New Methodology for Estimating Emissions from Residential Wood
Combustion

Paper #180

Roy Huntley

Emission Inventory and Analysis Group (EIAG), Office of Air Quality
Planning and Standards, U.S. Environmental Protection Agency, Research
Triangle Park, NC 27711 

John Van Bruggen, Stephen Colodner, Frank Divita

E.H. Pechan & Associates, Inc., 5528-B Hempstead Way, Springfield, VA
22151 

ABSTRACT

The Emissions Inventory and Analysis Group (EIAG) of the U.S.
Environmental Protection Agency’s (EPA’s) Office of Air Quality
Planning and Standards is in the process of updating the methodology it
uses to estimate emissions from residential wood combustion (RWC) for
the National Emissions Inventory (NEI).  The reasons for doing this are:
 1) to account for appliances not accounted for in the old methodology
(e.g., outdoor hydronic heaters), 2) to make the NEI methodology easier
for state, local, and tribal agencies (S/L/T) to input their own
location-specific knowledge (e.g., wood density, estimates of amount of
wood burned per appliance, appliance populations per county, etc.) into
the estimation methodology, 3) to correct some known deficiencies in the
existing NEI methodology, and 4) to make the methodology easier for some
users of the NEI (e.g., emission modelers) to extract appliance
population data from the NEI.

EPA worked with a group of state, local, and regional planning
organization representatives to create the new methodology.  This paper
describes the new methodology and introduces a Microsoft Access Tool
that was developed to allow S/L/T agencies to calculate annual emissions
from RWC sources in the future.  The Tool is designed to allow users to
update county-level input parameter (based on local survey or like
information) and then easily recalculate county-level emissions by
running a query.

INTRODUCTION

Residential wood combustion (RWC) appliances like fireplaces, fireplace
inserts, woodstoves, and hydronic heaters are significant sources of air
pollution in the United States - especially during winter months.  RWC
emits large amounts of fine particulate matter (PM), volatile organic
compounds (VOCs), and hazardous air pollutants that in several studies
have been shown to contribute to poor human health, air quality, and
visibility.  RWC PM2.5 emissions make up about 7 percent of the national
inventory mass according to the 2002 National Emission Inventory (NEI)
on an annual basis; however, local areas can experience much higher
concentrations in the air during the heating season.1

In recent years, sales and use of residential woodstoves have been
increasing due to rising costs of home heating fuel.  In addition and
over the past 5-years, sales of outdoor hydronic heaters (OHH) are also
strong in the Northern U.S. where they are used as the primary heating
source for some homes.  These outdoor wood boilers are essentially
unregulated by Federal air pollution regulations and emit, on an average
per hour basis, about four times as much fine PM as conventional
woodstoves and about 12 times as much as U.S. Environmental Protection
Agency (EPA)-certified stoves.2  These OHHs have not been included in
past inventories because, until recently, there were not very many of
them in operation.

EPA’s NEI includes emissions data from RWC, which in large part, are
submitted by state, local, and tribal (S/L/T) agencies to EPA for
inclusion in the NEI.  However, when a state does not submit RWC
emissions data, the EPA estimates those emissions.  For the 2002 NEI,
approximately one-half of the states did not submit emissions data from
RWC even though EPA estimates that this category contributes
significantly in many areas to annual PM2.5 and VOC emissions on an
annual basis.

The Emissions Inventory and Analysis Group (EIAG) is in the process of
updating the methodology it uses to estimate emissions from RWC for the
NEI.  The EIAG assembled a task force consisting of members of state and
local air pollution agencies.  See table for names and contact
information.  Members of the task force are identified in this paper in
the Acknowledgements. The reasons for doing this are:  1) to account for
appliances not accounted for in the old methodology (e.g., outdoor
hydronic heaters), 2) to make the NEI methodology easier for states,
local agencies, and tribes to input their own location-specific
knowledge (e.g., wood density, estimates of amount of wood burned per
appliance) into the estimation methodology, 3) to correct some known
deficiencies in the existing NEI methodology, and 4) to make the
methodology easier for modelers and others of like mind to pull
appliance population data from the NEI.  In addition, the 2002 NEI
methodology used to estimate the percent of conventional wood stoves
versus EPA certified woodstoves was determined in the late 1990’s and
is in need of updating.  Also, the percentage of certified woodstoves
can vary widely geographically due to local regulations or local
woodstove change out programs.

Our approach to performing this update is to develop a MS Access Tool
that houses county-level activity and parameters with queries that
allows recalculation of emissions.  The Tool has been developed to:

Include previously ignored appliance types, like outdoor hydronic
heaters, pellet stoves, wax logs, and indoor wood furnaces.

Provide more transparency.

Allow flexibility.

Allow stakeholders to substitute their own tool inputs

Allow easy updates for future years.

Use consistent and new Source Classification Codes (SCCs).

Make use of readily available data.

Be available from the internet.

Table 1 lists the appliance categories and updated SCCs for RWC sources.
 The first column of Table 1 indicates which source categories were
included in the RWC Tool developed for this project.  Note that these
SCCs are different from the codes used in the 2002 NEI.  The SCC update
was developed by the RWC work group to add missing appliance types and
to avoid double counting of emissions.

