The Bureau of Economic Geology Gulf Coast Carbon Center is currently conducting a $34 million multi-year field study of sequestration and monitoring strategies for long-term storage of carbon dioxide. This work is being performed in conjunction with the Southeast Regional Carbon Sequestration Partnership (SECARB) with support from the National Energy Technology Laboratory (NETL) and the U.S. Department of Energy (DOE) and managed by the Southern States Energy Board (SSEB).
Planning for the project began in November 2006 and we are now reaching the point where data collection begins. The project involves numerous industrial partners, scientific collaborators, and technical subcontractors, each of whom is responsible for different aspects of the study:
The Gulf Coast Carbon Center is responsible for geologic characterization, monitoring design, integration, and near-surface monitoring.
Denbury Resources, Inc. is the site host. They are also responsible for the well preparation and will supply the CO2 for storage and sequestration.
Sandia Technologies LLC is responsible for subsurface monitoring systems design and deployment.
Schlumberger Carbon Services will perform wireline logging and interpretation.
Lawrence Berkeley National Lab is responsible for cross-well, VSP and Continuous Active Seismic Source Monitoring (CASSM). Implementation and evaluation of of noble gas and other tracers and the U-tube and Distributed Temperature System (DTTS).
Lawrence Livermore National Labs and Promore will perform cross well Electrical Resistance Tomography (ERT).
University of Mississippi and Mississippi State University are responsible for groundwater monitoring.
Our aim in presenting this log of real-time updates about the project is to document our research, which is one step in the progress toward developing a process for safe, long-term, subsurface sequestration of carbon. We are aware that our work will involve successes as well as practical problems that must be overcome along the way. No other project in the United States has incorporated so many different geological and technical measurements and we expect to accumulate significant information and experience that will increase confidence and decrease costs of future geologic carbon sequestration projects.
The Study Site
The site for the project is Denbury Resources Cranfield Field in southwest Mississippi. Cranfield field, Mississippi, was discovered in 1946 and abandoned in 1965 at the end of primary oil and gas production. Denbury Resources, Inc. currently operates the field and is a partner in the study.
The study site in Cranfield field, MS includes an injection well and two observation wells.
The detailed area of study (DAS) includes three wells. The westernmost well (F-1) is operated by Denbury and is the site of injection of carbon dioxide into the subsurface geologic reservoir. Two observations wells are approximately 300 ft and 400 ft to the southeast and are referred to as F-2 and F-3, respectively. The observation wells will be equipped with state-of the-art monitoring systems that will track and monitor the carbon dioxide plume as it moves through the reservoir more than 10,000 feet below the surface.
February 2013Update of Accomplishments at Cranfield
Four million metric tons of CO2 have been stored at Cranfield over the past 4 years. This is an important demonstration that large scale CO2 injection is a viable and commercially available technology. Note: Recycling of CO2 at the field for EOR is 6.5 million metric tons, so the 4 million tons of CO2 stored represents actual saline storage.
Modeled lab, and field tests have simulated leakage to groundwater to define the signal that would be detected should this occur. Monitoring of Cranfield groundwater by BEG, Ol' Miss and Mississippi State have detected no CO2 related change to groundwater even in an area with many dozens of 1940s era wells, and plugged and abandoned (P&A) wells.
Pressure in Cranfield field has been elevated above hydrostatic (natural pressure) as much as 1200 psi, which is not insignificant. Microseismic monitoring by RITE (Japan) has detected no microseismicity. This is an important contribution in the context of concern about seismicity related to deep underground injection of CO2.
August, 2012 Like Champagne Bubbles in the Wellbore
We are assessing a new and elegantly simple technique to study the timing and chemical composition of breakthrough. This work was primarily developed by GCCC’s Seyyed Hosseini and Sandeep Verma of Schlumberger.
At breakthrough, supercritical carbon dioxide first enters the well-bore. Because CO2 is lower density than the surrounding brine (about 0.6 compared to 1.1 g/cm3), bubbles of CO2 form and rise to the top of the well, just like bubbles in a champagne flute. The first bubbles to arrive at the top of the well contain the first breakthrough fluid. The next bubbles record the chemical composition of the next bit of fluid, and so on. If the well is shut-in, the bubbles rise in the well-bore and collect in a narrow tubing in which mixing is suppressed. From the top down, the collected fluid represent a time series of breakthrough chemistry.
