17 Basic Ecosystem Dynamics in the Red Sea as Seen by Sundry . . .
341
interface, can create a deeply mixed layer (normally around 300 m, but mixing depths
in excess of 800 m have been measured in unusually cold winters) and increase nutrient concentration in surface waters (Genin et al. 1995). At the basin scale, nutrient
concentration generally increases from north to south, and so does, consistantly with
the nutrient distribution, primary production (Weikert 1987).
According to the environmental traits outlined above, the Red Sea can be divided
in various geographical provinces bounded by bathymetry, physical properties and
dynamics, and wind patterns (as e.g. in Siedler 1969). In the following, the general
trends of the Red Sea pigment field were analyzed at the basin scale, as well as in the
northern and southern regions separated by the critical area around 20
◦ N, where the
seasonal monsoon wind shift reaches its limit and the surface circulation appears to
be dominated by anticyclonic eddies (Johns et al. 2000, Acker et al. 2008). Similarly,
the relationship to wind speed was considered over the same scales (full, northern
and southern basin), in order to try and couple bio-optical patterns to their potential
atmospheric forcing.
17.2 SeaWiFS and QuikSCAT Historial Data Record
The present work is based on a time series collected by the SeaWiFS, a dedicated
ocean colour imager that operated between 1997 and 2010, acquiring Earth radiance
data in 8 spectral bands (ranging from 402 to 885 nm), with a spatial resolution of
∼ 1.1 km at nadir. Concurrent data collected by the MODIS-Aqua, a multispectral
imager active since 1999 (on board the Aqua orbital platform) and equipped with 36
spectral bands, 9 of which (ranging from 405 to 877 nm) dedicated to acquire ocean
radiance data at ∼ 1 km resolution, were also used for comparison. In order to avoid
incomplete coverage due to cloud cover, in particular during the summer monsoon, or
to desert dust aerosols, rather common in this region–or again to processing algorithm
failure due to unaccounted air or water constituents–only monthly composites (and
climatologies) of the original data were considered. As the number of valid pixels
increases with larger and longer binning intervals, the compositing process generates
complete, cloud-free images. However, the variability due to highly dynamical events
in a given area is averaged out too, and the composites retain only those patterns that
persist over significant space and time scale.
The monthly composite (and climatological) data were obtained directly from
web-based services
2 , which provide a suite of data products that includes SeaWiFS
2 i.e. the Goddard Earth Sciences (GES) and Data Information Services Center (DISC) Interactive Online Visualization ANd aNalysis Infrastructure, (GIOVANNI), available for public
access and use at http://disc.sci.gsfc.nasa.gov/giovanni/overview/, by the National Aeronautic and
Space Administration (NASA); as well as the Global Marine Information System (GMIS), at
http://gmis.jrc.ec.europa.eu/, by the Joint Research Centre (JRC) of the European Commission
(EC). Both web services allow for the extraction of fully processed data, from multiple sensors, and
the use of various data analysis and visualization options.
Précédent

- 351/436

Suivant