Search Documents

By Ihsan Berk Tulu, Ted Klemetti, Deniz Tuncay

Accurately estimating load distributions and ground responses around underground openings plays a significant role in the safety of the operations in underground mines. Adequately designing pillars and other support measures relies highly on the accurate assessment of the mining-induced loads, as well as the load-bearing capacities of the supports. There are various methods that can be used to approximate mining-induced loads in stratified rock masses, both empirical and numerical. In this study, the numerical modeling approach recently developed at West Virginia University, which is based on the modeling approach developed by the National Institute for Occupational Safety and Health (NIOSH), is investigated using the finite difference software FLAC3D. The model includes the longwall panels, the adjacent chain pillar systems, and the different stratigraphic layers of the overburden. Using the 3D model, changes in loading conditions and deformations on the areas of interest, induced by an approaching longwall face, can be examined. This paper details the 3D modeling of a longwall panel utilizing this approach, and the verification of the results against field observations. The studied panel was 360 m wide with a 3-entry chain pillar system and about a 160-m average overburden depth around the studied area. The overburden strata consist of alternating layers of shale, sandstone, and limestone. The FLAC3D results were compared against field measurements from the mine site. The stress change values measured in the chain pillars were comparable with the modeling results. The model also replicated the surface subsidence profile obtained from field measurements fairly well. Overall, the 3D modeling approach was found to be successful for the case study longwall panel.

Sep 23, 2020

By Liming Yuan, James Rowland, Lihong Zhou, RICK THOMAS, DAVOOD BAHRAMI

Continuous monitoring of carbon monoxide and other fire-related parameters by means of an atmospheric monitoring system (AMS) has been used by the mining industry for early fire detection in underground mines. The National Institute for Occupational Safety and Health (NIOSH) initiated a project to integrate real-time AMS sensor data with NIOSH’s mine fire simulation program, MFIRE 3.0, to simulate and predict the spread of smoke that would provide assistance to mine fire emergency response personnel. Determining the heat release rate of a fire using the monitored sensor data was a critical component of the successful completion of this project. NIOSH researchers developed a direct method to calculate the heat release rate when a fire is within close range of sensors. However, this method is only applicable to the case where a fire occurs in AMS-monitored airways. This paper presents an improved method for determining the fire heat release rate for complicated scenarios where a fire is distant from sensors and airflow splits and merges are present. The method was validated using a fullscale diesel fuel fire test conducted in the Safety Research Coal Mine at the Pittsburgh Mining Research Division and can help mine operators and safety personnel make informed decisions during a fire emergency.

Jul 31, 2020

By Herry Poernomo, Agus Prasetya, Iga Trisnawati, Himawan Tri Bayu Murti Petrus, Gyan Prameswara, Panut Mulyono

We studied yttrium extraction from untreated zircon sand processing waste tailings. Xenotime mineral as the rare earth element source with an abundance of yttrium (REY) was detected in sufficient grade for extraction. As much as 9.03% yttrium exists in zircon tailings (analysed by X-ray fluorescence spectrometry). The presence of yttrium was confirmed by X-ray diffractometry in xenotime minerals, which are yttrium carriers (Y-PO4). The purpose of this research was to determine the effect of leaching conditions on yttrium recovery from zircon sand after alkaline fusion treatment. Alkaline fusion was chosen to decompose phosphate into hydroxide in the xenotimemineral,which will reduce further required hydrometallurgical processing.Alkaline fusionwas carried out for 3 h at 450 °C, with a ratio of sodium hydroxide to zircon tailings sand of ~ 1:1. The alkaline fusion product was leached with water, followed by hydrochloric acid treatment to leach the yttrium. Yttrium recovery reached 87% under optimum conditions (60 °C, 1MHCl, and solid-to-liquid ratio = 1/10 for 7.5 min). A suitable model for yttrium dissolution with hydrochloric acid was diffusion through a solid particle ash layer. The calculated activation energy (EA) for this model was 20.21 kJ/mol.

