0
Research Papers

Seepage Monitoring of an Embankment Dam Based on Hydro-Thermal Coupled Analysis OPEN ACCESS

[+] Author and Article Information
Chung R. Song

Civil Engineering Department,
University of Nebraska-Lincoln,
Lincoln, NE 68583
e-mail: csong8@unl.edu

Tewodros Y. Yosef

Civil Engineering Department,
University of Nebraska-Lincoln,
Lincoln, NE 68583
e-mail: tyyosef@huskers.unl.edu

1Corresponding author.

Contributed by the Materials Division of ASME for publication in the JOURNAL OF ENGINEERING MATERIALS AND TECHNOLOGY. Manuscript received May 31, 2016; final manuscript received February 16, 2017; published online March 2, 2017. Assoc. Editor: Taehyo Park.

J. Eng. Mater. Technol 139(2), 021024 (Mar 02, 2017) (9 pages) Paper No: MATS-16-1160; doi: 10.1115/1.4036020 History: Received May 31, 2016; Revised February 16, 2017

Distributed temperature sensing (DTS)-based fiber optic sensors are widely used for monitoring spatially continuous temperature distribution in structures. In this research, hydro-thermal (H-T) coupled analysis is used to monitor seepage conditions in an embankment dam. Variably saturated two-dimensional heat transport (VS2DHI), a computer code developed by the U.S. Geological Survey, was used for this coupled analysis. From the coupled analysis, the temperature profile for a dam with an artificially generated crack clearly showed the location of the crack. In addition, it turned out that the temperature change in the dam took much longer than the seepage time due to the additional time required for heat transfer. The study shows that temperature variation in the dam is comparable to the seepage condition with time delay for heat transfer. This study also shows the possibility that temperature data may serve as a tool to diagnose prior seepage conditions and past incidents of a dam.

FIGURES IN THIS ARTICLE
<>

Seepage monitoring has been an important part of safety evaluation for embankment dams. Monitoring seepage condition enables engineers to predict seepage abnormalities, overall health, and integrity of embankment dams. Traditional seepage monitoring techniques include measuring pore pressure within the dam using piezometers. The spatial resolution of these piezometers may not be refined enough to detect local defects. According to Johansson [1], this is one of the main reasons why most seepage problems such as internal erosion incidents have been detected by visual inspections, not by monitoring systems.

Temperature measurement as a general tool for investigating the interaction between surface water and ground water, flow patterns in aquifer, infiltration of water, and leakage detection in embankment dams has been suggested by a number of researchers [27]. Even though there are copious theoretical studies and many benefits of temperature measurement, the method is not yet developed well enough to serve as a popular method to assess the health condition of geotechnical structures.

Seepage monitoring for embankment dams through temperature measurement, however, is gaining attention in recent years due to the development of a DTS technology based on fiber optic sensors [8]. DTS using fiber optics has provided significant advantages in overcoming the discrete nature of the point measuring technique using temperature probes. In the DTS technique, a LASER pulse is sent to optical fibers, and Raman spectroscopy is used to analyze the backscattered light. Based on Raman spectroscopy, the temperature at a particular point can be related to the ratio of anti-Stoke to Stoke intensity [9].

Among the different researchers who worked with distributed fiber optic temperature sensing, Dornstädter [9] reported the possibility of measuring and acquiring high resolution temperature data at different points of interest up to a length of 12 km. This versatility and accuracy in measuring continuous temperature data makes this technique desirable for the health monitoring and surveillance of geotechnical structures, such as embankment dams. Distributed temperature sensing based on fiber optics was first implemented during the year 1996 [10]. After that time, the DTS technique based on fiber optics has been applied for different continuous or occasional health monitoring schemes in levees and embankment dams [10].

The theoretical framework that uses temperature measurement as a basis for seepage monitoring starts from the consideration of natural and seasonal temperature variations of water in the reservoirs. Temperature change inside the dam, particularly in deep depth, may change when reservoir water seeps through. This temperature change can be used to analyze the seepage condition. Generally, a smooth temperature variation indicates small seepage, while large seasonal variation may indicate significant seepage or possibly concentrated under-seepage [1]. One cause for the existence of the temperature field inside the dam is advective heat transport and associated diffusion to the surrounding area as a result of seeping water from the upstream reservoir. Another cause is conductive heat transport from the surface of the dam. The techniques to analyze the temperature data, however, are not well developed yet. Instead, they are typically analyzed by visual inspection of spatial domain data and time domain data. This study applied the hydro-thermal coupled analysis to more rationally analyze temperature inside an embankment dam.

