Our study analyzed empirical evidence for pressure communication with the Austin Chalk Formation during the stimulation of two new (2017 child) wells in the Eagle Ford shale, which caused the pressure to locally surge in both reservoirs. The conductive fracture network formed in the Austin Chalk is assumed to have transmitted some fluid pressure from the Eagle Ford to the annulus of five monitoring wells in the Austin Chalk. Pressure gauge responses in the Austin Chalk wells were measured during the fracture treatment in the Eagle Ford. The magnitude of pressure response in the Austin Chalk wells is only a fraction of the Eagle Ford injection pressures. The pressure stimulation of the two Eagle Ford wells occurred in alternating stages (zipper fracking).
The Riverside wells began responding to the Eagle Ford frack job through an initial increase in pressure as hydraulic fractures begin to propagate outwards and connect to the naturally fractured network of the Austin Chalk. The initial rise in pressure is followed by a plateau region, which is limited by the stresses induced in a zipper fracking operation that causes fractures to propagate towards each other, in a direction perpendicular to the wellbore.
Immediately after the fracture treatment in Well H3 was completed, a second pressure rise occurred, which persisted until after the operation in Well H2 was completed. The pressure response rapidly declined to its pre-treatment state, which confirms that pressure communication was temporary in nature. The observed time delay was a result of stress shadows closing the induced/connected fractures, after which frack fluid is no longer forced into the Austin Chalk.
Our conceptual model of pressure communication takes into account the pressure depletion of the Austin Chalk reservoir due to decades of production, prior to the fracture treatment in the nearby Eagle Ford wells. The depleted, average reservoir pressure near the Austin Chalk wells is estimated to be 2354 psi. Based on history matching of earlier Eagle Ford (2014 parent) wells, the pressure in the Eagle Ford landing zone immediately prior to the fracture treatment was 4891 psi.
The initial pressure in the annulus of the five Austin Chalk observation wells was approximately 20 psi immediately prior to the Eagle Ford fracture treatment. However, the pressure rose to 265.8 psi in Riverside 1, 378 psi in Riverside 4, and 63.3 psi in Riverside 6. Pressure response profiles of wells Riverside 2 and 5 show similar trends.
We summarize conclusions, based on our interpretations of the pressure response profiles in wells R1–R6, as follows:
(1) Pressure communication between the two well sets (Eagle Ford-Austin Chalk) is a temporary phenomenon, taking approximately up to three days (from the start of the zipper fracking of the H2–H3 Well pair) to establish, lasting approximately 11 to 12 days to reach a plateau, which is followed by a brief final screen out peak that drops off nearly instantaneously. In all the pressure response profiles considered, the pressure rise in the Austin Chalk wells rapidly declines back to pre-treatment annulus pressures.
(2) The magnitude of pressure rise in the Austin Chalk is significantly lower than Eagle Ford fracture treatment injection pressures (about 6%)
(3) Pressure communication is thought to occur due to pressurization of isolated fracture stages in the Eagle Ford wells, which temporarily increases the pressure differential between the Austin Chalk and Eagle Ford shale.
(4) Coeval production uplifts in the months following the fracking of the Eagle Ford wells were observed in the offset Austin Chalk wells, in addition to the Riverside wells themselves, and are associated with the natural fractured network of the Austin Chalk, that is further activated through the fracking of the Eagle Ford wells.
(5) Hydraulic fractures from the Eagle Ford open due to fluid injection during fracture treatment and are assumed to temporarily connect with the natural fracture system in the Austin Chalk reservoir. Poroelastic effects are not considered in our work due to the nature of the Austin Chalk.
(6) The pressure response in the annulus of the Austin Chalk wells is characterized by a time delay, because the pressure surges in Wells Riverside 1 and 4 ceased shortly after (35.2 h in Riverside 1 and 38.7 h in Riverside 4) the zipper fracking operation of Eagle Ford Wells H2 and H3 was completed.
Data curation, S.S.; Investigation, S.S.; Methodology, R.W., I.A. and S.N.; Supervision, R.W.; Writing—review & editing, R.W. and S.S.
This project was sponsored by startup funds of the senior author (R.W.) from the Texas A&M Engineering Experiment Station (TEES).
