Content uploaded by C. H. Scholz
Author content
All content in this area was uploaded by C. H. Scholz on Oct 28, 2017
Content may be subject to copyright.
A preview of the PDF is not available
Two types of reservoir-induced seismicity
Abstract
The temporal distribution of induced seismicity following the filling of large reservoirs shows two types of response. At some reservoirs, seismicity begins almost immediately following the first filling of the reservoir. At others, pronounced increases in seismicity are not observed until a number of seasonal filling cycles have passed. These differences in response may correspond to two fundamental mechanisms by which a reservoir can modify the strength of the crust<>9d\one related to rapid increases in elastic stress due to the load of the reservoir and the other to the more gradual diffusion of water from the reservoir to hypocentral depths. Decreased strength can arise from changes in either elastic stress (decreased normal stress or increased shear stress) or from decreased effective normal stress due to increased pore pressure. Pore pressure at hypocentral depths can rise rapidly, from a coupled elastic response due to compaction of pore space, or more slowly, with the diffusion of water from the surface. -Authors
... With the construction of reservoirs, researchers found that impoundment and periodical variations in water levels are strongly correlated with seismicity in the reservoir area. Seismic activity in the reservoir and surrounding areas became abnormal after impoundment, and in some cases, it even increased [1][2][3][4] (Carder, 1945;Simpson et al., 1988;Talwani, 1997;Cao et al., 2015). The earliest seismicity associated with reservoir impoundment occurred near the Kremasta Reservoir in north Greece [5]. ...
... With the construction of reservoirs, researchers found that impoundment and periodical variations in water levels are strongly correlated with seismicity in the reservoir area. Seismic activity in the reservoir and surrounding areas became abnormal after impoundment, and in some cases, it even increased [1][2][3][4] (Carder, 1945;Simpson et al., 1988;Talwani, 1997;Cao et al., 2015). The earliest seismicity associated with reservoir impoundment occurred near the Kremasta Reservoir in north Greece [5]. ...
The Baihetan Reservoir is built for hydropower in China. The rise of the reservoir water leads to a series of earthquakes in the surrounding area. This study proposes fully coupled equations of pore-viscoelasticity and a parallel partition mesh model to study the short- and long-term effects of the Baihetan Reservoir and further calculate the changes in stress, pore pressure, and Coulomb failure stress with time on the major faults. Based on the calculation results, impoundment increases regional seismicity, which is consistent with the seismic catalog. The reservoir impoundment causes an increase in pore pressure in the crust, primarily enhancing Coulomb failure stress beneath the reservoir center. This effect extends to approximately 60 km in length and 20 km in width at a depth layer of 5–10 km. Seismicity varies greatly among different faults. Coulomb failure stress increases on the northern part of the Xiaojiang Fault and Zhaotong-Ludian Fault, and decreases on the southern part of the Xiaojiang Fault and Zemuhe Fault. The Coulomb failure stress is highly correlated with the number of earthquakes along the Xiaojiang Fault. The influence of the reservoir on the local seismicity is mainly limited to several months, and it has a slight effect later on. The focal depth of the induced earthquakes increases while the magnitude decreases. The earthquakes caused by the impoundment all have a small magnitude, and the Ms4.3 Qiaojia earthquake on 30 March 2022, was more likely a natural event.
... The influence of reservoir impoundment on seismicity is complex and depends on reservoir operation, regional tectonic setting, and hydromechanical properties of the rock mass. The underlying physical processes typically include the loading effect of impounded water and pore fluid phenomena (Simpson et al., 1988;Talwani, 1997;Gupta, 2002;Zhang et al., 2021). ...
... Beneath the riverbed, all areas generally correspond to high V P and low V S values, except for the head area of the XJB Reservoir, which exhibits characteristically low V P and low V S values. Gupta (2002) synthesized opinions from Simpson et al. (1988) and Talwani (1997) to categorize RIS into three types based on the occurrence time: rapid response seismicity, delayed response seismicity, and continued seismicity. Rapid response and delayed response earthquakes are associated with the initial filling or impoundment of a given reservoir. ...
