Measurement of Radon Gas in the Vicinity of the Adalar Segment of the North Anatolian Fault

F4-1 2nd KOCAELI SCIENCE CONGRESS (KOSC-2025) Kocaeli University, Faculty of Arts and Sciences November 19-21, 2025, İzmit, Kocaeli, Türkiye Measurement of Radon Gas in the Vicinity of the Adalar Segment of the North Anatolian Fault Ali Beyaz1, Caner Yalçın1, Ezgi Tantoğlu1, Osman Günay2 1Department of Physics, Kocaeli University, 41001, Kocaeli, TÜRKİYE 2Department of Biomedical Engineering, Yıldız Technical University, 34220 Istanbul, Turkey Corresponding author: [email protected] ORCID IDs: First Author: 0009-0007-2927-2279 Second Author: 0000-0002-3105-7267 Third Author: 0009-0009-4905-5735 Fourth Author: 0000-0003-0760-554X DOI : 10.5281/zenodo.18038250 Abstract The prediction of seismic events encompasses a range of parameters, including magnitude, geographical location, and temporal considerations. While the determination of location and magnitude can be made to a certain degree in the present, temporal predictions must be expressed in long intervals. This underscores the critical importance of short-term prediction. A substantial body of research has identified a correlation between variations in radon gas concentrations, both prior to and following seismic events, and seismic activity. A major earthquake is anticipated in the Adalar and Avcılar segments of the North Anatolian Fault, situated within the Sea of Marmara (a region previously impacted by the 1912 Ganos and 1999 Kocaeli earthquakes). In this study, radon concentration and temperature, humidity, and pressure data were collected in a continuous manner. This data was collected using a station that was to be established on the Anatolian plate in the Adalar segment. The relationship between radon anomalies and earthquakes was investigated using these data. The cause of the radon anomalies that occurred was determined, and their relationship with geological activity was analyzed. The rise in radon concentration that began on August 11, 2024, and persisted until the 15 August 2024, 06:20 magnitude 3.8 earthquake in Mudanya, Bursa, has been linked to the seismic event. Keywords: Soil radon concentration, Earthquake, Alpha guard radon monitor. 1. Introduction and Motivation According to the prevailing methodology for identifying precursors to earthquakes, the utilization of any physical or geochemical parameter in earthquake prediction is contingent upon the manifestation of continuous and reproducible anomalies in seismically active regions prior to seismic events [1]. In this context, a range of parameters representing diverse physical processes have been evaluated in earthquake prediction studies. Among geochemical parameters, soil gas emissions, particularly radon (Rn-222), are among the most commonly studied variables as earthquake precursors [2,3]. Radon (Rn) F4-2 2nd Kocaeli Science Congress, November 19-21, 2025 is a radioactive noble gas that is chemically inert and has a half-life of 3.8 days. It is formed by the decay of 226Ra, which is found in soil and rock [4]. Its capacity to be transported through diffusion and convection along cracks in rocks, fault zones, and porous environments renders radon a sensitive tracer of seismotectonic processes. Consequently, radon gas has become a prevalent method for the observation of stress–strain processes within the Earth's crust, particularly in active fault zones, geothermal fields, and seismic regions [5,6]. In this study, continuous radon measurements were taken at a depth of 1 meter in Yalova, Türkiye between July 23, 2024, and December 23, 2024, along with meteorological parameters. The relationship between meteorological data and radon was observed, and the effects of earthquakes greater than magnitude 3 that occurred during this period on radon concentration were investigated. 2. Material and Methods The radon measurement station utilized in this study is situated in Yalova (40.651187°N, 29.225013°E), which is located in a seismically active region of the Marmara Region. The surface geology of the study area is characterized by a sandy-loamy structure, which renders the area particularly conducive to soil-gas radon studies due to its relatively high gas permeability. Radon measurements were performed at a depth of 1 meter using a soil gas probe compatible with the AlphaGUARD radon monitor. This measurement geometry enables direct observation of the soil gas regime, independent of surface ventilation and indoor environment effects. Seismic activity data is comprised of earthquakes occurring within a radius of 100 km from the measuring station. In the course of the evaluation, events with magnitude information available in the earthquake catalog were included, and an M ≥ 3.0 threshold was used in the analyses. The temporal parameters of seismic occurrences, including time of day, magnitude, and geographical location, were meticulously matched with radon time series for the purpose of