Earth dams on complex geology without proper foundation treatment often face the risk of seepage problems. Sufficient installation and interpretation of field instruments are essential for monitoring dam behavior. Three indicators are introduced for assessment of seepage behavior: time lag (TL), pore pressure ratio (PR), and trigger water level (HW). The normalized TL reflects the washing out and plugging of rock cracks, as well as the progression of internal erosion. The foundation of the studied dam consisted of foliated rocks that were highly fractured, with the axis of the foliations aligned almost in the upstream-downstream direction, with a possible low stress zone on the syncline axis. The existing crack easily opened in the concave section of the syncline when the reservoir had risen to a curtain elevation, resulting in increased permeability and a higher flow to the downstream area, known as “hydraulic fracturing” (HF). The piezometer TL clearly indicated a shorter response time as the operating period progressed. The study dam showed the possibility of HF in the foundation, as observed during 2003–2022. The progression of HF was also confirmed by the increase in PR levels toward downstream. This revealed that the ongoing progression of HF had occurred at sta.2 + 700, which agreed well with the location of the slip zone that had occurred in 1993. HW was activated by the reservoir water level response also decreasing with time from 2003 to 2022, confirming that water infiltration through the rock crack progressed with time. These three indicators could act as good warning indices for seepage problems. This compiled knowledge could be transformed to a flowchart to identify the possible risks of hydraulic fracturing in the dam. If the three indices all showed the same trend the potential for hydraulic fracturing and internal erosion would be very high.
Research Article
Piezometer Time-Lag and Pore Pressure Ratio for Identification of Dam Internal Erosion
https://doi.org/10.21203/rs.3.rs-4815725/v1
This work is licensed under a CC BY 4.0 License
You are reading this latest preprint version
Time lag
Pore pressure ratio
Hydraulic fracturing
Internal erosion
Dam monitoring based on instruments provides essential data for the assessment of dam safety, since such information can identify probable defects and potential causes of failure. Dam safety levels can be evaluated when appropriate data are used in combination with visual inspection, testing, and computer modelling. [1–5]
Information and statistical analyses published by [6–9] showed that about 30–50% of earth dam failures are the result of piping and internal erosion. However, internal erosion may reach an advanced stage before there are any visible signs. In such cases, monitoring data, indicating high seepage rates and high pore pressure, are the first warning signs of seepage erosion. Proper interpretation of instrument data can be the first evidence of abnormal dam performance. [10–13] Furthermore, good interpretation can predict the level of dam defects and the estimated time to failure. [14-16]
Leakage detection using the seepage monitoring data method has been interpreted on many occasions [17-24, 3,] This approach is based on time lag and seepage length. Very few studies have evaluated dam long-term behavior based on piezometer time lag, pore pressure ratio, and trigger level.
An earthen dam in northern Thailand was used in the current case study. This dam had been evaluated based on semi-quantitative risk assessments in 2020 [25]. The results indicated there was a possible risk of concentrated seepage and internal erosion. Currently, corrective maintenance for this dam is ongoing. Previously, a concentrated leak in the downstream area was observed in a period of high reservoir water levels. [25-29], reported that the seepage problems were due to the complex geology of the dam foundation, where stress arching in rock foliation had initiated hydraulic fracturing (HF) and internal erosion. The HF was triggered when the reservoir water level reached the curtain level, at which stage, the water pressure started to open the rock crack and the seepage flow increased [27, 30-32]. This behavior was detected based on readings from piezometers installed in the foundation, using the lag time and pore pressure ratio.
Once HF had commenced, a continuous flow path gradually formed, resulting in more rapid seepage and the time lag (TL) from the reservoir water rising until the response based on the piezometer readings became shorter. In addition, the head loss along the seepage path decreased as a result of increasing permeability. The ratio of the piezometer head increase to the rise in the reservoir water is defined as the pressure ratio (PR). Therefore, a decreasing TL and an increasing PR can be good indicators of the progression of internal erosion. Variation in these two parameters can be used as indices for the evaluation of seepage erosion in the dam [20-21].