The seeping water from the reservoir into the dam body generates temperature front that travels together with the seeping water. The temperature in the shallow part of the dam is mostly affected by ambient air temperature, and the influence of atmospheric temperature is negligible in the deeper part of the dam, particularly for soils under the phreatic line. For high seepage rate, natural and seasonal temperature fluctuation in the reservoir water may primarily control the temperature variation inside the dam.

The hydro-thermal process includes such basic thermal processes as heat conduction, advection, and radiation due to the seeping water. Coupling behavior exists between advective–conductive heat transport, because physical parameters like density and viscosity are temperature dependent. The problem is further complicated by consideration of anisotropy in dam materials and the variety of different degrees of saturation of the dam body. In order to analyze the problem and for the convenience of computation, certain assumptions have to be made. In this research, isotropic porous material with steady-state flow is assumed. In addition, geothermal effects are ignored.

Kipp [11] addressed the heat-seepage coupled process in porous-media by the following coupling equation. The equation was first formulated for advective–dispersive solute transport problems Display Formula

(1)t(θCW+(1ϕ)Cs)T=KT(θ)T+θCWDHTθCWvT+qCWT*

where t is time in second (s); θ is volumetric moisture content of the soil; CW is heat capacity of water, in J/m3 °C; ϕ is porosity; Cs is heat capacity of the dry soil, in J/m3 °C; T is temperature, in °C; KT is thermal conductivity of the soil and water matrix, in W/m °C; DH is thermomechanical dispersion tensor, in m2/s; v is water velocity, in m/s; q is rate of fluid into or out of the domain, in s−1; and T* is temperature of fluid into or out of the domain, in °C. The left-hand side of Eq. (1) represents the temperature change in a dam body. The first term on the right-hand side of Eq. (1) accounts for the heat transport by thermal conduction. The second term accounts for the heat transfer by dispersion. The third term accounts for the heat transport as a result of the movement of water of different temperature, which is actually advective heat transport. The final term is for heat sources or sinks [11]. Heat sources or sinks take into account the mass added or removed from the domain by flowing water into and out of the domain.

Hydrodynamic dispersion accounts for heat transport due to the movement of water of different temperatures. The tensor quantity that accounts for hydrodynamic dispersion is given by Healy [12] Display Formula

(2)DHi,j=αT|v|δij+(αLαT)vivj/|v|

where αT represents dispersivity in the transverse direction, in m; vi is the ith component of the velocity vector; δij is the Kronecker delta; αL denotes the dispersivity in the longitudinal direction, in m; and |v| is the norm of the velocity vector. Equations (1) and (2), therefore, constitute a form of conductive–dispersive equation resulted from the conservation of internal energy in porous-media.

In addition, the flow velocity in Eqs. (1) and (2) is heavily governed by the hydraulic conductivity of soils. The hydraulic conductivity for unsaturated soils is affected by the water content. Buckingham [13] realized this dependence of water flow on moisture content and proposed the term called “conductivity,” similar to what is now referred to as “hydraulic conductivity.” Nearly 20 years later, Richards [14] introduced a comprehensive differential equation for water flow in unsaturated porous-media applying Buckingham's [13] concept to continuity equation.

Essentially, the general combination of continuity equation with Darcy's law gives the Richards equation. In order to derive the equation, one has to write the two relations first. Seepage field in unsaturated regime is given by Richards equation Display Formula

(3)θt=z(Dθz)+Kz

where θ is the volumetric water content of the soil; K is the hydraulic conductivity; t is time; D is pore water diffusivity; and z is the distance. It should be noted that both D and K are highly dependent on water content.

In order to solve Eq. (3), one must first properly address the task of determining D and K, both of which depend on water content. Several models based on theoretical and empirical evidence have been suggested for determining these parameters. Brooks and Corey model [15,16] and Van Genuchten model [17] are more widely used models. Relative saturation rate was the concept introduced in Van Genutchten model, which relates pressure, water content, and hydraulic conductivity of unsaturated soil. This model matches with experimental data and can be implemented in closed-form solution schemes. Thus, Van Genutchten model is adopted in the present research. The model uses the following relations to define hydraulic conductivity and water diffusivity: Display Formula

(4)D(θ)=(1m)Ksαm(θsθr)θ(1/2)(1/m)[(1θ1/m)m+(1θ1/m)2]

where θr is the residual moisture content in the soil; θs is saturated soil moisture content; and Se is effective saturation. Effective saturation is given by the following equation: Display Formula

(5)Se=1(1+|αHp|n)m

where Hp is the pressure head in unsaturated zone; α, n, and m are Van Genuchten model parameters; α is the reciprocal value of the intake line in moisture characteristics curve; and n is a parameter for water characteristics curve slope number, m=1(1/n).

