Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 10
Encontrada(s) 1 página(s)

DIGITAL.CSIC. Repositorio Institucional del CSIC
Datos de investigación Set de datos (Dataset)

[Dataset] Global physics-based database of injection-induced seismicity

  • Kivi, Iman Rahimzadeh
  • Boyet, Auregan
  • Wu, Haiqing
  • Walter, Linus
  • Hanson-Hedgecock, Sara
  • Parisio, Francesco
  • Vilarrasa, Víctor
The database and its accompanying files are as follows: GEoREST_database_Readme (.txt): provides an overview of the database goals and content, sharing/accessing information and methodologies used to develop the database. GEoREST_induced_seismicity_database_v20.11.2022 (.xlsx): database in a single Microsoft Excel spreadsheet. GEoREST_induced_seismicity_database_v20.11.2022 (.csv): database in ‎.csv format, considered as a standard machine-readable format for direct implementation of data in model developments. GEoREST_database_bibliography_v20.11.2022 (.docx): provides the full list of references used to collect data included in the database. GEoREST_database_dictionary_v20.11.2022 (.docx): presents unified, self-explanatory acronyms together with concise definitions for all variables included in the database., We present a comprehensive, publicly accessible database of injection-induced seismicity. The database is sourced from a comprehensive review of more than 500 published documents and contains information for 158 cases of induced earthquakes from around the world. The collected earthquakes are associated with a variety of geoenergy applications, which can be broadly categorized into geologic gas storage, geothermal energy development, shale gas fracturing, research projects and wastewater disposal. The compilation comprises more than 70 variables, including general project information, host rock properties, ‎in situ site characteristics, fault attributes, operational parameters, and recorded seismicity data (both overall seismic activities and the largest event sequence). The developed database opens up opportunities for improved understanding of the causative mechanisms of injection-induced seismicity and advancing on seismic hazard forecasting and mitigation., This research is funded by (1) the European Research Council (ERC) under ‎the ‎‎European Union’s Horizon ‎‎2020 Research and Innovation Program through the ‎Starting Grant ‎‎GEoREST (www.georest.eu) under Grant ‎agreement No. 801809, (2) MCIN/AEI/‎‎10.13039/501100011033 and the ‎European ‎Union ‎NextGenerationEU/PRTR through the international collaboration project EASYGEOCARBON ‎‎‎‎(www.easygeocarbon.com) under Grant ‎agreement No. PCI2021-122077-2B, (3) the European Union’s Horizon 2020 Research and Innovation ‎Programme through the Marie Sklodowska-Curie Action ARMISTICE (www.armistice-energy.eu) under grant â €Žagreement No. 882733. (4) the Secretariat for Universities and ‎Research of the Ministry of Business and Knowledge of the Government of Catalonia (AGAUR) and the European ‎Social Fund (FI-2019), (5) MCIN/AEI/ ‎‎10.13039/501100011033‎ through the Excellence Servero Ochoa (IDAEA-CSIC) under Grant agreement No. ‎CEX2018-000794-S, Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Datos de investigación Set de datos (Dataset)

Dataset for Numerical modeling of hydraulic stimulation of fractured crystalline rock at the Bedretto Underground Laboratory for Geosciences and Geoenergies

