Bathymetric and Sediment Survey of Kanopolis Reservoir, Ellsworth County, Kansas

Transcription

1 Bathymetric and Sediment Survey of Kanopolis Reservoir, Ellsworth County, Kansas Kansas Biological Survey Applied Science and Technology for Reservoir Assessment (ASTRA) Program Report (April 2009)

2 This work was funded by the Kansas Water Office through the State Water Plan Fund in support of the Reservoir Sustainability Initiative.

3 SUMMARY During October 2007, the Kansas Biological Survey (KBS) performed a bathymetric survey of Kanopolis Reservoir in Ellsworth County, Kansas. The survey was carried out using acoustic echosounding apparatus linked to a global positioning system. The bathymetric survey was georeferenced to both horizontal and vertical reference datums, allowing the 2007 lake depth data to be compared to a 1939 US Army Corps of Engineers pre-impoundment topographic map for an estimate of sediment accumulation. Comparison of the 2007 bathymetric survey data to a 1939 pre-impoundment map suggests that the capacity of the reservoir at the 1063 elevation pool has been reduced from 57,625 acre-feet to 48,784 acre-feet. Ten sediment cores were extracted from the lake to determine accumulated sediment thickness at locations distributed across the reservoir. Sediment samples were taken from the top six inches of each core and analyzed for particle size distributions. Additional sediment samples were taken in April 2009 and also analyzed for particle size distributions. Summary Data: Bathymetric Survey: Dates of survey: Water elevation on dates of survey: October 19, 2007 October 20, ft ft. Reservoir Statistics: Elevation at conservation pool 1463 ft. Area at conservation pool (acres): 3011 Volume at conservation pool (acre-ft): 48,784 Maximum depth at conservation pool: 30.6 ft. Year constructed: Sediment Survey: Date of sediment survey: October 29, 2008

4 TABLE OF CONTENTS SUMMARY...i TABLE OF CONTENTS...ii LIST OF FIGURES...iii LIST OF TABLES...iv LAKE HISTORY AND PERTINENT INFORMATION... 1 BATHYMETRIC SURVEYING PROCEDURE Pre-survey preparation:... 3 Survey procedures:... 3 Establishment of lake level on survey date:... 4 Post-processing... 6 BATHYMETRIC SURVEY RESULTS Area-volume-elevation tables... 9 PRE-IMPOUNDMENT MAP Pre-impoundment area-volume-elevation tables Comparison of pre-impoundment and 2007 area-elevation curves SEDIMENT CORING AND SAMPLING Sediment coring and sampling results ii

5 LIST OF FIGURES Figure 1. Kanopolis Lake Figure 2. Location of Kanopolis Lake in Ellsworth County, Kansas Figure 3. Bathymetric survey transects... 5 Figure 4. Reservoir depth map... 8 Figure 5. Cumulative area-elevation curve Figure 6. Cumulative volume-elevation curve Figure 7. Pre-impoundment contour lines map Figure 8. Digitized pre-impoundment contour lines map Figure 9. Pre-impoundment digital elevation model Figure 10. Pre-impoundment cumulative area-elevation curve Figure 11. Pre-impoundment cumulative volume-elevation curve Figure 12. Changes in lake-bottom elevation, Figure 13. Comparison of 2007 and pre-impoundment cumulative areaelevation curves Figure 14. Comparison of 2007 and pre-impoundment cumulative volumeelevation curves Figure 15. Sediment coring sites Figure 16. Map of sediment thickness in centimeters at coring sites Figure 17. Sediment particle size analysis Figure 18. Particle size distributions at sampling and coring sites iii

6 LIST OF TABLES Table 1. Cumulative area in acres by tenth foot elevation increments... 9 Table 2. Cumulative volume in acre-feet by tenth foot elevation increments...10 Table 3. Table 4. Pre-impoundment cumulative area and volume by one-foot elevation increments...16 Kanopolis Reservoir sediment coring/sampling site data...24 iv

