The Toro Company
Abstract: Landscape irrigation strategies usually rely on calendar or evapotranspiration (ET) schedules that completely ignore soil moisture. Soil moisture sensors (SMS) may improve irrigation accuracy but are not widely used. Our proposed objectives are to: 1) Evaluate irrigation trigger thresholds by measuring turf canopy responses to soil moisture using SMS; and 2) Compare SMS-based irrigation scheduling effectiveness among several SMS types and with traditional and ET-based irrigation scheduling. The study would be comprised of three field studies, two under rainout shelters and a third in a larger, uncovered area, and would be conducted on three turfgrasses. The new, low-cost SMS with internet of things (IoT) connectivity (wireless underground sensor network, or WUSN), developed by Dr. Ham, CSU, will also be evaluated in our proposed studies. Our overall goal/outcome is to conserve water by accelerating wide-spread use of SMS for landscape irrigation control by reducing costs and complexity in their use. Letters are included that demonstrate strong industry support for this project, including matching funds from the USGA.
Expanding and Operating the Parallel 41 Flux Network to Support Real-time Evapotranspiration Estimates from Remote Sensing
Christopher Neale, Daugherty Water for Food Global Institute at the University of Nebraska
Industry partner: LICOR
Abstract: The objective of this project is to augment the Parallel 41 network of eddy covariance flux stations being implemented initially in the Central Plains of the US. The purpose of this network is to provide real-time, quality controlled and processed crop and natural vegetation evapotranspiration, an important parameter for irrigation water management and water balance studies in watersheds and groundwater recharge estimations. The network is being partially established with 2018 first-year funding from the IIC with 7 towers located in Iowa (2), Nebraska (4) and Colorado (1). The 2019 funding will facilitate 2 additional towers, one in Iowa and one in Kansas. The eddy covariance flux towers will be networked together using the new LICOR FluxSuite software app and SmartFlux hardware installed at each tower, that conducts real-time processing and corrections. The real-time evapotranspiration data will be used to anchor satellite-based estimates of evapotranspiration using energy balance modeling approaches, allowing the distribution of the real time estimates to multiple irrigated fields and watersheds. The goal is to produce a reliable satellite-based spatial product that will be used by water managers in the participating states as well as farmers and irrigators through online and cell phone apps.
AI Model for Estimating Crop Water Demand: An Artificial Intelligence (AI) Model to Improve Agricultural Water Use Efficiency Using Field, Plant, and Weather Data
Fayzul Pasha, Ph.D., P.E., California State University Fresno
Industry partners: WiseConn, PowWow, AgH2O, Dynamax, Jain Irrigation, and Irrometer
Abstract: California State University, Fresno (Fresno State)—in partnership with six technology and service companies including WiseConn, PowWow, AgH2O, Dynamax, Jain Irrigation, and Irrometer—proposes AI Model for Estimating Crop Water Demand: An Artificial Intelligence (AI) Model to Improve Agricultural Water Use Efficiency Using Field, Plant, and Weather Data – Case Study (AI model). In this project, the Project Team will develop, validate, and test an AI model using previously collected plant-water-soil continuum data from a well-equipped olive orchard located at the University Agricultural Lab at Fresno State. An innovative quasi real-time AI model will be developed to estimate crop water demand with higher accuracy by utilizing both field and data science. Numerical analysis including sensitivity and parametric analysis will be conducted and machine learning approaches will be used to develop the model. The model will be executed and tested to real-world application for benefit analysis. Using farmer-defined and science-based parameters, this AI model will improve water use efficiency associated with crop production. Project Team anticipates that this system will reduce farm-wide water consumption by at least 10 percent or more. The AI model will offer a baseline to the growers to measure and compare their decision-making process and will provide a basis to improve their water management efficiency.
Development of a Low Pressure and Low Flow Water and Energy Efficient Media Filtration System
Kaomine S. Vang, Center for Irrigation Technology-Fresno State
Industry partner: Perigo LLC
Abstract: One of the most commonly used filtration systems in agriculture is sand media filtration. They are large tanks made of metal or other material that have large water flow capacities. Sand media filters require very little maintenance, which has made them one of most widely used filtration devices for farmers. However, media tanks and other filtration devices on the market require large amounts of pressure to sufficiently filter out contaminants. In general, after filtration there is some quantity of media and contaminants that is discharged with water. The water is lost and so is the energy used for pressure. This project intends to research new technology that can perform the same function with less pressure and water loss, thus being more efficient.