Ks is the hydraulic conductivity at saturation, and Kr (=(K/Ks)) is the relative hydraulic conductivity implying that it is the hydraulic conductivity of unsaturated soils normalized with that of saturated soils. By Van Genuchten, Kr is further expressed as Display Formula

(6)Kr=θ1/2[1(1θ1/m)m]2,m=11n

It is also noted that the thermal conductivity of soil (KT KT in Eq. (1), but the symbol λ is used in the following formulation in line with existing literature) depends on porosity (n), bulk density (ρb), soil water content (θ), temperature, and mineral composition. It is predominantly governed by the water content or saturation state of the soil. Adopting the concept of normalized thermal conductivity to account for this invariance is important for solving practical problems. Johansen [18] first introduced the concept of normalized thermal conductivity (Ke) and established a relationship between different states of soil moisture condition (λ), which is Display Formula

(7)λ=(λsatλdry)Ke+λdry

where λsat and λdry are the thermal conductivity of dry and saturated soils (W m−1 K−1), respectively.

Sen et al. [19] proposed a new model for determining normalized thermal conductivity and the thermal conductivity of dry and saturated soils for entire range of soil water content, which are Display Formula

(8)Ke=exp{α[1Srα1.33]}
Display Formula
(9)λsat=λs1nλwn
Display Formula
(10)λdry=kx+y

where α is soil texture-dependent parameter; 1.33 is a shape parameter which accounts for particle shape effect; λs is effective thermal conductivity of soil solids; λw is thermal conductivity of water (λw  = 0.594 W m−1 K−1 at 20 °C); n is porosity; and k,x, and y are empirical parameters.

All the models described above can be used as a framework for analyzing a coupled hydro-thermal mechanism and process for the following rationale: (1) considers temperature field of advective–conductive thermal process, (2) considers seepage and heat flow in the unsaturated zone, (3) considers heterogeneous heat transfer in saturated–unsaturated regime, and (4) takes into account seasonal fluctuation of temperature. This research followed a computer code which accommodates most of the previous models, variably saturated two-dimensional heat transport (VS2DHI) published by U.S. Geological Survey [20].

Model Specification.

Based on the numerical model that solves heat transport in variably saturated porous-media, a number of numerical experiments can be conducted for the dam. A finite difference numerical technique is utilized, and a two-dimensional dam region is simulated using a Cartesian coordinate system.

Dam Geometry.

A typical cross section of a dam is used with slight modification to generalize it. The embankment dam shown in Fig. 1 has a crest width of 8.0 m, a crest elevation of 215.05 m, and a dam height of 11.25 m. The clay core has an elevation of 212.5 m, a top width of 3 m, and a slope of 1:1.15. Both the upstream and downstream surfaces have 1(V):3(H) slopes, and a gravel filter on the downstream side has 2 m thickness. To illustrate the impact of defect in the dam, 0.02 m thick crack is introduced in the clay core, as shown in Fig. 1.

Computation Parameters.

Referring the relevant literatures [12,15,19,2130], computational parameters shown in Table 1 are used.

Thermal dispersivity is taken as 0.01 m, heat conductivity of air and water is considered to be 0.0204 W/(m °C) and 0.58 W/(m °C), respectively, air and water heat capacity is taken as 1002 J/(kg °C) and 4179 J/(kg °C), respectively, and density of air and water is taken as 1.205 kg/m3 and 1000 kg/m3, respectively [1,20,31].

Boundary and Initial Conditions.

For the variably saturated seepage field, the line AIH along upstream side of the dam is a total head boundary, with the reservoir elevation of 212.50 m and the datum elevation of 203.8 m. The total head boundary is the one to allow the water head to mimic typical piezometer head. Boundary HGF is a zero flux boundary. A zero flux boundary is a condition in which the component of flux in the direction normal to the boundary is zero. The boundary along line CDEF is seepage boundary. In a seepage boundary, water is allowed to flow out of the domain along the seepage face. The pressure head along the seepage face is set to zero. Ground water table is at the elevation of 203.8 m.

For a temperature field boundary, the line AIH is taken as a temperature boundary for the reservoir water. Seasonal temperature fluctuation is provided by Chang [26], which is measured data for an actual dam in a cold region as shown in Fig. 2.

Boundary AB is assumed to be an adiabatic boundary, meaning that boundary AB does not permit heat to pass through it. Boundary CDEFGH is open to the atmosphere, so it is affected by ambient air temperature fluctuation. Similarly, data acquired from a weather agency used for incorporation of ambient air temperature fluctuation into the numerical model, as shown in Fig. 3. The ambient air temperature was also provided by Chang [26].