7 LAKE HISTORY AND PERTINENT INFORMATION (This section summarized from National Dam Inventory data, and Kansas Water Office fact sheet descriptions) Kanopolis Lake was authorized by the Flood Control Act of Construction of the project began on 8 June Diversion of the river through the tunnel was made on 29 July Storage of water in the lake began on 17 February Kanopolis Lake is a multi-purpose project with storage allocated for flood control, water supply, water quality and recreation. Kanopolis Lake was constructed and is operated by the U.S. Army Corps of Engineers. The reservoir is located on the Smoky Hill River approximately 31 miles southwest of Salina, Kansas in Ellsworth County. It is one of the oldest lakes in Kansas. The Smoky Hill River is the major source of water flowing into Kanopolis Lake. The Smoky Hill river basin runs generally from west to east. Kanopolis Lake has 41 miles of Figure 1. Kanopolis Dam under construction, 1940s. shoreline at the top of multipurpose pool elevation. At flood control pool elevation, the lake has 135 miles of shoreline and extends generally westward. The main embankment is approximately 15,360 feet long, including a 4,070-foot long left abutment and a 2,550-foot long right abutment dike sections. The maximum height above the streambed is 131 feet. The top of dam is at elevation , which includes a freeboard allowance of 5.2 feet above the maximum spillway design flood. The rolled fill dam embankment consists of impervious, pervious, random, berm fill and blanket fill zones. The upstream face of the dam is protected by an 18-inch thick layer of riprap overlying a 6-inch layer of spalls overlying a 9-inch layer of sand and gravel. The slope protection materials were placed in The downstream face of the dam is protected with native grass cover. 1

8 Ellsworth County, Kansas Wilson Ellsworth Kanopolis Holyrood Lorraine - Miles Figure 2. Location of Kanopolis Reservoir in Ellsworth County, Kansas 2

9 Reservoir Bathymetric (Depth) Surveying Procedures KBS operates a Biosonics DT-X acoustic echosounding system ( with a 200 khz split-beam transducer and a 38-kHz singlebeam transducer. In addition to providing basic information on reservoir depth profiles, the Biosonics system also permits the assessment of bottom sediment composition. Latitude-longitude information is provided by a JRC global positioning system (GPS) that interfaces with the Biosonics system. ESRI s ArcGIS is used for on-lake navigation and positioning, with GPS data feeds provided by the Biosonics unit through a serial cable. Power is provided to the echosounding unit, command/navigation computer, and auxiliary monitor by means of a inverter and battery backup device that in turn draw power from the 12-volt boat battery. Pre-survey preparation: Geospatial reference data: Prior to conducting the survey, geospatial data of the target lake is acquired, including georeferenced National Agricultural Imagery Project (NAIP) photography. The lake boundary is digitized as a polygon shapefile from the FSA NAIP georeferenced aerial photography obtained online from the Data Access and Service Center (DASC) at the Kansas Geological Survey. Prior to the lake survey, a series of transect lines are created as a shapefile in ArcGIS for guiding the boat during the survey. Transect lines are spaced more closely (25-50 meters separation) on smaller state/local lakes, while a spacing of meters is used for federal reservoirs. Survey procedures: Calibration (Temperature and ball check): After boat launch and initialization of the Biosonics system and command computer, system parameters are set in the Biosonics Visual Acquisition software. The temperature of the lake at 1-2 meters is taken with a research-grade metric electronic thermometer. This temperature, in degrees Celsius, is input to the Biosonics Visual Acquisition software to calculate the speed of sound in water at the given temperature at the given depth. Start range, end range, ping duration, and ping interval are also set at this time. A ball check is performed using a tungsten-carbide sphere supplied by Biosonics for this purpose. The ball is lowered to a known distance (1.0 meter) below the transducer faces. The position of the ball in the water column (distance from the transducer face to the ball) is clearly visible on the echogram. The echogram distance is compared to the known distance to assure that parameters are properly set and the system is operating correctly. On-lake survey procedures: Using the GPS Extension of ArcGIS, the GPS data feed from the GPS receiver via the Biosonics echosounder, and the pre-planned transect pattern, the location of the boat on the lake in real-time is shown on the command/navigation computer screen. To assist the boat operator in navigation, an auxiliary LCD monitor is connected to the computer and placed within the easy view of the boat operator. Transducer face depth on all dates is 0.5 meters below the water surface. The transect pattern is maintained except when modified by obstructions in the lake (e.g., partially submerged trees) or shallow water and mudflats. Data are automatically logged in new files every half-hour (approximately 9000-ping files) by the Biosonics system. 3

