OpenET Launch Gives EROS Scientists—and Water Managers—Reason to Celebrate

By Earth Resources Observation and Science (EROS) Center October 21, 2021

The Landsat 9 satellite’s launch September 27 will ensure the continuity of thermal imaging for the work of the evapotranspiration (ET) team at the USGS Earth Resources Observation and Science (EROS) Center.

But that’s not the only launch the team is celebrating this fall. The online platform OpenET also opens to the public October 21. OpenET will provide free, easily accessible field-scale data about water consumption by crops and other vegetation in the western United States.

EROS ET colleagues Gabriel Senay, Mac Friedrichs, and Gabe Parrish have been part of the public-private collaboration that uses satellite and weather station data combined with modern computing capabilities to create OpenET. OpenET is a tool to help with decisions about water supplies that face increasing pressure from changes in weather and climate, drought impacts on agriculture, and growing population needs.

The launch comes after years of effort from partnering Federal agencies, universities, environmental groups, water managers, and farmers who use water to irrigate crops.

Evapotranspiration is the movement of water into the atmosphere via evaporation from surfaces and transpiration from plants. Senay’s Operational Simplified Surface Energy Balance (SSEBop) model, originally developed at EROS, estimates field-scale ET based mainly on Landsat images, where irrigated fields with more ET have a cooler surface temperature. Friedrichs and Parrish helped implement the ET model at regional scales using Google Earth Engine, a planetary-scale platform for Earth science data and analysis.

Six Models Add Up to One Ensemble

Senay and Friedrichs became involved before the project even had a name, offering their insights about ET modeling with Google Earth Engine as a demo of sorts. SSEBop is included in the ensemble of six models that OpenET ultimately developed to estimate the best available ET information for various land cover types and regions across the Western U.S.

The development process was not simple or brief, but it was beneficial, said Senay and Friedrichs. They and Parrish participated in a team of more than 30 people with expertise in remote sensing science and technology, spearheaded by NASA, the Desert Research Institute, the Environmental Defense Fund, and Google Earth Engine. Funding has come from NASA, philanthropic organizations, and water agencies.

The collaboration actually improved all of the ET models, Senay said—including SSEBop.

“We had the chance to critically evaluate the model performance in ways that were not possible five years ago, and we found ways to improve it as a result. … We made improvements to our model algorithm, based on constructive comments and feedback from the collaborative effort and user community. So, it’s satisfying to get here,” Senay said.

“The idea that an agreed-upon ET ensemble is providing the best information … after all this rigorous testing and development is a relief of sorts, but it’s also a success story for each of the contributing modeling teams,” Friedrichs said. “It also reinforces the idea that remote sensing science as a whole, especially in the water use and ag sector, is providing this really tremendous and relatively untapped value from a big data standpoint.”

A model intercomparison and accuracy assessment by the OpenET team will give farmers and water managers more confidence in the information they receive about their fields and less confusion about the various models, Senay said, highlighting the value of an OpenET ensemble product.

Irrigators like the E&J Gallo Winery in California already use ET data to help manage irrigation schedules and save water and money. OpenET data will be used as the basis of an online water accounting and trading platform by farmers concerned about groundwater in a water storage district in California’s Central Valley, and data will also help Colorado ranchers explore irrigation strategies to conserve water.

Landsat and Cloud Computing Are Essential

Landsat’s scale and its deep archive—nearly 50 years of data—are key to the project.

“Without Landsat, we wouldn’t have been able to even start OpenET,” Senay said. “Landsat captures meaningful field-scale resolution so that farmers can benefit from information about their fields. And also, when you want to study water use, it’s not only how much water is being used today, but you want to know the history—how that field has been using water over the years.”

Cloud computing is also an essential component. Recent advances in that technology, like Google Earth Engine, enable the operational processing of the quantities of data needed to produce information for 17 States over time. Those States are Arizona, California, Colorado, Idaho, Kansas, Montana, Nebraska, Nevada, New Mexico, North Dakota, Oklahoma, Oregon, South Dakota, Texas, Utah, Washington, and Wyoming.

The need is great for high-quality water data in a changing climate. Senay thinks the OpenET platform, with its solid science and technology framework, could expand to the rest of the United States and even the world.

After spending years meeting with other OpenET team members to turn the complex project into a reality, Senay is “pleasantly surprised” it all came together as it has.

“It’s such an ambitious project and an incredible job done by the leadership group and technical team to reach this accomplishment,” Friedrichs said, summarizing it as a “multimillion-dollar project with the goal of making timely crop water use information data freely available and accessible on a web application platform for the western States using the Landsat archive, all while involving numerous users and stakeholders from the start.”

And now with the launch, they hope people use the platform to help address water scarcity.

“It’s tough to quantify just how important this information is,” Friedrichs said.