An improved algorithm to retrieve Chlorophyll-a
After validation of various satellite-based algorithms with in situ data from several Flemish and Dutch water bodies, we have implemented an improved algorithm to retrieve Chlorophyll-a products from Sentinel-2. This machine learning approach, developed by Pahlevan et al. (2020), makes use of Mixture Density Networks (MDN) and is trained on a global in situ data set.
The updated Algorithm Theoretical Base Document is available here.
Terrascope systematically provides Chlorophyll-a products of Belgium via the Terrascope viewer, Virtual Machines, Notebooks, web services and also via the openEO Platform. Because of the automated and operational processing workflow, additional Chlorophyll-a products of other regions can be processed upon request.
The new and improved Chlorophyll-a products (Sentinel-2 archive from July 2015) are available in the Terrascope viewer. The Turbidity (TUR) and Suspended Particulate Matter (SPM) products in Terrascope are also available and remain unchanged. The previous version of Chlorophyll-a products was supported till 3.10.2022 and will remain available on the Virtual Machines till 31.01.2023.
Pahlevan, N., Smith, B., Schalles, J., Binding, C., Cao, Z., Ma, R., Alikas, K., Kangro, K., Gurlin, D., Hà, N., Matsushita, B., Moses, W., Greb, S., Lehmann, M. K., Ondrusek, M., Oppelt, N., & Stumpf, R. (2020). Seamless retrievals of chlorophyll-a from Sentinel-2 (MSI) and Sentinel-3 (OLCI) in inland and coastal waters: A machine-learning approach. Remote Sensing of Environment, 240, . https://doi.org/10.1016/j.rse.2019.111604