Recent Submissions

  • Dynamic Modeling of Buoyant Fluids: Implications for Hydrocarbon Distribution and Potential Carbon Capture and Storage (CCS)

    OConnor, Darragh; Richards, Bill; Angel, Max; Wach, Grant (2020-1-27) [Poster]
    SUMMARY: Many intergovernmental organizations, international panels, and global scientific institutions have come to an objective, scientific based understanding that anthropogenic release of CO2 to the atmosphere is one of the major contributors to human-induced climate change. The 2005 IPCC report on CCS identified the Scotian Margin as one of few world class locations for storage of CO2 in deep saline aquifers. Here we present a series of static and dynamic fluid flow models to illustrate the trapping mechanisms, or lack thereof, of hydrocarbons in the Sable Subbasin (Scotian Margin). Following this we present our evaluation of carbon capture and storage (CCS) potential in regional aquifers in the subsurface Sable Subbasin region. We conclude that CCS in depleted gas fields carries the least risk of leakage but has limited potential due to the small size and low relief of the structures. In contrast, CCS in regional aquifers offers huge storage potential but there are serious concerns regarding leakage through their updip subcrop near the seabed.   DATA & METHODS: Data: Data used in this project included pressure, geochemistry, temperature, maturity, lithostrat, and biostrat basin data collected from the GSC “BASIN” database. Formation structure maps, well logs, and seismic cross sections were collected from the GSC 2011 East Coast Basin Atlas Series. Seismic data, both 2D and 3D, as well as associated maps and reports were collected from the Canadian Nova Scotia Offshore Petroleum Board and their Data Management Center website. This included the Penobscot 3D seismic survey and wells L-30 and B-41. Offshore well data for wells Migrant N-20 and South Venture O-59 are part of a larger dataset purchased from Divestco.Methods: The above data were combined in order to build representative 3D geocellular models of structural closures of reservoirs and regional aquifers (Missisauga Fm.) in the offshore Scotian Margin. These models were constructed in Petrel 2018. Petrophysical properties were calculated at the wells (porosity and permeability) and were propagated through the models using the nearest point algorithms. Gas injection wells were inserted into the models. Using ECLIPSE simulation software, gas was injected into the base of each well for 50 or 100 years and then injection was stopped. Following this, each model was allowed to equilibrate for thousands of years. A detailed overview of the methods is too extensive for this presentation. Please contact the author for additional details.
  • Detecting Wellbore Shape and Cavities, and Cement Imperfections in Fiberglass Cased Wells

    Eltsov, Timofey; W. Patzek, Tadeusz (2020-1-27) [Poster]
    •One of the main cost of geothermal energy production is the electrical power consumption of the submersible pumps •Pumps are forced to overcome losses in a closed environment •Corrosion and mineral precipitation is often observed in geothermal wells •Bad cement job can result in wellbore breakdown and power plant shutdown Iron powder and fiberglass composite pipes (   •Electrically resistive composite casing materials are being introduced to the industry •Composite tubulars are 4 times lighter than made of steel, withstand same pressure and load, resistant to corrosion and lasts more than twice longer •The use of EM transparent casing materials requires the development of appropriate geophysical methods •Presence of conductive particles in the cement allows to increase its displacement and check its quality using EM •We use magnetic sensors to verify the quality of the cement behind the casing   Schematic representation of an idealized 4 layered geoelectric model Helmholtz equation in cylindrical coordinates (r, j, z):  w, circular frequency; m= mrm0, magnetic permeability (r - relative, 0 - permeability of vacuum) and s is electrical conductivity Tangential components of magnetic (H) and electrical (E) fields at the boundaries are continuous: Vector potential in the wellbore: M – magnetic moment of the coil, I0 – modified Bessel function, l – separation variable, z – axial distance, r – radial distance, C1 – is a constant obtained from boundary conditions that carry information about the borehole environment. Modeling parameters: Effect of magnetic permeability of cement on the magnetic field at frequency = 1 kHz. Left to right: real part of secondary field, Hz,s, imaginary part of Hz,s We assume that the tool is properly calibrated when the primary magnetic field has been excluded from the measurements. Logging signals at the frequency = 1 kHz cross a cement lift, different tool lengths are considered Logging signals at the frequency = 1 kHz cross a crack filled with cement, different tool lengths are considered The wellbore model with different resistivities of the cement and logging signals at the frequency of 200 MHz for real and imaginary part of Hz,s The eccentricity of cement can be determined using a magnetic sensor tool Conclusion Tools with lengths 0.25 – 0.6 m are most sensitive to the magnetic properties of cement The real part of secondary magnetic field, Hz,s, is more sensitive to magnetic cements than the imaginary part The most suitable frequencies are in the range 0.1 to 10 kHz, typical of magnetic susceptibility logging, inhomogeneities filled with cement are visible on the logs 200 MHz induction tool can be used for cement solidification detection The proposed method can determine poor-quality cementation through a non-conductive casing for low enthalpy geothermal applications
  • NubiaResource: A multidisciplinary investigation for implementing CPG in the Gulf of Suez (Egypt)

