Hydrocarbon Migration and Trapping in Unconventional Plays*

Zhiyong He1 and Daniel Xia2

Search and Discovery Article #10968 (2017)**

Posted July 3, 2017

*Adapted from oral presentation given at 2017 AAPG Annual Convention & Exhibition, Houston, Texas, April 2-5, 2017
**Datapages © 2017 Serial rights given by author. For all other rights contact author directly.
1
ZetaWare, Inc., Sugar Land, Texas ([email protected])
2
Petralloc Geochemistry, Sugar Land, Texas

Abstract

We will show in this presentation a collection of evidence for hydrocarbon migration at various scales in unconventional plays. This includes
spatial variations in fluid properties (API, GOR, H2S) which conform to structure, faults, and depositional facies boundaries, production of
higher maturity fluids from low maturity, or immature strata, and production of different fluids from adjacent, or interlaced zones. Mixing
and/or interlacing of high maturity and low maturity fluids due to differential migration often results in higher saturation pressure. This can lead
to excessive gas production in an oil play or liquids drop out near the well bore in a gas play. Migration distances are estimated to be one to
tens of kilometer laterally and hundreds of meters vertically.

Traditional sweet spot predictions have been mainly focused on source rock maturity. This may have led to incorrect predictions due to lack of
consideration of migration as a factor. Companies who acquired leases based on maturity estimates alone have sometimes found themselves
producing more gas than expected, and sometimes even dry gas from rocks currently in the “oil window”. We will also demonstrate that seals
are very important and are a required element for unconventional plays to be successful. Homogeneous shales without sealing intervals will not
retain sufficient hydrocarbons, even where maturity, TOC, porosity, and clay content, etc. are favorable. Exploration for unconventional plays
therefore also needs to investigate structure, depositional environments and potential sealing units. In other words, the same evaluation criteria
apply to conventional and unconventional plays, just at smaller scales in the latter case. In our examples, we analyzed data from ~ 2 million
wells to recognize and demonstrate regional migration patterns in several basins.
Understanding Migration and
Trapping in Unconventional Plays
through 3D Geospatial Data Analytics
2017 AAPG ACE, Houston, Texas, April 2-5, 2017

Zhiyong He
ZetaWare, Inc.
Daniel Xia
Petralloc Geochemistry

© Zetaware Inc.
Introduction
• Seals are required for trapping hydrocarbons, in both conventional and
unconventional plays.
• Capillary displacement pressure of the seals control HC saturation in the
reservoirs, as well as pressure, probably the two most important parameters in
shale plays.
• This study utilized “Big data” (from ~2 million wells) to geospatially visualize and
analyze the regional petroleum system behavior - migration patterns, regional
seals, phase separation etc., ie “Geospatial Petroleum System Analytics - GPSA”
• 3D visualization of such data in the context of the geological framework is
essential as many of the important patterns are stratigraphic and follows
geological controls such as faults and facies boundaries.
• The results led to the understanding of several aspects of petroleum system in
the region, previous unrecognized and or unappreciated.

© Zetaware Inc.
 100000 0.002
Seals Control HC Saturation & Pressure
Capillary Pressure

Migration
Pd

Primary
Austin 10000
B 0.02
Chalk

Pore throat diameter (μm)

Eagle ford

Saturation Pd

MICP (psi)
1000 0.2
A
Buda

26% Saturation 73% Saturation

A B
900 psi MICP 13,000 psi MICP 100 2
56 psi Po-Pw 800 psi Po-Pw
HC generation causes saturation and in turn capillary pressure (Po-
Pw) to increase. Primary migration occurs when capillary pressure
reaches the displacement pressure (Pd) of the seal. Higher Pd means © Zetaware Inc.
10 20
higher retained saturation, as well as higher reservoir pressure. 100 80 60 40 20 0
This is a generalized/idealized model – most systems are more Non-wetting phase saturation
heterogeneous but the principles are the same.
 Midland HC Generation Pore volume vs.
Basin (mass) Generated HC vol.
Migration in the Stacked 0 1 At Reservoir PT
Shale Plays:

Kinetic & PVT modeling

At higher maturity, generated HC volume
at reservoir PT conditions is 3-5 times the
available pore volume as HC density
decreases from 0.6-0.8 g/cc (low to Pore Vol.
moderate GOR oil) to 0.1 to 0.2 g/cc (for φ = 4-8%
gas condensates). The excess volume of
HC is forced to migrate up stratigraphy
and/or up dip.
Generated
For example, the Wolfcamp may contain a / Cracked
mixed fluid of local generated oil and gas HC Vol.
condensate from deeper source rocks
(Barnett, Woodford etc.).

