Checking for non-preferred file/folder path names (may take a long time depending on the number of files/folders) ...

Simulation Data from Rice et al. (2020), Numerical investigation of wellbore methane leakage from a dual-porosity reservoir and subsequent transport in groundwater


Authors:
Owners: This resource does not have an owner who is an active HydroShare user. Contact CUAHSI (help@cuahsi.org) for information on this resource.
Type: Resource
Storage: The size of this resource is 95.1 MB
Created: Aug 21, 2020 at 10:39 p.m.
Last updated: Nov 23, 2020 at 11 p.m. (Metadata update)
Published date: Nov 23, 2020 at 11 p.m.
DOI: 10.4211/hs.290ba9fc5b654f9f90ff59bafa5bfe98
Citation: See how to cite this resource
Sharing Status: Published
Views: 1109
Downloads: 51
+1 Votes: Be the first one to 
 this.
Comments: No comments (yet)

Abstract

Three‐dimensional, multiphase simulations are used to analyze migration of methane leakage from a hydrocarbon wellbore. The objective is to evaluate the relevance and importance of coupling fast, advective transport of methane through fractures with slower, diffusive transport in the shale matrix below a freshwater aquifer on water quality assuming dual‐domain mass transfer (DDMT) in the reservoir by using the Multiple Interacting Continua (MINC) as implemented in TOUGH2. The conceptual model includes a methane gas‐phase leak from a wellbore 20‐30 m below an aquifer; multiphase, buoyant transport through shale partially saturated with brine; and, after methane leakage reaches groundwater, multiphase transport under varying lateral groundwater flow gradients. Results suggest that DDMT affects the rate of methane reaching groundwater by (i) providing long‐time secondary storage in less‐mobile pore space and (ii) creating larger methane‐plume diameters than those predicted by a single‐domain advection‐diffusion equation. Compared to models without DDMT, these factors combine to increase methane flow rates by an order of magnitude across the base of the aquifer 100 years after leakage begins. In the simulated aquifer, dissolution of gas‐phase plumes leads to bimodal aqueous‐phase methane breakthrough curves in a simulated water well 100 m downstream from leakage, with peak concentrations appearing decades after a one‐year pulse of leakage. The major implication is that DDMT in the reservoir can explain newly discovered methane concentrations in water wells attributable to older leakage events. Therefore, remediation of abandoned or legacy wells with wellbore integrity loss may be necessary to prevent future incidents of groundwater contamination.

Subject Keywords

Coverage

Temporal

Start Date:
End Date:

Content

README.txt

This archive contains files related to the manuscript, NUMERICAL INVESTIGATION OF WELLBORE METHANE LEAKAGE IN DUAL-DOMAIN POROUS MEDIA AND SUBSEQUENT TRANSPORT IN GROUNDWATER by Amy Rice, John McCray, and Kamini Singha, for Water Resources Research.

It contains two types of files:
(1) Selected TOUGH2 input files (.dat file extension)
There are four input files. They pertain to simulations of methane transfer from a leaking hydrocarbon production well below groundwater. They are:
reservoir_DDMT.dat
reservoir_noDDMT.dat
aquifer_gwv2_DDMTinreservoir.dat
aquifer_gwv2_noDDMTinreservoir.dat

The first and second .dat files are used to simulate methane leakage in a shale reservoir, parameterized as the Pierre Shale underlying the Denver Basin aquifer system. The first file includes dual-domain mass transfer (DDMT), via the MINC method, in the shale. The second .dat file does not include DDMT.
The third and fourth .dat files are used to simulate methane migration in an aquifer, parameterized as the Laramie-Fox Hill Aquifer, which is the deepest productive aquifer of the Denver Basin aquifer system.  


(2) Selected files used to make figures in the paper (.mat and .txt file extensions) 
There are two .mat files. They each contain two 5000101x1 double matrices, named "DG" and "TIME", with DG providing density of methane (kg/m^3) everywhere in the modeled domain of the aquifer and TIME providing model time in seconds for simulations with groundwater velocity=0.01 m/d. They can be opened using MATLAB or Octave.

The .mat files are:

DDMT_LFH.mat (the filename stands for "Dual Domain Mass Transfer, Laramie-Fox Hills") was used to plot the solid black line on Figure 7b. DDMT_LFH.mat was also used to as a base case in the paper's Supplementary Information, Figure S2 top panel. 
noDDMT_LFH.mat (the filename stands for "No Dual Domain Mass Transfer, Laramie-Fox Hills") was used to plot the dashed red line on Figure 7b. noDDMT_LFH.mat was also used to as a base case in the paper's Supplementary Information, Figure S2 bottom panel.

The .txt files contain selections from the .mat files. Instead of methane concentrations for the entire domain, the .txt files show methane concentration through time for an observation well 100-m downstream from a leaking well screened in the bottom 10 m of the aquifer, 90-100 m below ground surface. They can be opened using any text editor.

The .txt files are:

DDMT_LFH.txt 
noDDMT_LFH.txt (following the same naming convention as the .mat files)

To extract specific well locations from the .mat files, it is necessary to know the total number of cells in the simulated aquifer, which is 50001. Also, to extract model data for a specific observation well, the index number of the bottom interval of the observation wells must be known. These are:

index=2361; %(index of cell O10, obs 1, not including header cell in output file)
index=2381; %(index of cell O30, obs 2, not including header cell in output file)
index=2489; %(index of cell P40, obs 3, not including header cell in output file)
index=2499; %(index of cell P50, obs 4, not including header cell in output file)
index=2421; %(index of cell O70, obs 5, not including header cell in output file)
index=2441; %(index of cell O90, obs 6, not including header cell in output file)

After loading the .mat files, one index is selected, and the following code may be using in MATLAB:

total_cells=50001; %(total models cells output, including extra (source) cell)
gas=DDMT_DG;
time=DDMT_t;
for ii=1:(length(gas)-1)/total_cells
    gas_DDMT(ii)=gas(index+1+total_cells*(ii-1)); 
    t_DDMT(ii)=time((total_cells)*ii)/(60*60*24*365.25);
end
plot(t_DDMT,gas_DDMT)

Related Resources

This resource is referenced by Rice, A.K., McCray, J.E. and Singha, K. (2020). Numerical investigation of wellbore methane leakage from a dual-porosity reservoir and subsequent transport in groundwater. Water Resources Research, https://doi.org/10.1029/2019WR026991

How to Cite

Rice, A., K. Singha (2020). Simulation Data from Rice et al. (2020), Numerical investigation of wellbore methane leakage from a dual-porosity reservoir and subsequent transport in groundwater, HydroShare, https://doi.org/10.4211/hs.290ba9fc5b654f9f90ff59bafa5bfe98

This resource is shared under the Creative Commons Attribution-NoCommercial CC BY-NC.

http://creativecommons.org/licenses/by-nc/4.0/
CC-BY-NC

Comments

There are currently no comments

New Comment

required