    Hefny, Mahmoud; Ebigbo, Anozie; Qin, Chao-Zhong; Adam, Benjamin; O. Saar, Martin (2020-1-27) [Poster]
    Introduction The injection of CO2 into the highly permeable Nubian Sandstone of a depleted oilfield in the central Gulf of Suez basin (Egypt) can be an effective way to extract geothermal energy from deep sedimentary basins while geologically sequestering a substantial amount of CO2 (main green-house gas driver), forming a so-called CO2-Plume Geothermal (CPG) system (Adams et al., 2015, Randolph and Saar (2011) and Garapati et al. (2015)). A CO2 thermosiphon makes CO2 a favorable subsurface working fluid as will be shown next.
  • Reducing Future Carbon Emissions: A Scoping Analysis of Carbon Capture and Storage (CCS) Costs and Funding

    Angel, Max (2020-1-27) [Poster]
    Summary This presentation outlines the costs and revenue associated with a CCS pilot project offshore Nova Scotia. This pilot project consists of two wells each injecting 1.25Mt/yr CO2 into the subsurface Misissauga Formation. Costs include capital expenditures ($2.2B with +$1B sensitivity) and annual operation expenditures ($40/tonne to $60/tonne CO2). Revenue is estimated using the sale of carbon credits ($50/tonne) and a surcharge applied to the sale of refined petroleum products ($0.05 to $0.20/L) in Nova Scotia. This project identifies commercial potential and regulatory framework for future CCS.
  • A Systematic Workflow for Geothermal Exploration in Carbonates: Lessons from the Upper Jurassic of the Molasse Basin/Germany

    Wolpert, Philipp; Beichel, Killian; Bendias, Daniel (2020-1-27) [Poster]
    WORKFLOW: (1) The core description reveals 22 facies types and 4 facies associations. (2) Not all core-based depositional facies can be differentiated unambiguously from the BHI. A rock catalog helps to capture diagnostic criteria of facies from core and  to link it with the BHI. Based on this, genetically related depositional sequences can be delineated. The vertical stacking pattern of image facies and marker beds allows ,therefore establishing a robust sequence stratigraphic framework. (3) Borehole image facies (facies associations) and sequences are integrated into the seismo-stratigraphic interpretation. The seismic interpretation reveals the mayor depositional elements and reflects the schematic cross-section shown in (1). The borehole image facies and facies associations are used to validate the seismic facies interpretation.  (4) Internal heterogeneities (like baffles, compartmentalization, karst) within the biohermal/reef reservoirs, which are below seismic resolution, can be recognized on the BHI and are key for the geothermal exploration strategy.  (5) Seven geothermal reservoir types are identified and linked with dynamic data. This highlights the most important flow zones, explains their origin and predictability in a sequence stratigraphic context.   CONCLUSION: •THE UPPER JURASSIC GEOTHERMAL SYSTEM IS A COMBINATION OF STRATIGRAPHIC AND STRUCTURAL TRAPS. •A SEQUENCE STRATIGRAPHIC FRAMEWORK IS PARAMOUNT TO UNDERSTAND GEOTHERMAL SWEET SPOTS. •INTEGRATION OF DYNAMIC DATA AND RANKING OF RESERVOIR DRIVERS HELPS TO UNDERSTAND FLUID FLOW BEHAVIOR. •THE CAPACITY OF THE PRODUCER WELL IS ONLY AS GOOD AS THE CAPACITY OF INJECTOR WELL (OFTEN THE BOTTLENECK). •IMPORTANT TO AVOID INTER-WELL CONNECTIVITY VIA HIGH-K STREAKS TO PREVENT EARLY THERMAL BREAKTHROUGH.
  • Temperature - Dependent Properties of Intact Rocks

