Airborne Sensor Shows Promise for Monitoring Water Quality

To demonstrate the potential of using remote sensing to monitor water quality across broad areas, U.S Geological Survey and NASA scientists teamed up for the simultaneous collection of water quality measurements from the air and in the water. While USGS scientists collected water quality measurements from a boat in the northeastern part of San Francisco Bay (Grizzly Bay and Suisun Marsh), scientists from NASA’s Jet Propulsion Laboratory simultaneously flew a small airplane above them, carrying an experimental sensor called PRISM (the Portable Remote Imaging SpectroMeter). The study successfully demonstrated the potential of remote sensing to facilitate the detection and sources of pollution and to assess the complex impacts of wetland restoration and climate change on water quality and ecosystem productivity.

High-resolution imaging spectroscopy can help monitor and understand the effects of wetland restoration on the distribution of methylmercury and potentially of some important contaminants in the Bay-Delta. 

The San Francisco Bay-Delta Estuary and its watershed are a major source for freshwater for California, and the ecosystems have been profoundly altered by humans over the past 150 years. Water quality monitoring is critical to the wise management of this important water resource and assessing the health of this ecosystem.

Evaluating the value and accuracy of PRISM for this purpose was only made possible by correlating the remotely-sensed spectroscopy data collected during the overflight with the water quality data collected by the USGS during a high-speed transect across Grizzly Bay, and a secondary correlation of the transect data with laboratory measurements. This experiment is the first time this technology has been successfully used in this manner.

“In estuaries, where the tides cause water quality to change over timescales of minutes and spatial scales of feet, we need to come up with better techniques for calibrating airborne and satellite sensor data to in-the-water conditions. This approach of using a high-speed boat to map conditions in the water across a broad area as the sensor is flown overhead worked well as a way to connect the laboratory measurements directly to airborne sensor data. We are now using it to calibrate satellite observations so we can expand the view to broader areas,” said Brian Bergamaschi, USGS biogeochemist and coauthor of the study.

USGS scientists tested the surface water for the presence of methylmercury (a potent neurotoxin), dissolved organic carbon (because methylmercury binds strongly with and is correlated to dissolved organic matter), chlorophyll (an indicator of phytoplankton in the water), turbidity (water cloudiness), and suspended sediment in the water column (to understand sediment transport in estuaries). These key water quality indicators were measured on a real-time, flow-through system on the boat, and in collected samples sent to USGS laboratories, at the same time the PRISM sensor was collecting data overhead, looking for the same information.

“While turbidity has been mapped remotely for years with satellites, this time we got the individual components of turbidity: suspended sediments, DOC and chlorophyll. Mapping those individually in this cocktail of bay water is the real strength of the PRISM sensor and the supporting field sample campaign. And the fact that organic carbon is both a contaminant in itself for drinking water and a tracer for methylmercury was another plus,” said Lisamarie Windham-Meyers, USGS ecologist and coauthor of the study.

The portable PRISM sensor is a remote-imaging spectrometer, that measures the amount and wavelength of visible light and near-infrared radiation reflected back from the Earth’s surface, both water and land. The correlation between what PRISM recorded, and the laboratory analysis of the water samples collected by the scientists in the boat, matched with 80 to 95 percent accuracy. The successful and accurate detection of these water quality indicators by an airborne sensor across an environmentally relevant, narrow window of concentrations is very important because some of the quality indicators are particularly difficult to measure due to sampling precision and/or technical costs.

“One of the most exciting things about this study was that it demonstrated our ability to take a mile-high view of methylmercury concentrations across a complex mosaic of wetlands and open water. This represents the first imaging for this toxic substance done at this resolution and spatial scale,” said Mark Marvin-DiPasquale, USGS microbiologist and coauthor of the study.

USGS has been consistently monitoring water quality in San Francisco Bay for almost 50 years. Vessel-based water quality monitoring programs are time-consuming and labor-intensive work. PRISM has the potential to greatly expand the spatial coverage of the traditional vessel-based and fixed monitoring station-based approaches used for water quality monitoring.

“This remote sensing technology holds great promise for efficiently collecting water quality data over a large area, at high resolution and with great accuracy,” said JPL scientist Cédric G. Fichot, who led the NASA-funded part of the study.

The technical report, “High-Resolution Remote Sensing of Water Quality in San Francisco Bay-Delta Estuary,” is published in the current issue of the journal, Environmental Science and Technology.

USGS scientists (from left) Judy Drexler, Brian Bergamaschi, Tamara Kraus, Bryan Downing and Katy O’Donnell in route to a sampling station aboard the USGS R/V Mary Landsteiner. Photo credit: Stephen de Ropp

USGS scientists (from left) Judy Drexler, Tamara Kraus and Bryan Downing and Katy O’Donnell preparing to take spot field measurements in the San Francisco estuary. Photo credit: Stephen de Ropp


The USGS R/V Mary Landsteiner shown at a brief stop during a study in the Northern San Francisco estuary. USGS scientist Bryan Downing is shown calibrating the on-board real-time underway measurement system. Photo credit: Stephen de Ropp