Recent advances in processing multibeam sonar data brought about by the Combined Uncertainty and Bathymetric Estimator (CUBE) [1] have demonstrated the value of identifying and tracking survey uncertainties. Most of these uncertainties were outlined in Hare, Godin, and Mayer uncertainty model developed in 1995 [2]. That report identified the uncertainties in the various electronic systems used to acquire the bathymetric data. However, one of the largest contributors to the overall error budge t in a near coastal hydrographic survey is that contributed by water level uncertainty. As the ocean mapping industry pushes for ever finer spatial details in its data, the traditional method of discrete tide zoning [3] must be abandoned for a more robust method that can match the requirements of the data. The method currently under investigation by the National Oceanic and Atmospheric Administration is the Tidal Constituent And Residual Interpolation (TCARI) method [4]. TCARI has the ability to interpolate the water level at a vessel’s position for any location and instance in time. It can also produce a gridded water level surface of the entire survey area. While the potential of this method is encouraging, a rigorous investigation of the uncertainties associated with it has yet to be completed. This research seeks to close that gap by examining the uncertainties in this method, using both observed water level information from around the country as well as data acquired during the original 1995 NOS Kinematic GPS experiment in Galveston Bay, Texas [5].


Center for Coastal and Ocean Mapping

Publication Date


Journal or Conference Title

US Hydrographic Conference

Conference Date

Mar 29 - Mar 31, 2005

Publisher Place

San Diego, California, USA


Hydrographic Society of America

Document Type

Conference Proceeding