Next Article in Journal
Index-Based Alteration of Long-Term River Flow Regimes Influenced by Land Use Change and Dam Regulation
Previous Article in Journal
Using Public Participation GIS to Assess Effects of Industrial Zones on Risk and Landscape Perception: A Case Study of Tehran Oil Refinery, Iran
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Earth
Volume 5
Issue 3
10.3390/earth5030022
earth-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Using the Contrast Boundary Concentration of LST for the Earthquake Approach Assessment in Turkey, 6–8 February 2023

by
Serhii Nikulin
,
Kateryna Sergieieva
*,
Olga Korobko
and
Vita Kashtan
Faculty of Information Technologies, Dnipro University of Technology, 19 Avenue Dmytra Yavornytskoho, 49005 Dnipro, Ukraine
*
Author to whom correspondence should be addressed.
Earth 2024, 5(3), 388-403; https://doi.org/10.3390/earth5030022
Submission received: 22 July 2024 / Revised: 12 August 2024 / Accepted: 17 August 2024 / Published: 18 August 2024
Browse Figures
Versions Notes

Abstract

:
Land surface temperature (LST) variations and anomalies associated with tectonic plate movements have been documented before large earthquakes. In this work, we propose that spatially extended and dynamic linear zones of high temperature anomalies at the Earth’s surface coinciding with faults in the Earth’s crust may be used as a predictor of an approaching earthquake. LST contrast boundary concentration maps are suggested to be a possible indicator for analyzing temperature changes before and after seismic sequences. Here, we analyze the concentration of LST contrast boundaries estimated from Landsat 8–9 data for the East Anatolian Fault Zone in the vicinity of epicenters of the destructive earthquakes with magnitudes up to 7.8 Mw that occurred in February 2023. A spatial relationship between earthquake epicenters and the maximum concentration of LST boundaries at azimuths of 0° and 90° was found to strengthen as the earthquake approaches and weaken after it. It was found that 92% of epicenters are located at up to 5 km distance from zones of maximum LST boundary concentration. The evidence presented in this work supports the idea that LST may provide valuable information for seismic hazard assessment before large earthquakes.

1. Introduction

On 6–8 February 2023, a series of catastrophic earthquakes struck southern and central Turkey (Kahramanmaraş Province), and northern and western Syria, causing widespread destruction and loss of life [1]. Two shocks of magnitude 7.8 and 7.5 Mw were particularly strong along the East Anatolian Fault System, which is known for its seismic activity [2,3]. As a result, more than 53,000 people died in Turkey and neighboring Syria [4]. These facts make the question of the search for new methods of earthquake prediction and indicators of increased seismic activity crucial.
Earthquake is a natural phenomenon occurring as a sudden shaking or shifting of the Earth’s surface caused by geological or tectonic forces [5]. In most cases, earthquakes occur at the boundaries of tectonic plates, where the contact of the plates prevents their free movement [6]. As a result of tectonic activity, plate boundaries exert pressure on each other and, when released, cause sudden quakes accompanied by strong energy release and potentially severe destruction [7]. As the earthquake approaches, high stresses are observed along the tectonic plates, accompanied by an increase in temperature. The surface temperature rises due to the energy released during fault slip with fluid transfer. The mechanical impact and resistance of rocks exceeding their strength threshold, causes the formation of cracks, faults and fractures grouped into larger structures—lineaments—along which there is an increased release of heat, water vapor and gases from the subsurface to the Earth’s surface [8,9,10].
Several possible seismic precursors have been proposed, including seismicity, mechanical deformation of the lithosphere and the Earth’s surface, changes in the level and composition of groundwater in the time interval from a few weeks to a few hours before the event, increased emission of gasses, in particular radon or radioactive ions, into the atmosphere, propagation of acoustic gravity waves, fluctuations in electric and magnetic field components over a wide frequency range, and changes in ionospheric parameters, in particular the total electron content (TEC) [11,12,13,14,15,16,17]. However, until now, none has passed the statistical tests to prove its strong reliability. Thus, the research continues in the field of fluctuations (primarily an increase) in the land surface temperature (LST) [18,19]. The LST is of particular interest because temperature variations are themselves the causes of geological and tectonic forces that trigger earthquakes [20]. The temperature rise in the Earth’s interior is accompanied by changes in ground heat flux and ground temperature rise, forming areas of increased thermal background—LST anomalies [10,21]. Temperature anomalies appear at the Earth’s surface as linear structures of elevated temperature along crustal faults, characterized by large spatial extent and high temporal dynamics [16]. LST anomalies are maximal in pre-earthquake periods and can be accompanied by a substantial increase in precipitable water (total atmospheric water column precipitated as rain) vapor immediately after the event [22,23].
Global and local networks of ground-based monitoring stations are organized in earthquake-prone areas to monitor earthquake precursors. For example, the Global Seismographic Network, operated jointly by the National Science Foundation (NSF) and the U.S. Geological Survey (USGS), is a global network of 152 seismological and geophysical sensors that provides free, open access to real-time data [24]. The International Federation of Digital Seismograph Networks (FDSN) provides assistance in organizing observing networks for partner countries around the world [25]. Changes in geological, hydrogeological and temperature conditions in earthquake-prone areas are monitored by local ground observation networks organized at the national, regional or local level, such as the Southern California Seismic Network (SCSN) in the USA, the Monitoring of Waves on Land and Seafloor (MOWLAS) in Japan, etc. [26,27,28]. According to FDSN, the Turkish National Seismic Network, which consists of 267 stations with a mix of surface and borehole broadband seismometers and accelerometers (260 broadband seismometers, 12 surface accelerometers, 14 borehole instruments), provides official earthquake data for Turkey [29]. The main limitations of ground-based networks are the high cost of sensors, low coverage density, increased requirements for network stability, and the need to store and process large amounts of data [30]. Conversely, remote sensing data from space provide additional information on the Earth’s surface state prior to an earthquake in areas with a low density of ground sensor networks.
Satellite observations of seismic activity zones include assessing the vertical displacement of the Earth’s surface, tracking changes in the pattern of surface occurrence of deep fractures and faults, and monitoring changes in LST. Pre-earthquake surface deformation and vertical displacements (particularly subsidence) can be detected from time series of satellite Synthetic Aperture Radar (SAR) data, such as Sentinel-1, using multi-temporal Interferometric Synthetic Aperture Radar (InSAR) techniques [19,31]. Data from Landsat 8 and Landsat 9 satellites provide regular images of the Earth’s surface at medium spatial resolution (100 m) in the far infrared band [32]. In the absence of cloud cover, LST maps based on satellite data allow us to detect temperature anomalies preceding earthquakes of magnitude 5.5 Mw and above, observed in open areas without vegetation cover or buildings [11,33,34,35,36]. Temperature anomalies are reliable but not sufficient precursors of earthquakes because temperature extremes can be masked by vegetation cover, terrain effects, or climatic conditions [11].
Methods for studying lineaments in satellite images play an important role in earthquake prediction. Lineaments are linear features or patterns on the Earth’s surface that can be observed in satellite images. These linear features may vary in length, orientation and width and may be caused by a wide range of geological and tectonic factors. Lineaments can be the result of fault zones or zones of structural weakness, fractures or joints in the rock formations. Remote sensing techniques based on image analysis methods may be used for the detection and mapping of lineaments.
Prior to an earthquake, geological faults and their associated lineaments can undergo various changes and processes. Lineaments may widen or deform due to stress accumulation along the fault zone, intensifying the upward fluid movement. This can cause various changes in surrounding rock formations, including bending, fracturing and displacement. Such lineament changes can potentially precede earthquakes [37,38]. The study described below examines these changes by analyzing Landsat 8–9 satellite images.
The objective of this study is to evaluate an additional indicator of an approaching earthquake based on the analysis of multi-temporal multispectral satellite images. Landsat 8–9 images were used as input data, which have two considerable advantages—they contain the thermal infrared bands and are in the open access, allowing a wide range of specialists to use the method described below.
The article is organized as follows. The study area and input data are described in Section 2.1, Section 2.2 and Section 2.3 presents a proposed method of Landsat 8–9 LST contrast boundary detection and boundary concentration analysis as an indicator of an approaching earthquake. In Section 3, the performance of the proposed indicator is evaluated, and the maximum values of contrast boundary concentration are compared with earthquake epicenters. The peculiarities of the proposed method are analyzed in Section 4. Section 5 presents the conclusions.

