**5. Implications and deductions**

that there appear to be sweet spots even within the core area, but there are substantial outliers

0 0.25 0.5 0.75 1 1.25

Production Coefficient, (bcf/year^0.5))

**Figure 10.** Distribution of values of the Production Coefficient for horizontal and vertical wells in the Core region, Tier

having a production coefficient in excess of 0.75 (bcf/yr^0.5) is about 8%.

outer areas. At the date of these figures (2009) Tier 2 was relatively unexploited.

provides more productive fracture surface area than does vertical well technology).

It is apparent form this cursory division that the fraction of wells that were completed horizontally shifts from 43% in the Core area to 94% in the Tier 1 area, which reflects the development of technology with time and the spread of drilling with time to the

The result of this detailed analysis (Figure 10) allows us to quantify the production variations in the Barnett Core, Tier 1 and Tier 2 areas and to distinguish the impact of horizontal and vertical well completions on the productivity. In a sense this represents a first, somewhat crude, pass at distinguishing the impact of nature (in the sense that reservoir properties depend on location, with the core area providing more fertile ground than Tier 1 or Tier 2) and nurture (in the anticipation that horizontal well technology

core area with high values of the production coefficient (in green), 10 of the wells with medium values (in blue) and 10 wells with low values (in red). It is apparent that there appear to be sweet spots even within the core area, but there are substantial outliers

> core horizontal‐2974 wells core vertical‐3886 wells tier 1 horizontal‐ 3865 wells tier 1 vertical‐251 wells tier 2 horizontal‐687 wells tier 2 vertical‐401 wells

and there are some relatively poor wells close to better wells.

**Figure 9.** Development of the Barnett shale in North Texas

348 Effective and Sustainable Hydraulic Fracturing

the Barnett shale.

1 region and Tier 2 region of the Barnett shale.

0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00 100.00

Cumulative

Probability

 (%)

and there are some relatively poor wells close to better wells.

Figure 9. Development of the Barnett shale in North Texas

One of the advantages of the semi-analytic method outlined in this paper is that it enables us to make certain deductions about the magnitude of the parameters that drive the productivity of the well. In particular we can make some inferences about the magnitude of the fracture surface area through which the gas is produced and about the likely spacing of the productive fractures.

#### **5.1. Productive fracture surface area**

Figure 10.Distribution of values of the Production Coefficient for horizontal and vertical wells in the Core region, Tier 1 region and Tier 2 region of Our analytic solution allows us to relate the Productivity Coefficient *CP* to a group of parameters that may be roughly divided into those that characterize the nature of the reservoir and those that characterize the impact of the completion and stimulation strategy. In our formula for *CP* (equation (16)) perhaps the parameter that has the greatest uncertainty is the productive fracture surface area, A. This is the area of the fractures in contact with the reservoir that serve as the channels that convey gas from the matrix to the wellbore. We can make no assertion at this time point about whether these fractures are natural fractures or propped or unpropped hydraulic fractures and nor can we say anything (yet) about their spacing or their lengths or indeed their number and location. We can, however, make an estimate from the production data analysis of the total surface area of these productive fractures.

In Figure 12 we show estimates of the productive fracture surface area for typical strong, medium and weak performing wells in the core area of the Barnett in terms of values of the matrix permeability. In this calculation we have made reasonable estimate of the other parameters that impact the productivity coefficient such as the gas viscosity and compressi‐ bility and the matrix porosity.

Several features of this plot merit discussion:


Figure 12.Estimate of productive fracture surface area for specified matrix permeability. The three curves were developed based on analysis of Barnett shale well data using the EGI semi-analytic production model. **Figure 12.** Estimate of productive fracture surface area for specified matrix permeability. The three curves were devel‐ oped based on analysis of Barnett shale well data using the EGI semi-analytic production model.

To place these numbers in context we note that a fracture of height 200 ft and half length 200 ft has surface area of 0.16 Msqft. Thus, 20 of these fractures would have a fracture surface area of 3.2 Msqft, which is a perfectly plausible estimate of the hydraulic fracture surface area created with modern multi-stage fracturing techniques. (Note that 20 such fractures would be spaced about 150 ft apart in a 3000 ft lateral.) Figure 12 was developed on the basis of production data from wells in the core area of the Barnett shale, but the results apply, at To place these numbers in context we note that a fracture of height 200 ft and half length 200 ft has surface area of 0.16 Msqft. Thus, 20 of these fractures would have a fracture surface area of 3.2 Msqft, which is a perfectly plausible estimate of the hydraulic fracture surface area created with modern multi-stage fracturing techniques. (Note that 20 such fractures would be spaced about 150 ft apart in a 3000 ft lateral.)

