Downloaded from http://onepetro.org/spesgpc/proceedings-pdf/08SGPC/All-08SGPC/SPE-119890-MS/2756107/spe-119890-ms.pdf/1 by Robert Durham on 12 August 2022
`
`SPE 119890
`
`What is Stimulated Reservoir Volume (SRV)?
`M.J. Mayerhofer, E.P. Lolon, N.R. Warpinski, C.L. Cipolla, and D. Walser, Pinnacle Technologies,
`and C.M. Rightmire, Forrest A. Garb and Associates
`
`
`
`Copyright 2008, Society of Petroleum Engineers
`
`This paper was prepared for presentation at the 2008 SPE Shale Gas Production Conference held in Fort Worth, Texas, U.S.A., 16–18 November 2008.
`
`This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been
`reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its
`officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission to
`reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.
`
`
`Abstract
`Ultra-low permeability shale reservoirs require a large fracture network to maximize well performance. Microseismic
`fracture mapping has shown that large fracture networks can be generated in many shale reservoirs. In conventional
`reservoirs and tight gas sands, single-plane fracture half-length and conductivity are the key drivers for stimulation
`performance. In shale reservoirs, where complex network structures in multiple planes are created, the concept of a single
`fracture half-length and conductivity are insufficient to describe stimulation performance. This is the reason for the concept
`of using stimulated reservoir volume as a correlation parameter for well performance. The size of the created fracture
`network can be approximated as the 3-D volume (Stimulated Reservoir Volume or SRV) of the microseismic event cloud.
`This paper briefly illustrates how the Stimulated Reservoir Volume (SRV) can be estimated from microseismic mapping data
`and is then related to total injected fluid volume and well performance. While the effectively producing network could be
`smaller by some proportion, it is assumed that created and effective network are directly related. However, SRV is not the
`only driver of well performance. Fracture spacing and conductivity within a given SRV are just as important and this paper
`illustrates how both SRV and fracture spacing for a given conductivity can affect production acceleration and ultimate
`recovery. The effect of fracture conductivity is discussed separately in a series of companion papers. Simulated production
`data is then compared with actual field results to demonstrate variability in well performance and how this concept can be
`used to improve completion design, and well spacing and placement strategies.
`
`
`Introduction
`Fisher et al. (2002), Maxwell et al. (2002), and Fischer et al. (2004) were the first papers to discuss the creation of large
`fracture networks in the Barnett shale and show initial relationships between treatment size, network size and shape, and
`production response. Microseismic fracture mapping results indicated that the fracture network size was related to the
`stimulation treatment volume. Figure 1 shows the relationship between treatment volume and fracture network size for five
`vertical Barnett wells, showing that large treatment sizes resulted in larger fracture networks. It was observed that as fracture
`network size and complexity increase, the volume of reservoir stimulated also increases. Fisher et al. (2004) detailed
`microseismic fracture mapping results for horizontal wells in the Barnett shale. This work illustrated that production is
`directly related to the reservoir volume stimulated during the fracture treatments. In vertical wells, larger treatments are the
`primary way to increase fracture network size and complexity. Horizontal well geometry provides other optimization
`opportunities. Longer laterals and more stimulation stages can also be used to increase fracture network size and stimulated
`reservoir volume. Mayerhofer et al. (2006) performed numerical reservoir simulations to understand the impact of fracture
`network properties such as SRV on well performance. The paper also showed that well performance can be related to very
`long effective fractures forming a network inside a very tight shale matrix of 100 nano-darcies or less.
`
`IWS EXHIBIT 1050
`
`EX_1050_001
`
`