The procedure of stage construction and impoundment of the dam is not incorporated in this study, because this study is primarily a proof of concept. It is, however, noted that the application to the actual dam will need to follow the actual procedure of stage construction so that the proper initial condition is considered.

Time Discretization and Calculation Condition.

The numerical analysis was conducted in two stages. First, the effect of water seepage into the dam body temperature profile was simulated without the artificial defect, then the same simulation was conducted with the artificial defect. Both simulations were conducted for 1230 days lap time with 0.5 days of time step.

Seepage Profile.

Degree of saturation can be regarded as an indirect indicator of seepage condition in embankment dams. Particularly, the top of the saturated zone may reasonably represent the location of the phreatic line. Distribution of degree of saturation across the dam at different times is shown in Figs. 4 and 5 with (right side) and without (left side) the presence of artificial defect. Water flow through the clay core is small, since the core material has low permeability (0.052 m/day).

For the case of no artificial defect, the degree of saturation starts to change as the time after the impoundment increases. Particularly, the bottom foundation layer (SM, CL, SP-SM, SC, and SP-SM layer) was saturated in 10 days due to the high permeability. However, the filter material (SM-SC) for the dam was not saturated quickly due to the presence of the virtually impermeable concrete lining (facing) on the upstream slope of the dam. In 50 days, however, the top of the saturated zone starts to have a shape similar to the normal seepage lines. From 1000 days, the contour of saturation does not change noticeably, but shows an almost parallel degree of saturation contour.

For the case of the dam with the artificial defect in the core, the contour seems quite different. The degree of saturation at the shallower depth is lower than that for the dam without the artificial defect. It is believed that the artificial defect served as an underground aqueduct that collected most of the seepage water, and the soil above the artificial defect remained hardly saturated. This is interpreted to mean that the phreatic line could not be positioned above the artificial defect once the seepage water met the dam core, due to the artificial underground aqueduct. After 50 days, the contour is substantially different from the one for the dam without the artificial defect—the dam core is saturated only below the artificial defect. The contour did not change noticeably after 50 days—the seepage condition achieved the equilibrium earlier than the case for the dam without the artificial defect in the core.

Temperature Profile.

Figures 6 and 7 show the temperature profile in the dam body with and without the artificial defect in the core. The plot for 50th day after the impoundment with artificial defect shows a colder temperature at the surface due to the ambient temperature, but an almost constant shallow depth temperature that varied from 25 °C to 30 °C. When compared to the matching degree of saturation profile, the temperature profile at the foundation level below the dam body does not change significantly considering the degree of saturation is already 100%. Instead, it shows that the temperature continuously changes all the way up to 1230th day. This is due to the fact that the heat transfer takes extra time in addition to the seepage time. Of course, the layer with high permeability will allow faster water flow and the temperature change can appear faster as shown in plotting at 1000th day and 1230th day—the temperature profile at SM layer (permeability = 40 m/day) showed faster cooling effect.

Figures 6 and 7 show that the temperature profile with the artificial defect in the core is different from the one without the artificial defect in the core. The most notable difference is at 500th day after impoundment. It shows a thin but clear cold temperature zone at the exact location of the artificial defect. The temperature profiles other than 500th days did not clearly show the location of the artificial defect due to following two anticipated mechanisms.

  • (a)Before 500 days, the heat transfer to soils from seeping water was not fully developed due to the extra time required for heat transfer.
  • (b)After 500 days, this localized low temperature contrast smoothens out, implying that the contribution of concentrated seepage and associated heat flux is not high enough compared to the overall seepage and heat flux through surrounding area to cause clear contrast from the surrounding area.

This time-dependent behavior, however, is not the weakness of the continuous measurement technique. The beneficial feature of the temperature-based seepage monitoring is the ability to back-predict the time for the defect initiation. The defect zone by hydrofracturing or internal seepage may disappear, but the trace these anomalies can be identified by the thermal profiles at later days with DTS temperature sensor system installed months or years later. With the back-calculated time of incident occurrence, the condition of the dam at the time of defect initiation can be analyzed, and factors which caused the defect can be identified—a very useful capability for structural forensic science.

Figure 8 shows the comparison of the dam core temperature with the artificial defect can be more than 10 °C lower than the one without the artificial defect. These data show that the measured localized different temperature zone may indicate the possible existence of anomalies in a dam. However, it is noted that Fig. 8 itself may not tell the time in which the incident happened.

Figure 9 shows the longitudinal distribution of the temperature profile for the dam with an artificial defect in the core. It also shows measureable temperature difference around the location of the artificial defect. More importantly, it addresses that the temperature contrast in the defect zone compared to the surrounding area is very high when time is 360 days, but the contrast is not very high after that. However, the temperature in the defect zone continuously decreased, which agrees well with the explanation “b” for Figs. 6 and 7.