As a part of the final monitoring program at Ella G. Lees #7 (the Phase II observation well), we assessed the wellbore fluid in order to assess CO2-methane ratio at breakthrough. Reservoir fluid has not been sampled at this well. The Schlumberger Platform Basic Measurement Sonde (PBMS) measures wellbore pressure and temperature, and the Platform Gradiomanometer Sonde (PGMS) measures high resolution average density. Schlumberger-Doll labs analyzed the density of the wellbore fluid. We combined fluid density information calculated by subtracting bottom-hole pressure and temperature from surface pressure and temperature. This provides breakthrough composition data in the same way that on-site sampling with a U-tube does but a lower cost.
The timing of breakthrough can be easily resolved from change in rate of change of tubing pressure at the wellhead. Prior to breakthrough the wellbore is filled with brine. As the CO2 fills the idle tubing, pressure gauges at the wellhead record increase change of pressure as a result of the more less dense supercritcal CO2 displacing the dense brine. This method has been previously validated by comparison of real-time sampling at the DAS. Fluid composition was measured with the U-tube during breakthrough at the observation well F2, which is closer to the injector and began to fill with CO2 on December 16, as evidenced by the increased pressure at the wellhead. Breakthrough began at the second observation well F3 three days later. After the wellbores filled with low-density CO2, the pressure remained high and constant as shown in the graph.
Pressure increases as CO2 bubbles fill the observation wells. The bubbling takes 3-4 days.
Sampling fluid in the shut-in well bore allows for a one-time effort rather than continuous monitoring.
August, 2012 Update of Activities at Cranfield
At Cranfield Field, long-term and time-lapse surveillance of carbon dioxide injection continues, as does data analysis, presentation, and discussion. A special volume of the International Journal of Greenhouse Gas Control Technologies focused on Cranfield results is in preparation.
In order to assess risk and signal from unexpected CO2 leakage to fresh water, we have completed a first-of-its-kind push-pull test to test CO2 interaction with aquifer mineralogy and water under in-situ conditions. Aquifer water was produced, saturated with CO2, and reinjected in a closed look. Reacted brine was withdrawn and analyzed for chemical and isotopic composition. This work is being conducted partly under funding from the American Water Works Association (AWWA).
GCCC student Julie Ditkof traveled to Curtain University in Perth, Australia to improve the 4-D seismic signal using time-lapse cross-equalization. The work attempts to improve CO2 signal in this deep, thin, and complex reservoir.
The Research Institute of Innovative Technology for the Earth (RITE) started collecting data from a recently installed 6-site microseismic array over the whole Cranfield field (in collaboration with WESTCARB). The seismic array is now operational and, while it has recognized distant events, no local seismic events have been detected, however noise from unidentified sources has been noted. The objective of the project is to develop experience with prediction and monitoring methods for microseismicity related to CO2 injection.
The cumulative total stored CO2 mass monitored at Cranfield at the end of the quarter has reached 3,565,107 metric tons. Cranfield is the first of the DOE’s Storage Partnerships to exceed its storage goals.
May 3, 2011 Annual Conference on Carbon Capture and Sequestration
Katherine Romanak presented results from soil-gas monitoring work at Cranfield's P-site at the 10th Annual Conference on Carbon Capture and Sequestration in Pittsburgh, PA. To view the complete poster, "Method for Distinguishing Signal From Noise in the Near-Surface Using Simple Soil-Gas Measurements: Lessons from Natural and Industrial Analogues", please click here.
May 6-8, 2010 Monitoring Network Meeting
The GCCC sponsored a Monitoring Network meeting of the International Energy Agency Greenhouse Gas R&D Programme at Cranfield. During the meeting, engineers demonstrated the venting of CO2 from injection equipment. CO2 is colorless, but it appears as a white stream because it cools rapidly from very high temperature to ambient temperature, which causes water from the air to condense. These water droplets scatter light so that the stream appears white. The force of the release also demonstrates the high pressure at which CO2 is injected into containment systems.
April and May, 2010Tracer Monitoring Test at Cranfield
Researchers of Gulf Coast Carbon Center, Laurence Berkley National Lab, US Geological Survey, Oak Ridge National Lab, and University of Edinburgh conducted a reservoir fluid sampling and tracer monitoring program at Cranfield field, Mississippi. Production wells under CO2 flood were sampled to understand CO2 -related water-rock reactions and their impact on organic chemistry. Tracers consisting of krypton, xenon, a suite of perfluorocarbons, and sulfur hexafluoride were introduced to the CO2 stream of the injector at Detailed Area of Study (DAS). The tracers were detected in the two monitoring wells by frequent sampling using U-tube systems. Two such tracer tests were conducted at different CO2 injection rates. The collected datasets shed light on fluid migration rates and evolution of pathways in the injection zone.