2. Materials and Methods

2.1. Study Area

The East Anatolian Fault Zone is a major tectonic feature in eastern Turkey, marking the boundary between the Anatolian Plate and the Arabian Plate. The fault zone is a complex combination of normal and strike-slip faults making the East Anatolian Fault Zone particularly prone to producing large earthquakes.
The East Anatolian Fault Zone has been responsible for several major earthquakes, including the 1983 Erzurum earthquake (6.9 Mw), the 1992 Erzincan earthquake (6.8 Mw), the 2010 Elazığ earthquake (6.1 Mw) and the 2020 Sivrice earthquake (6.8 Mw). These earthquakes caused substantial damage and loss of life and had far-reaching social, economic and environmental impacts.
The study area of 6130 km2 (Figure 1b) is located near the Nurdagı Municipality of Gaziantep Province at the junction of the Anatolian Fault Zone (Figure 1a) and the Dead Sea Transform, making it extremely seismically active. The study area crosses a tectonic fault, which is part of the Anatolian Fault Zone (Figure 1a), in the north southwest direction along the Uzunziyaret mountain range.
Two epicenters of the strongest and most destructive earthquakes are located in the study area (Figure 1b):
  • Epicenter 1: Atalar (coordinates: 37.17° N, 37.03° E), 6 February 2023, 7.8 Mw;
  • Epicenter 2: Nurdagı (coordinates: 37.13° N, 36.94° E), 6 February 2023, 6.7 Mw.
From 3 to 8 February, 24 earthquakes with magnitudes between 4.2 and 5.6 Mw were observed in the study area. In the following month, 16 earthquakes with magnitudes between 4.0 and 4.7 Mw were observed (Figure 1b).

2.2. Data

LST maps derived from Landsat 8–9 data with cloud cover up to 10% were the main input data used in this study. LST is the Earth’s surface temperature measured to a few centimeters deep [39]. Landsat 8–9 satellites use the Thermal InfraRed Sensor (TIRS) that measure the amount of thermal radiation emitted from the Earth’s surface at 100 m spatial resolution in two spectral bands: Thermal Infrared 1 (10.60–11.19 μm) and Thermal Infrared 2 (11.50–12.51 μm) [40]. USGS Landsat 8 and Landsat 9 Level 2, Collection 2, Tier 1 datasets provided by Google Earth Engine service were used in this study. The LST product is generated using a single-channel algorithm developed jointly by the Rochester Institute of Technology (RIT) and the National Aeronautics and Space Administration (NASA) Jet Propulsion Laboratory (JPL) [41]. According to this algorithm, the temperature values recorded in each pixel of a thermal image are first converted to brightness values, which are then corrected for atmospheric effects and converted to temperature values using empirical or physical models. The LST values obtained can then be presented as a continuous map of the temperature values at the Earth’s surface. Landsat 8–9 images were acquired from January 2020 to January 2024 (Figure 2) and most of them were taken at 08:09 UTC or 11:09 UTC+3 local time. Sensing dates (Table 1) were chosen during the autumn-winter season to minimize the effect of vegetation cover on the spatial distribution of temperatures.
Ground data were collected at the closest meteorological station to verify the results of the satellite data analysis: OĞUZELI AIRPORT STATION (36.97° N, 37.51° E), observation time 11:20.
  • Ground-based temperature data are publicly available from the Weather Underground service [42]. Using ground-based temperature data, we aimed to minimize the influence of solar radiation and maximize the contribution of subsurface heat.
  • Information on the epicenters (coordinates, date and magnitude) of earthquakes that occurred on 6–8 February 2023 and in the subsequent period, provided by the USGS Earthquake Hazards Program, was used as input data [43].