Table 1.  Appliance Types and Source Category Codes for RWC Sources.

Included	SCC	Fuel	Appliance Type	Comment

Yes	2104008100	Wood	Fireplace: general



2104008110	Wood	Fireplace: open	Conventional fireplace with open hearth

	2104008120	Wood	Fireplace: enclosed (or otherwise modified)	Enclosed
with glass doors or other modifications to a conventional fireplace such
as devices to boost efficiency (heat exchangers)

	2104008130	Wood	Fireplace: qualified for EPA voluntary program



2104008200	Wood	Woodstove: fireplace inserts; general	Fireplace inserts
are similar to freestanding woodstove but they sit inside a fireplace.
Other types of inserts should use SCC = 2104008120

Yes	2104008210	Wood	Woodstove: fireplace inserts; non EPA-certified

	Yes	2104008220	Wood	Woodstove: fireplace inserts; EPA-certified;
non-catalytic

	Yes	2104008230	Wood	Woodstove: fireplace inserts; EPA-certified;
catalytic



2104008300	Wood	Woodstove: freestanding, general

	Yes	2104008310	Wood	Woodstove: freestanding, non-EPA certified

	Yes	2104008320	Wood	Woodstove: freestanding, EPA certified,
non-catalytic

	Yes	2104008330	Wood	Woodstove: freestanding, EPA certified, catalytic



2104008340	Wood	Woodstove: freestanding, masonry heater



2104008400	Wood	Woodstove: pellet-fired, general	Freestanding or
fireplace insert

	2104008410	Wood	Woodstove: pellet-fired, non-EPA certified	Freestanding
or fireplace insert

	2104008420	Wood	Woodstove: pellet-fired, EPA certified	Freestanding or
fireplace insert

	2104008500	Wood	Furnace: Indoor, general

	Yes	2104008510	Wood	Furnace: Indoor, cordwood-fired, non-EPA certified



2104008520	Wood	Furnace: Indoor, cordwood-fired, EPA certified



2104008530	Wood	Furnace: Indoor, pellet-fired, general



2104008540	Wood	Furnace: Indoor, pellet-fired, non-EPA certified



2104008550	Wood	Furnace: Indoor, pellet-fired, EPA certified



2104008600	Wood	Hydronic heater: general, all types

	Yes	2104008610	Wood	Hydronic heater: outdoor



2104008620	Wood	Hydronic heater: indoor



2104008630	Wood	Hydronic heater: pellet-fired



2104008640	Wood	Hydronic heater: meets NESCAUM phase II standards



2104008700	Wood	Outdoor wood burning device, NEC	Fire-pits and chimeas

Yes	2104009000	Firelog	Total: All Combustor Types



2104010000	Biomass; All Except Wood	Total: All Combustor Types

	

OVERVIEW

Calculation Methodology

The emissions from RWC are calculated using the equation below.

Ey = u × EFy × CFy	(Eq. 1)

where:

Ey = annual emissions (ton/year) for a specific appliance (or SCC)

u = annual activity (tons of fuel burned)

EFy = emission factor (tons of pollutant emitted/mass of fuel used)

CFy = control factor

y is a specific pollutant

Note that CFy was assumed to be 1 for all appliances because emission
improvements for RWC are represented by improved appliance designs. 
These were accounted for by applying appropriately adjusted emission
factors.  Each of the terms in the equation and their sources are
explained further below.

Activity (u)

Methodologies for determining the activity varied by the type of RWC
appliance.  Activity for the majority of appliance types was derived
from census data.

Fireplaces, Inserts, and Woodstoves

For the first method, which applies to fireplaces, fireplace inserts,
and woodstoves, activity in terms of tons of fuel burned was calculated
based on several factors as shown in the equation below.

u = P × AP × BR × D	(Eq. 2)

where:

P = Number of occupied housing units in a county in 2005

AP = Percentage of occupied housing units for a specific appliance
category

(e.g., catalytic woodstoves used as main heating equipment, fireplaces
without inserts used as supplemental heating equipment, etc.)

BR = Burn rate (cords/year)

D = average density of the wood fuel burned

Outdoor Hydronic Heaters, Indoor Furnaces, and Pellet Stoves

A second method, which applies to outdoor wood burning devices, indoor
furnaces, and pellet stoves, estimates the number of appliances per
county based on state level proportioned to the number of woodstoves per
county.  Activity is calculated using the following formula.

u = AN × BR × D	(Eq. 3)

where:

AN	=	Number of appliances in county

BR	=	Burn rate (cords/year)

D	=	average density of the wood fuel burned

Number of Occupied Housing Units in 2005

Since appliance profiles were estimated in terms of the percentage of
occupied housing units by appliance type, it is important that county
population also be based on number of occupied housing units.  However
these had to be calculated because intercensal estimates of the number
of occupied housing units by county were not available for 2005 from the
U.S. Census – only annual estimates of the total number of units by
county.  It was important to account for this difference because the
occupancy rate varies greatly throughout the country [e.g., from 98.5
percent in Anoka County, Minnesota to 23 percent in Franklin County,
Pennsylvania].3

The 2005 intercensal estimate was used to determine the number of
housing units in each county and then multiplied by the ratio of
occupied housing units to total housing units from the 2000 census to
find an estimate of the number of occupied housing units in 2005.  The
only exception to this procedure was the county of Broomfield, Colorado,
which was established in 2001 and therefore was not in the 2000 census. 
The percentage of occupied housing units for Boulder County was
substituted for Broomfield because in 2000 it was mostly contained in
Boulder County.

