Remote sensing of chlorophyll-a concentration for drinking water source using genetic algorithms (GA)-partial least square (PLS) modeling

► Phytoplankton pigment of Chl-a for water supply source is estimated using a large in situ hyperspectral data set. ► An adaptive genetic algorithm-partial least square (GA-PLS) and three-band algorithm (TBA) were applied for Chl-a estimation. ► GA-PLS outperforms TBA for Chla estimation and shows better temporal transferability. ► TBA shows stable performance for Chl-a estimation, and is affected by nonalgal particle abundance. ► Hyperspectral remote sensing via GA-PLS or TBA modeling provides useful tool for algae surveillance.