  • Vaezi, Iman
  • Alcolea, Andrés
  • Meier, Peter
  • Parisio, Francesco
  • Carrera, Jesús
  • Vilarrasa, Víctor
[Date of data collection] Feb. 2020 for field data and Jan. 2022- Jan. 2023 for numerical models., This dataset includes the input files of the numerical models used in the manuscript for simulating the hydraulic stimulation at the Bedretto lab. Each folder is named after the corresponding model in the manuscript, i.e., EP (Elastic Prescribed), EE (Elastic Embedded), and VE (Viscoplastic Embedded). In each folder, there is a file with the name of the folder ended as “_gen.dat” which contains the input data of the model, including material properties, initial and boundary conditions and the time intervals. There is also a file ended as “_gri.dat” that includes the information on the mesh. The file “root.dat” includes the name of the model. The file ended as “_bcf.dat” contains injection rate input., European Research Council (ERC) under the European Union’s Horizon 2020 Research and Innovation Program through the Starting Grant GEoREST (www.georest.eu), under grant agreement No. 801809. Swiss Federal Office of Energy within the framework of the ZoDrEx GEOTHERMICA project under contract SI/501720-01. European Union’s Horizon 2020 Research and Innovation Program through the Marie Sklowdowska-Curie Individual Fellowship ARMISTICE, under Grant Agreement No. 882733. The “ZoDrEx” project, which has been subsidized through the ERANET Cofund GEOTHERMICA (Project No. 731117), from the European Commission and the Spanish Ministry of Economy, Industry and Competitiveness (MINECO). IMEDEA is an accredited "Maria de Maeztu Excellence Unit" (Grant CEX2021-001198, funded by MCIN/AEI/10.13039/501100011033)., With funding from the Spanish government through the "Severo Ochoa Centre of Excellence" accreditation (CEX2021-001198)., 1-EP_bcf.dat 1-EP_gen.dat 1-EP_gri.dat root.dat 2-EE_bcf.dat 2-EE_gen.dat 2-EE_gri.dat root.dat 3-VE_bcf.dat 3-VE_gen.dat 3-VE_gri.dat root.dat 3-VE-Extended_bcf.dat 3-VE-Extended_gen.dat 3-VE-Extended_gri.dat root.dat, Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Datos de investigación Set de datos (Dataset)

Dataset for Effectiveness of Injection protocols for hydraulic stimulation in Enhanced Geothermal Systems

  • Tangirala, Sri Kalyan
  • Parisio, Francesco
  • Vaezi, Iman
  • Vilarrasa, Víctor
This dataset includes the input files of the numerical models used in the manuscript for simulating the hydraulic stimulation in Enhanced Geothermal System with a single fracture setup. Each folder is named after the corresponding injection protocol in the manuscript, i.e., Constant, Step, and Cyclic. In each of these folders, there are two subfolders namely- Without Bleedoff and With Bleedoff. In the subfolders, the file ending with “_gen.dat” contains the input data of the model, including material properties, initial and boundary conditions and the time intervals. There is also a file ending with “_gri.dat” that includes the information on the mesh. The file “root.dat” includes the name of the model., European Research Council (ERC) under the European Union’s Horizon 2020 Research and Innovation Program through the Starting Grant GEoREST (www.georest.eu), under grant agreement No. 801809. European Union’s Horizon 2020 Research and Innovation Program through the Marie Sklowdowska-Curie Individual Fellowship ARMISTICE, under Grant Agreement No. 882733. IMEDEA is an accredited "Maria de Maeztu Excellence Unit" (Grant CEX2021-001198, funded by MCIN/AEI/10.13039/501100011033)., With funding from the Spanish government through the "Severo Ochoa Centre of Excellence" accreditation (Grant CEX2021-001198)., README.txt (this file) Constant_0p3_gen.dat Constant_0p3_gri.dat root.dat Constant_0p3_bleedoff_gen.dat Constant_0p3_bleedoff_gri.dat root.dat Step_0p3_gen.dat Step_0p3_gri.dat root.dat Step_0p3_bleedoff_gen.dat Step_0p3_bleedoff_gri.dat root.dat Cyclic_0p3_gen.dat Cyclic_0p3_gri.dat root.dat Cyclic_0p3_bleedoff_gen.dat Cyclic_0p3_bleedoff_gri.dat root.dat, Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Datos de investigación Set de datos (Dataset)

Dataset for analytical and numerical solutions of cyclic pore pressure diffusion equation

