While scientists have been working on satellite-based approaches for mapping evapotranspiration (ET) for more than 30 years, the data have historically been difficult for agricultural producers and irrigation professionals to access, notes Forrest Melton, senior research scientist in the earth sciences division at NASA’s Ames Research Center in California.
To address this data gap, a team of scientists and software engineers from NASA, the U.S. Geological Survey (USGS), the U.S. Department of Agriculture (USDA) and multiple U.S. universities have worked for the past eight years to develop a fully automated system for producing and distributing satellite-based ET data for the United States.
The system, OpenET, uses satellite observations from Landsat and other satellites in combination with gridded meteorological data to calculate ET at daily, monthly, seasonal and annual timescales using six well-established, satellite-driven ET models, computing a single ET value for every location and time step.
OpenET, an open science effort coordinated by the nonprofit OpenET Inc., was developed to address the lack of easily accessible, consistent and reproducible information on ET by making it freely available via multiple open data services.
OpenET uses Google Earth Engine as a shared computing platform, allowing the science team to work together to improve the consistency of model inputs and evaluate and improve the satellite-driven models.
Data are publicly available from October 1999 to present day. Current geographic coverage includes the 23 westernmost U.S. states, with work ongoing to provide data for the 48 contiguous states. Data availability, at press time, is anticipated later this year.
Melton notes that satellites operated by NASA, USGS and the National Oceanic and Atmospheric Administration circle the Earth, collecting invaluable information on crop and vegetation conditions, atmospheric conditions and weather patterns, natural disasters, and other factors directly affecting agriculture and the economy.
The data are freely available to the public and increasingly being used by U.S. agricultural producers and water managers to improve the reliability of water supplies for agricultural production.
The data provided by OpenET offers measurements of the amount of water moving from soil and plants into the atmosphere as water vapor, along with information on the health of crops and other vegetation.
The OpenET Application Programming Interface, or API, supports automated data retrievals and integration with other water data and farm management decision support systems.
The OpenET Consortium also completed the largest published accuracy assessments to date for field-scale satellite-based ET data, including data from more than 70 agricultural fields and dozens of crop types.
Melton says the Landsat series of satellites and the ECOSTRESS mission aboard the International Space Station collect measurements of land surface temperature, allowing biophysical models to compute ET across the globe at spatial scales as small as a quarter-acre of land.
“Many models used by scientists to compute ET from satellite data operate on the principle that the process of ET consumes energy,” Melton notes. “As water evaporates, it consumes energy and cools the surface from which it is evaporating. This process occurs continuously across the landscape.”
Areas with more ET have more evaporative cooling and lower land surface temperatures, while areas with less ET have higher land surface temperatures.
“These patterns are clearly evident in satellite imagery of land surface temperature in places like the arid western U.S., with wetlands and well-irrigated agricultural fields standing out against the surrounding landscape during the growing season,” says Melton. “Scientists use physically based models to combine satellite observations of land surface temperature, along with additional satellite measurements of vegetation extent and condition, with gridded weather data to compute ET for every pixel in a satellite image.”
Last March, OpenET released the OpenET FARMS mobile-friendly user interface — designed in partnership with more than 50 agricultural producers and ag-tech companies — to make OpenET data even more accessible and easy to use for the agricultural and water resources management communities.
Users can view or download data in multiple formats to increase field usability for select regions of interest, time periods and variables, and create automated recurring reports for areas of up to 150,000 acres in size.
The Landsat satellites, built and launched by NASA and managed by the USGS, are particularly important for mapping ET because the instruments provide optical and thermal infrared measurements at field scales (30 meters to 100 meters, or 0.22 acres to 2.4 acres per pixel). “This unique combination of observations provides the key measurements of vegetation condition and land surface temperature the physically based models use to compute ET,” Melton says.
NASA, USGS and USDA also support research that develops and advances the satellite-driven ET models to ensure they continue to provide data that meet user requirements for accuracy, timeliness and consistency across large geographic areas.
Actual ET data available in near real time and the recent past can be highly valuable in making irrigation decisions, as ET represents the amount of water that has been transferred from the soil to the atmosphere and that needs to be replaced through irrigation or precipitation to maintain soil moisture and avoid unintentional crop water stress, says Melton.