`

`Downloaded from http://onepetro.org/spesgpc/proceedings-pdf/08SGPC/All-08SGPC/SPE-119890-MS/2756107/spe-119890-ms.pdf/1 by Robert Durham on 12 August 2022
`
`
`
`SPE 119890
`
`
`
`
`
`450004500045000
`
`
`
`
`
`400004000040000
`
`
`
`
`
`350003500035000
`
`
`
`
`
`300003000030000
`
`
`
`
`
`250002500025000
`
`
`
`
`
`200002000020000
`
`
`
`
`
`150001500015000
`
`
`
`
`
`100001000010000
`
`
`
`
`
`500050005000
`
`
`
`
`
`000
`
`
`
`
`
`000
`
`
`
`
`
`200020002000
`
`
`
`
`
`400040004000
`
`
`
`
`
`600060006000
`
`
`
`
`
`800080008000
`
`100001000010000
`
`120001200012000
`
`
`Fluid Volume, bblFluid Volume, bblFluid Volume, bbl
`
`
`
`
`
`140001400014000
`
`
`
`
`
`160001600016000
`
`
`
`
`
`180001800018000
`
`
`
`
`
`200002000020000
`
`
`
`Fracture Network Length, ft
`Fracture Network Length, ft
`Fracture Network Length, ft
`
`2
`
`
`
`Figure 1. Relationship of total fracture network length as a function of total job fluid volume pumped (SPE 77441)
`
`Figure 2 illustrates the various types of fracture growth ranging from simple fractures to very complex fracture networks.
`Complex fracture networks are desirable in “super-tight” shale reservoirs since they maximize fracture surface contact area
`with the shale through both size and fracture density (spacing). Chances for creating large tensile fracture networks are
`increased by pre-existing healed or open natural fractures and favorable stress-field conditions such as a small difference in
`principal horizontal stresses. Figure 3 shows an example of a complex fracture network in a vertical well measured with
`microseismic mapping. The figure illustrates the development of a large-scale network with two distinct, orthogonal fracture
`orientations. This paper aims to expand on the previous papers by discussing in more detail the concept of stimulated
`reservoir volume and its relationship with shale well performance.
`
`
`
`
`
`
`
`Simple FractureSimple Fracture
`
`
`
`Complex FractureComplex Fracture
`
`
`Complex FractureComplex Fracture
`
`With Fissure OpeningWith Fissure Opening
`
`
`Complex FractureComplex Fracture
`
`NetworkNetwork
`
`Figure 2. Types of fracture growth (from SPE 114173)
`
`
`
`
`
`IWS EXHIBIT 1050
`
`EX_1050_002
`
`

`

`Downloaded from http://onepetro.org/spesgpc/proceedings-pdf/08SGPC/All-08SGPC/SPE-119890-MS/2756107/spe-119890-ms.pdf/1 by Robert Durham on 12 August 2022
`
`SPE 119890
`
`
`
`3
`
`All Microseisms
`
`~2700 ft
`
`Frac
`Well
`
`~1200 ft
`
`Possible
`Aligned
`Features
`
`Observation Well
`
`1500
`
`1000
`
`500
`
`0
`
`-500
`
`-1000
`
`-1500
`
`Northing (ft)
`
`-2000
`-1000
`
`-500
`
`500
`0
`Easting (ft)
`
`Figure 3. Microseismic fracture mapping shows complex network growth in shales (SPE 114173)
`
`1000
`
`1500
`
`
`Estimating Stimulated Reservoir Volume (SRV)
`
`It has been well documented by Albright and Pearson (1982), Warpinski et al. (2001), and Rutledge et al. (2003) that
`microseismic events are mainly created as a result of shear slippages around the hydraulic fractures. The mechanisms include
`shear slippages induced by altered stresses near the tip of the fractures as well as shear slippages related to leakoff induced
`pore pressure changes. In conventional reservoirs with higher reservoir permeabilities and/or oil/water reservoirs, the
`microseismic event cloud can be fairly wide but may not be related to the generation of complex fracture networks. In this
`case most of the events may be related to pore pressure increase as a result of rapidly moving pressure transients in high
`permeability formations and/or reservoirs with fairly incompressible fluids. In “super-tight” shale reservoirs diffusivity
`related pore pressure changes cannot move very far from the actual fracture planes, unless natural fractures in alternate
`directions are opened and hydraulically enhanced as a network structure, thus serving as a conduit for fluid movement. This
`means that a large event cloud structure must be approximately equivalent to the actual fracture network size. Thus, the
`microseismic event cloud structure observed by microseismic fracture mapping provides a means to estimate the SRV in very
`tight reservoirs.
`Figure 4 shows an automated method using discrete bins to estimate SRV from microseismic mapping data in a
`horizontal well. Constant width bins (e.g., 100 ft wide) are drawn in the principal fracture direction from the wellbore to the
`furthest event in the specific bin on both sides of the wellbore. The individual bin areas are then summed up to approximate
`the total stimulated reservoir area (SRA). The calculation of SRV (a 3-dimensional structure) also requires an estimate of the
`stimulated fracture network height in each discrete bin within the contacted shale section. This calculation is also performed
`within the selected bins and is performed by calculating the network height as the difference between the shallowest and
`deepest event within the specific bin and top and bottom of the shale section (Figure 5). While this method is not an
`analytically exact calculation, it does provide a fast automated method to approximate a very complex 3-dimensional
`structure, while honoring the contact with the actual shale section. The SRV is in this paper is specified in millions of cubic
`feet or acre-ft.
`An important aspect of the SRV measurement is the proper setup of observation well or multiple observation wells to
`guarantee that the entire SRV can in fact be observed. The proper design of a microseismic mapping setup takes into account
`maximum observation distance to ensure that the entire SRV can be imaged. Microseismic moment magnitude versus
`distance plots (Zimmer et al. (2007)) can be used to ascertain if all events were within observable range or if the SRV could
`in fact be larger than imaged. Other issues that affect correlations between SRV and well performance include associated
`formation water production, and condensate yield, which becomes relevant in less mature shales close to the oil window.
`Although general guidelines are applicable for SRV measurements, it is important to evaluate any given SRV measurement
`in conjunction with the particular reservoir setting.
`
`IWS EXHIBIT 1050
`
`EX_1050_003
`
`

