Methodology

The data were collected using Real-Time Kinematic (RTK) global positioning system (GPS) technology. Dual frequency Trimble® 5700 and R8 GPS receivers and Trimble Survey Controller® (TSC) and Trimble® GPS antennas were used for all the GPS work. All horizontal coordinates are in meters in the Washington State Plane South coordinate system and datum NAD 83. Geoid model G03US was used to convert GPS-derived heights-above-ellipsoid to orthometric heights or elevations above the NAVD 88 datum.

Several established GPS control points and two points established during the last USGS river survey in the area were used (table 2). Two base stations, one for the Selah Gap survey (fig. 2) and one for the Union Gap survey, were set up during the survey to broadcast radio signals of real-time GPS corrections to the roving GPS receivers. Horizontal and vertical data from the GPS rovers were compared to established control points to check the accuracy of the rovers.

The intent of the survey was to provide sufficient topographic information of the channel bottom wetted during average flow levels and as much of the streambanks (top-of-bank and toe-of-bank elevations) as possible. The shallow river depths made the survey difficult because the navigable area sometimes was limited to only part of the width of the channel. Most of the Naches River reach was too shallow for navigation; therefore, walking surveys were made. Both types of surveys, boat and walking surveys, used roving GPS units that received satellite GPS signals and radio RTK GPS correction signals from the base station. For the boat survey, the roving GPS unit was mounted directly above an echo sounder that was mounted to a 4.9 m (16 ft) jet boat. For the walking survey, a GPS receiver was mounted on a 2-m fiberglass rod and data were collected with the Trimble Survey Controller® (fig. 3).

Boat Surveys

The elevation of the streambed was determined from RTK GPS-derived elevation minus the antenna-to-echo-sounder length minus the depth recorded by the echo sounder (fig. 4). The GPS-derived elevation is the orthometric height computed by the Trimble software by subtracting the geoid height from the ellipsoid height. The echo sounder was an Innerspace® model 455 with an 8 degree signal angle. Adjusting the speed of sound to the water temperature yielded minimal precision gains because the depths sounded were shallow. As a result, the speed of sound was set to 4,800 ft/s for all the surveys and the speed of sound in water was set at a temperature of 57.5°F. The echo sounder resolves soundings to the nearest 0.1 ft or 0.1 m, but the manufacturer does not provide precision information. There was no compensation for roll and pitch of the boat. Horizontal and vertical positions of the echo sounder were computed every 1 second by the RTK GPS when adequate satellite and radio signals were received by the rover. The echo sounder gathered stream depths every 200 milliseconds. The echo sounder and GPS data were fed to an onboard laptop computer and the data were managed with HyPak 2008®, a PC-based Windows navigational and bathymetric mapping software. HyPak determined the location between successive GPS points by projecting the path between the points.

The general survey pattern used to collect bathymetry data from the boat was a zigzag path across the channel followed by five parallel, longitudinal surveys along the channel (fig. 5). Because of the shallow waters, vegetation along the banks, debris, and other obstacles, the survey pattern was modified at times. Data often were gathered at the only navigable part of the reach. (For example, notice the excursion of the survey around the shallow bar in the center of the map in figure 5 and the partial survey of a small side channel just northwest of the shallow bar.) The result was a high density of points along survey lines (about two or more soundings per meter depending on boat speed) with relatively large spaces not surveyed between the lines. Spaces of about 15 m between survey lines were common; the distances sometimes were greater around obstacles and shallow waters.

Walking Surveys

Walking surveys (figs. 6 and 7) were made on the banks and in wadable parts of the rivers by collecting elevation points using a roving GPS receiver mounted on a 2-m pole (fig. 3). The walking surveys provided better definition of the river banks than the definition provided by the LIDAR surveys. The walking surveys also augmented the boat surveys by providing data in the shallow areas that were not navigable or, in the case of the Union Gap survey, in areas with macrophytes that inhibited the sounding of depths by the boat-mounted echo sounder. The controller mounted on the 2-m pole was used to initiate a GPS reading, assess the GPS satellite coverage, and log the readings. When good satellite coverage was available, a reading took about 5 seconds to acquire. Each reading was the average of five readings. Points often were taken in pairs, one point at the top of the bank and one point at the toe of the bank. Attributes were attached to each point as a feature code: in the river (STR), at the left edge (LEW) or right edge (REW) of the water surface at the time of the survey (left and right are relative to looking downstream), at the top of a topographic bank (TOP), at the toe of a bank (TOE), or at random topographic elevation points (GRN). Excessive tree coverage, highway overpasses, and steep canyon walls prevented readings at times. Thick brush or steep banks also limited access to most of the river banks, so readings usually were taken only in locations with access to the river bank or in shallow areas. Most of the lower Naches River was not accessible by boat so extensive walking surveys were made along lines perpendicular to the direction of flow (fig. 6). The walking-survey points are dense in an area on the inside of the river bend shown in the lower left side of figure 7. Thick macrophyte growth along that part of the river bank limited proper sounding readings from the boat.

Post-Processing of Survey Data

Erroneous boat-survey points were deleted during post-processing of the GPS data using the HyPak 2008® single-beam editor. All soundings made by the boat surveys were visually inspected by viewing one trackline at a time. A trackline is the course that the boat takes once the sounding is initiated until the sounding is terminated. There were 125 tracklines recorded over the 3-day period of this survey. Little or no deleting of points was required for many tracklines; however, extensive editing was required for some tracklines. The criteria used to delete points were somewhat subjective. Extraneous points that were deleted included echoes that reflected off objects floating above the channel bottom and the tops of rooted vegetation, that penetrated the soft substrate below the bottom, or that reflected off multiple surfaces. A typical trackline of data before and after editing is shown in figure 8A and B. The deleted points were those immediately above the general trend of the channel bottom. If it was not clear whether the points were erroneous or true, the points were not deleted. A trackline of data is shown in figure 9Aand B before and after the boat likely tracked over heavy macrophyte growth that reflected the signal from the upper surface of the plant. After editing the tracklines, the remaining points were exported into an Excel file and GIS coverages.