Excess
Spraberry/Bone Spring/Clear fork

Excess
reservoirs are filled mostly with migrated
fluid.

© Zetaware Inc.
Effects of Mixing Oil with Gas Condensate in Subsurface
When locally generated low maturity oil is mixed with equal volume of migrated gas condensate in subsurface conditions,
GOR more than doubles, but oil API gravity increases only by about 10%. Biomarkers & ratios are not affected at all.
Incompatible GOR and API gravity is a sign of mixing. This can result in higher bubble point pressure and cause in-situ phase
separation – and production of gas in oil maturity formations.

GOR API Gravity BM Maturity

(scf/bbl) % VRE
Local Oil 1000 35 0.8
Gas condensate 10000 55 1.2
Mixed: 2700 39 0.8

Observations:
• API gravity reflects source rock maturity trend
better than GOR.
• Produced GOR typically higher than maturity
model “predicts”.
© Zetaware Inc.
 Reservoir and Seal Pairs for Major US Liquid Plays
Eagle Ford Bakken Woodford Midland Delaware

Clear fork
Spraberry
Wolfcamp
Typical shale plays are interbedded limestone,
shale, and siltstones. Pay (reservoir) can be
within, above, below or in between the
source intervals.

Sources: Pioneer, Matador, Continental Resources, ShaleExperts.com, Ajit K. Sahoo et al © Zetaware Inc.
 Top and Bottom Seals of Major Shale Plays
Reservoir Top Seal Bottom Seal
Eagle Ford shale Eagle Ford marl Austin Chalk Buda limestone

Marcellus shale Siliceous shale Tully limestone Onondaga limestone

Middle Bakken, Tree forks

Bakken Lodgepole limestone Nisku limestone
siltstones.
Woodford Siliceous shale Mississippi/Osage limestone Hunton limestone
Meramec Calcareous shale Chester shale/lime Osage limestone
Barnett Siliceous shale Marble falls/Forestburg limestone Chappel/Viola limestone

Haynesville Marls Interbedded limestones Smackover lime

Niobrara Marls/sands/chalk Interbedded shales/Chalk Fort Hays limestone

Spraberry, Bone
Sandstone & siltstones Interbedded limestone Interbedded limestone
Springs
Wolfcamp Marls, silts Interbedded limestone Interbedded limestone

The general observation is that majority of working shale plays are marl/shale/silt reservoirs sandwiched in between
limestone units. Limestones are likely (not necessarily exclusively) the master seals for these plays?
© Zetaware Inc.
Examples of Lateral Seals/Trapping
Midland Basin

Wolfcamp / Spraberry
Pinch outs at shelf edge

Williston Basin
Pearsall field

Top Oil Window

A geological definition of unconventional plays may be simply “stratigraphic Woodford shale isopach, Anadarko basin
traps”, that require fracking to be economical? Source: Continental resources
© Zetaware Inc.
Faults Provide Lateral Trapping In Eagle Ford ?
 Sweet spots down dip from and in between
major faults.
 Migration shadows up dip from faults
A
 Stepping up stratigraphy up dip from faults
 “Gas caps” observed
 Better IP rates on down thrown side of faults
(Drilling info)
A’

A Karnes Trough A’
0

8 Surf ace
Wilcox
10
Navarro
12 Austin chalk
10 miles
Eagle Ford
14

16
0 50 100 150 200 250 300 350 400

Faults from Hammes et al. 2016 Distance in f eet (1000)

© Zetaware Inc.
 Geological Controls on Fluid Migration

(b) (a)

Top Oil window

Top Gas window

Oil production above oil window (a) and gas production in oil window (b), with higher than expected
GORs for the measured or modeled maturity. Expected GOR (<500 scf/bbl) at the low maturity is much
lower than observed (1000-2000 scf/bbl).