`

`4
`
`
`
`SPE 119890
`
`Downloaded from http://onepetro.org/spesgpc/proceedings-pdf/08SGPC/All-08SGPC/SPE-119890-MS/2756107/spe-119890-ms.pdf/1 by Robert Durham on 12 August 2022
`
`Figure 4. Estimating SRA from microseismic mapping data
`
`
`
`400
`
`500
`
`600
`
`700
`
`800
`
`900
`
`000
`
`100
`
`200
`
`300
`
`400
`
`500
`
`600
`
`700
`
`800
`
`Events outside of shale section
`
`0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
`
`Figure 5. Conceptual sketch of SRV height calculation
`
`
`
`IWS EXHIBIT 1050
`
`EX_1050_004
`
`

`

`Downloaded from http://onepetro.org/spesgpc/proceedings-pdf/08SGPC/All-08SGPC/SPE-119890-MS/2756107/spe-119890-ms.pdf/1 by Robert Durham on 12 August 2022
`
`SPE 119890
`
`
`
`5
`
`L
`
`ftotal
`
` for the entire SRV as a function of fracture network half-
`
`SRV and Fracture Spacing in Shales
`
`It is important to note that the SRV is just the reservoir volume affected by the stimulation. It does not provide any details
`of the effectively producing fracture structure or spacing. Maxwell et al. (2006) introduced a concept that could eventually be
`used to characterize fracture density. In this approach “additional seismic signal characteristics allow investigation of the
`source of the mechanical deformation resulting in the microseisms. In particular, the seismic moment, a robust measure of the
`strength of an earthquake or microearthquake, can be used to quantify the seismic deformation.” Besides this potential
`geophysical approach, reservoir modeling also provides an avenue to better evaluate the effectively producing network. The
`details of this approach will be described in the following section. As an introduction Figure 6 shows an identical SRV with
`different fracture spacings (densities) and the effect of fracture spacing on the gas recovery factor. The SRV in this graph is
`about 2,000 x 106 ft3 (for h=300 ft). In contrast to conventional single fracture modeling using fracture half-length, the SRA
`(Stimulated Reservoir Area) dimensions are given by the total fracture network length (2xf) and width (xn). However, the key
`property for well production is really the total sum of all fracture network segments (linear feet) within the SRA, which is a
`strong function of fracture spacing (density).
`Equation 1 shows the calculation of total fracture length,
`length xf, width xn and fracture spacing Δxs.
`L
`
`=
`
`ftotal
`
`4
`
`n
`
`xx
`f
`x
`Δ
`
`s
`
`+
`
`2
`
`x
`
`f
`
`+
`
`x
`
`n
`
`......(1)
`
`
`
`Figure 7 shows a plot of Eq. 1 on a log-log plot for the given SRV in Figure 6. The resulting curve forms an approximate
`straight-line on a log-log plot. As an example, reducing fracture spacing from 300 ft to 50 ft would result in a more than 5-
`fold increase in total fracture length (from 48,600 ft to 264,600 ft) and would accelerate the 3-year cumulative gas recovery
`by about the same multiplier. This illustrates the importance of viewing SRV in context with the potential fracture spacing.
`A dual porosity approach is not utilized at this point since the fracture spacing is still relatively sparse and the numerical
`approach better illustrates the linear flow patterns in a low permeability shale system.
`
`
`Xf
`
`Xn
`
`After 1 Year Production
`
`
`Figure 6. Effect of fracture spacing (Δxs) on well performance and recovery factor for SRV~2,000 x 106 ft3 (from SPE
`102103 & 114173)
`
`IWS EXHIBIT 1050
`
`EX_1050_005
`
`