We would like to thank Hawkwood Energy, Exponent Energy and the Texas A&M University System for providing access to the RELLIS lease wells and data sets, which facilitated our study. Tyler Moehlman is thanked for help as an undergraduate student collecting pressure data in the field.
Conflicts of Interest
The authors declare no conflict of interest.
Appendix A. Pressure Depletion Calculation in Austin Chalk
When the pressure in the Austin Chalk is below its bubble point pressure, the oil formation volume factor is given as a function of pressure as shown in Equation (A1). The formation volume factor B0 is an important parameter in calculating pressure depletion. Equation (A2) shows the calculation of pressure depletion from initial reservoir pressures based on production data. Equation (A2) is derived from material balance .
The pressure in the reservoir can be iteratively calculated using monthly production data by solving Equation (A3) which is obtained from substituting Equation (A1) into Equation (A2) and solving for pressure P. Gas production is not a part of Equation (A2) (and hence Equation (A3)) since there is no free gas initially in the reservoir:
When the reservoir pressure is above the bubble point, the expression for formation volume factor changes, requiring the use of Equation (A4), which calculates the formation volume factor in an undersaturated reservoir. As in the below bubble point case, we substitute Equation (A2) to obtain the expression for pressure depletion, shown in Equation (A5), which needs to be solved using iterative techniques. The bisection method, which gives results to any desired level of precision (we use 0.01%)  was used to solve Equation (A5):
The methodology adopted was to first calculate pressure depletion using Equation (A5) from monthly production data, replacing the initial reservoir pressure term Pri with the calculated pressure P after every interval. When pressure falls below the bubble point, Equation (A3) would need to be used instead. Using monthly production data (from Texas RRC online), we can calculate the resulting reservoir pressure at the time of the fracture treatment in Wells H2 and H3. All the parameters required for the above calculations (with nomenclature) are shown in Table A1.
The original reservoir pressure is first calculated from pressure gradient of 0.45 psi/ft and the average true vertical depth of the six Riverside wells. Petrophysical parameters are needed to perform original oil and water in place calculation. Although there is understandably a large uncertainty in values, we consider a scenario, in which the water saturation is the highest and therefore the percentage of water produced (to that of oil produced) is an arbitrarily chosen high value. The bubble point pressure and composition of the reservoir fluids present in the Eagle Ford are the same as that for the Austin Chalk vertically above, which is reasonable since the Austin Chalk and Eagle Ford shale likely form a single hydrocarbon production system .
Table A1. Variables used for pressure depletion model for Austin Chalk.
The pressure depletion curves obtained are shown in Figure 4a Keeping all other variables constant, we perform a sensitivity analysis for the drainage area and present the results in Figure 4b, which shows that the effect of depletion is stronger for small drainage areas. For small drainage areas, Pri falls below the bubble point pressure, requiring use of Equation (A3) (see Section 2.2.1 for Figure 4a,b). The initial reservoir pressures plotted refer to the average of the reservoir pressures calculated for the duration of the fracture treatment.
When the drainage area becomes infinitely large, the pressure depletion effect becomes negligibly small. Likewise, the reservoir pressure goes to zero for small drainage areas. The relationship is logarithmic in nature, with the correlations shown on the plot. Given that the RELLIS field spans over an area of 2000 acres , we will use the corresponding reservoir pressure from Figure 4a (2354 psi) in building our pressure response model.
Appendix B. Pumping Schedule of Well H1 in Fracture Treatment of 2014
This appendix presents pumping schedule used for hydraulic fracture treatment in Well H1, which is representative for all present wells (Table 1) of the RELLIS area. Common terminology used in stimulation reports is also shown. The pumping process of 2014 fracture treatment of Well H1 was also similar to that used in 2017 fracture treatment of Wells H2 and H3. Wells were acidized, padded, and circulated with increasing levels of proppant concentration and decreasing proppant size as the stage progressed. Figure A1 shows the treatment graph for Well H1 Stage 7 for comparison with pumping schedules of Wells H2 and H3, in Figure 7. (shown in Section 3.1). The most common terms used in a post-stimulation report are summarized in Table A2.