The long- term seepage behavior of an instrument-monitored earth dam can be detected based on piezometer responses [25, 33-34]. TL, PR and the trigger level are three parameters used for evaluating the potential of internal erosion. With an established dam instrument database, a dam owner can carry out systematic seepage evaluation for a group of dams. The results can be a part of dam safety evaluation and included in the maintenance program for dam safety management.
2.1 Time lag
The piezometer response normally shows a time lag from the rise in the reservoir water level. Furthermore, it takes longer for the piezometer head to sense the peak or steady state, as shown in Fig. 1. This delay period is called “time lag or lag time” [3].
The two components of time lag in Fig. 2 are: 1) “traveling time”, which is the time for the pressure change in the reservoir to travel to the area surrounding the piezometer tip; and 2) “activated time”, which is the time for the water pressure to change around the tip flow into the piezometer sensor or measuring system. The traveling time is dependent on the seepage distance and the soil permeability along the seepage path. The activated time lag is dependent on the intake area of the filter element and the volume of water required to flow in or out of the tip to register a pressure change.
The basic equations for the determination of time lag are based on Darcy’s Law. Basic traveling and activated time lags for a standpipe piezometer, as shown in Fig. 2 can be defined using Eq. 1. and Eq. 2. [17]
where Ls is the seepage length for each soil, k is the soil permeability of each soil, F is a factor which depends on the shape and dimensions of the filter, and A is the cross-sectional area of the standpipe.
A normal long seepage length in a dam results in a long traveling time, while based on recent sensitive piezometers, the activated time is very short, so that for simplification, Tactivated can be ignored.
2.2 Pore pressure ratio
The pore pressure ratio or PR is the proportion of the reading at the piezometer head to the reservoir water level (RWL), indicating the remaining pressure after seepage loss along the seepage path. If PR = 1, then water can easily seep through the cracks or erosion channel to the piezometer tip without any head loss. The pore pressure ratio can increase with time as internal erosion progresses [20], [26], [29]. PR can be calculated based on the slopes of the plots between the piezometer reading and the rising reservoir water level, as shown in Fig. 3a). [25]. used PR to indicate the level of the head loss from upstream to downstream as detected by piezometers at different locations, as shown on Fig. 3b).
2.3 Study dam
The study dam was a saddle dam in a large hydropower project that was constructed during 1968 to 1972. The dam was placed directly on stripping ground with practically no foundation treatment. The upstream residual soil was assumed to acts as a natural blanket. The crest elevation and maximum dam height were 168 and 30 m. MSL, respectively. The geology of the dam consisted of quite complex metamorphic rocks. Intrusion of nearby igneous rocks had created lateral tectonic forces, resulting in rock foliation, as shown in Fig. 4. The foliation structures were aligned in a northeast-southwest direction and almost perpendicular to the dam axis. The dam foundation consisted of residual soil, phyllite, and quartz schist. [35]
[26] developed a longitudinal profile along the dam axis using bore holes and field investigation, as illustrated in Fig. 6. The current study used geological information from previous investigations, outcrop inspection and mapping, regional geology, and a core log.
[28] used finite element method to model the hydraulic fracturing (HF) of the dam. The results indicated that the potential for HF mainly occurred at the bottom of the syncline on the residual soil and phyllite layers, which were steep, narrow structures, resulting in low minor stress (σ3). Where the water pressure could overcome the perpendicular normal stress, the cracks increased and had higher permeability. The field outcrop investigation and regional geology revealed that the foliation axes were aligned almost perpendicularly to the dam axis. [35], [19].
Cross sections were considered based on the location of the HF occurrence in the deepest foliation at sta. 2 + 700, as indicated in Fig. 6.
2.4 Seepage incident
The dam was impounded in 1972 with no dam instrumentation installed. Later seepage points were observed at the downstream toe in 1994. Subsequently, a slip surface on the downstream dam slope developed above the wet areas, as shown in Fig. 5. During that time, the reservoir was filled above the normal high-water level (162.00 m. MSL.) due to flooding. Emergency repair work was done by laying geotextile and a rock berm on the downstream toe. However, high pore pressure and wet zones in downstream area were still recorded by piezometers. Several piezometers were installed on the foundation at locations shown in Fig. 6.
