HOW TO GET THE MOST OUT OF YOUR

MULTI-STAGE UNCONVENTIONAL
FRACTURE DESIGN
Based on SPE-184816

Bob Shelley, PE
Brian Davidson
Koras Shah
Amir Mohammadnejad, PhD
Stanislav Sheludko
McGuire and Sikora Post Fracture Productivity Increase

• For high-permeability reservoirs,

fracture conductivity is more
important than fracture length.
• For low-permeability reservoirs,
fracture length is more important than
fracture conductivity.
• For a given fracture length, there is an
optimum value of conductivity ratio
• For horizontal wellbores, the number
of fractures propagated can multiply
the stimulation ratio for the well.
𝑤𝑓𝑘𝑝𝑟𝑜𝑝
𝑘𝑓𝑜𝑟𝑚
2
 3 Well Utica Pad, Monroe Co. OH - TVD 10,600 ft.
• Direct Offsets
– Similar Prod Start Dates
– 220 Days Production
• Similar Lateral Lengths
• Different Frac Designs
– Proppant Volume & Selection
– Frac Volume
– Frac Stages Completed

Prop. Wt. Fluid Vol. Lateral Length Frac Stages Perf Clusters
Well Prop Description
(Million Lb.) (Million gal.) (ft.) Completed per Stage

A Resin Coated Sand (RCS) - Sand 10.3 9.3 5,800 24 5

B Low Density Ceramic (LDC) - Sand 11.5 10.3 6,100 24 5
C Low Density Ceramic (LDC) 7.1 8.8 6,200 21 5

3
Well Production Comparison
RCS-Sand
LDC-Sand
LDC

4
Well Performance Modeling Approach
Production History Match Drilling, Reservoir & Geology Frac Pressure Match

Property (Unit) FP1 FP2 FP3

Pore Pressure (psi) 9,074 9,074 9,074

Pore Pressure Gradient (psi/ft) 0.85 0.85 0.85

Reservoir Permeability (mD) 0.0035 0.0035 0.0035

Effective Fracture Half-length (ft) 170 135 100

Effective Fracture Height (ft) 110 75 75

# Fractures 114 114 114

Effective Fracture Conductivity (mD-ft) 11.0 6.0 3.0

Dimensionless Conductivity [FCD] 5.9 4.0 2.8

Calibrated Reservoir Model Calibrated Frac Model

Compare Fracture
Characteristics
Frac Model Geometry - Wellbore Profile View

RCS-Sand; 120 Fractures

LDC-Sand; 114 Fractures

LDC; 94 Fractures
 Reservoir Model - Production History Match (Well B)

FP 1 FP 2 FP 3

Property (Unit) FP1 FP2 FP3

Pore Pressure (psi) 9,074 9,074 9,074

Pore Pressure Gradient (psi/ft) 0.85 0.85 0.85

Reservoir Permeability (mD) 0.0035 0.0035 0.0035

# Fractures 114 114 114

Effective Fracture Half-length (ft) 170 135 100

Effective Fracture Height (ft) 110 75 75

Effective Fracture Conductivity (mD-ft) 11.0 6.0 3.0

Dimensionless Conductivity [Fcd] 5.9 4.0 2.8

7
Surface Flowing Pressure & Stress on Proppant (Well B)
Actual Flowing Pressure (psi) Stress on Proppant (psi)
8000

7000
6,800 psi
6000

5000
4,500 psi
4000

3000

2000

1000 Flow Flow Flow

Period 1 Period 2 Period 3
0
0 20 40 60 80 100 120 140 160 180 200 220
Producing Time (Days)
8
 Stress vs. Proppant Conductivity
20
LDC 30/50
White Sand 30/50
Conductivity (md*ft) 15
4,500 psi 6,800 psi

-36%
10

-58%
5
-60%

-90%
0
1000 3000 5000 7000 9000 11000
Source: Dynamic Conductivity from PredictK (StimLab)
Stress (psi)
9
 Fracture Modeling
RCS - Sand
Fracture RCS- LDC-
LDC
Most Degradation Characteristics Sand Sand

Proppant (lb) 85,800 100,900 75,500

LDC - Sand
Fluid Volume (gal) 77,500 90,400 93,600
Avg Created Half-Length (ft) 650 730 680
Most Effective Avg. Propped Half-Length (ft) 540 630 530
Avg Propped height (ft) 200 230 190
Effective Half-Length (ft) 150 - 75 170 - 100 165 - 100
LDC
Effective Height (ft) 85 - 60 110 - 75 65 - 60
Effective Cond. (md-ft) 10 - 2 11 - 3 10 - 5
Most Efficient
Contributing Area (Msqft/frac) 25.5 – 9.0 37.4 – 15.0 21.5 – 12.0
Fracture Efficiency % 12.0 - 4.2 12.9 - 5.3 10.5 - 6.1

8
 Fracture Efficiency vs. Conductivity
Fracture Efficiency (FE) = Effective Area (Aeff)/Propped Area (Aprop)

LDC
LDC-Sand
RCS-Sand
r2=0.965
𝐹𝐸 = 0.01 ∗ 𝐹𝑐 + 0.02

9
 Frac Design Production Forecasts
4

3.5

3
Cumulative Gas (BCF)

2.5

1.5

Lateral Frac Cost Total Proppant Fluid Volume 1Y Cum Gas

1 Case
Length (ft) (Millon $)
Frac Stages
(Million lb) (Thousand BBL) (BCF)
Resin Coated Sand-Sand 5,800 $1.91 24 10.3 221 2.8
Sand 5,800 $1.51 24 10.3 221 2.3
0.5
Large Sand 5,800 $2.50 24 21.3 389 3.2
Low Density Ceramic-Sand 5,800 $2.64 24 12.1 258 3.7
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Time (Years)