The number of occupied housing units by county appears in the Tool in
the table named County Populations.

Development of Appliance Profiles

County-level appliance profiles for RWC sources were estimated for each
appliance type listed as included in Table 1.  In order to estimate
county level appliance profiles, several steps were needed because
county specific appliance counts for the entire nation were not
available.  Our goal was to use easily accessible and publically
available data to estimate county-level activity by appliance type for
the entire United States, and let that be the default.  Then use state
or local data as submitted to improve on the default values. 

Fireplace and Wood Stove

The American Housing Survey (AHS) conducts national and metropolitan
area surveys on the Nation’s housing, including household
characteristics and heating equipment and fuels.4  Both the national and
metropolitan statistical area (MSA) surveys are conducted during a 3- to
7- month period.  The national survey, which gathers information on
housing throughout the country, conducts interviews at about 55,000
housing units every 2 years, in odd-numbered years.  The metropolitan
area survey consists of 47 metropolitan areas, where householders are
interviewed every 6 years.5  Data is gathered for about 14 metropolitan
areas on an even numbered year until all 47 metropolitan areas are
surveyed.  Table 2 lists the MSAs surveyed in the AHS and the last year
a survey was completed in each area.  Table 3 provides the 2005 AHS
appliance counts by region.

Table 2.  MSAs Gathered in the AHS.

MSA	Year of AHS

Anaheim-Santa Ana, CA PMSA	2002

Atlanta, GA MSA	2004

Baltimore, MD MSA	1998

Birmingham, AL MSA	1998

Boston, MA-NH CMSA	1998

Buffalo, NY CMSA	2002

Charlotte, NC-SC MSA	2002

Chicago, IL PMSA	2003

Cincinnati, OH-KY-IN PMSA	1998

Cleveland, OH PMSA	2004

Columbus, OH MSA	2002

Dallas, TX PMSA	2002

Denver, CO MSA	2004

Detroit, MI PMSA	2003

Fort Worth-Arlington, TX PMSA	2002

Hartford, CT MSA	2004

Houston, TX PMSA	1998

Indianapolis, IN MSA	2004

Kansas City, MO-KS MSA	2002

Los Angeles-Long Beach, CA PMSA	2003

Memphis, TN-AR-MS MSA	2004

Miami-Ft. Lauderdale, FL CMSA	2002

Milwaukee, WI PMSA	2002

Minneapolis-St. Paul, MN-WI MSA	1998

New Orleans, LA MSA	2004

New York-Nassau-Suffolk-Orange, NY PMSAs	2003

Norfolk-Virginia Beach-Newport News, VA-NC	1998

Northern NJ PMSAs	2003

Oakland, CA PMSA	1998

Oklahoma City, OK MSA	2004

Philadelphia, PA-NJ PMSA	2003

Phoenix, AZ MSA	2002

Pittsburgh, PA MSA	2004

Portland, OR-WA PMSA	2002

Providence-Pawtucket-Warwick, RI-MA PMSAs	1998

Riverside-San Bernardino-Ontario, CA PMSA	2002

Rochester, NY MSA	1998

Sacramento, CA PMSA	2004

St. Louis, MO-IL MSA	2004

Salt Lake City, UT MSA	1998

San Antonio, TX MSA	2004

San Diego, CA MSA	2002

San Francisco, CA PMSA	1998

San Jose, CA PMSA	1998

Seattle-Everett, WA PMSA	2004

Tampa-St. Petersburg, FL MSA	1998

Washington, DC-MD-VA MSA	1998



Table 3.  Example of Regional 2005 AHS Data in Thousands and Percentage.