    Guerrero, Camilo (2020-1-27) [Poster]
    Temperature – Dependent Properties of Intact Rocks   INTRODUCTION Geothermal energy can be a cost effective and sustainable   Rock behavior impacts the production efficiency of geothermal reservoirs → geothermal characterization is critical.   OBJECTIVE To propose physics-inspired and asymptotically correct models for the specific heat capacity, thermal conductivity and thermal diffusivity in intact rocks.   METHODOLOGY We compiled thermal property data from over 20 published papers for sedimentary, metamorphic and igneous rocks. This research presents data where the:   Confining pressures were negligible. Maximum temperatures were below the phase change transition.   RESULTS Specific heat capacity Thermal conductivity Thermal diffusivity   CONCLUSIONS AND FUTURE WORK External temperature increases require a higher internal energy gradient to produce a temperature change in intact rocks   The hyperbolic model shows a clear correlation between temperature and specific heat capacity   Results indicate that the inverse of the specific heat capacity when divided by temperature, collapses to approximately the same value for all types of rocks at absolute zero temperature. Ongoing research: Future research will investigate potential correlations between temperature, thermal conductivity and thermal diffusivity of intact rocks.   ACKNOWLEDGEMENT This research is a continuation of the database started by Professor Lucio Cruz (Universidad del Cauca, Colombia).
  • 1D Basin Modeling of Western Arabian Sedimentary Basins using Satellite Gravity Data for low-to-medium Enthalpy Geothermal Systems: Initial Case Study Midyan Area

    Arifianto, Indra (2020-1-27) [Poster]
    1D Basin Modeling of Western Arabian Sedimentary Basins using Satellite Gravity Data for low-to-medium Enthalpy Geothermal Systems: Initial Case Study Midyan Area Background Energy Issue and Diversification in KSA NEOM City and Renewable energy Water resource and consumption in KSA Direct-use of geothermal (low to medium) Objectives Determine and potential of geothermal low to medium enthalpy in the Red-sea Tertiary Rift Basin and the location of the well based on subsurface geological condition Determine the Geothermal Gradient of the basin based on Basin Modeling from Gravity Data dan Regional Heatflow Map. Why Midyan area is interesting : •Neom City Project •Exploration targeted area •Some gas wells •Some drilled exploration wells •Some publications on seismic interpretation and well information   Based on the literature review, gravity analysis, and gravity modeling, the Ifal basin has thick tertiary sedimentary rocks and the basin experiencing rifting from Oligocene until Late Miocene. Therefore it has high present heat flow, moreover, the Ifal basin has a thick evaporite salt diapir that has high specific heat capacity. Overall, the ifal basin has good potential of low to medium enthalpy, however, the best location is in the basin depocenter that has thick of salt deposit. The geothermal gradient of the Ifal basin has ranged between ~32.5oC/km in the basin edge and up to ~35oC/km in the basin depocenter. In the basement high area, the temperature is ~100oC at 2300 meter and in the basin depocenter reach ~290oC at 7200 meters. We will apply this kind of method to evaluate to other sedimentary basins in the west of the Arabian Peninsula, and since It is difficult to determine the residual anomaly depth that corresponds to salt diapir without seismic data. we need a geophysical data that can detect the evaporite deposits such as electric data or seismic data. We will do some simulation of heat transfer efficiency of some heat mining technique (single, doublet or triplet system) on particular subsurface condition.
  • Physics-based modeling of large triggered seismic event in an Enhanced Geothermal System: the 2017 Mw 5.5 Pohang Earthquake