Currently, the piezometers records about 70% of reservoir pressure at the dam axis, which gradually reduces toward downstream, as shown in Fig. 3b). This reading is triggered when the reservoir levels exceed 148.00–150.00 m MSL.
Field piezometer records from 2003 to 2022 in the residual soil and phyllite layers were collected and analyzed for TL and PR.
3.1 Time lag evaluation
The RWL and piezometer response for every peak flood are plotted in Fig. 7A. From the initial time at which the reservoir level rose or fell, the piezometer reacted with a time delay (time lag). The RWL reached the peak pressure first, with the piezometer response recording the peak later at the lower pressure. The two graphs were normalized by assigning both peaks as 100% of the pressure increase. Then every level of pressure increment was lotted as a percentage of each peak to obtain the normalized graphs, as shown in Fig. 7B. Then TL was measured on the time-axis at every 10% of pressure increment.
3.2 Analysis of pore pressure ratio
PR was calculated as the ratio of the piezometer level to RWL. A statistical trend line was fitted to each part of graph based on regression equations. The slope of the linear fit after the trigger point (Hw) is PR, as shown in Fig. 8.
Time lag (TL), pore pressure ratio (PR), and pressure trigger level (HW) can be classified to several risk levels. Then, decision rules from monitoring data can be applied according to the previous study by [26–29]. These rules can serve as partial decision factors for risk assessment of internal erosion and HF for the foundation of an earthen dam. However, an existing dam with sufficient piezometer readings and TL, PR, and HW data can be initially evaluated for internal erosion and HF.
4.1 Time lag versus internal erosion
Several piezometers were installed in the dam foundation on critical cross sections. The normalized time lag (TL) in the residual soil above the phyllite layer is shown in Fig. 9 and Fig. 10. The readings from 2003 to 2022 indicated the progression of erosion as TL gradually decreased, which could be divided into three periods: 2003–2008, 2008–2013, and 2013–2022. The maximum TL of OSP10 at the offset distance 15 m. to the downstream area provides a good example—it decreased from 50 to 20 to 0 days for periods 1–3 respectively. These values indicated that rock/soil cracks had gradually opened up, resulting in high permeability and possible internal erosion. In addition, in the dam axis area, the TL values of OSP8 and 8/1 indicated very shorter time lags of less than 10 days. The erosion in this area toward upstream might have occurred before the first installation of piezometers.
The potential for erosion and plugging was detected along the seepage paths in the residual soil layer, as shown in Fig. 10. The first period of erosion and plugging behavior is shown in Fig. 10A. Specifically, on OSP10 at the offset distance 15 m. to downstream, there was plugging as TL increased up to 50 days. Then, during the second period (Fig. 10B), the maximum TL reduced from 50 to 14 days. Finally, in the third period, TL was nearly 0 days, with no sign of the plugging area (Fig. 10C). These results indicated that soil cleavages could open with increasing seepage pressure, known as “hydraulic fracturing” (HF). Subsequently, internal erosion occurred through the seepage path. However, TL in the phyllite layer (Fig. 11) showed no decreasing trend related to internal erosion.
The TL values at different locations fluctuated with the reservoir water level. Figure 10 shows the TL values at OSP8, OSP8/1, and OSP10 in the residual soil. During the RWL increase from + 148 to + 162 m. MSL, TL presented alternate high to low time lags, revealing the alternate plugging and erosion of the surrounding soil. These phenomena proceeded until all the filler material eroded out, resulting in a continuous increase in permeability. In the phyllite layer (Fig. 11), because the permeability was higher than in the residual soil. A reduction in the time lag is not clearly shown, with the exception being in the lower part of the residual soil, where TL is very low when RWL exceeds + 158.5 m. MSL, indicating the erosion progression, as shown in Fig. 12. At OSP 10/1 and OSP11/1, as shown in Fig. 13, there was no apparent internal erosion until the RWL increased to + 160.0 m. MSL. These results agreed with [27–28] who reported that the HF on a study dam started from the lower hinges of the residual soil in the syncline area.
