(RYDER SCOTT COMPANY PETROLEUM CONSULTANTS LETTERHEAD)
EXHIBIT 99.3
February 20, 2003
Anadarko Petroleum Corporation
1201 Lake Robbins Drive
The Woodlands, Texas 77380
RE: Proved Reserves Status
Marco Polo Discovery
Gentlemen:
Ryder Scott was asked by Anadarko Petroleum Corporation to perform an
updated review of your internal oil and gas reserves studies of their Marco Polo
Field (Green Canyon Block 608) located in federal waters offshore Louisiana.
Ryder Scott's initial review of Marco Polo occurred in August 2002.
SCOPE OF ASSIGNMENT
The question Anadarko has asked is whether the technical data that was
gathered and then used for these reserve studies is sufficient in quality and
quantity to support an SEC Proved reserves classification. Our opinion, based
upon our reserve work in over 2,000 Gulf of Mexico (GOM) blocks, is that the
logs, cores, seismic and MDT data conclusively indicate high quality reservoir
rocks and that this aggregated and complementary data is sufficient to form a
"compelling case" to classify Marco Polo as having Proved reserves. Below is an
outline of our general methodology for reserve analysis in the Gulf Coast / Gulf
of Mexico and how the data in Marco Polo Field compares to our GOM database.
MARCO POLO DISCOVERY
DEEPWATER GULF OF MEXICO
ANADARKO PETROLEUM CORPORATION
Ryder Scott Company review - August 2002
RYDER SCOTT COMPANY REVIEW TEAM:
Ronald Harrell - CEO
John Hodgin - Exec VP & Chief Geologist
Joe Magoto - Sr. VP Reservoir Engineering & Head of Management Advisory
services
Ryder Scott Company review - February 12, 2003
RYDER SCOTT COMPANY REVIEW TEAM:
Joe Magoto - Sr. VP Reservoir Engineering & Head of Management Advisory
services
Anadarko Petroleum Corporation
February 20, 2003
Page 2
GENERAL DISCUSSION ON RYDER SCOTT COMPANY RESERVES BOOKING
Key parameters used for typical Ryder Scott Company's (RSC) Gulf of Mexico (GOM)
Analysis
1) Porosity from Logs
2) Porosity from side-wall Cores
3) Porosity & Permeability from whole Cores
4) Water saturation from open hole resistivity, Neutron-density, Gamma
ray & SP logs
5) Hydrocarbon PVT fluid analysis obtained from multiple MDT's
6) Reservoir continuity from Log correlations
7) Reservoir continuity from MDT pressure analysis
8) Recovery Factors based on Aquifer size
9) Comparison of all above and to RSC data base in Miocene Gulf Coast
sands
Ryder Scott Company approach to above parameters in approximate order of
importance to the analysis.
1) Primary tool used - Log Analysis
a) Does the reservoir meet Porosity requirements for GOM based upon
RSC data base
b) Does the reservoir meet Salt Water Saturation cutoff
requirements for GOM
2) Core Analysis - Side-wall cores calibration of Porosity to Logs
3) Core Analysis - Whole core sometimes taken for increased
certainty-more important in lower quality, higher shale content rock
than present in subject field.
4) MDT tests - how we use the data:
a) Fluid samples for Hydrocarbon type and Oil quality including
GOR.
b) Bottom hole pressure measurements
c) Multiple MDT pressures taken to establish pressure gradients
with reservoir depths
1) Used between wells to establish reservoir continuity between
updip and downdip wells
2) Used to estimate Oil/Water contacts in conjunction with
Seismic data
5) DST or Conventional test through perforations after well
completed-used in GOM primarily for Facility design or in shallow
water (~ 50') where exploratory wells are saved to become producing
wells. In that environment, tests have minimal additional costs or
safety / EPA issues that are overriding problems in normal GOM shelf
waters and especially deep water (> 400 meters).
