This article presents a recent project performed by undergraduate students at The University of Texas at San Antonio. Their project is the culmination their degrees in a capstone class. The objective of the class was to present an opportunity to apply design skills to execution of an open-ended integrated civil engineering design project, including field and laboratory investigations, numerical and scale modeling, design, and formal oral and written presentation of results. Their project considers safety, reliability, environmental, economic, and other constraints, as well as ethical and social impacts.
The students chose a rehabilitation and expansion of an existing 6 acre reservoir, dam and site appurtenances. The project site is located in Bell County, Texas just outside the Holland city limits (Figure 1). Existing site features include a reservoir with an approximate 6 acre surface area, and earthen dam and spillway on the north end of the reservoir.
The site area is known to consist of mainly Alluvium Sandy Clay on top of Uvalde Gravel on top of Weathered Navarro with underlying thick Clay Shale. Figure 2 shows a historical geological map of the area where the dam will be located. The area for the proposed dam is located in a sparsely vegetated area within a gentle valley with gradual to semi steep slopes. The historical data was used in lieu of subsurface investigation to generalize the site into four major strata as seen in Figure 3. From this generalization the students were able find typical values, calculate, and use engineering judgment based on knowledge of the site to produce in-situ soil conditions. Tests that were simulated to arrive at final soil conditions are as follows:
- Atterberg Limits ASTM418-10
- 200 Sieve Analysis ASTM1921
- Unconsolidated and Undrained (UU) ASTM D2850-03a obtained by a triaxial shear test
- Density ASTM D1557-09
- Specific Gravity ASTM D854 - 10
- Unified Soil Classifications
In addition several publications were used in order to ensure realistic values were achieved.
The study involved a watershed analysis to determine the maximum expected water levels of the reservoir. The maximum water surface elevation was determined to be:
- 100 year Storm = 495.6', and
- Design 50% PMF = 496.75'.
The proposed dam would have a watershed of approximately 390 acres. A peak discharge of 1035 cubic feet per-second was used for design. Hydrologic modeling was done with HEC-HMS, while hydraulics calculations were completed with HEC-RAS, Hydraflow Express, and various hand calculations.
The initial student design for the dam involved a 20ft top for a roadway. 3:1 side slopes were designed with a core minimum width of 10ft for compaction equipment. Iterative trials were then used to determine the final design. Original design was done with AutoCAD Civil 3D 2011. The student created profiles at 50 ft intervals from the centerline for analysis. A spillway analysis was performed in order to determine the potential draw-down rates for the reservoir. The AutoCAD drawing for the section Station 3+00 may be seen in Figure 4. The geometry of multiple cross-sections was subsequently imported into the SVSLOPE / SVFLUX software for analysis through the DXF import feature. This feature allowed the students to easily go from their AutoCAD Civil3D model outputs to the Soil Vision software.
All earth filled dams must be safe and stable during the phases of construction and during the operation of the reservoir. To accomplish this, certain criteria must be met, they are as follows.
- The embankment must be safe against over topping during the occurrence of an inflow at design flood level, this is accomplished by providing sufficient spillway and outflow structures that could handle this capacity.
- The slopes of the embankment must maintain stability during construction and under all reservoir operations, including rapid drawdown of the reservoir.
- The embankment must be designed so as to not impose an excessive stress upon the foundation.
- Seepage flow through the embankment, foundation, and abutments must be controlled so that no internal erosion takes place, and there is no sloughing in the area where the seepage emerges. The amount of seepage must be kept to a low enough level to ensure that it does not interfere with regular planned functions
- The embankment must be safe against over topping by wave action.
- The upstream slope must be designed to resist erosion against wave action, and the downstream slope should be able to resist erosion due to rain and wind.
- If the dam is located in an area subject to regular earthquakes, the design must be such that it can withstand the strongest reasonable earthquake that is expected in the region.
Calculations and models were ran to ensure against these conditions, except for (7) because the dam is not located in a seismically active region. The following conditions were analyzed with the Soil Vision numerical modeling program suite. Sections at 50 ft offsets from center were analyzed for the following:
- Analyzed Conditions
- Post-fill (full pond)
- Low pond
- Rapid drawdown
The SVSLOPE® and SVFLUX™ software packages created and marketed by SoilVision Systems Ltd. were utilized for the study. The answers from the software were subsequently checked using the United States Army Corps of Engineers (USACE) engineering methods.
The creation of the seepage numerical model required the setup of initial conditions and boundary conditions as seen in Figure 5. Flux sections were inserted in the model to track flows through various parts of the model.
Subsequent seepage analysis of multiple cross-sections was carried out in a steady-state and transient-state fashion in order to determine pore-water pressure distributions, flows through the drains, and phreatic conditions at various points in time. An example of the results of the seepage analysis may be seen in Figure 6. Of particular note is the influence of the drain on the flow regime. The filter layer was sized per USACE manuals.
SLOPE STABILITY ANALYSIS
The slope stability analysis on both the upstream and downstream sides of the earth dam for multiple cross-sections was carried out. Pore-water pressures were input from the related SVFLUX seepage models previously created. By using the Soil Vision software suite results from the SVFLUX model could be accessed in SVSLOPE. In order to determine the most critical slip surface location a number of different slip surface shapes (circular, non-circular, trapezoidal) were tested alongside multiple searching methods.