American Oil and Gas Reporter - April 2021 - 47

pages 47-55_Layout 1 4/8/2021 2:37 PM Page 47

SpecialReport: Frac Tech
Following the V1 frac job, the H1 experienced a significant
production uplift. This is particularly notable because of V1's
distance from H1 and the fact that it was fractured with a small
(18,000 pounds of ceramic 30/50-mesh proppant) treatment. It
is very likely that the uplift was caused by production from
rock around the hydraulically fractured V1 given:
· A strong frac hit was seen at H1 while fracturing the V1;
· The second V1 DFIT showed depletion of pore pressure
and stress (even though the V1 was never produced); and
· Similar Bakken wells that undergo MDD injection do not
see a production uplift, ruling out the alternative possibility that
MDD injection caused the H1's uplift.
Integrated Simulator
The study used a fully integrated hydraulic fracturing/reservoir
simulator. The simulation was set up with nearly identical parameters to a previous simulation that had been used to match
10 years of production data in the H1. The new simulation
results closely matched the original simulation results, except
for a moderate deviation in water production. This discrepancy
could have been resolved with minor changes to the relative
permeability curves, but the same curves were used to maintain
consistency with the earlier modeling work.
Following the H1's 12-year production period, the simulation
modeled the first V1 DFIT, the MDD injection in the H1, the
hydraulic fracturing treatment in V1, the V2 DFIT and the

second V1 DFIT. The simulation was history matched to a
broad set of data, including:
· Production from H1;
· The pressure transient response in the three DFITs;
· The pressure trend in the H1 during MDD injection and
during injections in the V1 and V2;
· Fracture geometry inferred from the microseismicity; and
· The observed production uplift in the H1 from the
fracturing of V1.
Static model properties (permeability, porosity, saturation,
Young's modulus, Poisson's ratio and stress) were populated
with the same 18-layer property model that had been used in
the previous simulation. The integrated simulator uses three
nonconforming meshes: matrix, wellbore and fracture. The
matrix mesh is rectilinear, with local refinement around the V1
fracture azimuth. Wellbore elements are line segments of equal
length. Fractures are meshed using square elements of equal
height and length. Fracture element aperture is recalculated
constantly in each element as a function of proppant concentration
and net stress.
The H1 fractures were created using " pre-existing fractures "
in the model, which allows it to exactly replicate the fracture
geometry used in the prior modeling with one exception: the
fracture at the H1, along strike with the V1, was shortened to
weaken the frac hit that occurred during the first V1 DFIT.
While it would have been possible to simulate the H1 hydraulic

FIGURE 1
Pressure and Conductivity Distribution Before First V1 DFIT

Fracture conductivity (md-ft)

Pressure (psi)
1040

2701

4363

11 y, 52 d, 8 h, 38 m, 24 s

6024

7686

0.1000

0.5623

3.162

17.78

100.0

Stretch: Horiz = 5, Direction = 90 deg, Vert = 1

Note: To improve visibility, images have been stretched five times in the direction perpendicular to fractures.

APRIL 2021 47

American Oil and Gas Reporter - April 2021

Table of Contents for the Digital Edition of American Oil and Gas Reporter - April 2021