SPECIFIC DISCUSSION OF ANADARKO'S MARCO POLO FIELD IN GULF OF MEXICO
<Table>
<Caption>
Status of development @ August 2002: Status of Development @ February 2003:
- ------------------------------------ --------------------------------------
4 Expendable open hole logged wells 4 Expendable open hole logged wells
4 Development wells 6 Development wells for production
</Table>
Anadarko Petroleum Corporation
February 20, 2003
Page 3
Ryder Scott Company performed a limited review of Anadarko's
Original-Oil-in-Place (OOIP). We had previously concluded that the structures
are sufficiently well defined by the 4 Expendable wells and the 3-D Seismic
data, and now in February 2003, 6 development wells have defined additional
reserves. However, we have not performed an independent estimate of the pay
counts nor constructed isopach maps. The sands are massive and clean enough that
a visual inspection of the logs is sufficient for us to audit Anadarko's
porosity and water saturations. After an additional audit of Anadarko's log
analysis, we concluded that if RSC independently analyzed this field for
official reserves certification, our porosity values would be approximately the
same average as Anadarko's, which was 30 percent. This 30 percent average
porosity (and permeability confirmed by whole cores) is sufficiently high that
sands 50 to 100 feet thick typically flow 10,000 to 15,000 bopd. However
Anadarko has used only 2,900 bopd per well for their SEC reserve forecast. It is
our opinion, based upon both Darcy's Law flow rate calculations and analogies to
other GOM fields, that the actual producing rates will be much higher.
We have taken Anadarko's reservoir parameters (see attached) and have
independently calculated flow rates of 4,000 bopd for 50 foot sands and up to
8,000 bopd for 100 foot +/- sands using only 10 percent bottom hole pressure
draw-down. Additionally based upon analogies, we would expect actual, possibly
commingled flow rates to exceed 10,000 bopd. As part of our GOM database, we
have had occasion to audit recently 2 of the nearest offset discoveries, which
happen to be located in the same geologic basin. Those discoveries have
porosities generally around 26 to 28 percent. Wells in those 2 discoveries are
forecasted to flow at 15,000 to 17,000 bopd due to tubing restrictions, but
potentially calculate as high as 30,000 bopd without restriction. Both of those
discoveries have lower GOR's and less mobile oil than Marco Polo.
CONCLUSIONS
The 1978 SEC reserves definitions clearly state that "Reservoirs are
considered proved if economic producibility is supported by either actual
production or conclusive formation test". Other parts of the definitions relate
to fluid contacts, pricing, costs and development status. Anadarko has not
conducted any traditional flow tests in their evaluation of the Marco Polo Field
but has acquired all of the information obtainable from a flow test through
other more safe, cost effective and environmentally sensitive means. All of
these techniques, which include advanced well log suites, core analyses and
down-hole sampling and flow testing, represent proven technologies developed or
advanced beyond those available in 1978.
After a careful review of the data provided by Anadarko, both raw and
interpreted, we conclude that you have acquired the necessary information to
support economic producibility without conducting the type of flow testing
envisioned in 1978 and that you have effectively met the SEC reserves
definitions in that regard. Attached is our comparison of multiple MDT testing
versus a single conventional flow test that was presented at the October 2002
SPEE conference on SEC reserves definitions. From this table it can be clearly
deduced there is an overwhelming amount of additional data gained from multiple
MDT testing versus a single conventional flow test. It is our opinion that an
extensive MDT testing program provides a more compelling case of reasonable
certainty for proved reserves than a single conventional flow test.
Furthermore, our internal industry survey in October 2002 indicated only 6
out of 45 fields or 13 percent have been traditionally flow tested in the
deepwater Gulf of Mexico. It is also our experience that the practice of
conducting a traditional flow test in the U.S. Gulf Coast and the Gulf of Mexico
is limited to those circumstances where down-hole data are (1) not sufficiently
conclusive to meet the "economic producibility" required or (2) where a level of
refinement is needed in designing facilities to meet the expected rates of flow.
