Optical remote detection of surface water quality for open ocean waters has been ongoing since the 1960's. This method of detection is much more challenging, and not yet standard practice, for the more optically complex inland and coastal waters. A series of empirically developed models were built relating remote sensing reflectance (Rrs) for Compact Airborne Spectrographic Imagery (CASI), acquired over both inland and coastal water bodies of the Cowichan River watershed in British Columbia, to concurrently sampled water parameter concentrations. The models predict concentrations of chlorophyll-a and its pheopigments (as a proxy signal of phytoplankton), coloured dissolved organic matter (CDOM) and total suspended solids (TSS), using techniques including simple regression, step-wise regression, band ratios, derivative analysis, and fluorescence line height (FLH). The most statistically significant models were ultimately applied. Each of the models was found to be statistically significant, although confidence in the lake models and the CDOM model for the bay are limited due to potential chance correlations. The Chl-a and TSS models for Cowichan Bay were found to be acceptable based upon an understanding of the phenomenon being investigated. These two parameters were correlated to short blue wavelengths, where Chl-a is primarily correlated with absorption at 451 nm and TSS with scattering at 462nm. These findings provide support of the application of the standard configuration of MERIS bands as the common configuration for further coastal and inland water remote sensing studies.