Cam McBurney and Richard Baker GCA Canada (a division of Baker Hughes)

Field Performance of the Weyburn CO2 Miscible Flood and Factors

Affecting CO2 Performance (Naturally Fractured Reservoirs)

Cam McBurney and Richard Baker GCA Canada (a division of Baker Hughes)

Presented at the 17th Annual CO2 Flooding Conference

December 8-9, 2011

Midland, Texas

2

Overview

• Executive Summary• CO2 miscible floods in NFR

– Midale/Weyburn Canada (horizontal CO2 + waterflood)

• Spraberry Tracer Tests and Waterflooding• What is critical knowledge needed to

characterize NFR for CO2 ?• Conclusions

Executive Summary

• In Weyburn CO2 and waterflood (Canadian) –– moderatemoderate fracture connectivityWaterflooding, horizontal wells and CO2 flooding are

successful.• In Spraberry

– very long range long range high fracture connectivity and stress sensitivity

• Most injection processes in NFR are controlled by;

Long range fracture connectivity

3

4

Weyburn / Midale Fracture Characterization

Horizontal CO2 miscible floodSPE 63286

JPT OCTOBER 1993CIM 98-88

CIM 96-03-03CIM 2007-31WPC 18-986

Cenovus Web site

5

Calgary, Alberta

Weyburn field

Cenovus web site

Weyburn Midale Location

Current well countsTotal Well Count = 807Hz Well Count = 488Injection well Count = 283

NE – SW fracture trends

Weyburn Unit Reservoir Parameters

* Matrix ** variation within unit

9

More intenselyfractured

NE to SW fractures

~3ft fracture Spacing; Marly

~1ft fracture Spacing; Vuggy

Weyburn Unit - Midale Formation Full Field

Start of CO2flood

CO2 +Hz well

Hz well

Weyburn; Midale Formation Full Field Inj

Start of CO2 flood

1980’s vert infill

1990’s horizontalwells

Weyburn; Full Field Well Counts

CO2Injection

13

Weyburn; Full Field – Injector Type

EPIC’s analysis of CO2Incremental Oil Rate~ 25 MSTB/d

Original Oil in PlacePhase OOIP from

Application(e6m3)

OOIP from Application

(MMbbl)

Phase 1A 29.0 183Phase 1B 7.6 48Phase 1C 17 107Phase 1D 18.5 117

Weyburn Midale Phase 1A

Map of Weyburn CO2 PhasesPhase 1A

Phase 1B

Phase 1C

Phase 1D

Phase 1A Composite Plot

Bubble Plots-Phase 1 AWater Cum Produced = 470 MMBblOil Cum Produced = 135 MMBbl

Water Cum Injected = 580 MMBblCO2 Cum Injected = 46,000 MMBbl (257 Bcf)(Injected water bubbles have a multiplier of 100)

Weyburn Phase 1A Production

Weyburn Phase 1B Production

21

Start of CO2 Flood

Start of Water Flood

Weyburn Phase 1C Production

22

Phase 1D (not enough data to do a decline)

Start of Water FloodStart of CO2 Flood

Incremental Oil Recovery (Epic’s interpretation) @ initial reservoir pressure and temperature

CO2 Response Examples

90’s Infill Hz Well• Drop in watercut/increase

in oil rate with CO2 b/t

Original Vertical Well• Drop in watercut/increase

in oil rate with CO2 b/t

27

Recap for Weyburn Unit

Successful Waterflood≈

1980’s vertical well infills

Horizontal wells CO2 flooding

Incremental recovery factor due to CO2 flooding 12-20% OOIP

Total UFR >50%

What was the critical data? (in my opinion)

Lack of high connectivity fractures determined from production + injection testing

Interwell tracer

28

Spraberry West Texas Waterflood CO2 pilot

(DOE/Pioneer Natural Resources)

SPE-39623SPE – 35469SPE – 71635SPE – 71605

David SchechterSchechter, D.: “Waterflooding and CO2 Injection in the Naturally Fractured

Spraberry Trend Area,” JCPT (October 2002)John Lorentz

Glasscock Co

Reagan CoUpton Co

Midland Co

Martin Co

Spraberry Trend Area

8-10 MMMbbl=OOIPDiscovered in 1949Sandstone Stress sensitiveTight low matrix permeability (keff <0.1 md)URF <15%

30

Tests Conducted

Vertical and horizontal well core

• Outcrop• Pulse, PBU and fall off test• FMI (fracure logs)

Production decline analysis

Step rate

CO2 pilot

Interwell tracer

• ( Pilot - 9 years)

31

O'Daniel O'Daniel O'Brian

O'DanielO'Daniel Brunson Boone

O'BrianPowellFloyd

35

33

B-1

D-1

C-1

5

7

12

9 13

10

63

9

8 12

7

A-1

B-11

A-5

BooneE-1 A-6

A-3 A-7

32

12

8W

34

6

1

24

31

36

26

D-1

19

13

30G-1

C-2E-1

BooneA-1

D-1 A-1

F-1

C-1

C-11

D-1

3 2

47

25

29

37

46

C-1

28

45

38

39

40

48

14

41

43

4950

44

42

N

Pilot Area Map

Pilot Area

Well Oil Response to Waterflood

1 mile

32

33

SPE 71635

Tracer Test Results

Section lines~1 mile

Tracer Travelled2 miles in(1-3 days)

WHAT DO I NEED TO DO TO UNDERSTAND RESERVOIR FLOW AND CO2 FLOW?

34

Communication Analysis

– We use Communication Analysis (signal processing) to see fracture connectivity

35

objective

Communication Analysis (signal processing)

But the Extremes of the Data Set can Sometimes Control Injection Process

36Fracture connectivity (distance)

Number ofOccurrences

Average “bulk” properties

“outliers” or extremes

Communication Map

• Need a communication map or if too many zoom in view

• Example;

Early Waterflood History

We know from tracer tests offset Field that lateral fracture Connectivity is not that long <300ft

Weyburn Midale Phase 1A Communicating Well Pair

Injector

Producer

Injector

Producer

Weyburn Midale Phase 1A Communicating Well Pair

Producer

Injector

Weyburn Midale Phase 1A Communicating Well Pair

41

Conclusions from Oil Recovery Point of View for CO2 Injection Processes

Spraberry (West Texas)Long connectivityStress sensitivity

WeyburnModerate fracture connectivitySuccessful; waterflood, hz wells, CO2 flood

The best way for brownfields is through communication analysisPressure pulseProduction data