  • Sciandra, Dario
  • Kivi, Iman Rahimzadeh
  • Makhnenko, Roman Y.
  • Rebscher, Dorothee
  • Vilarrasa, Víctor
This dataset encompasses the input files associated with diverse analytical and numerical models investigating cyclic pressure diffusion within uniform poroelastic materials. The dataset is organized into two primary folders. The "Analytical Solutions" folder comprises three Python libraries containing the material properties and functions for calculating pressure amplitude at various distances. The "Numerical Solutions" folder contains simulations for both Hydraulic ("01_") and Hydro-Mechanical ("HM_") problems. The simulations are conducted for representative values corresponding to Berea sandstone ("BS_"), Opalinus Clay ("OPA_"), and Westerly granite ("WG_"), considering both monodimensional ("1D") and axisymmetric ("Ax") cases. Within each subfolder, files with the suffix "_gen.dat" provide input data for the model, encompassing material properties, initial and boundary conditions, and time intervals. Correspondingly, files concluding with "_gri.dat" contain information of the mesh employed. The "_bfc.dat" file documents the periodic injection pressure applied for each simulation, while the "root.dat" file specifies the model's nomenclature. To execute a model, launching the executable "Cb_v9_1_3par4.exe" within each ".gid" folder is sufficient., European Research Council (ERC) under the European Union’s Horizon 2020 Research and Innovation Program through the Starting Grant GEoREST (www.georest.eu), under grant agreement No. 801809. IMEDEA is an accredited "Maria de Maeztu Excellence Unit" (Grant CEX2021-001198, funded by MCIN/AEI/10.13039/501100011033)., With funding from the Spanish government through the "Severo Ochoa Centre of Excellence" accreditation (CEX2021-001198)., File List: README.txt (this file) \Analytical Solutions\CyclicDiffusion.py \Analytical Solutions\CyclicDiffusion_Adimensional.py \Analytical Solutions\Materials.py \Numerical Solutions\01_BS_1D_g0_1yr.gid \Numerical Solutions\01_BS_Ax_g0_1yr.gid \Numerical Solutions\01_OPA_1D_g0_1w.gid \Numerical Solutions\01_OPA_Ax_g0_1w.gid \Numerical Solutions\01_WG_1D.gid \Numerical Solutions\01_WG_Ax.gid \Numerical Solutions\HM_BS_1D_g0_1yr.gid \Numerical Solutions\HM_BS_Ax_g0_1yr.gid \Numerical Solutions\HM_OPA_1D_g0_1w.gid \Numerical Solutions\HM_OPA_Ax_g0_1w.gid \Numerical Solutions\HM_WG_1D.gid \Numerical Solutions\HM_WG_Ax.gid \Numerical Solutions\Cb_v9_1_3par4.exe, Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Datos de investigación Set de datos (Dataset)

Implicit hydromechanical upscaling of fractures using a continuum approach [Dataset]

  • Vaezi, Iman
  • Parisio, Francesco
  • Yoshioka, Keita
  • Alcolea, Andrés
  • Meier, Peter
  • Carrera, Jesús
  • Olivella, Sebastià
  • Vilarrasa, Víctor
Any usage of field data should aquire permission from Geo-Energie Suisse., This dataset includes the input files of the numerical models used in the manuscript for simulating the hydraulic stimulation in the conceptual models as well as the Bedretto model with the executable file of the numerical code (CODE_BRIGHT). Each folder is named after the corresponding model in the manuscript described in the sections 3.1.1 and 3.2.1. In each folder, there is a file with the name of the folder ended as “_gen.dat” which contains the input data of the model, including material properties, initial and boundary conditions and the time intervals. There is also a file ended as “_gri.dat” that includes the information on the mesh. The file “root.dat” includes the name of the model. The file ended as “_bcf.dat” contains injection rate input for Bedretto model. To run the simulation, copy and paste Code_Bright executable file, i.e., “Cb_2021.exe”, in a folder that contains the input files and execute it., European Research Council (ERC) under the European Union’s Horizon 2020 Research and Innovation Program through the Starting Grant GEoREST (www.georest.eu), under grant agreement No. 801809. Swiss Federal Office of Energy within the framework of the ZoDrEx GEOTHERMICA project under contract SI/501720-01. European Union’s Horizon 2020 Research and Innovation Program through the Marie Sklowdowska-Curie Individual Fellowship ARMISTICE, under Grant Agreement No. 882733. The “ZoDrEx” project, which has been subsidized through the ERANET Cofund GEOTHERMICA (Project No. 731117), from the European Commission and the Spanish Ministry of Economy, Industry and Competitiveness (MINECO). EURAD, the European Joint Programme on Radioactive Waste Management through the project DONUT (Grant No 847593), Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Datos de investigación Set de datos (Dataset)