`

`Downloaded from http://onepetro.org/spesgpc/proceedings-pdf/08SGPC/All-08SGPC/SPE-119890-MS/2756107/spe-119890-ms.pdf/1 by Robert Durham on 12 August 2022
`
`6
`
`
`
`SPE 119890
`
`100
`Fracture Spacing (ft)
`
`1000
`
`1.E+06
`
`1.E+05
`
`Total Fracture Length (ft)
`
`1.E+04
`
`10
`
`
`Figure 7. Fracture spacing versus total fracture length based on Eq. 1 for SRV of 2,000 x 106 ft3
`
`
`
`SRV and Well Performance
`
`Previous publications have briefly discussed the relationship between actual SRV and well performance. Here we will
`expand on the subject with more details and comparisons with longer-term production data from the Barnett shale. Figure 8
`shows a plot of SRV measured from microseismic fracture mapping versus 6-month cumulative gas production from a group
`of horizontal wells in one specific county in the Barnett shale. The plot shows that SRV’s in this county range from as low as
`120 x 106 ft3 (2,755 acre-ft) to about 1,900 x 106 ft3 (43,618 acre-ft). Small SRV’s were partly generated as a result of shorter
`laterals and smaller treatment sizes (fluid volumes). While the data shows some scatter, it clearly indicates a trend that large
`SRV’s will result in better well performance. A similar trend is still valid when plotting SRV versus longer term 3-year
`cumulative gas production (Figure 9). The scatter in the data is to be expected as a result of variations in fracture spacing
`within the SRV and potential differences in effective network size and conductivity. While the presented trends in Figure 8
`and Figure 9 show a general relationship between generated SRV from microseismic mapping and production, it does not
`provide any deterministic quantification of how the actual effective network (portion of the network contributing to gas
`production) is structured to provide the specific gas production. The quantification of the effectively producing fracture
`network can only be performed with flow simulations such as provided by a numerical reservoir simulator.
`
`
`
`IWS EXHIBIT 1050
`
`EX_1050_006
`
`

`

`Downloaded from http://onepetro.org/spesgpc/proceedings-pdf/08SGPC/All-08SGPC/SPE-119890-MS/2756107/spe-119890-ms.pdf/1 by Robert Durham on 12 August 2022
`
`SPE 119890
`
`
`
`7
`
`200
`
`400
`
`600
`
`800
`
`1000
`SRV (Mio ft3)
`
`1200
`
`1400
`
`1600
`
`1800
`
`2000
`
`1000
`
`900
`
`800
`
`700
`
`600
`
`500
`
`400
`
`300
`
`200
`
`100
`
`0
`
`0
`
`6-month Cum Gas (MMCF)
`
`
`Figure 8. SRV trend versus 6-month cumulative horizontal well production for one Barnett shale county
`
`400
`
`800
`
`1200
`
`1600
`
`2000
`
`SRV (Mio ft3)
`
`2500
`
`2000
`
`1500
`
`1000
`
`500
`
`3 year Cum Prod (MMcf)
`
`0
`
`0
`
`
`Figure 9. SRV trend versus 3-year cumulative horizontal well production for one Barnett shale county
`
`
`
`Previous work has provided extensive numerical reservoir modeling (Mayerhofer et al. 2006, Warpinski et al. 2008,
`Cipolla et al. 2008) to show how specific fracture network properties can affect well performance. The work demonstrated
`that numerical modeling using a blackoil reservoir simulator can adequately history-match well performance. The numerical
`modeling describes the fracture network as a discrete set of high permeability fractures in two orthogonal directions, with one
`direction representing the principal hydraulic fracture direction as a result of newly created fractures or opening of natural
`fractures and the orthogonal set representing opening of conjugate natural fracture sets. Besides geochemical shale properties
`and pore pressure, the key fracture network properties that affect well performance include SRV, shale matrix permeability,
`fracture spacing and fracture conductivity. It was discussed in these papers that once the value for shale permeability is fixed,
`the solutions for total cumulative fracture length (the sum of all fracture segments in the network) and conductivity become
`fairly unique. Figure 10 shows the results from a vertical well history match (Mayerhofer et al. 2006), where the network
`was approximated with an effectively producing network having almost 8,000 ft of total fracture length within a
`microseismically mapped SRV of 219 x 106 ft3 (gray dots in graph). Fracture spacing was modeled to be about 180 ft in the
`principal fracture direction with a uniform fracture conductivity of 4 md-ft, and shale matrix permeability of 100 nano-darcy
`(0.0001 md). The fracture network is larger on the southwest side of the wellbore with only a few microseismic events to the
`northeast.
`
`IWS EXHIBIT 1050
`
`EX_1050_007
`
`

