This study examines reservoir-triggered seismicity (RTS) in Poland and Vietnam. The current state of individual RTS seismic networks necessitates detecting earthquakes from only a few stations. The number of P waves is often inadequate for phase association and event location, which underscores the importance of identifying S waves. Given that individual RTS cases may consist of only hundreds of events, it is crucial for algorithms to be trained on small datasets or to detect effectively using external, global training data. To evaluate this, we compared the efficiency of a deep learning global detection model, transfer learning to the RTS database, a specialized neural network designed for RTS, and manual detection of seismic signals. Transfer learning efficiency was database dependent. Additional interpretation and parametrization of detection results are assumed. Therefore, the emphasis is on phase detection, rather than phase picking accuracy, and detection sensitivity is more important than its specificity. Phase association plays a vital role in detecting seismic signals, facilitating the elimination of most false picks. As a result, the comparisons of detections were based on parameters related to the location of seismic events. The findings indicate that neither the automatic signal detection methods nor the manual methods alone are sufficient. However, their combination significantly enhances detectability. The final catalogs cover up to 30% more events compared to the previous manual. It fulfills the main aim of applying a neural network detector, which is to increase the number of seismic events in the catalog. It may also be further utilized in the research of the triggering process, such as identifying fluid paths and determining fault geometry.

The strongest in Western Carpathians (WC) in XXI century M5 earthquake occurred on 9 October 2023 near Humenne (Slovakia). It was widely felt in Slovakia and Poland, what is rare. It occurred in complex tectonic setting formed with overthrusted frontal nappes and rotated internal lithospheric units. The present tectonic regime of the WC is resulting with vertical movements related to convergence of the WC and of the stable European Platform. Since the complex tectonic setting may influence the estimates of the source parameters, we propose to use local velocity model and regional and temporary seismic stations available within the time of the event occurrence from different projects: AdriaArray, Polish Geological Institute—National Research Institute monitoring network and broadband stations available from Polish, Slovak and Hungarian national seismological networks. Basing on the local velocity model derived from the earlier seismic experiments we obtained similar location and magnitude estimates as EMSC and NEIC, however our focal mechanism is significantly different. Usage of the above-described data improved the precision of the focal mechanism solution and of the depth location of the studied event. Obtained solutions suggest that focus was located at 10–15 km depth and had orientation of strike-slip fault with significant reverse fault component of strike parallel to the main discontinuities in this region and to the PKB (Pieniny Klippen Belt), which follow the trend of the Carpathian Mountains arc or perpendicular to above-mentioned main structures NE-SW strike in agreement with minor discontinuities and main compressional trend in this area.

Poland is characterised by weak natural seismicity. However, the last analysis of the natural seismic hazard in the country was carried out 24 years ago. Therefore, a significant fraction of the recorded seismicity is not included in the hazard estimates currently used, either because recent observations are not taken into account or because of improved seismic network capabilities. Furthermore, Podhale, in the Tatra Mountains, is the only region with recorded permanent natural seismicity. This study aims to create new seismic hazard maps of the Podhale region from a newly compiled database containing information on historical events and two complete instrumental catalogues (regional and local), each at a different level of completeness. The local catalogue was recorded over the last few years. Two seismic hazard assessment techniques were applied, namely the conventional (zone-based) (Cornell in Bull Seismol Soc Am 58(5): 1583–1606, 1968) and the smoothed seismicity model, based on the spatial distribution of seismicity. The earthquake recurrence parameters were estimated using the methodology developed by Kijko et al. (Bull Seismol Soc Am 106: 1210–1222, 2016). The new seismic hazard model incorporates several improvements, such as a comprehensive logic tree and a new set of ground motion models. The new maps provide a more detailed assessment of the seismic hazards of the investigated area. Moreover, they predict higher PGA than previous seismic hazard maps covering Podhale, like global European Seismic Hazard Maps 2013 and 2020.