March 7-10, 2010GCCC Collaborates with Department of Energy’s National Energy Technology Laboratory (DOE NETL) on CO2-Rock-Water Interaction Experiments
GCCC’s Jiemin Lu traveled to NETL in Pittsburgh to work with scientists who have developed two autoclave reactors to study rock-water interactions. The system precisely controls both temperature and pressure to carefully replicate the conditions that occur deep underground in the formations used for geologic sequestration of carbon dioxide.
The system’s inner vessel is a gold bag that serves as the reaction container. Brine and rock samples are sealed in this bag. One tube is connected to allow water sample to be taken during the reaction. The gold bag is then placed inside a steel container, which is filled with water. Pressure is applied to the water to change the pressure inside the gold bag. The entire apparatus is placed in a heat chamber, which controls the temperature. The system is closed, so that the samples are never exposed to oxygen or atmospheric pressure.
Lu’s initial experiment will involve Tuscaloosa sandstone and brine collected from the U-tube at Cranfield field during the SEACARB III experiments. Rock and brine will equilibrate for one month, after which CO2 will be added to the system to simulate injection of carbon dioxide. Samples of brine will be taken periodically to study changes in acidity, concentrations of anions and cations, redox, and isotopes.
Lu explains that the study will both “help refine our questions and bound model conditions,” providing new insights into the interactions that occur during geological sequestration.
December 25, 2009 Holiday Season at Cranfield
GCCC scientists maintained an around the clock work schedule throughout the holidays, monitoring the development of the injection experiment. For the full story, click here.
December 6, 2009 Tour of P-Site
Katherine Romanak hosted a tour of P-site, which is the study site established by GCCC for determining how environmental factors affect near-surface monitoring at an engineered location. Romanak demonstrated methods used by BEG scientists to monitor the site and showed the scientists the first soil-gas data set that collected. The scientists discussed preliminary interpretations of the data, as well as numerous areas for future collaboration.
The visitors to the P-site are (from left to right) Stuart Gilfillan (University of Edinburgh, Scotland) Susan Pfiffner (University of Tennessee) Yousif Kharaka (U.S. Geological Survey, Menlo Park, California) Tommy Phelps (Oak Ridge National Laboratory) Kim Gilbert (Ph.D. student, The University of Texas at Austin) and GCCC’s Katherine Romanak.
December 5, 2009 Update – First Week of Injection at SECARB III: Detailed Area of Study (DAS)
We have begun injection in the Detailed Area of Study (DAS) of the SECARB project at Cranfield. The DAS consists of three wells drilled at close spacing: one injection well (CFU 31-F1) and two observation wells (CFU 31-F2 and CFU 31-F3). The spacing from F1 to F2 is 227 feet (69 m) and from F1 to F3 367 ft (111 m). The observation wells are east and downdip of F1. The three wells are completed in the lower Tuscaloosa with a 65 ft (20 m) perforated interval at a top average depth of 10,445 feet. The downdip is 1 to 2 degrees.
A dense array of subsurface instruments have been deployed in the DAS wells designed to: (1) test novel MVA (monitoring, verification, and accounting) tools and improve tool performance evaluation and (2) improve capacity assessment via measurement of saturation and sweep efficiency as a CO2 flood develops.
The instrumentation in the DAS wells allows us to simultaneously collect numerous physical measurements to assess the spatial extent and concentration of CO2. Electrical resistance tomography (ERT) receivers are installed on the casing, above-zone pressure gages, and distributed temperature sensors were installed on both F2 and F3. Baseline geophysical surveys requiring full well diameter, such as cross-well seismic measurements, gravity measurements, and VSP (vertical seismic profiles), were run several months ago prior to perforation. During tubing completion, additional tools were strapped onto the tubing and lowered into the wells, including reservoir pressure and temperature gages on wireline read-outs, tubing conveyed piezoelectric sources fabricated by LBNL (Lawrence Berkeley National Laboratory) in F2 and two receiver strings in F3. The installed seismic system is known as continuous active seismic source monitoring (CASSM).
The DAS is the last portion of the Phase III Early Test to start injection. Injection has been underway at wells to the north of the DAS since April as part of the Phase III high volume injection test (HiVIT). We have observed a relatively strong pressure increase in the DAS as a result of this injection.
The background pressure in the DAS has been increasing as a result of injection in the high volume test at Cranfield.
In mid November, we completed a hydrologic interference test, which demonstrated that the wells are in good hydrologic continuity. We performed a step rate field brine injection into F1, during which the rate of brine injected into the well was repeatedly increased over time. We estimated that the permeability over the perforated interval is 216 mD at F1 and 411 mD at F2.