“Scheduling irrigation amounts to replace this volume of water that is being consumed through ET is one of the best ways to maximize irrigation efficiency and minimize unintentional leaching of nutrients below the bottom of the root zone,” he says, adding that the data also enable irrigation professionals to respond to the effects of heat waves and flash droughts.
OpenET will soon make data available as far back as October 1985, near the start of the Landsat 5 data record, providing invaluable information in understanding changes in crop water needs and updating irrigation systems. The information can also be helpful in designing new irrigation systems in regions like the Southeastern U.S. that have not historically needed irrigation but are beginning to experience more frequent flash droughts that can negatively affect crop yields without irrigation, says Melton.
“Maximizing irrigation efficiency can have multiple benefits, including reducing costs for electricity for pumping water, minimizing unintentional leaching of fertilizers below the root zone and maximizing crop yields by ensuring crops receive sufficient irrigation while minimizing periods of soil saturation,” he says.
Research in vegetable crops in the Salinas Valley in California, led by Michael Cahn, PhD, and Richard Smith, PhD, of the University of California Cooperative Extension, has shown that use of ET-based irrigation management can reduce applied water by up to 27% and fertilizer applications by up to 31% while sustaining yields.
Recent studies by RTI International, USDA and the Almond Board of California estimated that the value of ET-based irrigation scheduling tools, like the ET Toolkit, to almond farmers is up to $45.5 million over five years, while the total economic benefits are projected to reach $127.6 million over the period from 2028 to 2033.
OpenET currently provides data on total ET from all sources, including irrigation, precipitation, access to shallow groundwater and soil moisture stored in the root zone. “It is important to account for effective precipitation in converting total actual ET to the ET from water applied for irrigation,” Melton says. “Additional calculations are required to convert a daily or weekly ET value into an irrigation system run time. It is important for irrigation professionals to account for system application rates, distribution uniformity, salinity management and other site-specific considerations.”
One challenge associated with satellite instruments mapping ET is that they cannot see through clouds, which can increase the error in the near-real-time satellite-derived ET data during cloudy periods with long gaps between cloud-free satellite observations.
The Landsat 8 and Landsat 9 satellites currently provide an observation every eight days for most locations in the U.S., excepting clouds, which can result in a lag of a few days to a few weeks — cloud-free observation is needed to account for rapid changes in field conditions due to events such as a recent rainstorm, harvest or change in irrigation management, Melton says.
The Landsat overpass usually occurs between 10 a.m. and 11 a.m. local time, which can potentially introduce a bias in ET values for crops with significant daily variability in stomatal conductance during some parts of the growing season.
The OpenET Consortium is working to integrate data from additional satellites including ECOSTRESS and Sentinel-2 to help overcome these challenges.
Kendall DeJonge, PhD, research agricultural engineer at the USDA Agricultural Research Service (ARS) Water Management and Systems Research Unit in Fort Collins, Colorado, notes “there is more ET when it is sunny, hot, dry and/or windy. A larger, more mature plant with many leaves will have more ET than a young crop.”
DeJonge says crop productivity is maximized by meeting ET demands through supplying water to the crop through rainfall and/or irrigation. “The soil acts as a reservoir to hold water for the plant, which may be adequate for several days or more than a week depending on soil properties,” he says. “If the plant uses up too much water in the soil, it will exhibit signs of water stress, which could impact plant health and crop yields.
“Because irrigation systems vary in frequency, capacity and efficiency, having knowledge of crop ET needs helps with irrigation timing and amount. In operations with multiple fields, the workflow for irrigation management should consider the crop’s continuous ET needs.”
Several states’ weather station network reports reference ET amounts that can be adjusted to specific crops. “Irrigators can avoid overwatering, which can save money on pumping costs or water if irrigation allocations are limited, as well as minimize runoff and nutrient loss,” DeJonge adds. “Precise, ET-based irrigation paired with a good knowledge of soil water-holding capacity can also help avoid costly yield losses due to water stress.”
“Precise, ET-based irrigation paired with a good knowledge of soil water-holding capacity can … help avoid costly yield losses due to water stress.”
