Simple methods for satellite identification of algal blooms and species using 10-year time series data from the East China Sea
Fang Shen,Rugang Tang,Xuerong Sun,Dongyan Liu
DOI: https://doi.org/10.1016/j.rse.2019.111484
IF: 13.5
2019-01-01
Remote Sensing of Environment
Abstract:Long-term ocean color satellite missions have the ability to help monitor algal blooms. However, satellite identification of algal blooms and species in turbid coastal waters has been challenging. There is an urgent need for simple and effective methods to identify locations, areas, durations, and species present in algal blooms through satellite observation, to aid in the operational and emergency monitoring of the marine environment. In this study, based on a three-band blended model, we propose an indicator (Red tide Detection Index, RDI) that is suitable for the purpose of detecting algal blooms in turbid coastal waters using multi-source ocean color data. MERIS, MODIS, and GOCI data used for the detection of algal blooms demonstrate consistent results using the RDI indicator, and these results correlate with in situ investigations. Furthermore, based on a green-red spectral slope, we propose a method that uses MERIS data to identify dominant species of diatoms and dinoflagellates in algal blooms in the East China Sea (ECS). The 10-year time series MERIS data collected between 2003 and 2012 indicates that algal bloom occurrences are mainly distributed in the nearshore areas of the ECS, and possess a distinct climatological cycle. The 10-year time series MERIS data help discriminate diatom and dinoflagellate blooms in the ECS, and show the dissimilarity in the seasonal timing of their life cycles. We find that these two dominant species in algal blooms are usually distributed in different spatial locations in the ECS. Additionally, by defining a divergence index (DI), the limitations of spectral resolutions of satellite data used for algal bloom species differentiation are quantitatively assessed.