VS2DHI, a computer code that accommodates variably saturated–unsaturated seepage coupled with heat transport is used to analyze an artificial embankment dam in a cold region with and without the artificial defect in the dam core. The conclusions from this research are as follows:

  1. (1)The degree of saturation profile and probable location of the phreatic line after a certain time (50 days in this research) showed a clear difference between the dam with the artificial defect (0.02 m thickness) and the dam without the artificial defect. Phreatic line was not extended above this artificial defect.
  2. (2)This difference was permanent—the defect affected the stabilized phreatic line.
  3. (3)The temperature profile in the dam, however, showed a substantial time lag compared with the degree of saturation profile. It took as long as 500 days for the model dam in this study. The duration of the time delay is due to the needed time for heat transfer in addition to the time for seepage. Before 500 days, the heat transfer to soils from seeping water was not fully developed due to the extra time required for heat transfer. After 500 days, this localized low temperature contrast smooths out, implying that the contribution of concentrated seepage and associated heat flux is not high enough compared to the overall seepage and heat flux through the surrounding area to cause a clear contrast from the surrounding area.
  4. (4)The beneficial feature of this time delay behavior in temperature-based seepage monitoring is the ability to back-predict the time of the defect initiation. The defect zone may disappear after some time due to hydrofracturing or internal seepage, but the trace of these anomalies can be identified by the thermal profiles at later days with a temperature sensor system installed months or years later. With the back-calculated time of incident occurrence, the condition of the dam at the time of defect initiation can be analyzed, and factors which caused the defect can be identified—very important for structural forensic science.
  5. (5)This result showed the possibility of a new temperature-based dam health monitoring system using a spatially continuous temperature monitoring system such as distributed temperature sensors (DTS). The system may bring the best results when used with the traditional piezometers that monitor the condition of phreatic line. However, it can also be used as a stand-alone device, because it can pinpoint the location of the temperature anomalies—and it is conveniently analyzed by hydro-thermal coupled analysis.

The motivation of this research was intrigued by Dr. K. T. Chang at Kumoh University in South Korea. The reservoir water level, water temperature, and ambient temperature were provided by Dr. K. T. Chang. However, the cross section of the dam is the one appeared in Ref. [27].