4.2 Pore pressure ratio versus internal erosion
Figure 14 provides a plot of PR in the foundation layer. The highest PR values were located mostly on the upstream and lower parts of the residual soil. The pressures had no impact until reaching the trigger level of + 148 to + 149 m. MSL. The high range for PR was 0.95–0.70, indicating. These indicated hydraulic fracturing and internal erosion phenomena. In the phyllite layer at a distance of 45 m. from the dam axis to downstream, the PR values were lower (0.50–0.15), which are considered as normal seepage, compared to the analyses reported by [27]
[29] proposed risk levels of internal erosion using PR, as shown in Fig. 15. The contour maps of internal erosion indicated by PR levels show the progression with time in Fig. 16.
During the first monitoring period from 2003 to 2022, the level of internal erosion of the very likely (red-shaded) area located upstream to about the dam axis is shown in Fig. 16a. The advancement proceeded to the downstream area during the second and third periods (Fig. 16b and c). The greatest increase in HF was at Sta. 2 + 700, where the rock fractures gradually opened and erosion developed along this path, which corresponded to the old stream channel. In addition, the field observations showed that the downstream area at sta.2 + 650 to 2 + 700 had wet areas when the reservoir level rose above 160 m. MSL. This result was in agreement with [36], who used ground aeration sound, showing the possibility of a high seepage flow zone.
The trend in internal erosion can be observed in Fig. 17. Piezometer OSP8 had PR values in the range 0.9–1.0 from 2003 to 2022. Seepage from upstream to the dam axis had almost no head loss. The PR values at OSP 8/1 and OSP 10 from 2003 to 2022 increased from 0.7 to 0.8 (OSP 8/1) and from 0.6 to 0.8 (OSP 10). In contrast, at OSP11/2 located near downstream, the PR value remained constant at 0.6, indicating no erosion. Soil particles could migrate from the upstream area to downstream and at the curtain distance in the downstream area, resulting in plugging and self-filtering. This blocking area was temporary and tended to move downstream.
4.3 Triggered water levels versus internal erosion
Linear regression was used to evaluate the yearly triggered water levels (HW), as they related to the level of internal erosion. In general, the piezometer readings showed triggered levels at 148.00–150.00 m. MSL, as shown in Fig. 18. The HW values from the dam axis to about 45 m downstream presented a decreasing trend with time, as shown in Figs. 19 and 20. There were decreasing values of ΔHw = 1.6 m, 0.8 m, 0.5m, and 0.2m for OSP8, OSP8/1, OSP10, and OSP11/2, respectively, 2003 to 2022. The Hw values of the piezometers located in the phyllite (Fig. 20) did not decrease with time, indicating that seepage in the residual soil layer flowed more easily than in the phyllite. As the cracks opened, the trigger levels decreased. OSP 8 was vulnerable to erosion because the trigger level decreased about 1.6 m., as shown in Fig. 18A.
The changes in HW can be used as criteria for determining HF occurrence in the study case. Four levels of HF could be identified: HW decreasing: >1 m. (very likely), 0.25–1 m. (likely), < 0.25 m. (unlikely), and constant (very unlikely).
The relationship between HW and PR is presented in Fig. 21. The residual soil layer had a greater seepage problem than the phyllite layer, since it had a lower HW with a higher PR. In the residual soil, the range in HW was from + 148 to + 150 m. MSL. with a PR range from 0.7 to 1.0, whereas in the phyllite layer, HW is the range from 149 to + 156 m. MSL, with low PR values in the range from 0.2 to 0.6.
