Eric Lindsey, an assistant professor in the Department of Earth and Planetary Sciences at the University of New Mexico, is concerned about the sustainability of aquifers around the world. Aquifers, natural stores of groundwater, are vital to human life on this planet. Worldwide, aquifers are key sources of drinking water and crop irrigation water. Unfortunately, many aquifers are in trouble due to excessive removal of their water.
Excessive water removal, also called excessive extraction, can result in severe consequences such as loss of available drinking water and irrigation. Excessive extraction can also lead to extreme consequences in coastal areas when aquifer soil compacts to the point that the soil surface falls below sea level. This can be seen in places like Jakarta, Indonesia, where some neighborhoods are now a few feet below sea level, leading to floods at high tide.
Aquifers fill with groundwater when rain or snow falls in large enough quantities to penetrate the ground and accumulate in small, empty spaces below the soil surface. When too much water is removed from the aquifer, the porous soil dries out and begins to compact. This phenomenon, known as “soil subsidence”, causes a drop in the altitude of the soil surface. Eventually the soil can compact to a point of no return. This point of no return, called “permanent subsidence”, has serious consequences. Once permanent soil subsidence occurs, the capacity of the aquifer to store water decreases forever.
Among other research interests, Lindsey specializes in monitoring soil subsidence rates. He recently joined the faculty at UNM after spending several years in Southeast Asia studying the movement of the Earth’s surface to monitor the dangers of the underground that could lead to disasters such as the floods in Jakarta. To continue its soil subsidence studies, Lindsey is partnering with the UNM’s Center for Advanced Research Computing (CARC) to host and process data to help researchers determine the true state of aquifers around the world with a focus on the state of New Mexico.
To obtain accurate measurements of soil subsidence, Lindsey collects data collected by Sentinel-1 satellites operated by the European Space Agency (ESA). Since the launch of the first Sentinel-1 satellite in 2014, they have been orbiting the Earth using radar to measure subtle changes in surface altitude using a technique known as Synthetic Aperture Radar or InSAR. Each Sentinel-1 satellite takes 90 minutes to complete one orbit. However, due to the Earth’s rotation, it can take up to 12 days for a satellite to pass over the same location twice. This results in twice a month observing the Earth’s surface for every point on the planet. ESA collects, stores and allows free global access to raw data transmitted back to Earth.
To make his landslide maps, Lindsey first compiles all the satellite images collected in an area. This means that it routinely downloads more than a hundred high-resolution images covering approximately 24,000 square miles each. Each image is about five gigabytes in size, which when fully processed in datasets can become tens of terabytes of data. This huge amount of data can be problematic and requires huge storage capacity.
To address these storage issues, Lindsey recently received a grant from the National Science Foundation (NSF) to establish new storage nodes at CARC. This increased storage capacity combined with CARC’s high-speed parallel computing capabilities will allow it to process the hundreds of terabytes of data needed to produce a current land settlement map for the entire state of New Mexico. And Lindsey doesn’t want to stop there. Future goals include mapping not only the current landslide in New Mexico, but also creating maps that go back to the beginning of the first available InSAR data in 1992. With enough time and resources, he hopes to eventually create landslide maps for the entire Western United States.
Processing this vast amount of data is complicated. Lindsey is currently working on developing methods for efficient data processing. To create his maps, all downloaded satellite images must be overlapped and carefully aligned with each other. This process requires the alignment of images based on satellite geometry and orbital position, as well as the removal of harmful atmospheric effects, which can distort images. Once this process is complete, Lindsey creates pairs of matched images and calculates measurements for all height changes at a particular location over a period of time. These measurements are very accurate and can detect changes of a few centimeters for a single compiled image. If they average together over time, they can be accurate to millimeter levels per year.
After mapping, the next step is to compare these measurements with water measurements in wells from the United States Geological Survey (USGS) to infer aquifer properties. In a sustainably managed aquifer, the aquifer must be allowed a recovery period to ensure that the soil remains “elastic”. This means that after water withdrawal during the growing season, the aquifer must be left to rest to refill with water. This results in what Lindsey calls a “subtle pattern of seasonal subsidence and uplift” and points to sustainable aquifer circulation. Unfortunately, what Lindsey and others most often notice is that the soil loses elasticity and continues to descend over the long term. This is worrying for the health of the aquifers of our planet.
Lindsey’s accurate measurements combined with USGS data will allow him to map areas that damage their aquifers due to excessive groundwater abstraction. Albuquerque aquifers are currently stable or recovering since the city reduced groundwater use in 2008, although that recovery is again threatened by drought. However, preliminary data from InSAR’s observations of cities surrounding Albuquerque show signs of constant soil subsidence. Lindsey hopes to use her data for information and education before too much damage occurs and more aquifers settle permanently.
Land subsidence is not Lindsey’s only focus on UNM. It applies the same processes and data types to study soil motion caused by earthquakes, volcanoes, and tectonic plate movements. Lindsey and his associates at UNM, including Lindsay Worthington, Brandon Schmandt and Mousumi Roy, plan to lead a workshop this summer to introduce undergraduate students to fast computing and geophysical research topics such as subsidence monitoring and seismic surveying. . data. They hope participants will learn about high-performance computing, geophysical data analysis, and programming in a fun, inclusive, collaborative environment. For more information, visit the workshop website: ESCAPE: Earth Sciences Computing & Programming Experience.
To learn more about Lindsey and his projects, visit his faculty profile and his personal profile. Lindsey also has code available on Github.