Anadarko Petroleum Corporation
February 20, 2003
Page 4
CONSENT
Anadarko is authorized to incorporate this letter in its required financial
reporting or in communications with the U.S. SEC as long as the entire letter
report is used. We would grant our consent for Anadarko to use selected portions
of this report only upon specific review of the passages sought to be included.
This consent is being granted for a limited time period not to exceed
approximately 90 days from the effective date of this report.
RYDER SCOTT COMPANY, L.P.
/s/ Ronald Harrell
-----------------------------------
Ronald Harrell, P.E.
Chairman and CEO
DRH/sw
MARCO POLO DISCOVERY
DEEPWATER GULF OF MEXICO
ANADARKO PETROLEUM CORPORATION
RYDER SCOTT COMPANY FLOW RATE CALCULATIONS - AUGUST 2002
<Table>
ROCK DATA
Porosity 29.9 %
Permeability - k 215 md
Water saturation - Sw 32.9 %
RESERVOIR DATA
Pressure - BHP initial 7433 psia
Temperature - reservoir 121 F
Pay -average net feet 113 feet
FLUID PROPERTIES
Oil gravity - API 32
Solution Gas GOR 920 scf / bbl
viscosity 1.2 cp
bubble point 2750 #
PRODUCING CONDITIONS
Drawdown ~ 10% 750
skin factor gravel-packed 2
-----
CALCULATED INITIAL RATE 8,200 bopd
-----
</Table>
MDT MULTIPLE TESTS VS. ACTUAL PERFORATION / DST TESTS
FEBRUARY 18, 2003
<Table>
<Caption>
MDT TOOL
-----------------------------------------------------------------
TYPE OF MEASUREMENT ADVANTAGES DISADVANTAGES
------------------- ------------------------------- ------------------------------
Drawdown & commercial rate multiple depth contributions limited sample size
measured
Permeability measurements multiple depth mobilities (k) shorter / limited time of test
can capture foot by foot in the past with older tools
hetrogenity (see Future Trends)
Pressure measurements multiple depth tests reveal
actual fluid gradients, I.e. gas
vs. Oil vs. Water (validates
some PVT data)
fluid gradients at depth can
define and confirm Gas/water or
Oil/Water contacts
Industry now determining Cond
yields from press gradients
EXTENT OF FLOW DATA
Multiple zone tests In one pass, can take 50 +/-
tests which usually covers many
individual pays
Multiple well tests across the MDT costs low enough to run tool
structure in most, if not all wells.
Provides continuity between
wells validating the size of
area that is productive
FUTURE TRENDS
Maxi-MDT tool and bypass flow
before fluid sample capture
approximate longer time actual
prod tests
<Caption>
PROD. TEST - PERF OR DST
-------------------------------------------------------------------------
TYPE OF MEASUREMENT ADVANTAGES DISADVANTAGES
------------------- --------------------------------- -------------------------------------
Drawdown & commercial rate fluid to surface producing DST test not representative of
charteristics measured completed well with typical gravel
pack but skin damage
Permeability measurements longer drawdown period (perceived) only average Permeability (k)
measured over the perforation / DST
test interval
Pressure measurements long term buildups could multiple $$$ million dollar costs &
conceptually be possible, but at EPA / MMS environmental risks
prohibitively high costs & lack
of storage (especially for Oil)
EXTENT OF FLOW DATA
Multiple zone tests Good comparison of actual Costs are so high for one DST, or
flowrates between different especially perforation tests, that
quality rocks only 1 or 2 zones are usually
tested.
Multiple well tests across the -------------- High costs that typically range
structure from $5 - 15MM limits this type of
test to almost always a single well.
FUTURE TRENDS
-------------- EPA requirements and increased
pressures on Oil industry in
general increasing rapidly, making
future risks of Oil spills or
incomplete burn even more
politically difficult.
</Table>