    Hendrawan Palgunadi, Kadek (2020-1-27) [Poster]
    Physics-based modeling large triggered seismic event in an Enhanced Geothermal System:the 2017 Mw 5.5 Pohang Earthquake   The Mw 5.5 Pohang earthquake was recently confirmed (KoreanGovernment Commission, 2019) to have occurred due to theinjection activity of an Enhanced Geothermal System (EGS),making this the largest induced event associated with EGS.Based on moment tensor solutions (Grigoli et al., 2018), thePohang earthquake features both strike-slip and thrust faulting aswell as a pronounced non-double couple component. Grigoli et al.(2018) proposed the non-double couple solution as a combinationof two possible fault planes.We investigate the Pohang earthquake using high-resolution 3Ddynamic rupture simulations which coupled seismic wavepropagation and frictional shear failure across predefined complexfault system. A local stress orientation constrained by wellbreakouts leads to mixed strike-slip/thrust faulting for both thesingle and multiple fault geometries with moment release close tothe observations. This dynamic rupture simulation shows spatiotemporal of rupture processes to understand better the Pohang(induced) earthquake.Objectives:How significant is the observed change of local maximum horizontal stress orientation withrespect to faulting mechanism and dynamic of the Mw 5.5 Pohang Earthquake?Can the reduction in effective normal stress due to hydraulic stimulation causes run-away rupture?What are the physical processes that lead to a pronounced compensated linear vector dipole(CLVD) moment tensor rupture of the Mw 5.5 Pohang Earthquake?How dynamic rupture simulation can contribute to the seismic hazard assessment in EGS?What can seismograms provide to the source characteristics from near-distance monitoring?PohangMap of Korea Peninsula associated with the closest regionalseismic stations. (Inset) Interested area with 3 hours aftershocks.Fault reconstruction. Looking at a) normal to thefault, b) back-normal to the fault, c) map view.Material properties and friction parameters. a) 1D velocity,b) density, Lamé’s parameters, d) friction parameters.We cluster earthquakes by modifying technique introduced by Wang et al. (2013). Guided anisotropic locationuncertainty distribution (g-ACLUD) accounts for the uncertainty of the hypocenters and the direction ofmaximum horizontal stress.We apply principal component analysis to fit the corresponding seismic cluster with a plane.g-ACLUD recognized two fault planes.We employ material properties analyzed by Woo et al. (2019).We use laboratory-based friction law (rate-and-state with strong velocity weakening).We solve the coupled dynamic rupture and wave propagation problem using open-source softwareSeisSol ( domain of 54km x 40km x 20 km is discretized into an unstructured mesh containing 4.2 milliontetrahedral elements. We incorporate topography at high resolution (200m edge length). The meshresolution is refined to 50m close to the fault plane.We account for off-fault plasticity and locally inferred fault strength.We examine 180 numerical experiments for fault rupture by testing permutation of Shmaxorientation, prestress, and fluid-pressure ratio.Outcome of numerical experiments:We prescribe stress ratio to be 0.8The fault is easier to rupture if SHmax = 120° with combination of lower prestress ratio and fluidpressure ratioWith hydrostatic fluid pressure, except for 120°, none of the permutations lead to have run-awayrupture.We create rupture nucleation by overstressing an area centered