Lake Observations from Citizen Scientists and Satellites
Tamlin Pavelsky, University of North Carolina, Chapel Hill
Lakes are the primary source of irrigation and drinking water in many parts of the world. They are used to provide food and generate electricity. They can also lessen the impact of flooding due to natural disasters, and they provide habitats for fish, plants, and other wildlife.
Lake Observations by Citizen Scientists and Satellites (LOCSS) is an effort to better understand how the water volume in lakes is changing. LOCSS is working with a network of citizen scientists who report lake height by reading simple lake gauges. LOCSS also uses satellite images to measure the surface area of the lake. By using the measured lake height and lake surface area, researchers can understand and calculate the volume of water in a given lake, understand how the lake volume is changing over time, and predict how those changes in lake volume may impact humans and wildlife. However, this data will need to be validated with ground measurements. LOCSS creates a framework to provide such measurements.
There are tens of millions of lakes in the world, but only a fraction are monitored. This project uses the power of citizen science to gather large amounts of lake volume data that a single research team would be unable to obtain. These lake volume measurements will help NASA prepare for the upcoming Surface Water and Ocean Topography (SWOT) satellite mission. SWOT will be able to measure lake height and surface area simultaneously allowing for global measurements of lake water storage. However, this data will need to be validated with ground measurements.
During the prototype phase, the project focused on lakes in North Carolina. In the implementation phase, LOCSS will expand the project to new lakes including those in Washington, Wisconsin, Bangladesh. For information on how you can volunteer visit the LOCSS site.
One of the first installations of the implementation phase saw gauges installed in Northeastern Bangladesh at water bodies called "haors." The local partnering institutions for this project are the Bangladesh Water Development Board and Hydrometerology Research and Consulting.
From the months of April through October, thousands of km2 of low-elevation land is inundated with transboundary runoff generated by the monsoon rains from mountains in India in the Assam province. These water bodies become a productive fisheries ecosystem for six months. As the waters recede in the post-monsoon season, the soil becomes rich in nutrients and organic matter. The land is then primed for rice cultivation from groundwater. Rice cultivation during this season, known as the Boro season, is critical to food security of Bangladesh.
Although the extent and duration of inundation are well known, the volume of water storage and changes through the system of haors are not well known. Through the citizen science project, the team is quantifying the seasonal surface water storage component of the water cycle. This understanding may one day inform policy decisions for managing post monsoon water availability, pre-monsoon flash floods and rice cultivation.
Update October 2019
LOCSS has continued the expansion of its network around the world. The project has installed new water level gauges in 24 lakes in Washington State, seventeen lakes in Illinois, and four lakes in Bangladesh (in collaboration with the Bangladesh Water Development Board).
Google Earth Engine software was used in the prototype phase to automatically calculate lake surface areas and tested in Washington and Wisconsin lakes. Limitations related to topography and vegetation were identified, however, so a new algorithm was developed based on USGS Landsat Digital Surface Water Extent (DSWE) product. A paper on using a combination of DSWE and ESA Sentinel 1 and 2 data to identify lake area variations in Bangladesh is currently in preparation.
A second algorithm was developed through LOCSS to estimate inundation extent over Bangladesh lakes, which are rarely observed by optical imagery during the monsoon season due to clouds. This algorithm combines measurements from Landsat 8, Sentinel 2, and Sentinel 1 of inundation extent to produce monthly estimates of inundation for three Bangladesh lakes. It was validated against high-resolution measurements of inundation extent from Planet Cubesat data and shows a strong accuracy. A paper on this algorithm, currently in review, will be published in IEEE Transactions in Geoscience and Remote Sensing.
LOCSS has also developed a prototype app for the project. The app is in revision currently to improve the user interface and provide more functionality. The expected release date is prior to AGU’s Fall Meeting 2019.
This CSESP project is a collaboration between the Department of Geological Sciences at UNC-Chapel Hill, the UNC-Chapel Hill Institute for the Environment, the University of Washington Department of Civil and Environmental Engineering, and the Tennessee Technological University Department of Computer Science. Expert advice has been provided by the NC Museum of Natural Sciences and the Wisconsin Citizen Lake Monitoring Network.
Ahmad, S., Hossain, F., Eldardiry, H., & Pavelsky, T.M. (in review). A Fusion Approach for Water Area Classification using Visible, Near Infrared and Synthetic Aperture Radar for South Asian Conditions. IEEE Transactions on Geoscience and Remote Sensing.
Pavelsky, T.M., Ghafoor, S., Hossain, F., Parkins, G., Rogers, M., Topp, S., & Yelton, S. (2018). Developing a Citizen Science Network to Validate Lake Water Storage Measurements from the SWOT Satellite Mission. Presented at the American Geophysical Union (AGU) Fall Meeting.
Last Updated: Nov 8, 2019 at 1:20 PM EST