10
 Frac Design Economic Forecasts

1 Year NPV Fracturing Cost Total Proppant Fluid Volume Trk Loads 1 Y Cum Gas
Case
(Million $) (Million $) (Million lb) (Thousand BBL) Proppant (BCF)
Low Density Ceramic-Sand $7.92 $2.64 12.1 258 270 3.7
Large Sand $6.69 $2.50 21.3 389 473 3.2
Resin Coated Sand-Sand $6.20 $1.91 10.3 221 229 2.8
Sand $5.03 $1.51 10.3 221 229 2.3

11
2 Well Utica Pad, Belmont Co. OH - TVD 8,600 ft.
o Each Well used Different Frac Designs

o Large volume 100% sand

o Smaller volume 100% ceramic

o Frac Stage Cost Neutral

Parameter (Unit) Ceramic Sand

Horizontal Length (ft) 6,840 9,110
# of Stages 27 38
# of Clusters/Stage 5 5
Cluster Spacing (ft) 48 48
Total Proppant (MMlb) 5.6 17.0
Total Fluid (Mbbl) 143 258
2017 Frac Cost (MM$) 2.6 3.5
Well Production Comparison
Fracture Modeling
Sand Frac 100 Day Production History Match
Frac Type Sand Ceramic
Sand Well
Frac Stages 38 27
Leff – 188’ 𝑘𝑝𝑟𝑜𝑝∗𝑤𝑓
Fcd = 2.1 𝐹𝑐𝑑 =
Prop Wt (lb)/Stage 450,000 200,000
𝜋∗𝑘∗𝐿
Leff Frac Vol (bbl)/Stage 8,100 5,400

Fractures/Stage 5 5
Lprop
Created Half Length 469 342
(ft)
Ceramic Well
Prop HalfLLength (ft) 447 332
Ceramic Frac eff– 206’
Prop Height (ft) 212 202
𝑘𝑝𝑟𝑜𝑝
103 ∗𝑤𝑓 158
𝐿
Frac Conductivity
(md-ft) 𝑒𝑓𝑓 ≈
Fcd = 3.6 𝜋∗𝑘∗𝐹𝑐𝑑

Leff

Lprop
Fracture Effectiveness over Time
230 Day Production History Matches

Sand Ceramic
Leff – 140 ft Leff – 190 ft

100 day Cum Gas

Formation Pressure 230 day Est
100 Day 100 day Est Stress 230 Day 230 Day Change Change per Frac
Frac Type Perm Gradient Stress On
Leff (ft) md-ft On Prop Leff (ft) md-ft Leff md-ft Stage
(mD) (psi/ft) Prop (psi)
(psi) (MCF)
Sand 0.0087 0.8 188 11 3,383 140 2.25 -26% -80% 4,517 56,157
Ceramic 0.0087 0.8 206 20 3,745 190 5.00 -8% -75% 4,985 78,351

40% More
Proppant Characteristics Comparison
Laboratory Conductivity Measurements at 1 lb/ft2 Concentration

Day 0 Day 100 Day 230

Fc Ratio≈1.5
Conductivity (md-ft)

Fc Ratio≈2.1

Fc Ratio≈2.8

Stress (psi)
Utica Operator Perspective

• Consol Energy, September 2016, SPE 184078, Dry Utica Proppant and Frac Fluid Design
Optimization; “for the ceramic well to be cost effective, between 20% to 30% uplift in
production is needed to justify the incremental capex in todays markets. Based on the
current production data and BHFP, the ceramic well is on the path to reaching those
expectations.”
• EQT, February 2017, World Oil Shaletech; “We had a view that maybe sand would
work, and at the time would be significantly cheaper than ceramics, so we switched
over,” says new CEO Steven T. Schlotterbeck. “Those next couple of wells were
significant underperformers from the Scotts Run. And then we switched back to
ceramics for the last couple of wells, and they were significantly better than the wells
with sand. Those two wells have gotten us much closer to the target recoveries that
we think we need. Our current plans is to use ceramics for all wells in the future.”
 5 Well Eagle Ford Pad, McMullen Co. TX - TVD 10,600 ft.
Fracture Efficiency (FE) = Effective Area (Aeff)/Propped Area (Aprop)
• Significant fracture
inefficiency.
• Apparent fracture
degradation with
aggressive production
drawdown.
• Data indicates that
there is a positive
relationship between
fracture conductivity
and efficiency.

20
 3 Well Eagle Ford Pad, McMullen Co. TX - TVD 11,600 ft.
• Significant fracture Fracture Efficiency (FE) = Effective Area (Aeff)/Propped Area (Aprop)

inefficiency
• Apparent fracture
degradation with
aggressive production
drawdown.
• Data indicates that there is
a positive relationship
between fracture
conductivity and
efficiency.

21
 Summary
 All of the cases presented indicate significant hydraulic fracture inefficiency.
 Due to stranding of large portions of the propped fracture area which consequently do not
contribute to well performance.
 Increasing fracture conductivity appears to mitigate this issue resulting in improved
fracture effectiveness, greater effective frac length and area.
 Proppant placement difficulties which reduce cluster efficiency, proppant and
treatment volumes placed; decrease fracture effectiveness and well
production.
 These issues can be caused by formation and/or completion/frac design issues.
 This data indicates that a hydraulic fracture’s effectiveness degrades over time.
It was necessary to incorporate fracture degradation to match the production
performance of these wells.
 Pressure drawdown due to production which increases the stress on proppants appears to
reduce fracture conductivity and effective fracture area.
22
Thank You!