EROS Discussion Focuses on Irrigation as Part of the National Water Census

Sources/Usage: Public Domain. View Media Details

Data are compiled and estimated for the U.S. every five years as part of the National Water Census, which offers a treasure-trove report on water-use information in America—everything from irrigation to domestic, power production to aquaculture. Where the water is. How it’s being used. Even how that use is changing over time.

The question before a small group of hydrologists meeting at the Earth Resources Observation and Science (EROS) Center near Sioux Falls, SD, in July then was this: How can methods using Landsat and other remote-sensing tools estimate consumptive water use for irrigation in the country at higher temporal and spatial scales, and more frequently? And how can that help the USGS improve on water-use estimations for the National Water Census?

A lot of people are interested in those answers, said Molly Maupin, the National Water User Coordinator for the USGS’ NWUSP, and one of the participants at the EROS meeting. Academia are paying attention to that discussion, Maupin said. So are international users, and government officials at the Federal, state and community levels. Even local farmers and water managers.

Hydrologists from the Water Mission Area and USGS Water Science Centers got together at EROS for a planning meeting to talk specifically about methods to improve the computation of consumptive water use, particularly through irrigation, and what the available tools are for measuring that in even better ways going forward.

“Thanks to the Landsat missions and EROS, these very valuable remotely sensed satellite images are available,” Maupin said. “They have a lot of value in that they are regularly acquired over the same geographic places on Earth, so you can get this stream of data that helps us to show what’s changing on the landscape.

Those landscape changes are often the driving force of changes in water use. So, the value of looking at Landsat data over time, Maupin said, “is to help us get a time series of the picture of what’s happening on the landscape, and then to determine how that impacts the water resources based on how man is using water.

Maupin, along with Julie Kiang, who is chief of the Analysis and Prediction Branch of the Water Mission Area, and hydrologists from the Water Science Centers were briefed by EROS Scientist Jess Brown on new methods for mapping irrigation areas across the country. They also heard about how EROS Scientist Gabriel Senay has improved his Operational Simplified Surface Energy Balance (SSEBop) model, which is used to estimate water consumption by plants in irrigated fields. Water consumption by crops is a major component to estimating irrigation water use.

The fact is, Senay’s model actually estimates consumptive water use across the entire landscape—from irrigated agriculture to forests to urban lands. That makes the irrigation maps from Brown and others important in order to mask out everything else except the irrigated lands.

Crop irrigation has the largest percentages of total consumptive water use among all types of water uses, Maupin said. Water that is consumptively used is lost to evapotranspiration or locked up in crop plant material, and is not immediately available to be used again.  Other water uses, such as for power plant cooling purposes in thermoelectric plants, actually require more water, though usually much of that water ends up being returned to the hydrologic system and thus is not consumptively used. The bottom line is, irrigation may use less total water compared to thermoelectric power generation, but it has a much higher percentage of consumptive use.

“That’s why we have to look at the irrigated lands, especially in the West,” Maupin said.

And it’s not just the water being sprinkled on farm fields that these hydrologists are interested in. The group wants to know more about how to measure the amount of water being lost as its withdrawn from its source and conveyed along canals or other channels to the fields.

While satellite data can quantify consumptive plant use, it doesn’t reveal much about how water is actually delivered to irrigated fields. Or how much is lost on the way. That’s important information for the National Water Census to have, Senay said.

With a canal system, “maybe the water takes 10 miles, or 20 miles before it reaches the point of use,” especially in the West, Senay said. “Some of the loss depends on whether the canal is lined or not. So, finding out how much of the canals are lined … this is the kind of information we need to estimate the total withdrawals from a surface-water source.”

It’s all pertinent in a world where water is becoming more and more precious with each passing year, Senay said. Population growth and the related user demand that comes with it continues to grow even though the amount of the resource does not. The growing population needs to be fed. That means groundwater supplies are being depleted by excessive extraction of water for irrigation in some countries of the world. In other places, land degradation and changing climates are leading to the kind of rainfall variability that is spawning drought and famine more often.

Those realities make keeping tabs on the sources of water, its use, and the trends in its availability and consumption all the more important. With satellite data, it’s possible to monitor rainfall variability, Senay said. It’s possible with his SSEBop model to quantify how much water is being evaporated from the landscape and transpired by plants on that landscape—what’s known as evapotranspiration, or ET.

Sources/Usage: Public Domain. View Media Details

“ET is strongly, linearly correlated to food production,” Senay said. “The more plants transpire, the more likely that more grain is being produced. So, by understanding how much ET and rainfall variability is occurring over the landscape, value-added information can be generated for drought monitoring and food security alerts.”

Senay’s work on ET makes him a leader in his field, Maupin said. “We wouldn’t be where we are today; I can say that pretty clearly,” she said. “I mean, I don’t know how we would do what we’re doing today without satellite imagery, and good methods and tools to use them.”

At the same time, she sees promise in what Cloud computing and platforms like Google Earth Engine bring to the National Water Census when it comes to such things as computing consumptive use based on Landsat data. The ability for such platforms to host massive amounts of data, and to provide considerable computing power that can be made available to everybody “is really advantageous,” Maupin said.

“So, I think the challenges are the ability to convey the (Cloud-computing) method, to keep pushing those frontiers, and to get the information we are gleaning from these methods … out to the public,” she said. “The push right now, I think, is for faster and more.”

For more information:

About the SSEBop model: 2018

https://doi.org/10.13031/aea.12614