This report documents the data collection, post-processing, and uncertainty estimation methods for bathymetric surveys conducted during July–September 2024 on Council Grove Lake and Marion Reservoir, Kans., and Pine Creek Lake, Okla. This report also presents bathymetric contour maps and elevation-area-capacity tables for those lakes calculated from the processed bathymetric survey data and supporting data in an associated data release (Smith and others, 2026).

Council Grove dam, which is approximately 1.5 miles north of Council Grove, Kans. (fig. 1B), was completed in 1964 on the Neosho River (USACE, 2025a) to form Council Grove Lake. The lake is used for flood control, water supply, water-quality control, fish and wildlife, and recreation and has a conservation pool elevation of 1,274.0 feet (ft) above the National Geodetic Vertical Datum of 1929 (NGVD 29) and flood-control pool elevation of 1,289.0 ft above NGVD 29 (USACE, 2024a). For this report, which references elevations to the North American Vertical Datum of 1988 (NAVD 88), an offset of 0.5 ft was added to convert NGVD 29 elevations to NAVD 88 elevations at Council Grove Lake (National Oceanic and Atmospheric Administration [NOAA], 2025).

Marion dam, which is approximately 3.5 miles northwest of Marion, Kans. (fig. 1C), was completed in 1968 on the Cottonwood River (USACE, 2025b) to form Marion Reservoir. The reservoir is used for flood control, water supply, water-quality control, and recreation and has a conservation pool elevation of 1,350.5 ft above NGVD 29 and flood-control pool elevation of 1,358.5 ft above NGVD 29 (USACE, 2024b). For this report, an offset of 0.5 ft was added to convert NGVD 29 elevations to NAVD 88 elevations at Marion Reservoir (NOAA, 2025).

Pine Creek dam, which is approximately 5.0 miles northwest of Wright City, Okla. (fig. 1D), was completed in 1969 on the Little River (USACE, 2025c) to form Pine Creek Lake. The lake is used for flood control, water supply, water-quality control, recreation, and fish and wildlife and has a conservation pool elevation of 438.0 ft above NGVD 29 and flood-control pool elevation of 480.0 ft above NGVD 29 (USACE, 2024c). For this report, an offset of 0.1 ft was added to convert NGVD 29 elevations to NAVD 88 elevations at Pine Creek Lake (NOAA, 2025).

Bathymetric surveys of Council Grove Lake, Marion Reservoir, and Pine Creek Lake were conducted from July 8 to September 20, 2024, when the lakes were above their respective conservation pool elevations. The surveying vessel was a 22-ft, flat-bottom, aluminum-hull boat with an attached outboard motor. The vessel was equipped with two 12-volt, 100-ampere-hour (360,000-coulomb) marine deep-cycle batteries wired in parallel to provide power to a multibeam mapping system (MBMS) consisting of three components: (1) a multibeam echosounder (MBES), (2) an inertial navigation system (INS), and (3) a laptop computer for data collection and processing. The MBES was a 400-kilohertz NORBIT iWBMSh (NORBIT, 2014) mounted on the port side of the survey vessel. The MBES had a curved array that allowed data collection within a 210-degree beam angle (NORBIT, 2014). The beam angle was limited to 140 degrees (70 degrees each side of nadir, or directly below the MBES) to reduce noise in the collected data. The MBES array can be tilted to more accurately survey sloped banks. The tilted array was used during surveying to collect data along banks except where heavy vegetation limited the advantage of this array configuration.

The INS used for the bathymetric surveys was a Trimble Applanix OceanMaster (Trimble Inc., 2023). The INS positions the MBES in three-dimensional space based on the measured heave, pitch, roll, and heading of the survey vessel. The location of the vessel was determined by using a global navigation satellite system (GNSS) with two Trimble AT1675-540TS GPS antennas mounted to a NORBIT PORTUS Pole (a rigid carbon-fiber pole resistant to thermal expansion or contraction; NORBIT, 2021). During each day of surveying, a third GPS unit, the Trimble R12i GNSS (Trimble Inc., 2025), was used to collect GPS base data at a stable position on land. The GPS base was located on the dam intake structures at Council Grove and Pine Creek Lakes and on a concrete slab near a boat ramp at Marion Reservoir. The GPS base data were processed through the NOAA (2024) Online Positioning User Service to tie the local surveys to the National Spatial Reference System, thereby improving the raw GPS results collected by the base station. The GPS collected data were referenced to NAVD 88.