[Dataset] To bleed-off or not to bleed-off?

  • Boyet, Auregan
  • De Simone, Silvia
  • Vilarrasa, Víctor
This dataset corresponds to the model made from the EGS project of Basel (2006, Switzerland). The model is solving coupled hydromechanical problem for 2D fault network on a surface of 1 km2, located at 4630-meters depth coinciding with the injection depth in the crystalline basement at Basel. A set of faults are embedded in a rock matrix, with the fault network derived from the induced seismicity that was monitored in the range of 3750 and 4750 meters deep. The maximum principal stress S_Hmax is aligned with y-axis (S_Hmax=160 MPa, S_hmin=84 MPa, S_v=115 MPa). The hydrostatic pressure is set at 45 MPa following a hydrostatic profile and the temperature at the depth of the reservoir is set at 190°C. Stimulation parameters are inputs as wellhead pressure based on the injection strategy from Häring et al. (2008), injection fluid is water. - “ .gid” is the Code_Bright file with the model of Basel. The file “_gen.dat” contains the input data of the model (including material properties, initial and boundary conditions and the time intervals). The file “_gri.dat” includes the information on the mesh. The “root.dat” includes the name of the model. To run simulations, execute the Code_Bright executable “Cb_2020_21.exe” in a folder that contains the three input files and the executable, where X and Y denote the used version of the executable. • V13_C_NW_longbleed.gid corresponds to the model in which the bleed-off of the well is imposed. • V13_C_NW_shutin.gid corresponds to the model in which the injection well is shut-in., Horizon 2020 Research and Innovation Program through the ‎Starting Grant ‎GEoREST under Grant agreement No. 801809, - V13_c_NW_longbleed.gid - V13_c_NW_shutin.gid - Readme.docx, Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Datos de investigación Set de datos (Dataset)

Dataset for 3D numerical simulations of fractured specimen

  • Sciandra, Dario
  • Kim, Hyunbin
  • Makhnenko, Roman Y.
  • Kivi, Iman Rahimzadeh
  • Vilarrasa, Víctor
This dataset encompasses the input files associated with diverse numerical models investigating the hydro-mechanical behavior of a fractured specimen under different mean effective stresses. The folder contains simulations for Hydro-Mechanical ("HM_") problems. The simulations are conducted for different mean effective stresses ("P03"; "P05"; "P10"; "P15"; and "P20"). Within each subfolder, files with the suffix "_gen.dat" provide input data for the model, encompassing material properties, initial and boundary conditions, and time intervals. Correspondingly, files concluding with "_gri.dat" contain information on the mesh employed. To execute a model, launching the executable "Cb_v9_1_3par4.exe" within each ".gid" folder is sufficient., European Research Council (ERC) under the European Union’s Horizon 2020 Research and Innovation Program through the Starting Grant GEoREST (www.georest.eu), under grant agreement No. 801809. IMEDEA is an accredited "Maria de Maeztu Excellence Unit" (Grant CEX2021-001198, funded by MCIN/AEI/10.13039/501100011033)., With funding from the Spanish government through the "Severo Ochoa Centre of Excellence" accreditation (CEX2021-001198)., README.txt (this file) \HM_P03.gid \HM_P05.gid \HM_P10.gid \HM_P15.gid \HM_P20.gid \HM_P20_higherKn.gid, Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Datos de investigación Set de datos (Dataset)

