lacking of continuous column of CO2 from the injection point to the caprock in a sandstone reservoir model
Hi,
I am injecting CO2 in a high porosity and permeability sandstone reservoir during a period of 30 years. It is an isothermal model.
I have a problem with the CO2 plume. It does not go up along the well. As it can be seen in the attached picture, there is a clear CO2 injection at the injection point and an accumulation of CO2 below the caprock but there is a discontinuity among both. I am missing a continuous column of CO2 going up along the well from the injection point to the caprock.
The reservoir is divided in several layers but all of them have the same porosity and permeability values. The relative permeability and capillary pressure are also the same, and both follow Van Genuchten functions.
Initially, the model was run with TOUGHREACT but once reactive transport was removed and the model ran only with TOUGH2, the problem persisted, so it has to be a physical problem.
May anyone help me, please? Did anyone experience something similar?
Thanks a lot!
Cheers,
Laura
3 replies
-
Laura,
Can you describe you how simulated the well itself (if flow within the well rather than within the formation is your concern; I'm not clear)?
Stefan
-
Hello Stefan,
The well is 1050 m long but only active in the reservoir from -1010 m to -1030 m. The injection zone is then open 20 m. Supercritical CO2 comes out from that zone. After 30 years of injection, it can be seen in the picture that there is a large accumulation of CO2 around the injection zone and below the caprock but it is missing a column of supercritical CO2 going up close to the well till the caprock. I think it is a physical problem of the model but I don't know what mistake can be.I have run the model with both only TOUGH2, as well as with reactive transport (TOUGH2 + TOUGHREACT), obtaining the same CO2 plume shape (without connexion between injection zone and caprock accumulations). Therefore, I think it is not a chemical problem but a physical one.
As I said in my previous entrance, the reservoir has several layers but all of them with the same physical and chemical conditions. However, it seems that CO2 doesn't flow up in a continuous plume.
I hope I made me clear now. However, in case you have any doubt, please, don't hesitate to contact me; I will explain further any detail that you consider relevant for solving the problem/s of my model.
Thanks a lot for your help!!
Kind regards,
Laura
-
Laura,
I understand what it is you wanted to simulate. However, my question is how you actually implemented the well in your model (is it a string of high-permeability elements? With what relative permeability and capillary pressure functions? How did you simulate the injection zone? Did you explicitly include the packers at -1010 and -1030 m? What boundary conditions at the top? How is it connected to the formation? Etc.). You also want to check in the output file whether you get a lateral pressure gradient and flux from the formation to the well elements above Z=-1010 m, and if not, why not.
Please try to answer these questions - if you cannot figure it out, send me your input files (SAFinsterle@lbl.gov).
Good luck!
Stefan