Hyperspectral and multi-mission high resolution optical remote sensing of aquatic environments (HYPERMAQ)

Start-End 01/12/2016 -  30/11/2020
Programme STEREO 3
Contract SR/00/335
Objective Algorithms for remote sensing of suspended particulate matter and chlorophyll concentration are quite mature and applications relating to coastal sediment transport and environmental monitoring are routinely using satellite data. The challenge for researchers is now to estimate more than just concentration. Sediment transport users want spatial and temporal information on particle size and composition. Marine biologists and water quality managers want information on phytoplankton and benthic community composition, which is particularly difficult to obtain from remote sensing in turbid waters. The advent of spaceborne hyperspectral instruments offers the potential to yield more information on aquatic particles, both algal and non-algal.

Dedicated ocean colour missions provide daily data but only at moderate spatial resolution (~300-1000m). Many high spatial resolution (~10-100m) spaceborne instruments designed for land applications are also suitable for aquatic applications, although algorithm developments are needed for atmospheric correction and for aquatic product retrievals. For patchy distributions of micro- and macro-algae this high spatial resolution will be important to ensure that spectral features are not missed because of under-resolution of patches.

The general objective of HYPERMAQ is to develop and test new algorithms for aquatic remote sensing of coastal and inland waters, using both hyperspectral and high resolution multispectral satellite data to provide more than “just” concentration of suspended particulate matter and chlorophyll. Test sites focus particularly on turbid waters.

Specific remote sensing objectives include:
  • Design and testing of hyperspectral algorithms for phytoplankton species composition in turbid waters and identification of their limitations.
  • Design and testing of hyperspectral algorithms for macro-algae cover and type.
  • Design and testing of new algorithms for retrieval of particulate backscatter spectral slope and particulate backscatter: absorption for non-algal particle size/type in turbid waters.
  • Refinement of algorithms for high spatial resolution missions with limited spectral bands
  • Design of an in situ instrument system for a hyperspectral validation network
  • Validation and exploitation of high spatial resolution data as a tool to study plankton and benthic algal dynamics and bloom formation in coastal and inland waters
  • Provide recommendations for future satellite missions as regards design of spectral bands and signal:noise for aquatic applications
Specific scientific exploitation objectives include:
  • Increase our understanding of multi-species micro-algal blooms in a shallow turbid coastal area characterized by pronounced spatial and temporal variation in turbidity and nutrients related to weather and climate variability.
  • Increase our understanding of micro- and macro-algal blooms in shallow freshwater and brackish inland waters, with focus on benthic-pelagic interaction and algal groups.
  • Increase our understanding of sediment transport in coastal and estuarine systems by supplementing existing remotely sensed concentration data with particle size information.
Method Bio-optical properties of micro-algae will be analysed from cultures and field phytoplankton samples, including scattering, backscattering, fluorescence and absorption properties.

Hyperspectral algorithms will be developed for micro-algae in turbid waters firstly from field absorption next from field reflectance spectra. The developed algorithms will be tested using simulated spectra derived from sample spectra with varying added types and concentrations of non-algal particles, and using measured field data. All algorithms will also be tested for applicability for data with limited spectral bands.

Bio-optical properties of macro-algae
will be analysed using cultures of macro-algae from the Spuikom. Reflectance spectra of macro-algae will be measured above-water using a hyperspectral camera, with a spectral range of 300–1100 nm every 5 nm, with relative calibration using a Spectralon plaque. Spectra will be measured for seaweeds grown under different light and nutrient levels to accommodate for physiological status.

Hyperspectral algorithms will be developed for macro-algae based on the field and the hyperspectral camera measurements. Algorithms will also be tested with data simulated from the lab measurements and including varying concentrations of CDOM, algal and non-algal particles. Sensitivity studies will show the impact of water depth for submerged macro-
algae. The algorithms will be refined, when possible, to use high spatial resolution missions with limited spectral bands.

Algorithms will be developed for particle size and type in turbid waters based on estimating the
spectral slope of the particulate backscattering coefficient. The particulate absorption coefficient, whose spectral variations contain information of the particulate composition (mineralogy, trace metals, etc.), can also be retrieved. Simulations and measurements will be made to determine the influence of specific inherent optical properties, and particularly the ratio of particulate backscatter:absorption, on reflectance spectra and hence provide a basis for retrieving particulate backscatter:absorption.

A "pan and tilt" radiometer pointing system will be tested and developed in the frameworok of a future "AERONET-OC" style hyperspectral validation network.
Result Expected Scientific Results

It is expected that the project will:
  • Develop and validate new algorithms for phytoplankton species composition in turbid waters based on hyperspectral data
  • Develop and validate new algorithms for floating and submerged macro-algae based on hyperspectral data
  • Develop and validate new algorithms for suspended particle size/type in turbid waters based on hyperspectral data
  • Improve algorithms for and exploitation of high spatial resolution missions
  • Develop and test extensively an automated in situ spectroradiometer system for deployment on platforms and buoys to provide validation data for all optical missions as a precursor for a future “HYPERNET-OC” international network, similar to the successful AERONET-OC network but hyperspectral.
  • Provide quantified signal:noise requirements for future hyperspectral satellite missions and optimal “aquatic application” band sets for future multispectral satellite missions with.

Expected Products and Services

Algorithms will be developed for the following new products:
  • Micro-algae species composition (in turbid waters)
  • Macro-algae coverage and, if possible, composition or physiological state
  • Non-algae particle type and/or size

Potential Users
Potential users include Water quality managers, aquatic scientists including biologists and sediment transport specialists, dredging companies, space agencies, industrial designers of hyperspectral space missions as well as other remote sensing scientists.
Website link
Team Member: CATTRIJSSE Andre VLIZ (Vlaams Instituut voor de Zee)
Team Member: SABBE Koen Ugent (Universiteit Gent)
Project Leader: RUDDICK Kevin Institut royal des Sciences naturelles de Belgique/Koninklijk Belgisch Instituut voor Natuurwetensch
Team Member: DOXARAN David Université Pierre et Marie Curie Paris VI
Sensors used
Related presentations
Related publications
Datasets used