`

`Downloaded from http://onepetro.org/spesgpc/proceedings-pdf/08SGPC/All-08SGPC/SPE-119890-MS/2756107/spe-119890-ms.pdf/1 by Robert Durham on 12 August 2022
`
`SPE 119890
`
`Casing Pressure
`
`k=0.0001 md
`wkf= 4md-ft
`Lf=7,720 ft
`
`Model
`
`
`
`Gas Rate (Actual versus Predicted)
`
`Actual
`Predicted
`
`
`
`1600.00
`
`1400.00
`
`1200.00
`
`1000.00
`
`800.00
`
`600.00
`
`400.00
`
`200.00
`
`Gas Rate, Mcf/d
`
`0.00
`04/01
`
`11/01
`
`05/02
`
`12/02
`
`06/03
`Date
`
`01/04
`
`08/04
`
`02/05
`
`09/05
`
`8
`
`SRV= 219 x 106 ft3
`180 ft frac spacing in frac direction
`Total Segment Length =7,720 ft
`
`Figure 10. Reservoir simulation results for vertical well from SPE 102103
`
`
`
`Figure 11 shows the simulation setup for a 3,000-ft long horizontal well, from which a SRV of 5000 x106 ft3 with
`equidistant frac spacing of 400 ft was generated. The plot clearly shows that the ultimate drainage area is constrained to the
`created SRV in “super-tight” shale reservoirs. Using this approach, a series of simulations was generated with different SRV
`sizes and fracture spacings to evaluate the effect of these properties on well performance. The SRV was varied from about
`800 x106 ft3 to 6,000 x106 ft3 and equidistant fracture spacing from 50 to 800 ft.
`
`
`After 1 Year
`
`SRV = 5000 x 106 ft^3
`
`After 15 Years
`
`k (matrix) = 0.0001 md
`
`3,000 ft
`
`2,450 ft
`
`Treatment Horizontal Well
`
`Treatment Horizontal Well
`
`
`Figure 11. Simulation of horizontal well fracture network (SRV=5000 x106 ft3, Frac Spacing= 400 ft, Lateral Length~
`3,000 ft)
`
`To crosscheck the generic simulations with actual well production data and a more detailed production history match, the
`vertical well example in Figure 10 was plotted against the more generic simulation results. Although the well configuration is
`different and flowing pressures are not exactly the same, the correlation between SRV, well performance and fracture spacing
`is approximately consistent, showing that the 6-month cumulative production of this vertical well would fall on the 200-ft
`frac spacing curve (Figure 12). Figure 13 shows a plot of the same generic simulations with actual horizontal well Barnett
`shale data throughout the producing areas in North Texas. The figure shows a wide scatter of well results with cumulative gas
`production ranging from less than 100 MMscf to 500 MMscf for SRV’s ranging from 100 x106 ft3, to 3,000 x106 ft3. The data
`indicate that most of the wells plot at a fracture spacing of 200 ft or larger, with many wells at fracture spacings larger than
`800 ft, indicating a sparse fracture network (smaller total fracture length) within the measured SRV (Figure 7) and therefore a
`less effective fracture network. The plot also shows the significant acceleration component of smaller fracture spacings on
`short-term production. For example, at an SRV=1,000 x106 ft3, reducing fracture spacing from 400 ft to 100 ft would result in
`
`IWS EXHIBIT 1050
`
`EX_1050_008
`
`

