Estimation of chlorophyll-a concentrations in diverse water bodies using ratio-based NIRRed indices Remote Sensing Applications: Society and Environment

Numerous studies indicate that biological and chemical water quality parameters such as chlorophyll-a have distinctive spectral characteristics and can thus be measured using spectral indices. But these indices appear to be less reliable in diverse water bodies including lakes, ponds, rivers and streams in coastal regions. The purpose of this study was to develop a new ratio-based two-band spectral index to estimate chlorophyll-a concentrations in diverse water bodies. In this study, an Ocean Optics Spectroradiometer (300–900 nm) was used to collect hyperspectral data from ponds, lakes, streams and rivers in the coastal St. Jones River Watershed, DE from March 2006 to May 2007, while independent in situ water samples were collected concurrently and analyzed for chlorophyll-a and total suspended solids (TSS). The relationship between the reflectance value of individual bands and in situ chlorophyll-a concentration was first examined to identify bands sensitive to chlorophyll-a but not to TSS. The most sensitive band was then selected as the numerator to build a series of ratio-based indices and each of them was examined for its correlation with in situ chlorophyll-a concentration. The ratio showing the highest correlation coefficient was determined as the new ratio-based spectral index. Also, the performance of ratio-based spectral indices from the literature was examined to compare with this newly developed spectral index. It was found that the new spectral index R716/R667 showed high correlation (r =0.82) with in situ chlorophyll-a concentration and it is not sensitive to TSS in diverse water bodies and cross different seasons. This new spectral index could be used as an alternative of two popular indices (R700/R670 and R700/R675) in diverse water bodies with large spatial variation in concentration of chlorophyll-a and TSS. Findings from this study demonstrate the suitability of hyper-spectral remote sensing techniques for mapping chlorophyll-a concentrations in diverse water bodies. Grant no. NA17AE1624