2.3. Methodology

This study investigates the relationship between variations in temperature distribution and earthquake epicenters in three observation periods: three years to a few months before the earthquake, immediately before or after the earthquake (January and February 2023), and one year after the earthquake (January 2024).
There have been several previous studies using Landsat thermal imagery to investigate the relationship between earthquakes and surface temperature changes in Turkey. For example, thermal anomalies from Landsat and MODIS thermal data are associated with the 2011 Van earthquake in Turkey with a magnitude of 7.2 Mw [44]. Severe temperature anomalies were found to be associated with the earthquake based on changes in LST before and after the earthquake. The LST increased substantially in the days before the earthquake and remained high for several days after the earthquake. Thermal anomalies were found to be consistent with fault location and earthquake seismic activity. The earthquake was preceded by a substantial increase in LST, which may have been related to stress accumulation and deformation of the Earth’s crust prior to the earthquake.
Landsat 8–9 Thermal Infrared 1 data were pre-processed to mask cloud cover and cloud shadows and to fill in the gaps [45]. Ground-based data (Table 1) were used for an additional temperature correction. This correction is necessary to remove as much of the influence of solar heating as possible, so that only the subsurface temperature is used.
To detect temperature variations of the Earth’s surface, this study proposes the creation of concentration maps of the LST contrast boundary as a tool to analyze temperature changes before and after a series of earthquakes (Figure 3).
Computer detection of LST contrast boundaries from satellite images is a complex process that usually consists of two stages: identifying the contrast boundaries of the image and combining their linear fragments into extended lineaments.
Algorithms for automatic lineaments detection require the specification of many parameters that radically affect the results.
In this regard, it is suggested to use the first, simpler stage, which includes the detection and analysis of contrast boundaries. This makes lineament analysis faster and less subjective, without losing important information.
When processing satellite images, the Canny algorithm can be used to detect the edges of natural features such as geological formations, vegetation, lakes, as well as linear features such as rivers or geological boundaries. The optimal Canny algorithm was developed by John Canny in 1986 and is widely used in computer vision and image processing applications [46]. Since all these objects are controlled by faults in the Earth’s crust, the Canny algorithm makes it possible to identify linear boundaries associated with fault structures.
The concentration of LST contrast boundaries is the total length in meters of LST boundaries in the central pixel of a sliding neighborhood of a certain size divided by the neighborhood area. A window of 2.5 × 2.5 km has been used in this study. The contrast boundary concentration is estimated assuming that it is maximum in the vicinity of earthquake epicenters, allowing further use of this indicator for predicting approaching earthquakes.
The computational steps were implemented using the GIS RAPID 3.2 software [47].

3. Results

3.1. Studying the Heating of the Earth’s Crustal Surface

A rise in the Earth’s surface temperature is one of the most reasonable and intuitively expected indicators of an approaching earthquake. The accumulation of stress in the Earth’s crust and an increase in its permeability can result in surface heating.
In general, Landsat 8–9 thermal data can be useful for detecting and monitoring thermal anomalies associated with earthquakes in Turkey and other regions. The February 2023 earthquakes were also accompanied by an increase in surface temperature. Figure 4 shows the increase in mean LST values from 13.70 °C on 5 January 2023 (Figure 4a) to 17.59 °C on 21 January 2023 (Figure 4b).
At the same time, following a series of large earthquakes on 6–8 February 2023 and the release of stresses in the Earth’s crust, the surface temperature continued to rise to mean LST values of 18.96 °C on 22 February 2023, as shown in Figure 4c. On the one hand, this may indicate a potential threat of new destructive earthquakes in this region. On the other hand, during the month after 8 February 2023, seismic activity in the region gradually decreased, despite the increase in surface temperature, which is probably related to ongoing tectonic movements that do not cause strong earthquakes but are accompanied by fluid transfer from the depths.
There is therefore a need for another indicator that is more flexible and reflects the situation before and after earthquakes.

3.2. Changes of LST Contrast Boundaries Concentration

Figure 5 shows a fragment of the Landsat thermal infrared band, and the contrast boundaries detected on it.
By analyzing these contrast boundaries over time, we will try to identify patterns of change that can be used to predict seismic activity.
Before major earthquakes, tectonic plates move, and stresses build up in the Earth’s crust, changing groundwater levels, increasing pressure and fluid flow, slight uplift or subsidence of the Earth’s surface, and other geophysical changes that can affect the stability of geological structures and cause changes in the number, length and orientation of lineaments [48,49]. In addition to analyzing the entire set of contrast boundaries, their oriented fragments were also studied, because differently oriented geological faults and their corresponding lineaments can undergo varying changes prior to earthquakes.
Figure 6 shows the total length of oriented contrast boundaries in Landsat thermal infrared images taken before and after a series of catastrophic earthquakes on 6–8 February 2023.
Calculation results showed that the cumulative length of contrast boundaries with azimuths of 0° and 90° gradually increased by 14,335 and 10,162 km, respectively, as earthquakes approached and decreased sharply (by 5674 and 32,132 km, respectively) after seismic events. At the same time, the cumulative length of boundary fragments with other azimuths changed less.
For further analysis, maps were produced showing the concentration (total length in meters) of detected boundaries within a sliding window of 2.5 × 2.5 km (Figure 7).
As earthquakes approach, the concentration of contrast boundaries at 0° azimuth (Figure 7a–c) and 90° azimuth (Figure 7d–f) gradually increase, while the number of boundaries of the other orientation remains almost stable (Figure 6). Combining these maps gives the generalized maps of Figure 7g–i. The subsequent analysis was based on these maps. They expressively show an increase in the number of distinct contrast boundaries as a series of catastrophic earthquake approaches on 6–8 February 2023.
The accuracy of the result was evaluated as the percentage of earthquake epicenters within a radius of up to 5 km from the zones of maximum concentration of contrast boundaries at azimuths of 0° and 90°. In anticipation of the major earthquakes on 6–8 February 2023, the accuracy was 92%.
The concentration of contrast boundaries was calculated for three samples of 16 2.5 × 2.5 km sites: (1) centered on earthquake epicenters, (2) within areas more than 5 km away from epicenters, and (3) within randomly selected areas. Figure 8 shows the average concentration of contrast boundaries (in meters) at earthquake epicenters (orange line), at distances greater than 5 km from earthquake epicenters (green line), and at random points uniformly distributed over the study area (blue line) for the time series of LST data. The red bar chart in Figure 8 shows the number of earthquake epicenters of magnitude greater than 4 Mw that occurred during one month after the corresponding Landsat image acquisition date.
When earthquakes of a magnitude greater than 4 Mw approach, the concentration of contrast boundaries increase over the entire adjacent area (Figure 8). From the point of view of earthquake forecasting, this fact has both positive and negative sides: it is impossible to precisely determine the location of the epicenter of the approaching earthquake, but at the same time, to establish the fact of increasing seismic activity, it is sufficient to conduct regular studies at a small number of random sites in the area.
Despite the impossibility of reliably locating the epicenter of a future earthquake, we can conclude from the analysis of Figure 8 that the higher the average value of the contrast boundaries concentration with azimuths of 0° and 90° at a particular point, the higher the hazard of seismic events.

