[Note: The first part of this abstract was not received electronically] ...el entorno de Cuba", "Descripcion de particularidades de la Corriente del Lazo", entre otros.

Se propone la conformacion de un grupo de trabajo regional para la implementacion de las tecnicas y metodos de Teledeteccion como parte de un sistema de observacion del oceano y en particular del Canal de Yucatan.

Synoptic realizations are presented of the simulated circulation of the IAS using the Princeton Ocean Model (POM), called IAS-POM in this implementation, as forced by monthly mean ECMWF winds. For example, interactions of the meandering Caribbean Current with the wind-driven upwelling regime along the Venezuelan coast give rise to characteristic filamentous surface anomalies, perhaps most strikingly in plankton concentration fields calculated with an embedded marine ecosystem model.

Such simulations serve to help define the phenomena and scales of variability which must be accounted for in a regional GOOS system; such as, IAS-GOOS. To bring the design of IAS-GOOS to a high level, data assimilation experiemnts; such as, IAS-RODAE, are needed with researchobserving and modeling sysems.

Present thinking on the design of IAS-RODAE is reviewed. Included is the need for participation by IAS scientists as critical-minded users of IAS-RODAE information products.

Understanding and predicting the connectivity of coral reef organisms linked by larval dispersal is a key goal of tropical coastal ecosystem science and management. As oceanographers make advances modelling transport processes between reefs, ecologists should be prepared to embrace the population dynamics of organisms at larger metacommunity scales. A metacommunity is demographically closed but contains multiple open local populations. Metacommunity models of coral dynamics would aid the identification of larval source and sink areas and help identify the boundaries of demographically closed populations. Metacommunity models would also aid the understanding of species extinctions and help formulate transboundary management strategies to conserve ecosystem function. Existing metacommunity models are not spatially realistic or proven to represent physical and biological processes at appropriate spatial and temporal scales. Before more realistic alternatives can be established, further research is needed into (i) the larval transport between reefs (i.e. the coupling of reef-scale and oceanic models of water circulation, the pre settlement mortality rates of larvae, and larval mobility in the water column), (ii) the influence of larval supply on coral population dynamics at local (tens of kilometres) scales (i.e. processes of settlement behaviour, post-settlement mortality, and the pre-emption of space by algae), and (iii) processes affecting the net fecundity of local populations (i.e. interactions of colony size, partial mortality rate, competition with algae, and the influence of habitat, physical disturbance, herbivore pressure, nutrification and sedimentation).

Further, most coastal habitat mapping is conducted on an ad-hoc basis with little consistency in terminology and ambiguous documentation. These limitations obstruct interpretation and integration of maps for parameterising population models and coastal management applications, particularly at regional (international) scales where standardisation is urgently required. Habitat classification methods are discussed briefly.