Bathymetry and Acoustic Backscatter—Elwha River Delta, Washington

Survey Results

Elwha River Delta

The Elwha River Delta survey area consisted of 192 survey lines totaling more than 349 linear kilometers of data. Depths ranged from 0 m to 25 m relative to NAVD88 (fig. 5). The normalized acoustic backscatter shows that most of the tidal flat in front of the delta presents a relatively hard acoustic signature, with the only significant area of soft acoustic reflections found along the southwest shoreline (fig. 6).

After filtering the data to remove obvious outliers, the standard deviation of the remaining sounding elevations was calculated for each 1 m x 1 m cell (each containing 28 soundings on average) in CARIS. The mean standard deviation for all cells in the Elwha survey is 0.13 m, and 95 percent of the cells in the data set have a standard deviation of less than 0.25 m (fig. 7). There are a few outlier cells with variations as high as 1.61 m.

A map of the survey area colored by standard deviation (fig. 8) indicates that most areas with high TVU are found on the northeast side of the survey area, where the survey vessel was exposed to larger northeasterly wind waves. Higher bathymetric uncertainty tends to concentrate in the nadir region of individual tracklines—a common artifact in SWATHplus interferometric data that is difficult to mitigate.

Vertical uncertainty increases with depth, and IHO standards account for this in their specifications of survey quality. A comparison of the estimated bathymetric uncertainty (TVU) in this survey against IHO standards can be seen in figure 9. Most of the elevation uncertainty in the dataset satisfies the IHO standard for first-order surveys; however, there are outlier cells that exceed both first- and even second-order standard thresholds. The TVU map can be used to generate a map of the IHO survey order each cell in the DEM satisfies, as shown in figure 10.

Rayonier Pier

On February 26, 2010, weather conditions were too rough for surveying outside Port Angeles harbor and two alternate locations within the harbor were surveyed insteadz; the Rayonier Pier site and the south side (inside) of Ediz Hook.

The Rayonier Pier location was near the range limit of the GPS base station to the west (fig. 1) and a few lines were compromised due to loss of RTK lock. Time constraints also limited the number of passes the boat was able to make over the study area. The combination of poor weather, the lower quality navigation, and the relatively large line spacing resulted in higher-than-normal artifacts in the dataset, including an along-track "rail" along the nadir of each trackline and scalloping artifacts on the eastern side of the pier due to uncompensated pitch and roll, presumably a result of the choppy conditions. In spite of these problems, the data showed remarkable details of hundreds of sunken logs (particularly in the backscatter), submerged pipelines, and evidence of dredging alongside of the pier (fig. 11 and fig. 12).

A histogram of the standard deviations of elevation values in each 1 m x 1 m cell in the raster is shown in fig. 13. Approximately 95 percent of the data had standard deviations < 25 cm. When these values are converted to TVU according to equation 2, as shown in figure 14 and figure 15, most of the dataset meets IHO special-order specifications.

Ediz Hook (Inside Harbor)

The Ediz Hook survey area inside the harbor was formerly the site of a semipermanent log raft. Hundreds of logs sank over the years, and many of these are visible in the sonar bathymetry and backscatter (fig. 16 and fig. 17).

A histogram of the standard deviations of elevation values in each 1 m x 1 m cell in the raster is shown in figure 18. Approximately 95 percent of the data had standard deviations less than 50 cm. When these values are converted to TVU according to equation 2, as shown in figure 19, most of the dataset meets IHO first-order specifications. This uncertainty is higher than normally observed with the SWATHplus system in these water depths and may be attributed to the higher-than-normal seafloor roughness (due to the sunken logs) observed in the area. Some of the individual logs are visible in a planview map of TVU values shown in figure 20.

Ediz Hook (Outside Harbor)

In 2005, a USGS Bathymetric survey of the north side of Ediz Hook (Cruise ID: K-1-05-PS) revealed several slide features on the spit. As time allowed during transits to and from the primary survey area on the Elwha River Delta, 4 survey lines were collected to inspect what, if any, changes had occurred along the north side of the spit during the intervening 5 years. These survey lines were reconnaissance in nature. The bathymetry and backscatter are shown in figure 21 and figure 22, respectively.

A histogram of the standard deviations of elevation values in each 1 m x 1 m cell in the raster is shown in figure 23. Approximately 95 percent of the data had standard deviations less than 50 cm. When these values are converted to TVU according to equation 2, as shown in figure 24, most of the dataset meets IHO Second order specifications. Towards the edges of the swaths the data quality deteriorates as shown in figure 25.

Top of Page

For more information, contact the PCMSC team.