EPA Contract No. 68-D7-0068
Work Assignment No. 2-09
ERG No. 0101-01-009
ESTIMATING PARTICULATE MATTER EMISSIONS
FROM CONSTRUCTION OPERATIONS
FINAL REPORT
Prepared for:
Emission Factor and Inventory Group
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Prepared by:
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
Under Subcontract to:
Eastern Research Group, Inc.
1600 Perimeter Park
P.O. Box 2010
Morrisville, North Carolina 27560
September 30,1999
EASTERN RESEARCH GROUP, INC
-------
EPA Contract No. 68-D7-0068
Work Assignment No. 2-09
ERG No. 0101-01-009
ESTIMATING PARTICIPATE MATTER EMISSIONS
FROM CONSTRUCTION OPERATIONS
FINAL REPORT
Prepared for:
Emission Factor and Inventory Group
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Prepared by:
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
Under Subcontract to:
Eastern Research Group, Inc.
1600 Perimeter Park
P.O. Box 2010
Morrisville, North Carolina 27560
September 30,1999
-------
Preface
This report was prepared by Midwest Research Institute (MRI) for the U.S.
Environmental Protection Agency, Office of Air Quality Planning and Standards under
Purchase Order No. EFIG-0068 from Eastern Research Group (ERG). Mr. Garry Brooks
was the Work Assignment Leader for ERG. This work was performed under EPA Prime
Contract No. 68-D7-0068 with ERG.
The report summarizes the methods that have been used to develop inventories of
fugitive dust and exhaust particulate matter (PM) emissions from construction activities,
identifies surrogate data sources for PM emission calculations, and proposes a preferred
methodology to estimate county level emissions. Mrs. Mary Ann Grelinger was the MRI
Project Leader for this assignment. Dr. Chatten Cowherd and Dr. Greg Muleski served as
technical consultants on this project. This report was prepared by Mrs. Grelinger,
Ms. Courtney Kies, and Dr. Cowherd.
MIDWEST RESEARCH INSTITUTE
_J
Chatten Cowherd, Jr., Ph.D.
Principal Advisor
Approved:
/ri- RoyNeulicht
Program Manager
Technical Support Contact
September 15,1999
in
-------
Contents
Preface iii
Figures vii
Tables vii
Section 1. Introduction 1-1
Section 2. PM Emissions from Construction Activities 2-1
2.1 Information Sources—Construction Activity Levels 2-1
2.2 Information Sources—Construction Emission Factors ............ 2-2
2.3 Emission Calculations 2-2
2.4 Factors Influencing Construction Emissions 2-3
Section 3. Categories of Construction 3-1
3.1 Road Construction 3-1
3.2 Residential Construction 3-1
3.3 Nonresidential Construction ; 3-2
3.4 Other Construction 3-2
Section 4. Existing Methodologies for Estimating Construction Emissions 4-1
4.1 Methodology 1: General "Top-Down" Emission Inventory 4-1
4.2 Methodology 2: NET Inventory 4-4
4.3 Methodology 3: California Emission Inventory Procedure 4-7
4.4 Methodology 4: National Particulate Inventory—Phase I — 4-8
4.5 Methodology 5: Regional Emission Inventories 4-9
4.6 Methodology 6: Major Construction Project Inventory 4-16
4.7 Methodology 7: U.S. EPA NONROAD Model 4-17
Section 5. Recommended Methodologies and Data Sources 5-1
5.1 Assumptions and Limitations of Current Methodologies 5-1
5.2 Recommended Changes to the NET Methodology 5-1
5.3 General Emission Factor for Construction 5-2
5.4 Residential Construction Emissions 5-2
5.5 Nonresidential Construction Emissions 5-6
5.6 Roadway Construction Emissions 5-8
5.7 Correction Parameters 5-11
5.8 PM10 Emissions from Combustion of Cleared Materials 5-14
Section 6. References .6-1
-------
Figures
Figure 5-1. Residential Construction Emissions Flowchart 5,-3
Figure 5-2. Nonresidential Construction Emissions Flowchart 5-7
Figure 5-3. Road Construction Emissions Flowchart 5-9
Figure 5-4. Map of PE Values for State Climatic Divisions 5-12
Tables
Table 2-1. Types of Construction Equipment 2-1
Table 4-1. Construction Dollars-To-Acres Conversion Factors (MRI, 1974) 4-3
Table 4-2. Estimation of Construction Emissions—National Inventory by MRI 4-4
Table 4-3. Estimation of Construction Emissions—EPA
National Emission Trends Analysis by E.H. Pechan and Associates 4-6
Table 4-4. Estimation of Construction Emissions—California Methodology 4-7
Table 4-5. Estimation of Construction Emissions—SJV Methodology 4-10
Table 4-6. Estimation of Construction Emissions—SCAQMD Methodology 4-10
Table 4-7. AP-42 Recommended PM10 Emission Factors for
Construction Operations 4-11
Table 4-8. Recommended PM10 Emission Factors for Construction Operations 4-12