Dataset for On the limitations of closed loop geothermal systems for electricity generation outside high geothermal gradient field

  • Tangirala, Sri Kalyan
  • Vilarrasa, Víctor
This dataset includes the input files of the numerical models used in the manuscript for simulating the injection well and the horizontal wells of the Closed Loop Geothermal System (CLGS) setup. There are two main folders named Vertical and Horizontal that contain their respective models. In the vertical folder, there are 4 folders that contain the input files of the models. In the horizontal folder, there are 5 subfolders named with their corresponding surface flow rates (12.5, 25, 50 and 75 kg/s) and one with 30yr as the title that has models with 30 year duration. Each of these subfolders have the specific model folders containing their input files. In each of the model folder, there are the input files ending with “_gen.dat” containing the input data of the model, including material properties, initial and boundary conditions and the time intervals. There are also files ending with “_gri.dat” that include information on the mesh. The file “root.dat” includes the name of the model., European Research Council (ERC) under the European Union’s Horizon 2020 Research and Innovation Program through the Starting Grant GEoREST (www.georest.eu), under grant agreement No. 801809. IMEDEA is an accredited "Maria de Maeztu Excellence Unit" (Grant CEX2021-001198, funded by MCIN/AEI/10.13039/501100011033)., With funding from the Spanish government through the "Severo Ochoa Centre of Excellence" accreditation (CEX2021-001198)., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Datos de investigación Set de datos (Dataset)

Numerical models for the simulation of bleed-off control on post-injection seismicity in ‎enhanced geothermal systems