10 Establishment Of Lake Level On Survey Dates: Federal Reservoirs: Lake levels on the survey dates are obtained from the US Army Corps of Engineers web sites for those lakes: Reservoir Names Clinton Lake, Hillsdale Lake, Kanopolis Lake, Melvern Lake, Milford Lake, Perry Lake, Pomona Lake, Tuttle Creek Lake, Wilson Lake Corps of Engineers District Kansas City Website for Lake Level t/eightdayreservoirreport.cfm Big Hill Lake, Council Grove Reservoir, El Dorado Lake, Elk City Lake, Fall River Lake, John Redmond Reservoir, Marion Reservoir, Toronto Lake Tulsa Kanopolis Reservoir Water Surface Elevations: Survey Date Elevation (feet) Elevation (meters) October 19, October 20, Aerial Photography date Elevation (feet) Elevation (meters) April 1, Reservoir shoreline perimeters were digitized off aerial photography, and the elevation of the reservoir on the date of aerial photography was used as the water surface elevation in all productions of TINs or interpolations from point data to raster data. 4

11 Miles Date of Survey 10/19/ /20/2007 Figure 3. Bathymetric survey transects in Kanopolis Reservoir 5

12 Post-processing (Visual Bottom Typer) The Biosonics DT-X system produces data files in a proprietary DT4 file format containing acoustic and GPS data. To extract the bottom position from the acoustic data, each DT4 file is processed through the Biosonics Visual Bottom Typer (VBT) software. The processing algorithm is described as follows: The BioSonics, Inc. bottom tracker is an end_up" algorithm, in that it begins searching for the bottom echo portion of a ping from the last sample toward the first sample. The bottom tracker tracks the bottom echo by isolating the region(s) where the data exceeds a peak threshold for N consecutive samples, then drops below a surface threshold for M samples. Once a bottom echo has been identified, a bottom sampling window is used to find the next echo. The bottom echo is first isolated by user_defined threshold values that indicate (1) the lowest energy to include in the bottom echo (bottom detection threshold) and (2) the lowest energy to start looking for a bottom peak (peak threshold). The bottom detection threshold allows the user to filter out noise caused by a low data acquisition threshold. The peak threshold prevents the algorithm from identifying the small energy echoes (due to fish, sediment or plant life) as a bottom echo. (Biosonics Visual Bottom Typer User s Manual, Version 1.10, p. 70). Data is output as a comma-delimited (*.csv) text file. A set number of qualifying pings are averaged to produce a single report (for example, the output for ping 31 {when pings per report is 20} is the average of all values for pings 12-31). Standard analysis procedure for all 2008 and later data is to use the average of 7 pings to produce one output value. All raw *.csv files are merged into one master *.csv file using the shareware program File Append and Split Tool (FAST) by Boxer Software (Ver. 1.0, 2006). Post-processing (Excel) The master *.csv file created by the FAST utility is imported into Microsoft Excel. Excess header lines are deleted (each input CSV file has its own header), and the header file is edited to change the column headers #Ping to Ping and E1 to E11, characters that are not ingestable by ArcGIS. Entries with depth values of zero (0) are deleted, as are any entries with depth values less than the start range of the data acquisition parameter (typically 0.49 meters or less) (indicating areas where the water was too shallow to record a depth reading). In Excel, depth adjustments are made. A new field Adj_Depth is created. The value for AdjDepth is calculated as AdjDepth = Depth + (Transducer Face Depth), where the Transducer Face Depth represents the depth of the transducer face below water level in meters (Typically, this value is 0.5 meters). Four values are computed in Excel: DepthM, DepthFt, ElevM and ElevFt, where: 6

