This data release contains model input and output files for simulating and predicting thermal spring flows at Hot Springs National Park (HOSP), Hot Springs, Arkansas. A three-dimensional hydrogeologic framework of the Hot Springs anticlinorium beneath Hot Springs National Park was constructed to represent the complex hydrogeology of HOSP and surrounding areas to depths exceeding 9,000 feet below ground surface. The framework, composed of 6 rock formations and 1 vertical fault emplaced beneath the thermal springs, was discretized into 19 layers, 429 rows, and 576 columns and incorporated into a 3-dimensional steady-state groundwater-flow model constructed in MODFLOW-2005. Historical daily mean thermal spring flows were simulated for one stress period of approximately 34 years (1980–2014), chosen to represent the period of record for historical climate data used in the quantification of the boundary conditions. The groundwater-flow model was manually calibrated to historical daily mean thermal spring flows of 88,000 cubic feet per day observed over a 12-year period of record (1990–1995 and 1998–2005) at the thermal springs collection system. Calibration was achieved by calculating starting heads and general head boundary conditions from the Bernoulli equation and then adjusting the horizontal and vertical hydraulic conductivities of the rock formations and vertical fault and the hydraulic conductance of head-dependent flux boundaries. The groundwater-flow model was coupled to a surface-water model developed in the Precipitation-Runoff Modeling System (PRMS) by using PRMS-simulated gravity drainage as a specified flux recharge boundary condition in the groundwater-flow model. Together, the MODFLOW and PRMS models were used to (1) locate the areas of groundwater recharge to the thermal springs by using forward and reverse particle-tracking capabilities of MODPATH, (2) simulate the effects of variable recharge rates on the spring flows at the thermal springs, and (3) assess possible effects of climate and land-use change on the long-term variability of spring flows at the thermal springs.

Geothermal power plants typically show decreasing heat and power production rates over time. Mitigation strategies include optimizing the management of existing wells - increasing or decreasing the fluid flow rates across the wells - and drilling new wells at appropriate locations. The latter is expensive, time-consuming, and subject to many engineering constraints, but the former is a viable mechanism for periodic adjustment of the available fluid allocations. Data and supporting literature from a study describing a new approach combining reservoir modeling and machine learning to produce models that enable strategies for the mitigation of decreased heat and power production rates over time for geothermal power plants. The computational approach used enables translation of sets of potential flow rates for the active wells into reservoir-wide estimates of produced energy and discovery of optimal flow allocations among the studied sets. In our computational experiments, we utilize collections of simulations for a specific reservoir (which capture subsurface characterization and realize history matching) along with machine learning models that predict temperature and pressure timeseries for production wells. We evaluate this approach using an "open-source" reservoir we have constructed that captures many of the characteristics of Brady Hot Springs, a commercially operational geothermal field in Nevada, USA. Selected results from a reservoir model of Brady Hot Springs itself are presented to show successful application to an existing system. In both cases, energy predictions prove to be highly accurate: all observed prediction errors do not exceed 3.68% for temperatures and 4.75% for pressures. In a cumulative energy estimation, we observe prediction errors that are less than 4.04%. A typical reservoir simulation for Brady Hot Springs completes in approximately 4 hours, whereas our machine learning models yield accurate 20-year predictions for temperatures, pressures, and produced energy in 0.9 seconds. This paper aims to demonstrate how the models and techniques from our study can be applied to achieve rapid exploration of controlled parameters and optimization of other geothermal reservoirs. Includes a synthetic, yet realistic, model of a geothermal reservoir, referred to as open-source reservoir (OSR). OSR is a 10-well (4 injection wells and 6 production wells) system that resembles Brady Hot Springs (a commercially operational geothermal field in Nevada, USA) at a high level but has a number of sufficiently modified characteristics (which renders any possible similarity between specific characteristics like temperatures and pressures as purely random). We study OSR through CMG simulations with a wide range of flow allocation scenarios. Includes a dataset with 101 simulated scenarios that cover the period of time between 2020 and 2040 and a link to the published paper about this project, where we focus on the Machine Learning work for predicting OSR's energy production based on the simulation data, as well as a link to the GitHub repository where we have published the code we have developed (please refer to the repository's readme file to see instructions on how to run the code). Additional links are included to associated work led by the USGS to identify geologic factors associated with well productivity in geothermal fields. Below are the high-level steps for applying the same modeling + ML process to other geothermal reservoirs: 1. Develop a geologic model of the geothermal field. The location of faults, upflow zones, aquifers, etc. need to be accounted for as accurately as possible 2. The geologic model needs to be converted to a reservoir model that can be used in a reservoir simulator, such as, for instance, CMG STARS, TETRAD, or FALCON 3. Using native state modeling, the initial temperature and pressure distributions are evaluated, and they become the initial conditions for dynamic reservoir simulations 4. Using history

