Determination of permeability data and 3-D modelling of the host rock and sinters from a geothermal field: Los Geysers, northern Trans-Mexican Volcanic Field

This data article describes the connected pore cluster data from segmented nano-images of rocks related to a geothermal system. The collected samples include two (2) vesicle-amygdaloidal basalt (host rock) and four (4) horizons collected from a siliceous sinter mound (rock precipitated from hot waters). All the samples have undergone computed tomography scanning using a SkyScan 2211 multiscale X-ray nano-CT system (Bruker®), and the slices were analyzed using the Digital Rock Physics (DRP) approach. Pore volume and fluid permeability in the three directions were calculated with scripts of Python (v.3.9) and the visualizations of the 3D models were run with Paraview (v.5.10) software. The petrophysical properties, diagrams, and figures were produced by stacking the 2D projections (8-bit grayscale *.png images format) from the scanning. Raw data (images) were deposited in a repository, which has granted a persistent identifier (Mendeley Data: https://data.mendeley.com/datasets/srpxhpd37p/2). This article provides a study case to handle the data that test the interconnectivity and ability to transport fluids and/or exogenous matter carried during high-flow events in rocks outcropping at the surface level of a geothermal system.


Value of the Data
• The provided data is of invaluable importance as it represents non distructive 3D volumes for hydrothermal precipitates along with their host rock where both show the internal structure. Its importance also exist in being a representative for similar deposits around the world or share similar physico-chemical conditions of similar system, as reveald by González-Guzmán et al. (2022). • Petrophysicits and geologists are the main researchers who can reuse these data.
• The provided 3D nano-CT data can be used for petrophysical comparative studies on similar hydrothermal precipitations/ systems. • The 2-phase flow of the hydrothermal fluid along with its contained gas can be modelled for environmental history purposes. • The provided data for the host rock can be used for studying the cementation processes by modelling the filling of pore by silicious precipitations through what is called the processbased rock modelling.
The processed experimental data and their construction strategy are fully detailed in the next section. Table 1 presents the calculated coefficient of permeability (Darcy) of the six samples .  Fig. 11 ), was simulated on the six samples using the percolation algorithm in the OpenPNM python package [2] . The Hg saturation is relative to the sample's porosity meaning the Hg saturation in Fig. 11 is normalized between 0-1, regardless of the Hg intruded quantity. The MICP values of both ARU-144 and ARU-145 represent only the Hg volume intruded into the surficial pore spaces on the sample's six sides.

Experimental Design, Materials and Methods
Six samples were collected from Los Geysers geothermal field (20.53624 °, -100.54657 °; ∼1800 m.a.s.l.; Fig. 1 ) and scanned using n-XRT computed tomography, in order to test their interconnectivity and their ability to transport fluids and/or organic matter. The analyzed samples include two aliquots collected from the local host rock and four others collected from a sinter sample. The sinter specimen corresponds to the wall of a relict sinter mound. From the whole mass, four subsamples were taken and target individual horizons. These subsamples are representative of the geothermal field. They were labeled alphabetically in the same order (from a[bottom] to d [top]). Carbonaceous material present in the pores of samples CIC7a and CIC7d was extracted for radiocarbon dating [3] . The bulk organic material extracted from the sample CIC7a yielded an age range between 6776 and 6673 cal yr B.P. (1 σ ), whereas the organic matter ith a rotation step of ∼0.2 °. The resolution is controlled by sample position between the X-Ray tube (with submicron spot size) and the detector. Three-D reconstruction from a set of CT raw images employs several computational technologies to tackle the inverse problem going from 2D images to 3D visualizations and models. Here, diverse aspects of the modelling protocol and its application for calculating the poro-perm properties of the analyzed rocks is presented. All the samples have undergone CT scanning using the SkyScan 2211 multiscale X-ray nano-CT system, then analyzed using the Digital Rock Physics (DRP) approach to test their permeability, effective porosity, and capillary pressure.
DRP concerns modelling and computing the rock's physical and petrophysical properties through, mostly 2-phases (voids and solids), segmented 3D CT images of that rock by using an adequate algorithm or solver [6] . DRP workflow begins with 3D CT image acquisition, image processing, and finally computing and simulating the desired physical properties. The DRP method incorporate two categories: direct pore-scale modelling methods, where physical models or solvers, such as Lattice-Boltzmann Method (LBM) (e.g. [7 , 8] , are used directly on the segmented image to calculate the desired properties; and the Pore Network Modelling (PNM) technique (e.g. [9][10][11][12] , where the pore space is simplified into pores and throats which are in turn represented by simple geometrical shapes, especially, spheres and cylinders. For an overview, the author recommends [13 , 14] . In this work, the PNM approach was used, as it is computationally affordable, in comparison to the direct simulation techniques that would take days to calculate these petrophysical properties. The PNM technique has two major steps: simplifying the irregular pore space into regular spheres and cylinders and this is called the pore network extraction; and calculating the rock properties. Considering the first step, the SNOW algorithm [11] which is implemented in the Porespy [15] Python package was used. This algorithm uses the distance map and watershed concepts to simplify the pore space into regular spheres and cylinders. The second step, therefore, that includes calculating the fluid permeability of rock samples in the three directions, was accomplished using the OpenPNM [2] Python package on the extracted pore network from the first step. Both the extracted pore networks and the 3D porous rock samples were then exported to be visualized in the Paraview ( [16] ; v.5.10) software.
Firstly, the six samples were segmented using the Otsu's method. The sizes and porosities of the six samples are shown in Table 1 . Then, the homogeneity of the rock samples is measured using the two-point correlation function reported by [17] . This is to test the ability of the scanned sample for having a representative elementary volume (REV). Additionally, this porepore, two-point correlation function describes the probability of finding every possible pair of voxels, with a known distance between each other, belonging to the same phase (pore) [9] . The 2-points correlation functions of the samples are shown in Fig. 2 which reveals that ARU-144 and CIC7d samples are the only ones that may have a representative elementary volume, while the other samples are heterogenous. Consequently, all the samples were downscaled by a factor 0.5 for our workstation to be able to handle these sizes in a short time. Nonetheless, porositybased REVs were extracted from ARU-144 and CIC7d and compared with the resized data in order to assure that the downscaling reserves the same morphology of the original images. We use Google Colab [18] cloud as our workstation for that purpose. All the samples were then undergone network extraction, permeability estimation, and finally both image datasets and their pore networks were exported to Paraview (v.5.10) for 3D visualization.
In the PNM methodology, the pore space is classified into pore bodies, represented by spheres, and pore throats which connect 2 pore bodies and are represented by tubes. Due to the computational cost, the data was resized by a factor 0.5 then the pore network was extracted and finally the permeability was calculated. To test the reliability of the resizing method, the porosity ( ϕTot) was tested and found to be like the original. Thus, downscaling the big image

Ethics Statements
Ethics statements are not required for the presented data. Our work did not involve human subjects, animal experiments, nor collect data from social media platforms.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data Availability
Two-D CT images of the host rock and layers of a sinter mound from Los Geysers (northern TMVB) (Original data) (Mendeley Data).