The Xiluodu Reservoir (XLD), the third hydropower station developed downstream of the Jinsha River, is currently the fourth largest hydropower station worldwide. The reservoir has been impounded for 10 years, and numerous earthquakes have occurred, thus indicating an increased seismic risk. In existing studies, the deep tectonics and fine-scale structure in the reservoir area have not been sufficiently explained. Thus, the mechanism underlying reservoir-induced seismicity in this area remains unclear. In this study, we utilized a microseismic catalogue compiled based on the LOC-FLOW workflow (Li et al., 2023), alongside long-term observation data from a dense seismic array in the Jinsha River downstream, to obtain high-precision 3D V P , V S and V P /V S structures, as well as refined seismic relocations in the XLD Reservoir area. Additionally, we incorporated petrological, geomagnetic, gravity, crustal density, geodetic observation and numerical simulation data, which demonstrate that seismicity in the reservoir head area and near the Manao Fault is influenced by fluids. Moreover , the two M4 earthquakes that occurred east of the Lianfeng Fault are indicative of tectonic activity. Of these earthquakes, the Tianba earthquake cluster in the reservoir head area is located in a high-V P /V S zone, while a low-V S anomaly channel is observed here, which suggests that large amounts of fluid infiltrated along the fault, increasing the pore pressure and reducing the effective stress on the fault. The Wuji earthquake cluster in the head area is also located in a low-V P /V S zone, which indicates a low porosity and possibly undrained conditions. Thus, the earthquakes in this area are the result of coupled elastic stress-pore water pressure interactions.
... [32] presented a framework to model surface deformation and forecast induced seismicity for reservoir operations. [33] - [40], reviewed challenges offered by reservoir induced seismicity. [41] studied regional seismicity of reservoir area using digital seismometric observations. ...
In this article, impact of climatic anomalies and artificial hydraulic loading on earthquake generation has been studied using federated learning (FL) technique and a model for the prediction of earthquake has been proposed. Federated Learning being one of the most recent techniques of machine learning (ML) guarantees that the proposed model possesses the intrinsic ability to handle all concerns related to data involving data privacy, data availability, data security, and network latency glitches involved in earthquake prediction by restricting data transmission to the network during different stages of model training. The main objective of this study is to determine the impact of artificial stresses and climatic anomalies on increase and decrease in regional seismicity. Experimental verification of proposed model has been carried out within 100 km radial area from 34.708o N, 72.5478o E in Western Himalayan region. Regional data of atmospheric temperature, air pressure, rainfall, water level of reservoir and seismicity has been collected on hourly bases from 1985 till 2022. In this research, four client stations at different points within the selected area have been established to train local models by calculating time lag correlation between multiple data parameters. These local models are transmitted to central server where global model is trained for generating earthquake alert with ten days lead time alarming a specific client that reported high correlation among all selected parameters about expected earthquake.
... The elastic stress resulting from the water load leads to pore compression in the bottom rocks of the reservoir, causing an increase in pore pressure and a reduction in rock strength, which in turn triggers fast-response type earthquakes. In the case of delayed response types, their activity and lag time are associated with the diffusion of pore water pressure (Simpson et al., 1988). Research has indicated that the mechanism of reservoir-induced seismicity comprises reservoir load and pore pressure diffusion categories, with reservoir load mechanisms typically leading to fast responses and pore pressure diffusion mechanisms tending to induce delayed responses (Kinscher et al., 2023). ...
High-resolution three-dimensional VP, VS and VP/VS images in the Houziyan Reservoir Area were obtained by using Fast Marching Tomography Package (FMTOMO) with the travel time data from 6330 seismic events monitored by the Houziyan Reservoir Seismic Network. This analysis yielded the 3-D velocity structure, including longitudinal wave velocity (VP), shear wave velocity (VS), and the ratio of longitudinal and shear wave velocity (VP/VS) at different impoundment stages. The data changes at various impoundment times, depths of sections, and directions of profiles were analyzed to obtain these results. The final findings demonstrate the following results: 1) Through tomographic analysis, it was determined that the underground velocity structure in the Houziyan reservoir area was anisotropic before impoundment. 2) The area of high wave velocity increases in stage 1, stage 3, and stage 4. The area of low wave velocity increases in stage 2, especially in depth, indicating significant changes in the underground velocity structure at different impoundment stages. 3) Compared to the changes in underground velocity structures in other reservoirs after impoundment, the Houziyan reservoir exhibited a unique pattern. 4) In general, the underground velocity structure displayed an overall increasing trend after impoundment. However, it also exhibited instances of decreasing velocity, reflecting continuous dynamic adjustments to the underground velocity structure after impoundment. These conclusions highlight the impact of impoundment in the reservoir area on the underground velocity structure and provide scientific theoretical support for seismic risk assessment following impoundment in the reservoir area.