13 DepthM = Adj_Depth DepthFt = Adj_Depth * These water depths are RELATIVE water depths that can vary from day-to-day based on the elevation of the water surface. In order to normalize all depth measurements to an absolute reference, water depths must be subtracted from an established value for the elevation of the water surface at the time of the bathymetric survey. Determination of water surface elevation has been described in an earlier section on establishment of lake levels. To set depths relative to lake elevation, another field is added to the attribute table of the point shapefile, ElevM. The value for this attribute is then computed as Depth_ElevM = (Elevation of the Water Surface in meters above sea level) - Adj_Depth. Elevation of the water surface in feet above sea level is also computed by converting ElevM to elevation in feet (ElevM * ). Particularly for multi-day surveys, ADJ_DEPTH, Depth_M, and Depth_Ft should NOT be used for further analysis or interpolation. If water depth is desired, it should be produced by subtracting Elev_M or Elev_Ft from the reference elevation used for interpolation purposes (for federal reservoirs, the elevation of the water surface on the day that the aerial photography from which the lake perimeter polygon was digitized). Post-processing (ArcGIS): Ingest to ArcGIS is accomplished by using the Tools Add XY Data option. The projection information is specified at this time (WGS84). Point files are displayed as Event files, and are then exported as a shapefile (filename convention: ALLPOINTS_WGS84.shp). The pointfile is then reprojected to the UTM coordinate system of the appropriate zone (14 or 15) (filename convention ALLPOINTS_UTM.shp). Raster interpolation of the point data is performed using the same input data and the Topo to Raster option within the 3D Extension of ArcGIS. The elevation of the reservoir on the date of aerial photography used to create the perimeter/shoreline shapefile was used as the water surface elevation in all interpolations from point data to raster data. Contour line files are derived from the raster interpolation files using the ArcGIS command under 3D Analyst Raster Surface Contour. Area-elevation-volume tables are derived using an ArcGIS extension custom written for and available from the ASTRA Program. Summarized, the extension calculates the area and volume of the reservoir at 1/10-foot elevation increments from the raster data for a series of water surfaces beginning at the lowest elevation recorded and progressing upward in 1/10-foot elevation increments to the reference water surface. Cumulative volume is also computed in acre-feet. 7

14 DepthinFeet Ü Figure 4. Water depth based on October 2007 bathymetric survey. Depths are based on a pool elevation of feet. 8

15 Table 1 Cumulative area in acres by tenth foot elevation increments Elevation (ft NGVD)

16 Table 2 Cumulative volume in acre-feet by tenth foot elevation increments Elevation (ft NGVD)

17 Cumulative Area (acres) Elevation (feet) Figure 5. Cumulative area-elevation curve Cumulative Volume (acre-feet) Elevation (feet) Figure 6. Cumulative volume-elevation curve 11