plays described earlier in this paper) cannot be achieved by producing from the hydraulic fractures alone.

productivity is concerned. All we have demanded is that their conductivity is sufficiently large that

that a propped fracture surface area of the order of a few million square feet is quite plausible.

is that area created and what are the implications of this figure?

surface area may be estimated at

� Inline formula

**5.2. Productive fracture spacing** 

��� �

project.

least qualitatively, to other shale or tight gas plays. For example for more conventional tight gas plays for which the permeability is of the order of 1 μd or more, we should expect respectable productivity with only one such hydraulic fracture, which reinforces our experience that a vertical well with a single bi-wing fracture may be adequate for those reservoirs, but not for shale gas plays. Conversely the productive fracture surface area for economic production from ultra-tight shale plays (such as the shallow shale

We have been careful so far to make no formal distinction between the natural fractures and the hydraulic fractures in so far as

principle present too great a restriction on the fracture conductivity whether it is associated with propped fractures or unpropped fractures. On the basis of our data analysis we should expect a productive fracture surface area in the range of 1-6 Msqft. How then

A crude estimate of the fracture surface area that is created by pumping large volumes of frac fluid may be made by performing a mass balance and assuming that none of the fluid has leaked off or imbibed into the formation over the time in which the fracture network is created. For a total fluid volume V and assuming a created average fracture width w during pumping, the total fracture

(using any consistent set of units of course). If, for example, 100 million gallons of frac fluid were pumped and the assumed frac width was 0.2 inches, then the total surface area would be about 100 Msq ft. Clearly, this is far in excess of our estimate of the productive fracture surface area and would suggest that less than 10% of the created fracture surface area is actually productive. Naturally, this raises all sorts of other questions concerning the efficiency of this process, which we plan to address in a future

A similar mass balance for the proppant placed in a typical job enables an estimate to be made of the surface area of propped fractures. If we make some estimate of the likely width (0.1 in) and porosity (0.4) of a propped fracture (after closure), it appears

Some insight about the spacing of these productive fractures can be obtained by examining the time scale of pressure diffusion in the matrix. We demonstrated earlier that we may expect the root-time solution to be valid until neighboring fractures begin to compete with one another for production. In other words until, pressure diffusion into a fracture can no longer be considered to be

1 , which should not in

Figure 12 was developed on the basis of production data from wells in the core area of the Barnett shale, but the results apply, at least qualitatively, to other shale or tight gas plays. For example for more conventional tight gas plays for which the permeability is of the order of 1 μd or more, we should expect respectable productivity with only one such hydraulic fracture, which reinforces our experience that a vertical well with a single bi-wing fracture may be adequate for those reservoirs, but not for shale gas plays. Conversely the productive fracture surface area for economic production from ultra-tight shale plays (such as the shallow shale plays described earlier in this paper) cannot be achieved by producing from the hydraulic

The Role of Natural Fractures in Shale Gas Production

http://dx.doi.org/10.5772/56404

351

We have been careful so far to make no formal distinction between the natural fractures and the hydraulic fractures in so far as productivity is concerned. All we have demanded is that their conductivity is sufficiently large that *λ* < <1 , which should not in principle present too great a restriction on the fracture conductivity whether it is associated with propped fractures or unpropped fractures. On the basis of our data analysis we should expect a productive fracture surface area in the range of 1-6 Msqft. How then is that area created and what are the

A crude estimate of the fracture surface area that is created by pumping large volumes of frac fluid may be made by performing a mass balance and assuming that none of the fluid has leaked off or imbibed into the formation over the time in which the fracture network is created. For a total fluid volume V and assuming a created average fracture width w during pumping,

(using any consistent set of units of course). If, for example, 100 million gallons of frac fluid were pumped and the assumed frac width was 0.2 inches, then the total surface area would be about 100 Msq ft. Clearly, this is far in excess of our estimate of the productive fracture surface area and would suggest that less than 10% of the created fracture surface area is actually productive. Naturally, this raises all sorts of other questions concerning the efficiency of this

A similar mass balance for the proppant placed in a typical job enables an estimate to be made of the surface area of propped fractures. If we make some estimate of the likely width (0.1 in) and porosity (0.4) of a propped fracture (after closure), it appears that a propped fracture

Some insight about the spacing of these productive fractures can be obtained by examining the time scale of pressure diffusion in the matrix. We demonstrated earlier that we may expect the root-time solution to be valid until neighboring fractures begin to compete with one another for production. In other words until, pressure diffusion into a fracture can no longer be considered to be independent of the fracture spacing. According to the analysis presented earlier we should expect the cumulative production data to deviate from a straight-line in the

fractures alone.