Table 4-9. Estimation of Construction Emissions—Phoenix Methodology 4-13
Table 4-10. Estimation of Construction Emissions 1991
Las Vegas Methodology 4-14
Table 4-11. Emission Inventory Methodologies 4-20
Table 5-1. Example Annual PM10 Emissions from Residential Construction
in a Hypothetical County 5-5
Table 5-2. Example 1992 PM10 Emissions for Nonresidential Construction
in a Hypothetical County ...; 5-8
Table 5-3. Road Miles-to-Acres Conversion Calculation 5-10
Table 5-4 Example PM10 Emissions from Road Construction in a
Hypothetical County : 5-11
Table 5-5. Dry Silt Content by Soil Type 5-13
Table 5-6. Recommended Methodology 5-14
Table 5-7. Combustion of Cleared Materials Emission Factors by Region 5-15
Table 5-8. Example Calculation of PM10 Emissions from the Burning of Vegetative
Residues 5-16
vn
-------
Section 1.
Introduction
This report was prepared as part of a study to develop an improved method for
estimating particulate matter (PM) emissions from construction operations.
A new methodology is needed to improve emission estimates on a national county-by-
county basis for the National Emission Trends (NET) inventory. Construction operations
can substantially impact local air quality from suspended dust, equipment exhaust, and
burning emissions. The majority of PM emissions originates from sources that suspend dust
from soil and construction materials, especially from equipment travel. PM emissions are
released into ambient air from the following construction activities:
• Equipment movement on unpaved surfaces (suspended dust and exhaust emissions)
• Earthmoving (cut and fill operations, and excavation activities)
• Material transfer operations, including loading/unloading activities
• Material alterations, including drilling, crushing, screening, cutting, blasting, and
surface cleaning activities
• Portable plant crushing and screening
• Track-out of dirt to nearby paved roads for subsequent dust resuspension by traffic
• Land clearing, including demolition/burning of existing structures and vegetative
residues
• Wind erosion of soil exposed by construction activities
The activities performed in this study included:
• Identification of readily available national and regional information sources that
can be used to prepare an inventory of PM emissions from construction activity
• Identification of categories of construction that can be expected to have different
emission characteristics (e.g., highway, commercial, housing)
• Characterization of factors that impact construction emissions (e.g., meteorological
parameters, regional differences in construction, soil types, economic conditions)
• Development of a methodology to estimate county-level emissions of fugitive dust
from construction activities
This report is organized as follows. Section 2 provides information on the calculation
of PM10 and PM2 5 components of fugitive dust and exhaust emissions generated during
construction operations. Section 3 identifies the categories of construction that are believed
to have different dust emitting characteristics and levels of activity and in turn produce
1-1
-------
different amounts of PM emissions. Section 4 presents existing methodologies used to
calculate PM10 emissions from construction activities. Section 5 presents an assessment of
the California methodology and the NET methodology, recommended changes to the Trends
procedure, an updated methodology for calculating emissions for the county-level on a
national basis, and a review of the data sources needed to develop such an inventory.
1-2
-------
Section 2.
PM Emissions from Construction Activities
Particulate matter emissions from construction activities are produced from equipment
exhaust (primarily from diesel-fueled engines), equipment travel and activity on unpaved
surfaces, on-site material handling operations (e.g., temporary on-site crushing/screening),
and track-out of dirt onto adjacent paved roads with subsequent resuspension by traffic.
Equipment exhaust emissions consist of finer, combustion aerosols, while fugitive dust
emissions consist mostly of coarser crustal particles.
Conditions that influence construction PM emissions include equipment type, size, and
travel speed; engine type, size, and load; soil type and moisture content; and wind
conditions. For example, exhaust emissions are high when excavating soil and engines are
under load; fugitive dust emissions are high when dry surface dust is disturbed and
suspended by construction equipment travel.