Automatic detection of seismic events is a useful tool for routine data processing. Effective detection saves time and effort in phase picking and events' location, especially in areas with moderate seismicity at regional and local scales. The Lai Chau area in northern Vietnam is a good example of such a region. An additional difficulty in detection is the anthropogenic origin of reservoir-triggered seismicity observed in this region, where seismicity is non-stationary and there was no prior seismic activity. Neural network event detection was prepared to aid event identifications and further processing of seismic data. An automatic detection system was utilized to reduce the effort of manual interpretation of seismic signals in the region of the Lai Chau dam in North Vietnam while maintaining the detection of weak events at the same level. For this reason, a Single Layer Recurrent Neural Network (SLRNN) was applied. Compared to deep learning algorithms, fewer examples were needed to train the SLRNN. This paper presents a modified version of SLRNN, which additionally uses polarization analysis and the multistage learning process. In the first stage, the training data consists of events detected manually and disturbances selected visually by the operator. In the next stages, the earlier trained detection is validated in the successive recording periods. False detections together with new seismic events are added to the training set and the detection is retrained. The multi-stage process significantly reduces false detections. The software allows SLRNN to be used for routine seismic data processing.

Reservoir-triggered seismicity (RTS) has the potential to generate disastrous seismic events of M6 and bigger. Song Tranh 2 (STR2) is an artificial water reservoir located in Central Vietnam. High seismic activity has been observed in this area since the reservoir was first filled in 2011. The relation between water level and seismic activity in the Song Tranh area is complex, and previous studies have led to the conclusion that ongoing STR2 seismic activity is an example of the delayed response type of RTS. However, the first phase of the activity observed after impoundment has been deemed a rapid response type. There were three stages of the reservoir filling periods: first, a period of initial impoundment, hereinafter referred to as pre-gap period (from 05/01/2011 to 10/06/2012), then a gap period (from 10/06/2012 to 31/08/2013) where reservoir impoundment stopped and water was drained to minimum exploitation level, and finally, a third post-gap period (from 31/08/2013 to 19/06/2017). In this work, we prove that the gap in the filling of reservoir results to a 2-fold rise of seismicity rate. The re-filling of the reservoir results to a drop of activity rate, roughly equal to the pre-gap period, accompanied by a significant increase of b-value. As a consequence, after the gap, the exceedance probability is significantly lower in comparison to pre-gap and gap periods. We also proved that the seismicity recorded between 2013 and 2016 manifested seasonal trends related to water level changes during wet and dry seasons. The response of activity and its delay with respect to water level changes suggest that the main triggering factor is pore pressure change due to the significant water level changes observed. The findings indicate that water load and related pore pressure changes considerably influence seismic activity and stress orientation in this area.

The Song Tranh 2 hydropower construction is located in the Quang Nam province (central Vietnam), it has a reservoir volume of 740 million cubic meters of water and a dam height of 96 m. The reservoir was filled to capacity for the first time in February 2011. The seismicity in the vicinity of reservoir is example of reservoir triggered seismicity(RTS). The natural seismic activity of the Song Tranh 2 reservoir is very low. After the reservoir was filled, the seismic activity increased, and the number and frequency of the tremors also changed as the water level changed. Water level changes are accelerating the tectonic process leading the critically stressed faults to slip. Data suggest that reservoir exploitation stress field changes as triggering origin of this seismicity. The stress inversion method was used to check if there were any seasonal trends. The inverted stress tensor and, in particular, the stress ratio, which is very sensitive to data quality and scope and difficult to accurately retrieve, can be influenced by porous pressure changes. Has been checked, how the average annual seismic activity is related to the change of the water level and if it implies the orientation of the principal stress during high and low water levels in the reservoir. The pore pressure changes and the stress ratio changes were also estimated in relation to the high and low water level periods.

Reservoir-triggered seismic activity depends not only on the technical characteristics of the future reservoir (filling volume, the height of water column) but also on the seismo-tectonics and the natural seismic processes occurring in the area before construction of an artificial reservoir. Passive seismic monitoring was realised near Lai Chau (Vietnam) before the impoundment started. It allowed exploration of the natural seismicity in the area of the future dam. Locations of seismic events several months prior to the reservoir impoundment were observed with ten stations installed in the reservoir vicinity. Events were mainly located near the dam along the Da river headwaters fault. However, only four stations were available for the entire period before the impoundment. Despite the network limitations, completeness of seismic catalogue and b value were determined and may be used as a baseline for analysis of the seismicity in this area after impoundment. The magnitude completeness level is significantly smaller than in the broader seismogenic zones covering an area of the dam. The b value differs from the results obtained for regional seismicity of Northeastern Vietnam. The capability of the local network for moment tensor inversion was estimated with the use of synthetic data tests. Test results provided the requirements for the station number according to azimuthal coverage of the network to obtain the reliable full moment tensor (MT) solution. Preliminary analysis of the seismic activity after Lai Chau reservoir impoundment indicates some changes in activity related to the impoundment and reservoir exploitation.