Heating Category	Appliance type	Total	Northeast	Midwest	South	West

Main Heating	Wood Stove	896	150	143	304	299

Main Heating	Fireplace with Inserts	131	14	22	47	47

Main Heating	Fireplace without Inserts	59	10	0	22	27

Pleasure Heating	Wood Stove	985	258	181	260	286

Pleasure Heating	Fireplace with Inserts	655	74	99	220	263

Pleasure Heating	Fireplace without Inserts	391	30	70	133	157

Secondary Heating	Wood Stove	3,096 	920	635	626	915

Secondary Heating	Fireplace with Inserts	4,323 	486	829	1,673 	1,334

Secondary Heating	Fireplace without Inserts	4,080 	371	643	1,686 	1,381

All Houses	Total	108,871	20,337	24,955	39,722	23,858

Main Heating	Wood Stove	0.82%	0.74%	0.57%	0.77%	1.25%

Main Heating	Fireplace with Inserts	0.12%	0.07%	0.09%	0.12%	0.20%

Main Heating	Fireplace without Inserts	0.05%	0.05%	0.00%	0.06%	0.11%

Pleasure Heating	Wood Stove	0.90%	1.27%	0.73%	0.65%	1.20%

Pleasure Heating	Fireplace with Inserts	0.60%	0.36%	0.40%	0.55%	1.10%

Pleasure Heating	Fireplace without Inserts	0.36%	0.15%	0.28%	0.33%	0.66%

Secondary Heating	Wood Stove	2.84%	4.52%	2.54%	1.58%	3.84%

Secondary Heating	Fireplace with Inserts	3.97%	2.39%	3.32%	4.21%	5.59%

Secondary Heating	Fireplace without Inserts	3.75%	1.82%	2.58%	4.24%
5.79%



The AHS provides RWC appliance information for three different appliance
types:  fireplaces with inserts, wood stoves, and fireplaces without
inserts.  In addition, AHS data provide information on the purpose of
each household heating appliance.  These are separated into three
categories of heating appliances use - main, secondary, or pleasure.

The following are the definitions used by the AHS to differentiate
between different types of equipment ownership, by usage
characteristics.

Main heating equipment – Only one type of equipment was reported as
the “Main heating equipment,” as understood by survey respondents. 
More than one category of “Other heating equipment,” which includes
both pleasure and secondary heating equipment, could be reported for the
same household.

Supplemental heating equipment – This is additional heating equipment
for a heated area of the housing unit.  (AHS Appendix A.   HYPERLINK
"http://www.census.gov/hhes/www/housing/ahs/definitions.html" 
http://www.census.gov/hhes/www/housing/ahs/definitions.html .)  For
example, if a home is usually heated with main equipment which is oil or
natural gas, and the homeowner uses a fireplace at times in winter, the
fireplace is considered to be supplemental equipment.

Parallel heating equipment – This is additional heating equipment for
an area not heated by the main heating equipment.6  For example, if the
main house is heated with a wood stove, and a garage is heated by an
electric heat pump, the heat pump would be considered parallel heating
equipment.

The following are the definitions used by the AHS to differentiate
between different types of RWC Appliances.

Fireplaces without inserts - refers to glass door fire screens or fire
backs inserted in the back of the fireplace to passively reflect heat.

Fireplaces with inserts - have a fan-forced air circulation system to
force the heat into the room.

Wood stove - refers to any range or stove that burns solid fuel
including wood burning, pot belly, and Franklin stoves.

These population percentages were applied to county-level occupied
housing units as described above in order to estimate the number of
appliances in each county.

The Hearth and Patio Association estimates the proportion of EPA
certified wood stoves and inserts to uncertified wood stoves and
inserts.7  In 1999 the estimate was 92 percent uncertified, 5.7 percent
certified using a non-catalytic design, and 2.3 percent certified using
a catalytic design.  EPA estimates an annual growth rate of 1 percent of
certified stoves in operation so a case could be made for altering the
mix of appliances slightly depending on the age of the latest Hearth and
Patio report.8  One estimate of the ratio in 2005 may be as much as 20
percent certified.  For this study, we used a distribution of 80 percent
uncertified, 15 percent certified, and 5 percent certified catalytic for
both fireplace inserts and woodstoves.

Fireplaces without inserts were assumed to burn cordwood 78.3 percent of
the time, and wax/sawdust firelogs 21.7 percent of the time, and the AHS
data was split between the two relevant SCCs accordingly.  This was
based on the proportion of estimated cordwood equivalent for firelogs to
cords consumed in fireplaces without inserts in the 2002 NEI
documentation.1  The most recent analysis performed by E.H. Pechan &
Associates, Inc. (Pechan) resulted in an improvement that we plan to
implement in the 2008 version of this tool.  The updated distribution
between cordwood and wax/sawdust firelogs is assumed to be 86.5 percent
cordwood and 13.5 percent wax/sawdust firelogs.  This was based on
survey data reflecting the number of households which use wax firelogs
exclusively.9

These appliance profiles are in the Tool as a table called Appliance
Profiles.

The product of these appliance profile percentages with the number of
occupied housing units in a county was used to find the total number of
appliances in the county for the majority of appliance types.

Outdoor Hydronic Heaters and Indoor Furnaces

For OHHs and indoor furnaces, a different approach was used to find the
number of appliances.  Sales data was able to provide estimates of the
number of appliances within each county.

Outdoor Hydronic Heaters

For Outdoor Hydronic Heaters, the population of these appliances were
obtained by using sales data compiled by Northeast States for
Coordinated Air Use Management (NESCAUM) that was provided by Hearth,
Patio, and Barbeque Association.  The counties populations were
estimated by multiplying the state totals by the fraction of woodstoves
in the county relative to the state.  Sales data were available for all
50 states.

For outdoor hydronic heaters, state level sales data by distributing the
sales within the state to all counties not in MSAs.  This distribution
was in proportion to the number of occupied housing units.  The counties
in the MSAs were not included in the distribution to avoid assigning
high numbers of these units, which are typically found in more rural
environments, to cities, based solely on population density.