A wide variety of equipment classes, sizes, and engine types are used in construction
activities. Construction equipment includes motor graders, trucks, scrapers, and other
equipment types. General construction equipment is outlined in Table 2-1.
Table 2-1. Types of Construction Equipment
Motor araders
Loaders (track- and
wheel-type)
Road wideners
Windrow elevators
Trucks
Tractors (track- and wheel-type)
Compactors (pneumatic and
vibratory)
Cold planers
Scraoers
Excavators (track- and wheel-
type)
Road reclaimers/ Soil
stabilizers
Power shovels
2.1 Information Sources—Construction Activity Levels
Many data sources are available that provide construction statistics for the national,
regional, state, and county levels. This study identified information sources that can be
used to develop a county-by-county inventory of PM emissions associated with
construction activities. The available information sources determine the form of
methodology that is used to develop the inventory.
Due to variations in the type of data that local governmental agencies can provide
(construction permits and/or compiled local construction data), methods for determining
construction activity levels differ by area. Many areas have high quality measures of
construction activity levels resulting from local government requirements for construction
permitting; however, only lower quality (less resolved) data may be available for other
areas.
2-1
-------
Two widely used references for national construction statistics are the F.W. Dodge
Reports published by McGraw Hill, Inc. and the U.S. Bureau of Census, Construction
Statistics Division. The F.W. Dodge Group publishes the monthly Dodge Construction
Potentials Bulletin, and the Dodge Local Construction Potentials Bulletin providing the
dollar value spent on various types of construction and also the number of buildings
constructed. Annual reports and other supporting databases are also available from F.W.
Dodge. All information is provided for a fee. The U.S. Bureau of Census publishes yearly
the Statistical Abstract of the United States. This publication includes statistics on various
aspects of construction. The Census of Construction Industries Division produces monthly
statistics on construction activities including the number of housing starts. Most
information from the F.W. Dodge group and the U.S. Census Bureau is available on a state
basis.
Transportation statistics are published yearly by the Federal Highway Administration
(FHWA) in Highway Statistics. The publication includes roadway characteristics and
extent along with other roadway statistics. The data provided by the FHWA is useful in
determining the new miles of roadway constructed on a yearly basis.
2.2 Information Sources—Construction Emission Factors
Two chapters of the U.S. EPA handbook, "Compilation of Air Emission Factors" [AP-
42]l apply to particulate matter emissions from construction activities. Chapter 7 relates to
emissions from the mineral products industry, including construction aggregate processing
and crushed stone processing. Chapter 13 contains relevant emission factors for prescribed
burning, unpaved road traffic, aggregate handling and storage piles, industrial wind
erosion, abrasive blasting, and explosives detonation. Section 13.2.3, "Heavy Construction
Operations," contains PM emission factors specifically for emissions from heavy
construction. Exhaust emissions contains emission factors from diesel-fueled construction
equipment are separately estimated using EPA's NONROAD model.
2.3 Emission Calculations
Emissions from construction operations are related to three phases of a project.
Demolition and debris removal includes removal of old structures or brush collection and
transport/burning. Site preparation involves cut-and-fill, grading, and compaction
activities (i.e., all earthmoving operations). General construction includes material
handling operations for construction of structures and roads. Under some local PM
estimation methodologies, construction equipment activity is allocated to road
construction, building construction, and miscellaneous land-moving operations. Emissions
are calculated for specific periods and time intervals. Inventories can be developed for
annual, seasonal, monthly, and for worst-case, twenty-four hour periods.
2-2
-------
Estimates of PM10 and PM2 5a emissions from construction activities are developed
using emission factors, activity level (source extent) datab, and control efficiencies (if
applicable). Historically, the primary emission factor for construction activities has been:
e = 1.2 tons/acre/month of activity
This factor was based on early (i.e., 1970's) upwind/downwind tests of construction site
impacts on ambient particulate levels. It refers to total suspended paniculate (TSP) matter
emissions represented by particles no greater than 30 fim in aerodynamic diameter.
Additional emission factors for earthmoving and other activities associated with
construction operations can be borrowed from other AP-42 chapters, but certain differences
exist between construction operation emissions and emissions from other fugitive dust
sources. These additional factors were derived from field testing using the MRI exposure
(plume) profiling method that determines the downwind transport of PM flux.
Consequently, these emission factors combine exhaust with fugitive dust emissions. PM
emission factors for fugitive dust are available in AP-42 Chapters 7 and 13 and are related
to soil silt and moisture contents.
