Title: Hyperspectral imaging of chlorophyll in the Caballo Reservoir in New Mexico

Authors: Alfonso Blanco, Foudan Salem, Richard Gomez

Addresses: George Mason University, 4400 University Drive, MSN 3D1 Fairfax, Virginia 22030-4444, USA. ' George Mason University, 4400 University Drive, MSN 3D1 Fairfax, Virginia 22030-4444, USA. ' George Mason University, 4400 University Drive, MSN 3D1 Fairfax, Virginia 22030-4444, USA

Abstract: Chlorophyll is a good water quality indicator for determining other pollutants in inland water bodies. Chlorophyll concentrations can be measured by remote sensing techniques; however, studies show that the broad wavelength spectral data available on current satellites do not permit discrimination between chlorophyll and suspended sediments. Hyperspectral imaging by satellite and aircraft sensors provide significant relationships between chlorophyll and radiance/reflectance measurements. This paper describes the identification of chlorophyll concentrations using hyperspectral imagery in inland water bodies. This hyperspectral study site is located in the Caballo Reservoir along the Rio Grande in the State of New Mexico. Run off from agricultural fields, failing septic tanks, and other sources of pollutants enter this water body. NASA|s Jet Propulsion Laboratory flew the flight line on June 30, 1999 and acquired the images in the study with an Advanced Visible infrared Imaging Spectrometer (AVIRIS). No groundtruth measurements were available at the time of the AVIRIS flight mission to compare to the images of the flight line. The images were atmospherically corrected for distortions using the ACORN software. Hyperspectral image analysis techniques using ENVI software were used for identifying potential chlorophyll areas. The spectral matching signature method was used to identify chlorophyll near the shores of the Caballo Reservoir. This method allows mapping of the chlorophyll constituent using the Spectral Angle Mapper (SAM) classification technique. The clean-up tasks afterwards are easier for eliminating any point or non-point pollutant sources because it reduces equipment and field sampling costs by obtaining water quality pollutant concentrations over a large geographical area.

Keywords: hyperspectral imaging; advanced visible infrared imaging spectrometer; AVIRIS; water quality; ACORN; ENVI software; remote sensing; Rio Grande; Caballo Reservoir; USA; United States; water pollution; chlorophyll; runoff; image analysis; pollutant concentrations.

DOI: 10.1504/IER.2004.053928

Interdisciplinary Environmental Review, 2004 Vol.6 No.2, pp.85 - 100

Published online: 13 May 2013 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article