May 26, 2020

By Hemant Agrawal, Arvind Kumar Mishra

Rock breakage using blasting is one of the important operations of the mining industry. The delay timing decides the initiation sequence of holes in the blasts and is a crucial factor to improve the overall blast performance. Blast-induced ground vibration is an undesired yet unavoidable outcome of blasting. Peak particle velocity (PPV) is commonly used to measure the magnitude of blast-induced ground vibration. Channelization and better utilization of explosive energy delay are provided between blast holes using initiating systems. The use of different initiation systems even on the same site results in a very significant difference in the values of PPV generated. Due to scattered delay timings of pyrotechnic detonators, the scaled distance predictor equations obtained become unreliable. The error/scatter in delay timing leads to overlapping of holes and hence increases in maximum charge per delay. In this paper, an analytical approach has been discussed to measure the overlapping of holes during blasting with initiation systems having inherent cap scatter. The analytical approach will be calculating the probable overlapping of holes due to scatter in delay timing and will help blasting engineers to understand the variation in maximum charge per delay due to pyrotechnic based initiating systems. In this study, a total number of 102 blasts were conducted using different initiation systems at two different opencast coal mine and 221 corresponding blast vibrations events were recorded.

Nov 10, 2020

By Jon A. Hummer, MICHAEL R. SHAHAN, Aleksandar D. Bugarski, Shawn Vanderslice

A study was conducted in an undergroundminewith the objective to identify, characterize, and source apportion airborne aerosols at the setup face and recovery room during longwall move operations. The focus was on contributions of diesel- and battery-powered heavy-duty vehicles used to transfer equipment between the depleted and new longwall panels and diesel-powered light-duty vehicles used to transport personnel and materials to various locations within the mine. Aerosols at the setup face were found to be distributed among diesel combustion-generated submicrometer and mechanically generated coarse aerosols. According to the data, the submicrometer aerosols downstream of the setup face were sourced to diesel exhaust emitted by vehicles operated inside and outside of the panel. Depending on the intensity of the activities on the panel, the outby sources contributed between 12.5 and 99.6% to the average elemental carbon mass flow at the setup face and recovery room. Extensively used light-duty vehicles contributedmeasurably to the elemental carbon concentrations at the setup face. The number concentrations of aerosols downstream of the setup face were associated with aerosols generated by combustion in diesel engines operated in the shield haulage loop and/or outside of the longwall panels. Entrainment of road dust by diesel or battery-powered load-haul-dump vehicles operated near the measurement site appears to be the primary source of mass concentrations of aerosols. The findings of this study should help the underground mining industry in its efforts to reduce exposures of miners to diesel and coarse aerosols.

Apr 14, 2020

By MARK J. VANARELLI

Estimating groundwater inflows in rock deep tunnels is critical to the safety of personnel constructing the tunnel and the cost of construction. This paper presents two case studies for estimating steady-state groundwater inflows into deep rock tunnels. The case studies included the Chattahoochee and Nancy Creek tunnels. In this paper, statistical methods were applied to packer test data obtained from exploratory borings during geotechnical site investigations. A semi-empirical procedure was utilized for estimating groundwater inflows into these deep rock tunnels. It is the most widely used method/procedure in the tunnel industry. The Chattahoochee and Nancy Creek tunnels are deep rock tunnels which exhibit radial flow conditions. For the radial flow condition, a statistical analysis presented in this paper indicates that the semi-empirical procedure works well; however, variations in the procedure and/or an insufficient amount of testing could lead to underestimations of the inflow quantities. Packer test data plotted in histograms were observed in all cases to be log-normally distributed for the radial flow condition. Modeling using Monte Carlo simulations was observed to be an effective tool for removing irregularities in the distribution of the data and incorporating high-end permeability data to derive reasonable groundwater inflow estimates. This paper is based on the PhD thesis entitled “Improvements to the Current State of the Practice in Steady-State, Groundwater Inflow Estimates into Rock Tunnels based on the Application of Statistical Analysis Methods of Packer Test Data obtained from Exploratory Borings During Geotechnical Site Investigations.”

Oct 9, 2020

By Liaqat Ali Shah, Muhammad Farooq, Zait Ullah Shah, Francisco Rolando Valenzuela-Díaz

Bentonite collected from Nowshera (Pakistan) was treated with NaCl, and the < 2-μm clay fractions were separated by sedimentation process. The textural changes in NaCl treated purified bentonite were compared with raw and physically purified sample. It was detected that the NaCl-treated purified bentonite and the sample purified by simple sedimentation, respectively, consist of sodium-montmorillonite and calcium-montmorillonite as a major clay mineral. The honeycombed-type morphology was observed for both the purified samples; however, the particles of NaCl treated purified bentonite were relatively in a compact form. The particle sizes of both the purified samples were found considerably smaller. The high volume percent of the fine particles was noticed in the NaCl-treated purified bentonite. This attributed that the sodium exchange has a valuable impact on the bentonite dispersion in aqueous media during particle size measurement. The relatively high mesopore area/volume and micropore area/volume and specific surface area (SSA) were detected in purified samples. The NaCl-treated purified sample showed relatively high mesopores area/volume than the sample obtained by simple sedimentation. These changes seem to be originated from either high montmorillonite contents and/or the smaller particle size, which might be produced after the layer rearrangement during the interlayer cation replacement by Na+ ions. The large SSA and high porosity of purified samples particularly the NaCl-treated bentonite suggest their potential use as adsorbents, catalysts, pharmaceuticals, and cosmetics.