... Because of the long history and well-researched hydropower, it has become a mature, predictable and price-competitive technology [20]. It is currently the most widely used renewable energy in human society [21]. Among all known energy sources, hydroelectric power has the highest conversion efficiency -90% and its power generation does not require any fuel, and the cost will not be affected in any way. ...
Since the construction of the first thermal power plant in 1875, mankind has achieved great success in the field of energy. Today, the pursuit of clean energy is also a major breakthrough. Through examples and comparisons, this paper classifies human energy by means of power generation, and expounds the evolution, history and development experience of energy. Mankind is now entering the era of nuclear energy on a large scale, which is an important node in the history of power generation. This article summarises the past, reviews the era of nuclear energy, and discusses the future development of energy. In the history of energy evolution, from steam power to clean energy, mankind has experienced constant innovation and progress. The advent of the nuclear energy era has provided mankind with more efficient and sustainable energy choices, and also led the development direction of future energy.
... 12,43 Relevant studies show that friction coefficient for pure clays may be <0.2 under water action if the fractures were initially dry. 44,45 This increased pore pressure (P 1 ) and decreased friction coefficient (μ 1 ) causes a clockwise rotation of the Mohr-Coulomb criterion line. The final result during reservoir filling is that the Mohr circle intersects the Mohr-Coulomb criterion line, and therefore result in additional earthquakes ( Figures 5B & 8B). ...
The transition to low-carbon energy generation requires a diverse portfolio of energy sources. Large hydropower stations play a crucial role in providing eco-friendly energy but can lead to crustal deformation and seismic activity due to reservoir impoundment and discharge. The mechanism behind how impoundment-induced deformation influences reservoir-triggered seismicity remains unclear. We investigate this at the world’s 4th largest hydropower station in Xiluodu, China, using interdisciplinary analyses of relocated earthquakes, geodetic data, and numerical simulations. Geodetic observations confirm both a linear elastic response to hydrologic loading and the transition towards a relaxed and flexural response. Notably, approximately 81% of the earthquakes since impoundment occurred within ~15 km of the subsidence center. These events temporally relate to rapid hydrologic loading/unloading periods, with their spatiotemporal outward migration potentially caused by viscous relaxation of load stresses in the lower crust and upper mantle. The reservoir-induced flexural deformation and subsequent viscous relaxation may control the distribution and expansion of triggered seismicity. This research highlights the urgent need to enhance geohazard monitoring efforts around the subsidence center and emphasizes the vital importance of carefully considering this factor during the construction of large reservoirs to mitigate potential seismic risks.
... Furthermore, the occurrence and distribution of seismicity in time and space were found to follow complex patterns: earthquakes may occur at the early stages of reservoir impoundment or be delayed by several years or decades 2 . The latter case commonly corresponds to deeper and larger seismic events 4 . These complexities in the spatiotemporal distribution of the seismic events depend primarily on the water level in the reservoir, reservoir filling rate, water level fluctuations, in situ stress state and hydrogeological characteristics of the underground formations, among others [13][14][15] . ...