All motion-sensing and navigation data were processed through the Trimble Applanix POSPac Mobile Mapping Suite (Trimble Inc., 2024). This software suite is a post-processing software package designed to import and process position data. The software utilizes tools to identify and compensate for sensor and environmental uncertainties and compute an optimally accurate blended navigation solution. After post-processing, the INS and the GNSS data were blended to produce a smoothed best estimate of trajectory (SBET) file. The SBET file was used as the navigation data from each day to provide the best possible trajectory of the vessel. The file was then applied to the bathymetric data to georeference the survey data.

The MBES measures water depth using ultrasonic sound waves; however, the sound wave velocity is affected by changes in salinity and temperature through the water column (Wilson and Richards, 2006). The MBES has an integrated sound velocity probe that continuously measures sound velocity at the sonar head (lake surface), but sound velocity profiles (SVPs) of the water column were periodically collected by using an AML Oceanographic Base X2 sound velocity profiler (AML Oceanographic Ltd., 2024). A minimum of two SVPs were collected each day of the survey and covered the maximum survey depth for that day. The downloaded SVPs were analyzed with HYPACK/HYSWEEP software (Xylem, Inc., 2024) to verify that the profiles were correctly georeferenced by checking them against the created target in the software at the time and location that each SVP was collected. The SVP data were visually inspected for spikes or anomalies, which were subsequently removed. The SVPs were then incorporated into the bathymetric dataset during processing in the software as water-depth corrections. For some locations, slight temperature adjustments were made to the SVPs to correct mismatches of overlapping MBES water-depth data (Bian and others, 2018).

Water-depth data, SVP corrections, and SBET files were combined in HYPACK/HYSWEEP software to create bathymetric surface elevation points. Three filters were used to assist in removing anomalous elevation point values in the software. These filters included (1) a minimum or maximum elevation filter to remove any elevation point values above the water surface or below the lowest portions of the survey, (2) a matrix filter with a four-sigma limit to remove points with elevation values greater than four standard deviations from the cell statistic, and (3) an over or under filter to remove any elevation points that either overhang or undercut the previous or next beam. After the initial filtering, all data were visually inspected and any anomalous values that did not match the general trend from neighboring data points were removed.

A bathymetric surface was developed for each lake from the corrected and filtered elevation point values by using the triangulated irregular network algorithm method with the Create Elevation Grid from 3D Vector Data tool in Global Mapper version 26.0.0 (Blue Marble Geographics, 2025) to generate digital terrain models (DTM). Light detection and ranging (lidar) point-cloud data classified as ground points (USGS, 2018, 2022a, b; downloaded tiles listed in Smith and others [2026]) were used to extend survey coverage to inundated, shallow, or debris-obstructed areas that could not be surveyed with the MBES. To reduce noise and improve the appearance of contours, the lidar point-cloud data were gridded to a 2-meter cell size by using the median elevation within the grid cell before input to the triangulated irregular network algorithm.