Indoor Furnaces

For indoor furnaces, sales data was able to provide county level
estimates of the number of appliances.8  The number of appliances for
both indoor furnaces and outdoor hydronic heaters is listed in the table
Other Appliance Populations in the Tool.

Sales of indoor furnaces were very strong in the 1970’s and 1980’s,
with the strongest sales in the upper Midwest, the Great Lake states,
and New York and Pennsylvania.  After 1988 when the EPA’s NSPS
Woodstove rule was promulgated, sales of indoor furnaces have been weak.
(Roy Huntley email to Frank Divita of Pechan).  Sales in the southern
states and in the Pacific Northwest have traditionally not been strong.

Several approaches were used to estimate the county level appliance
population of indoor furnaces.  For the Mid-Atlantic Northeast
Visibility Union (MANE-VU) states (Maine, Vermont, New Hampshire,
Massachusetts, Connecticut, New York, Delaware, Maryland, New Jersey,
Pennsylvania, Rhode Island, and the District of Colombia), we used the
MANE-VU report to obtain the number of furnaces.9  The MANE-VU report
provides the number of centralized heaters/furnaces by state.  This
figure includes both indoor furnaces and OHH, so taking the difference
between this figure and the NESCAUM data for OHH, we were able to
estimate the number of indoor furnaces by state.  We then allocated
these units to the county level by ratioing the state total by the
fraction of woodstoves in the county.

For Minnesota, we used information in a recent report from Minnesota
Department of Natural Resources that provided the number of furnaces by
five Minnesota regions.10  Furnaces were allocated to the county level
by ratioing the state total by the fraction of woodstoves in the county.

For the rest of the Great Lakes states (Wisconsin, Michigan, Illinois,
Indiana, and Ohio), we generated an average factor from the Minnesota
report (38 furnaces per 100 woodstoves) for the number of furnaces per
woodstove and used the number of woodstoves in each county to estimate
the number of furnaces.

For the rest of the states, we developed a factor from the 2000
Oregon/Washington/Idaho survey provided by the Oregon Department of
Environmental Quality (DEQ).  The survey provided the percentages of
device type by region for 4 of the 5 Oregon regions (we were unable to
use data from the southeast region due to incomplete central furnace
data).  The central region furnace data as assumed to be represented by
indoor furnaces and we therefore used the ratio of the percent of
furnaces to the percent of woodstoves by the 4 regions.  The average of
the ratios was used to determine the number of furnaces in the remaining
state not mentioned above.  As a last step, we assumed no outdoor
furnaces in counties that are in climate zones 4 and 5

Burn Rates

Burn rate estimates with the level of detail needed for inventory
purposes was difficult to come by.  National average burn rates were
assumed based on the U.S. Department of Agriculture (USDA) Forest
Service documents from Midwest or Great Plainstate surveys performed in
the 1990s (publications available from   HYPERLINK
"http://www.ncrs.fs.fed.us/pubs/"  http://www.ncrs.fs.fed.us/pubs/ ). 
Burn rate was differentiated by appliance type (woodstove, fireplace,
fireplace with insert, furnace/boiler, and firepit) and burning purpose
(Main, Secondary, and Pleasure).  The Access Tool includes the average
number of cords of wood burned in a year by appliance type.  It was
assumed that these burn rates are based on most of the appliances not
being EPA certified.  To account for the greater efficiencies of EPA
certified appliances, a discount factor of 79.4 percent was applied. 
This means that if an uncertified stove burns 7 cords of wood in a
season, a certified stove would burn only 5.55 cords of wood.

Wax/Sawdust firelogs also heat more efficiently than wood, and it is
estimated that their use displaces more than their volume in wood that
would have been burned.  Therefore, the burn rates for wax/sawdust
firelogs were reduced from those of fireplaces without inserts by a
factor of 4.49, which accounts for the increased volume of wood that
would be burnt to release the equivalent amount of heat.11, 12

County level climate zone were used to adjust burn rate profiles to
account for the fact that less wood is burned in warmer states.  The
Commercial Buildings Energy Consumption Survey (CBECS) climate zones are
groups of climate divisions, as defined by the National Oceanic and
Atmospheric Administration (NOAA), which are regions within a state that
are as climatically homogeneous as possible.  Each NOAA climate division
is placed into one of five zones based on its 30-year average heating
degree-days (HDD) and cooling degree-days (CDD) for the period 1971
through 2000.13  Climate zone is listed in the County Population table. 
Burn rates for all SCCs in the national default were multiplied by the
ratio of the average British thermal unit (Btu) consumption to heat a
house each climate zone to the average Btu consumption in climate zone
1.  The ratios were 0.30 for climate zone 5, 0.44 for climate zone 4,
and 0.77 for climate zone 3.8

Modifications by S/L/T agencies with more detailed information may be
made by appending a new burn profile to the Burn Rates table in the
Access Tool and changing the burn profile number of the appropriate
counties in the County Populations table.  Each burn rate profile should
address all the RWC SCCs that are considered by this Tool, and can
distinguish between Main, Secondary, and Pleasure burn rates.  In the
event that appliance populations and burn rates are not distinguished by
burn type (main, secondary, or pleasure), insert the overall average
burn rate in the burn type for which appliance populations are
calculated.  For example, in Minnesota, appliance populations were all
put under the secondary burn type, consequently, burn rates were only
populated in the secondary burn type.