4.4 Indication criteria for internal erosion
FEM was applied to set benchmarks for TL and PR based on piezometer readings for normal seepage. The results are shown in Fig. 22, where the average TL piezometer value near the dam axis was 100 days and the PR values were in the range 0.53–0.62. The TL values recorded between 2003 and 2022 were much lower than 55 days, compared to the benchmark. In addition, the PR range of 0.73–1.00 was higher than the benchmark. All the TL and PR data were analyzed based on regression equation.
The percentage decreases in TL and PR are plotted in Fig. 23. Based on this information, the risk levels for HF occurrence were proposed for the very likely, likely, unlikely, and very unlikely scenarios, according to [29]. The risk criteria are presented in Table 1, classified according to the three main indicators (PR, TL, and HW).
The criteria in Table 1 can be used to construct a decision tree for use in an expert system, which would be beneficial for dam owner monitoring a large number of dams. The decision tree could also combine the dam database to monitor the risk condition of each dam after each piezometer reading.
Table 1. TL, PR, HW criteria for HF risk assessment
| Risk Level | PR [29] | TL* Decreasing | HW** Increasing |
|---|---|---|---|
| Very unlikely | 0.0–0.3 | < 5% | Constant |
| Unlikely | 0.3–0.6 | 5–50% | < 0.25 m. |
| Likely | 0.6–0.8 | 50–80% | 0.25–1 m. |
| Very likely | 0.8–1.0 | > 80% | > 1 m |
Note
TL* is percentage decrease from analytical or first reading after construction. HW** is head decrease from first reading after construction. The values are applied from study case only.
Based on this study, the following conclusions can be drawn:
1. Dam piezometer readings can be used for interpretation of potential for internal erosion and hydraulic fracturing. Three parameters from piezometer readings were analyzed: time lag (TL), pore pressure ratio (PR), and reservoir trigger level (HW).
2. The study dam was a saddle type forming part of a large hydropower project and was constructed during 1968–1972. After 22 years of operation, the dam experienced a sliding mass above the dam toe with high pressure and wet areas.
3. After emergency repairs using a downstream rock berm, piezometers were installed to monitor the pressure in the dam foundation. High pressure readings were observed in the lower part of the residual soil just above the underlying phyllite.
4. The time lag (TL) behavior could be divided into three periods. During the first period from 2003 to 2008, TL values remained high (up to 50 days). During the second period from 2008 to 2013, the TL values reduced substantially to less than 12 days. During the third period from 2013 to 2022, the TL values were close to 0 days, indicating that along the seepage path, the rock/soil cracks were continuously open, resulting in greater permeability and possible internal erosion.
5. The TL values at the different locations fluctuated with the reservoir water level, reflecting the potential of erosion and plugging effects along the seepage paths.
6. The pore pressure ratio (PR) was very high in the residual soil from upstream to 15 m. downstream. There was a high range in the PR values (0.7–0.95) compared to the FEM analytical range (0.5–0.6), indicating that the seepage resistance had reduced and internal erosion may have also lessened.
7. The expansion of the hydraulic fracturing zone plotted using PR-contours showed advancement in the downstream direction. The greatest advancement was close to the old buried stream channel at about Sta. 2+700.
8. The values for the trigger water level (HW) in the potential internal erosion risk zone tended to decrease with time. On risk zone of study dam the decreasing of 1.5 m. was founded.
9. The time lag decrease (in days) determined using the analytical time lag versus the pore pressure ratio could be applied to establish the risk zones of internal erosion/hydraulic fracturing. This process indicated that the study dam shows was likely to very likely to experience hydraulic fracturing.
10. TL, PR, and HW can be compiled to provide HF risk criteria and a decision tree. The decision tree could be applied in an expert system for use in monitoring a group of dams, with dam owners who operate many dam benefitting by making real-time judgments of the risk situation of each dam.
Competing interests:
The authors declare no competing interests.
