Development of optical models for assessing the trophic status of coastal waters

The present work aims to develop the water quality index for the optically complex turbid coastal waters of Bay of Bengal based on the optical properties of water column. In this study, a new trophic index was derived as a function of the ratio of vertical diffuse attenuation coefficient (Kd) at 531 nm to 490 nm, expressed as Kd(531/490), to obtain the predictive scaling for monitoring the eutrophication status. The trophic index was also expressed in terms of Turbidity (NTU) which employs the Secchi Depth (Zsd) as a crucial intermediate parameter to derive the desired product. It was found that the ratio of Kd in the green-blue region of the visible spectra plays a paramount role in predicting the eutrophic conditions. Moreover, turbidity can be leveraged to establish the bio-optical link-pin between chlorophyll concentration and Zsd to determine the water quality in terms of trophic state index (TSI). The chlorophyll concentration was fixed as the primary classifier to derive water quality index on the scale of 0 to 100 denoted by non-dimensional reference scale of TSI (referred as TSIref). TSIref was then, used as standard scale to predict the accuracy of the modelled TSI products obtained by correlation of Kd(531/490) and Zsd with chlorophyll concentration, referred as TSIKd and TSIZsd respectively. The performance of the modelled TSI products was tested over a wide variety of water types, ranging from relatively clear to highly turbid waters. The statistical measures calculated between the modelled TSI and the TSIref revealed that the trophic state of coastal and estuarine waters can be estimated with a percent mean relative error of ±20% and RMSE∼0.015 with R2 of approximately 0.9. It was found that the newly derived TSI are in good agreement with TSIref. The spatial contours for modelled and reference TSI were plotted for different seasons (i.e. for the month of August 2013 and January 2014) which further revealed the potential capability of the modelled TSI to closely predict TSIref. Thus, it can be implied that the proposed optical parameters can be adopted as crucial inputs to determine the eutrophication level of relatively turbid and turbid productive waters.