Wood Density

To compute average density of wood by county, we use the density of oven
dried wood because emission factors developed by EPA are based on oven
dried wood mass units.  Dried wood density data was obtained from the
U.S. Forest Service for various wood species.14  The Forest Service
developed a database (called the timber products output) that contained
survey results of sawmill operators that includes the volume of wood by
species for several different categories of use - one of the uses being
fuel wood.

Using the oven dried density by species multiplied by the per-species
volumes gave a per species weight which was summed to calculate the
total weight for the county.  This was then divided by the total volume
of wood in the county to get the average density by county.  This
average is stored in the Density by County table under the pounds per
cubic feet (lbs/ft3) column.  If a county specific density was not
available, regional averages were used instead.

The calculated density by county from the Forest Service data was then
converted to tons/cords.  Officially a cord is defined as a stack of
wood 4 feet wide, 8 feet long, and 4 feet tall or 128 ft3.  However, to
account for air spaces in the stack, we assumed 80 ft3 per cord instead.

For wax firelogs, density was assumed to not vary from county to county,
and a density of 4.005 tons per cord was used.  This was based on the
volume of a typical 5 pound firelog.  For wax firelogs, a cord was
assumed to be actually 128 ft3 because air spaces assumptions are
applicable.  The density figure was arrived at based on a Duraflame
wax/sawdust firelog, which was measured to have a volume of ~138 cubic
inches or ~.0799 cubic feet.  From this a density of ~62.58 lb/ft3  was
derived.  Assuming that for this source type, the equivalent of one cord
was 128 cubic feet, we arrived at a density of ~8010 lb/cord or 4.005
tons/cord.

The wood density by county is in the Tool as in the table Density by
County.  S/L/T agencies can update the appropriate counties with the
density of cordwood consumed, filling in not only the Density (lbs/ft3),
but converting and filling in the Density (tons/cord) column by
multiplying by the factor:  (80 ft3/cord)/(2000 lbs/ton).  The
data_source column should be updated appropriately as well.  When
updating or revising the wood density, users should be sure to use
oven-dried density numbers.

Emission Factors

The emission factors used are expressed as tons of pollutant produced
for every ton of wood burned.  The emission factors were reviewed by the
task force and for the most part are the same as the ones used in past
inventories.  Some minor changes were implemented.  The most significant
change was to the VOC emissions factor.  The old factor was from AP-42
and it was 229 lb VOC /ton of wood burned.  The new factor is 18.9 lb
VOC/ton of wood burned and it was pulled from the Mid-Atlantic Regional
Air Management Association report.  The complete list of the revised
emission factors is in the tool.

DESCRIPTION AND USE OF THE RWC TOOL

The RWC Tool was developed in Microsoft Access to allow S/L/T agencies
to calculate annual emissions from RWC sources.  The Tool is designed to
allow users to update county-level input parameter (based on local
survey or like information) and then calculate county-level emissions by
running a query named “Create Calculated Inventory” to generate a
table called Calculated Inventory.  Figure 1 provides the flow of the
RWC Tool calculations.  The shaded boxes in Figure 1 show calculated
values which the non-shaded boxes show data inputs.

As shown in Table 4, the RWC Tool consists of several tables of
parameters used to calculate emissions associated with burning.  Each
are described below.

Figure 1.  RWC Tool Data Flow Diagram.

 

Table 4.  Listing and Descriptions of the Tables Included in the RWC
Tool.

Table Name	Table Description

County Populations	Entries contain the county, the number of occupied
housing units in 2005, the appliance profile, the burn profile, and the
climate zone.

Appliance Profiles:	Entries contain the appliance profile, the SCC, the
burn purpose (Main, Secondary, or pleasure) and the percentage of
households with an appliance of the type corresponding to the SCC.

Burn Rates	Entries contain the burn profile, the SCC, the burn purpose,
and the cords burned per year per appliance.

Density by County	Entries contain the county, the density in lb/ft3, the
density in tons/cord, and the data source.

Other Appliance Populations	Entries contain the county, the SCC, the
burn purpose, and the number of appliances in the county with an
appliance of the type corresponding to the SCC.

Emission Factor by SCC	Entries contain the SCC, the Pollutant, the
emission factor with units, the emission factor converted to tons
pollutant/tons of wood combusted, and the data source for the emission
factor.



County Populations - This table lists the number of occupied housing
units in each county in the United States.  These estimates are based on
the 2005 intercensal estimate of total housing units and the percentage
of units occupied in 2000.  If state or local agencies have performed
counts of the number of occupied households per county, these numbers
can be substituted into the county populations table and the query will
recalculate the emissions accordingly.  More accurate counts of the
occupied housing units can be submitted to EPA along with a data source
for reference, and they will be used to replace the current estimates.

