• Home
• Introduction
• Sedimentation History
• Geophysical Studies
• Results
• Summary
• References
• Acknowledgments
• Figures
• CMG Home Page
• Search CMG

Distribution, thickness, and volume of fine-grained sediment from precipitation of metals from acid-mine waters in Keswick Reservoir, Shasta County, California

Geophysical Studies

In 1993, the USGS used a Global Positioning System (GPS) navigation system and marine high-resolution seismic-reflection equipment to delineate the extent and thickness of sediments deposited into the lower Spring Creek Arm and the adjacent main Keswick Reservoir (Figures 2, 3A-E). The boat used was a houseboat obtained from a rental agency on Shasta Lake.

Navigation

Navigation for the survey was determined using a Differential Global Positioning System (DGPS) composed of a shore-based receiver mounted in a known position, and a second receiver mounted on the boat. We used the USGS-developed navigation software, “Yo-Nav,” to receive uncorrected GPS positions and to apply the differential correction based on the shore receiver (Gann, 1992). These data were digitally recorded by Yo-Nav, and were edited after the fieldwork to remove spurious navigation points or abrupt changes in location.

Three problems are known to exist with this data set. First, the northern land-receiver station was not on an existing USGS benchmark, but was surveyed in with the available GPS equipment; this station could have a plus or minus 5 m absolute location error. Such an error would in turn create a static shift in all of the survey data. The southern land-receiver station was on an existing benchmark, and data from Keswick dam 1.5 km north to the westward bend of the reservoir do not show any static offset with respect to the shoreline.

The second problem is that, after the cruise, one of the navigation heads was found to be malfunctioning, creating 5-10 m location jumps in the data set. During post-fieldwork processing of the data, most of these jumps were readily apparent, and were corrected simply by moving the offset line segment back onto the main line. In areas where the correction was not clear, mainly where signal reception was poor, either navigation and seismic data were eliminated from the data set, or the navigation position was estimated from both the plotted position and field notes kept during data collection.

Finally, three areas were significantly affected by poor signal reception; in these areas the navigation accuracy must be considered as relatively poor. These areas are: (1) in the segment of upper Keswick Reservoir that trends northeast beginning about 1 km north of Spring Creek Arm (area north of line 51, Figure 3A), (2) at the westernmost end of the Spring Creek arm (western 300 m of lines 53, 54, Figure 3B-1), and (3) at the westward bend in lower Keswick Reservoir about 0.5 km south of Spring Creek Arm (mainly including the area of Figure 3D, lines 56-58, 63-65, and 79-82).

Seismic-reflection geophysical equipment

High-resolution seismic-reflection data (Figures 3A-E and 4) were acquired using two systems. A sled-mounted Geopulse system consisting of a boomer-plate source and a short single-channel hydrophone streamer receiver was towed just aft of the survey boat at 1-4 knots. This system was fired at 1/8-second (s) intervals and digitized at 16 kiloHertz (kHz), with a usable frequency range of about 2000-6000 Hertz (Hz). A USGS-developed acquisition system, “Mudseis”, was used to digitize and record the data. This system allowed imaging of sedimentary units to bedrock, including presumed alluvial units at the bottom of the creek or river beds, and the overlying material composed of mine-waste debris or chemically precipitated sediment. Penetration was on the order of 15-25 m subbottom, where sediments that thick were present.

An analog 200-kHz Lowrance bathymetric system was run concurrently with the Geopulse system, also at a 1/8-s firing interval. This system provided a high-resolution bathymetric profile, with some minor subbottom penetration. However, the bathymetric profiles were critical for areas where chemically-precipitated sediment had accumulated. The top of this sediment layer was imaged well by the Lowrance system, but only poorly imaged or not imaged at all in some areas on the Geopulse data.