SiPLABoratory

The background

The disagregation of the Eastern block in 1990 shifted the concept of global war to that of local war. Local war is generically defined as a regional conflict aiming at the invasion of land or strategic infrastructures by non governamentally controlled armed groups. The contention of such conflict often requires the disembark of small number of troups in hostile remote regions or the control of sea accessible areas from enemy troups, ships, submarines or any other platform from where attacks could be made. These operations require the rapid assessment of the environmental conditions in a given region to which there is no direct (man) access. By region we mean beaches for troups landing and corresponding access sea corridors as well as the coastal area in front of it.

The concept

Classical means involved historical information, deploying of air sensors (sonobuoys) and remote sensing. Nowadays the development and readily availability of underwater platforms such as gliders and autonomous underwater vehicles (AUVs) drastically changed the possibilities of assessing remote information without the necessity to deploy man forces. Deployments can be made in two phases: phase one, where traditional means such as remote sensing, archival data and airborne sensors are used to assess the overall environmental picture up to, say, 10 kms from shore and the phase two, where ships may approach the area and deploy moving unmanned platforms to cover in detail the coast up to the beach. In both phases the data collected is assimilated into environmental models to be able to predict not only general circulation models tracing the evolution of currents, temperature and surface agitation in the zone but also, beach models of the surf zone. Moving platforms will be used not only to acquire environmental data (water column and bottom) but also to "clean the area" from possible obstacles and mines.

One of the main concerns during phase one is to determine the conditions of acoustic detection in the objective area so as to determine the possibilities of being detected or of detecting ennemy assets possibly hiden in the region or closeby. This involves the assessment of actual environmental conditions (nowcast) and predict those to be encountered in the near future (forecast). The classical approach is to deploy as much sensors as possible to initialize appropriately nested ocean circulation models so as to be able to predict water column and surface evolution in the next 12, 24 or 48 hours. These environmental conditions are then feeded in acoustic numerical propagation models to determine acoustic transmission loss (TL) and therefore compute actual performance surfaces for the area at hand. The main drawback of this aproach is that the optimization is done in the oceanographic but not in the acoustic information. Therefore there is no garantee that the best oceanographic prediction produces the best acoustic prediction. One way of acoustically constraining the predicted acoustic field is to acquire acoustic data together with oceanographic data during the survey process. The technology and the methodology associated with the acquisition of acoustic data for Rapid Environmental Assessment (REA) is called Acoustic REA (AREA).

Technological tools

The technological tools being developed at SiPLAB include easy-to-deploy and easy-to-use Acoustic Oceanographic Buoys (AOB) (figure on the left). AOBs have been successfully tested in several sea trials since 2003 (version 1) and then after 2005 (actual version 2). See sea trials for a complete list of experiments where AOBs have been used. See various publications on the subject such as A. Silva [abstract, PDF] for a full description of the AOB2 testing during MakaiEX'2005 in Hawai, and N. Martins [abstract, PDF] for an example of the usage of the data provided by the AOB2 for AREA objectives.

Transmission Loss at 800 Hz without acoustic data assimilation (above) and with acoustic data assimilation(below)

As an example, this figure shows the TL error in depth and range at 800 Hz measured between the true TL and the predicted TL using the standard procedure of feeding the best oceanographic prediction into the acoustic model for a given water depth mismatch (above). The same TL error but now measured between the true TL field and that obtained from the assimilation of the acoustic data is shown below [drawn from N.Martins et al.].