Geomatics Research GroupResearch ProjectsWRIScS Remote Sensing Pilot Study wriscs-ad.gif (6918 bytes)

Watershed-Reef Interconnectivity Scientific Study (WRIScS):

A Pilot study of the use of satellite Ocean Colour Imagery for Water Quality Monitoring in Optically Shallow Tropical Coastal Environmnets.

The Watershed-Reef Interconnectivity Scientific Study (WRIScS, 1997 - 2001) had the objective of promoting a balance between sustainable land-use development and reef conservation in Belize, Central America. The pilot study deals with a one year marine research programme funded by the European Union as an extension to the main project.   A full overview of WRIScS can be found at the main site: www.wriscs.org.

The pilot study addressed the use of ocean colour satellite imagery for effective monitoring of water quality in the coastal zone. Water colour is the result of the action of light on the materials suspended and dissolved in it. The relation between water quality and water optical properties is a long-established open-ocean science. The benefits to Belize of the development of such a tool would be the daily acquisition of data relating to a range of water quality issues, covering the whole of the Belize marine zone, and at minimal cost.

Remote sensing is not yet used in water quality assessment in the coastal zone because of the problem of optical shallowness. This term describes the condition where sunlight can penetrate all the way to the seabed and be reflected back upwards and out again through the water surface. This bed-reflected light adds another element of colour to the water surface, and masks the colour due to water quality.

This project has addressed a range of feasibility issues concerned with setting up a marine remote sensing facility in Belize (including awareness and training, use of image acquisition software, the impact of cloud cover, and validation of the image-derived data). It has been primarily involved however with an examination of the bed reflectance problem. A basic strategy has been to move away from 'whole image' analysis, which is the major advantage of satellite imagery, but which presents an unresolvable complexity when dealing with bed reflectance. The classical spot sampling approach has been used to create time series data over a six month period at some 40 stations along W-E aligned transects at contrasting sites in north and south Belize coastal waters. Each 'site' is one satellite image pixel, 1km2. SeaWiFS has been used as it provides daily images, at no cost to developmental projects.

At each of these sites, water quality has been recorded at 1-2 week intervals, and the seabed mapped twice. The field measurements were carried out by Raleigh International volunteers, based on two island 'dive-sites' from March to June and July to September 2001. Water quality measurements included secchi-depth observations and sampling for suspended particulate matter, plankton, chlorophyll-a and coloured dissolved organic matter. Samples were preserved and sent to the UK for analysis. SCUBA transect surveys were used to map the bed types and depth, and to collect information on bed colour using digital cameras and Munsell cards. In addition to this basic measurements, scientists used the dive-team facilities to make more complex observations, including RoxAnn acoustic mapping of pixel zones, CTD profiles of salinity, temperature and turbidity, and a Profiling Reflectance Radiometer to measure light attenuation through the water column.

From the wealth of data generated over two three-month expeditions (covering both the wet and dry seasons) , descriptive models have been developed of the light-reflective properties of the seabed within each pixel-station, and of the seasonal variation of water quality within zones incorporating stations of similar conditions.

The three types of data have been interrelated using a simple optical equation developed by earlier researchers, which breaks down the light reflecting from the ocean surface into water-column (quality) and bed- sourced components. The equation has been developed into a simple interactive model which can be used to test the sensitivity of the reflected light output to the effects of the various contributing components. This model has underlined the importance of the attenuation properties of the water column in determining the influence of bed-reflected light. It has also demonstrated that peak reflection from bed occurs between 490 and 555nm, and that hardly any light is reflected back at the spectral extremes of 412 and 665nm. Detailed inspection of the output from the model at a range of sites of different bed conditions show that although the general responsiveness of the model is good, it gives unrealistic results under certain conditions. At the close of the investigation, three new objectives have been prioritised to carry the project forward. These tasks are to:

  1. develop an algorithm for reliably predicting the diffuse attenuation coefficient of water at 412 or 665nm from satellite data
  2. better characterise the reflectance properties of biological surfaces, including the shading effects of relief and vegetation under-storeys and
  3. investigate and develop the optical theory within the light model.

The project concludes on an optimistic note that although a solution to water quality monitoring in optically shallow water has not been fully revealed in this pilot study, an excellent database has been generated for continuance of the investigation, and there is much enthusiasm and good-intention to continue with the research. Actual steps forward have been made in the revelation that the 665 and 412nm bands are effectively optically deep in most of the Belize coastal zone and can be used for monitoring for suspended sediment and Coloured Dissolved Organic Matter respectively. Hardware and software for acquiring and processing SeaWiFS images have been donated to the Coastal Zone Management Institute. The WRIScS team recommend that all who can should take advantage of this facility, with the intention that familiarity with the data and system will give confidence for the future development of this tool for water quality monitoring.

SeaWiFS Chlorophyll Imagery

Level 2 SeaWiFS OC4v4 chlorophyll images processed from Level1 HRPT data (1km resolution) using SeaDAS v4.1 at the University of Plymouth.  The numbers in the table represent the Julian day and if you click on the image you will see the full sized version.   Below is also the chlorophyll colour palette with cocentrations given in mg m-3.

We would like to thank the SeaWiFS Project (Code 970.2) and the Goddard Earth Sciences Data and Information Services Center/Distributed Active Archive Center (Code 902) at the Goddard Space Flight Center, Greenbelt, MD 20771, for the production and distribution of these data, respectively. These activities are sponsored by NASA's Earth Science Enterprise.  The SeaWiFS imagery provided here are only available for research and education and you should also be SeaWiFS Authorised.

chl-scalebar.gif (4435 bytes)

S2001201181946chlor_a.gif

201

202

203

216

218

219

223

228

231

237

239

240

242

244

245

246

247

249

252

253

254

255

256

258

261

     

 

Publications

  • End of project report:
  • Case Study on the Geomatics Awareness WWW site
  • Goodban, I. 2002. Mapping of Spatial / Temporal Variations of In-Situ Optical Properties in the Belezian Coastal Zone. BSc Hydrography, University of Plymouth.
  • Gunn, J. 2002. Remote Sensing of Water Turbidity Monitoring in the Coastal Zone. BSc Hydrography, University of Plymouth.
  • Kirkwood, D. 2002. The development of a low-cost approach for the measurement of in-situ seabed reflectance. MRes Applied Marine Science, University of Plymouth.
  • Lavender, S., Nunny, R., Nagur, C. and Tillett, D. 2002. Remote sensing of water quality in the Belzian coastal zone using satellite ocean colour imagery. European Geophysiacl Society Annual Conference, Nice, 21 - 26 April. [Oral presentation]
  • Nunny, R. and Lavender, S.  2002. Investigation of a practical method for removing the bed reflectance signal from ocean colour images. Challenger Centenary Conference: Marine Science 2002, Plymouth, 9-13 September. [Oral presentation]
  • Lavender, S., Nunny, R., Tillett, D., Kirkwood, D., Miller, A., Moore, G. and Cook, J. 2001. Monitoring of the Belizian coastal zone using satellite ocean colour imagery. Proceedings of RSPS 2001: Geomatics, Earth Observation and the Information Society, London, United Kingdom, pp 154-163. [PDF file 1.50 MB]

Web page maintained by Sam Lavender. Last updated on 8 November 2002