Modeling water quality using Terra/MODIS 500m satellite images

A study was conducted in Hong Kong with the aim of deriving algorithms for the retrieval of turbidity, chlorophyII a and suspended solids concentrations from Terra/MODIS 500m level 1B reflectance data. Rigorous atmospheric correction using Radiative Transfer Model coupled with the inputs of AERONET data and MODIS atmospheric products was carried out for deriving the water leaving reflectances. The in-situ measurements were compared with coincident water leaving reflectances using different sets of empirical algorithms. Due to the high variation of water quality in Hong Kong, the in-situ data were divided into three groups based on the spatial locations of the measurements: east zone (similar to case I water), west zone (similar to case II water) and harbour zone (in between case I and case II water). Significant correlations were observed between water-leaving reflectances and in-situ marine data on the three zones while only moderate correlations can be found on those with the entire dataset. These suggest the single empirical algorithm may not be applicable on the areas with high variation of water quality and MODIS 500m data are feasible for water quality retrieval in a local scale.