  • Kivi, Iman Rahimzadeh
  • Vilarrasa, Víctor
[Description of methods used for collection/generation of data] The models are 2D plane strain including a homogeneous and isotropic granitic reservoir rock ‎intersected by an interconnected fracture network and a distant fault to the wellbore. Generic hydromechanical properties are assumed for the reservoir rock and the intersecting fractures and fault. The fault ‎in the models is either hydraulically isolated from the fracture network, i.e., our "basecase" scenario, or intersected by them, a scenario referred to as "connected". Post-‎injection protocols to control seismicity include well shut-in or bleed-off. During bleed-off, the ‎wellhead is opened to impose hydrostatic pressure conditions along the well over a period of ‎either 24 h or 1 h. The ‎hydrostatic pressure profile is maintained afterwards. Different injection ‎periods are considered to inspect the effect of this parameter on the effectiveness of bleed-off ‎in controlling post-injection seismicity. ‎, [Methods for processing the data] The Code_Bright executables "Cb_v9_3.exe" or "Cb_v9_1_3par4", i.e., the parallel version of the code running on 4 CPU cores, can be used to run the numerical models. Upon running the executables, the code generates several outputs, including "root.post.msh" and "root.post.res". These two outputs include the mesh structure and the simulation results, respectively. Any postprocessing software supporting the output file formats can be used to visualize the data., The Global physics-based database of injection-induced seismicity is licensed under Creative Commons CC BY 4.0 International License. Users are free to copy and redistribute the database material in any medium or format, and remix, transform, and build upon the material for any purpose, even commercially., This dataset comprises the input files for numerical simulation of stimulating an enhanced geothermal system and the associated risk of induced seismicity. Numerical models explore the hydromechanical mechanisms of post-injection induced seismicity and the potential impact of bleed-off to control seismicity. The models are 2D plane strain including a homogeneous and isotropic granitic reservoir rock intersected by an interconnected fracture network and a distant fault to the wellbore. The fault in our numerical models is either hydraulically isolated from the fracture network, i.e., our “basecase” scenario, or intersected by them, a scenario referred to as “connected”. Post-injection protocols to control seismicity include well shut-in or bleed-off. During bleed-off, the wellhead is opened to impose hydrostatic pressure conditions along the well over a period of either 24 h or 1 h. The ‎hydrostatic pressure profile is maintained afterwards. Different injection periods are considered to inspect the effect of this parameter on the effectiveness of bleed-off in controlling post-injection seismicity. Following this brief explanation, the dataset includes seven models provided in seven distinct folders: “Basecase_Inj250h_shutin”, basecase scenario of 250 h injection followed by well shut-in, “Basecase_Inj250h_bleedoff_1h”, basecase scenario of 250 h injection followed by well bleed-off over 1 h, “Basecase_Inj250h_bleedoff_24h”, basecase scenario of 250 h injection followed by well bleed-off over 24 h, “Connected_Inj84h_shutin”, connected scenario of 84 h injection followed by well shut-in, “Connected_Inj84h_bleedoff_1h”, connected scenario of 84 h injection followed by well bleed-off over 1 h, “Connected_Inj200h_shutin”, connected scenario of 200 h injection followed by well shut-in, “Connected_Inj200h_bleedoff_1h”, connected scenario of 200 h injection followed by well bleed-off over 1 h. In each folder, there is a file with the name of the folder ending as “_gen.dat” which contains the input data of the model, including material properties, initial and boundary conditions and the time intervals. There is also a file ending as “_gri.dat” that includes the information on the mesh. The file “root.dat” includes the name of the model. To run the simulation, execute the Code_Bright executables “Cb_v9_3.exe” or “Cb_v9_1_3par4”, i.e., the parallel version of the code running on 4 CPU cores, in each folder., - Project EASY GEO-CARBON ‎‎(www.easygeocarbon.com)‎ funded by MCIN/AEI‎ through grant number PCI2021-122077-2B‎. - The UKRI Postdoc Guarantee ‎Award THMC4CCS funded by the Engineering and Physical Sciences Research Council through grant number EP/X026019/1. - The ‎Starting Grant ‎‎GEoREST (www.georest.eu) funded by the European Research Council (ERC) under ‎the ‎‎European Union’s Horizon 2020 Research and ‎Innovation Program through under Grant ‎agreement No. 801809. ‎‎, This dataset comprises the input files for numerical simulation of stimulating an enhanced ‎geothermal system and the associated risk of induced seismicity. Numerical models explore the ‎hydromechanical mechanisms of post-injection induced seismicity and the potential impact of ‎bleed-off to control seismicity.‎ The dataset includes seven models provided in seven distinct folders: ‎ ‎"Basecase_Inj250h_shutin", basecase scenario of 250 h injection followed by well shut-in,‎ ‎"Basecase_Inj250h_bleedoff_1h", basecase scenario of 250 h injection followed by well bleed-‎off over 1 h,‎ ‎"Basecase_Inj250h_bleedoff_24h", basecase scenario of 250 h injection followed by well ‎bleed-off over 24 h, ‎ ‎"Connected_Inj84h_shutin", connected scenario of 84 h injection followed by well shut-in,‎ ‎"Connected_Inj84h_bleedoff_1h", connected scenario of 84 h injection followed by well bleed-‎off over 1 h,‎ ‎"Connected_Inj200h_shutin", connected scenario of 200 h injection followed by well shut-in,‎ ‎"Connected_Inj200h_bleedoff_1h", connected scenario of 200 h injection followed by well ‎bleed-off over 1 h.‎, Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Datos de investigación Set de datos (Dataset)

Dataset, code, and figure outputs for “What controls maximum injection-induced earthquake magnitude?”

