Prediction of Earthquake Magnitude Using Tree-Based Ensemble Learning

Abstract

The present paper aims at predicting earthquake magnitude (Mw) from soil gas radon concentration (CRn) and three meteorological parameters (hourly humidity (H), hourly temperature (T), and hourly air pressure (P)). To accomplish this, three tree-based ensemble machine learning approaches, namely gradient boosting (GBoost), extreme gradient boosting (XGBoost), and random forest (RF), were created. A total of 386 datasets including recorded Mw values and measured soil gas CRn, H, T, and P values were used to develop the models. The models were then verified using statistics such as relative absolute error (RAE), root mean square error (RMSE), mean absolute error (MAE), and the ratio of RMSE to data standard deviation (RSR). A comparison of the performance metrics reveals that the GBoost model predicted the Mw value with lower MAE, RMSE, RSR, and RAE values than both the XGBoost and RF models. Performance was also verified using rank analysis and plots of Taylor and scaled percentage error (SPE). Rank analysis showed that the GBoost model received higher overall scores than both the XGBoost and RF models, indicating that the GBoost model achieved better prediction accuracy than both the XGBoost and RF models in predicting the Mw values. Both Taylor and SPE plots showed that GBoost model predicted the Mw values more accurately than XGBoost and RF models. According to the results of the study, the GBoost model can be used to predict Mw reliably and quickly, provided that three meteorological factors (H, T, and P) and soil gas CRn values are available.

Keywords

• Earthquake magnitude
• Soil gas radon concentration
• Extreme gradient boosting
• Gradient boosting
• Random forest.

References

1. [1]. Jilani Z, Mehmood T, Alam A, Awais M, Iqbal T (2017) Monitoring and descriptive analysis of radon in relation to seismic activity of Northern Pakistan Journal of Environmental Radioactivity; 172:43-51.
2. [2]. Asim KM, Martínez-Álvarez F, Basit A, Iqbal T (2017) Earthquake magnitude prediction in Hindukush region using machine learning techniques. Natural Hazards; 85:471-486.
3. [3]. Martinelli G, Dadomo A (2017) Factors constraining the geographic distribution of earthquake geochemical and fluid-related precursors Chemical Geology; 469:176-184.
4. [4]. Cothern C R, Smith JE (Eds.) (1987) Environmental radon (Vol. 35). Springer Science & Business Media.
5. [5]. Wilkening M (1990) Radon in the Environment (Vol. 40). Elsevier.
6. [6]. Reddy DV, Sukhija BS, Nagabhushanam P, Kumar D (2004) A clear case of radon anomaly associated with a micro‐earthquake event in a Stable Continental Region. Geophysical Research Letters; 31(10).
7. [7]. Moussa MM, El Arabi AGM (2003) Soil radon survey for tracing active fault: a case study along Qena-Safaga road, Eastern Desert, Egypt. Radiation Measurements; 37(3):211-216.
8. [8]. Wakita H (1996) Geochemical challenge to earthquake prediction. Proceedings of the National Academy of Sciences; 93(9):3781-3786.