*<sup>A</sup>*=2 *<sup>V</sup> w*

implications of this figure?

the total fracture surface area may be estimated at

process, which we plan to address in a future project.

**5.2. Productive fracture spacing**

surface area of the order of a few million square feet is quite plausible.

Figure 12 was developed on the basis of production data from wells in the core area of the Barnett shale, but the results apply, at least qualitatively, to other shale or tight gas plays. For example for more conventional tight gas plays for which the permeability is of the order of 1 μd or more, we should expect respectable productivity with only one such hydraulic fracture, which reinforces our experience that a vertical well with a single bi-wing fracture may be adequate for those reservoirs, but not for shale gas plays. Conversely the productive fracture surface area for economic production from ultra-tight shale plays (such as the shallow shale plays described earlier in this paper) cannot be achieved by producing from the hydraulic fractures alone.

We have been careful so far to make no formal distinction between the natural fractures and the hydraulic fractures in so far as productivity is concerned. All we have demanded is that their conductivity is sufficiently large that *λ* < <1 , which should not in principle present too great a restriction on the fracture conductivity whether it is associated with propped fractures or unpropped fractures. On the basis of our data analysis we should expect a productive fracture surface area in the range of 1-6 Msqft. How then is that area created and what are the implications of this figure?

A crude estimate of the fracture surface area that is created by pumping large volumes of frac fluid may be made by performing a mass balance and assuming that none of the fluid has leaked off or imbibed into the formation over the time in which the fracture network is created. For a total fluid volume V and assuming a created average fracture width w during pumping, the total fracture surface area may be estimated at

*<sup>A</sup>*=2 *<sup>V</sup> w*

In Figure 12 we show estimates of the productive fracture surface area for typical strong, medium and weak performing wells in the core area of the Barnett in terms of values of the matrix permeability. In this calculation we have made reasonable estimate of the other parameters that impact the productivity coefficient such as the gas viscosity and compressi‐

**•** Well productivity increases with productive fracture surface area and with matrix perme‐

**•** Wells in this group typically have matrix permeabilities in the range 100-300 nd and this implies that the productive fracture surface area is in the range 1-6 million square feet (Msqft): the higher the permeability, the less fracture surface area is needed to achieve the

**•** If the matrix permeability was as low as even 10 nd, the required fracture surface area would approach 100 Msqft. On the other hand, matrix permeabilities of the order of 1 μd would

Typical Productive Fracture Surface

Figure 12.Estimate of productive fracture surface area for specified matrix permeability. The three curves were developed based on analysis of

poor producer average producer good producer

To place these numbers in context we note that a fracture of height 200 ft and half length 200 ft has surface area of 0.16 Msqft. Thus, 20 of these fractures would have a fracture surface area of 3.2 Msqft, which is a perfectly plausible estimate of the hydraulic fracture surface area created with modern multi-stage fracturing techniques. (Note that 20 such fractures would be spaced about

Figure 12 was developed on the basis of production data from wells in the core area of the Barnett shale, but the results apply, at least qualitatively, to other shale or tight gas plays. For example for more conventional tight gas plays for which the permeability is of the order of 1 μd or more, we should expect respectable productivity with only one such hydraulic fracture, which reinforces our experience that a vertical well with a single bi-wing fracture may be adequate for those reservoirs, but not for shale gas plays. Conversely the productive fracture surface area for economic production from ultra-tight shale plays (such as the shallow shale

We have been careful so far to make no formal distinction between the natural fractures and the hydraulic fractures in so far as

principle present too great a restriction on the fracture conductivity whether it is associated with propped fractures or unpropped fractures. On the basis of our data analysis we should expect a productive fracture surface area in the range of 1-6 Msqft. How then

A crude estimate of the fracture surface area that is created by pumping large volumes of frac fluid may be made by performing a mass balance and assuming that none of the fluid has leaked off or imbibed into the formation over the time in which the fracture network is created. For a total fluid volume V and assuming a created average fracture width w during pumping, the total fracture

(using any consistent set of units of course). If, for example, 100 million gallons of frac fluid were pumped and the assumed frac width was 0.2 inches, then the total surface area would be about 100 Msq ft. Clearly, this is far in excess of our estimate of the productive fracture surface area and would suggest that less than 10% of the created fracture surface area is actually productive. Naturally, this raises all sorts of other questions concerning the efficiency of this process, which we plan to address in a future