Pressure gauges have been collecting baseline data since early November. ERT began on November 26, 2009. ERT is returning consistent data, with some intermittent noise issues near the packer and in the F2 well. The theoretical depth of penetration of ERT in monopole mode is 4 meters away from the well. CASSM began operations incrementally and at first performed very well.
On November 30, 2009 the project collected the baseline pulsed neutron log Schlumberger Reservoir Saturation Tool (RST) in F2 and F1. F3 still contains dense workover fluid, that will be produced next week and baseline runs will be conducted at the same time as the first repeat in F2. F1 repeat RST will be at the same time as repeat cross well seismic, gravity, and VSP, as injection will continue into this well in the near term.
At 8:40 AM December 1, 2009, CO2 injection began into CFU 31 F1. The injection rate was initially set by a 10/64 choke on a needle valve at the wellhead to assess the well and formation response. The pressure in the reservoir responded immediately and as expected, stabilizing about 550 psi higher than the initial pressure. The pressure in F2 and F3 has been rising to about 25 psi above initial conditions.
As the well and near-well bore saturated and the back-pressure decreased over the first few hours, we stepped up the rate to 240 tons/day. This was determined to be a moderate rate for the experiment. We plan to step up the injection rate at this well over the next few weeks to double or more, and then step up rates at nearby wells to produce the Phase III million-ton-per-year rate.
Following injection at well F1 (red), pressure at the F2 observation well (blue) and at the F3 observation well (green) increased as expected.
A program of pulsed tracers including 5 PFT’s, SF6, and Xe and Kr was introduced into the injection well CO2 stream over the first 4 days of injection and will be repeated when the injection rate is increased. These tracers will be used to: (1) measure changes in velocity as the plume builds, (2) observe transport mechanism, and (3) estimate dissolution into brine.
To measure these tracer concentrations, we have developed a program of high frequency sampling employing the U-tube in-situ fluid sampler constructed by LBNL and operated in teamwork by the Bureau. Samples are extracted at a rate of one sample every 2 to 4 hours. The USGS operates an in-line high pressure pH meter. The tracer and gas composition of every sample is measured with an LBNL quadrupole mass spectrometer in the field, along with lab duplicates of selected samples.
As of December 5, 2009, the gas species measured in the U-tube is methane at concentrations consistent with dissolved phase, and no increase in CO2 or tracer has been detected. The USGS conducts additional assessments of selected samples for aqueous geochemistry. Baseline alkalinity of 370mg/L as HCO3 has been measured. Fluorescent dye placed in 900 barrels of injection brine at step-rate test has been detected at F2, confirming that the wells are in good connection for single phase transport. Rhodamine was used as a tag in workover water prior to perforation and is an indication of a contaminant that is present but of low quantity. The USGS is also resampling and adding to the HiVIT monitoring program.
The U-tube operation is staffed around the clock by a minimum of two staff per shift. Two additional geochemists are also processing samples. A site supervisor from Sandia Technology manages well issues, and the U-tube N2 and compressor issues on a 3 to 5 days/week schedule. This schedule will continue though arrival of the last tracer estimated to occur mid January. Denbury provides 24-hour oversight of injection.
Example model result prepared by Christine Dougty, LBNL using TOUGH 2 showing the spatial extent of carbon dioxide 10 and 30 days after the initiation of injection. The model predits breakthrough at F2 after 9.5 days and F3 after 22.7 days.
BEG and LBNL have recently completed several models using our new data to estimate the time it will take CO2 to move from F1 to F2 and then F3. The model results are sensitive to saturation and sweep efficiency (the variables that the experiment is designed to assess), also to assumptions about interaction of pressure and gravity on transport. The modeling provides a range of estimates of plume building performance. The fastest estimate of breakthrough (detection of supercritical CO2 arrival at the F2 observation well) is 7 days, with an average of about two weeks. Our past experience during the Frio pilot injection tests showed breakthrough at the observation well on the early end of estimates. Therefore, we began the well-based detection programs prior to injection with gradually increasing intensity during the first week. The average estimated arrival time for F3 is January 4, 2010.
October 13, 2009 Instrumentation Installed at Soil-Gas Monitoring Site
GCCC researchers Changbing Yang and Katherine Romanak continue their work to understand how soil-gas monitoring can best be used to document storage effectiveness. The study site at Cranfield has a suite of the complexities that must be studied and analyzed in order to separate any possible subsurface-sources leakage signal from background variability. These complexities include a constructed gravel pad that contains a plugged and abandoned (P&A) oil well and a 1950’s era pit filled with water and possibly some legacy hydrocarbons. The entire area is surrounded by forest, and the plants are expected to play an important role in the dynamics of carbon dioxide and other gases in the area. The relative importance of the contribution of the pad, the pit, and the plants will be assessed in our analysis, so we therefore refer to the site as the P-site. The Jackson School of Geosciences at the University of Texas generously provided matching funding to GCCC for the instrumentation.