Emission factors for PM from construction equipment exhaust are available in the
NONROAD model produced by EPA, Office of Mobile Sources (QMS), and are related to
engine type, size, and load. The EPA QMS has developed a second draft of the
NONROAD Emission Inventory Model. The NONROAD model calculates emissions of
criteria and HAP pollutants, including PM emissions.
Control efficiency data for construction equipment engines is built into the
NONROAD model for future diesel engine rules that will affect PM emissions. Control
efficiencies for fugitive dust are published in AP-42 and are primarily related to watering
or chemical suppression of surface soils at construction sites.
2.4 Factors Influencing Construction Emissions
The factors that influence construction emissions represent meteorological parameters,
regional construction .differences (e.g., basement/no basement for residential housing), soil
types, and economic growth. Construction activity is related to climate, terrain, and
economic conditions. For example, residential foundations differ between northern and
southern states in the U.S. (e.g., fewer basements are excavated in southern states).
Regional terrain and soil variations are also important (e.g., highway construction in
mountains, or rocky vs. silty soils).
a PM10 and PM25 refer to particulate matter no greater than 10 (am and 2.5 urn in aerodynamic
diameter, respectively.
In most cases emissions are proportional to activity level.
2-3
-------
Regional economic cycles in the construction industry impact construction PM
emission inventories. The factors that will cause the highest activity levels for construction
are low real interest rates, increasing economic growth, and some need for housing and
commercial structures (population growth is a strong predictor of need). A prediction of
future emissions must rely on economic and demographic forecasts for the inventoried
area.
Construction activity also varies temporally according to meteorology (rainfall stops
work), climate (unfavorable winter conditions impact work schedules), soil
characterization (compacted, rocky areas slow construction), workforce availability (labor
disputes halt construction), and economic conditions (effective demand).
Effective demand is defined as the combination of need for structures and roads, and
affordable resources (capital). Several socioeconomic forces affect the need for
construction, and are likely to impact regions and sub-regions unequally. Residential
construction is driven by localized population growth, low interest rates, and the quality of
current housing; on the other hand industrial construction is driven largely by economic
growth. In turn, highway construction is frequently driven by new residential and
commercial/industrial construction.
2-4
-------
Section 3.
Categories of Construction
Construction activities can be distinguished by three classes: (1) road construction,
(2) residential construction, and (3) nonresidential construction. Each is discussed below
to show the variations in emission producing activities.
3.1 Road Construction
Road construction includes the building of new roadways from all the functional
classes. The FHWA divides roads by purpose, lane width, number of lanes, surface type,
location (including urban, rural, state), and other roadway characteristics. The
characteristics of roadways vary depending on the type of roadway being constructed.
The road characteristics along with the new miles of roadway built on an annual basis
are used to determine the land area that is affected by construction for the type of road
being built. The three primary functional classes, arterials, collectors, and local roads, vary
in width, lanes, and may have further variations depending on whether the road is located
in an urban or rural area. Four divisions of roadways were made by functional class and
demographic type in order to group the roads by similar characteristics.
3.2 Residential Construction
The construction of houses and apartment buildings is included as a separate category
than other building construction primarily because of the statistics available for residential
construction. Statistics are available for the number of housing units constructed and also
the value of the construction.
Another variation is the level of activity that occurs at a residential construction site as
compared to other forms of building construction. Housing construction does not normally
require a large amount of earthmoving and occurs during a shorter time period, producing
less emissions per unit area than would be seen at a nonresidential construction site.
Apartment building construction lasts longer than housing construction.
MR[-AED\SECT-3.WPD
3-1
-------
3.3 Nonresidential Construction
Office buildings, warehouses, manufacturing facilities, schools, public works, and
hospitals are all included in nonresidential construction. Construction on nonresidential
sites is normally more involved and lasts longer than housing construction. It varies in the
amount of earthmoving that takes place but most nonresidential construction impacts a
similar amount of land on a per dollar basis.
3.4 Other Construction
Almost all construction activity can be included in either road, residential, or
nonresidential construction. Public projects in which a large amount of earthmoving and
building activity occurs (e.g., an expansive project such as a stadium or airport), should be
considered separately and emissions should be estimated using detailed construction data
from the engineering plans.
MRI-AED\SECT-3.WPD
3-2
-------
Section 4.
