A digital elevation model (DEM) is an essential component of any hydrodynamic model. The Delta Modeling Section (Section) has maintained a database of bathymetry soundings and levee surveys for decades and published a 10-meter (10m) DEM for the San Francisco Bay and Sacramento-San Joaquin Delta (Delta) (California Department of Water Resources 2012). In collaboration with the U.S. Geological Survey (USGS) Pacific Coastal and Marine Science Center, the California Department of Water Resources (DWR) has continued to upgrade these DEMs based on newer survey data and improved interpolation methodologies. An updated San Francisco Bay-Delta bathymetric/topographic digital elevation model was published by the USGS (Fregoso, Wang, Ateljevich, and Jaffe 2017).
Both DWR and USGS continue to work on the elevation models for several reasons. First, high-resolution multibeam bathymetry data continues to become available. A good portion of the newer collections are performed at locations where bathymetry data is lacking, or of poor quality, or where model sensitivity to bathymetry is known to be high, so the effort has a high return on investment. Recent high-resolution multibeam datasets easily support the development of accurate 2-meter (2m) DEMs, although shallow water, turbidity, vegetation, and the gap between the multibeam data and terrestrial data from light detection and ranging (LiDAR) remain vexing issues that fuel development of enhanced techniques.
In some locations, migration toward 2m resolution models is motivated by geographical structure even where there has been little improvement in the underlying data. DEMs at 10m resolution are insufficient to adequately describe small-scale terrain features, such as levee crests or the main conveyance channel through a narrow reach, such as Middle River. As a result, elevation modelers have traditionally needed to perform feature enforcement for 10m DEMs, but not for the Section’s 2m DEMs which do resolve these features (California Department of Resources 2012). There are several other problems associated with a coarse target resolution, including reduced conveyance and rough contours where channels run oblique to the elevation model. In places where the topography demands a finer elevation model, the Section and others have adopted improved interpolation techniques for single-beam data to produce reasonable 2m DEMs that preserve features on appropriate scales.