Romanak and Yang with the help of researcher Bob Reedy established four soil-gas monitoring transects within P-site to help distinguish the various influences of the pad, pit, plants, and P&A well on near-surface soil-gas compositions as shown in the diagram of the site.
Schematic diagram of the P-site shows the locations of twelve soil- gas stations relative to the gravel pad, the P&A well, and the mud pit.
Each transect is composed of three to six soil-gas stations. A single soil-gas station includes three 1/8-inch stainless steel gas wells with 6-inch screens set at depths ranging from 2 to 15 feet. The wells are packed with sand in the regions where gas is to be sampled and bentonite clay between sample regions. In addition, a soil-gas well was installed in the detailed area of study (DAS).
A soil-gas well is constructed of 1/8 inch stainless steel tubing and fitted with a 6-inch screen for sampling of gas. These types of wells are permanently installed at various depths within one gas station.
PVC protector pipes emerging from the ground at P-site indicate the locations of soil-gas stations.
Sediment samples were collected from the soil-gas stations. They have been brought back to the lab at GCCC where they will be analyzed for water content, matric potential (an indicator of the potential of water to move through soils), bulk density, organic matter, and texture. These measurements will be used in modeling of the movement of gases at the P-site.
Sediment samples were collected in clear plastic tubes. The different colors visible through the tubes suggest significant complexity of the soils at P-site. The orange and reddish colors may indicate oxidation and reductive processes.
A near-surface observatory, an integrated system for continuously connecting above- and below-ground information was also installed at the P-site. The observatory collects above-ground information, such as air temperature, barometric pressure, precipitation, air CO2 concentration, solar radiation, wind speed and directions, and relative humidity and also subsurface data including soil water content, matric potential, temperature, and soil CO2 concentration at different depths. A solar panel provides power to the observatory. Data from the observatory and soil gas stations will also be used for numerical modeling exercises.
Romanak and Reedy program the data collection system to continuously collect measurements at the weather observatory.
A water well with a depth of 34 feet was also drilled at the P-site. The diameter of the borehole was 3 inches. It was completed with 1-inch diameter PVC pipe with10 feet screen at the bottom. The depth of the water level at P-site is around 30 feet. This is a perched aquifer, as the regional aquifer used for industrial and domestic water supply is roughly 200 to 300 feet below the ground surface.
September 21, 2009 Geomechanical Test begins at Cranfield
With his colleagues at the GCCC, Lawrence Berkeley National Laboratory, and Weston Geophysical, Tip Meckel recently initiated a subsurface geomechanical experiment at Cranfield. The study site for the project is approximately one mile from the DAS site and contains an observation well located about 1,200 feet from an injection well.
The observation well is equipped with downhole tool string on which twelve high-resolution geophones are attached. As CO2 is pumped into the injection well, pressure will increase as anticipated in the subsurface formations. The geophones will continuously monitor for and record any deep-subsurface small-scale acoustic signals that are generated within the injection reservoir and overlying seals. The number and frequency of the acoustic signals will indicate how changes in pressure affect the rock formations.
On September 14th a tool string was lowered more than 10,000 feet underground in the observation well, and initiation of CO2 injection began on September 21st. However, normal operational problems delayed initial data acquisition. A valve problem at the injection wellhead at the surface was detected, which delayed pumping for a few days. In the meantime, some of the geophones near the bottom of the observation well stopped communicating with the surface. Repairs to the problems are underway, and injection and monitoring are expected to continue imminently.
Video: Microseismic tool string is installed in observation well.
September 14, 2009: Work Plan for Analysis of the Sample Cores
GCCC Post Doc Jiemin Lu has begun analyzing the three cores taken from the observations wells at the Detailed Area of Study (DAS). The work plan for the cores involves four major categories of analysis: petrology and mineralogy, chemistry, petrophysics, and geomechanics Many of the tests or analyses will be performed at BEG. In addition other institutions and subcontractors are being contacted to perform specialized analyses.
