Searching for the German U-559 Submarine with GIS

By Michael Cooper, Laura Crenshaw, and Tracie Penman, Veridian, Chantilly, Virginia, and Edward Saade, Thales GeoSolutions (Pacific), San Diego, California

The search area for the U-559, shown relative to the claims and boundaries of littoral nations as maintained in Veridian's Global Maritime Boundaries Database using Esri's ArcPlot.

Recent advances in deepwater search and salvage have put within reach shipwrecks once thought to be inaccessible. The well-publicized search for and location of the Titanic is one example. The most important step in successfully locating a wreck is defining the search box, the targeted area inside which the wreck is most likely to be found. Establishment of a search box calls for the pooling and reconciliation of information from multiple sources.

Use of a GIS to plan wreck searches helps to organize and make use of the complicated sets of data and, ultimately, to define the search box. Once defined, the characteristics of the bottom, including depth, type and consistency of sediment, and other area features, such as canyons or known debris, help the sidescan sonar operators plan specific search parameters such as two-direction, tow fish altitude, and potential for false targets.

The World War II Setting

As documented by Rohwer and Hummelchen (1992), in the fall of 1942, British destroyers supported by a Welsley aircraft forced the German submarine U-559 to the surface in the eastern Mediterranean Sea. British troops successfully boarded the submarine and managed to seize equipment, codebooks, and other documents vital to the decoding of German transmissions, but two men were lost when the submarine's captain scuttled his vessel. The important M-4 deciphering machine also sank with the submarine.

The search area for the U-559, superimposed on Veridian's global ocean sediment and global bathymetry databases.

Decades later, analyses of surface currents, sea state, and prevailing winds at the time of the boarding indicated a more northerly search box was desirable than the U-559 wreck location as presumed from the British destroyers' location and the location as recorded in Veridian's Global Maritime Wrecks Database of more than 250,000 wrecks. The following is a summary of the GIS methodology used by Veridian to organize and present data pertinent to the establishment of an exact search box. The study is of a hypothetical nature, similar to actual proprietary studies that are currently being performed by Veridian for its customers today.

GIS Methodology

Veridian, an Esri Business Partner, recognized more than 15 years ago the savings and power that these GIS tools could bring to the capture, manipulation, and analysis of data sets, large and small. In particular, Veridian chose the ArcInfo software to build and maintain a number of global environmental and cultural databases at multiple scales and, today, is continuing to migrate this data to ArcGIS and the geodatabase for use with ArcIMS, ArcView, and ArcInfo. The marine databases in this GIS support deepwater search and salvage studies (e.g., Cimino, Pruett, and Palmer, 2000; Palmer and Pruett, 2000). These databases can generally be used as is but are often updated with higher resolution site-specific data for a particular study area.

The first step in the analysis is to determine the general area in which the lost vessel (in this case the U-559) is likely located, and whether or not this is in an area claimed by a country. Using Veridian's Global Maritime Boundaries Database (GMBD), the last known position of the lost ship is compared with the Territorial Sea, the Contiguous Zone, and the Exclusive Economic Zone claims of the Mediterranean Sea littoral nations. The GMBD was built using the ArcInfo Regions model to portray overlapping polygon claims in the GIS database. If the shipwreck is believed to lie within such an area, a permit is required from the claiming country for a search to begin. It was fortunate in this case that only Egypt had claims in the vicinity and the wreck site was outside of the claimed region. A permit for shipwreck search was therefore not required.

Technical details of the U-559, commissioned 27 February, 1941. German U-boat type, VIIC built by Blohm & Voss, Hamburg, Germany between 1939 and 1942 (used with permission courtesy www.uboat.net).

Additional data retrieved from Veridian's global GIS archives was then used to plan the search effort. The search area was defined based on the analysis of this data with two parallel goals: to allow the most efficient search methodology to be employed and to minimize the impact on the ocean bottom during the search.

This additional GIS data also provides a historical look at the weather that can be expected in the search area, for example, showing a graphical representation of conditions for the period during which the hypothetical at-sea portion of the search of the U-559 took place. Time spent at sea for such a deepwater search is expensive (sometimes exceeding $40,000/day for the lease of a vessel, the navigation system, and a side-scan sonar for mapping the seafloor). A tool such as GIS that enables spatial review and analysis of all data can contribute to a successful and cost-effective search.