Appliance Profiles - The Appliance Profiles table has a number of
existing appliance profiles.  Each profile provides the percentage of
occupied households which utilize each type of wood burning appliance
(SCC) and burn purpose combination.

If more accurate appliance percentages for any or all such combinations
are available for a state or number of counties, a new appliance profile
can be created to accommodate the data.  For example, if state or local
woodstove change-out programs result in a higher concentration of
certified woodstoves in a county or counties, this can be accounted for
by editing the relative proportion of certified to uncertified
woodstoves.  It should be noted in the appliance_profile column which
appliance profile is currently assigned to the state or counties.  If
improved percentages are known only for some appliance/burn type
combinations, the rest should be copied from the original, otherwise all
emissions from these sources will be calculated with a value of zero. 
After adding a new appliance profile with a new number, the
appliance_profile column in the county populations table should be
updated for the affected counties.

If local agencies have more accurate data, EPA should be provided with a
table of both the updated and copied percentages (so as to ensure a
complete profile.)  The Federal information processing standard (FIPS)
of the counties to which these percentages apply should also be
provided, along with a reference for the source of the updated
quantities.

In the event that actual equipment populations are known, rather than
percentages, the percentages must be back-calculated by dividing the
number of appliances per each burn type by the county populations.  This
will result in a separate new appliance profile for each geographic area
in which appliance populations are known.  The same steps apply as above
for supplementing incomplete population data with the existing profile
data and providing references for data sources.

Burn Rates - The Burn Rates table contains a number of existing burn
profiles.  Each profile provides the cords burned per year per household
for each type of wood burning appliance (e.g., SCC) and burn purpose
combination.

If more accurate burn rates are known for any or all SCC/burn purpose
combinations, a new burn profile can be created to accommodate this
data.  It should be noted in the County Populations table burn_profile
column which profile is currently assigned to the state or counties.  If
improved burn rates are only known for some appliance/burn type
combinations, the rest should be copied from the original burn profile,
or the amount of wood burned, and consequently the emissions, will be
calculated as zero.  After the addition of a new burn profile with a new
number, the affected counties should have the burn_profile column in
County Populations updated.

If local agencies have more accurate data than is currently in the tool,
EPA should be provided with a table of both the updated and copied burn
rates in cords/year/household.  The FIPS of all the counties to which
these burn rates should apply should also be provided, along with a
reference for the source of the updated burn rates.

If the rate at which cordwood is burned is known in terms of cubic feet
rather than cords, a cord should be assumed to be 80 cubic feet of
stacked dry cordwood when converting the burn rates.  The same steps
apply as above for supplementing incomplete burn rate data with rates
from the currently used existing burn profile and providing references
for data sources.

Density by County - The Density by County table contains estimates of
the average density of dry cordwood that is used in residential
combustion for every county.  Each County has the density listed in both
lbs/ft3 and in tons/cord.  The density can be updated for any county or
counties and the query will recalculate the emissions accordingly.

If more accurate estimates of the average wood density exist for one or
more counties, the improved data, along with a source reference can be
sent to EPA.  If this data is used, the data_source column in the
Density by County table will be updated with this information.  Please
note that all densities must be for dry cordwood.  Also, the conversion
between lbs/ft3 and tons/cord assumes 2000 lbs in one ton and 80 cubic
feet in one cord.

Other Appliance Populations - The Other Appliance Populations table
contains data on the number of appliances for SCCs which were not
covered by the appliances in the American Housing Survey.  These include
pellet-fire woodstoves, indoor furnaces, and outdoor hydronic heaters. 
Each county is listed with the number of appliances of each appliance
type and burn purpose combination.

To model different estimates of the appliance population in one or more
counties, this data can be entered by changing the Number of Appliances
column for all affected counties.  If the query is run, it will
calculate the updated emissions accordingly.

If more accurate estimates of the appliance population in one or more
counties are known, EPA should be provided a table of the changes or
additions to the table, including all the columns that appear in the
table in the tool.

Emission Factor by SCC - The Emission Factor by SCC table contains a
list of the emission factors that were used to calculate emissions.  It
also cites the source of each emission factor.  Any changes or additions
to the emission factors will change the results for every county when
the query is run.  Users are not encouraged to change this table.  Any
errors noticed in the current emission factor table should be reported
to EPA.

CONCLUSIONS

An updated methodology was developed as collaboration between EPA,
Pechan, and a group of state, local and regional planning organization
representatives.  In addition, a Microsoft Access Tool was developed in
a way that allows users to update county-level input parameter (based on
local survey or like information) and then recalculate county-level
emissions by running a query.  The Tool is transparent so that inputs
can be easily identified.

EPA is interested in obtaining local data that can be used to calculate
RWC emissions.  Improved data sources in the format used by the Access
Tool can be provided to the authors via email.  For everyone’s
convenience, Pechan has developed a template that will aid in uploading
state specific data to the tool.  The template can be obtained by
contacting the authors of this paper.  As of this writing, EPA is
updating the tool for the 2008 calendar year.

