You are here:
OHIO RIVER WATER QUALITY ASSESSMENT USING LANDSAT-7 DATA
FROHN, R. C. AND B. C. AUTREY. OHIO RIVER WATER QUALITY ASSESSMENT USING LANDSAT-7 DATA. Presented at SWIMS Conference, Chicago, IL, January 30 - February 01, 2007.
The goal of this research is to develop methods and indicators that are useful for evaluating the condition of aquatic communities, for assessing the restoration of aquatic communities in response to mitigation and best management practices, and for determining the exposure of aquatic communities to different classes of stressors (i.e., pesticides, sedimentation, habitat alteration).
The objectives of this project were (1) to develop a universal index for measuring Turbidity and Chlorophyll-A from remote sensing data and (2) to correlate satellite image parameters from Landsat-7 data with field measurements of water quality for five parameters: Chlorophyll-A (Chl-A); Turbidity; Total Suspended Solids; Dissolved Oxygen; and Secchi Depth. The study area was a 95 km stretch of the Ohio River near Cincinnati, Ohio. Within this stretch, 23 samples of water quality were collected within 24 hours of a the capture of the Landsat-7 image. Two new image indices were successfully developed for estimating Chl-A and Turbidity. The Chl-A index had a 0.85 correlation and the Turbidity a 0.89 correlation with the water samples. Thirty-four other Landsat image parameters were also tested including individual bands, band ratios, principal components, and minimum noise fraction transformations but none had higher correlations than the Chl-A and Turbidity indices. Linear regression models were developed to quantify Turbidity (NTU) and Chl-A (ug/l) from the Turbidity and Chl-A indices respectively. The regression models had R2 values of 0.86 for Turbidity and 0.81 for Chl-A indicating good fits. A multiple regression model was developed for Turbidity using the Turbidity Index and five other image parameters and yielded an R2 of 0.90. Multiple regression models were also developed for Total Suspended Solids (R2 = 0.89) and Dissolved Oxygen (R2 = 0.72) using the Turbidity and Chl-A indices with other Landsat image parameters. The high correlation and simplicity of the new Turbidity and Chl-A indices indicate that they may be applicable to other rivers on a regional or national scale for rapid and cost-effective water quality monitoring.