at the hypocenter.Numerical experimentsSlip-rate evolutionSlip distribution (a) and b)) and rake (c) and d))Spontaneous complex rupture propagation occurs at t < 0.8 sMultiple reflections between two fault planes generate at least twoupcoming rupture fronts at t = 0.65 sThe secondary fault plane ruptures due to dynamic triggering almostsimultaneously after earthquake nucleationComplex rupture provides complicated rake distribution.Average rupture speed = 2,250 m/sAverage slip = 32 cmResulting moment release = 2.76 x 1017Nm = Mw 5.59(b)Surface deformation (LoS displacement) for (a) observed (Choi et al., 2019)and (b) syntheticsPeak ground velocity (PGV) shakemap (GMRotD50)We generate two fault planes using g-ACLUD to constrainpossible fault segmentation based on high resolution seismicdata.Run-away rupture occurred due to overpressure fluid inPohang EGS site.Resolving the regional stress field on both fault geometriesleads to pure strike-slip faulting as only mechanical viablesolution. Local stress regime contributes to the thrust-faultingcomponent.Simultaneous rupture on two fault planes and complexruptures contribute to -5.05% CLVD component out of -37%CLVD from observed data.The remaining differences potentially come from the existenceof fracture/fault networks, tensile faulting, poroelasticrheology, and source or medium anisotropy.The physics-based rupture simulation is able to reproducesurface deformation but has larger displacement.Based on PGV map, the North-West and South-East regionrelative to the epicenter are prone to have higher shaking.Local high topography is responsible for the high PGV. Thisresult is in accordance with field survey.Dynamic rupture simulation provides first insight forseismic hazard assessment given fault geometry and localstress condition.Synthetics from two different scenarios (one and two faultplanes) produce significant differences in waveformsignatures for near-distance stations, especially forhorizontal components. Mostly located on the hanging wall.Near-distance seismic monitoring is essential for accurateseismic hazard assessment. Far-distance seismicmonitoring could not resolve the complexity of theearthquake source.Dummy stations for synthetic seismogramMoment tensor solution for a) one andb) two fault planesChoi, Jin‐Hyuck, et al. “Surface Deformations and Rupture Processes Associated with the 2017 Mw 5.4 Pohang, Korea, Earthquake.” Bulletin of the Seismological Society of America, vol. 109, no. 2, 2019, pp. 756–769., doi:10.1785/0120180167.Ellsworth, W., D. Giardini, J. Townend, S. Ge, and T. Shimamoto (2019) Triggering of the Pohang, Korea, Earthquake (Mw 5.5) by Enhance Geothermal System Stimulation. SRL Vol. 90 No. 5, doi: 10.1785/0220190102Grigoli, F., S. Cesca, A. P. Rinaldi, A. Manconi, J. A. López-Comino, J. F. Clinton, R. Westaway, C. Cauzzi, T. Dahm, and S. Wiemer (2018). The November 2017 Mw5.5 Pohang earthquake: A possible case of induced seismicity in South Korea, Science 360, no. 6392, 1003–1006, doi:10.1126/science.aat2010Korean Government Commission (2019). Summary report of the Korean Government Commission on relations between the 2017 Pohang Earthquake and EGS Project, Geological Society of Korea, Seoul, South Korea, doi: 10.22719/KETEP-20183010111860.Wang, Y., G. Ouillon, J. Woessner, D. Sornette, and S. Husen (2013). Automatic reconstruction of fault networks from seismicity catalogs including location uncertainty. JGR: Solid Earth. Vol 118, pp 5956-5975, doi: 10.1002/2013JB010164
  • Hydraulic Fracturing in Pre-fractured Media

