Improving Emission Inventories in North America

1
Improving Emission Inventories inNorth America

• NERAM V
• October 16, 2006
• William T. Pennell
• NARSTO Management Coordinator

2
Outline

• Emission Inventory Development
• Methods for Evaluation
• Strengths and Weaknesses of North American
  Emission Inventories
• Actions for Addressing Weaknesses
• Costs of Emission Inventory Development and
  Improvement

3
First of All, What Is NARSTO?

• A multi-stakeholder, public-private partnership
  among government, the private sector, and
  academia throughout Canada, Mexico, and the
  United States that collaborates to improve air
  quality management science in North America.
• NARSTO has recently completed an assessment of
  emission inventories for the three countries of
  North America.
• This presentation is based on that assessment.

4
How Are Emission InventoriesConstructed?
E EF A (1 - ER) Where E are the emissions
(e.g., kg/hr or tonnes/yr) EF is the emission
factor (mass emitted per unit process variable) A
is the related process variable (e.g., mass of
fuel consumed ER is an emission reduction factor
(1 gt ER gt 0) But this simple relationship hides
many complexities!
5
How Are Uncertainties Estimated?

• Top-down methods
• Comparison with ambient measurements
• Alternative estimation using different activity
  factors
• Inverse modeling
• Bottom-up methods
• Direct measurements
• Model uncertainty analysis
• Model sensitivity analysis

6
Consequences of Errors?
Consequences of errors in emission
inventories (as well as lack of knowledge of
their uncertainties) can range from wrongly
identifying a pollutant that should be controlled
to overlooking source categories whose control
could result in more cost-effective emission
reductions.
7
An Example Houston, Texas

• Houston is currently in non-attainment for ozone
  NAAQS.
• An initial proposed management plan, based on
  then-current inventories, proposed reducing NOx
  emissions by 90 percent.
• However, a subsequent field experiment discovered
  significant sources of highly reactive VOCs not
  included in the inventory.
• Subsequent modeling indicated a revised strategy
  of VOC reductions coupled with 80 percent NOx
  reductions.
• Economic modeling suggested substantial economic
  savings with the revised strategy (10 years after
  implementation, 65,000 fewer jobs and a regional
  economy 9B smaller with 90 NOx reduction).

8
Strengths and Weaknesses of Current North
American Inventories

• Strengths
• Good understanding of major point sources
• Emissions models have continuously improved
• Importance of natural emissions recognized
• Quantitative estimates of emissions available for
  many source categories
• Improved understanding of relative importance of
  source categories
• Emission trends can be tracked

9
Strengths and Weaknesses of Current North
American Inventories

• Weaknesses
• Significant uncertainties in mobile source
  inventories (e.g., speciation of VOCs, CO
  emissions, trends in NOx)
• Emissions for many important categories are
  uncertain (e.g., fine PM, biogenics, ammonia, air
  toxics, etc.)
• Quality assurance is uneven
• Estimates often based on insufficient sample
  sizes
• Temporal and spatial resolution inadequate for
  modeling
• Speciation profiles out of date
• Inventories not timely
• Differences in methods and standards between
  countries complicate joint air quality management

10
What Is The Case for Improving Emission
Inventories?

• Current inventories may support todays needs,
  but they have significant shortcomings
• As emissions from major sources decline,
  emissions from sources that are more difficult to
  characterize will become more important
• Errors in emission estimates from these sources
  will have increasing consequences
• These consequences can result in ineffective air
  quality management actions that affect the
  credibility of the air quality management system.

11
How Can Needed Improvements Be Achieved?
12
Key Findings And Recommendations

• 1. Focus immediate attention on reducing
  uncertainties for key under-characterized sources
• Fine particles and their precursors
• Toxic and hazardous air pollutants
• Onroad motor vehicles
• Agricultural sources, especially ammonia
• Biogenic sources
• Petrochemical and other industrial facilities
• Offroad mobile sources
• Open biomass burning
• Residential wood combustion
• Paved and unpaved road dust

13
Within This Focus

• Improve Emission Inventory Speciation Estimates
• Improve Existing and Develop New Emission
  Inventory Tools
• Quantify and Report Uncertainty
• Increase Emission Inventory Compatibility and
  Comparability
• Improve User Accessibility
• Improve Timeliness
• Assess and Improve Emission Projections

14
Implementation Issues

• To achieve these recommendations
• They must be addressed over the next 10 years
• However, significant additional resources -- on
  the order of 2-10 times current expenditures --
  will be required
• Long-term commitments to improving tools and
  techniques and for increasing the emission
  inventory workforce are necessary
• Cooperation, collaboration and strong backing
  from all stakeholders is essential

15
An Initial (3-5 Year) Action PlanFor Canada

• Improve the emission inventory for PM2.5 and its
  precursors
• Improve speciation profiles of PM and VOCs
• Improve point source emission estimates
• Improve the timeliness of the dissemination of
  emission inventory trends and projections.
• Engage appropriate stakeholder groups to develop
  a national strategy to implement emission
  inventory improvements

Estimated cost? About 6M/yr. (US) for the next 5
years.
16
An Initial (3-5 Year) Action PlanFor The United
States

• Enhance the emission inventories and associated
  tools (e.g., SPECIATE) for PM2.5 and its
  precursors
• Establish emission inventory reporting
  requirements for hazardous air pollutants and
  integrate these data into the NEI.
• Improve the capacity of state, local, and tribal
  agencies to develop inventories needed to meet
  regulatory requirements.
• On a regional basis, engage appropriate
  stakeholder groups to develop action plans to
  implement the full range of recommendations.
• Increase support of research to develop and
  improve emission inventories.

Estimated Cost? Approximately 35M/year above
current expenditures, or about 0.2 of the
current annual cost of CAA compliance.
17
An Initial (3-5 Year) Action PlanFor Mexico

• Complete the National Emission Inventory for
  Mexico (released September 2006)
• Develop and implement a communications strategy
  to disseminate results of the NEI
• Develop and fulfill requirements at the national
  level to enable emission inventory updates on a
  three-year cycle
• Build emission inventory development capacity in
  state environmental agencies
• Expand technical capabilities in Mexican agencies
• Continue to improve inventory development
  capabilities through partnerships with Canada and
  the U.S.
• Conduct direct emission measurements and develop
  Mexico-specific emission factors
• Develop a national data system
• Increase human resources at federal and local
  levels

Estimated Cost? About 7M/yr. (U.S.) for the next
3-5 years.
18
Summary

• Accurate emission inventories are the foundation
  of effective air quality management.
• Uncertainties in emission inventories must be
  addressed to avoid the potentially significant
  consequences of inaccurate information.
• NARSTO Emission Inventory Assessment provides a
  road map for improving emission inventories that
  will improve quality, enhance timeliness, and
  increase the cost-effectiveness of air quality
  management programs.
• Significant investment by government agencies and
  the private sector is needed.
• For more information and a PDF version of the
  assessment, go to http//www.narsto.org. Also
  see August 2006 issue of JAWMA.