Many other data sets are used in planning an operation. In this instance, historical Notices to Mariners (NOTOM) chart correction and broadcast warnings from the National Imagery and Mapping Agency were reviewed to determine past activities in the vicinity of the proposed operation. This data can reveal whether previous surveys of the area have been performed and who performed them. These may turn out to be useful sources of new or higher resolution data. Further, and perhaps more important from a safety standpoint, Veridian's historical archive of NOTOM can indicate whether this area is presently being, or in the past has been, used for military exercises or as an ocean dumping ground. Either usage could affect the safety of personnel and the viability of the operation.

Another use of GIS is to create a chart in which the positional data of the tow fish is entered into the GIS database to ensure that the sidescan sonar vehicle has fully covered the search grid. This is to ensure that the swaths of the vehicle have not missed the area where the submarine might actually be located.

Acoustic Mapping

In addition to making possible the location of sunken vessels such as the U-559, recent developments in high-resolution acoustic sonar systems make possible an integrated systems approach to mapping seafloor features that can help researchers understand the geological setting in which a shipwreck or artifact is located. These applications include mapping mobile sediments that may cover a wreck during investigation and salvage and identifying the slump potential of regional sediment. This information can also aid sidescan sonar operators in planning specific search parameters such as two-direction, tow fish altitude, and potential for false targets. For Veridian in collaboration with Thales GeoSolutions (Pacific), the process often begins with high-frequency sidescan sonar imagery data collected and processed using TRITON ISIS software and mapped to support mosaic generation of the seafloor morphology. Mosaics are generated using DELPH MAP software after all sonar data has been collected and integrated with differential GPS data (which tracks the surface vessel's position and, by extension, the position of the sidescan sonar vehicle). The mosaic functions as the basemap for determining areas of interest and high importance.

A 100-kHz acoustic image showing high relief and detailed rock texture. Dark reflections likely result from steep slope or coarse-grained sediment (cobbles and pebbles).

Designated areas are then mapped with a high-frequency multibeam echo sounder to ensure the highest possible resolution and smallest individual beam-pattern footprint. Multiple swaths are collected across the feature to provide 100 percent coverage and a minimum of 10 percent overlap. Using a remotely operated vehicle to deploy the multibeam echo sounder, this method can be used in water as deep as 2,500 meters. The multibeam data is collected using WinFrog Multibeam software, then cleaned, processed, and displayed with hydrographic information processing software, applying sun illumination results in shaded relief images and allowing output in CAD format. The data sets can be overlaid in various formats for analysis and GIS support. The mapped regions indicate where future sediment slumps will likely occur and where the seafloor is currently mobile.

Conclusion

Ownership of, and rights to recover, antiquities and objects of value from the seafloor remain controversial issues. Knowledge of the extent of maritime boundaries, a coastal state's jurisdiction over waters, and the character of the seafloor are essential considerations for those seeking to legally salvage such objects. A GIS can greatly assist in the planning of an at-sea operation, reducing costs, avoiding hazardous consequences, and ensuring that an area has been thoroughly covered. Esri's GIS software suite, in concert with combined geophysical and geological methodologies, can constitute an integrated systems approach to mapping seafloor features, providing an efficient and economical way to image the shallow seafloor, and producing data that can be used to address many problems associated with deepwater search, salvage, and marine archaeology.

For more information, contact Lorin Pruett, manager, Global Maritime Boundaries Database Project, Veridian (e-mail: Lorin.Pruett@veridian.com), or Dawn Wright, associate professor, Department of Geosciences, Oregon State University (e-mail: dawn@dusk.geo.orst.edu).

This article is derived from a chapter in the Esri Press book Undersea with GIS (ISBN: 1-58948-016-3) edited by marine geographer Dawn Wright of Oregon State University. Esri Press books are available at better bookstores, online (www.esri.com/esripress), or by calling 1-800-447-9778. Outside the United States, please contact your local Esri distributor.