18 PRE-IMPOUNDMENT MAP Caution should be exercised in drawing conclusions based on comparison between two maps of different scales, dates, and production methods. A pre-impoundment topographic map dated 1939 with a contour interval of ten feet (10 ) was obtained in digital form from the US Army Corps of Engineers (USACE) via the Kansas Water Office (Figure 7). The two map panels were georeferenced to the Universal Transverse Mercator (UTM) projection, NAD83, Zone 15, using the ArcGIS Georeferencing Tool. Control points were located on the 1939 Corps maps at section corners and referenced to corresponding locations on a UTM-georeferenced USGS Digital Raster Graphic (DRG) topographic map. A second-order polynomial transformation was computed from the coordinate pairs, and the 1939 maps were resampled to the UTM coordinate system using a nearest-neighbor algorithm. Contour lines were manually digitized to a polyline shapefile and attributed (Figure 8). All contour intervals at elevations 1520 feet and below were digitized (every ten feet of vertical). Lines representing streams were digitized as a separate polyline shapefile and used as breaklines to force valley bottoms to their true locations. The ArcGIS TIN tool was then used to generate a Triangulated Irregular Network (TIN) for the reservoir. The TIN file was then converted to a raster file to facilitate comparison of elevations to the 2007 bathymetric data (present-day lake bottom elevations) (Figure 9). The pre-impoundment cumulative area-elevation curve exhibited a stair-step effect as a result of the 10-foot contour interval used to create the preimpoundment digitial elevation model (Figure 10, orange line). A cubic spline function was applied to the cumulative area-elevation curve, using the contour line intervals (e.g., 1220, 1230, 1240, 1250) as input and splining to one-foot intervals to eliminate the stair-step effect (Figure 10, blue line). No spline function was applied to the cumulative volumeelevation curve data (Figure 11). Area-volume-elevation curves for the 2007 data were also plotted with the preimpoundment area-volume-elevation curves as a graphical representation of changes in area and volume since construction. Changes in lake bottom elevation between 1939 and 2007 were computed by digitally subtracting the 1939 digital elevation model from the 2007 digital elevation model. Negative numbers on the resulting output indicate 2007 elevation lower than 1939 elevation (loss of material during the 64-year period); positive numbers indicate 2007 lake bottom higher than 1939 (accumulated material, or likely siltation) (Figure 12). As the contour interval is ten (10) feet, areas of ± 5 feet difference are not shown on the difference map. The difference map suggests that the greatest sedimentation has occurred in the former river channel, as might be expected; furthermore, the majority of the non-river channel silt accumulation has occurred in the upper part of the reservoir (Figure 12; orange and yellow colors). 12

19 Ü Legend 2002 DOQQ Perimeter Miles Figure 7. Scanned preimpoundment contour lines map 13

20 Ü Miles Legend 2002 DOQQ Perimeter 10-ft Contours Streams Figure 8. Digitized preimpoundment contour lines map 14

21 Ü Miles Legend 2002 DOQQ Perimeter Elevation in feet AMSL (NGVD29) High : Low : 1430 Figure 9. Preimpoundment digital elevation model. 15

22 Table 3 Cumulative pre-impoundment area and volume by one-foot elevation increments Elevation (ft NGVD29) Splined pre-impoundment cumulative area (acres) Volume (acre-feet)

23 3500 Cumulative Area (acres) Splined data Unsplined Data Elevation (feet) Figure 10. Raw and splined pre-impoundment cumulative area-elevation curve Cumulative Volume (acre-feet) Elevation (feet) Figure 11. Pre-impoundment cumulative volume-elevation curve 17

24 Change in lake-bottom elevation, (values in feet) Figure 12. Change in lake-bottom elevation, 1939 (pre-impoundment) to Negative values indicate 2007 elevation lower than 1939 elevation; Positive values indicate 2007 lake bottom higher than

25 3500 Cumulative Area (acres) Elevation (feet) Figure 13. Comparison of splined pre-impoundment and 2007 survey cumulative areaelevation curve Cumulative Volume (acre-feet) Elevation (feet) Figure 14. Comparison of pre-impoundment and 2008 survey cumulative volume-elevation curve 19