Neutron scanning (National Energy Technology Labs)
Rock-water reactions (National Energy Technology Labs)
Stable isotope analysis
Permeability and porosity (CoreLabs)
Micro computer tomography (Lawrence Berkley National Labs)
Seismic velocity (Lawrence Berkley National Labs)
CO2 saturation (Lawrence Berkley National Labs)
Residual saturation (Stanford)
CO2 diffusion in seal
Triaxial compression test (CoreLabs, Petroleum Geosciences Engineering at UT)
XRD mineralogy is a specialized form of X-ray diffraction analysis. X-ray diffraction is a common technique used by geologists to identify mineral phases, such as those in the sandstone and shale of the cores retrieved from Cranfield. After the analysis is performed, the diffraction patterns are collected from samples. They are then compared against a database maintained by the International Centre for Diffraction Data and used to identify substances and quantify mineral compositions.
In order to perform quantitative XRD, samples must be ground to a powder, which is then mounted onto a slide for analysis. A major assumption of the technique is that all crystalline orientations in the sample are represented equally on the slide. A problem with some XRD techniques is that pressing powders onto sample holder results in biases the orientation of the crystals.
Lu is the first user a new system purchased by the BEG Mudrocks System Research Laboratory that avoids the types of artifacts that often plague XRD. The sample preparation system consists of a McCrone Micronizing Mill (which allows wet grinding to produce fine and homogenized powder samples in resulting slurries), a Spay drier (which makes truly randomly oriented powder samples), and a centrifuge (to separate clay minerals).
The new equipment at the Mudrocks System Research laboratory that is being used for sample analysis includes a large silver spray dryer, which produces truly random powder samples. The rectangular black equipment on the bench on the left is McCrone MicroNizing mill.
August 31, 2009: Preliminary Results from Environmental Monitoring Survey
During the initial survey of plugged and abandoned (P&A) wells around the detailed area of study (DAS), GCCC researchers Katherine Romanak and Changbing Yang performed reconnaissance soil-gas measurements with a soil probe and a portable gas chromatograph at three sites. A portable gas chromatograph measures the volume percent concentrations of methane (CH4), carbon dioxide (CO2), oxygen (O2), and nitrogen (N2) gas trapped between particles of soil.
A soil probe inserted through the gravel pad at a study site is used to sample soil gases.
The soil-gas probe is typically inserted up to 1 meter into the ground in a region called the vadose zone or the unsaturated zone. In this region, which lies above the aquifer, the soils are not fully saturated by water. The unsaturated zone is a region of significant environmental complexity as a result of microbial activity, water flux, gas transport and exchange with the atmosphere, chemical reactions, and plant root respiration.
Initial measurements showed mostly concentrations of gases that are typically found in air at the sites. However, to separate near-surface gas generation from a deep source that might indicate leakage takes more work. Romanak and Yang will study in detail an old abandoned well site situated near the center of a newly-constructed gravel pad. A swampy 1950’s era mud pit lies to the south of the site. The area around the gravel pad is heavily wooded.
The study site is a plugged and abandoned well at the center of a newly constructed gravel pad.
A 1950’s era mud pit and dense woods are adjacent to the study site
Prior to the trip, the DAS experienced a week of heavy rainfall. Such environmental variability has been shown to influence methane and carbon dioxide measurements. Romanak returned to the site ten days later, during which time the soils had dried. When soil-gas measurements were repeated, soil-gas concentrations had changed. This initial study highlights the environmental complexity involved with developing an accurate environmental monitoring program.
In order to evaluate the chemical, physical, and biological processes at the site in more detail, Romanak and Yang plan to install more permanent soil-gas monitoring stations and sensors to monitor environmental factors such as barometric pressure, rainfall, water flux, and subsurface pressure and temperature. They will also measure the isotopic composition of gases in order to trace their diverse sources. By unraveling the pathways of gas evolution they hope to develop a framework for environmental monitoring at industrial sites. Some of the questions Romanak and Yang hope to address include:
What is the possible contribution of methane from bacteria in the soil, in the mud pit, and from the well, if any?
What is the possible contribution of carbon dioxide from the soil, the mud pit, the surrounding plants, and the well, if any?
How do environmental impacts, such as rain, affect the concentrations of carbon dioxide and methane?
How can we identify and monitor impacts from industrial factors, such as pits, P&A wells, and pads?
How can we engineer sites that sufficiently reduce environmental complexities to allow accurate near-surface monitoring of carbon sequestration?
A preliminary conceptual model of the study site highlights some of the many pathways in which carbon dioxide and methane influence environmental monitoring.
August 19, 2009: Completion of the Second Observation Well F-2 We completed logging of the middle observation well F-2 today. The big questions are: How will the CO2 move through this rock? Will it occupy a lot of the pores or will it look for a “shortcut”, and just move through the shortcut? If the CO2 occupies many available pores, this is known as “good sweep”. In the F-2 well, we collected dense data to characterize the rock volume to improve our prediction and eventual understanding of this process. In addition to conventional porosity logs, we are trying the Combinable Magnetic Resonance” CMR tool that shows where moveable water occurs in the rock.