The seismicity in the vicinity of the Song Tranh 2 reservoir that commenced in late 2010 is an example of reservoir-induced seismicity. Moment tensor (MT) solutions are mostly normal-faulting mechanisms with shearing as the main component of full MT, which suggest reservoir exploitation stress field changes as a triggering origin of this seismicity. However, the local tectonic stress field plays a role in the seismogenic process, which is accelerated by the reservoir exploitation. Two main orientations of discontinuities reactivated by reservoir exploitation are identified, with clustering of events located in two main areas: northern and southern. In addition to the northern cluster (NC) and southern cluster (SC), further smaller clusters are distinguishable, assuming both clustering of epicenters and similarity of focal mechanisms as the clustering criteria. Maximum seismic activities in clusters appear in different periods, with high seismicity in the SC preceding that in the NC. The results indicate significant differences between the NC and SC. It is shown that the SC is an example of rapid triggering, whereas the NC is a delayed triggering type. Also, magnitude distribution differs between the NC (b approximate to 1.1) and SC (b approximate to 1.3).

The Institute of Geophysics Vietnam Academy of Science and Technology (IGP VAST) and the Institute of Geophysics Polish Academy of Sciences (IG PAS) has been cooperating for decades in the field of seismology. It started in the International Geophysical Year 1957-1958. For a long time, the cooperation was based on a seismic analysis of shocks. An example of such cooperation was the earthquake in 1989 in Hoa Binh, North Vietnam. New horizons of cooperation appeared in 2013, when an Agreement for Research Co-operation between the IGP VAST and the IG PAS was concluded to provide a framework for co-pperative research in the field of Reservoir Triggered Seismicity (RTS) research in Song Tranh 2 (STR2) area. The cooperation was expanded in 2015, when next agreement covering second reservoir in Lai Chau was concluded. The measurement and results of common investigations in STR2 and Lai Chau are the scope of this work. In both areas dense local networks were installed which allowed estimate focal parameters. The Lai Chau case is under initial study now. Contrary to most cases of RTS, the local network in Lai Chau had been built before the filling of the reservoir started. It allows to study full cycle of changes of seismogenic processes. The seismicity in STR2 began in 2010, just after the dam was filled. Result of investigations were already published. We discovered there two regimes of seismic response to the filling of the reservoir, which occur in clusters. They differ in periods of seismic activity and frequency- magnitude distributions.

PASSEQ 2006–2008 (Passive Seismic Experiment in TESZ; Wilde-Piórko et al. 2008) was the biggest passive seismic experiment carried out so far in the area of Central Europe (Poland, Germany, the Czech Republic and Lithuania). 196 seismic stations (including 49 broadband seismometers) worked simultaneously for over two years. During the experiment, multiple types of data recorders and seismometers were used, making the analysis more complex and time consuming. The dataset was unified and repaired to start the detection of local seismic events. Two different approaches for detection were applied for stations located in Poland. The first one used standard STA/LTA triggers (Carl Johnson's STA/LTA algorithm) and grid search to classify and locate the events. The result was manually verified. The second approach used Real Time Recurrent Network (RTRN) detection (Wiszniowski et al. 2014). Both methods gave similar results, showing four previously unknown seismic events located in the Gulf of Gdańsk area, situated in the southern Baltic Sea. In this paper we discuss both detection methods with their pros and cons (accuracy, efficiency, manual work required, scalability). We also show details of all detected and previously unknown events in the discussed area.