A similar mass balance for the proppant placed in a typical job enables an estimate to be made of the surface area of propped fractures. If we make some estimate of the likely width (0.1 in) and porosity (0.4) of a propped fracture (after closure), it appears

Some insight about the spacing of these productive fractures can be obtained by examining the time scale of pressure diffusion in the matrix. We demonstrated earlier that we may expect the root-time solution to be valid until neighboring fractures begin to compete with one another for production. In other words until, pressure diffusion into a fracture can no longer be considered to be

1 , which should not in

plays described earlier in this paper) cannot be achieved by producing from the hydraulic fractures alone.

productivity is concerned. All we have demanded is that their conductivity is sufficiently large that

0 100 200 300 400 500

**matrix permeability (nD)**

**Figure 12.** Estimate of productive fracture surface area for specified matrix permeability. The three curves were devel‐

To place these numbers in context we note that a fracture of height 200 ft and half length 200 ft has surface area of 0.16 Msqft. Thus, 20 of these fractures would have a fracture surface area of 3.2 Msqft, which is a perfectly plausible estimate of the hydraulic fracture surface area created with modern multi-stage fracturing techniques. (Note that 20 such fractures would be

that a propped fracture surface area of the order of a few million square feet is quite plausible.

require less than about 1 Msqft of productive fracture surface area.

Barnett shale well data using the EGI semi-analytic production model.

oped based on analysis of Barnett shale well data using the EGI semi-analytic production model.

is that area created and what are the implications of this figure?

150 ft apart in a 3000 ft lateral.)

spaced about 150 ft apart in a 3000 ft lateral.)

**Surface area (MM sq ft)**

surface area may be estimated at

� Inline formula

**5.2. Productive fracture spacing** 

��� �

project.

bility and the matrix porosity.

350 Effective and Sustainable Hydraulic Fracturing

ability, as expected.

given productivity.

Several features of this plot merit discussion:

(using any consistent set of units of course). If, for example, 100 million gallons of frac fluid were pumped and the assumed frac width was 0.2 inches, then the total surface area would be about 100 Msq ft. Clearly, this is far in excess of our estimate of the productive fracture surface area and would suggest that less than 10% of the created fracture surface area is actually productive. Naturally, this raises all sorts of other questions concerning the efficiency of this process, which we plan to address in a future project.

A similar mass balance for the proppant placed in a typical job enables an estimate to be made of the surface area of propped fractures. If we make some estimate of the likely width (0.1 in) and porosity (0.4) of a propped fracture (after closure), it appears that a propped fracture surface area of the order of a few million square feet is quite plausible.

#### **5.2. Productive fracture spacing**

Some insight about the spacing of these productive fractures can be obtained by examining the time scale of pressure diffusion in the matrix. We demonstrated earlier that we may expect the root-time solution to be valid until neighboring fractures begin to compete with one another for production. In other words until, pressure diffusion into a fracture can no longer be considered to be independent of the fracture spacing. According to the analysis presented earlier we should expect the cumulative production data to deviate from a straight-line in the root-time plot for *t* >0.15*Tm* . If we could detect the time at which this departure occurs then, we have some information with which to estimate the productive fracture spacing. Even if the entire production history to date is in the linear flow regime, we can at least make an estimate of a lower bound on the fracture spacing. in the linear flow regime, we can at least make an estimate of a lower bound on the fracture spacing. It is instructive to estimate the matrix diffusion time for typical values of the fracture spacing and matrix permeability. The results are shown in Figure 13. The diffusion time increases quadratically with the fracture spacing and inversely with the matrix permeability. Typical values for the diffusion time are quite low. For example, if, as we expect, linear flow continues for at least 3

we have some information with which to estimate the productive fracture spacing. Even if the entire production history to date is

**6. Conclusions**

this respect.

pumped, but

**Nomenclature**

k – permeability

*c* – gas compressibility

*cf* – fracture conductivity

(17))

A common view of production mechanisms in shales is "because the formations are so tight gas can be produced only when extensive networks of natural fractures exist" [6]. To this extent gas production from some of the shallower (Devonian) shales is similar to gas production from coal. As we have discussed earlier in this paper, we expect that the deeper gas shales differ in