4. Discussion

Figure 6 shows that the cumulative length of contrast boundaries increases as the earthquake approaches and then decreases. This generally corresponds to a gradual decrease in seismic activity for at least three months after 6–8 February 2023. Unfortunately, after February 2023, Landsat images are covered by a thick cloud cover, making it impossible to extract contrast boundaries with acceptable accuracy, and summer images are difficult to compare with winter images due to the seasonal factor.
There are at least four important questions to be answered in the discussion of the results:
  • Is there a spatial relationship between the localization of earthquake epicenters and anomalies in the contrast boundaries concentration map?
  • Are there any other indicators calculated from Landsat satellite images (e.g., spectral indices) that also increased before the February 2023 earthquakes and then decreased?
  • Do maps show the location of aftershock epicenters?
  • What is the hazard of new major earthquakes one year on?
However, it is possible to analyze more recent images acquired in autumn 2023 and winter 2024 to help consider possible answers.
Figure 9 helps to answer the first question. For each of the multi-temporal images, two indicators were calculated: the average concentration of the contrast boundaries only in pixels above the epicenters and in all pixels of the image.
Both these two indicators themselves and the difference between them increase with the approach of earthquakes and decrease sharply after them. In other words, the concentration of contrast boundaries at the locations where earthquakes occurred on 6–8 February 2023 increased faster than in the study area as a whole. The concentration of boundaries, both above the epicenters and across the study area, decreases sharply after stress relief.
Answering the second question requires too much calculation, given that several hundred spectral indices are known (https://www.indexdatabase.de/). One of the indicators based on them is the Soil Moisture Index (SMI) [50].
SMI is modelled based on a linear relationship between the Normalized Difference Vegetation Index (NDVI) and LST derived from Landsat 8–9 images (Figure 10). It is calculated from the Thermal–Optical TRAapezoid Model (TOTRAM). SMI values range from 0 (dry soil) to 1 (wet soil) [50]. Increased rock fracturing causes increased rock moisture [51]. The index is highly dependent on vegetation conditions, but images taken in January can be compared correctly.
The difference between the maps generated from the data of 5 January 2023 (Figure 10a) and 21 January 2023 (Figure 10b) is quite noticeable, but it is of the opposite nature to that expected: as the date of the earthquake approaches, the SMI values do not increase but decrease.
Calculations with some other vegetation indices did not give convincing results either.
Figure 11 compares concentration maps of contrast boundary fragments with azimuths of 0° and 90° from 21 January 2023 and 22 February 2023, with plotted epicenters of aftershocks. The dots indicate an aftershock magnitude greater than 4 Mw between 9 February 2023 and 9 March 2023.
Both maps show the spatial association of aftershocks with areas of increased concentration of contrast boundaries. Despite a decrease in the concentration values after the earthquake, the aftershock epicenters are clearly located in the zones of local maxima of contrast boundary concentration. This may provide indirect evidence for the proposed indicator.
Finally, an image of the study area taken one year after the series of earthquakes (24 January 2024) was examined. Figure 12 shows the LST values obtained on 21 January 2023 and 24 January 2024. There is a noticeable decrease in surface temperature, but this cannot be an indicator of a reduced hazard of new earthquakes (as mentioned above, LST values on 22 February 2023 are even higher than on 21 January 2023, just before the earthquakes).
The concentration map of boundaries with azimuths of 0° and 90° is much more informative and its mean values decrease from 554.40 to 509.47 m. In 2024, severe hotspots of seismic activity are found only in the northern part of the study area, while they are almost absent in the south-western part. It shows that, in general, the hazard of new large earthquakes in the study area has decreased during the year. If, as suggested above, the concentration of contrast boundaries increases as earthquakes approach, their decrease in 2024 may indicate a reduced hazard of new large earthquakes in the study area.
However, Figure 7 shows that there was a slight increase in the hazard of new earthquakes in the year following the 6–8 February 2023 events.
The results obtained were compared with previous studies of the 2023 Turkish earthquakes. An important precursor of the 2023 Kahramanmarash earthquake doublet of magnitude 7.8–7.6 Mw is the difference of the b-values of the Gutenberg–Richter law parameter from the background values (Δb) in combination with the cumulative migration method (CMP), which quantitatively describes the migration of seismic activity [52,53]. A decrease in b-values is observed a decade before the earthquake doublet, and a high probability distribution of CMP is present near the points of earthquake doublet nucleation, allowing this approach to be used as a reliable predictor of increased differential crustal stress in the preparatory process of earthquake doublet [52]. Combined with the fractal dimension and energy rate, the b-values reveal the existence of a long-lasting preparatory phase that starts approximately eight months before the 2023 Kahramanmarash earthquake. Determining the beginning of such a phase, characterized by steady fault behavior, helps to mitigate seismic hazards [54]. Cluster analysis of Turkish seismicity based on the Turkish Homogenised Earthquake Catalogue (TURHEC) with comparison of local and global coefficients of variation of inter-event times of crustal seismicity has revealed that areas subjected to large seismic events in the last century tend to have globally clustered and locally Poissonian seismic activity. Higher values of the global coefficient of variation of inter-event times can be used to predict the occurrence of large earthquakes [55]. The potential role of fluid chemistry in contributing to seismic activity processes in Turkey is also validated. It is confirmed that according to the lithosphere-atmosphere-ionosphere coupling (LAIC) models, the precursors of earthquakes on 6 February 2023, Turkey, were anomalies that appeared to progress from the lithosphere upward through the atmosphere to the ionosphere [56].
In combination with existing approaches, high LST contrast boundaries may be used as a predictive indicator of approaching earthquakes.