Existing Methodologies for Estimating Construction
Emissions
Many methodologies have been developed to calculate PM emissions from construction
activity. The basic limitations to developing a construction emissions methodology are how to
estimate the level of activity that occurs at a construction site and what emission factor is
appropriate to use to calculate PM emissions.
Two basic approaches are used in collecting data for the development of emission inventories:
(a) "top down" methodology; and (b) "bottom up" methodology. The "top down" method uses
national and state data resources to estimate activity levels that are multiplied by general
emission factors to calculate emissions for a large region. The calculated emissions are then
apportioned to more resolved areas, such as county and sub-county levels using surrogate activity
level data, such as population or affected land area. The "top down" method for estimating
construction operation emissions uses a single-valued, composite emission factor of 1.2 tons
TSP/acre/month, multiplied by estimated acres of construction (derived from construction cost
data) and an average duration for construction. The "top down" method is cost-effective, but
does not usually provide an accurate reflection of emissions when broken down into the county
and subcounty levels.
The "bottom up" methodology may use multiple emission factors (for specific construction
phases and activities) and local activity data to calculate emissions. Local data includes
equipment population levels, construction permit information, and specific factors that affect
construction activity for that area, including construction equipment usage. "Bottom up"
methods more accurately reflect the actual construction emissions than is represented using a
"top down" method, but are labor-intensive and costly. A "bottom up" emission inventory is
preferred for spatial and temporal allocation needed by modeling applications.
Existing methodologies for estimating PM emissions from construction activities are
described below and are mostly "top down" methods. Their advantages and limitations are also
explained.
4.1 Methodology 1: General "Top-Down" Emission Inventory
Most "top down" emission inventories of PM emissions from construction activities have
utilized the current composite AP-42 emission factor as follows:
EFPM-k - k x EFTSP
4-1
-------
where: k = fraction of TSP that is PM-k
EF = emission factor, 1.2 tons TSP/acre/month
This emission factor requires only that the activity level (acres of construction and duration of
the construction activity) be known for each type of construction. If construction activities are
controlled, a fractional control efficiency is utilized:
PM-k emissions = EFPM.k x acres of construction x months of activity x (1 - CE)
where: CE = fractional control efficiency
The acres of construction are determined, usually from a published relationship of
construction cost to acres disturbed. PM-k emissions are calculated by multiplying the TSP
emission factor of 1.2 tons/acre/month by the PM-k/TSP ratio, the total acres disturbed by the
construction activity and the months of activity. A control efficiency may be applied to reduce
emissions.
For example, the PM10 emissions inventory for the Southern California Air Quality
Management District's (SCAQMD) 1991 and subsequent 1994 Air Quality Management Plan
used a PM10 emission factor of 0.31 tons/acre/month. This factor was based on the TSP
emission factor of 1.2 tons/acre/month, a PM10 /TSP ratio of 0.52 (SCAQMD, 1991 and 1994),
and a 50% emission reduction to account for watering as a dust control measure.2
The ratios of PM10 /TSP and PM2 5/PM10 are important because of their use to project PM10
and PM2 5 emissions from TSP estimates. A typical ratio of 0.30 is used for PM10 /TSP. The
Criteria Document for Paniculate Matter (USEPA, 1996)3 indicates a ratio for PM2 5/PM10 of
approximately 0.15 for construction sites in Fresno, California. Other laboratory and field tests
have indicated ratios of crustal PM2 5 to PMjg in the range of 0.05 to 0.20, and are documented
by Cowherd and Kuykendal.4 They recommended a PM25/PM10 ratio of 0.15 for construction
operations because of the typical dominance of unpaved road emissions.
The information on the acres of land disturbed by construction activity can be obtained from
local government agencies and regional planning councils. Building permits usually specify the
area of land and/or the cost of the construction. Permits are typically issued by city or county
governments and require different levels of activity information.
The duration for an individual construction activity is likely to be identified in the building
permit. An average duration can also be estimated using the MRI-developed values of 6 months
for residential, 11 months for nonresidential, and 18 months for non-building construction.5
Construction activity information can also be obtained from two major national sources, the
U.S. Bureau of Census and from the McGraw-Hill Construction Information Group's Dodge
Construction Analysis System, an on-line service that provides monthly-updated construction
data for a fee.
