NL FR EN
www.belgium.be

Remote sensing for characterization of intertidal sediments and microphytobenthic algae (ALGASED)

Research project SR/00/109 (Research action SR)

Persons :

  • Dhr.  MONBALIU Jaak - Katholieke Universiteit Leuven (KU Leuven)
    Coordinator of the project
    Financed belgian partner
    Duration: 1/12/2006-30/9/2009
  • Prof. dr.  SABBE Koen - Universiteit Gent (UGent)
    Financed belgian partner
    Duration: 1/12/2006-30/9/2009
  • Dr.  FORSTER Rodney - Center for environment, fisheries and aquaculture sciences (CEFAS)
    Financed foreign partner
    Duration: 1/12/2006-30/9/2009
  • Dr.  VAN der WAL Daphné - Netherlands Institute for Ecology (NIOO)
    Financed foreign partner
    Duration: 1/12/2006-30/9/2009

Description :

Context and objectives

The upper mm’s of intertidal sediments harbour dense microalgal biofilms (microphytobenthos - MPB) which together with the phytoplankton support the estuarine food webs. MPB composition, biomass and production display marked spatial heterogeneity and temporal variability at different spatial and temporal scales. Understanding the spatial and temporal variability of MPB at different time- and space-scales requires a combination of observational, remote sensing and modelling approaches.

The main objectives of ALGASED are:

- improvement and fine-tuning of biomass estimates and modelling of primary production (PP)
- assessment of the performance of various types of satellite data for the quantification of MPB biomass and sediment physical properties
- multi-scale analysis incorporating both ground, airborne and satellite data
- accuracy assessment and improvement of supervised and unsupervised classification methods for hyperspectral imagery


Methodology

Usually change in chlorophyll a (chl a) absorbance is used as a biomass estimator. This estimator however shows reflectance saturation at high biomass values, and also does not allow to distinguish between algal groups. It will be investigated how other pigments with non saturated absorption features in other bands can provide additional information on the composition of the MPB and how this can lead to better modelling of PP.
Match-ups (within a few weeks of airborne hyperspectral imagery) of past and future satellite data will be collected from high resolution satellite sensors with four spectral bands. It will be investigated in how far these images can provide information on MPB biomass quantification and on sediment properties.
Apart from traditional spatial analysis methods such as correlograms and semivariograms, newly developed techniques, such as Principal Coordinates of Neighbor Matrices, will be applied.
New, intensive and hopefully successful field campaigns combined with the acquisition of new hyperspectral images will provide independent measurements to test supervised and unsupervised classification methods.


Results expected

- algorithms to estimate the amount and composition of biofilms using several existing and new indices and for the detection of nuisance algal blooms plus maps of MPB biomass and composition
- a validated and calibrated Fluorescence and Grazing Index that predicts the physiological status of the MPB
- description of the influence of moisture content and mud fraction on absorption features of pigments
- a model for the net primary production, which considers the primary production of MPB, the losses due to respiration, the availability of nutrients, the losses due to grazing animals plus maps of PP
- guidelines about the usefulness of satellite data to study intertidal sediment and ecosystems. These guidelines will be general so that they can be used to study other intertidal flats and to choose the most suitable sensor.
- description of the interactions between bio-physical sediment parameters, topography and hydrodynamics and of the organization of these interactions at multiple scales
- determination of the main variables influencing the spatial distribution of MPB and sediment properties
- construction of a validated historical archive for our test sites, and advice on band selection for future missions
- models at multiple scales of bio-physical sediment properties
- accuracy assessment and improvement of supervised and unsupervised classification methods based on clustering techniques
- a database that integrates the available data and metadata on the study sites

Documentation :