26 SEDIMENT CORING/SAMPLING PROCEDURES KBS operates a Specialty Devices Inc. sediment vibracorer mounted on a dedicated 24 pontoon boat. The vibracorer uses 3 diameter aluminum thinwall pipe in user-specified lengths (KBS has used up to 10 sections). The vibracorer runs off 24-volt batteries, and uses an electric motor with counter-rotating weights in the vibracorer head unit to create a high-frequency vibration in the pipe, allowing the pipe to penetrate even solidly packed sediments and substrate as it is lowered into the lake using a manually operated winch system. Once the open end of the core pipe has penetrated to the substrate, the unit is turned off and the unit is raised to the surface using the winch. At the surface, the pipe containing the sediment core is disconnected from the vibracore head for further onboard processing. The sediment core can be cut into sections while in the pipe, the pipe bisected longitudinally for taking samples along the length of the core, or it can be extruded from the tube and measured. KBS vibe-core system. At each site, determined using GPS, the core boat is anchored and the vibracore system used to extract a sediment core down to and including the upper several inches of pre-impoundment soil (substrate). The location of each core site is recorded using a GPS linked to a laptop running ArcGIS and the ArcGIS GPS extension. Cores are carefully extruded from the core pipe, and the interface between sediment and substrate identified. Typically, this identification is relatively easy, with the interface being identifiable by changes in material density and color, and the presence of roots or sticks in the substrate. For most analyses, the top six inches of sediment are collected and sealed in a sampling container. Sediment re-sampling: Several samples were damaged in shipping for analysis. On April 14, 2009, the sites were re-sampled. A GPS linked to ArcGIS and the map of original coring sites was used to locate the boat within ±5 meters of the original site. Several additional sites in the upper end were taken as well. At each location, a Ponar dredge was used to take a sediment sample from the top 3-5 inches of sediment. The sample was manually mixed to ensure uniformity and a sample amount of 32 volumetric ounces (~940 cubic centimeters) was taken. The samples were then sealed and shipped to MidWest Laboratories for texture analysis. 20

27 Sediment Coring and Sampling Results: Sampling sites were distributed across the length and breadth of the reservoir (Figure 15). Generally, an effort was made to avoid the original stream channel, which would have likely yielded higher sediment thicknesses not representative of the overall reservoir bottom sediment thickness. Sediment thickness ranged from a low of 34 centimeters at site K01 (located in the shallow southwestern part of the reservoir)(figure 16, Table 4) to greater than 250 centimeters at sites K04 and K09. Although high sediment thicknesses were found in the upper end of the reservoir near the inflow (Sites K09 and K10), all three sites nearest the dam had sediment thicknesses exceeding two meters (200 cm) (sites K04, K05, and K06)(Figure 16; Table 4). Clay dominates the particle size analysis in samples taken from Kanopolis Reservoir (Table 4; Figure 17; Figure 18). Of the thirteen sites sampled, only three samples had less than 50% clay content (sites K01, K02, and K07). Sites K01 and K02 are both located in the western part of the lower end of the reservoir, where the lake bottom topography forms a shallow shelf before descending into the main stem of the reservoir (refer back to Figure 4 for lake bottom topography). These two sites also exhibited higher sand percentages (K01: 33%; K02: 18%) than most of the other sites in the reservoir (Figure 17; Figure 18). 21

28 K13 K10 K09 K08 K12 K02 K07 K11 K03 K06 K04 K01 K Miles Ü Figure 15. Sediment coring and sampling sites in Kanopolis Reservoir 22

29 Miles Ü Figure 16. Sediment thickness in centimeters at coring sites in Kanopolis Reservoir 23

30 Table 4 Kanopolis Reservoir Sediment Coring/Sampling Data Code UTMX UTMY Sediment Thickness (cm) Sand % Silt % Clay % K K Ind K K K K K K K K K n/m K n/m K n/m Notes: 1. Ind. = Indeterminate sediment-substrate interface. Unable to estimate sediment thickness. 2. Sediment thickness exceeded the 250 cm (8 feet) length of the core tube. 3. Sediment resampled for texture analysis on 4/14/ n/m = Sediment thickness not measured. Sample taken for texture analysis only. Coordinates are Universal Transverse Mercator (UTM), NAD83, Zone 14 North 24

31 Kanopolis Lake 2008 Sediment Particle Size Analysis 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% K01 K02 K03 K04 K05 K06 K07 K08 K09 K10 K11 K12 K13 Sample Site % Clay % Silt %Sand Figure 17. Sediment particle size analysis. 25

32 Particle Size Distributions Sand K13 K10 Silt Clay K09 K08 K12 K07 K02 K11 K03 K06 K04 K01 K05 Ü Miles Figure 18. Particle size distributions of samples taken from coring and sampling sites in Kanopolis Reservoir 26