`

`Downloaded from http://onepetro.org/spesgpc/proceedings-pdf/08SGPC/All-08SGPC/SPE-119890-MS/2756107/spe-119890-ms.pdf/1 by Robert Durham on 12 August 2022
`
`SPE 119890
`
`
`
`9
`
`a 3.5-fold increase in production. The larger the SRV, the more significant the fold of increase in production with smaller
`fracture spacing (due to much longer total fracture length with larger SRV). On the other hand, generating a larger SRV with
`large 800-ft fracture spacing will not provide increased production, which can be explained by the low fracture conductivity
`(and associated increasing pressure drop with distance from the well) in a sparsely fractured network.
`
`
`
`
`Figure 12. Simulated SRV and actual 6-month vertical well production response from SPE 102103
`
`
`
`
`Figure 13. Simulated SRV and simulated 6-month cumulative production versus actual Barnett shale data (entire
`North Texas area)
`
`IWS EXHIBIT 1050
`
`EX_1050_009
`
`

`

`Downloaded from http://onepetro.org/spesgpc/proceedings-pdf/08SGPC/All-08SGPC/SPE-119890-MS/2756107/spe-119890-ms.pdf/1 by Robert Durham on 12 August 2022
`
`10
`
`
`
`SPE 119890
`
`After comparing short-term production, the discussion will now focus on longer-term production (3 to 15 years). Figure
`14 shows a plot of the SRV from the same generic horizontal well simulations versus 3-year cumulative gas production,
`fracture spacing and actual horizontal well Barnett shale data. The data indicate a slight shift of the real data to smaller
`fracture spacings, which may indicate improved cleanup for the longer-term production. The actual cumulative production
`ranges from as low as 100 MMscf to 2,000 MMscf. The simulations also show that the folds of increase in production as a
`function of smaller fracture spacing become less for longer-term recovery as previously illustrated in Figure 6 (early-time
`acceleration). The plot also clearly shows that significant improvements in well performance by generating larger SRV’s is
`more significant at low to moderate SRV’s (less than 1,000 x106 ft3). For larger SRV’s the benefits can only be achieved with
`a simultaneous decrease of fracture spacing. For example, increasing SRV from 1,500 x106 ft3 to 3,000 x106 ft3 with 800 ft
`fracture spacing will have no meaningful increase in gas production.
`
`
`Figure 15 shows the results for 15-year cumulative production. In this case no actual production data is available, since
`horizontal well development started in 2002 in the Barnett shale. The plot shows that for longer-term production the impact
`of SRV becomes more significant even at larger fracture spacing since SRV provides the ultimate drainage area and diffusion
`has had time to access the entire network. In this case, increasing SRV from 1,500 x106 ft3 to 3,000 x106 ft3 with 800 ft
`fracture spacing will result in a 50% cumulative production increase from 2,000 MMscf to 3,000 MMscf (compared to no
`meaningful increase after 3 years of production).
`
`
`
`3-Year Cumulative Gas vs. SRV
`
`Inputs:
`Lateral length, ft:
`Reservoir permeability, md:
`Pore pressure, psi
`Net pay thickness, ft:
`Frac conductivity, md-ft:
`Minimum BHP, psi:
`
`3000
`0.0001
`3000
`300
`4
`1000
`
`Largest Observed SRV per Well
`
`10000
`
`9000
`
`8000
`
`7000
`
`6000
`
`5000
`
`4000
`
`3000
`
`2000
`
`1000
`
`3-Year Cumulative Gas (MMscf)
`
`0
`
`0
`
`500
`
`1000
`
`1500
`
`2000
`
` Network Spacing = 800 ft
` Network Spacing = 100 ft
`
`3500
`3000
`2500
`SRV (Million cuft)
` Network Spacing = 400 ft
` Network Spacing = 50 ft
`
`4000
`
`4500
`
`5000
`
`5500
`
`6000
`
` Network Spacing = 200 ft
` Actual Data
`
`
`
`
`
`IWS EXHIBIT 1050
`
`EX_1050_010
`
`

