Water quality monitoring and harmful algal blooms
Figure 1: HAB detection using a multivariate spectral classifier approach showing a (automatically identified) Karen mikimotoi bloom off the U.K. coastline. Areas of green, yellow and orange represent an estimate of its intensity. Areas of blue and purple are clear ocean.
Among the many species of marine phytoplankton some produce toxins that can have serious impacts on fish or human health, and economic consequences if fish farms or shell fisheries need to be closed. These harmful algae blooms (HABs) can occur with very high biomass or may comprise a small part of the flora. Monitoring for HABs traditionally has entailed collection of sea-water samples for species identification or toxicity analysis. However, in situ monitoring has limitations as sampling is expensive and may need to be undertaken over large geographical areas: where sampling is infrequent, detection may happen once the bloom has had an adverse effect. Hence, there has been considerable interest in using satellites to provide synoptic observations of the algal chlorophyll concentrations.
The main research staff envolved in this area are Steve Groom, Peter Miller, Jamie Shutler and Mike Grant with financial support from the BNSC, NERC and the EU.
Try our new MultiView portal developed for near-real time water quality monitoring.
Figure 2: Automatically detected perimeter of a Coccolithphore phytoplankton bloom using computer vision techniques (Shutler et al 2005).
Miller, P.I., J. D. Shutler, G. F. Moore, and S. B. Groom, (2006) SeaWiFS discrimination of harmful algal bloom evolution, International Journal of Remote Sensing, 27 (11), 2287-2301.
Shutler, J. D., P. I. Miller, S. B. Groom and J. Aiken (2005) Automatic near-real time MERIS data as input to a phytoplankton classifier, European Space Agency MERIS and (A)ATSR Workshop 2005, Frascati, Italy.
Shutler, J D., M. G. Grant and P. I. Miller (2005) Towards spatial localisation of harmful algal blooms: statistics-based spatial anomaly detection, SPIE proc. Image and Signal Processing for Remote Sensing XI, 5982, 218-228.
All materials copyright © Plymouth Marine Laboratory 2013.