The photo below shows the view from inside the wireline rig truck. This work requires that we spend lot of time looking out this window – it takes at least a 45 minute “trip” time to lower a tool on wireline cable to the bottom of the well more than 10,700 feet below the surface. Think of it as a commute to the jobsite, but downward more than two miles into the ground.
View from inside the wireline truck.
The photo below is shows our first image of the properties of the rocks in the well. The bright areas show the locations of pore volumes that the first analysis suggests CO2 can access. In the detail view of the image on the screen, we indicated with arrows the possible path that the carbon dioxide will access most quickly after injection. We have a lot more work to do before we decide if CO2 will use this zone.
A first look at geophysical properties of the well.
We conclude this entry with a poetic look at our new powerline, which is less beautiful by day, but provides us with the ability to operate our monitoring equipment.
A nighttime view of the drill rig.
August 18, 2009: Slabbing the Upper Core From the Second Observation Well F-2 The upper core from observation well F-2 arrived at GCCC yesterday morning and was slabbed this morning at the Bureau of Economic Geology facility in Austin. This core is from the layer that will serve as the ultimate seal to the carbon dioxide reservoir. When core was extruded from the tubes in which it was shipped, it was still coated in the thick drilling mud.
A section of the upper core from observation well F-2 is extruded from its casing.
The shale, which is thick and hard to cut underground, becomes weak when stress is released and water content changes and can easily break especially across the core. BEG staff had to handle the core extremely carefully to maintain its structure. In order to successfully slab the delicate core, it was wrapped foil for stability.
The core was wrapped in foil to stabilize it and sawed into two sections.
When the core is cut, geologists can examine many features of the rock. This core is from a geologic layer called the marine member of Tuscaloosa Formation, which is a dark mudstone and shale. The smaller section of core is retained and archived at the Bureau of Economic Geology. The larger section of core, shown at right, will be used for physical and chemical analysis by GCCC and many collaborators. It is rare to cut a shale core, but we expect it to provide information about how this common rock type can assure that CO2 storage is permanent.
The dark shale of the core is exposed after it is slabbed.
August 17, 2009: The Second Core From the Second Observation Well F-2
Over the weekend, rock from within the formation that will serve as the carbon dioxide reservoir was cored. GCCC planned to retrieve 120 feet of rock from more than 10,500 feet below the surface. This includes the top seal, the reservoir, and the bottom seal. Because the rig can pick up 90 feet of pipe at a time, we planned to core in two 60-foot sections.
In the upper section of the core 56 feet of rock were successfully retrieved. This section contains the seal immediately above the reservoir as well as a portion of the reservoir.
During drilling of the lower section of core, the core barrel became jammed about half way through the core. Because of cost considerations, the GCCC decided not to make a fourth core trip into this well. Denbury has cored the bottom-seal in other wells in the field. These data can be used in combination with wireline log information to provide rock properties
Besides cost, one of the factors that has to be considered with coring is the stability of the well. Some of the geologic layers the well cuts through contain rock that are weak and can crumble or swell and slump into the well. Well drillers constantly circulate a thick liquid called drilling mud to temporarily stabilize the walls, however it important to move as quickly as possible to permanently stabilize the well by installing and cementing in the well casing.
After coring was completed, the drillers removed the core bit from the well and replaced it with a drill bit, which will be used to finish drilling the well. Drilling to total depth (TD) is expected later today.
A drill bit used to drill the observation wells.
August 10, 2009:Survey of Sites for Environmental Monitoring The Federal Safe Drinking Water Act of 1974 requires that before any fluid (gas or liquid) is injected underground, proof be provided that the injectate will be retained in the injection zone, and that no harm can come to underground sources of drinking water. Because the Tuscaloosa Reservoir at Cranfield has held oil and natural gas for 100’s of millions years, we do not expect that there is any risk of CO2 leakage. Most of our work is within the reservoir to confirm that CO2 movement and eventual stabilization is occurring as expected and therefore there is no possibility for injected CO2 to reach the atmosphere. While we do not expect the CO2 to leak from the reservoir, part of our research is to determine methods to confirm where the CO2 is located in the reservoir and that CO2 will not leak from the reservoir and reach the atmosphere.
How can we make sure that there is no risk to water or to surface ecosystems from CO2 injected beneath? Environmental monitoring is one tool that can be used to provide this information. We plant to look in detail at natural variations in the ecosystem, and then consider how we could find changes that might indicate leakage and quickly and confidently separate real leaks from other kinds of changes.