comparison. The notion of an earthquake preparedness zone was initially delineated in 1979 by Russian scientist Ilya Dobrovolsky and his associates [7]. This concept refers to a spatial area that is tectonically stressed and where precursor processes associated with an impending earthquake can be observed. It has been posited that alterations in the physical and chemical parameters of groundwater (level, flow rate, ionic composition, and temperature), irregularities in the Earth's magnetic field, various gas emissions, and analogous geophysical and geochemical anomalies may occur within this zone. Dobrovolsky and colleagues proposed the following empirical relationship for the radius, which varies depending on the moment magnitude of the earthquake, in order to define the characteristic size of the earthquake preparation zone: R=100.43M (1) where R represents the radius of the earthquake preparation zone (km) and M represents the magnitude of the earthquake. 3. Results and Discussion During the measurement period, a total of three earthquakes with a magnitude of M ≥ 3.0 were recorded within a 100 km radius area centered on the measurement station. The temporal, spatial, and seismological parameters of these earthquakes, along with the radius of the impact area for each earthquake and the hypocenter distances from the measurement station, are presented in detail in Table 1. A thorough examination of Table 1 indicates that only one of these earthquakes (occurring on August 15, 2024, at 3:20:31 a.m.) is situated within the designated earthquake impact area, as delineated by the established magnitude-distance relationship. The epicenter location of the earthquake is illustrated in Figure 1. Consequently, the direct seismic effect on radon concentration observations made at the measuring station is assessed to be limited (with the exception of the earthquake on F4 - 3 August 15, 2024, at 03:20:31). It is concluded that the observed radon changes are predominantly influenced by local geological conditions and meteorological parameters. Table 1. Earthquakes with a magnitude greater than 3 were recorded during the measurement pe Time M Latitude Longitude 2024-08-03 02:05:35 4.2 39.8697 28.7602 2024-08-15 03:20:31 3.8 40.3943 28.9335 2024-11-05 12:00:00 3.3 40.4007 28.8472 Fig. 1. The epicenter of the seismic event that took place on August 15th, 2024, at 3:20:31 a.m., was located at Mudanya, Bursa. The measuring station is located at the The radon concentration obtained during the measurement period and the simultaneous meteorological parameters (temperature, relative humidity, and atmospheric pressure) are presented in Figure 2 along with their temporal variat ions. A thorough examination of Figure 2 reveals a discernible diurnal fluctuation in radon concentrations. However, it has been observed that radon concentration displays an inverse relationship with temperature and a parallel relationship with atmospheri relative humidity. To quantitatively evaluate these relationships, the linear relationship between radon and meteorological variables was calculated using the Pearson correlation coefficient. The results are presented in Table 2. In addition , fundamental statistical analyses were conducted on the data set to ascertain its general characteristics. These analyses included the calculation of the mean, standard deviation, minimum, and maximum values, among others. The results of these analyses ar comprehensively outlined in Table 3. 2nd Kocaeli Science Congress, November 19 15, 2024, at 03:20:31). It is concluded that the observed radon changes are predominantly influenced by local geological conditions and meteorological parameters. Earthquakes with a magnitude greater than 3 were recorded during the measurement pe Longitude Depth (km) Location Dobrovolsky R (km) Hypocenter 28.7602 6.3 GaziolukOrhaneli (Bursa) 64 28.9335 12.1 Mudanya (Bursa) 43 28.8472 5.8 KumyakaMudanya (Bursa) 26 The epicenter of the seismic event that took place on August 15th, 2024, at 3:20:31 a.m., was located at Mudanya, Bursa. The measuring station is located at the center of the blue circles. The radon concentration obtained during the measurement period and the simultaneous meteorological parameters (temperature, relative humidity, and atmospheric pressure) are presented in Figure 2 along ions. A thorough examination of Figure 2 reveals a discernible diurnal fluctuation in radon concentrations. However, it has been observed that radon concentration displays an inverse relationship with temperature and a parallel relationship with atmospheri relative humidity. To quantitatively evaluate these relationships, the linear relationship between radon and meteorological variables was calculated using the Pearson correlation coefficient. The results are , fundamental statistical analyses were conducted on the data set to ascertain its general characteristics. These analyses included the calculation of the mean, standard deviation, minimum, and