Jan 7, 2021

By Patrick Linzer, Lindsay Linzer

By its nature, seismology has been a data intensive field since the first digital computers were introduced in the 1950s. The first empirical models were developed in the same period and are still widely used today, for instance, the Gutenberg-Richter frequency-magnitude relationship (Gutenberg and Richter, 1956). Machine learning (ML) methods make it possible to redefine these empirical models using large datasets and examining considerably more parameters, where their significance to the problem being addressed is methodically and repeatably evaluated using robust statistical methods. In this paper we develop a simple logistic model for the short-term prediction of seismic events above a threshold magnitude from a mine’s event catalogue comprising more than a million events with 14 basic parameters from which a pool of over 830 derived parameters was developed, covering a series of 11 precursory time windows. The logistic model was fitted stepwise using SAS with α = 0.025 for both entry and acceptance. 25 variables from the pool met the acceptance criterion. The findings were surprising in that parameters a rock engineer or seismologist might have expected to be the most significant in the logistic model were not; instead, it was an increase in standard deviation of the immediately-prior seismic magnitudes that carried the most weight, followed by the average maximum source displacement and median apparent volume change. These findings make intuitive sense because the increasing scatter of the magnitude shows an increasing number of large events being recorded that could be indicative of instability in the system that could lead to large events. The model is rudimentary and enormous scope exists for more exploration in terms of data pre-processing, model types, and response variable definitions.

Apr 26, 2022

By Zong-Xian Zhang, Charlie C. Li

It has been observed that the rock debris of a strain burst event are often in the mode of tensile fracture. Seismic waves could create tensile fractures in the material nearby reflecting surfaces, such as at the boundary of the fracture zone and the rock surface of an underground space. The fractured rock could be ejected to the space when the seismic waves are powerful enough. The average ejection velocity of the ejected rock is a measure of the magnitude of a burst event. The ejection velocity is directly associated with the released energy from the seismic waves and with the strain energy stored in the rock prior to the burst. In some cases, the seismic wave induced tensile fracture and the energy conversion occur in successive moments during bursting. In this paper, tensile fracture caused by seismic waves is investigated through an 1D model. The relationship of the average ejection velocity of the ejected rock in a rockburst event is analytically established with the energy of the seismic waves and the strain energy stored in the rock.

Apr 26, 2022

By B. Simser, T. Butler, A. Hall

The longterm future of Glencores’ mining operations in the Sudbury basin area of Ontario, Canada likely depends on mining deposits at depths below two kilometres. To gain a better insight into the rockmass response to mining, a geological scouthole program has been implemented at Glencores’ Nickel Rim Deep and Onaping Depth Project to gain insight into rockmass conditions ahead of the devel-opment face. The scouthole program involves drilling long (up to 600 m) subhorizontal boreholes drilled ahead of the development end and collecting data about the rockmass conditions from various geophysi-cal and mechanical tools (in addition to conventional core logging). Once all data has been collected and the borehole is available, seismic sensors are installed up to 450 m down the hole as part of the micro-seismic monitoring program with the leading sensor always located ahead of the development end. The recorded seismic data is used to provide realtime insight into the rockmass response to mining. On May 20, 2017 a flurry of seismicity with a maximum moment magnitude (Mw) of 0.7 occurred causing damage to the ramp heading. Data from the microseismic monitoring system had indicated that conditions at the development end were changing prior to the rockburst. A back analysis of the scouthole data indicated a clear anomaly in the rockmass at the location of the rockburst. The scouthole data also provided sufficient information to suggest that once this anomalous zone was exited, mining conditions would improve. With the microseismic monitoring system providing realtime information from the rockmass, the area was re-habilitated and ramp development continued successfully. Details of the rehabilitation and adjustments to the ramp development are discussed. Some correlations between the scouthole data and recorded micro-seismicity are also presented.

