Keeping Water on the Radar: Machine Learning to Facilitate Essential Water Cycle Measurement | Computing
Department of Computer Science Assistant Professor Chris Heckman and CIRES Research Hydrologist Toby Minear have been awarded a Grand Challenge Research & Innovation Seed Grant to create an instrument that could revolutionize our understanding of the amount of water in our rivers, lakes, wetlands and coastal areas dramatically increasing where we measure it.
The new, low-cost instrument would use radar and machine learning to quickly and safely measure water levels in a variety of scenarios.
This work could prove vital as the USDA recently proclaimed the entire state of Colorado a “primary natural disaster zone” due to an ongoing drought that has made the American West potentially the driest for more than a millennium. Other climate records around the world also continue to be broken, year after year. Our understanding of the evolution of the water cycle has never been more essential at local, national and global levels.
A fundamental element to develop this understanding is to know the changes in the surface height of water bodies. Currently, measuring changing surface water levels involves expensive sensors that are easily damaged by flooding, difficult to install, and time-consuming to maintain.
“One of the big issues is that we have limited places where we take measurements of surface water heights,” Minear said.
A new method
Heckman and Minear aim to change that by building an inexpensive instrument that doesn’t need to be in a body of water to read its average water surface level. Instead, it can be placed several feet away – safely elevated from flooding.
The instrument, roughly the size of two credit cards stacked on top of each other, relies on high frequency radio waves, often referred to as “millimeter waves”, which have only been made commercially available to the course of the last decade.
Thanks to radar, these short waves can be used to measure the distance between the sensor and the surface of a body of water with high specificity. As the water surface level rises or falls over time, the distance between the sensor and the water surface level changes.
The instrument’s small form factor and potential off-the-shelf ease of use set it apart from previous efforts to identify water by radar.
It also streamlines data transmitted over often limited and expensive cellular and satellite networks, reducing costs.
Additionally, the instrument will use machine learning to determine whether a change in the measurements might be a temporary outlier, such as a bird swimming, and whether or not a surface is liquid water.
Machine learning is a form of data analysis that seeks to identify patterns in data to make decisions with little human intervention.
While radar has traditionally been used to detect solid objects, liquids require different considerations to avoid being misidentified. Heckman believes traditional radar processing methods may not be sufficient to measure liquid surfaces at such close proximity.
“We’re looking at going further up the radar processing chain and reconsidering how some of these algorithms were developed in light of new techniques in this type of signal processing,” Heckman said.
In addition to possible fundamental changes in radar processing, the project could empower communities of citizen scientists, according to Minear.
“Right now, many systems we use need an expert installer. Our idea is to internalize some of these expert decisions, which reduces a lot of the cost and makes this instrument more user-friendly for a citizen science approach,” he said. .
By lowering the barrier of entry to surface water level measurement through low-cost devices with smaller data requirements, researchers are expanding opportunities for communities, even in areas where networks cellular are limited, to measure their own water sources.
The team is also committed to open source principles to ensure everyone can use and build on the technology, allowing new innovations to happen faster and more democratically.
Minear, who is a member of the science team and the Cal/Val team for NASA’s upcoming Surface Water and Ocean Topography (SWOT) mission, also hopes the new instrument could help verify the accuracy of level measurements. of the surface of the water carried out by the satellites.
These sensors could also give local, regional and national communities better insight into their water use and supply over time and could be used to help make evidence-based policy decisions regarding rights and use of water.
“I’m very excited about the opportunities that arise by getting data in places where we currently don’t get it. I anticipate this could give us a better insight into what’s happening with our water sources, even in our backyard,” Heckman says.