9. [9]. Hartmann J, Levy JK (2005) Hydrogeological and gasgeochemical earthquake precursors–A review for application. Natural Hazards; 34:279-304.
10. [10]. Singh S, Jaishi HP, Tiwari RP, Tiwari RC (2016) A study of variation in soil gas concentration associated with earthquakes near Indo-Burma Subduction zone. Geoenvironmental Disasters; 3:1-8.
11. [11]. Zmazek B, Todorovski L, Džeroski S, Vaupotič J, Kobal I (2003) Application of decision trees to the analysis of soil radon data for earthquake prediction. Applied Radiation and Isotopes; 58(6):697-706.
12. [12]. Walia V, Virk HS, Bajwa BS (2006) Radon precursory signals for some earthquakes of magnitude> 5 occurred in NW Himalaya: An overview. Pure and Applied Geophysics; 163:711-721.
13. [13]. Iwata D, Nagahama H, Muto J, Yasuoka Y (2018) Non-parametric detection of atmospheric radon concentration anomalies related to earthquakes. Scientific Reports; 8: 3028.
14. [14]. Kawabata K, Sato T, Takahashi HA, Tsunomori F, Hosono T, Takahashi M, Kitamura Y (2020) Changes in groundwater radon concentrations caused by the 2016 Kumamoto earthquake. Journal of Hydrology; 584: 124712.
15. [15]. Chowdhury S, Deb A, Barman C, Nurujjaman M, Bora DK (2022) Simultaneous monitoring of soil 222Rn in the Eastern Himalayas and the geothermal region of eastern India: an earthquake precursor. Natural Hazards;112: 1477–1502.
16. [16]. Galiana-Merino JJ, Molina S, Kharazian A, Toader VE, Moldovan IA, Gomez I (2022) Analysis of radon measurements in relation to daily seismic activity rates in the vrancea region, romania. Sensors; 22: 4160.
17. [17]. Karastathis VK, Eleftheriou G, Kafatos M, Tsinganos K, Tselentis GA, Mouzakiotis E, Ouzounov D (2022) Observations on the stress related variations of soil radon concentration in the Gulf of Corinth, Greece Scientific Reports;12: 5442.
18. [18]. Manisa K, Erdogan M, Zedef V, Bircan H, Bicer A (2022) Variations of 222Rn concentrations over active fault system in Simav, Kutahya, Western Turkey: Possible causes for soil-gas 222Rn anomalies. Applied Radiation Isotopes ;190: 110484.
19. [19]. Alam A, Nikolopoulos D, Wang N (2023) Fractal Patterns in Groundwater Radon Disturbances Prior to the Great 7.9 Mw Wenchuan Earthquake, China. Geosciences;13(9):268.
20. [20]. Jaishi HP, Singh S, Tiwari RP, Tiwari RC (2023) Analysis of Subsurface Soil Radon with the Environmental Parameters and Its Relation with Seismic Events. Journal of the Geological Society of India; 99: 847–858.
21. [21]. Romano D, Sabatino G, Magazù S, Di Bella M, Tripodo A, Gattuso A, Italiano F (2023). Distribution of soil gas radon concentration in north-eastern Sicily (Italy): hazard evaluation and tectonic implications. Environmental Earth Sciences; 82(11): 273.
22. [22]. Walia V, Kumar A, Chowdhury S, Lin SJ, Lee HF, Fu CC (2024) Earthquake precursory study using decomposition technique: time series soil radon monitoring data from the San-Jie Station in Northern Taiwan. Journal of Radioanalytical and Nuclear Chemistry; 333(6): 3047-3054.
23. [23]. Baykut S, Akgül T, İnan S, Seyis C (2010) Observation and removal of daily quasi-periodic components in soil radon data. Radiation Measurements; 45(7):872-879.
24. [24]. Oh Y, Kim G (2015) A radon-thoron isotope pair as a reliable earthquake precursor. Scientific Reports; 5:13084.
25. [25]. Saç MM, Harmansah C, Camgoz B, Sozbilir H (2011) Radon monitoring as the earthquake precursor in fault line in Western Turkey. Ekoloji; 20(79):93-98.
26. [26]. Yang TF, Walia V, Chyi LL, Fu CC, Chen CH, Liu TK, ... Lee M (2005) Variations of soil radon and thoron concentrations in a fault zone and prospective earthquakes in SW Taiwan. Radiation Measurements; 40(2-6): 496-502.
27. [27]. Friedman JH (2001) Greedy function approximation: a gradient boosting machine. Annals of Statistics; 1189-1232.
28. [28]. Huang S (2024) Processing and Comparison of GBoost, XGBoost, and Random Forest in Titanic Survival Prediction. Applied and Computational Engineering; 102:175-182.
29. [29]. Chen T, Guestrin C (2016) Xgboost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 785-794).
30. [30]. Breiman L (2001) Random forests. Machine Learning; 45:5-32.
31. [31]. AFAD (2021) Disaster and Emergency Management, (n.d.). http://www.afad.gov.tr
32. [32]. TEC (2018) Turkish Earthquake Code: Specifications for Building Design Under Earthquake Effects, Regulations on structures constructed in disaster regions, Ankara
33. [33]. Freund Y, Schapire R, Abe N (1999) A short introduction to boosting. Journal-Japanese Society for Artificial Intelligence; 14(5):771-780.
34. [34]. Persson C, Bacher P, Shiga T, Madsen H (2017) Multi-site solar power forecasting using gradient boosted regression trees. Solar Energy; 150:423-436.