`

`Downloaded from http://onepetro.org/spesgpc/proceedings-pdf/08SGPC/All-08SGPC/SPE-119890-MS/2756107/spe-119890-ms.pdf/1 by Robert Durham on 12 August 2022
`
`SPE 119890
`
`
`
`11
`
`3-Year Cumulative Gas vs. SRV
`
`2000
`
`1600
`
`1200
`
`800
`
`400
`
`3-Year Cumulative Gas (MMscf)
`
`0
`
`0
`
`500
`
`1000
`
` Network Spacing = 800 ft
` Network Spacing = 100 ft
`
`1500
`SRV (Million cuft)
` Network Spacing = 400 ft
` Network Spacing = 50 ft
`
`2000
`
`2500
`
`3000
`
` Network Spacing = 200 ft
` Actual Data
`
`
`
`Figure 14. Simulated SRV and simulated 3-year cumulative production versus actual Barnett shale data (entire North
`Texas area)
`
`
`
`15-Year Cumulative Gas vs. SRV
`
`Inputs:
`Lateral length, ft:
`Reservoir permeability, md:
`Pore pressure, psi
`Net pay thickness, ft:
`Frac conductivity, md-ft:
`Minimum BHP, psi:
`
`3000
`0.0001
`3000
`300
`4
`1000
`
`Largest Observed SRV per Well
`
`16000
`
`14000
`
`12000
`
`10000
`
`8000
`
`6000
`
`4000
`
`2000
`
`15-Year Cumulative Gas (MMscf)
`
`
`
`0
`
`0
`
`500
`
`1000
`
`1500
`
`2000
`
` Network Spacing = 800 ft
` Network Spacing = 100 ft
`
`3500
`3000
`2500
`SRV (Million cuft)
` Network Spacing = 400 ft
` Network Spacing = 50 ft
`
`4000
`
`4500
`
`5000
`
`5500
`
`6000
`
` Network Spacing = 200 ft
`
`
`
`Figure 15. Simulated SRV versus 15-Year cumulative production
`
`
`Applying SRV and Network Azimuth to Well Placement and Spacing Strategies
`
`
`Figure 16 and Figure 17 show conceptual diagrams illustrating how SRV and fracture network azimuth can be used for
`well placement and spacing considerations. Figure 16 shows the importance of lateral orientation with respect to fracture
`network azimuth and SRV. The graph assumes that the lateral will be drilled in a NE-SW orientation. If fracture direction
`follows the same SW to NE trend, the scenario on the left-side of picture will promote longitudinal fractures with a very
`small SRV. If wells are to be drilled in this orientation, more wells with closer spacing will be required to drain a given
`section of reservoir. The fracture orientation provided in the center of the picture provides an oblique orientation of the
`fracture direction with respect to the lateral and also results in less efficient reservoir coverage but SRV is now larger than in
`the longitudinal case. The best scenario with largest SRV is illustrated on the right-side with transverse network orientation
`providing the largest SRV per lateral.
`
`IWS EXHIBIT 1050
`
`EX_1050_011
`
`

`

`Downloaded from http://onepetro.org/spesgpc/proceedings-pdf/08SGPC/All-08SGPC/SPE-119890-MS/2756107/spe-119890-ms.pdf/1 by Robert Durham on 12 August 2022
`
`12
`
`
`
`SPE 119890
`
`Keeping these considerations in mind, one could postulate that well spacing (distance between laterals) should roughly
`coincide with the extent of the SRV perpendicular to the wellbore as illustrated in Figure 17 on the left-hand side (SRV from
`each lateral is adjacent to the offset well). This approach may appear to be the optimum configuration. While the areal
`coverage of SRV is definitely optimized with this approach, gas recovery factors within the SRV can only be optimum with
`sufficiently close fracture spacing as illustrated in Figure 6. If fracture spacing cannot be optimized with large SRV’s, it may
`be more beneficial to drill laterals at closer spacing, with the chance of overlapping SRV’s (right-hand picture in Figure 17).
`This is the reason for the so-called “simul-fracs” and “zipper-fracs” in shale reservoirs where multiple parallel wellbores are
`either stimulated simultaneously or in short alternating sequence from one lateral to the next. The hypothesis in these
`strategies is to maximize fracture density between laterals as a result of stress diversion from simultaneously growing
`opposing fractures (simul-fracs) and/or creating new branches as a result of increased stresses around newly opened fractures.
`Real data from Barnett shale completions may suggest that this approach has merits but no specific study has been published
`on this topic. The challenge lies in understanding the practical and physical limitations of what is possible in terms of
`network size and fracture spacing at reasonable completion costs (i.e., lateral length, number of stages, perforations and
`treatment size).
`
`
`
`
`Figure 16. Importance of SRV and network azimuth for lateral orientation and well placement
`
`
`
`
`Adjacent SRV = Optimizing areal
`coverage but not recovery factor?
`
`SRV 3
`
`SRV 2
`
`Well 3
`
`SRV 1
`
`Well 2
`
`Well 1
`
`SRV 3
`
`Overlapping SRV = Higher recovery factor
`(smaller frac spacing)
`
`SRV 2
`
`SRV 1
`
`Well 3
`
`Well 2
`
`Well 1
`
`
`Figure 17. Horizontal well placement strategy considerations in conjunction with SRV and fracture spacing in shale
`reservoirs
`
`IWS EXHIBIT 1050
`
`EX_1050_012
`
`

