Directional/complex wells

Unlocking Egypt’s Unconventional-Resource Potential

The Apollonia tight-gas chalk play is located in the Abu Gharadig Basin in the Western Desert of Egypt. This has long been ignored as a gas play in the overburden, while the Jurassic and Cretaceous oil fields deeper in the basin have been explored and developed.

jpt-2018-07-188587-hero.jpg

The Apollonia tight-gas chalk play is located in the Abu Gharadig Basin in the Western Desert of Egypt. This has long been ignored as a gas play in the overburden, while the Jurassic and Cretaceous oil fields deeper in the basin have been explored and developed. However, several structures in the Apollonia are known to contain potentially significant hydrocarbon volumes, although a potential Apollonia full-field development is challenging because of regulated gas prices in the Western Desert and low-productivity formations. This paper discusses the process of developing the first unconventional-gas opportunity in Egypt.

Introduction

Vertical appraisal wells show that low production rates and low estimated ultimate recoveries (EURs) present a challenge for cost-effective development of tight gas in Apollonia. With the play’s decreasing levels of permeability, long-reach horizontal wells are needed with induced stimulation. The optimized technique of deploying multistage hydraulic-fracture stimulation efficiently has been documented and applied successfully in North America and has potential for success in Apollonia. Shell and Apache created a joint-development proposal to unlock the significant stranded gas in Apollonia. The proposal consisted of a staged development, starting with a three-­horizontal-well pilot followed by an optional full-field development.

Geology

Apollonia is a homogeneous reservoir; however, it is very tight, and induced stimulation by hydraulic fracturing is required to produce a commercial and sustainable production rate. Smectite and illite contribute to reservoir quality and can be predicated by conventional logs. Fracture densities in Apollonia are low. The fractures are either closed or only partially open, and their contribution to production is perceived to be low. In addition to these factors, development may require drilling many wells (low spacing) with induced stimulation in order to deliver cost-effective production rates. This requires lower well costs than currently exist. While production from the three existing vertical wells continues, EURs from these wells are suboptimal.

Apollonia comprises tight, micro­porous chalky carbonates that are proved to contain movable hydrocarbons. The formation is subdivided into four members, Apollonia A (top layer) through D (bottom layer). Apollonia A and C are composed of thick massive limestones (chalk) with minor marly and shaley intervals, while Apollonia B and D are dominated by shale. Most of the porous intervals occur within Apollonia A and C. Regional correlations have shown that most of the thickness variations are confined to Apollonia C and, to a lesser extent, D. However, recent seismic interpretation has shown that there are also thickness variations in Apollonia A and B associated with Eocene inversion. The individual porous zones within Apollonia A and, to a lesser extent, C are laterally correlatable over large distances.

Pilot: Horizontal Wells With Multistage Hydraulic Fracturing

The pilot-phase execution began with the drilling of the first horizontal well (1000-m horizontal section) in Q2 2016. The well was drilled vertically to the kickoff point then began building angle until 9⅝-in. casing was landed and cemented back to surface. Drilling continued with an 8¾-in. bit, and the 7-in. casing was cemented back in the 9⅝-in. casing. 5-in. liner casing was set back to the surface and cemented. The well was then perforated and completed with an eight-stage fracture completion. The flowback was controlled so that it would not carry or draw sand out of the fractures. The second well was drilled in Q4 2016 and started production by early 2017. The second well was perpendicular to the first well, with a high-angle/oblique-fracture orientation.

The six vertical wells (appraisal) were shut in to accommodate the production of the horizontal pilot-phase wells. Commingling the production of the vertical and horizontal wells in the same pipeline to the facility faced challenges. Thus, the decision was made to close the low-producing vertical wells until sufficient data were gathered from the pilot phase. The challenges implied that the two horizontal wells were not produced together, and the first well had to be shut in for 3 months while the second well was closely& monitored.