Geothermal power plants typically show decreasing heat and power production rates over time. Mitigation strategies include optimizing the management of existing wells - increasing or decreasing the fluid flow rates across the wells - and drilling new wells at appropriate locations. The latter is expensive, time-consuming, and subject to many engineering constraints, but the former is a viable mechanism for periodic adjustment of the available fluid allocations. Data and supporting literature from a study describing a new approach combining reservoir modeling and machine learning to produce models that enable strategies for the mitigation of decreased heat and power production rates over time for geothermal power plants. The computational approach used enables translation of sets of potential flow rates for the active wells into reservoir-wide estimates of produced energy and discovery of optimal flow allocations among the studied sets. In our computational experiments, we utilize collections of simulations for a specific reservoir (which capture subsurface characterization and realize history matching) along with machine learning models that predict temperature and pressure timeseries for production wells. We evaluate this approach using an "open-source" reservoir we have constructed that captures many of the characteristics of Brady Hot Springs, a commercially operational geothermal field in Nevada, USA. Selected results from a reservoir model of Brady Hot Springs itself are presented to show successful application to an existing system. In both cases, energy predictions prove to be highly accurate: all observed prediction errors do not exceed 3.68% for temperatures and 4.75% for pressures. In a cumulative energy estimation, we observe prediction errors that are less than 4.04%. A typical reservoir simulation for Brady Hot Springs completes in approximately 4 hours, whereas our machine learning models yield accurate 20-year predictions for temperatures, pressures, and produced energy in 0.9 seconds. This paper aims to demonstrate how the models and techniques from our study can be applied to achieve rapid exploration of controlled parameters and optimization of other geothermal reservoirs. Includes a synthetic, yet realistic, model of a geothermal reservoir, referred to as open-source reservoir (OSR). OSR is a 10-well (4 injection wells and 6 production wells) system that resembles Brady Hot Springs (a commercially operational geothermal field in Nevada, USA) at a high level but has a number of sufficiently modified characteristics (which renders any possible similarity between specific characteristics like temperatures and pressures as purely random). We study OSR through CMG simulations with a wide range of flow allocation scenarios. Includes a dataset with 101 simulated scenarios that cover the period of time between 2020 and 2040 and a link to the published paper about this project, where we focus on the Machine Learning work for predicting OSR's energy production based on the simulation data, as well as a link to the GitHub repository where we have published the code we have developed (please refer to the repository's readme file to see instructions on how to run the code). Additional links are included to associated work led by the USGS to identify geologic factors associated with well productivity in geothermal fields. Below are the high-level steps for applying the same modeling + ML process to other geothermal reservoirs: 1. Develop a geologic model of the geothermal field. The location of faults, upflow zones, aquifers, etc. need to be accounted for as accurately as possible 2. The geologic model needs to be converted to a reservoir model that can be used in a reservoir simulator, such as, for instance, CMG STARS, TETRAD, or FALCON 3. Using native state modeling, the initial temperature and pressure distributions are evaluated, and they become the initial conditions for dynamic reservoir simulations 4. Using history