— Kendall DeJonge, PhD, agricultural engineer, USDA ARS Water Management and Systems Research Unit
Kyle Knipper, physical research scientist at the USDA ARS Sustainable Agricultural Water Systems Research Unit in Davis, California, notes satellites add field-level measurements of actual ET (ETa) — the water a crop and soil really use — by blending thermal and optical imagery with weather data in physically based models.
ETa reflects real conditions — partial canopy, cover crops, stress, wind exposure, system nonuniformity — not just a typical crop under ideal water supply and conditions. “OpenET’s ETa is the consumptive use signal that captures those differences, so you can calibrate run times to what happened in each block and verify whether deficit strategies are truly occurring on the ground,” Knipper says.
Knipper says his unit is testing OpenET in California specialty crop systems through the USDA-led GRAPEX (vineyards) and T-REX (almonds and olives) projects, evaluating it against the eddy-covariance towers of ETa across the Central Valley and diverse management styles such as cover crops, row orientation and advective conditions with promising evaluation metrics capturing seasonal patterns and spatial differences at field scale within expected uncertainty ranges.
John Volk, PhD, associate research scientist in hydrology at the Desert Research Institute in Nevada, led the research and writing on a paper, “Assessing the accuracy of OpenET satellite-based evapotranspiration data to support water resource and land management applications,” published in Nature Water.
Researchers compared OpenET’s ensemble estimates data from six independent Landsat-based models to ground-based measurements from more than 150 eddy-covariance flux towers. The main finding supports OpenET accuracy, demonstrating remotely sensed ET data can reflect actual crop water use under field conditions.
Volk notes that while irrigation scheduling has often relied on weather station data combined with crop coefficients (the “ETc” approach) adjusted for crop type and growth stage, the data are generalized, while satellite-based ET shows actual water use of each field by integrating the combined effects of crop growth, management, soil and local weather, allowing irrigation managers to distinguish which fields are using more or less water at any given time.
While return on investment is ultimately crop- and farm-specific, ET data provide a foundation for evaluating where water is going and how efficiently it’s used, says Volk. The open nature of OpenET “helps build trust in the data by allowing anyone to see how the estimates are produced,” he says. “For end users, it means they can understand the basis for decisions and compare results across different systems using a transparent, shared information source.
“What really makes OpenET unique other than being open-source is its ease of access or accessibility — users can click on a field and get ET data.”
“What really makes OpenET unique other than being open-source is its ease of access or accessibility — users can click on a field and get ET data.”
— John Volk, PhD, associate research scientist, hydrology, Desert Research Institute
Elwyn Pratt, data management at Sun Pacific Farming, notes that his team builds in-house user interfaces to help its farmers manage more than 45,000 acres of crops in California, including citrus, table grapes, kiwi, tree fruit and nuts. “The scale of our agricultural program requires us to apply data-driven decisions to our irrigation, which is complicated by many factors, including fruit species, the geography of each farm and the weather,” Pratt says. “OpenET has enabled us to fine-tune our irrigation schedule — volume and duration — by estimating daily evapotranspiration at each plot of land and applying precisely enough water to replenish the water being consumed by the plants or evaporated into the air.”
“OpenET has enabled us to fine-tune our irrigation schedule — volume and duration — by estimating daily evapotranspiration at each plot of land and applying precisely enough water to replenish the water being consumed by the plants or evaporated into the air.”
— Elwyn Pratt, data management, Sun Pacific Farming
Pratt notes that this conserves more water and enables a higher crop yield of finer quality while reducing company costs.
Ultimately, “when growers can access reliable, transparent data about crop water use, they are empowered to make more informed irrigation decisions — saving water, improving yields and strengthening their operations,” notes Sara Larsen, CEO of OpenET Inc. “This not only benefits individual farms but also helps communities and water managers sustain their limited water resources.”
8280 Willow Oaks Corporate Drive | Suite 630 | Fairfax, VA 22031
Tel: 703.536.7080 | Fax: 703.536.7019
HOME | ABOUT US | ADVERTISE | SUBSCRIBE | CONTACT | PRIVACY POLICY | IA ANTITRUST STATEMENT