Reservoir-triggered seismicity commonly occurs as a result of reservoir impoundment. In particular, the Nova Ponte reservoir triggered a series of earthquakes, including the 1998 M4.0 earthquake, which represents the second-largest earthquake triggered by reservoir impoundment in Brazil. The earthquake occurred after prolonged seismic activity following reservoir impoundment starting in 1993. After more than two decades, the mechanisms governing these earthquakes and their relation with the spatiotemporal evolution of the seismic events are still poorly understood. Here, we explain the causal mechanisms of the two largest earthquakes: an initial response M3.5 in 1995 and the delayed M4.0 in 1998. To this end, we numerically simulate the poromechanical subsurface response to reservoir impoundment using a 3D model that includes three geological layers down to 10 km depth. From the proposed potential nodal planes of the 1995 M3.5 earthquake, we show that the earthquake has most likely occurred on a vertical, E–W-oriented strike-slip fault with a reverse-displacement component. Deviatoric stresses generated by the water column loading on the surface, superimposed by undrained pore pressure enhancement in deep low-permeability layers can explain the fault reactivation. We find that for the 1998 M4.0 earthquake to occur, conductive flow pathways with permeability as high as 6.6·10⁻¹⁵ m² should exist to transmit pore pressure to a deep critically oriented fault. Our analysis raises the importance of accounting for coupled poromechanical mechanisms controlling fault stability, hydromechanical properties of different rock layers, and realistic shape of the reservoir to accurately assess the potential for reservoir-triggered seismicity. We conclude that reliable forecasting models require accurate subsurface characterization before reservoir filling to enable managing the associated reservoir-triggered seismicity.
The detection of microseismic events with low signal-to-noise ratios (SNRs) can expand the seismic catalog and provide opportunities for a deeper understanding of subsurface reservoir features. We propose a novel polarization analysis method for comprehensively detecting S-wave arrival and P–S travel time of low-SNR events from the particle motion of P and S waves in the time and frequency domain. In most circumstances, the direct S-wave particle motion shows a flat plane, and that is perpendicular to the direct P-wave motion direction. We combine these two properties to detect the S-wave arrival of low-SNR events. Our previous study applied spectral matrix (SPM) analysis to characterize the 3D particle motion of P waves. However, SPM analysis had limitations in detecting S-wave arrivals. We then introduce the time-delay components of the SPM (complex spectral matrix [cSPM]) to characterize the S-wave particle motion, separate the S-wave from the noise, and detect S-wave arrivals. Using the cSPM analysis method, we assess the planarity and perpendicularity of the S-wave polarization in the time and frequency domains. We then define a characteristic function that detects S-wave arrivals by combining two properties, planarity and perpendicularity, to detect more low-SNR events. The P–S travel time is obtained by setting the threshold values for the P- and S-wave characteristic functions. We apply our method to 4 hr and 2 months of field data recorded at the Groningen field in the Netherlands. Our method successfully detects the P–S travel time of all catalog events and several additional undetected events. We locate the hypocenter of all events using the detected P–S travel times with a grid-based search method.
Earthquake recurrence intervals, surface-rupture extents, and interactions between faults provide insight into how faults behave and are critical for seismic hazard mitigation and earthquake forecasting. Investigating the paleoseismology of spatially related faults can reveal strain distribution and whether faults rupture as a system or independently. Summer Lake basin, a graben in the northwestern Basin and Range with four active faults (three of which have prior paleoseismic investigations), provides an opportunity to investigate fault interactions. To expand the paleoseismic record, two trenches were excavated across the previously undocumented Thousand Springs fault, exposing a normal fault zone that offsets a sequence of deep- to shallow-water lake sediments, sand dunes containing reworked Mazama ash, and other Cascades-sourced tephra. Tephra units were correlated to known units by their physical characteristics, stratigraphic sequence, glass chemistry, and two new radiocarbon dates from the uppermost lake sediments. Using trench exposures, measured vertical separations through auguring, colluvial wedges, and extrapolated offsets based on a constant sedimentation rate, we identified at least five surface-rupturing earthquakes with a total offset of 3.4 + 2/−1 m in the past ∼65 ka. The oldest event (EH5) occurred at 63.8 ± 1.5 ka, event horizon 4 at 36.2 ± 12.7 ka (which could be more than one event), and event horizon 3 at 24.6 ± 0.3 ka. Event horizon 2, a warping event at our site, is likely more than one event and occurred between 7.5 and 10 ka; and the most recent event (EH1+), most likely more than one event, occurred between 3.3 and 7.7 ka. Several events correlate, within error, with events on other faults in the Summer Lake basin, suggesting that (1) the faults generally rupture together as a system, (2) the most recent earthquake may have ruptured all faults in the region, and (3) fault rupture is influenced by the rapid regression of Lake Chewaucan (∼13 ka).