Subsurface data analysis, reservoir modeling, and machine learning (ML) techniques have been applied to the Brady Hot Springs (BHS) geothermal field in Nevada, USA to further characterize the subsurface and assist with optimizing reservoir management. Hundreds of reservoir simulations have been conducted in TETRAD-G and CMG STARS to explore different injection and production fluid flow rates and allocations and to develop a training data set for ML. This process included simulating the historical injection and production since 1979 and prediction of future performance through 2040. ML networks were created and trained using TensorFlow based on multilayer perceptron, long short-term memory, and convolutional neural network architectures. These networks took as input selected flow rates, injection temperatures, and historical field operation data and produced estimates of future production temperatures. This approach was first successfully tested on a simplified single-fracture doublet system, followed by the application to the BHS reservoir. Using an initial BHS data set with 37 simulated scenarios, the trained and validated network predicted the production temperature for six production wells with the mean absolute percentage error of less than 8%. In a complementary analysis effort, the principal component analysis applied to 13 BHS geological parameters revealed that vertical fracture permeability shows the strongest correlation with fault density and fault intersection density. A new BHS reservoir model was developed considering the fault intersection density as proxy for permeability. This new reservoir model helps to explore underexploited zones in the reservoir. A data gathering plan to obtain additional subsurface data was developed; it includes temperature surveying for three idle injection wells at which the reservoir simulations indicate high bottom-hole temperatures. The collected data assist with calibrating the reservoir model. Data gathering activities are planned for the first quarter of 2021. This GDR submission includes a preprint of the paper titled "Subsurface Characterization and Machine Learning Predictions at Brady Hot Springs" presented at the 46th Stanford Geothermal Workshop (SGW) on Geothermal Reservoir Engineering from February 16-18, 2021.

Subsurface data analysis, reservoir modeling, and machine learning (ML) techniques have been applied to the Brady Hot Springs (BHS) geothermal field in Nevada, USA to further characterize the subsurface and assist with optimizing reservoir management. Hundreds of reservoir simulations have been conducted in TETRAD-G and CMG STARS to explore different injection and production fluid flow rates and allocations and to develop a training data set for ML. This process included simulating the historical injection and production since 1979 and prediction of future performance through 2040. ML networks were created and trained using TensorFlow based on multilayer perceptron, long short-term memory, and convolutional neural network architectures. These networks took as input selected flow rates, injection temperatures, and historical field operation data and produced estimates of future production temperatures. This approach was first successfully tested on a simplified single-fracture doublet system, followed by the application to the BHS reservoir. Using an initial BHS data set with 37 simulated scenarios, the trained and validated network predicted the production temperature for six production wells with the mean absolute percentage error of less than 8%. In a complementary analysis effort, the principal component analysis applied to 13 BHS geological parameters revealed that vertical fracture permeability shows the strongest correlation with fault density and fault intersection density. A new BHS reservoir model was developed considering the fault intersection density as proxy for permeability. This new reservoir model helps to explore underexploited zones in the reservoir. A data gathering plan to obtain additional subsurface data was developed; it includes temperature surveying for three idle injection wells at which the reservoir simulations indicate high bottom-hole temperatures. The collected data assist with calibrating the reservoir model. Data gathering activities are planned for the first quarter of 2021. This GDR submission includes a preprint of the paper titled "Subsurface Characterization and Machine Learning Predictions at Brady Hot Springs" presented at the 46th Stanford Geothermal Workshop (SGW) on Geothermal Reservoir Engineering from February 16-18, 2021.

The files and folders in this data release contain the input and output files and MATLAB algorithms used for simulations described in the associated journal article (https://doi.org/10.1007/s10040-024-02868-x). The algorithms implement a data-driven, mechanistic model of vertical infiltration through the unsaturated zone and recharge to the water table that is developed from water-balance concepts. The model of infiltration and recharge is defined in terms of observed states (such as, the water-table altitude) and unobserved states (such as, fluxes through the unsaturated zone and recharge to the water table) and includes both diffuse and preferential flow through the unsaturated zone to the water table. Estimates of the daily contributions to recharge at the water table from diffuse and preferential flow are performed by interpreting daily time-series records of observations of water-table altitude and meteorological inputs (such as, the liquid precipitation rate, snowmelt rate, and the Potential Evapotranspiration (PET) rate). The modeling approach used here is an extension of concepts of modeling infiltration and rapid recharge originally presented in Shapiro and Day-Lewis (2021) https://doi.org/10.1029/2020WR029110 and Shapiro and others (2022) (https://doi.org/10.1111/gwat.13206). The model of infiltration and recharge to the water table is applied to daily records available at 32 U.S. Geological Survey (USGS) Climate Response Network (CRN) wells located in the Delaware River Basin (DRB) in the eastern United States from January 1, 2005, through December 31, 2021. The daily water-table altitude and the meteorological records described in the associated journal article (https://doi.org/10.1007/s10040-024-02868-x) are included as input files to the MATLAB algorithms described in this data release.