Figure A1. Well H1 Stage 7. The same properties apply as in Wells H2 and H3 (Figure 7) except there is much more variation in Well H1, with regards to the proppants used, both in terms of grain and mesh size, which explains the high increase in proppant concentration (red line). In H2 and H3, 100 mesh is mostly used, with 40/70 being employed for select stages, so the corresponding graph would have a lower gradient or be flatter. Type of proppant and gels used are labeled.
Table A2. Common fracturing terminology.
Appendix C. Pressure Response Profiles for Riverside 2 and Riverside 5
This appendix presents the pressure response profiles of Wells R2 and R5 in the same fashion as the profiles for Wells R1, R4 and R6 (shown in Section 3.2) Not enough reliable data was collected for Wells R2 and R5 to produce a complete correlation, but results are still included since valuable insights can be gained from even the limited data available.
The data collected for Riverside 2 is very limited (as discussed in Section 2.5 and shown in Figure 6). As a result, the pressure response profile shown in Figure A2 consists mostly of interpolations (indicated by thin black lines). The data is sparse due to equipment failure, however Figure A2 still shows that the gauges did detect an increased pressure in R2 during the treatment, even if for a short time. The data for Riverside 2 was unavailable after 11/27/2017, so B2 does not necessarily represent a maximum pressure. This would be since the gauge ran out of battery at that time or was otherwise unable to continue taking accurate readings.
Table A3 discusses the estimated point of increase and the approximate timings of events A2 and B2. The important observation from Well R2 response lies not in the correlation trend, but rather in the magnitude. Even for the minimal data connected, the figure shows the data in region B2 has magnitudes in regions of 3000 to 5000 psi, which is significantly more than Wells R1 and R4. Interestingly, there was no pressure response detected for the duration of the DFIT Test, even though pressure data was collected during this time (11/02/2017 to 11/03/2017).
Figure A2. Correlated plots for Riverside 2. The left vertical axis is the pressure in Wells H2 and H3. The right axis is the pressure in the annulus of the Riverside wellhead. The significance of each labeled box is discussed in Table 1. The lighter black lines in the plot are interpolations that connect between the missing data.
Table A3. Observations and interpretations for Riverside 2.
The response profile for Riverside 5 presents almost no viable continuous data set, as shown in Figure A3. While it is easy to assume this was caused by faulty gauges, the fact remains that there still does exist a pressure communication signal on the time scale of major events. Data is unavailable for Riverside 5 after 11/22/2017 05:21 AM. This well does not show any of the trends as the other Riverside wells studied. This would be since the gauge ran out of battery at that time or were otherwise unable to continue taking accurate readings.
Figure A3. Correlated plots for Riverside 5. The left vertical axis is the pressure in Wells H2 and H3. The right axis is the pressure in the annulus of the Riverside wellhead. The significance of each labeled box is discussed in Table A3.
Table A4. Observations and interpretations for Riverside 5.
The pressure gauges installed could read a maximum value of 2000 psi, which means the responses recorded for this well could exceed this value, since the highlighted regions indicate that the gauges were recording values beyond its maximum, which is what killed the gauge to begin with. As identified in Table A4, Riverside 5 shows the strongest response to the DFIT test and fracture treatment of all the wells studied, which is identified even in the case of faulty equipment and an incomplete data set.
Appendix D. Austin Chalk Pressure Response Calculations
Table A5 describes in detail the pressure response calculations made for each Riverside observation well. Magnitude, duration and rate of pressure response in each of the Austin Chalk wells were determined from observations of the pressure response profiles of each well (shown in Figure 9, Figure 10 and Figure 11, Figure A2 and Figure A3 respectively)
Table A5. Rate, duration and intensity calculations for communications measured from available wellhead data. Timings are based on accurate date values.
1. Rodgers, L. Subsurface Trespass by Hydraulic Fracturing: Escaping Coastal v. Garza’s Disparate Jurisprudence through Equitable Compromise. Tex. Tech Law Rev. Online Ed. 99. 2013, 45. Available online: http://texastechlawreview.org/volume-45/ (accessed on 2 November 2018).
2. Khanal, A.; Weijermars, R. Pressure depletion and drained rock volume near hydraulically fractured parent and child wells. J. Pet. Sci. Eng. 2019, 172, 607–626.
3. Weijermars, R.; Alves, I.N. High-resolution visualization of flow velocities near frac-tips and flow interference of multi-fracked Eagle Ford wells, Brazos County, Texas. J. Pet. Sci. Eng. 2018, 165, 946–961.