ACKNOWLEDGEMENTS

The authors would like to thank all of the following individuals who
were members of the EPA RWC Workgroup who provided data and guidance
during the development of the RWC Access Tool.  These include:

Scott DiBiase, Pinal County, Arizona

Di Tian, State of Georgia

David Fees, State of Delaware

Julie McDill, Mid-Atlantic Regional Air Management Association

Sally Otterson, State of Washington

Bart Sponsellar, State of Wisconsin

Christal Thompson, State of Alabama

Chris Swab, State of Oregon

David Wright, State of Maine

REFERENCES

1.	U.S. Environmental Protection Agency, “Documentation for the Final
2002 Nonpoint Sector (Feb 06 Version) National Emission Inventory for
Criteria and Hazardous Air Pollutants,” Office of Air Quality Planning
and Standards, Air Quality Assessment Division, Research Triangle Park,
NC, July 2006.

2.	Office of the Attorney General Environmental Protection Bureau,
“Smoke Gets in Your Lungs:  Outdoor Wood Boilers in New York State,”
March 2008.

3.	U.S. Census Bureau, American FactFinder, Data Sets, Census 2000
Summary File 3, Quick Tables, Internet address:    HYPERLINK
"http://factfinder.census.gov"  http://factfinder.census.gov , 2000.

4.	United States Census Bureau, “American Housing Survey for the
United States:  2005, Table 2-4: Selected Equipment and Plumbing,”
Current Housing Reports,   HYPERLINK
"http://www.census.gov/hhes/www/housing/ahs/ahs05/ahs05.html" 
http://www.census.gov/hhes/www/housing/ahs/ahs05/ahs05.html , U.S.
Government Printing Office, Washington, DC, 2006.

5.	United States Census Bureau, “American Housing Survey Metropolitan
Data,” Current Housing Reports, various years, retrieved from  
HYPERLINK
"http://www.census.gov/hhes/www/housing/ahs/metropolitandata.html" 
http://www.census.gov/hhes/www/housing/ahs/metropolitandata.html ,
November 23, 2007.

6.	U.S. Census Bureau, “2007 American Housing Survey data” retrieved
from   HYPERLINK
"http://www.census.gov/hhes/www/housing/ahs/ahs07/ahs07.html" 
http://www.census.gov/hhes/www/housing/ahs/ahs07/ahs07.html , 2007.

7.	Hearth & Home, “Hearth, Barbeque, and Patio Retailing, 2000,”
Pages 26-34, April 2000.

8.	Energy Information Administration, Table CE2-1c, Space-Heating Energy
Consumption in US Households by Climate Zone, 2001,
http://www.eia.doe.gov/emeu/recs/recs2001/ce_pdf/spaceheat/ce2-1c_climat
e2001.pdf

9.	Houck, J.E.; Eagle, B.N., “Control Analysis and Documentation for
Residential Wood Combustion in the MANE-VU Region,” prepared for the
Mid-Atlantic Regional Air Management Association by OMNI Environmental
Services, Inc., December 2006.

10.	Minnesota Department of Natural Resources, “Residential Fuel Wood
Assessment – State of Minnesota, 2002-2003 Heating Season,” Division
of Forestry, 2005.

11.	Houck, J.E., “Recommended Procedure for Compiling Emission
Inventory Data for Manufactured Wax/Sawdust Fireplace Logs,” OMNI
Consulting for the U.S. Environmental Protection Agency, October 2001.

12.	Houck, J.E., “Addendum to: Recommended Procedure for Compiling
Emission Inventory Data for Manufactured Wax/Sawdust Fireplace Logs,”
OMNI Consulting for the U.S. Environmental Protection Agency, November,
2001.

13.	Energy Information Administration, “U.S. Climate Zones for
Commercial Buildings,” retrieved from   HYPERLINK
"http://www.eia.doe.gov/emeu/cbecs/climate_zones.html" 
http://www.eia.doe.gov/emeu/cbecs/climate_zones.html , 2007.

14.	U.S. Department of Agriculture, “Timber Products Output Survey,”
Forestry Service, retrieved via query from   HYPERLINK
"http://ncrs2.fs.fed.us/4801/fiadb/rpa_tpo/wc_rpa_tpo.ASP" 
http://ncrs2.fs.fed.us/4801/fiadb/rpa_tpo/wc_rpa_tpo.ASP , November 19,
2007.

KEYWORDS

Residential wood combustion, National Emissions Inventory, woodstove
emissions, fireplace emissions, wax firelog emissions

  PAGE  1 

 Occupied

Units; P

(Households)

Appliance Profile; AP

x

(% of Occupied Units)

 Number of

appliances; AN

x

(Sales Data)

Number of

appliances

(Census Data)

Burn Rate per

Appliance; BR

x

(Cords/year)

Total Volume of

Wood

(Cords)

Wood

Density; D

(Tons/cord)

Total Mass

of Wood

(Tons)

Emission Factor;

EF

y

(Tons Pollutant/

Ton Wood)

Emissions; E

x

y

Census Data Approach

(Fireplaces and Woodstoves)

Appliance Sales Approach

(OHHs and Indoor Furnaces)