5. Conclusions

For the study area of the East Anatolian Fault Zone, the use of land surface temperature boundary concentration maps from Landsat 8–9 satellite images, especially at 0° and 90° azimuth, is proposed and justified as an indicator of earthquake hazard. This indicator has a simple and clear physical meaning: as earthquakes approach, crustal stress and seismic activity increase, leading to an increase in heat flows from the depths to the surface. As a result, the pressure on near-surface regions increases, resulting in the extension of geological faults and fractures and their improved delineation on images.
Of interest is the fact that the main fault structures of the study area have subdiagonal azimuths, but the change in the number of contrast boundaries of latitudinal and meridional directions is more pronounced. This issue requires further study.
We propose that this indicator may be used for seismic hazard assessment in combination with other indicators currently in use (except for land surface temperature, which has proven to be insufficiently reliable at this location).
From the results it can be concluded that the hazard of new large earthquakes in the study area is gradually increasing but has not yet reached the recorded earthquake in January 2023. Analysis of the results (Figure 8 and Figure 12) suggests that the hazard of new large earthquakes at the beginning of 2024 was low.

Author Contributions

Conceptualization, S.N., K.S., O.K. and V.K.; methodology, S.N. and K.S.; software, S.N.; validation, S.N., K.S., O.K. and V.K.; formal analysis, S.N.; investigation, S.N., K.S., O.K. and V.K.; data curation, K.S.; writing—original draft preparation, S.N. and K.S.; writing—review and editing, S.N., K.S., O.K. and V.K.; visualization, S.N. and K.S. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

The data used to support the findings of this study are described in this article.