Johansson, S. , 1997, “ Seepage Monitoring in Embankment Dams,” Ph.D. dissertation, Royal Institute of Technology, Stockholm, Sweden.
Kappelmeyer, O. , 1957, “ The Use of Near Surface Temperature Measurements for Discovering Anomalies Due to Causes at Depths,” Geophys. Prospect., 1(3), pp. 239–258. [CrossRef]
Merkler, G. P. , Armbruster, H. , Blind, A. , and Dösche, H. D. , 1985, “ Field Investigation for the Assessment of Permeability and Identification of Leakage in Dams and Dam Foundations,” 15th ICOLD Congress, Lausanne, Switzerland, pp. 125–141.
Stallman, R. W. , 1960, “ Notes on the Use of Temperature Data for Computing Groundwater Velocity,” Sixth Assembly on Hydraulics, Nancy, France, Report No. 3, pp. 1–7. (Reproduced in “Methods of Collecting and Interpreting Ground-Water Data,” Compiled by Ray Bentall, U.S. Geol. Surv. Water Supply Paper 1544-H, pp. 36–46.)
Birman, J. H. , 1968, “ Leak Detection Method,” U.S. Patent No. 3,375,702.
Cartwright, K. , 1968, “ Thermal Prospecting for Groundwater,” Water Resour. Res., 4(2), pp. 395–401. [CrossRef]
Cartwright, K. , 1974, “ Tracing Shallow Groundwater Systems by Soil Temperatures,” Water Resour. Res., 10(4), pp. 847–855. [CrossRef]
Pontus, S. , and Johansson, J. , 2012, “ Experiences From Internal Erosion Detection and Seepage Monitoring Based on Temperature Measurement on Swedish Embankment Dams,” Sixth International Conference on Scour and Erosion (ICSE6), Paris, Aug. 27–31, pp. 1361–1368.
Dornstädter, J. , 2000, “ Leakage Detection in Embankment Dams—Sate of the Art,” International Commission on Large Dams (ICOLD), Florence, pp. 77–86.
Aufleger, M. , Dornstadter, J. , Fabritius, A. , and Strobl, T. , 1998, “ Fibre Optic Temperature Measurements for Leakage Detection,” Applications in the Reconstruction of Dams—66th International Commission on Large Dams Annual Meeting, New Delhi, India, pp. 181–189.
Kipp, K. , 1987, “ HST3D: A Computer Code for Simulation of Heat and Solute Transport in Three-Dimensional Ground-Water Flow Systems,” Water-Resources Investigations Report, U.S. Geological Survey, Denver, CO, Report No. 86-4095.
Healy, R. , 1990, “ Simulation of Solute Transport in Variably Saturated Porous Media With Supplemental Information on Modifications to the U.S. Geological Survey's Computer Program VS2D,” Water-Resources Investigations Report, U.S. Geological Survey, Denver, CO, Report No. 90-4025.
Buckingham, E. , 1907, “ Studies on the Movement of Soil Moisture,” Bull. 38. USDA, Bureau of Soils, Washington, DC.
Richards, L. , 1931, “ Capillary Conduction of Liquid Through Porous Mediums,” J. Appl. Phys., 1(5), pp. 318–333.
Brooks, R. , and Corey, A. , 1964, “ Hydraulic Properties of Porous Media,” Hydrology, Vol. 3, Colorado State University, Fort Collins, CO, pp. 24–27.
Corey, A. T. , 1994, Mechanics of Immiscible Fluids in Porous Media, Water Resource Publication, Highlands Ranch, CO.
Van Genuchten, M. T. H. , 1980, “ A Closed-Form Equation for Predicting the Hydraulic Conductivity of Unsaturated Soils,” Soil Sci. Soc. Am. J., 44(1), pp. 892–898. [CrossRef]
Johansen, O. , 1975, “ Thermal Conductivity of Soils,” Ph.D. thesis, Norwegian University of Science and Technology, Trondheim, Norway (CRREL Draft Translation 637, 1977).
Sen, L. , Ren, T. , and Gong, Y. , 2007, “ An Improved Model for Predicting Soil Thermal Conductivity From Water Content at Room Temperature,” Soil Sci. Soc. Am. J., 71(1), pp. 8–14. [CrossRef]
Healy, R. W. , and Ronan, A. D. , 1996, “ Documentation of Computer Program VS2DH for Simulation of Energy Transport in Variably Saturated Porous Media—Modification of the U.S. Geological Survey's Computer Program VS2DT,” Water Resources Investigation Report, U.S. Geological Survey, Denver, CO, Report No. 96-4230.
Abdelkabir, M. , Bruno, B. , Michel, A. , and Mbonimpa, M. , 2013, “ Conversion of the Modified Kovács Model Parameters to the Brooks and Corey and Van Genuchten Model Parameters for the Water Retention Curve of Sandy and Silty Soils,” J. Irrig. Drain Eng., 139(5) pp. 388–398. [CrossRef]
Alavijeh, B. G. , Liaghat, A. , Huang, G. G. , and Van Genuchten, M. T. , 2010, “ Estimation of the Van Genuchten Soil Water Retention Properties From Soil Textural Data,” Pedopshere, 20(4), pp. 456–465.
Benson, C. , Chiang, I. , Chalermyanont, T. , and Sawangsuriya, A. , 2014, “ Estimating van Genuchten Parameters α and n for Clean Sands From Particle Size Distribution Data,” Geo-Congress 2014, Atlanta, GA, Feb. 23–26.
Das, B. M. , and Sobhan, K. , 2014, Principles of Geotechnical Engineering, Cengage Learning, Stamford, CT, pp. 131–135.
Campbell, G. S. , 1974, “ A Simple Method for Determining Unsaturated Hydraulic Conductivity From Moisture Retention Data,” Soil Sci., 117(1), pp. 311–314. [CrossRef]
Chang, K. T. , 2014, personal communication.
Hamil, C. A. , 2015, “ The Investigation of Okhissa Dam Using a Real-Time Monitoring System,” MS thesis, University of Mississippi, Oxford, MS.
Lu, N. , and Dong, Y. , 2015, “ Closed-Form Equation for Thermal Conductivity of Unsaturated Soils at Room Temperature,” J. Geotech. Geoenviron. Eng., 141(6), p. 04015016. [CrossRef]
Shao, M. , and Horton, R. , 1998, “ Integral Method for Estimating Soil Hydraulic Properties,” Soil Sci. Soc. Am. J., 62(3), pp. 585–592. [CrossRef]
Stankovich, J. , and Lockington, D. , 1995, “ Brooks-Corey and Van Genuchten Soil-Water-Retention Models,” J. Irrig. Drain Eng., 1(1), pp. 1–7. [CrossRef]
Song, C. R. , and Yosef, T. Y. , 2015, “ Seepage-Heat Coupled Analysis for Estimating Phreatic Line of an Earth Dam From Temperature Profile,” 28th Annual Symposium on the Application of Geophysics to Engineering and Environmental Problems (SAGEEP), Austin, TX, Mar. 22–26, pp. 252–256.
Copyright © 2017 by ASME
View article in PDF format.

