This supporting information includes additional text, figures, and tables regarding the newly derived sodium-chloride fluid resistivity model and the magnetotelluric (MT) inversion methodology. It also includes additional simulated electrical resistivity results that are not explicitly presented in the main text; the three simulations featured in the main text are selected to represent this larger set of simulations.

Six supporting datasets are also described in this document. The first is a compilation of previously published laboratory-measured electrical resistivity data taken at various temperatures and salinities for sodium-chloride fluids. While all of this data is considered reliable, the measurement accuracy is variable, since the data were compiled from work that was published over a time period ranging from 1907 to 2009. These data are used to derive a thin-plate spline model that permits estimation of sodium-chloride fluid resistivity over an extensive range of salinities (6 to 321,420 mg/L) and temperatures (0 to 309°C). A table of fluid resistivity, as estimated by the spline model, that covers salinities from 0.001 to 5.5 mol/L over a temperature range of 0 to 300°C is also provided. A zip folder containing an R script and the necessary input files to use this spline model independently is included as well. Tables containing the hydrologic modeling results are provided for the three simulations featured in the main text. Lastly, the MT forward responses and calculated inversion fits to those responses are enclosed. The forward responses include both noise-free MT curves and those with 2% Gaussian noise added; the noisy curves were used exclusively for the inverse analysis.

Dataset S1.

Compilation of published sodium-chloride resistivity data measured at numerous temperatures (0 to 309°C) and salinities (6 to 321,420 mg/L). These data are used for the fitting of the spline fluid resistivity surface derived in this study. Measurements reported at non-unique temperature and salinity are averaged; these data are flagged in the “notes” column of the dataset. Literature sources for the data include Noyes [1907]; National Research Council [1930]; Quist and Marshall [1968]; Ucok et al. [1980]; Zimmerman et al. [1995]; Ho et al. [2000]; and Lide and Haynes [2009]. This dataset includes the following fields:

ID: Unique data identification number.

Conc_mol_L: Fluid concentration, in moles per liter.

Temp_C: Fluid temperature, in degrees Celsius.

Resistivity_ohm_m: Fluid resistivity, in ohm-meters.

Source: Data reference.

Conc_g_L: Fluid concentration, in grams per liter.

Conc_mg_L: Fluid concentration, in milligrams per liter.

Notes: Comments regarding the data.

Dataset S2.

Table presenting sodium-chloride fluid resistivity as a function of temperature (0 to 300°C) and salinity (0.001 to 5.5 mol/L). This table is produced by using the thin-plate spline model that is derived as part of this study. Temperatures increase by column while salinities increase by row.

Dataset S3.

A .zip folder including an R script and associated data files required to estimate resistivity as a function of temperature and salinity by using the thin-plate spline model derived in this study. This model is developed using R version 3.4.1. [R Core Team, 2017] and RStudio version 1.0.153. Reference the “READ ME.txt” file that is included in the .zip folder for instructions on using the script. The subfolders and files included in the .zip are as follows:

NaCl_Fluid_Resistivity_Model: Master folder that contains all other files and folders.

input: Subfolder containing files that are read in by the model, including calibration data (Literature_Resistivity_Data.csv), an example of input data (Input_Example.csv), and an input data template (Input_Template.csv). The calibration data is identical to Dataset S1 described above. The Input_Example.csv includes fields for a unique identification number (ID), spatial location (X and Y), fluid temperature (temp_C), and fluid concentration (conc_mol_L). The Input_Template.csv file only includes fields for fluid temperature (temp_C) and fluid concentration (conc_mol_L). Additional columns can be added to the Input_Template.csv as needed; model results will be appended after the last column of the input data.

NaCl_Fluid_Resistivity_Model.Rproj: RStudio project file that is used to set the working directory for the model.

output: Subfolder containing an example of model output data (Resistivity_Results_Input_Example.csv). The fields for this example data are the same as Input_Example.csv, except the model’s resistivity results have been appended to the input example data (fluid_resis_ohm_m). The model will automatically generate model output to this folder with a naming convention as follows: Resistivity_Results_<Input Data Filename>

READ_ME.txt: Text file containing instructions for using the model.

scripts: Subfolder containing the R script used for running for the model (NaCl_Fluid_Resistivity_Model.R).

Dataset S4.

A .zip folder including tables that present the hydrologic modeling results from the final timestep of the three simulations (Damköhler number = 0.2, 1, and 99) featured in main text. The fields in these files are identical to each other and include:

x_m: Horizontal distance for model node location, in meters.

elevation_m: Elevation for model node location, in meters.

conc_kgsolute_kg_solution: Fluid concentration, in kilograms of solute per kilogram of solution.

density_g_L: Fluid density, in grams per liter.

viscosity_Pa_s: Fluid viscosity, in Pascals per second.

temp_C: Fluid temperature, in degrees Celsius.

conc_mol_L: Fluid concentration, in moles per liter.

conc_mg_L: Fluid concentration, in milligrams per liter.

fluid_resis_ohm_m: Fluid resistivity, in ohm-meters.

archie_ohm_m: Effective resistivity from Archie’s law, in ohm-meters.

glover_ohm_m: Effective resistivity from the Glover equation, in ohm-meters.

Dataset S5.

A .zip folder containing MT forward model responses at synthetic stations for the three simulations that are featured in the main text. These files are in the industry-standard EDI file format as generated by WinGLink. EDI files for noise-free forward modeled responses are included in addition to that with 2% Gaussian noise added. Gaussian noise is not added to the tipper data, as the tipper is not utilized in the inversions in order to be more conservative. These files are organized by the Damköhler number associated with each simulation (see Figure 5 and Table S1). Horizontal distance and elevation of the stations are indicated in the “Synthetic_Station_Locations.csv” file. The subfolders and files included in this .zip folder are:

Folder_Model_EDI_Files: Master folder that contains all other files and folder.

No_noise_EDI_Files: Noise-free EDI files for all synthetic stations (stations 1 through 43). These files are organized in subfolders that are named after the Damköhler number of the corresponding simulation. The fields in these files are discussed in the WinGLink documentation and software.

Noisy_EDI_Files: Noisy (2% Gaussian noise) EDI files for all synthetic stations (stations 1 through 43). These files are organized in subfolders that are named after the Damköhler number of the corresponding simulation. The fields in these files are discussed in the WinGLink documentation and software.

Synthetic_Station_Locations.csv: Synthetic station locations within the model domain. Fields include a station identification number (ID), horizontal distance location in meters (X_meters), and an elevation in meters (Z_meters) for each station.

Dataset S6.

A .pdf presenting two-percent Gaussian noise forward modeled MT responses (Original) with inversion fits (Calculated) for each synthetic station. The apparent resistivity and phase of the transverse electric (TE) and transverse magnetic (TM) modes of the MT response are shown. Inversion root-mean-squared error (rms) is shown for each station. The three simulations featured in the main text (Damköhler number = 0.2, 1, and 99) are all represented.

All related notes are included in the methods section above or in above referenced readme files included in the individual code/dataset files in the collection.