The Koyna earthquake of December 10, 1967 was the most damaging reservoir-induced earthquake. It was followed by a long sequence of earthquakes which is still continuing. Precise locations of the Koyna earthquakes have been very much disputed as different locations of the main earthquake and stronger aftershocks were obtained by various workers. Over 1,500 epicenters of Koyna earthquakes through 1973 were obtained by Guha et al. (1974). They cover a large area in a diffused pattern. In view of the continuing seismicity and a recently obtained seismic velocity model, the larger events (ML ≧ 4.0) and about 300 selected smaller events (ML < 4.0) were relocated.
The relocated epicenters show some concentration and suggest the possibility of two trends in the NNE and NW directions. There is a NNE trend of epicenters near the dam and another about 20 km west of the reservoir. The NW trend cuts through these NNE trends. The events were grouped to obtain their composite fault-plane solutions which indicate left-lateral strike-slip faulting along the NNE faults and normal faulting in the NW direction. Faults observed in the LANDSAT imagery match with these trends.
The characteristics of the Oroville, California earthquake sequence of 1975 are presented. Historically, the Oroville area is one of low seismicity; the largest earthquake in the region occurred with a magnitude (ML) 5.7 in 1940 some 50 km north of Oroville. The first foreshock of the sequence occurred on June 28, 1975. Twnety-one foreshocks (ML ≧ 1.6), the largest of magnitude 4.7, preceded the magnitude 5.7 main shock of August 1. All foreshocks and aftershocks of ML ≧ 3.0 were located using seismographs operated by the University of California at Berkeley, USGS, and the California Department of Water Resources. The aftershock region covers an area approximately 14 by 10 km southeast of the city of Oroville. The depth distribution of the earthquakes indicates a west dipping fault plane. The b value of 0.61 shows the sequence to be rich in larger magnitude aftershocks. Similar b values have been determined for other aftershock sequences in California, such as a sequence near Coalinga in August 1975. The aftershock occurrence rate follows an Omori relation with n(t∞t−0.70. Apparent variability in the earthquake mechanisms of the series makes interpretation of composite fault-plane solutions difficult, but the data indicate normal faulting striking NNW and a west dip of about 30°–40°, the Sierra Nevada moving up with respect to the Great Valley.
Six weeks after the start of filling of the Manic 3 reservoir a long sequence of induced earthquakes began. The main shock, mbLg = 4.1, occurred on October 23, 1975, preceded by 1 month of foreshock activity and followed by more than 1,000 aftershocks in the next 4 months. After 2 years, the activity still persists, although the microshocks are substantially reduced in frequency and magnitude. Station MNQ, about 80 km from the activated source, served as an efficient monitor providing data from which increasing induced seismicity was predicted and for evaluating the subsequent activity. Two portable networks were deployed around the source, one in 1975 and the other in 1976. The activity was very shallow, with an average value of 1.5 km, and the cluster of microearthquakes spread over an area of 4 by 4 km. No systematic spatial migration was observed. The activity appeared to have been triggered by the water percolating through joint systems and along northwest oriented planes. The presence of regional stresses and local inhomogeneities, structural and lithological, are suggested as the principal causes, and not the dimensions of the reservoir and the water height.
The 1975 Oroville earthquake of M5.7 occurred almost eight years after the creation of Lake Oroville. This earthquake immediately followed the largest seasonal fluctuation in the lake level to that date. Seismic monitoring at Oroville since 1975 shows that the local seismicity decreases as the lake fills during winter and spring, and that the strongest earthquakes occur as the lake empties during summer and fall. This pattern has been remarkably consistent during the past seven years, and indicates that the seasonal fluctuations in water depth at Lake Oroville control the earthquake occurrences. The magnitude of the earthquakes triggered by Lake Oroville during each seasonal drawdown has generally become smaller since 1975, suggesting that the 1975 rupture zone is being progressively relieved of stress. However, stress has not necessarily been relieved outside the 1975 rupture zone.-Authors
A model of orthogonal, deformable fractures under plane-strain loading is assumed. Only if fractures are sound, tight and wide-spaced (as in deep hypocentral but not weathered surficial zones), can there occur large changes of horizontal effective stresses. Reservoir filling promotes failure on normal and wrench faults, but stabilizes thrusts. Fracture system anisotropy is unimportant. Transients promote failure upon reservoir drainage, but not water table lowering. Reservoirs may be aseismic because loading history may cause non-critical stresses, a strength threshold. Refs.