The horizontal wells were drilled successfully, and the horizontal sections were placed in the thin Apollonia A5 reservoir. The horizontal multistage fracturing wells showed much higher initial gas rates compared with the vertical foam-fractured wells (Fig. 1). The production of both wells slowly declined over time; however, their production trend has begun to flatten out as of the writing of this paper. The two wells are producing at very high water/gas ratios (WGRs) compared with the vertical wells (not all vertical wells produced water), and the former have been producing water from Day 1. The second well (with an oblique fracturing orientation) was producing at double the WGR of the first horizontal well at the time of writing.

jpt-2018-07-188587f1.jpg
Fig. 1—Comparison of vertical and horizontal multistage-fracturing-well performance.

 

Two theories could explain this water production. The first is that the water production is coming from the matrix itself as the reservoir quality quickly deteriorates with increased clay content. The second theory suggests that the water is coming through faults either intersected by the wells or connected to natural fractures that are connected to the aquifer. The fact that gas should be much more mobile than water suggests that the latter theory is more plausible. The image logs of the two wells show many such events interpreted as faults and fractures.

The actual cause of the high water production from Day 1 is still under investigation. It is believed that the reason behind the relatively sustainable flow of the horizontal multistage-fracturing wells compared with that of the vertical wells is the longer surface area of the horizontal wells.

Estimation of EUR per Well

Here, a method of production forecasting for tight-gas reservoirs without the use of numerical modeling is outlined. The method uses a stochastic technique that couples data from the long-­producing vertical wells (4 years of production) with the initial results of the horizontal wells to estimate a range of EUR per well for the horizontal multistage-fracturing wells in Apollonia while considering the wide range of uncertainties.

Three-segment decline-curve analysis (DCA) is applicable for wells that transition from years of transient flow into years of boundary-dominated flow. A probabilistic DCA work flow was implemented for this project in four steps:

  1. The time was estimated at which the wells would show the true decline. This was estimated by plotting wellhead pressure against time to find the point at which the tubinghead pressure would be the dominating factor in the decline. The initial decline rate from that period also was estimated.
  2. Segment 1 decline parameters were calculated. The data from the horizontal wells were insufficient and did not yet show that decline; thus, the decline parameters were estimated from the six vertical wells. Multiple fits were used on each of the six wells, which led to a range of decline parameters, and then a distribution was fitted.  
  3. The function of porosity, formation and fluid compressibilities, fluid viscosity, drainage area, and permeability were calculated. Each of these parameters has its own distribution derived either from core data or pressure/volume/temperature (PVT) analysis. A Monte Carlo simulation derived a distribution of time for boundary-dominated flow.
  4. The Monte Carlo simulation was run. The final distribution was then discretized to determine the EUR per well.

The midcase EUR per well was validated by using rate transient analysis. A model was built with the first horizontal-well data available at that point in time—post-fracturing model results (fracturing dimensions and conductivity), PVT data, core data, and available gas rate vs. wellhead pressure (WHP). The WHP was converted to bottomhole pressure. The reliability is subjective because of the high levels of water produced.

Conclusion

The Apollonia tight-chalk formation was targeted as a candidate for long horizontal completion with multistage fracture stimulation to enhance the reservoir’s commercial potential after an appraisal phase with six vertical foam-fractured wells. A pilot phase was designed to demonstrate the cost-effective execution of drilling horizontal laterals with multistage fractures, establish initial production rates of the horizontal wells, and estimate EUR per well.

At the time of writing, two of the three planned pilot horizontal wells have been drilled and completed with eight stages of fracture stimulation. The initial results of these two wells show much better sustainable relatively higher gas rates compared with stimulated vertical wells in a tight gas reservoir (0.1–1 md).

Stochastic DCA was used to couple the data from the long-producing vertical wells and initial results from the horizontal wells to estimate a range of EUR per well for the horizontal wells, which will later help in defining full-field development better.

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 188587, “Unlocking Egypt’s Unconventional Potential,” by Amr Zaher, Etienne Loubens, Mohamed Zayed, SPE, Nicholas Gill, SPE, Oneil Sadhu, SPE, and Layla El Hares, SPE, Shell, prepared for the 2017 Abu Dhabi International Petroleum Exhibition and Conference, Abu Dhabi, 13–16 November. The paper has not been peer reviewed.