Linear enforcement was implemented by using methods described by Wilson and Richards (2006) to define bathymetric elevations in areas with gaps between the lidar data and the MBES data. Linear enforcement is the use of breaklines to force linear interpolation of elevations in the DTM to ensure linearity of valleys, ridges, channels, banks, and steep slopes (Wilson and Richards, 2006). USGS satellite imagery (USGS, 2009) was used in determining linear enforcement of portions of the shoreline where data gaps occurred in some shallow pools because the maximum depth was too shallow for the MBES survey or in areas obstructed by debris. The process of linear enforcement consisted of generating line features in the gap areas and interpolating elevations linearly at points along the line features based on the lidar data and MBES data. The DTM was then created from the interpolated elevations along the line features, the lidar data, and the MBES data. Bathymetric contours were then generated from the bathymetric surface DTM; the contours were generated at 2-ft intervals for Council Grove Lake and Marion Reservoir and at 10-ft intervals for Pine Creek Lake (Smith and others, 2026). Contours less than 1,000 ft in length were deleted to simplify the bathymetric maps. Some vertices of the Pine Creek Lake flood-control pool contour (480.1-ft above NAVD 88) were adjusted manually to prevent vector line intersections with the 480.0-ft (above NAVD 88) contour. Contours were smoothed for the bathymetry maps (shown later) utilizing the ArcGIS Pro 3.4.3 Smooth Line tool with the Polynomial Approximation with Exponential Kernel (PAEK) method using a smoothing tolerance of 30 meters and resolving topological errors with an input barrier of the top-most contour (Esri, 2025). Contours for the Council Grove Lake and Marion Reservoir bathymetry maps were filtered to 4-ft intervals.

Elevation-area-capacity tables indexed at 1-ft intervals were generated from the bathymetric surface DTM. The DTM elevation values (one per raster cell) were filtered to remove no-data values and sorted in descending order. A column of stage values was added by subtracting the DTM minimum elevation value from the sorted elevation values. Surface area at each indexed water-surface elevation was calculated as the count of all lesser sorted elevation values multiplied by the uniform DTM cell area. The storage capacity at each indexed water-surface elevation was calculated as the sum of all stage values associated with lesser sorted elevation values multiplied by the uniform DTM cell area.

The accuracy of depth readings obtained by the MBES outer beams are prone to greater error than readings obtained by the inner more nadir beams because of longer travel distances (Richards and others, 2019). A beam-angle test compares multibeam check lines to a reference surface and estimates the depth accuracy of the multibeam system at different angle limits (Xylem, Inc., 2024). The beam-angle test was performed by using USACE (2013) methods to verify the beams of the MBES were collecting MBES data at sufficient accuracy for the beam angles (up to 140 degrees) used in these lake bathymetric surveys. The beam-angle test results were within the recommended performance standards described by the USACE (2013).

Patch tests were conducted near the beginning and end of each lake bathymetric survey by using methods described in Xylem, Inc. (2024). Patch tests are dynamic calibration checks that determine systematic errors propagated by the mounting angle, timing, and position of the MBES in relation to the INS and real-world coordinates (Huizinga, 2017). The offsets estimated by the patch tests were used to correct for systematic errors in latency, roll, pitch, and yaw. The offsets used for each lake were the results of the patch tests performed on that lake; however, each lake’s patch tests resulted in similar offsets, indicating the consistency of the systematic errors throughout the project. Proper calibration will yield consistent bathymetric results under various vessel and MBES orientations (Huizinga, 2017).

Uncertainties in bathymetric elevation data were estimated by computing the total propagated uncertainty (TPU; sheet 1, figs. 2–4) as described in Huizinga (2017). The TPU was calculated for each grid cell in the surveyed area for each lake by using the Combined Uncertainty and Bathymetry Estimator (CUBE) method (Calder and Mayer, 2003; Richards and Huizinga, 2018) in the HYPACK software (Xylem, Inc., 2024). The CUBE method combines the uncertainty from all MBMS system-component and resolution effects to estimate the TPU.

The highest TPU uncertainties were located near high-relief features on the lakebed or along the outermost beam extents of the MBES swath, which typically yield the noisiest data (figs. 2–4). Some minor gaps in the data occurred where a segment of data collection was completed and the boat moved before the next segment was initiated. Other minor gaps in the data occurred because of swaths not overlapping. TPU statistics for each lake are described in table 1. The MBES bathymetric elevations estimated in this report were based on the CUBE “depth” as described in the HYPACK software (Xylem, Inc., 2024). The CUBE depth is the most probable elevation of an MBES survey point and is calculated using the TPU estimates (Calder and Mayer, 2003; Xylem, Inc., 2024). For this report, CUBE depths were reported as elevations above NAVD 88.