References

Johansson, S. , 1997, “ Seepage Monitoring in Embankment Dams,” Ph.D. dissertation, Royal Institute of Technology, Stockholm, Sweden.
Kappelmeyer, O. , 1957, “ The Use of Near Surface Temperature Measurements for Discovering Anomalies Due to Causes at Depths,” Geophys. Prospect., 1(3), pp. 239–258. [CrossRef]
Merkler, G. P. , Armbruster, H. , Blind, A. , and Dösche, H. D. , 1985, “ Field Investigation for the Assessment of Permeability and Identification of Leakage in Dams and Dam Foundations,” 15th ICOLD Congress, Lausanne, Switzerland, pp. 125–141.
Stallman, R. W. , 1960, “ Notes on the Use of Temperature Data for Computing Groundwater Velocity,” Sixth Assembly on Hydraulics, Nancy, France, Report No. 3, pp. 1–7. (Reproduced in “Methods of Collecting and Interpreting Ground-Water Data,” Compiled by Ray Bentall, U.S. Geol. Surv. Water Supply Paper 1544-H, pp. 36–46.)
Birman, J. H. , 1968, “ Leak Detection Method,” U.S. Patent No. 3,375,702.
Cartwright, K. , 1968, “ Thermal Prospecting for Groundwater,” Water Resour. Res., 4(2), pp. 395–401. [CrossRef]
Cartwright, K. , 1974, “ Tracing Shallow Groundwater Systems by Soil Temperatures,” Water Resour. Res., 10(4), pp. 847–855. [CrossRef]
Pontus, S. , and Johansson, J. , 2012, “ Experiences From Internal Erosion Detection and Seepage Monitoring Based on Temperature Measurement on Swedish Embankment Dams,” Sixth International Conference on Scour and Erosion (ICSE6), Paris, Aug. 27–31, pp. 1361–1368.
Dornstädter, J. , 2000, “ Leakage Detection in Embankment Dams—Sate of the Art,” International Commission on Large Dams (ICOLD), Florence, pp. 77–86.
Aufleger, M. , Dornstadter, J. , Fabritius, A. , and Strobl, T. , 1998, “ Fibre Optic Temperature Measurements for Leakage Detection,” Applications in the Reconstruction of Dams—66th International Commission on Large Dams Annual Meeting, New Delhi, India, pp. 181–189.
Kipp, K. , 1987, “ HST3D: A Computer Code for Simulation of Heat and Solute Transport in Three-Dimensional Ground-Water Flow Systems,” Water-Resources Investigations Report, U.S. Geological Survey, Denver, CO, Report No. 86-4095.
Healy, R. , 1990, “ Simulation of Solute Transport in Variably Saturated Porous Media With Supplemental Information on Modifications to the U.S. Geological Survey's Computer Program VS2D,” Water-Resources Investigations Report, U.S. Geological Survey, Denver, CO, Report No. 90-4025.
Buckingham, E. , 1907, “ Studies on the Movement of Soil Moisture,” Bull. 38. USDA, Bureau of Soils, Washington, DC.
Richards, L. , 1931, “ Capillary Conduction of Liquid Through Porous Mediums,” J. Appl. Phys., 1(5), pp. 318–333.
Brooks, R. , and Corey, A. , 1964, “ Hydraulic Properties of Porous Media,” Hydrology, Vol. 3, Colorado State University, Fort Collins, CO, pp. 24–27.
Corey, A. T. , 1994, Mechanics of Immiscible Fluids in Porous Media, Water Resource Publication, Highlands Ranch, CO.
Van Genuchten, M. T. H. , 1980, “ A Closed-Form Equation for Predicting the Hydraulic Conductivity of Unsaturated Soils,” Soil Sci. Soc. Am. J., 44(1), pp. 892–898. [CrossRef]
Johansen, O. , 1975, “ Thermal Conductivity of Soils,” Ph.D. thesis, Norwegian University of Science and Technology, Trondheim, Norway (CRREL Draft Translation 637, 1977).
Sen, L. , Ren, T. , and Gong, Y. , 2007, “ An Improved Model for Predicting Soil Thermal Conductivity From Water Content at Room Temperature,” Soil Sci. Soc. Am. J., 71(1), pp. 8–14. [CrossRef]
Healy, R. W. , and Ronan, A. D. , 1996, “ Documentation of Computer Program VS2DH for Simulation of Energy Transport in Variably Saturated Porous Media—Modification of the U.S. Geological Survey's Computer Program VS2DT,” Water Resources Investigation Report, U.S. Geological Survey, Denver, CO, Report No. 96-4230.
Abdelkabir, M. , Bruno, B. , Michel, A. , and Mbonimpa, M. , 2013, “ Conversion of the Modified Kovács Model Parameters to the Brooks and Corey and Van Genuchten Model Parameters for the Water Retention Curve of Sandy and Silty Soils,” J. Irrig. Drain Eng., 139(5) pp. 388–398. [CrossRef]
Alavijeh, B. G. , Liaghat, A. , Huang, G. G. , and Van Genuchten, M. T. , 2010, “ Estimation of the Van Genuchten Soil Water Retention Properties From Soil Textural Data,” Pedopshere, 20(4), pp. 456–465.
Benson, C. , Chiang, I. , Chalermyanont, T. , and Sawangsuriya, A. , 2014, “ Estimating van Genuchten Parameters α and n for Clean Sands From Particle Size Distribution Data,” Geo-Congress 2014, Atlanta, GA, Feb. 23–26.
Das, B. M. , and Sobhan, K. , 2014, Principles of Geotechnical Engineering, Cengage Learning, Stamford, CT, pp. 131–135.
Campbell, G. S. , 1974, “ A Simple Method for Determining Unsaturated Hydraulic Conductivity From Moisture Retention Data,” Soil Sci., 117(1), pp. 311–314. [CrossRef]
Chang, K. T. , 2014, personal communication.
Hamil, C. A. , 2015, “ The Investigation of Okhissa Dam Using a Real-Time Monitoring System,” MS thesis, University of Mississippi, Oxford, MS.
Lu, N. , and Dong, Y. , 2015, “ Closed-Form Equation for Thermal Conductivity of Unsaturated Soils at Room Temperature,” J. Geotech. Geoenviron. Eng., 141(6), p. 04015016. [CrossRef]
Shao, M. , and Horton, R. , 1998, “ Integral Method for Estimating Soil Hydraulic Properties,” Soil Sci. Soc. Am. J., 62(3), pp. 585–592. [CrossRef]
Stankovich, J. , and Lockington, D. , 1995, “ Brooks-Corey and Van Genuchten Soil-Water-Retention Models,” J. Irrig. Drain Eng., 1(1), pp. 1–7. [CrossRef]
Song, C. R. , and Yosef, T. Y. , 2015, “ Seepage-Heat Coupled Analysis for Estimating Phreatic Line of an Earth Dam From Temperature Profile,” 28th Annual Symposium on the Application of Geophysics to Engineering and Environmental Problems (SAGEEP), Austin, TX, Mar. 22–26, pp. 252–256.