Acknowledgments

The authors would like to thank the National Aeronautics and Space Administration (NASA), and the United States Geological Survey (USGS) for providing Landsat images and earthquake data.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Dal Zilio, L.; Ampuero, J.P. Earthquake doublet in Turkey and Syria. Commun. Earth Environ. 2023, 4, 71. [Google Scholar] [CrossRef]
  2. Karabulut, H.; Güvercin, S.E.; Hollingsworth, J.; Konca, A.Ö. Long silence on the East Anatolian Fault Zone (Southern Turkey) ends with devastating double earthquakes (6 February 2023) over a seismic gap: Implications for the seismic potential in the Eastern Mediterranean region. J. Geol. Soc. 2023, 180, jgs2023-021. [Google Scholar] [CrossRef]
  3. Provost, F.; Karabacak, V.; Malet, J.P.; Van der Woerd, J.; Meghraoui, M.; Masson, F.; Ferry, M.; Michéa, D.; Pointal, E. High-resolution co-seismic fault offsets of the 2023 Türkiye earthquake ruptures using satellite imagery. Sci. Rep. 2024, 14, 6834. [Google Scholar] [CrossRef] [PubMed]
  4. Scendoni, R.; Cingolani, M.; Tambone, V.; De Micco, F. Operational Health Pavilions in Mass Disasters: Lessons Learned from the 2023 Earthquake in Turkey and Syria. Healthcare 2023, 11, 2052. [Google Scholar] [CrossRef] [PubMed]
  5. Ghosh, A.; Holt, W.E.; Bahadori, A. Role of large-scale tectonic forces in intraplate earthquakes of central and eastern North America. Geochem. Geophys. Geosyst. 2019, 20, 2134–2156. [Google Scholar] [CrossRef]
  6. Cui, Y.; Huang, J.; Zeng, Z.; Zou, Z. CO Emissions Associated with Three Major Earthquakes Occurring in Diverse Tectonic Environments. Remote Sens. 2024, 16, 480. [Google Scholar] [CrossRef]
  7. Ning, L.; Hui, C.; Cheng, C. Exploring the Dynamics of Global Plate Motion Based on the Granger Causality Test. Appl. Sci. 2021, 11, 7853. [Google Scholar] [CrossRef]
  8. Nath, B.; Singh, R.P.; Gahalaut, V.K.; Singh, A.P. Dynamic Relationship Study between the Observed Seismicity and Spatiotemporal Pattern of Lineament Changes in Palghar, North Maharashtra (India). Remote Sens. 2022, 14, 135. [Google Scholar] [CrossRef]
  9. Das, D.; Mallik, J. Koyna earthquakes: A review of the mechanisms of reservoir-triggered seismicity and slip tendency analysis of subsurface faults. Acta Geophys. 2020, 68, 1097–1112. [Google Scholar] [CrossRef]
  10. Fitzenz, D.D.; Miller, S.A. A forward model for earthquake generation on interacting faults including tectonics, fluids, and stress transfer. J. Geophys. Res. Solid Earth 2001, 106, 26689–26706. [Google Scholar] [CrossRef]
  11. Nahornyi, V.V.; Pigulevskiy, P. Vibration forecast in Europe from the results of groundwater monitoring on the territory of Ukraine. MM Sci. J. 2022, 5926–5930. [Google Scholar] [CrossRef]
  12. Conti, L.; Picozza, P.; Sotgiu, A. A critical review of ground based observations of earthquake precursors. Front. Earth Sci. 2021, 9, 676766. [Google Scholar] [CrossRef]
  13. Picozza, P.; Conti, L.; Sotgiu, A. Looking for earthquake precursors from space: A critical review. Front. Earth Sci. 2021, 9, 676775. [Google Scholar] [CrossRef]
  14. Lee, H.A.; Hamm, S.-Y.; Woo, N.C. Pilot-Scale Groundwater Monitoring Network for Earthquake Surveillance and Forecasting Research in Korea. Water 2021, 13, 2448. [Google Scholar] [CrossRef]
  15. Sekertekin, A.; Inyurt, S.; Yaprak, S. Pre-seismic ionospheric anomalies and spatio-temporal analyses of MODIS Land surface temperature and aerosols associated with Sep, 24 2013 Pakistan Earthquake. J. Atmos. Sol.-Terr. Phys. 2020, 200, 105218. [Google Scholar] [CrossRef]
  16. Semenov, V.; Ladanivskyy, B.; Petrishchev, M. Emergence of earthquakes footprint in natural electromagnetic field variations. Geodynamics 2018, 2, 65–70. [Google Scholar] [CrossRef]
  17. Albano, M.; Chiaradonna, A.; Saroli, M.; Moro, M.; Pepe, A.; Solaro, G. InSAR Analysis of Post-Liquefaction Consolidation Subsidence after 2012 Emilia Earthquake Sequence (Italy). Remote Sens. 2024, 16, 2364. [Google Scholar] [CrossRef]
  18. Boudriki Semlali, B.-E.; Molina, C.; Park, H.; Camps, A. First Results on the Systematic Search of Land Surface Temperature Anomalies as Earthquakes Precursors. Remote Sens. 2023, 15, 1110. [Google Scholar] [CrossRef]
  19. Mahmood, I. Anomalous variations of air temperature prior to earthquakes. Geocarto Int. 2019, 36, 1396–1408. [Google Scholar] [CrossRef]
  20. Chalyi, O.; Diaconescu, M.; Gurova, I.; Lisovyi, Y.; Pigylevsky, P.; Shcherbina, S.; Shevtsov, A.; Shumlianska, L. The cause of high intensity of seismicity in Ukraine. Visnyk Taras Shevchenko Natl. Univ. Kyiv Geol. 2018, 4, 38–45. [Google Scholar] [CrossRef]
  21. Bhardwaj, A.; Singh, S.; Sam, L.; Joshi, P.K.; Bhardwaj, A.; Martín-Torres, F.J.; Kumar, R. A review on remotely sensed land surface temperature anomaly as an earthquake precursor. Int. J. Appl. Earth Obs. Geoinf. 2017, 63, 158–166. [Google Scholar] [CrossRef]
  22. Guo, A.; Xu, Y.; Jiang, N.; Wu, Y.; Gao, Z.; Li, S.; Xu, T.; Bastos, L. Analyzing correlations between GNSS retrieved precipitable water vapor and land surface temperature after earthquakes occurrence. Sci. Total Environ. 2023, 872, 162225. [Google Scholar] [CrossRef] [PubMed]
  23. Guo, A.; Jiang, N.; Xu, Y.; Xu, T.; Wu, Y.; Li, S.; Gao, Z. Co-seismic characterization analysis in PWV and land-atmospheric observations associated with Luding Ms 6.8 earthquake occurrence in China on September 5, 2022. Geomat. Nat. Hazards Risk 2023, 14, 2279494. [Google Scholar] [CrossRef]
  24. Ringler, A.T.; Steim, J.; Wilson, D.C.; Widmer-Schnidrig, R.; Anthony, R.E. Improvements in seismic resolution and current limitations in the Global Seismographic Network. Geophys. J. Int. 2020, 220, 508–521. [Google Scholar] [CrossRef]
  25. International Federation of Digital Seismograph Networks. Available online: https://www.fdsn.org/networks (accessed on 12 July 2024).
  26. Nolte, K.A.; Tsoflias, G.P.; Holubnyak, Y.; Raney, J.; Wreath, D. Designing monitoring networks for local earthquakes. J. Geophys. Eng. 2022, 19, 75–84. [Google Scholar] [CrossRef]
  27. Hauksson, E.; Yoon, C.; Yu, E.; Andrews, J.R.; Alvarez, M.; Bhadha, R.; Thomas, V. Caltech/USGS Southern California Seismic Network (SCSN) and Southern California Earthquake Data Center (SCEDC): Data availability for the 2019 Ridgecrest sequence. Seismol. Res. Lett. 2020, 91, 1961–1970. [Google Scholar] [CrossRef]
  28. Aoi, S.; Asano, Y.; Kunugi, T.; Kimura, T.; Uehira, K.; Takahashi, N.; Ueda, H.; Shiomi, K.; Matsumoto, T.; Fujiwara, H. MOWLAS: NIED observation network for earthquake, tsunami and volcano. Earth Planets Space 2020, 72, 126. [Google Scholar] [CrossRef]