4. Hu, Y.; Weijermars, R.; Zuo, L.; Yu, W. Benchmarking EUR estimates for hydraulically fractured wells with and without fracture hits using various DCA methods. J. Pet. Sci. Eng. 2018, 162, 617–632.
5. McGowen, H.E.; Krauhs, J. Development and Application of an Integrated Petroleum Engineering and Geologic Information System in the Giddings Austin Chalk Field. Soc. Pet. Eng. 1992.
6. Parshall, J. Can Austin Chalk Expansion Lead to Revival? Soc. Pet. Eng. 2017.
7. Hester, C.T.; Walker, J.W.; Sawyer, G.H. Oil Recovery by Imbibition Water Flooding in the Austin and Buda Formations. Soc. Pet. Eng. 1965.
8. Dake, L.P. Fundamentals of Reservoir Engineering, 8th ed.; Elsevier: Amsterdam, The Netherlands, 2001.
9. Weijermars, R.; Burnett, D.; Claridge, D.; Noynaert, S.; Pate, M.; Westphal, D.; Yu, W.; Zuo, L. Redeveloping depleted hydrocarbon wells in an enhanced geothermal system (EGS) for a university Campus: Progress report of a real-asset-based feasibility study. Energy Strategy Rev. 2018, 21, 191–203.
10. Rafiee, M.; Soliman, M.Y.; Pirayesh, E. Hydraulic Fracturing Design and Optimization: A Modification to Zipper Frac. Soc. Pet. Eng. 2012.
11. Vermylen, J.; Zoback, M.D. Hydraulic Fracturing, Microseismic Magnitudes, and Stress Evolution in the Barnett shale, Texas, USA. Soc. Pet. Eng. 2011.
12. Craig, D.P.; Barree, R.D.; Warpinski, N.R.; Blasingame, T.A. Fracture Closure Stress: Reexamining Field and Laboratory Experiments of Fracture Closure Using Modern Interpretation Methodologies. Soc. Pet. Eng. 2017.
13. Economides, M.J.; Hill, A.D.; Ehlig-Economides, C.; Zhu, D. Petroleum Production Systems, 2nd ed.; Prentice Hall: Upper Saddle River, NJ, USA, 2013; Available online: http://ezproxy.library.tamu.edu/login?url=http://search.ebscohost.com/login.aspx?direct=true&db=cat03318a&AN=tamug.4023753&site=eds-live (accessed on 2 November 2018).
14. Daneshy, A. Effect of Dynamic Active Fracture Interaction DAFI on Activation of Natural Fractures in Horizontal Wells. Soc. Pet. Eng. 2016.
15. Yu, W.; Wu, K.; Zuo, L.; Tan, X.; Weijermars, R. Physical Models for Inter-Well Interference in shale Reservoirs: Relative Impacts of Fracture Hits and Matrix Permeability. In Proceedings of the 4th Unconventional Resources Technology Conference, San Antonio, TX, USA, 1–3 August 2016.
16. Valko, P.P. Hydraulic Fracturing Short Course: Fracture Design, Fracture Dimensions, Fracture Modeling. 2005. Available online: http://slideplayer.com/slide/4948692/ (accessed on 10 November 2018).
17. Martin, R.; Baihly, J.D.; Malpani, R.; Lindsay, G.J.; Atwood, W.K. Understanding Production from Eagle Ford-Austin Chalk System. Soc. Pet. Eng. 2011.
18. Seth, P.; Manchanda, R.; Kumar, A.; Sharma, M. Estimating Hydraulic Fracture Geometry by Analyzing the Pressure Interference Between Fractured Horizontal Wells. Spe Annu. Tech. Conf. Exhib. 2018.
19. Seth, P.; Manchanda, R.; Zheng, S.; Gala, D.; Sharma, M. Poroelastic Pressure Transient Analysis: A New Method for Interpretation of Pressure Communication Between Wells During Hydraulic Fracturing. Soc. Pet. Eng. 2019.
20. Chapra, S.C.; Canale, R.P. Numerical Methods for Engineers; McGraw-Hill: New York, NY, USA, 2015.
Correspondence: [email protected]; Tel.: +1-979-845-4067
© 2019 by the authors. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).