Author Contribution
Assist. Prof. Bunpoat Kunsuwan, M.Eng: Conception, implementation and supporting of the modeling, interpretation and validation of results, reviewing, suggestion, visualization. Wiphada Thepjanthra, Ph.D.: Wrote the main manuscript text, verified the analytical methods, designed the computationalframework, and analysis and interpreting the results.Assoc. Prof. Warakorn Mairaing, Ph.D.: Conceived the study, conception, methodology, suggestion, supervision, critical reviewing and final approval of the version to be submitted. Montri Jinagoolwipat, Dam Engineer: Support the monitoring data, reviewing, and suggestionThawatchai Chalermpornchai: verified the analytical methods, designed the computationalframework, and aided in interpreting the results.Methodology, Software analysis, investigation, interpretation of results, writing and drafting the journal manuscript.
Acknowledgement
The authors are thankful for the data support and suggestions provided by Research Center for Sustainable Infrastructure Engineering, Kasetsart University and Electricity Generating Authority of Thailand. Thanks, the Research and Development Institute at Rajamangala University of Technology Suvarnabhumi for its support in the budgeting and publishing of this research.
- IEC (1985) Analysis techniques for system reliability-Procedure for failure mode and effects analysis, Geneva, Switzerland
- Peyras L, Royet P, Boissier D (2011) Dam ageing diagnosis and risk analysis: Development of methods to support expert judgment. Can Geotech J 43:169–186. 10.1139/t05-096
- Shao-wei Wang Ying-li, Xu Chong-shi, Gu Teng-fei, Bao (2018) Monitoring models for base flow effect and daily variation of dam seepage elements considering time lag effect. Water Sci Eng 11(4):344–354. 10.1016/j.wse.2018.12.004
- Zhen-xiang, Jiang H, Chen (2022) A new early warning method for dam displacement behavior based on non-normal distribution function. Water Sci Eng 15(2):170–178. 10.1016/j.wse.2022.04.001
- Zhao M, Liu P, Jiang L, Wang K (2021) The Influence of Internal Erosion in Earthen Dams on the Potential Difference Response to Applied Voltage. Water 13(23). doi.org/10.3390/w13233387
- Foster Mark F, Robin, Matt S (2000) The statistics of embankment dam failures and accidents. Can Geotech J 37(5):1000–1024
- Zhang LM, Xu Y, Jia JS (2007) Analysis of earth dam failures - A database approach. Paper presented at ISGSR2007 First International Symposium on Geotechnical Safety and Risk, Shanghai Tongji University, China, 18–19 October 2007
- ICOLD (2019) ICOLD Incident Database Bulletin 99 Update: Statistical Analysis of Dam Failures. FranceBulletin 188, Committee on Dam Safety. International Commission on Large Dams, Paris
- ICOLD (2020) General Synthesis, World Register of Dams. https://www.icold-cigb.org/GB/world_register/general_synthesis.asp
- Fu CJ, Yao XW, Li T, Shen HY, Wang ZY, Jiang JQ (2014) Investigation and evaluation of increasing uplift pressure in an arch dam: A case study of the Huaguangtan Dam. KSCE J Civil Eng 18(6):1858–1867. 10.1007/s12205-014-0432-
- Su HZ, Chen ZX, Wen ZP (2015a) Performance improvement method of support vector machine-based model monitoring dam safety. Struct Control Health Monit 23(2):252–266. 10.1002/stc.1767
- Su HZ, Hu J, Yang M (2015b) Dam seepage monitoring based on distributed optical fiber temperature system. IEEE Sens J 15(1):9–13. 10.1109/JSEN.2014.2335197
- Eduardo E, Alonsoa NúriaM, Pinyol (2016) Numerical analysis of rapid drawdown: Applications in real cases. Water Sci Eng 9(3):175–182