  29. Alver, F.; Kılıçarslan, Ö.; Kuterdem, K.; Türkoğlu, M.; Şentürk, M.D. Seismic Monitoring at the Turkish National Seismic Network (TNSN). Summ. Bull. Int. Seismol. Cent. 2019, 53, 41–58. [Google Scholar] [CrossRef]
  30. Li, Z. Recent advances in earthquake monitoring I: Ongoing revolution of seismic instrumentation. Earthq. Sci. 2021, 34, 177–188. [Google Scholar] [CrossRef]
  31. Panchal, H.; Saraf, A.K.; Das, J.; Dwivedi, D. Satellite based detection of pre-earthquake thermal anomaly, co-seismic deformation and source parameter modelling of past earthquakes. Nat. Hazards Res. 2022, 2, 287–303. [Google Scholar] [CrossRef]
  32. USGS. Landsat Missions. Available online: https://www.usgs.gov/landsat-missions (accessed on 12 July 2024).
  33. Huda, D.N.; Shidiq, I.P.A. Spatiotemporal analysis land surface temperature in relation to earthquake occurrence around the cimandiri fault. In Proceedings of the IOP Conference Series: Earth and Environmental Science, Kuala Lumpur, Malaysia, 20–21 October 2020; Volume 540, p. 012069. [Google Scholar] [CrossRef]
  34. Jiao, Z.; Shan, X. A Bayesian Approach for Forecasting the Probability of Large Earthquakes Using Thermal Anomalies from Satellite Observations. Remote Sens. 2024, 16, 1542. [Google Scholar] [CrossRef]
  35. Ghosh, S.; Sasmal, S.; Maity, S.K.; Potirakis, S.M.; Hayakawa, M. Thermal Anomalies Observed during the Crete Earthquake on 27 September 2021. Geosciences 2024, 14, 73. [Google Scholar] [CrossRef]
  36. Pavlidou, E.; Van der Meijde, M.; Van der Werff, H.; Hecker, C. Time Series Analysis of Land Surface Temperatures in 20 Earthquake Cases Worldwide. Remote Sens. 2019, 11, 61. [Google Scholar] [CrossRef]
  37. Sichugova, L.; Fazilova, D. Study of the seismic activity of the Almalyk-Angren industrial zone based on lineament analysis. Int. J. Eng. Geosci. 2024, 9, 1–11. [Google Scholar] [CrossRef]
  38. Pappachen, J.P.; Hamdan, H.A.; Sathiyaseelan, R.; Darya, A.M.; Shanableh, A. Possible seismo-ionospheric anomalies of Mw 6.0 and 6.4 south Iran twin earthquakes on 14 November 2021 from GPS and ionosonde observations. Arab. J. Geosci. 2024, 17, 201. [Google Scholar] [CrossRef]
  39. Li, Z.L.; Wu, H.; Duan, S.B.; Zhao, W.; Ren, H.; Liu, X.; Leng, P.; Tang, R.; Ye, X.; Zhu, J.; et al. Satellite remote sensing of global land surface temperature: Definition, methods, products, and applications. Rev. Geophys. 2023, 61, e2022RG000777. [Google Scholar] [CrossRef]
  40. Boudriki Semlali, B.E.; Molina, C.; Park, H.; Camps, A. Association of land surface temperature anomalies from GOES/ABI, MSG/SEVIRI, and Himawari-8/AHI with land earthquakes between 2010 and 2021. Geomat. Nat. Hazards Risk 2024, 15, 2324982. [Google Scholar] [CrossRef]
  41. Google Earth Engine USGS Landsat 8 Level 2, Collection 2, Tier 1. Available online: https://developers.google.com/earth-engine/datasets/catalog/LANDSAT_LC08_C02_T1_L2#description (accessed on 12 July 2024).
  42. Weather Underground. Available online: https://www.wunderground.com (accessed on 12 July 2024).
  43. Jones, L.; Bernknopf, R.; Cox, D.; Goltz, J.; Hudnut, K.; Mileti, D.; Perry, S.; Ponti, D.; Porter, K.; Reichle, M.; et al. Earthquake Hazards Program; US Geological Survey Open File Report 2008-1150; USGS: Reston, VA, USA, 2008. Available online: https://pubs.usgs.gov/of/2008/1150/of2008-1150small.pdf (accessed on 12 July 2024).
  44. Ouzounov, D.; Pulinets, S.; Kafatos, M.C.; Taylor, P. Thermal radiation anomalies associated with major earthquakes. In Pre-Earthquake Processes: A Multidisciplinary Approach to Earthquake Prediction Studies; American Geophysical Union: Washington, DC, USA, 2018; pp. 259–274. [Google Scholar] [CrossRef]
  45. Shedlovska, Y.I.; Hnatushenko, V.V. Shadow removal algorithm with shadow area border processing. In Proceedings of the 2016 II International Young Scientists Forum on Applied Physics and Engineering (YSF), Kharkiv, Ukraine, 10–14 October 2016; Volume 24, pp. 164–167. [Google Scholar] [CrossRef]
  46. Nikulin, S.L.; Sergieieva, K.L.; Korobko, O.V. Computer detection of the Earth’s crust blocks using satellite image lineaments. In Proceedings of the Geoinformatics: Theoretical and Applied Aspects, Kyiv, Ukraine, 11–14 May 2020; Volume 2020, pp. 1–5. [Google Scholar] [CrossRef]
  47. Busygin, B.; Nikulin, S.; Sergieieva, K. Solving the tasks of subsurface resources management in GIS RAPID environment. Min. Miner. Depos. 2019, 13, 49–57. [Google Scholar] [CrossRef]
  48. Nur, A. The origin of tensile fracture lineaments. J. Struct. Geol. 1982, 4, 31–40. [Google Scholar] [CrossRef]
  49. Tannock, L.; Herwegh, M.; Berger, A.; Liu, J.; Regenauer-Lieb, K. The effects of a tectonic stress regime change on crustal-scale fluid flow at the Heyuan geothermal fault system, South China. Tectonophysics 2020, 781, 228399. [Google Scholar] [CrossRef]
  50. Burdun, I.; Bechtold, M.; Sagris, V.; Komisarenko, V.; De Lannoy, G.; Mander, Ü. A Comparison of Three Trapezoid Models Using Optical and Thermal Satellite Imagery for Water Table Depth Monitoring in Estonian Bogs. Remote Sens. 2020, 12, 1980. [Google Scholar] [CrossRef]
  51. Baik, H.; Son, Y.-S.; Kim, K.-E. Detection of Liquefaction Phenomena from the 2017 Pohang (Korea) Earthquake Using Remote Sensing Data. Remote Sens. 2019, 11, 2184. [Google Scholar] [CrossRef]
  52. Yin, F.; Jiang, C. Unraveling the Preparatory Processes of the 2023 M w 7.8–7.6 Kahramanmaraş Earthquake Doublet. Seismol. Res. Lett. 2024, 95, 730–741. [Google Scholar] [CrossRef]
  53. Bilim, F. The correlation of b-value in the earthquake frequency-magnitude distribution, heat flow and gravity data in the Sivas Basin, central eastern Turkey. Bitlis Eren Univ. J. Sci. Technol. 2019, 9, 11–15. [Google Scholar] [CrossRef]
  54. Picozzi, M.; Iaccarino, A.G.; Spallarossa, D. The preparatory process of the 2023 Mw 7.8 Türkiye earthquake. Sci. Rep. 2023, 13, 17853. [Google Scholar] [CrossRef] [PubMed]
  55. Zaccagnino, D.; Telesca, L.; Tan, O.; Doglioni, C. Clustering Analysis of Seismicity in the Anatolian Region with Implications for Seismic Hazard. Entropy 2023, 25, 835. [Google Scholar] [CrossRef] [PubMed]
  56. Cianchini, G.; Calcara, M.; De Santis, A.; Piscini, A.; D’Arcangelo, S.; Fidani, C.; Sabbagh, D.; Orlando, M.; Perrone, L.; Campuzano, S.A.; et al. The Preparation Phase of the 2023 Kahramanmaraş (Turkey) Major Earthquakes from a Multidisciplinary and Comparative Perspective. Remote Sens. 2024, 16, 2766. [Google Scholar] [CrossRef]
