next up previous contents
Next: SEABED SAMPLING Up: INITIAL RESULTS Previous: INITIAL RESULTS   Contents

BATHYMETRY

Pattern of the multibeam lines during the 12 days of operation in the study area are presented in Fig.4. A surface of approximately 8500 Km$^2$ was covered in the Thetis, Conrad, Oceanographer Deeps and Coral Sea Peak areas. A partial processing of the acquired data by the RESON 8160 was carried-out during the cruise with the Caraibes and Neptune softwares. The quality of the data acquired varied from very good to acceptable, excluding a few lines obtained in bad weather conditions because of the rough seas. A detailed survey was carried-out in the Thetis Deep area extending the survey to the south in the transition zone to the Hadarba Deep and to north in the transition zone to the Nereus Deep, where the eastern margin was also mapped during the water sampling operations and transit to the Coral Seapeak.

The DTM and maps were produced by (a) filtering the UTM-37 25m PDS-2000 DTM data with a batch ISMAR procedure (filter_bat) to eliminate spikes and high-frequency noise, (b) creating NETCDF grids with the GMT nearneighbor or surface algorithms or straight to with xyz2grd, (c) gaussian filtering, when required, and (d) plotting the maps in the UTM and geographical domains (GMT routines complemented by ISMAR routines).

The bathymetry of the Thetis deep shows three asymmetric "subdeeps" which are smaller and shallower from north to south. The "subdeeps" are separated by transition zones and are characterised by several volcanoes and evaporite-mobilization structures. The structural architecture of the deep is generally formed by a series of fault segments with dominant NW orientation, sub-parallel to the rift axis. These are linked by relay ramps, some of which are breached by local transfer faults.

The Conrad Deep (Fig.16) shows an elongate shape, oriented parallel to the Aqaba-Dead Sea fault, and the Oceanographer Deep (Fig.17), seems to be in some way connected to the Brothers islands, localized and very close right west of the deep.

We have also produced a map of the Nereus Deep by merging the SEABEM data published by [Pautot (1983)] with the swaths we collected on the eastern flanks during the station approach, the brine sampling and the transit to N (Fig.(18). Since the SEABEAM data were apparently collected with a constant speed of sound of 1500 m/s, we tried to produce the best fit and applying the new function to the digitized contours, by

Figure 16: The Conrad Deep. Bathymetry from PDS-2000 DTM 25m.
\begin{figure}\centerline{\epsfig{
file=IMG/CONRAD_PDS2000.eps,width=15cm}}
\end{figure}

Figure 17: The Oceanographer Deep. Bathymetry from PDS-2000 DTM 25m.
\begin{figure}\centerline{\epsfig{
file=IMG/OCEANOGRAPHER_PDS2000.eps,width=12cm}}
\end{figure}

Figure 18: The Nereus Deep. Compilation with the data of [Pautot (1983)] and present survey.
\begin{figure}\centerline{\epsfig{
file=IMG/NEREUS.eps,width=12cm}}
\end{figure}


next up previous contents
Next: SEABED SAMPLING Up: INITIAL RESULTS Previous: INITIAL RESULTS   Contents
2005-11-05