    Rached, Rached; Garcia, Adrian; Carlos Santamarina, J. (2020-1-27) [Poster]
    Hydraulic Fracturing in Pre-fractured Media The orientation of fracture sets in relation to the in-situ stress field controls hydraulic fracturing in pre-fractured rocks. In many high-pressure injection operations, the reopening of preexisting fractures occurs instead of the initiation of new hydraulic fractures through the rock matrix. This study investigates the competing influences of state-of-stress and fracture set geometry on injected silicone sealant imprint morphologies.
  • Heat Flow in Fractured Rocks

    Garcia, Adrian; Carlos Santamarina, J. (2020-1-27) [Poster]
  • Staying Cool in A Warming Climate: Temperature, Electricity and Air Conditioning in Saudi Arabia

    Odnoletkova, Natalia; Patzek, Tadeusz; Howarth, Nicholas (2020-1-27) [Poster]
  • Proxy-model for Flow and Transport in Geothermal Reservoirs

    Tahir, Ahmad (2020-1-27) [Poster]
    INTRODUCTION: Geothermal cycle recurs until reservoir rock temperature drops consequently unable to thermally recharge the injected water sufficiently and maintain the produced water at initial reservoir temperature; as a result, cooler water breaks through into production well. Determination of this break-through time is crucial and performed using Reservoir Simulation. Accurate high-fidelity 3D models, despite being capable of predicting breakthrough, require significant time in uncertainty quantification and data assimilation analysis due to CPU demanding simulations. OBJECTIVE: Develop proxy model to predict flow and heat transport and consequently breakthrough time in low-enthalpy geothermal reservoirs using selected number of streamtubes, in order to minimize simulation time and CPU resources. This model is based on fact that streamlines, that describe flow in a system, are mostly controlled by steady-state pressure distribution. METHODOLOGY: • Full-scale simulation is run on a flow and heat transport solver for a set of parameters; the solver used is DARTS.• Once steady-state is achieved, streamlines are traced for a given pressure distribution.• Layer is divided into 2-D streamtubes (as illustrated by the centre image and arrows in figure 3).• Cell volume and transmissibility lists are generated for each of these regions.• Simulation is run on a flow and transport solver for each of these regions.• The pressure and temperature distributions results from each region are merged (figure 4b) based on weighted average of cell volume. The producer well temperatures from each region are averaged based on weighted volumetric flowrate produced in order to generate breakthrough curves for proxy model. RESULTS AND DISCUSSION: • The error is concentrated at the streamtube boundaries (as in figure 4c). While intra-region conduction still exists, this error occurs due to lack of inter-region conduction.• Maximum absolute error is high but the average error and breakthrough curve error is within tolerance.• As number of streamlines (consequently streamtubes) increase, the lack of inter-region conduction increases causing the error to amplify (as in figure 6).• All the regions simulated individually take significantly less time than the full-scale model (table 7a). If all the regions are simulated in parallel, the region with the highest simulation time will be the bottleneck (e.g. if 6 fastest regions are selected, it reduces simulation time down to 14 seconds). CONCLUSION AND FUTURE WORK: • Comparison reveals proxy model results are optimistic; error between p-T distributions (table 4d) and breakthrough curves (figure 5) is within acceptable tolerance.• Future work suggestions: 1. Find an optimal number of streamlines separation 2. Robustness at different thermal Peclet numbers 3. Energy transfer correction at region boundaries REFERENCES: 1 Khait, M., 2019. Delft Advanced Research Terra Simulator: General Purpose Reservoir Simulator with Operator-Based Linearization. PhD thesis. Delft University of Technology, The Netherlandsþ2 Shetty S., D. Voskov, and D. Bruhn, 2018. Numerical Strategy for Uncertainty Quantification in Low Enthalpy Geothermal Project, Proceedings 43rd Workshop on Geothermal Reservoir Engineering, Stanford University, Stanford, California, February 12-14, 2018, SGP-TR-213.þ3 Pollock, D.W. 1988. Semianalytical Computation of Path Lines for Finite-Difference Models, Groundwater, 26 (6), pp. 743-750. þCONTACT:þAhmad Mohammad TahirM.Sc. Petroleum EngineeringEmail: email protectedLinkedIn: ahmadmohammadtahirþDenis VoskovAssociate Professor (T.U. Delft)Email: email protectedLinkedIn: denis-voskov-823690a4þ