Relief Well Rehabilitation Options in Chief Joseph Dam Right Abutment Drainage Tunnel

This report summarizes the data, observations, methods, assumptions, and decisions for the design of the Relief Well Rehabilitation Project in the Right Abutment Drainage Tunnel at Chief Joseph Dam. Chief Joseph Dam (CJD) is a dam on the Columbia River and is owned and operated by the U.S. Army Corps of Engineers (USACE). It is the second only to Grand Coulee dam as the largest producer of hydropower in the United States. The right abutment drainage tunnel contains wooden stave relief wells. Water flows from these wells which reduces the hydrostatic pressure in the right abutment of the dam. The 22 wells in the floor of the tunnel are 60 years old and are in need of rehabilitation. The objective of this project is to control the groundwater gradient, prevent the movement of sediment, stop total screen collapse, and prevent initiation of backwards erosion and piping in the abutment. The rehabilitation solution is to install new stainless steel screens into the existing wells, backfill the annular space between the old wooden screen and the new stainless steel screens with a 3/8-inch pea gravel filter pack, and install a new top cap to hold the new screen in place. This report documents the data, observations, and methods used to complete the final design. During tunnel inspections USACE geologists observed dislodged end plugs and evidence of sediment movement out of the formation. The relief wells have historically high flows between 6,000 gallons per minute (gpm) to 9,000 gpm. New screens are designed based on as-built data and historic tunnel flow. The new screens are 8-in diameter, 100 slot (0.10-inch) screens. We found that screen diameter and slot size would provide adequate transmitting capacity for most of the relief wells. The filter pack gradation is based on descriptions from foundation construction reports. I found that 3/8-inch pea gravel is appropriate for the abutment material. During design, I also considered an option to install the screens into the relief wells without filter pack. I eliminated this option because it did not meet our rehabilitation objective to prevent total failure of the wooden screens.

Determining Spatial Distribution and Physical Properties of the Vashon Advance Outwash near Mountlake Terrace, Washington

This study is aimed at determining the spatial distribution, physical properties, and groundwater conditions of the Vashon advance outwash (Qva) in the Mountlake Terrace, WA area. The Qva is correlative with the Esperance Sand, as defined at its type section; however, local variations in the Qva are not well-characterized (Mullineaux, 1965). While the Qva is a dense glacial unit with low compressibility and high frictional shear strength (Gurtowski and Boirum, 1989), the strength of this unit can be reduced when it becomes saturated (Tubbs, 1974). This can lead to caving or flowing in excavations, and on a larger scale, can lead to slope failures and mass-wasting when intersected by steep slopes. By studying the Qva, we can better predict how it will behave under certain conditions, which will be beneficial to geologists, hydrogeologists, engineers, and environmental scientists during site assessments and early phases of project planning. In this study, I use data from 27 geotechnical borings from previous field investigations and C-Tech Corporation’s EnterVol software to create three-dimensional models of the subsurface geology in the study area. These models made it possible to visualize the spatial distribution of the Qva in relation to other geologic units. I also conducted a comparative study between data from the borings and generalized published data on the spatial distribution, relative density, soil classification, grain-size distribution, moisture content, groundwater conditions, and aquifer properties of the Qva. I found that the elevation of the top of the Qva ranges from 247 to 477 ft. I found that the Qva is thickest where the modern topography is high, and is thinnest where the topography is low. The thickness of the Qva ranges from absent to 242 ft. Along the northern, east-west trending transect, the Qva thins to the east as it rises above a ridge composed of Pre- Vashon glacial deposits. Along the southern, east-west trending transect, the Qva pinches out against a ridge composed of pre-Vashon interglacial deposits. Two plausible explanations for this ridge are paleotopography and active faulting associated with the Southern Whidbey Fault Zone. Further investigations should be done using geophysical methods and the modeling methods described in this study to determine the nature of this ridge. The relative density of the Qva in the study area ranges from loose to very dense, with the loose end of the spectrum probably relating to heave in saturated sands. I found subtle correlations between density and depth. Volumetric analysis of the soil groups listed in the boring logs indicate that the Qva in the study area is composed of approximately 9.5% gravel, 89.3% sand, and 1.2% silt and clay. The natural moisture content ranges from 3.0 to 35.4% in select samples from the Qva. The moisture content appears to increase with depth and fines content. The water table in the study area ranges in elevation from 231.9 to 458 ft, based on observations and measurements recorded in the boring logs. The results from rising-head and falling-head slug tests done at a single well in the study area indicate that the geometric mean of hydraulic conductivity is 15.93 ft/d (5.62 x 10-03 cm/s), the storativity is 3.28x10-03, and the estimated transmissivity is 738.58 ft2/d in the vicinity of this observation well. At this location, there was 1.73 ft of seasonal variation in groundwater elevation between August 2014 and March 2015.