Apr 26, 2022

By Liliana Rybarska-Rusinek, Aleksandr M. Linkov, Ewa Rejwer-Kosińska

Joined modeling of stressed state and induced seismicity is a means to join geomechanics with seismic observations. When applied to quantifying rockburst hazard, it provides, on one hand, estimations in terms of mechanics, such as stress concentration, and, on the other hand, in statistical terms of the risk analysis developed in mining seismology. The modeling became possible due to progress in computers and numerical methods. However, despite tremendous increase of computational abilities, the joined modeling is still quite time expensive. To make it available for wide use and real time simulations, there is need to reduce the time cost of its key stages: calculation of stresses, caused by mining, and simulation of seismic events, induced by these stresses. The present work aims to decrease the time cost. The stages are time costly because of the same reason, which is a very large number N (up to millions) of elements, representing rock structure, openings and flaws. This leads to immense computational complexity (roughly of order N2) of conventional accounting for pairwise interactions between the elements. The complexity may be notably reduced, when employing the fact that interactions decrease with growing distance between elements. Specifically, the fast multipole methods (FMM) avoid the complexity N2, by employing translations of far-fields from whole collections of sources. This reduces the complexity to the order 22, where p is the number of terms involved in a single translation, is the number of translations, and is the dimension of the quantity translated. Still, for conventional FMM, the complexity remains too high to routinely perform joined modeling of stresses and seismicity.

Apr 26, 2022

By Stephanie Singarayer, Damodara Reddy Chinnasane, Mike Yao, Danielle Pelletier

Mining at Vale’s Creighton Mine has been active for nearly 120 years in Sudbury, Ontario, Canada. Being the deepest mine in the Sudbury region, mining induced seismicity is a common occurrence at Creighton Mine. Over the past five years, 26 large magnitude seismic events ranging from 2.5 Mn to 3.7 Mn (Nuttli) have occurred. Out of these large magnitude events, three 3.0+ Mn events occurred between April 2020 and January 2021. The majority of the 3.0+ Mn events are believed to be the fault-slip type that could be linked to the known major geological structures and/or long-term effect of mining over many years. The recent large magnitude events are believed to be triggered by high induced stresses while mining the first few stopes of a new lower horizon in the top-down centre out mining sequence.

Apr 26, 2022

By Christina Dahnér, Kristina Jonsson, Savka Dineva

The iron-ore underground Kiirunavaara mine (Sweden) has been seismically active since 2008. In 2021 the production is on levels 1079 and 1108 m, at the deepest. The seismic activity has changed in character over time with one step change in 2015 when the number of large (ML>2) seismic events increased, and another one when the mine experienced a seismic event with Mw 4.2 in 2020. In 2017, the mine evaluated Hydraulic Fracturing (HF) as a possible mitigation technique for larger seismic events and a one-hole trial was performed. The trial proved promising, and it was decided to go forward with a large-scale trial covering an entire production area in the mine. The large-scale trial consists of 25 fracturing holes, 50 – 100 m long, running along 500 m out of the ~4500 m long orebody and spanning between production levels 1108 m and 1252 m. The fracturing process continued for a year, resulting in ~38,000 seismic events with −3.3<ML<1.4. The breakdown pressure varied between fractures in the same hole as well as between holes in the same area. Average breakdown pressure per hole was between 12 MPa and 50 MPa. The seismic response to HF in 2021 varied greatly from hole to hole and between fractures in the same hole. The variation in response could be related to the geological conditions and the infrastructure, such as large caving ore passes. In this paper some details about the experiments and broad preliminary analysis of the effect of the HF by using seismic source parameters of the induced seismic events are presented: 1) Distribution of seismic activity, 2) the ‘equivalent’ magnitudes for each hole calculated from the cumulative seismic energy and seismic potency released during the fracturing, compared with the local geological conditions; 3) the predominant failure mechanism of the seismic events caused by HF was defined. A difference in the behaviour between larger and smaller events was noticed, as ML>0 events seem to be compression type failures and ML<0 events seem to be related to slip on pre-existing structures. No seismic events related to creation of a new fracture have been identified yet.