One of the features that may increase the risk of leakage are wells that were drilled into the injection zone. When wells are no longer used, they are supposed to be properly plugged with cement and heavy mud, then capped and cut off below the surface before they are abandoned. In some other fields, injection was started, and then the operator has found out that some wells were forgotten or improperly plugged. Our team is assessing some of the plugged and abandoned (P&A) wells at Cranfield to determine if they are performing correctly as the injection starts.
One way to do this is though soil gas measuments. Using these measurements we determine if there are any changes or differences in the area where P&A wells are found.
GCCC Researcher Katherine Romanak sets up to perform surface monitoring of soil gas.
August 9, 2009: The First Core from the Second Observation Well
Last night, a 30-foot core was retrieved from observation well F-2 at Cranfield field. GCCC Project Manager Ramon Trevino and GCCC Post-Doctoral Fellow, Jiemin Lu oversaw removal of the core from about 200 feet above the depth that the core retrieved from F-3 in June was located. This layer will act as an ultimate seal to the carbon reservoir.
In order to drill a well to such a depth rigs similar to those used by oil and gas companies are erected on the site. A very heavy fluid called drilling mud is pumped into the well and constantly circulated in order to properly stabilize the well after it is drilled. A common problem with the mud is that it can cake up and stop circulating properly. When this occurs rig operators perform a “wiper trip” in order to recondition the mud in the well. Because the well extends roughly two miles below the surface, each wiper trip takes several hours.
The drill rig in position at F-2. Fiberglass casing is prepared to be inserted and cemented inside the well once the total depth (TD) is reached.
After the core was retrieved, a subcontractor, Corelabs, sawed the 60-foot core into 3-foot sections for transport to the University of Texas. This core is composed of rock known as marine Tuscaloosa. This marine Tuscaloosa is a black shale. Coring through this layer of rock was a slow process: coring a single foot required about an hour.
Corelabs saws the 30-foot core exposing the dark marine Tuscaloosa shale. The core was divided into three-foot sections for transport to the GCCC.
July 24, 2009: Slabbing the Core from the First Observation Well
GCCC Post Doc Jiemin Lu oversaw the slabbing of the core from observation well F-3 at the GCCC laboratory in Austin. His first step was to cut off eight six-inch long intervals, which were preserved by wax coating for future whole core analysis. He extruded the remaining cores out of the aluminum tubes in which they were shipped.
The core is from a geologic layer called Tuscaloosa Formation. It is composed of D and E sandstone and conglomerates units interbedded with mudstone layers. At the top of the D sand unit, there is about ten feet of red mudstone and siltstone. This interval will act as the immediate seal to prevent escape of the carbon dioxide from the reservoir.
The core shows that the lower Tuscaloosa Formation was deposited by gravelly rivers. The character of these rivers changes over short distances from gravel, to sand, to mud. When these various parts of the river are turned to rock, different lithologies are preserved. When the CO2 moves through these heterogeneous rocks, it will flow more quickly than some zones than in others.
GCCC will continue to analyze the slabs from the core to understand physical features of the rock including its permeability, strength, and porosity. In addition the rates of flow of carbon dioxide through the rock as well as its chemical composition will be analyzed. One of the key factors that GCCC will study is the amount of variability in both the physical and chemical structure of the core.
Slabs from the core retrieved from the first observation well consist of a reddish mudstone interbedded with sandstone. Some conglomerate is also visible in the core (click images to enlarge photo).
June 16, 2009: Coring the First Observation Well
Observation well F-3 was drilled, cored, and logged. A 90-foot core was retrieved from a depth greater than 10,000 feet below the surface. It was shipped to the Gulf Coast Carbon Center’s laboratory on the University of Texas campus for analysis. Initial observations of the core revealed nothing unexpected.
A 90-foot core was retrieved from the first observation well.
June 6, 2009: Choosing the Location for the Core from the First Observation Well
The same type of rig used to drill oil and gas wells is used to drill the observation wells. The procedures for drilling observation wells are very similar to those used by oil and gas companies for drilling wells, except that the diameter of the well is somewhat larger. This additional area is required to fit scientific instrumentation into the well.
This morning, the location of the core that will be drilled from observation well F-3 was determined. The core will be used to determine the physical and chemical properties of the geological formations. This information is critical for developing accurate computer models of the subsurface carbon dioxide reservoir.
Three hundred and fifty 30-foot joints of well casing arrive at the well pad for completion of the F-3 observation well.
GCCC seeks to apply its technical and educational resources to implement
geologic storage of anthropogenic carbon dioxide on an aggressive
time scale with a focus in a region where large-scale reduction
of atmospheric releases is needed and short term action is possible.