|A single algorithm to retrieve turbidity from remotely-sensed data in all coastal and estuarine waters|Dogliotti, A.I.; Ruddick, K.G.; Nechad, B.; Doxaran, D.; Knaeps, E. (2015). A single algorithm to retrieve turbidity from remotely-sensed data in all coastal and estuarine waters. Remote Sens. Environ. 156: 157-168. hdl.handle.net/10.1016/j.rse.2014.09.020
In: Remote Sensing of Environment. Elsevier: New York,. ISSN 0034-4257; e-ISSN 1879-0704, meer
Turbidity (T); Water reflectance; Radiative transfer simulations; Uncertainty analysis; T algorithm validation; Southern North Sea; Scheldt; Gironde; Río de la Plata; French Guyana
|Auteurs|| || Top |
- Dogliotti, A.I.
- Ruddick, K.G.
- Nechad, B.
Ocean color remote sensing has been shown to be a useful tool to map turbidity (T) and suspended particulate matter (SPM) concentration in turbid coastal waters. Different algorithms to retrieve T and/or SPM from water reflectance already exist, however there are important questions as to whether these algorithms need to be calibrated specifically for different regions. In the present work the potential generality of a semi-empirical single band turbidity retrieval algorithm using the near infrared (NIR) band at 859 nm in highly turbid waters is assessed. For completeness the use of 645 nm in medium to low turbidity waters is also proposed. Radiative transfer simulations and in situ measurements from various European and South American coastal and shallow estuarine environments characterized by high concentrations of suspended sediments are analyzed. Reflectance and turbidity measurements were performed in the southern North Sea (SNS) and French Guyana (FG) coastal waters, and Scheldt (SC), Gironde (GIR) and Río de la Plata (RdP) estuaries. Simulations showed that uncertainty for turbidity estimation associated with different particle types and bidirectional effects is typically less than 6%. When applied to field data from the five different sites, the semi-analytical algorithm performed well: turbidity estimates were within 12% and 22% of in situ values. A good performance was also found when the entire database was analyzed (n = 106) with a mean relative error of 13.7% and bias of 4.8%. The good performance of the algorithm for all these regions, despite differences in sediment characteristics, and the results of the radiative transfer simulations suggest the global applicability of the algorithm to map turbidity up to 1000 FNU. Consequently regional algorithms to retrieve SPM concentration from reflectance can be designed by combining this global algorithm to retrieve T from water reflectance with a regional relationship to convert T to SPM. This has the very practical advantage that the measurements needed to calibrate the latter T/SPM conversion for any new region are much easier and cheaper than in situ reflectance measurements.