The files and folders in this data release contain the input and output files and MATLAB algorithms used for simulations described in the associated journal article (https://doi.org/10.1007/s10040-024-02868-x). The algorithms implement a data-driven, mechanistic model of vertical infiltration through the unsaturated zone and recharge to the water table that is developed from water-balance concepts. The model of infiltration and recharge is defined in terms of observed states (such as, the water-table altitude) and unobserved states (such as, fluxes through the unsaturated zone and recharge to the water table) and includes both diffuse and preferential flow through the unsaturated zone to the water table. Estimates of the daily contributions to recharge at the water table from diffuse and preferential flow are performed by interpreting daily time-series records of observations of water-table altitude and meteorological inputs (such as, the liquid precipitation rate, snowmelt rate, and the Potential Evapotranspiration (PET) rate). The modeling approach used here is an extension of concepts of modeling infiltration and rapid recharge originally presented in Shapiro and Day-Lewis (2021) https://doi.org/10.1029/2020WR029110 and Shapiro and others (2022) (https://doi.org/10.1111/gwat.13206). The model of infiltration and recharge to the water table is applied to daily records available at 32 U.S. Geological Survey (USGS) Climate Response Network (CRN) wells located in the Delaware River Basin (DRB) in the eastern United States from January 1, 2005, through December 31, 2021. The daily water-table altitude and the meteorological records described in the associated journal article (https://doi.org/10.1007/s10040-024-02868-x) are included as input files to the MATLAB algorithms described in this data release.

This set of data includes four potential annual recharge grids for the State of Maine that were simulated using the Maine Soil-Water-Balance model for 1991 through 2015. The files include a grid representing the uncertainty in the potential recharge and a grid showing the annual average precipitation from the climate dataset that the simulation is based on. A 25-year simulation of potential recharge to groundwater from the Maine Soil-Water-Balance model for the years 1991 to 2015 produced annual results from which the four potential recharge grids were derived. The four are: 25-year mean, median, maximum, and minimum simulated annual potential. A data exclusion zone (see Scientific Investigations Report 2019–5125) has been applied to the recharge datasets, resulting in a dataset that covers most, but not all, of the State of Maine. The potential recharge grids are given in units of inches per year, with a raster grid cell size of 250 meters. The uncertainty in the simulated grid values is the standard deviation grid, which represents the standard deviation of the simulated median recharge grid. The precipitation data used in the 25-year simulation are from DayMet version 3 daily data. The average annual precipitation grid is the calculated annual average from those data. Further details about the generation and application of the data can be found in Scientific Investigations Report 2019–5125.

This set of data includes four potential annual recharge grids for the State of Maine that were simulated using the Maine Soil-Water-Balance model for 1991 through 2015. The files include a grid representing the uncertainty in the potential recharge and a grid showing the annual average precipitation from the climate dataset that the simulation is based on. A 25-year simulation of potential recharge to groundwater from the Maine Soil-Water-Balance model for the years 1991 to 2015 produced annual results from which the four potential recharge grids were derived. The four are: 25-year mean, median, maximum, and minimum simulated annual potential. A data exclusion zone (see Scientific Investigations Report 2019–5125) has been applied to the recharge datasets, resulting in a dataset that covers most, but not all, of the State of Maine. The potential recharge grids are given in units of inches per year, with a raster grid cell size of 250 meters. The uncertainty in the simulated grid values is the standard deviation grid, which represents the standard deviation of the simulated median recharge grid. The precipitation data used in the 25-year simulation are from DayMet version 3 daily data. The average annual precipitation grid is the calculated annual average from those data. Further details about the generation and application of the data can be found in Scientific Investigations Report 2019–5125.

This model archive contains the Soil-Water-Balance (SWB) model used to simulate potential mean annual recharge in the Grand Canyon region for 1981 through 2016. The simulated results are described in the associated U.S. Geological Survey Scientific Investigations Report 2022-xxxx. The model archive includes all the necessary files to document and run the SWB model and process the results as displayed in the accompanying report. The directories in the archive are presented each as a separate .zip file and include an "ancillary" directory, a "bin" directory, a "georef" directory, a "model directory, an "output" directory, and a "source" directory. There is a README file describing all the files and directories in the archive and information on how to run the model.