Earth 05 00022 g001
Figure 1. Study area in the East Anatolian Fault Zone: (a) major tectonic faults in the region. Basemap: Here Wego Terrain (EPSG:4326–WGS 84); (b) Landsat 8 Natural Color image of the study area acquired on 21 January 2023 and epicenters of the earthquakes in different time periods.
Figure 1. Study area in the East Anatolian Fault Zone: (a) major tectonic faults in the region. Basemap: Here Wego Terrain (EPSG:4326–WGS 84); (b) Landsat 8 Natural Color image of the study area acquired on 21 January 2023 and epicenters of the earthquakes in different time periods.
Earth 05 00022 g001
Earth 05 00022 g002
Figure 2. Landsat 8–9 image acquisition dates.
Figure 2. Landsat 8–9 image acquisition dates.
Earth 05 00022 g002
Earth 05 00022 g003
Figure 3. Constructing and analyzing LST contrast boundary concentration maps.
Figure 3. Constructing and analyzing LST contrast boundary concentration maps.
Earth 05 00022 g003
Earth 05 00022 g004
Figure 4. LST maps from Landsat 8 data before and after the earthquake 6–8 February 2023: (a) 5 January 2023; (b) 21 January 2023; and (c) 22 February 2023.
Figure 4. LST maps from Landsat 8 data before and after the earthquake 6–8 February 2023: (a) 5 January 2023; (b) 21 January 2023; and (c) 22 February 2023.
Earth 05 00022 g004
Earth 05 00022 g005
Figure 5. Example of contrast boundaries for Landsat 8 Thermal Infrared band 1 from 5 January 2023: (a) thermal infrared band 1; (b) contrast boundaries.
Figure 5. Example of contrast boundaries for Landsat 8 Thermal Infrared band 1 from 5 January 2023: (a) thermal infrared band 1; (b) contrast boundaries.
Earth 05 00022 g005
Earth 05 00022 g006
Figure 6. Cumulative length of oriented contrast boundaries (km).
Figure 6. Cumulative length of oriented contrast boundaries (km).
Earth 05 00022 g006
Earth 05 00022 g007
Figure 7. LST contrast boundary concentration maps at 0° and 90° azimuths before the 6–8 February 2023 earthquake: (a) 0°, 2 November 2022; (b) 0°, 5 January 2023; (c) 0°, 21 January 2023; (d) 90°, 2 November 2022; (e) 90°, 5 January 2023; (f) 90°, 21 January 2023; (g) 0° and 90°, 2 November 2022; (h) 0° and 90°, 5 January 2023; and (i) 0° and 90°, 21 January 2023.
Figure 7. LST contrast boundary concentration maps at 0° and 90° azimuths before the 6–8 February 2023 earthquake: (a) 0°, 2 November 2022; (b) 0°, 5 January 2023; (c) 0°, 21 January 2023; (d) 90°, 2 November 2022; (e) 90°, 5 January 2023; (f) 90°, 21 January 2023; (g) 0° and 90°, 2 November 2022; (h) 0° and 90°, 5 January 2023; and (i) 0° and 90°, 21 January 2023.
Earth 05 00022 g007
Earth 05 00022 g008
Figure 8. Concentration of contrast boundaries for 2.5 × 2.5 km areas centered on earthquake epicenters (orange line), at distances greater than 5 km from earthquake epicenters (green line), and at random sites uniformly distributed over the study area (blue line), and the number of earthquake epicenters of magnitude greater than 4 Mw that occurred within a month after the corresponding Landsat 8–9 image acquisition date (red bar chart).
Figure 8. Concentration of contrast boundaries for 2.5 × 2.5 km areas centered on earthquake epicenters (orange line), at distances greater than 5 km from earthquake epicenters (green line), and at random sites uniformly distributed over the study area (blue line), and the number of earthquake epicenters of magnitude greater than 4 Mw that occurred within a month after the corresponding Landsat 8–9 image acquisition date (red bar chart).
Earth 05 00022 g008
Earth 05 00022 g009
Figure 9. Average concentration of LST contrast boundaries at 0° and 90° azimuth (m).
Figure 9. Average concentration of LST contrast boundaries at 0° and 90° azimuth (m).
Earth 05 00022 g009
Earth 05 00022 g010
Figure 10. SMI maps before the 6–8 February 2023 earthquake: (a) 5 January 2023; (b) 21 January 2023.
Figure 10. SMI maps before the 6–8 February 2023 earthquake: (a) 5 January 2023; (b) 21 January 2023.
Earth 05 00022 g010
Earth 05 00022 g011
Figure 11. Fragment of concentration maps of contrast LST boundary at 0° and 90° azimuths: (a) 21 January 2023; (b) 21 February 2023.
Figure 11. Fragment of concentration maps of contrast LST boundary at 0° and 90° azimuths: (a) 21 January 2023; (b) 21 February 2023.
Earth 05 00022 g011
Earth 05 00022 g012
Figure 12. LST values for images from (a) 21 January 2023 and (b) 24 January 2024; contrast boundaries concentrations at 0° and 90° azimuth for images from (c) 21 January 2023 and (d) 24 January 2024.
Figure 12. LST values for images from (a) 21 January 2023 and (b) 24 January 2024; contrast boundaries concentrations at 0° and 90° azimuth for images from (c) 21 January 2023 and (d) 24 January 2024.
Earth 05 00022 g012
Table 1. Landsat 8–9 image acquisition date and time, and ground temperature from nearest meteorological stations.
Table 1. Landsat 8–9 image acquisition date and time, and ground temperature from nearest meteorological stations.
Landsat 8–9 Acquisition Date and TimeGround-Based Temperature (°C)
24 January 202408:09:32.80Z9.0
28 October 202308:09:42.73Z24.0
22 February 202308:09:37.88Z12.0
21 January 202308:09:50.18Z11.0
5 January 202308:09:55.33Z10.0
4 December 202208:10:04.26Z15.0
2 November 202208:10:09.37Z19.0
25 October 202208:09:57.91Z22.0
15 September 202208:15:58.31Z28.0
19 February 202208:09:32.84Z11.0
15 November 202108:09:51.53Z20.0
30 December 202008:09:48.01Z13.0
13 January 202008:09:41.57Z10.0
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Nikulin, S.; Sergieieva, K.; Korobko, O.; Kashtan, V. Using the Contrast Boundary Concentration of LST for the Earthquake Approach Assessment in Turkey, 6–8 February 2023. Earth 2024, 5, 388-403. https://doi.org/10.3390/earth5030022

AMA Style

Nikulin S, Sergieieva K, Korobko O, Kashtan V. Using the Contrast Boundary Concentration of LST for the Earthquake Approach Assessment in Turkey, 6–8 February 2023. Earth. 2024; 5(3):388-403. https://doi.org/10.3390/earth5030022

Chicago/Turabian Style

Nikulin, Serhii, Kateryna Sergieieva, Olga Korobko, and Vita Kashtan. 2024. "Using the Contrast Boundary Concentration of LST for the Earthquake Approach Assessment in Turkey, 6–8 February 2023" Earth 5, no. 3: 388-403. https://doi.org/10.3390/earth5030022

APA Style

Nikulin, S., Sergieieva, K., Korobko, O., & Kashtan, V. (2024). Using the Contrast Boundary Concentration of LST for the Earthquake Approach Assessment in Turkey, 6–8 February 2023. Earth, 5(3), 388-403. https://doi.org/10.3390/earth5030022

Article Metrics

Back to TopTop