DNV Certification, Inc. …

                                DNV Certification, Inc.


                              Chevron Corporation

           2004-2006 Worldwide Greenhouse Gas Inventory
          Greenhouse Gas Inventory Unqualified Opinion Statement

                                   October 26, 2007

Introduction

Chevron Corporation (Chevron) engaged Det Norske Veritas Certification, Inc (DNV) to
evaluate its carbon dioxide, methane and nitrous oxide (GHG) emissions data for the
years 2004-2006. Our review covered Chevron's total direct and indirect GHG emissions
on an equity and an operated basis, as defined by Chevron in its document entitled:

  ESTABLISHING THE EMISSIONS INVENTORY FOR CHEVRONTEXACO
              INVENTORY PROTOCOL FOR 2003-2005
                       (December 8, 2004)

The scope of this unqualified opinion statement excludes ChevronPhillips, Dynegy,
Chevron Australia Limited, and non-operated joint ventures (NOJV's), partially-owned
assets over which Chevron does not have full authority to implement its HE&S policies.

This report is made solely to Chevron in accordance with the terms of our engagement.
The DNV work has been undertaken so that DNV might state to Chevron those matters
we have been engaged to state in this report and for no other purpose. To the fullest
extent permitted by law, we do not accept or assume responsibility to anyone other than
Chevron, for our work, for this report, or for the conclusions DNV has reached.

The initial verification is based on the information made available to DNV and Chevron's
own requirements set forth in its request for proposal. DNV cannot guarantee the
accuracy or correctness of this information. Hence, DNV cannot be held liable by any
party for decisions made or not made based on this report.

The third-party verification process for Chevron's greenhouse gas emissions inventory
aimed to systematically evaluate:
   ·   the quality of the Chevron greenhouse gas emissions inventory protocol;
   ·   the execution of the inventory process; and
   ·   the quality of the underlying data gathered and the inventory numbers generated.
                                 DNV Certification, Inc.


The purpose of the verification process was to ensure that the level of quality for the
inventory can be expressed in the context of defined criteria:

   ·   Fully suitable for public reporting purposes; and
   ·   With a view towards defining areas for improvement.

Respective Responsibilities of Chevron and DNV

Chevron's management and staff was responsible for the preparation of the underlying
GHG emissions data based on the requirements set out within Chevron's Emissions
Inventory Protocol and for the development and maintenance of this Protocol. The
responsibility of DNV was to express our opinion with regards to the quality of
Chevron's GHG 2004-2006 emissions data.

Basis of DNV's work

At this time there are no universally accepted standards on GHG reporting. Guidance on
reporting GHG emissions for the oil and gas industry is available from the International
Petroleum Industry Environmental Conservation Association (IPIECA), the American
Petroleum Institute (API), and the International Association of Oil and Gas Producers
(OGP). The Chevron "Emissions Inventory Protocol" was developed based on this
guidance and has formed the primary basis of DNV's work.

Chevron Data Gathering Process

Chevron uses the SANGEATM system to compile inventories of its greenhouse gas
emissions. Data coordinators from across Chevron's worldwide operations enter fuel
consumption and other data into the ExcelTM-based SANGEATM software. The software
quantifies greenhouse gas emissions, and this information is sent to corporate
headquarters to allow for the development of a company-wide inventory.

DNV GHG Verification Methodology

The actual site entities (facilities) to be visited within each Chevron operating company
were determined through a dialog between Chevron Corporation and DNV and ISO
14064-3 Validation & Verification Process. The initial number of sites and locations was
based on the outcome of a "Phase I Risk Analyses," relevant costs considerations,
geographic location, and security considerations. DNV visited the 5 top
                                  DNV Certification, Inc.


GHG emitting operational entities (facilities) where it conducted a full data validation.
DNV visited three (3) additional sites to verify emissions estimates, only (5 + 3
approach). Based on the outcome of these audits and further evaluation of
risk/uncertainty, DNV audited a representative sample of the remaining sites within each
operating company (Chevron International Exploration and Production, Chevron North
American Exploration and Production, Global Refining, Global Gas, etc.) for a total of 30
sites. (i.e., 5 top emitters, 3 additional sites, plus 22) If DNV were to identify significant
material discrepancies in this sample set described above, DNV, upon approval by
Chevron, would expand the sample size.

The overall intent of the first round of eight (8) audits (5 + 3) was to assess a
representative sample from each operating company with special attention, as applicable,
on the "Top 5 emission sources". DNV utilized the first 8 site visits (5 + 3) to perform a
"risk assessment" (uncertainty evaluation) which informed the audit
process/methodology and sample approach for the remainder of the 22 sites. That is,
DNV chose the specific operations and locations within each operating company for the
remaining 22 sites based on the risk/uncertainty evaluation derived from the first group of
8.

Top 5-Full Data Validation (Focus on Full Systems and Calculations)

   1) Management Systems and data control (including documentation &
      recordkeeping)
   2) Relevance
   3) Completeness (all material emissions direct and indirect)
   4) Consistency
   5) Transparency
   6) Accuracy
   7) Evaluation execution of the Chevron GHG Emissions Protocol
   8) Assurance data checks Chevron GHG Emissions Reported Years 2004-06

Additional 3-Verify Emissions Estimates, only (Focus on Calculations)

   1) Management Systems and data control (very limited review of documentation
      and recordkeeping)
   2) Accuracy (Select limited sample of calculations to verify)
   3) Focus on all significant, direct process and combustion emission sources, only
      (e.g., process heaters, boilers, turbines, etc.)
   4) Assurance data checks on the above focus areas Chevron GHG Emissions
      Reported Years 2004-06
                                   DNV Certification, Inc.


Final Set 22-Limited Data Validation (Focus on 1 Year's Data set and execution of
the Chevron GHG Emissions Protocol)

    1)   Evaluate execution of Chevron GHG Emissions Protocol
    2)   Limited review of documentation and recordkeeping
    3)   Accuracy (Select limited sample of calculations to verify)
    4)   Focus on a limited sample of direct and indirect emission sources
    5)   Assurance data checks on one Year's data (The intent is to randomly select the
         `Years' to audit ­e.g., Site 1 2005, Site 2 2006, Site 3 2004, Site 5 2004, Site 6
         2006, Site 7 2005-----etc.)

Conclusions

Based on the above process, nothing has come to DNV's attention that causes DNV to
believe that Chevron's 2004-2006 greenhouse gas emissions data are not fairly stated.
Thus, DNV has determined that the Chevron greenhouse gas estimates for the years
2004-2006 are accurate based on a low level of uncertainty1. The GHG estimates are
generally consistent with voluntary reporting practices (e.g. WBCSD/WRI Greenhouse
Gas Protocol and the Voluntary Carbon Standard) and the Chevron Emissions Inventory
Protocol.
1
 Notes:
The DNV verification audits did include an assessment of a limited sample of non-
operated joint ventures. The limited verification activities undertaken by DNV for
NOJV's indicated that the GHG emissions estimates for non-operated joint ventures
generally conformed to the principles of voluntary reporting practices. However, while
DNV did not find material misstatements in the audit sample, there was enough
uncertainty to warrant a qualified, rather than an unqualified, opinion. Additionally, in
some cases DNV could not directly evaluate the accuracy of the emissions data.




                                      Houston, Texas
                                      26 October 2007