  • Sachdeva, Sanchit
  • Kivi, Iman Rahimzadeh
  • Mukuhira, Yusuke
  • Vilarrasa, Víctor
[Description of methods used for collection or generation of data] Selected variables from the open-access GEoREST database were characterized, and rock names were standardized into 12 categories. Missing values were filled using literature values, deterministic calculations, and random-forest regression. Moment_Max was neither imputed nor used for imputation. Two LightGBM models were compared: an injection-only model and a full 12-feature model. Data were divided into 90% training and 10% testing subsets. Predictors were normalized from 0 to 1 using training-set ranges only., [Description of data processing and preparation] 1. Select and filter the GEoREST variables. 2. Fill supported literature values. 3. Calculate deterministic elastic and stress properties. 4. Impute remaining eligible values using random forests. 5. Merge the completed data and train the LightGBM models. Excel colours show provenance: white = original, yellow = literature, orange = random forest, and green = deterministic. Full processing details are provided in the scripts., [Codebooks or data dictionaries] This README documents the final variables. The original GEoREST workbook and Kivi et al. (2023) provide the source-variable definitions., Fluid injection into the subsurface for energy and resource development has led to increased seismic activity in recent years. Understanding the key factors governing maximum induced earthquake magnitude remains critical for their improved prediction and risk assessment. In this study, we apply a machine-learning model using the LightGBM algorithm to a global induced seismicity database spanning multiple geoenergy applications and settings to identify major factors associated with maximum injection-induced moment magnitude. We find that incorporating operational parameters, site characteristics, and host rock properties as input variables, compared to only operational parameters, significantly increases the predictive model accuracy. Cumulative injected fluid volume and injection rate impose first-order controls on seismic magnitude, while rock porosity, permeability, pore pressure, and in-situ stress field are found to play important secondary roles. These results highlight the importance of physical site characteristics beyond injection parameters in enriching magnitude-scaling relations and developing more reliable, physics-based seismicity forecasting and risk mitigation strategies., - The authors acknowledge support for data collection from the European Research Council (ERC) under the European Union’s Horizon 2020 Research and Innovation Program through the starting grant GEoREST under grant agreement no. 801809. - S.S. and V.V. acknowledge funding from the European Research Council through the Horizon Europe Proof-of-Concept project NOSHAKE, Grant Agreement No. 101248689. - I.R.K. acknowledges funding from MICIU/AEI/10.13039/501100011033 and the European Social Fund Plus (FSE+) through the Ramón y Cajal fellowship RYC2024-048711-I. - IMEDEA is a María de Maeztu Excellence Unit supported by Grant CEX2021-001198, funded by MICIU/AEI/10.13039/501100011033., With funding from the Spanish government through the "Maria de Maeztu Centre of Excellence" accreditation (CEX2021-001198)., File list Main files - README.md: workflow guide. - README_DATASETS.txt: dataset metadata. - requirements.txt: Python dependencies. - rf_script.py: random-forest imputation. - principal_stress_rf_script.py: principal-stress imputation. - final_dataset_script.py: final data merge and deterministic recalculation. - lightgbm_base_script.py: baseline LightGBM workflow. - lightgbm_normalization_script.py: normalized LightGBM workflow and figures. 01_Data_Curation - GEoREST_induced_seismicity_database_v20.11.2022.xlsx/.csv: source dataset. - DataCuration.xlsx: selected, standardized, literature-filled, and deterministic data. - Literature/RockProperties_4.png to RockProperties_8.png: literature tables used for filling rock properties. 02_RF_Imputation - RF_Datacurated.xlsx: random-forest-imputed data. - Audit_Outputs/principal_stress_values.csv: imputed principal stresses. - Audit_Outputs/principal_stress_validation.csv: stress-imputation validation. 03_Model_Training - Final_Imputed_Dataset.xlsx/.csv: final model input. - Previous_Baseline/Mean_RockSiteInjection.xlsx: earlier comparison dataset. - Results/model_comparison.csv: model-performance summary. - Results/new_v2_all_row_predictions.csv: model predictions. - Results/normalization_ranges.csv: training-set normalization ranges. - Results/normalized_labelled_model_inputs.csv: normalized model inputs. - Results/comparison_report.json: model settings and validation report. - Results/LightGBM_V2_TestV1_model.txt: trained LightGBM model. - Figures/F10.png, F21.jpg, F31.png, and F32.png: manuscript figures., Peer reviewed