The paper describes a temporary seismic project aimed at developing the national database of natural seismic activity for seismic hazard assessment, officially called “Monitoring of Seismic Hazard of Territory of Poland” (MSHTP). Due to low seismicity of Poland, the project was focused on events of magnitude range 1–3 in selected regions in order to maximize the chance of recording any natural event. The project used mobile seismic stations and was divided into two stages.Five-year measurements brought over one hundred natural seismic events of magnitudes ML range 0.5–3.8. Most of them were located in the Podhale region in the Carpathians. Together with previously recorded events this made it possible to conduct a preliminary study on ground motion prediction equation for this region. Only one natural event, of magnitude ML = 3.8, was recorded outside the Carpathians in a surprising location in central-west Poland.

This chapter presents our up-to-date knowledge on seismicity of the Polish part of the Western Carpathians. It also gives a list of seismic sites and description of measurement procedures applied in the discussed region thus far. We mainly focus on seismicity of the Podhale region, the Pieniny Klippen Belt and the Beskids. At the moment there are 5 seismic stations in the Podhale region, which record the permanent weak seismic activity at the level of 0.5 <= M-L <= 2.3. In this paper we estimated the completeness of catalogs and the distribution of magnitude exceedance probability for events in the Podhale region.

Song Tranh 2 hydropower plant and the reservoir containing backed up water are located in the Quang Nam province (Central Vietnam). The region experiences unusual seismic activity related to the reservoir impoundment, with earthquakes of magnitude up to 4.7. In result of cooperation between the Institute of Geophysics, Vietnam Academy of Sciences and Technology and the Institute of Geophysics, Polish Academy of Sciences a seismic network has been built to facilitate seismic monitoring of the Song Tranh 2 area. The network, operating since August 2013, consists of 10 seismic stations. Here we show that the network is sufficient for advanced data processing. The first results of monitoring of the earthquake activity in Song Tranh 2 area in the period between 2012 and 2014, especially the completeness of catalogs, study and comparisons between water level and the seismic activity suggest direct connection between reservoir exploitation and anthropogenic seismicity.

On 19 March 2013, a tremor shook the surface of Polkowice town where the Rudna Mine is located. This event, of ML = 4.2, was the third most powerful seismic event recorded in the Legnica Głogów Copper District (LGCD). Inhabitants of the area reported that the felt tremor was bigger and lasted longer than any other ones felt in the last couple of years. Analysis of spectral parameters of the records from in-mine seismic system and surface LUMINEOS network along with broadband station KSP record were carried out. The location of the event was close to the Rudna Główna Fault zone; the nodal planes orientations determined with two different approaches were almost parallel to the strike of the fault. The mechanism solutions were also obtained as Full Moment Tensor from P-wave amplitude pulses of underground records and waveform inversion of surface network seismograms. The results from the seismic analysis along with macroseismic survey and observed effects from the destroyed part of the mining panel indicate that the mechanism of the event was complex rupture initiated as thrust faulting on an inactive tectonic normal fault zone. The results confirm that the fault zones are the areas of higher risk, even in case of carefully taken mining operations.

This study is an application of a Real Time Recurrent Neural Network (RTRN) in the detection of small natural seismic events in Poland. Most of the events studied are from the Podhale region with a magnitude of 0.4 to 2.5. The population distribution of the region required that seismic signals be recorded using temporary stations deployed in populated areas. As a consequence, the high level of seismic noise that cannot be removed by filtration made it impossible to detect small events by STA/LTA based algorithms. The presence of high noise requires an alternate method of seismic detection capable of recognizing small seismic events. We applied the RTRN, which potentially can detect seismic signals in the frequency domain as well as in the phase arrival times. Data results of small local seismic events showed that the RTRN has the ability to correctly detect most of the events with fewer false detections than STA/LTA methods.

The earthquake of magnitude M L = 3:8 (EMSC) took place on Friday, 6 January 2012, north-east of the town of Jarocin in Wielkopolska Region, Poland. The only historical information about past earthquakes in the region was found in the diary from 1824; apart of it, there was a seismic event noticed in the vicinity of Wielkopolska in 1606 (Pagaczewski 1982). The scope of this paper is to describe the 6 January 2012 event in view of instrumental seismology, macroseismic data analysis and known tectonics of the region, which should be useful in future seismic hazard analysis of Poland.