NL FR EN
www.belgium.be

Remote sensing of turbid waters in the Short Wave Infrared (SWIR: 1-3µm) (SEASWIR)

Research project SR/00/156 (Research action SR)

Persons :

Description :

Context and objectives

The limits of optical remote sensing of aquatic systems are being progressively pushed into more and more turbid water, because of growing interest in the environmental impacts of the world’s most important river estuaries/plumes and in the monitoring of inland waters.
This evolution has already led to refinement of the black pixel assumption for the near infrared (NIR: 700-1000nm) marine reflectance and to the development of NIR-based Total Suspended Matter (TSM) retrieval algorithms for moderately turbid waters (10-100 g/m3).
Extension of such approaches to extremely turbid waters (TSM> 100 g/m3) requires further refinement of theory and atmospheric correction algorithms. Use of the Short Wave Infrared (SWIR, 1-3 m) range then offers new opportunities, but requires fundamental research and
measurements for a sound theoretical basis. Providing this basis is the focus of the SeaSWIR proposal.

The main objectives are:

1. Determine the variability of marine reflectance in the Short Wave Infrared as function of Total Suspended Matter (TSM) concentration/turbidity for a range of study areas.
2. Analyze the effects of misapplication in extremely turbid waters of the SWIR black pixel assumption in atmospheric correction algorithms and propose a refinement of the SWIR black pixel method.
3. Provide information for the design and/or exploitation of SWIR bands on the next generation of ocean colour sensors (OLCI, SGLI, SABIAMAR, etc.).

Methodology

The black pixel assumption and the variability of the marine reflectance in the SWIR as function of TSM/turbidity will be assessed by performing and analyzing in situ water reflectance measurements in the SWIR, analyzing images, theory and radiative transfer code (Hydrolight) simulations.
An algorithm for estimating TSM will be developed for the SWIR based on the inversion of the theoretical model. We will further evaluate theoretically the implications of misapplication of the SWIR black pixel assumption on the water-leaving radiance using the
MODIS SWIR atmospheric correction algorithm.
We will theoretically determine whether the salinity-related variability of pure water refractive index and hence Fresnel reflectance coefficient in the SWIR can be detected by a (future) radiometer by use of sunglint.

Results expected

- In situ measurements of the spectral variability in the SWIR water reflectance as function of TSM concentration/turbidity for several study areas.
- In situ measurements of particulate backscattering in VNIR and SWIR.
- A model estimating marine reflectance as function of TSM or turbidity and wavelength and other parameters.
- An algorithm to invert this model to estimate TSM and turbidity from marine reflectance in extremely turbid waters.
- An evaluation of the effects of misapplication of the SWIR black pixel assumption in atmospheric correction and quantification of errors in the retrieval of the water-leaving radiance. An assessment from which TSM concentrations we have to account for SWIR ocean contributions.
- Theoretical base to evaluate whether the salinity-related variability of pure water refractive index and hence Fresnel reflectance coefficient in the SWIR can be detected by a (future) radiometer by use of sunglint.

Products and services expected

- Database of IOP data, water reflectance and concentration measurements
- APEX datasets
- Annual progress/activity report
- Website of the project
- Conference papers
- At least one journal paper

Documentation :