Apr 26, 2022

By Ehsan Moosavi, Kamyar Tolouei, Mehran Gholinejad

The literature defines rockbursts in slightly different ways based on their generation mechanisms, field phenomena, and degree of damage, but in a way that is almost consistent. Generally, a rockburst refers to seismic events that mainly happen in underground excavations, for purposes such as underground mining, road and railway tunnels, nuclear power, etc., and that are followed by the collapse and outburst of rocks. Long-term rockburst prediction can be abstracted as a classification problem mathematically, in which we can introduce deep learning to solve it. Further, using deep learning in rockburst prediction can overcome drawbacks such as subjectivity and inconsistency brought from previous traditional approaches. A rockburst data-set, was employed to evaluate the current method for predicting rockburst grade, and the good results of overall success rate were obtained. The results indicated that the proposed algorithms can speed up parameter optimization search, the proposed method is robust model and might hold a high potential to become a useful tool in rockburst prediction research.

Apr 26, 2022

By E Marambio, R Brändle, G Fischer, G von Rickenbach, M Fuentealba, L Burgos, R Luis Fonseca

In underground excavations, under high stress conditions, ground support systems must provide safe and effective response to dynamic events. These systems must be capable to resist dynamic impacts and yielding during the loading process. In this context dynamic testing of the reinforcement and retaining elements that compose the ground support system are required to study and improve the behaviour of these elements under dynamic load events. During the last years, Geobrugg has been working on the improvement of retaining products by testing them in a large-scale impact test facility located at Walenstadt, Switzerland. The test facility is composed of a double level platform with a square-shaped pyramidal trunk geometry, in the upper level housing a loading mass that drop from a height up to 5m. The loading mass is guided by one central steel pipe, and the impact occurs in the sample to be tested located at the lower level in a slab with an area of 3.6m × 3.6m where the ground support system is installed. During the last years, this innovative facility has been used to test several configurations of ground support systems. The results of these tests have enabled to improve the understating of the behaviour of ground support systems under dynamic loads. In this manuscript, the back analysis through numerical methods (central difference and Newmark’s method) of a ground support system tested with a double dynamic impact in 2019 is presented.

Apr 26, 2022

By Jake Beale, Joe Wickline, M. Van Dyke, Ted M. Klemetti

A longwall coal mine in southwestern Virginia mining the Pocahontas #3 Coal Seam experienced a history of mining-related seismic events. The events were never large enough to cause concern until three major events occurred recording 3.4, 4.3, and 3.4 ML in 2005, 2006, and 2007 respectively. Two of the seismic events generated an air blast which damaged stoppings which fueled a mine fire in 2005 and 2007. The mine responded to the events by narrowing the longwall widths, changing ventilation controls, installing a 7-sensor seismic system around the active panel, and creating a mining hazard map to avoid similar geology in the future. All of the actions taken have prevented seismic activity from impacting mining operations, but it did not prevent other large seismic events from occurring (Van Dyke et al., 2018). In 2016, another large seismic event, measured at 3.7 ML, occurred in the gob that was reported by the local media and recorded by the United States Geological Survey (USGS). While the event did not directly impact mining operations, mine management wanted to know where and under what conditions could large seismic events occur in the future. The National Institute for Occupational Safety and Health (NIOSH) became involved to help understand what conditions precipitated these events so that future mine design could more accurately assess seismic hazards and reduce the risk to mine worker safety. Researchers at NIOSH and mine management created a seismic forecasting map that combined three factors: (1) overburden thickness, (2) sandstone thickness, and (3) the expected caving height of the gob. The criteria were developed based on the geologic data the mine collected from core drilling, e-log analysis, and fiberscope observations. The forecast map was able to provide an accuracy of 72%–85% of events 1.5 ML or greater and 100% accuracy of events that were 3.0 ML or greater based on historical data. The seismic forecasting map was applied to the entire future mining area to show areas where seismic potential could be the greatest based on current mine design planning, ultimately improving the ability of the mine operator to assess the seismic hazards surrounding active and future mining areas (Van Dyke et al. 2018). Following this research, the seismic forecasting map was implemented into the mine’s hazard recognition program and is continuously updated and evaluated prior to mining each longwall panel. This paper will cover in more detail how the seismic and lithologic data was processed and how the mine implemented a new seismic mitigation program based on the forecasting results.

