News Release

Study helps satellites measure Great Lakes' water quality

Peer-Reviewed Publication

Ohio State University



The image above was created by a computer model that greatly underestimated the amount of algae in Lake Erie, based on satellite images. Engineers at Ohio State University recently determined which computer models did the best job. Here, algae blooms appear red. Image courtesy of Ohio State University.

Full size image available through contact

COLUMBUS, Ohio -- Ohio State University engineers are helping satellites form a clearer picture of water quality in the Great Lakes.

The study -- the first ever to rate the effectiveness of various computer models for monitoring the Great Lakes -- might also aid studies of global climate change.

As algae flourishes in the five freshwater lakes every summer, satellite images show the water changing color from blue to green, explained Carolyn Merry, professor of civil and environmental engineering and geodetic science.

When algae levels are too high, water takes on a foul taste and odor that isnt easily removed by traditional treatment methods. Some forms of algae, such as one called microcystis, are toxic when consumed in large quantities. Though it can be filtered out of drinking water, microcystis can kill fish and birds, and coastal communities often have to ban swimming and water skiing in the summer when the algae blooms.



This image of the eastern United States was taken with the Sea-viewing Wide Field-of-view Sensor (SeaWiFS) on board NASAs SeaStar spacecraft in spring of 2000. Using SeaWiFS images, Ohio State University engineers determined which computer algorithms did the best job of processing satellite data to measure the amount of algae in lake Erie. Image courtesy of NASA.

Full size image available through contact

Computer models enable scientists to measure the color of light reflected from the water to gauge how much algae is present in a lake, and where. The problem: all the available models of this type were originally designed for sea water, not lake water.

Theyve got it down pat for the ocean, Merry said of the various models developed by NASA and other agencies over the years. But lakes are shallower and have different water conditions that affect the wavelengths of light collected by the satellites, so we can get erroneous measurements.

Merry and masters degree student Raghavendra Mupparthy reported the results of an initial study of Lake Erie May 25 in Denver at the meeting of the American Society for Photogrammetry & Remote Sensing. They determined which four of the top ocean models may perform well for studies of the Great Lakes.



This image above was created by a computer model that accurately estimated the amount of algae in Lake Erie, based on satellite images. Engineers at Ohio State University recently determined which computer models did the best job. Here, algae blooms appear red. Image courtesy of Ohio State University.

Full size image available through contact

Most scientists look for evidence of climate change in the oceans rather than in lakes, but that may change in the future, the engineers said.

Lakes are important because they respond much faster to climate change than oceans do, Mupparthy said.

Erie is the warmest of the Great Lakes, the shallowest and richest in life, and has changed much in recent years. For instance, the lake has experienced a microcystis outbreak every summer since 1995. The warmer the temperature, the bigger the outbreak, Merry said.

Though lakeside towns routinely measure algae levels firsthand, satellite images give scientists a broader view of trends in the lakes.

If we want to really understand whats happening in Lake Erie, we can take these computer models and marry them to other data to get a more complete picture, Merry said.

The models could also apply to other large lakes and seas, such as the Mediterranean.

The key to proving the models valid, she said, is to compare them to actual measurements of algae in the water. Thats what she and Mupparthy did for Lake Erie, by collecting water samples from four sites in October 2000.

They compared the actual algae levels to the levels suggested by 17 different models, each using images of the lake taken by NASAs SeaStar spacecraft.

Model performance varied widely; one demonstrated an error of more than 250 percent, while several were off by less than one percent.

What set the top-performing models apart was how well they enabled the engineers to remove the interference of atmospheric molecules. Over the ocean, carbon dioxide, ozone, and humidity are fairly constant, but amounts of these molecules fluctuate dramatically over lakes depending on the weather and human activity.

The National Oceanic and Atmospheric Administrations Cooperative Institute for Limnology and Ecosystems Research funded this study. Merry and Mupparthy have applied for funding to collect water samples from more sites, with the hope of customizing a model for Lake Erie.

###

Contact: Carolyn Merry, 614-292-6889; Merry.1@osu.edu
Raghavendra Mupparthy, 614-688-4753; Mupparthy.1@osu.edu

Written by Pam Frost Gorder, 614-292-9475; Gorder.1@osu.edu


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.