Assessment of Coastal Water Quality Parameters Along Mangaluru Region from AVIRIS-NG Hyperspectral Remote Sensing Data

Under the ISRO-NASA joint program, airborne hyperspectral campaign and simultaneous in situ data collection were conducted along the Mangaluru coastal region with the objective to understand the water quality parameters along the coastal water of the region. A Semi-Analytical remote sensing reflectance model (SA model) has been used to simulate the reflectance spectra of the coastal waters. In situ measured suspended particulate matter (SPM) and bottom depth was given as input to the model, and look up table of chlorophyll-a (Chl-a), colour dissolve organic matter (CDOM), backscattering coefficient of SPM (bbSPM) and absorption coefficient of Chl-a (aChl-a) within an appropriate range were used to optimise the model. The model-derived spectra are then compared with the field-measured spectra. The set of water quality parameters are then optimised based on Spectral angle mapper (SAM) value and minimum euclidean distance (ED) between two spectra. A high variability of backscattering to scattering ratio of suspended sediments is observed and is inferred to play an important role in modulating reflectance spectra along the coastal water of Mangaluru. We then applied the parameterised SA model in an inverse mode on AVIRIS-NG image with input of spatially varying backscattering coefficient of SPM from Landsat 8 to generate map of optically active water quality parameters. The procedure is carried out without any further input of in situ data. The results clearly show the heterogeneous variations of coastal waters influenced by local mixing of river and ocean waters under the tidal and wind forces.