© Zetaware Inc.
 Geological Controls on Fluid Migration

Good top seal in deeper basin: The silty reservoirs of the Spraberry and the overlying
deep water calcareous facies present a good contrast for sealing/grapping mechanism.
The lack of structure relief may also be a factor. Wolfcamp zones may be productive
where it is overlain by limestone zones.

© Zetaware Inc.
 Geological Controls on Fluid Migration

Significant vertical migration occurs at shelf edge where deep reservoirs or seals may be
truncated due to facies change and/or faults.

© Zetaware Inc.
 Geological Controls on Fluid Migration

Phase separating due to fluid mixing? High maturity gas condensate may have migrated
into shallow reservoirs to mix with locally generated oil. Pressure is below bubble point
pressure for the mixture and phase separation occurs and forms “gas cap”. Such gassy
areas may expand downdip over time as production further draws down pressure.

© Zetaware Inc.
More Evidence of Upward Migration & Mixing
-5

W E Surface
Trassic-Cretaceous
Rustler-Salado-Castile
Tansill Yates Bell Canyon
San Andres Bushy Canyon
0 Clearfork 1st Bone Spring
Spraberry 2nd Bone Spring
Wolfcamp
Straw n
Morrow
Barnett

5
Mississippian

GOR3MAverage (sf/bbl)
0-300

300-1000

1000-3000
10
3000-5000

5000-10000

10000-50000
50000-1e+006

15
0 40 80 120 160 200 240 280
Distance in feet (1000)

Vertical migration of gas condensate seem to have affected a significant area in the Midland basin. The
Spraberry (yellow) is in early oil window maturity. The produced fluids seem to have relatively low API
gravity for the high gas oil ratio (5,000 to 20,000 scf/bbl).
© Zetaware Inc.
Long Distance Migration ?
Wolfcamp outcrop  Wolfcamp only matures in the deeper part of Midland
basin.
 Oil may have migrated from Midland basin over to the Fort
Worth basin > 100 miles.
 Production from the Spraberry, Bone Springs, Brushy
Canyon, Avalon formations are mixed with fluids from
Wolfcamp, and/or Woodford.
 “Gas caps” observed in several plays.
 Vertical migration near Permian shelf edge where deep
water reservoirs pinch out against carbonates.
Wolfcamp/Barnett/Eagle Ford depth map

Delaware basin Midland basin Wolfcamp outcrop Fort Worth basin East Texas basin

Wolfcamp
Barnett East Texas Field

100 miles
Bossier/Haynesville

© Zetaware Inc.
 The Natural PVT Model
GOR (scf/bbl) CGR (bbl/mmscf)
100 500 1000 3000 5000/200 100 50 10
0

1) During the burial and generation process, the

depth and pressure increase as maturity, and
gas oil ratio increases. Due to increasing
pressure the fluid is naturally single phase.
Dual Phase
INCREASING PRESSURE

2) Pressure drop due to uplift or migration into

shallow reservoirs may cause fluid to become
dual phase (gas exsolves from oil to form
Single Phase Single Phase vapor phase at reservoir pressure).

3) Mixing high maturity gas with low maturity

Uplift or oil can increase bubble point pressure to
Migration above reservoir pressure, giving a similar
effect.

Black Oil Volatile Oil Liquid Rich Gas Dry Gas

 The Natural PVT Model
GOR (scf/bbl) CGR (bbl/mmscf)
100 500 1000 3000 5000/200 100 50 10
0

1) Separate gas and oil phase causes higher gas

production and GOR increases over time as
pressure declines.

Dual Phase 2) Regions where reservoir is near bubble point

INCREASING PRESSURE

may initially produce oil, but soon becomes

gassier.

