Date of Award

5-2016

Document Type

Thesis

Degree Name

Master of Science (MS)

Legacy Department

Hydrogeology

Committee Member

Dr. Lawrence Murdoch, Committee Chair

Committee Member

Dr. Ronald Falta

Committee Member

Dr. Stephen Moysey

Abstract

An instrument, called a DELTA extensometer, has been developed to measure vertical displacement in the subsurface with a resolution of less than 0.01 μm. The instrument is 2-m-long, and it is pushed below a vertical boring where it is anchored to the soil at either end. Displacement of the anchors responds to changes in mass load caused by changes in water content as well as other factors. The goal of this study is to evaluate the feasibility of using subsurface displacement data to estimate average changes in moisture content and other hydrologic processes. Ten extensometers were successfully installed in saprolite, loess, and clay at depths from 3 m to 6 m. The focus of this study is on four extensometers installed in saprolite near Clemson, SC. The tops of three of the extensometers were at a depth of 6 m, and one was shallower at 3 m. The water table rose from 9 m to 7 m during the study. The extensometers were initially above the water table, but the water table was between the upper and lower anchors for most of the study. The field site has been monitored since early 2012, with a nearly continuous record of vertical displacements along with other data, including rainfall, water level in piezometers, barometric pressure, pan evaporation, soil moisture, wind speed and other weather data used to calculate evapotranspiration in the Penman Monteith equation. Four factors were identified that cause displacement: pore pressure, barometric pressure, temperature, and mass load from water and other sources. The effects of these variables on displacement were evaluated by establishing loading coefficients using environmental data from the field site. The loading coefficients at the four extensometers from pore pressure ranged from 12 (±1) μm/kPa to 32 (±11) μm/kPa and from mass loads were 8 (5 to 15) μm/kPa to 20 (14 to 34) μm/kPa. The numbers in parentheses are the uncertainty characterized by one standard deviation. These coefficients were within the range of uncertainty of the estimates, so it was concluded that they were essentially the same. The barometric signal had smaller loading coefficients ranging from 1.8 (±1) to 4.2 (±0.6) μm/kPa, and they changed as a function of time. The response to temperature fluctuations varied among the extensometers and ranged from -0.3 (±17) μm/kPa to -37 (±5.8) μm/kPa. The response of displacement to changes in surface moisture content was evaluated following the removal of the signals resulting from pore pressure, temperature and barometric pressure. The general behavior consisted of compressive displacements during rainfall, with extension between rainfalls. A loading coefficient was established by correlating rainfall amount to compressive displacement that occurred during individual rainfalls. The coefficient ranged from 13 (±4) μm/kPa to 30 (±9) μm/kPa at the different extensometers, where the units are converted to be similar to those used for the other loading coefficients. In general, the loading coefficients from rainfall were similar to those for other processes, except barometric loading. Displacement during most rainfalls was compressive and the magnitude was proportional to the amount of rainfall, but there were exceptions. Sudden extension occurred during or soon after some rainfalls. In these cases, the displacement was apparently a result of processes other than the progressively increasing load that occurred as water accumulated at the surface during rainfall. This response was particularly likely to occur during heavy rainfalls that were preceded by other large rainfalls. This association with antecedent moisture suggests that this response may be associated with rapid downward flow that could reduce the overlying load and increase the pressure between the anchors, both of which would cause extension. Extension occurred between rainfalls and the displacement rate was converted to a rate of change of water volume using the rainfall loading coefficient. The rates estimated from displacement at the 6-m-deep extensometers were generally similar to the rate of change in average surface moisture content measured using capacitance gauges, with correlation coefficients of 0.72, 0.73, and 0.66. The results indicate that the extensometers respond to changes in water content, but the details of their response differs from that of the capacitance gauges, probably because of the difference in the averaging scales of the instruments. The extensometers respond to changes in load averaged over a region that extends laterally approximately twice their depth, and vertically over approximately half their depth on average. This means that the 6-m-deep extensometers respond to changes in water content over 102 m3 to 103 m3. By comparison, capacitance gauges respond to changes in water content over approximately 10-1 m3. The overall findings of this study show that vertical displacements of approximately 100 microns occur on an annual cycle as a result of pore pressure and temperature changes, and smaller displacements occur with shorter periods due to barometric pressure changes. Rainfall causes compression whereas evapotranspiration and recharge cause extension, and these effects are characterized most effectively following the removal of displacements from pore pressure, temperature, and barometric pressure.

Share

COinS