- Harbi Leyla S, Nadia R Bouchrit (2023) Modeling and predictive analyses related to piezometric level in an earth dam using a back propagation neural network in comparison on non-linear regression. Model Earth Syst Environ 9:1169–1180
- Joaquim Tinoco M, de Granrut D, Dias TM, Alexandre-Gilles Simon (2019) Piezometric level prediction based on data mining techniques. Neural Comput Appl 32:4009–4024
- Fernando Salazar R, Morán, Miguel Á, Toledo E Oñate (2015) Data-Based Models for the Prediction of Dam Behaviour: A Review and Some Methodological Considerations. Arch Comput Methods Eng 24:1–21
- ICOLD (1951) Time lag and soil permeability in ground-water observations Bullentin No. 36. U. S. Army, Mississippi
- Stéphane Bonelli P, Royet (2008) Delayed response analysis of dam monitoring data. Paper presented at ICOLD European Symposium on Dams in a European Context, GEIRANGER, Norway. 1–4 April 2008
- Bahar Razavi1, Parehkar M, Ali Gholami (2011) Investigation on Pore Water Pressure in Core of Karkheh Dam. Eng Technol Int J Civil Environ Eng 59(8):2417
- Thongthamchart Chinoros, Brohmsubha P (2014) The Safety Criteria for Geotechnical Instruments on the Internal Erosion in Embankment Dams. Paper presented at International Symposium in ICOLD 2014 Annual Meeting, Bali, Indonesia, 1–6 June 2014
- Richard S, Olsen, Isaac J, Stephens (2016) Relearning how to look at piezometric data for seepage evaluation. Paper presented at USSD 2016 conference, Denver, 11–15 April 2016
- Wang SW, Bao TF (2013) Monitoring model for dam seepage based on lag effect. Applied Mechanics and Materials 353–356: 2456–2462 doi.10.4028/www.scientific.net/AMM.353-356.2456
- Rajkumar Prasad M Dixit (2020) Performance Monitoring of Dams through Piezometers - A Case Study. Int J Eng Appl Sci 7(2)
- Haghighi AT, Tuomela A, Ali Akbar Hekmatzadeh (2020) Assessing the Efficiency of Seepage Control Measures in Earthfill Dams. Geotech Geotech Geol Eng 38:5667–5680. 10.1007/s10706-020-01371
- Geotechnical Engineering Research and Development Center (2019) Final Report, Semi-Quantitative Risk Assessment on Stability of Saddle Dikes of Sirikit Dam. Electricity Generating Authority of Thailand, Bangkok
- Research Center for Sustainable Infrastructure Engineering (2020) Design Report Slurry Cutoff Wall Saddle Dike 4. Kasetsart University, Bangkok
- Chalermpornchai T, Kunsuwan B, Mairaing W (2021) Simulation of Rock Crack and Permeability in Dam Foundation During Hydraulic Fracturing. Int J GEOMATE 21(86):55–62
- Kunsuwan B, Chalermpornchai T, Mairaing W, Thepjunthra W (2023) Assessment of Hydraulic Fracturing in Earth Dams on Complex Foundations. J Disaster Res 18(3):270–279
- Saejiaw W, Kunsuwan B, Mairaing W, Chalermpornchai T (2023) Evaluation of Hydraulic Fracturing Phenomena in Earth Dam. UBU Eng J 16(1):44–56
- Wang JJ (2014) Hydraulic Fracturing in Earth-Rock Fill Dams. Wiley, New York
- Salari M, Akhtarpour A, Ekramifard A (2021) Hydraulic Fracturing: A Main Cause of Initiating Internal Erosion in a High Earth- Rock Fill Dam. Int J Geotech Eng 15(2):207–219
- James L, Sherard (1986) Hydraulic Fracturing in Embankment Dams. J Geotech Eng 112(10):905–927
- Mairaing W (1999) Earth Dam Engineering. Department of Civil Engineering. Kasetsart University, Bangkok
- FEMA (2015) Evaluation and Monitoring of Seepage and Internal Erosion. Interagency Committee on Dam Safety (ICODS), Washington
- Singharajwarapan S, Berry R (2000) Tectonic Implications of the Nan Suture Zone and its Relationship to the Sukhothai Fold Belt, Northern Thailand. J Asian Earth Sci 18(6):663–673
- Apiniti et al (2014) Preliminary results of Ground Aeration Sound (GAS) survey for SRK dike-4. Electricity Generating Authority of Thailand, Bangkok
No competing interests reported.