4-2
-------
The disturbed area can be determined by using the cost of the construction activity and
published conversion factors for several construction types. This simple method uses the
aggregated cost of construction in an area which is available from the U.S. Bureau of Census,
Construction Statistics Division or from the U.S. Census Bureau's annual publication, Privately
Owned Construction Authorized by Building Permits. The dollars-to-acres conversion factors are
presented in Table 4-1 and are from the MRI report, Emissions Inventory of Agricultural Tilling,
Unpaved Roads and Airstrips, and Construction Sites.
Table 4-1. Construction Dollars-To-Acres Conversion Factors (MRI, 1974)5
SIC
code
1521
1522
1531
1541
1542
1611
1622
1623
1629
SIC description
General Contractors-Single-Family Houses
General Contractors-Residential Buildings, Other Than Single-Family
Commercial, Institutional, Manufacturing, and Industrial Buildings
General Contractors- Industrial Buildings and Warehouses
General Contractors- Nonresidential Buildings, Other than Industrial
Buildings
Highway and Street Construction, Except Elevated Highways
Bridge, Tunnel, and Elevated Highway Construction
Water, Sewer, Pipeline, and Communications and Power Line
Construction
Heaw Construction; Non-buildinq Structures Construction
Factor
(acres/$106)
5
5
5
5
5
25
25
5
150
Reference: Cowherd, Chatten, Christine Guenther, and Dennis Wallace. Emissions Inventory of
Agricultural Tilling, Unpaved Roads and Airstrips, and Construction Sites. EPA-450/3-
74-085, U.S. Environmental Protection Agency, Research Triangle Park, NC, November
1974.
Acres under construction, if obtained from construction cost data, are usually temporally
resolvable only to a monthly level. It is possible to extrapolate to a daily emission estimate by
dividing either annual or monthly emission estimates by the appropriate number of workdays in a
month.
Table 4-2 identifies the original data resources used by MRI for the estimation of
construction activity variables to support the methodology developed in 1974 for estimating
county-by-county construction activity levels and emissions. Annual TSP emissions were
estimated by MRI by determining the average construction duration (in months) for each type of
construction and multiplying by the monthly emission estimate.
4-3
-------
Table 4-2. Estimation of Construction Emissions—National Inventory by MRI
Variable
Statewide dollars spent on
construction
Dollars-to-acres conversion factors
County acres under construction
Average duration of construction
Data resource
U.S. Bureau of Census, Census of Construction 1972,
Developed by MRI using Census of Construction 1972.
U.S. Bureau of Census, Census of Construction 1972,
construction employment data.
Developed by MRI economists; 6 months for residential,
1 1 months for nonresidential, and 18 months for nonbuilding
construction.
Reference: Cowherd, Chatten, Christine Guenther, and Dennis Wallace. Emissions Inventory of
Agricultural Tilling, Unpaved Roads and Airstrips, and Construction Sites. EPA-450/3-74-
085, U.S. Environmental Protection Agency, Research Triangle Park, NC, November 1974.
Summary. Using a composite emission factor of 1.2 tons TSP/acre/month is believed to
overestimate PM10 emissions from construction activities. The emission factor assumes all
construction produces emissions at the same level on a per acre basis. The indicator for the level
of activity that occurs at construction sites, dollar value of construction, is a good indicator of
activity but conversion factors may not be accurate for converting dollar value to acres for all
types of construction. The emission factor and the conversion factors were developed in 1974
and require changes to reflect current construction activity and economic factors.
4.2 Methodology 2: NET Inventory
E.H. Pechan and Associates based the National Emission Trends (NET) inventory
methodology on the general methodology developed by MRI in 1974 for a national inventory to
estimate construction PM10 emissions. The activity level is acres under construction and is
estimated using construction expenditures by SIC code. The NET methodology is described
below, and differences from the MRI method (described in Section 4.1) are identified.