35. [35]. Wang FK, Mamo T (2020) Gradient boosted regression model for the degradation analysis of prismatic cells. Computers & Industrial Engineering; 144:106494.
36. [36]. Kumar P, Alruqi M, Hanafi HA, Sharma P, Wanatasanappan VV (2024) Effect of particle size on second law of thermodynamics analysis of Al2O3 nanofluid: application of XGBoost and gradient boosting regression for prognostic analysis. International Journal of Thermal Sciences; 197:108825.
37. [37]. Pathy A, Meher S (2020) Predicting algal biochar yield using eXtreme Gradient Boosting (XGB) algorithm of machine learning methods. Algal Research; 50:102006.
38. [38]. Qiu Y, Zhou J, Khandelwal M, Yang H, Yang P, Li C (2022) Performance evaluation of hybrid WOA-XGBoost, GWO-XGBoost and BO-XGBoost models to predict blast-induced ground vibration. Engineering with Computers; 38(5):4145-4162.
39. [39]. Kumar D, Sood SK, Rawat KS (2023) Early health prediction framework using XGBoost ensemble algorithm in intelligent environment. Artificial Intelligence Review; 56(1):1591-1615.
40. [40]. Li X, Ma L, Chen P, Xu H, Xing Q, Yan J, ... Cheng Y (2022) Probabilistic solar irradiance forecasting based on XGBoost. Energy Reports; 8:1087-1095.
41. [41]. Sagi O, Rokach L (2021) Approximating XGBoost with an interpretable decision tree. Information Sciences; 572: 522-542.
42. [42]. Zhang P, Jia Y, Shang Y (2022) Research and application of XGBoost in imbalanced data. International Journal of Distributed Sensor Networks; 18(6):15501329221106935.
43. [43]. Sahin EK, Colkesen I, Kavzoglu T (2020) A comparative assessment of canonical correlation forest, random forest, rotation forest and logistic regression methods for landslide susceptibility mapping. Geocarto International; 35(4):341-363.
44. [44]. Wu Y, Liu H, Liu S, Lou C (2023) Estimate of near-surface NO2 concentrations in Fenwei Plain, China, based on TROPOMI data and random forest model. Environmental Monitoring and Assessment; 195(11):1379.
45. [45]. Khan M, Khan A U, Houda M, El Hachem C, Rasheed M, Anwar W (2023) Optimizing durability assessment: Machine learning models for depth of wear of environmentally-friendly concrete. Results in Engineering; 20:101625.
46. [46]. Sharma C, Ojha CSP (2019) Statistical parameters of hydrometeorological variables: standard deviation, SNR, skewness and kurtosis. In Advances in Water Resources Engineering and Management:select proceedings of TRACE 2018 (pp. 59-70). Singapore: Springer Singapore.
47. [47]. Cain MK, Zhang Z, Yuan KH (2017) Univariate and multivariate skewness and kurtosis for measuring nonnormality: Prevalence, influence, and estimation. Behavior Research Methods; 49: 1716-1735.
48. [48]. Benson CH (1993) Probability distributions for hydraulic conductivity of compacted soil liners. Journal of Geotechnical Engineering; 119(3):471-486.
49. [49]. Brown SC, Greene JA (2006) The wisdom development scale: Translating the conceptual to the concrete. Journal of College Student Development; 47(1):1-19.
50. [50]. Bulmer MG (1979) Principles of Statistics. Dover.
51. [51]. Tabachnick BG, Fidell LS (2013) Using Multivariate Statistics. Pearson.
52. [52]. Gravetter F, Wallnau L (2014) Essentials of statistics for the behavioral sciences. Wadsworth.
53. [53]. Byrne BM, Van de Vijver FJ (2010) Testing for measurement and structural equivalence in large-scale cross-cultural studies: Addressing the issue of nonequivalence. International Journal of Testing; 10(2):107-132.
54. [54]. Hair Jr JF, Black WC, Babin BJ, Anderson RE (2010) Multivariate data analysis. In Multivariate Data Analysis (pp. 785-785).
55. [55]. Hair Jr J, Wolfinbarger MC, Ortinau DJ, Bush RP (2013) Essentials of Marketing, New York, Mc Graw Hill.
56. [56]. Rabbani A, Samui P, Kumari S (2023c) Optimized ANN-based approach for estimation of shear strength of soil. Asian Journal of Civil Engineering; 24(8):3627-3640.
57. [57]. Nafouanti MB, Li J, Mustapha NA, Uwamungu P, Al-Alimi D (2021) Prediction on the fluoride contamination in groundwater at the Datong Basin, Northern China: comparison of random forest, logistic regression and artificial neural network. Applied Geochemistry; 132:105054.
58. [58]. Erzin S (2024a) Using radial basis artificial neural networks to predict radiation hazard indices in geological materials. Environmental Monitoring and Assessment; 196(3):315.
59. [59]. Erzin S (2024b) Prediction of the radon concentration in thermal waters using artificial neural networks. International Journal of Environmental Science and Technology; 21(10):7321-7328.
60. [60]. Erzin S (2025) Optimization of artificial neural network for predicting radon exhalation rates and effective radium content in soil samples. Earth Science Informatics; 18(1):170.
61. [61]. Erzin S, Yaprak G (2022) Prediction of the activity concentrations of 232Th, 238U and 40K in geological materials using radial basis function neural network. Journal of Radioanalytical and Nuclear Chemistry; 331(9):3525-3533.