maximum values, among others. The results of these analyses ar comprehensively outlined in Table 3. Congress, November 19 - 21, 2025 15, 2024, at 03:20:31). It is concluded that the observed radon changes are predominantly Earthquakes with a magnitude greater than 3 were recorded during the measurement pe riod. Hypocenter Distance (km) 95.4 37.7 42.4 The epicenter of the seismic event that took place on August 15th, 2024, at 3:20:31 a.m., was center of the blue circles. The radon concentration obtained during the measurement period and the simultaneous meteorological parameters (temperature, relative humidity, and atmospheric pressure) are presented in Figure 2 along ions. A thorough examination of Figure 2 reveals a discernible diurnal fluctuation in radon concentrations. However, it has been observed that radon concentration displays an inverse relationship with temperature and a parallel relationship with atmospheri c pressure and relative humidity. To quantitatively evaluate these relationships, the linear relationship between radon and meteorological variables was calculated using the Pearson correlation coefficient. The results are , fundamental statistical analyses were conducted on the data set to ascertain its general characteristics. These analyses included the calculation of the mean, standard deviation, minimum, and maximum values, among others. The results of these analyses ar e F4 - 4 Fig.2. Time series of (a) soil-gas 222 Rn concentration measured at 1 m depth, (b) air temperature, (c) air pressure, and (d) relative humidity recorded simultaneously by the AlphaGUARD system. Vertical dashed lines indicate earthquakes with magnitudes M the monitoring station. Table 2. Pearson correlation coefficients for radon and meteorological data. R adon Radon 1 Temperature - 0.659 Humidity 0.393 Pressure 0.47 Table3. F undamental statistical analysis of radon and meteorological data Count Mean S Radon(kBq/m 3 ) 2950 24.136 Temperature (C) 2950 24.244 Humidity (%) 2950 61.224 Pressure (mbar) 2950 1014.076 2nd Kocaeli Science Congress, November 19 Rn concentration measured at 1 m depth, (b) air temperature, (c) air pressure, and (d) relative humidity recorded simultaneously by the AlphaGUARD monitoring system. Vertical dashed lines indicate earthquakes with magnitudes M ≥ 3 occurring within 100 km of Pearson correlation coefficients for radon and meteorological data. adon Temperature Humidity P ressure -0.659 0.393 0.47 0.659 1 -0.725 - 0.693 0.393 -0.725 1 0.33 0.47 -0.693 0.33 1 undamental statistical analysis of radon and meteorological data S td. Dev. Min Max Variance S kewness 12.467 10.007 55.808 155.414 0.47 6.115 11.067 37.433 37.399 - 0.221 8.07 35.833 82 65.129 - 0.084 6.767 998.783 1031.85 45.791 0.713 Congress, November 19 - 21, 2025 Rn concentration measured at 1 m depth, (b) air temperature, (c) monitoring ≥ 3 occurring within 100 km of Pearson correlation coefficients for radon and meteorological data. ressure 0.693 undamental statistical analysis of radon and meteorological data kewness Kurtosis -0.685 0.221 -1.058 0.084 -0.166 0.713 -0.256 F4-5 2nd Kocaeli Science Congress, November 19-21, 2025 Pearson correlation analysis was applied to quantitatively evaluate the linear relationships between radon concentration and meteorological parameters. The correlation coefficients obtained are presented in Table 2. The analysis results indicate a moderate negative correlation (r=−0.659) between radon concentration and temperature. This finding suggests that gas transport mechanisms undergo changes in response to variations in temperature within surface and shallow subsurface environments. These changes can be associated with the weakening of convective transport with increasing temperature or alterations in permeability resulting from increased soil moisture. A positive yet relatively weak correlation (r=0.393) was observed between radon and relative humidity, while a moderate positive relationship (r=0.470) was found between atmospheric pressure and radon concentration. This phenomenon aligns with the barometric pumping effect, a widely documented phenomenon in the relevant literature. The barometric pumping effect posits that alterations in atmospheric pressure exert an influence on radon release, accomplished through the modulation of soil gas flow. Conversely, the strong correlations among meteorological parameters (e.g., the negative relationships between temperature–humidity and temperature–pressure) reveal the interdependent nature of these variables and underscore the necessity of multivariate analyses. The fundamental descriptive statistics conducted to elucidate the general distribution characteristics and statistical nature of the data set are summarized in Table 3. A total of 2950 data points were obtained during the measurement period, and the average value of radon concentration was calculated as 24.14 kBq/m³, with a standard deviation of 12.47 kBq/m³. The minimum and maximum