Single Phase Single Phase 3) Volatile oil region with significant erosion, or
gas migration into low maturity oil = higher
risk for phase separation.
Uplift or
Migration 4) Low maturity oil window and gas window
(CGR>100 bbl/mmcf) are less affected by
phase issues.

Black Oil Volatile Oil Liquid Rich Gas Dry Gas

Intermediate Seal Becomes Effective During Production
Capillary pressure
Intermediate seals may
“become” sealing during
production once pressure
drops below its displacement
pressure (Pd).

(a) Initial pressure at Pd of

main seals. Two reservoirs
Now are in pressure
“Sealing” communication.

(b) During production,

pressure drops below Pd of
intermediate seal. Oil
(a) (b) phase now snapped and
disconnected.© Zetaware Inc.
Some Important Observations
• Several of the stacked systems seem to show a pattern where if the upper reservoir performs well, the lower
reservoir is less productive, and vice versa, the Austin Chalk vs the Eagle Ford, Spraberry / Bone Spring vs the
Wolfcamp, and Clear Fork vs the Wolfberry system. In one example, the Chester shale, which is recognized as a
seal for the STACK play in Anadarko basin, is eroded to the north where water oil ratio is higher in the underlying
reservoirs.
• The sweet spots of the younger play tend to be offset to the up-dip direction of the lower play – indicating
migration up stratigraphy where the lower reservoir pinches out or lacks sealing capacity. Younger conventional
reservoirs are mainly found outside of the main kitchen in the Permian basin. Such relationships are very clear
when production data are visualized in 3D in the stratigraphic context.
• Significant conventional accumulations may be found up to 300 miles up dip from the source kitchen of the
unconventional reservoirs, such as the oils in Kansas and Nebraska that have migrated from the Anadarko basin
in Southern Oklahoma, and oil accumulations near Fort worth in the Wolfcamp formation that may have
migrated from the Midland basin.
• Along major trends/fairways of long distance migration, the near kitchen plays are more gassy – perhaps
because the increased efficiency allows oil to migrate further.
• Migration and/or uplift may cause reservoir fluids to go below bubble point pressure, resulting dual phase
reservoirs may produce fluids with significantly higher gas oil ratios.

© Zetaware Inc.
Conclusions:
• It is important to also consider seals in searching for the next unconventional play. Capillary seals
help retain saturation in the reservoirs.
• Seals are simply rock layers with smaller pore throat size than the reservoirs. For conventional
reservoirs, seals are typically shales. But for shale reservoirs, they may be typically limestones and
may often be a stack of interbedded reservoirs and seals, at different scales.
• Limestone formations seem to be regional top and bottom seals for most productive shale reservoirs.
Some plays may be described as self-sealing as the reservoirs are highly stacked sequence of
silt/shale/marl/limestones. Shales and other rock types can be seals as long as the pore throat size is
smaller than that of the reservoir.
• Point of likely misunderstanding: Seals do not imply no migration at all. Capillary seals are exactly like
pressure valves, bleeding off excess fluid while retaining a full capacity of the reservoirs. Maximum HC
Saturation depends on the difference in capillary displacement pressure of the seal and the reservoir
pore size distribution.
• Lateral seals for shale reservoirs are typically facies changes, and sometimes faults. Structure
relief/focusing can also be important. Viscosity at low maturity front may also be effective (?).
• Migration is observed at different scales. Therefore, unconventional play targets do not have to be
the source rock. Potential reservoirs can be above, below or up dip from the mature source rock, or
highly interbedded reservoir/seal packages. The potential reservoirs may not be limited to areas
where the source is mature.
THANK YOU!
Thanks to the following people and others for the many discussions on the
subject:
Harris Cander BP
Michael Cheng BHPBilliton
Steve Crews Apache
Bob Davis Mubadala Petroleum
Chanrui Gong Apache corporation
Jie Huang Northlight Resource Analytics LLC
Dan Jarvie Worldwide Geochem
Eric Michael ConocoPhillips
John Pantano ZetaWare, Inc.
Andrew Murray Murray Partners
Andy Pepper TIPS

Zhiyong He
ZetaWare, Inc.
© Zetaware Inc.