62. [62]. Rabbani A, Samui P, Kumari S (2023a) A novel hybrid model of augmented grey wolf optimizer and artificial neural network for predicting shear strength of soil. Modeling Earth Systems and Environment; 9(2):2327-2347.
63. [63]. Rabbani A, Samui P, Kumari S (2023b) Implementing ensemble learning models for the prediction of shear strength of soil. Asian Journal of Civil Engineering; 24(7):2103-2119.
64. [64]. Santos-Pereira J, Gruenwald L, Bernardino J (2022) Top data mining tools for the healthcare industry. Journal of King Saud University-Computer and Information Sciences; 34(8):4968-4982.
65. [65]. Thornton C, Hutter F, Hoos HH, Leyton-Brown K (2013) Auto-WEKA: Combined selection and hyperparameter optimization of classification algorithms. In Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 847-855).
66. [66]. DeCastro-García N, Muñoz Castañeda ÁL, Escudero García D, Carriegos MV (2019) Effect of the sampling of a dataset in the hyperparameter optimization phase over the efficiency of a machine learning algorithm. Complexity; 1:6278908.
67. [67]. Belete D M, Huchaiah M D (2022) Grid search in hyperparameter optimization of machine learning models for prediction of HIV/AIDS test results. International Journal of Computers and Applications; 44(9):875-886.
68. [68]. Zöller MA, Huber MF (2021) Benchmark and survey of automated machine learning frameworks. Journal of Artificial Intelligence Research; 70:409-472.
69. [69]. Khatti J, Grover KS (2024) Assessment of hydraulic conductivity of compacted clayey soil using artificial neural network: An investigation on structural and database multicollinearity. Earth Science Informatics; 17(4): 3287-3332.
70. [70]. Khatti J, Grover KS, Kim HJ, Mawuntu KBA, Park TW (2024) Prediction of ultimate bearing capacity of shallow foundations on cohesionless soil using hybrid LSTM and RVM approaches: An extended investigation of multicollinearity. Computers and Geotechnics; 165:105912.
71. [71]. Rabbani A, Samui P, Kumari S, Saraswat BK, Tiwari M, Rai A (2024) Optimization of an artificial neural network using three novel meta-heuristic algorithms for predicting the shear strength of soil. Transportation Infrastructure Geotechnology; 11(4):1708-1729.
72. [72]. Kumar M, Kumar DR, Khatti J, Samui P, Grover KS (2024b) Prediction of bearing capacity of pile foundation using deep learning approaches. Frontiers of Structural and Civil Engineering; 18(6):870-886.
73. [73]. Biswas R, Kumar M, Singh RK, Alzara M, El Sayed SBA, Abdelmongy M, ... Yousef SEA (2023) A novel integrated approach of RUNge Kutta optimizer and ANN for estimating compressive strength of self-compacting concrete. Case Studies in Construction Materials; 18: e02163.
74. [74]. Erzin Y, Ecemis N (2017) The use of neural networks for the prediction of cone penetration resistance of silty sands. Neural Computing Applications; 28:727-736.
75. [75]. Golbraikh A, Tropsha A (2002) Beware of q2!. Journal of Molecular Graphics and Modelling; 20(4):269-276.
76. [76]. Alavi AH, Ameri M, Gandomi AH, Mirzahosseini MR (2011) Formulation of flow number of asphalt mixes using a hybrid computational method. Construction and Building Materials; 25(3):1338-1355.
77. [77]. Roy PP, Roy K (2008) On some aspects of variable selection for partial least squares regression models. QSAR & Combinatorial Science; 27(3):302-313.
78. [78]. Khatti J, Grover KS (2023) Prediction of compaction parameters for fine-grained soil: Critical comparison of the deep learning and standalone models. Journal of Rock Mechanics and Geotechnical Engineering; 15(11):3010-3038.
79. [79]. Ardakani A, Kordnaeij A (2019) Soil compaction parameters prediction using GMDH-type neural network and genetic algorithm. European Journal of Environmental and Civil Engineering; 23(4):449-462.
80. [80]. Chen W, Hasanipanah M, Nikafshan Rad H, Jahed Armaghani D, Tahir MM (2021) A new design of evolutionary hybrid optimization of SVR model in predicting the blast-induced ground vibration. Engineering with Computers; 37:1455-1471.
81. [81]. Shirani Faradonbeh R, Jahed Armaghani D, Abd Majid MZ, Md Tahir M, Ramesh Murlidhar B, Monjezi M, Wong HM (2016) Prediction of ground vibration due to quarry blasting based on gene expression programming: a new model for peak particle velocity prediction. International Journal of Environmental Science and Technology; 13:1453-1464.
82. [82]. Wang HL, Yin ZY (2020) High performance prediction of soil compaction parameters using multi expression programming. Engineering Geology; 276:105758.
83. [83]. Momeni E, Nazir R, Armaghani DJ, Maziar H (2014) Prediction of pile bearing capacity using a hybrid genetic algorithm-based ANN. Measurement; 57:122-131.