values of the radon distribution were 10.01 kBq/m³ and 55.81 kBq/m³, respectively, indicating significant temporal variability. The positive skewness (skewness = 0.47) and negative kurtosis (kurtosis = −0.685) values of the distribution indicate that radon concentrations deviate from a normal distribution and that high values are observed with greater frequency. The statistical characteristics of radon data suggest that they are susceptible to extreme values, necessitating the use of robust statistical approaches in addition to classical parametric methods. A subsequent examination of the descriptive statistics of meteorological parameters reveals that the distributions of temperature and relative humidity are relatively symmetric, while the atmospheric pressure distribution exhibits a slight positive skewness. When evaluated in conjunction with the distribution characteristics of radon data, these results statistically substantiate the implementation of multivariate and machine learning-based anomaly detection methods to reliably discern potential radon anomalies independent of meteorological effects. 4. Conclusions In this study, continuous soil-gas 222Rn measurements and simultaneous meteorological observations were evaluated in a seismically active area of the Marmara Region over a five-month period. During the monitoring interval, three earthquakes with magnitudes M ≥ 3.0 occurred within a 100 km radius of the measurement station. However, only one event was located within the earthquake preparation zone as defined by the Dobrovolsky empirical relationship. In conclusion, the probability of a direct seismic effect on the recorded radon time series was negligible, and no significant change in radon that could be definitively attributed to seismic activity was considered. Only the increase in radon concentration between August 11, 2024, and August 15, 2024, was assessed as potentially related to the magnitude 3.8 earthquake that occurred in Mudanya, Bursa. The temporal behavior of radon concentrations exhibited a discernible diurnal pattern, suggesting a notable sensitivity to near-surface environmental conditions. Correlation analysis revealed a moderately systematic relationship between radon variability and meteorological parameters, particularly temperature and atmospheric pressure. This finding aligns with established soil gas transport mechanisms, such as barometric pumping. These findings underscore the critical importance of accounting for meteorological forcing when interpreting radon time series, particularly in the context of shortto medium-term monitoring studies. F4-6 2nd Kocaeli Science Congress, November 19-21, 2025 Descriptive statistical analysis further indicated that radon data exhibit non-normal distribution characteristics and substantial temporal variability, which complicates the identification of subtle anomalies potentially related to tectonic processes. In consideration of the constraints imposed by the current dataset, the findings indicate that the variations in radon observed at the station are primarily influenced by local environmental factors. Consequently, the implementation of extended monitoring periods, in conjunction with multivariate and robust statistical methodologies, is imperative to ensure the reliable evaluation of the potential role of soil-gas radon as a precursor signal within this specific region. Acknowledgments This study was supported by the Scientific and Technological Research Council of Türkiye (TÜBİTAK) under Project No. 123Y145. References [1] Khan, P. A., Tripathi, S.C., Mansoori, A.A., Bhawre, P., Purohit, P., Gwal, A. 2011. "Scientific efforts in the direction of successful Earthquake Prediction." International Journal of Geomatics and Geosciences, 1(4), 669-677. [2] Cicerone, R. D., Ebel, J.E., Britton, J. 2009. "A systematic compilation of earthquake precursors.", Tectonophysics 476(3-4), 371-396. [3] Author, D. (2021). Book Title: Mathematical Foundations. Publisher Name. [3] Chiodini, G., Caliro, S., Cardellini, C., Frondini, F., Inguaggiato, S., Matteucci F. 2011. "Geochemical evidence for and characterization of CO2 rich gas sources in the epicentral area of the Abruzzo 2009 earthquakes.", Earth and Planetary Science Letters, 304(3-4), 389-398. [4] Riggio, A., Santulin, M., 2015. "Earthquake forecasting: a review of radon as seismic precursor.", Bollettino Di GeofisicaTeorica e Applicata, 56(2), 95-114. [5] Hauksson, E. 1981. "Radon content of groundwater as an earthquake precursor: evaluation of worldwide data and physical basis.", Journal of Geophysical Research: Solid Earth, 86(B10), 93979410. [6] Etiope, G., Martinelli, G., 2002. "Migration of carrier and trace gases in the geosphere: an overview.", Physics of the Earth and Planetary Interiors, 129(3-4), 185-204. [7] Dobrovolsky, I.P., Zubkov, S.I., Miachkin, V. I. 1979. “Estimation of the size of earthquake preparation zones,” Pure and Applied Geophysics, 117(5), 1025-1044.