Figures

Grahic Jump Location
Fig. 1

Geometry of the dam

Grahic Jump Location
Fig. 2

Seasonal reservoir water temperature fluctuation

Grahic Jump Location
Fig. 3

Seasonal ambient air temperature fluctuation

Grahic Jump Location
Fig. 4

Contour of degree of saturation in the dam during initial time of simulation with (right side) and without (left side) the presence of artificial crack in the core

Grahic Jump Location
Fig. 5

Contour of degree of saturation in the dam during middle and end of simulation period with (right side) and without (left side) the presence of artificial crack in the core

Grahic Jump Location
Fig. 6

Contour of temperature distribution in the dam during initial time of simulation with (right side) and without (left side) the presence of artificial crack in the core

Grahic Jump Location
Fig. 7

Contour of temperature distribution in the dam during middle and end of simulation period with (right side) and without (left side) the presence of artificial crack in the core

Grahic Jump Location
Fig. 8

Temperature time series near the core with/without the presence of leakage

Grahic Jump Location
Fig. 9

Temperature profile along the center core wall while leaking

Tables

Table Grahic Jump Location
Table 1 Seepage and temperature field parameters used in the numerical simulation
Table Footer NoteNote: CL = clay of low plasticity (lean clay), SM = silty sand, CH = clay of high plasticity (fat clay), SP-SM = poorly graded sand with silt, SM-SC = silty sand with clay, SC = clayey sand, Ks = saturated hydraulic conductivity, RMC = residual moisture content, n = porosity, VG = Van Genuchten, α  = the reciprocal value of the intake line in moisture characteristics curve, nv = parameter for water characteristics curve slope number, KTr = residual thermal conductivity, and KTs = saturated thermal conductivity.
Table Footer NotebAlavijeh et al. [22].
Table Footer NotecBrooks and Corey [15].
Table Footer NoteeShao and Horton [29].
Table Footer NotefBenson et al. [23].
Table Footer NotegCampbell [25].
Table Footer NotehAbdelkabir et al. [21].
Table Footer NoteiLu and Dong [28].
Table Footer NotejSen et al. [19].
Table Footer NotekDas and Sobhan [24].

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In