2D Porosity Assessment in Randomly Generated 3D Savonnières Carbonate Samples
The high-value samples from random-3D-sample-generator were then analysed using open-source tools.
The first analysis was the porosity of each sample. Aim was to quantify the absolute and effective porosities of each sample. First utilizing 2D analysis i.e. by slice
First tool used was the BoneJ plugin in Fiji-ImageJ The following parameters were determined:
- Pore shape - using the surface area to volume ratio
- Connectivity - using the Euler characteristics
- Pore Size Distribution - using the difference between the largest and smallest pores
The table of results is shown below. 1st Column - Sample number and the coordinates for each sample 2nd Column - Number of voxels in largest connected pore space 3rd Column - Size of the smallest disconnected pore space 4th Column - The size of the median disconnected pore space 5th Column - The amount of pore space attributed to the largest interconnected space 6th Column - The pore distribution, with "W" denoting "Wide"
Second tool used was Python
Total Porosity for each sample was calculated as follows in porosity_calc: Total Porosity = Total Pore Voxels/Total Number of Voxels The results for each sample were then plotted on a graph.
I then set out to calculate the pore distribution along the length of each sample. This is outlined in pore_dis. I then incorporated the coordinates of each sample. This is outlined in por_dis_coor. I then combined each of these into a loop that calculated for each sample, This is outlined in por_dis_coor_comb. Finally, I separated the original core sample into 3 regions. From the top of the sample to 1500 micrometres was the "Top Region" From 1500 to 3000 micrometres was the "Middle Region" From 3000 to 4500 micrometres was the "Bottom Region" I then calculated the average porosity and standard deviation for all the samples by region and for the whole length This is outlined in pore_regions
The output is shown below:
