skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Assessing Uncertainty and Repeatability in Time-lapse Vertical Seismic Profile (VSP) Monitoring of CO2 Injection in a Brine Aquifer, Frio Formation, Texas— A Case Study

Abstract

NRAP TRS

Authors:
Publication Date:
Research Org.:
National Energy Technology Laboratory - Energy Data eXchange; NETL
Sponsoring Org.:
USDOE Office of Fossil Energy (FE)
OSTI Identifier:
1433171
Report Number(s):
e25963cf-d21a-4501-b651-b23287a8d32e
DOE Contract Number:
1022407
Resource Type:
Data
Data Type:
Figures/Plots
Country of Publication:
United States
Language:
English
Subject:
TRS

Citation Formats

Robert Dilmore. Assessing Uncertainty and Repeatability in Time-lapse Vertical Seismic Profile (VSP) Monitoring of CO2 Injection in a Brine Aquifer, Frio Formation, Texas— A Case Study. United States: N. p., 2017. Web. doi:10.18141/1433171.
Robert Dilmore. Assessing Uncertainty and Repeatability in Time-lapse Vertical Seismic Profile (VSP) Monitoring of CO2 Injection in a Brine Aquifer, Frio Formation, Texas— A Case Study. United States. doi:10.18141/1433171.
Robert Dilmore. Wed . "Assessing Uncertainty and Repeatability in Time-lapse Vertical Seismic Profile (VSP) Monitoring of CO2 Injection in a Brine Aquifer, Frio Formation, Texas— A Case Study". United States. doi:10.18141/1433171. https://www.osti.gov/servlets/purl/1433171.
@article{osti_1433171,
title = {Assessing Uncertainty and Repeatability in Time-lapse Vertical Seismic Profile (VSP) Monitoring of CO2 Injection in a Brine Aquifer, Frio Formation, Texas— A Case Study},
author = {Robert Dilmore},
abstractNote = {NRAP TRS},
doi = {10.18141/1433171},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Wed Jun 21 00:00:00 EDT 2017},
month = {Wed Jun 21 00:00:00 EDT 2017}
}

Dataset:

Save / Share:
  • This study was done to assess the repeatability and uncertainty of time-lapse VSP response to CO 2 injection in the Frio formation near Houston Texas. A work flow was built to assess the effect of time-lapse injected CO 2 into two Frio brine reservoir intervals, the ‘C’ sand (Frio1) and the ‘Blue sand’ (Frio2). The time-lapse seismic amplitude variations with sensor depth for both reservoirs Frio1 and Frio2 were computed by subtracting the seismic response of the base survey from each of the two monitor seismic surveys. Source site 1 has been considered as one of the best sites formore » evaluating the time-lapse response after injection. For site 1, the computed timelapse NRMS levels after processing had been compared to the estimated time-lapse NRMS level before processing for different control reflectors, and for brine aquifers Frio1, and Frio2 to quantify detectability of amplitude difference. As the main interest is to analyze the time-lapse amplitude variations, different scenarios have been considered. Three different survey scenarios were considered: the base survey which was performed before injection, monitor1 performed after the first injection operation, and monitor2 which was after the second injection. The first scenario was base-monitor1, the second was basemonitor2, and the third was monitor1-monitor2. We considered three ‘control’ reflections above the Frio to assist removal of overburden changes, and concluded that third control reflector (CR3) is the most favorable for the first scenario in terms of NRMS response, and first control reflector (CR1) is the most favorable for the second and third scenarios in terms of NRMS response. The NRMS parameter is shown to be a useful measure to assess the effect of processing on time-lapse data. The overall NRMS for the Frio VSP data set was found to be in the range of 30% to 80% following basic processing. This could be considered as an estimated baseline in assessing the utility of VSP for CO 2 monitoring. This study shows that the CO 2 injection in brine reservoir Frio1 (the ‘C’ sand unit) does induce a relative change in amplitude response, and for Frio2 (the ‘Blue’ sand unit) an amplitude change has been also detected, but in both cases the uncertainty, as measured by NRMS indicates the reservoir changes are, at best, only slightly above the noise level, and often below the noise level of the overall data set.« less
  • Seismic surveys successfully imaged a small scale C02injection (1,600 tons) conducted in a brine aquifer of the Frio Formationnear Houston, Texas. These time-lapse bore-hole seismic surveys,crosswell and vertical seismic profile (VSP), were acquired to monitorthe C02 distribution using two boreholes (the new injection well and apre-existing well used for monitoring) which are 30 m apart at a depth of1500 m. The crosswell survey provided a high-resolution image of the C02distribution between the wells via tomographic imaging of the P-wavevelocity decrease (up to 500 mls). The simultaneously acquired S-wavetomography showed little change in S-wave velocity, as expected for fluidsubstitution. A rockmore » physics model was used to estimate C02 saturationsof 10-20 percent from the P-wave velocity change. The VSP survey resolveda large (-70 percent) change in reflection amplitude for the Friohorizon. This C02 induced reflection amplitude change allowed estimationof the C02 extent beyond the monitor well and on 3 azimuths. The VSPresult is compared with numerical modeling of C02 saturations and isseismically modeled using the velocity change estimated in the crosswellsurvey.« less
  • Using an orbital vibrator source (2-components), and a 40 level 3-component geophone string, a 6-component crosswell survey was acquired before and after a CO2 injection in a saline aquifer. Decomposition of the two source components and component rotation of both source and sensors created good separation of P- and S-wave energy allowing independent analysis of travel time and reflectivity. A time-lapse VSP was also acquired.
  • Part 1 of this report focuses on results of the western Kentucky carbon storage test, and provides a basis for evaluating injection and storage of supercritical CO 2 in Cambro-Ordovician carbonate reservoirs throughout the U.S. Midcontinent. This test demonstrated that the Cambro- Ordovician Knox Group, including the Beekmantown Dolomite, Gunter Sandstone, and Copper Ridge Dolomite in stratigraphic succession from shallowest to deepest, had reservoir properties suitable for supercritical CO 2 storage in a deep saline reservoir hosted in carbonate rocks, and that strata with properties sufficient for long-term confinement of supercritical CO 2 were present in the deep subsurface. Injectionmore » testing with brine and CO 2 was completed in two phases. The first phase, a joint project by the Kentucky Geological Survey and the Western Kentucky Carbon Storage Foundation, drilled the Marvin Blan No. 1 carbon storage research well and tested the entire Knox Group section in the open borehole – including the Beekmantown Dolomite, Gunter Sandstone, and Copper Ridge Dolomite – at 1152–2255 m, below casing cemented at 1116 m. During Phase 1 injection testing, most of the 297 tonnes of supercritical CO 2 was displaced into porous and permeable sections of the lowermost Beekmantown below 1463 m and Gunter. The wellbore was then temporarily abandoned with a retrievable bridge plug in casing at 1105 m and two downhole pressure-temperature monitoring gauges below the bridge plug pending subsequent testing. Pressure and temperature data were recorded every minute for slightly more than a year, providing a unique record of subsurface reservoir conditions in the Knox. In contrast, Phase 2 testing, this study, tested a mechanically-isolated dolomitic-sandstone interval in the Gunter.« less