The Role of Natural Fractures in Shale Gas Production

http://dx.doi.org/10.5772/56404

353

Using a new semi-analytic production model, we have analyzed production data from a number of shale gas wells in several different North American shale gas plays. Interpretation of the results suggest that productivity is largely determined by a small group of parameters that may be decomposed into two sub-groups representing the nature of the reservoir (such as matrix permeability and porosity) and what we may term our (engineering) attempts at nurture (including completion and stimulation parameters). Of key importance is the produc‐ tive fracture surface, which unfortunately is difficult to estimate a priori. However, our

**•** The volume of these productive fractures is very much less than the volume of water

**•** Typically, there is no indication of fracture interference during production even after several

We are led to the conclusion that almost all the fracturing fluid pumped during a multi-stage horizontal well fracturing operation in the shales serves to open a vast, and possibly complex, network of natural fractures and that these fractures do not make a significant contribution to the well's productivity. We are led inevitably to questions concerning the conductivity of these, largely unpropped, fractures and to investigate the rock and fluid mechanisms that seemingly prevent them from being productive. The role of the fracturing fluid (usually slickwater) in

*φ* – porosity *μ* – gas viscosity *λ* – dual porosity transmissivity factor (defined in equation

**•** Productive fracture volume scales approximately with the volume of proppant placed.

interpretation of the production data suggest the following

**•** Productive fracture surface area ~1-6 Msqft and probably within 2-4 Msq ft.

years, which suggests that the productive fracture spacing is at least 100 ft.

**•** Time to drain 90% of the fractured region or matrix blocks: ~10-20 years

this process should now be investigated from this new perspectivel

It is instructive to estimate the matrix diffusion time for typical values of the fracture spacing and matrix permeability. The results are shown in Figure 13. The diffusion time increases quadratically with the fracture spacing and inversely with the matrix permeability. Typical values for the diffusion time are quite low. For example, if, as we expect, linear flow continues for at least 3 years, then we should expect to see a diffusion time of the order of 20 years. Figure 13 suggests that the productive fracture spacing is likely to be of the order of 100 ft or more. years, then we should expect to see a diffusion time of the order of 20 years. Figure 13 suggests that the productive fracture spacing is likely to be of the order of 100 ft or more. In figure 13 we have identified the diffusive time scale with the matrix drainage time. As we showed above, 90% of the total gas in the pore space between the fractures has been drained by this time. A time scale of about 20 years is at least consistent with the industry estimates of the effective production lifetime of these wells. It is worth noting here the consequences of much smaller fracture spacing. For a fracture spacing of only 10 ft, we estimate that 90% of the total gas production will have occurred within the first few months of production, which is quite unrealistic. Note also that the surface area of planar fractures only 10 ft apart in a

3000 ft lateral would be of the order of 150 Msq ft, which again is unreasonably large.

Figure 13.The impact of fracture spacing on the time to produce 90% of the gas in place. **Figure 13.** The impact of fracture spacing on the time to produce 90% of the gas in place.

our interpretation of the production data suggest the following

the productive fracture spacing is at least 100 ft.

**6. Conclusions**  A common view of production mechanisms in shales is "because the formations are so tight gas can be produced only when extensive networks of natural fractures exist" [6]. To this extent gas production from some of the shallower (Devonian) shales is similar to gas production from coal. As we have discussed earlier in this paper, we expect that the deeper gas shales differ in this respect. Using a new semi-analytic production model, we have analyzed production data from a number of shale gas wells in several different North American shale gas plays. Interpretation of the results suggest that productivity is largely determined by a small group of parameters that may be decomposed into two sub-groups representing the nature of the reservoir (such as matrix In figure 13 we have identified the diffusive time scale with the matrix drainage time. As we showed above, 90% of the total gas in the pore space between the fractures has been drained by this time. A time scale of about 20 years is at least consistent with the indus‐ try estimates of the effective production lifetime of these wells. It is worth noting here the consequences of much smaller fracture spacing. For a fracture spacing of only 10 ft, we estimate that 90% of the total gas production will have occurred within the first few months of production, which is quite unrealistic. Note also that the surface area of planar frac‐ tures only 10 ft apart in a 3000 ft lateral would be of the order of 150 Msq ft, which again is unreasonably large.

Productive fracture surface area ~1-6 Msqft and probably within 2-4 Msq ft.

Time to drain 90% of the fractured region or matrix blocks: ~10-20 years

The volume of these productive fractures is very much less than the volume of water pumped, but

Productive fracture volume scales approximately with the volume of proppant placed.

permeability and porosity) and what we may term our (engineering) attempts at nurture (including completion and stimulation parameters). Of key importance is the productive fracture surface, which unfortunately is difficult to estimate a priori. However,

Typically, there is no indication of fracture interference during production even after several years, which suggests that
