American Oil and Gas Reporter - May 2016 - 45

industry understands the high decline
rates in resource plays, and as long as
the reserve analyst is careful, available
decline curve models are well equipped
to handle the characteristic steep early
declines that flatten as wells age.
The question is where does the analyst
draw the line and say there are not enough
data available to extrapolate observed decline trends, and therefore, he must resort
to using type curves? Opinions vary, but
the more information and the longer the
time frame represented, the better. Many
analysts have the greatest confidence in
reserve evaluations that have at least three
years of observed decline trend data. That
may be asking too much in some cases,
but it is clear that there is substantial risk
in projecting decline trends with less than
six months of observed data.
For example, during the first several
weeks of production, wells in resource
plays typically go through a cleanup
process following fracture treatment. They
generally produce a lot of water during
this time, but this early production behavior
normally does not last very long. In many
cases, new wells initially produce high
oil or gas rates for a few days, but do not
decline nearly as rapidly as they will
after cleaning up. Estimating future production based on these early-stage decline
trends can lead to gross inaccuracies.
That brings up another important consideration: In many resource plays, new
wells are choked back deliberately for
several weeks while they are cleaning up
to avoid damaging the formation. The
Haynesville Shale is a classic example.
Operators have found that drawing down
new wells too quickly can damage productivity and limit EURs. With such sensitive formations, operators have learned
to be very careful in managing pressures
by keeping new wells choked back for
some period during early production.
This is problematic because the standard Arps' decline curve method assumes
that wells are produced at a constant bottom-hole pressure. If the BHP is gradually
decreasing with time (before finally reaching a plateau after several months), the
decline trend will be optimistic and lead
to overestimating reserves potential almost
all the time.
The analyst has to be aware of this
situation, but it can be very hard to spot
in public data (in fact, pressure data are
rarely even reported in public filings). If
possible, the production practices of operators in the area should be investigated
to determine whether wells were choked
back for sustained periods.
Even in a case where the analyst is
confident that BHP has plateaued, there
is another potential shortcoming when

using the traditional Arps' decline model
with early-life data before a well stabilizes
and begins producing from its entire
drainage area. The "b factor" hyperbolic
exponent in the Arps' equation tends to
decrease as more data are acquired over
the first several months of a well's productive life. This means the b factor parameter calculated from a best-fit of early
data is often too high, relative to a well's
performance over its full life cycle.
The solution is the modified Arps' decline
model, whereby Arps calculations are used
in combination with a simple exponential
decline rate. I have seen a number of cases
where analysts misapplied the b factor exponent over 40 or 50 years of well life,
almost certainly overestimating reserves.
That is a common error that oil and gas
companies can make in assessing acquisition
prospects, if they are not careful.
As important as it is to get an evaluation
done before the competition, there is never
a reason to sacrifice accuracy for speed.
The reliable technology method is no more
time consuming than alternative techniques.
To assure the highest degree of confidence in reserve projections in resource
plays, operators should seek the bestavailable expertise. That expertise comes
in two forms: advanced analytical and
forecasting software with demonstrated
successful applications, and skilled analysts
with proven track records.
❒

W. JOHN
LEE

W. John Lee is the Chevron faculty
fellow in petroleum engineering at
Texas A&M University. The author of
four textbooks published by the Society
of Petroleum Engineers, Lee is an
SPE honorary member and former
distinguished lecturer. He served as
an academic engineering fellow with
the Securities and Exchange Commission, and was a principal architect
of SEC's modernized reserves reporting
rules. Lee began his career at Exxon,
where he served in reservoir engineering roles for 18 years. He then
joined S.A. Holditch & Associates
Inc., where he retired as a vice president after 19 years. He holds a B.S.,
an M.S. and a Ph.D. in chemical engineering from Georgia Tech.
MAY 2016 45

American Oil and Gas Reporter - May 2016

Table of Contents for the Digital Edition of American Oil and Gas Reporter - May 2016