Apr 26, 2022

By M. Sears, W. H. Su, J. Wickline, M. A. Van Dyke, G. Esterhuizen, Z. Khademian

It is commonly recognized that seismicity can be induced by rock failures or slip along geologic structures around mining excavations. The unpredictable nature of these events and their proximity to the working area may raise safety concerns in underground mining operations. Modeling can be a useful tool for estimating seismic potential for a given mine layout, but the challenge is to detect a seismic event and estimate its intensity in the model. One well-documented approach is to monitor the variation in the energy content of the model and record potentially damaging seismic events when there is a significant release of energy. Researchers from the National Institute for Occupational Safety and Health (NIOSH) employed this approach using 3DEC modeling software to simulate the anomalous seismic events that occurred in a deep longwall mine in Virginia in 2005. A four-panel model of the mine was constructed utilizing a 2D cross section at the location of the mining face at the time of the event. Seismic potentials associated with the development of each panel are evaluated by recording the seismic energy released during the simulation. The results demonstrate the occurrence of two relatively significant events during the mining of the last panel, which are consistent with the historical observations at the mine.

Apr 26, 2022

By Kai Koosmen, Mehdi Serati, Tengzhou Wang

Spalling and rockburst are unwanted brittle failures at the boundaries of underground excavations that are commonly observed in hard rock and high-stress environments at significant depths, although, spalling has also been observed in comparatively shallow soft rocks. The crack initiation (CI) stress threshold (representing the stress level at which new micro cracks begin to form in an intact rock) is widely used to establish the in-situ spalling strength, and is generally regarded as the lower limit of the long-term rock strength. This study analyses the evolution of CI and other mechanical properties for a high-porosity sandstone with an average UCS of 38 MPa under unconfined cyclic compressive loading. The results indicate that rock properties are variably affected by cyclic loading with; no significant effect on Youngs Modulus; a moderate increase in Poisson’s ratio and fracture angle; a moderate decrease in UCS; and a major reduction in the CI stress threshold. All of these changes were generally amplified by cyclic loading at higher stresses. From the results, it is concluded that rockburst and spalling limit in underground structures may reduce over time if the rock is subject to repetitive cyclic loading, and that the reductions will be greater for higher cyclic stresses. This may be a significant factor in the design and operation of some deep underground structures.

Apr 26, 2022

By P. K. Kaiser, M. Cai

When designing an efficient and effective support system for burst-prone ground, it necessary to select the most appropriate, i.e., the ‘right’ support components, including areal and reinforcement components. For this paper, it is assumed that the areal support is not the weakest link and the support performance is controlled by the rock reinforcement and holding components (bolts). We discuss the loading modes of rockbolts under static and dynamic loadings and analyze published data from laboratory and field tests. The mechanisms of energy absorption of dynamic rockbolts are presented and rockbolts are categorized based on factors such as energy absorption capacity, energy absorption rate, direct and indirect energy absorption capacities, rock mass bulking or displacement control, and shearing resistance capacity. Guidelines for selecting the ‘right’ rockbolts to build an effective rock support system for burst-prone ground are provided.

Apr 26, 2022

By Ping Zhang, Shahram Mozaffari, Andreas Ylmefors

The risk associated with mining induced seismicity and the related rock bursts has become one of the major threats to the safety and sustainability of mining at the Kiirunavaara mine. The causes of mining induced seismicity at the Kiirunavaara mine are very complex and have affected our understanding on the source mechanisms of seismicity and further the mitigation of the seismic risk. Detailed analysis of historical mining induced seismicity at the Kiirunavaara mine was conducted. The focus of this analysis was placed on the classification of large seismic events and the evaluation of the influential factors on the classified seismic events. 145 events with local magnitude larger than 1.5 that occurred between January 1, 2011 and March 31, 2021 were analyzed. For this analysis, seismic events located in the hanging wall were not included, as these have not been observed to cause large damages to the mine itself. It was found that the magnitude of the events had good correlation with the stress change in relation to the production location. The seismic events were classified as mining-front associated, ore-pass associated, mining-front and ore-pass associated and others, according to their hypocenter location. Most of the mining-front associated events, concentrated around the production levels and slightly above and below them, generally correlated with shear cracks, and most of them occurred when the ore flow between two production levels started. The numerical modelling showed that slip of the geological discontinuity, due to unclamping between hanging wall and footwall, could be the major causative mechanisms, which further supports this finding. The ore-pass associated events are located close to the ore passes. Large seismic events located between ore passes have the source mechanism of both fault-slip and non-shear, but events located at the outer side of ore passes were all fault-slip type. By checking with the ore pass scanning data, it was concluded that the most of the events located between ore passes could be pillar bursts. An elastic core concept has been then used to define a critical pillar width. The paper is not intended to be prescriptive but rather to initiate discussion of the possible classification of seismic events at the Kiirunavaara mine and to present new perspectives of the causes of the seismicity for improving the present seismic risk management.

Apr 26, 2022