Remote chlorophyll-a retrieval in Case-II waters using an improved model and IRS-P6 satellite data

Abstract

In remote chlorophyll-a (chl-a) retrieval in Case-II waters, there always exists some limitations from empirical parameters in the estimating models and uncertainties from the chl-a specific absorption coefficient. In this study, we present a newly improved three-band model in a case study in Shitoukoumen Reservoir for direct calculation of chl-a without any empirical parameters derived from regression. Inherent optical properties of the reservoir were investigated to determine parameters in the improved model. Results show that taking into account variations from chl-a specific absorption, as well as absorption by pure water, the improved model performed well in the field study used for model calibration and also robustly in two other field studies. The findings underline the rationale behind the model and demonstrate a potentially general solution for assessing chl-a in Case-II waters. In the evaluation of P6 data, the traditional near-infrared-to-red reflectance ratio (band 4/band 3) could estimate chl-a with root mean square error below 2.17μg l−1, which confirms that P6 data could be used potentially to determine chl-a concentration.