`

`Downloaded from http://onepetro.org/spesgpc/proceedings-pdf/08SGPC/All-08SGPC/SPE-119890-MS/2756107/spe-119890-ms.pdf/1 by Robert Durham on 12 August 2022
`
`SPE 119890
`
`
`
`13
`
`Conclusions
`
`
`
`1. Low permeability shale reservoirs require a large fracture network with small fracture spacing and adequate
`conductivity to maximize well performance. The size of the SRV, fracture spacing and azimuth are key
`components for well placement and spacing strategies.
`
`
`2. Natural features controlling SRV and fracture spacing in shales include shale thickness, stress field and
`magnitude, presence of pre-existing open or healed natural fractures, fracture barriers, rock brittleness and
`geologic features such as faults and karsts. While these parameters are not under direct human control, it is
`critical to understand how they will impact well results, completion and well development strategies.
`
`
`3. Engineering measures to increase SRV and fracture spacing include lateral length and orientation, treatment
`sizes, number of stages, perforation clusters, diversion techniques and/or openhole packer completion systems.
`More perforation clusters and stages in cemented and cased completions increase the likelihood of dense
`fracturing.
`
`
`4. The key completion strategy is to balance the creation of a large SRV with maximum possible fracture density.
`Large SRV’s with small fracture spacing provide maximum well performance and gas recovery factors.
`However, the practical and physical limitations of achieving this goal will require an economic optimization
`process.
`
`
`
`
`
`
`Acknowledgments
`The authors would like to thank Pinnacle Technologies and Carbo Ceramics for supporting the publication of this work.
`Some of this work is also based on previous publications of Devon Energy Corporation data and we thank them for their
`support of those efforts.
`
`Nomenclature
`
`
` hf = fracture height, L
`xf = hydraulic fracture wing or half-length, L
`
`xn = hydraulic fracture network width (from microseismic event pattern), L
`
`Δxs= orthogonal fracture spacing, L
`
`L
` = Total composite fracture length
`
`ftotal
`
`SI Metric Conversion Factors
`acre
`x 4.046 873
`e+03 =
`bbl
`x 1.589 874
`e-01
`=
`cp
`x 1.0
`e-03
`
`=
`ft
`x 3.048 e-01
`
`=
`
`References
`Albright, J.N. and Pearson, C.F. 1982. Acoustic Emissions as a Tool for Hydraulic Fracture Location: Experience at the Fenton Hill Hot
`Dry Rock Site. SPEJ 22: 523-530.
`Cipolla, C.L., Warpinski, N.R., Mayerhofer, M.J., Lolon, E.P., and Vincent, M.C. 2008. The Relationship between Fracture Complexity,
`Reservoir Properties, and Fracture Treatment Design. Paper SPE 115769 presented at the SPE Annual Technical Conference and
`Exhibition, Denver, Colorado, 21-24 September.
`Fisher, M.K., Davidson, B.M., Goodwin, A.K., Fielder, E.O., Buckler, W.S., and Steinsberger, N.P. 2002. Integrating Fracture Mapping
`Technologies to Optimize Stimulations in the Barnett Shale. Paper SPE 77411 presented at the SPE Annual Techn