Section 4.8.2.7.1, "Construction Activities," of the National Air Pollution Emission Trends
Procedures Document for 1900-199$ gives the calculation methodologies for PM10 emissions
from construction activities for the years 1985 through 1996 and includes PM2 5 emissions for
1990 through 1996. In a manner patterned after Methodology 1, emissions were calculated from
the AP-42 composite emission factor, an estimate of the acres of land under construction, and the
average duration of construction activity. The acres of land under construction were estimated
from the dollars spent on construction. -
The 1985 through 1989 emission calculation procedure incorporated the general AP-42
emission factor for determining PM10 emissions for construction activities during that time
period:
E=Tx$xfxmxP
4-4
-------
where E = PM10 emissions
T = TSP emission factor (1.2 tons/acre)
$ = Dollars spent on construction ($ million)
F = Factor for converting dollars spent on construction to acres of construction
(varies by types of construction, acres/$ million)
M = Months of activity (varies by type of construction)
P = Dimensionless PM10/TSP ratio (0.22)
The 1990 through 1995 emission calculation procedure used the same basic equation but
also accounts for a control efficiency level and calculates both PM10 and PM2 5 emissions:
= Px$xfxmx(l-CE)
where E = PM emissions
P = PM emission factor (tons/acre of construction/month of activity)
(PM10 = 0.1 1;PM2>5 = 0.022)
$ = Dollars spent on construction ($ million)
F = Factor for converting dollars spent on construction to acres of construction
(varies by type of construction, acres/$ million)
M = Months of activity (varies by type of construction)
CE = Fractional control efficiency
Estimates for the dollars spent on various types of construction by EPA region for 1987 were
obtained from the Census Bureau. The fraction of the total U.S. dollars spent in 1987 for each : •
region for each construction type was calculated. Since the values from the Census Bureau .are
only available every five years, the Census dollars spent for the United States for construction
were normalized using estimates of the dollars spent on construction for the United States as
estimated by the F.W. Dodge Corporation for other years. This normalized Census value was
distributed by region and construction type using the previously calculated fractions.
Construction acres were calculated using the proportionality developed by MRI between the
number of acres and the dollars spent on that type of construction.5 This information
(proportioned to constant dollars using the method developed by Heisler)7 was utilized along
with total construction receipts to determine the total number of acres affected by each type of
construction type. Estimates of the duration (in months) for each type construction were derived
by MRI, from its 1974 report.5
The PM1(/TSP ratio for construction activities was derived from MRI research studies.
Pechan used PMjo/TSP ratios for 19 test sites for three different construction activities presented
in Table 9, "Net Particle Concentrations and Ratios" from the MRI Report "Gap Filling PM10
Emission Factors for Selected Open Area Dust Sources."8 This report suggests averaging the
ratios for the construction activity of interest. Since Pechan was looking at total construction
emissions, the average PM10/TSP ratios for all test sites were calculated and used for the
4-5
-------
PM10/TSP ratio. The PM10 emission factor 0.11 tons/acre/month is from the Best Available
Control Method (BACM) Report, Improvement of Specific Emission Factors.9 A particle size
adjustment of 0.2 was used to convert PM10 to PM2i5 emissions, after a review of PM2 5/PM10
ratios between EPA, Pechan, and MRL6 For the 1995 and 1996 NET inventories, the control
efficiencies used for PM10 and PM2 5 were 62.5 and 37.5 percent, respectively. No detail was
provided on the rationale for the control efficiencies. [Note: MRI has reviewed past test data and
found that the efficiency of watering, as a dust control method, is not related to the particle size
fraction (i.e., the control efficiency should be the same for both PMIO and for PM2 5).]
For the 1996 NET inventory, construction fugitive dust emissions were calculated from the
composite TSP emission factor prepared by MRI for EPA, with default EPA correction
parameters and 1996 Bureau of Census data. Controls were applied.10 The total emissions are
then allocated to the county level by county construction payrolls to develop a county-level
inventory. Table 4-3 summarizes the Pechan methodology to develop NET emissions from
construction activity.
Table 4-3. Estimation of Construction Emissions—EPA
National Emission Trends Analysis by E.H. Pechan and Associates
Variable
Statewide dollars spent on
construction
Dollars-to-acres conversion
factors
Average duration of
construction
Data resource
U.S. Bureau of the Census, Census of Construction Industries,
1987, and F.W. Dodge/McGraw Hill, Inc. construction data
(published annually).
Midwest Research Institute, Emissions Inventory of Agricultural
Tilling, Unpaved Roads and Airstrips, and Construction Sites,
November 1974.
Midwest Research Institute, Emissions Inventory of Agricultural
Tilling, Unpaved Roads and Airstrips, and Construction Sites,
November 1 974.
Reference: Barnard, William R., Allan Dean, and Patricia M. Carlson. Evaluation of Fugitive
Dust Emission Data, Draft Report, E.H. Pechan & Associates, October 11,1992.
Summary. The NET Inventory uses a "top-down" methodology and uses dollar value of
construction as an indicator of activity level. The dollar value is found for nine EPA regions, and
then emissions are allocated to the county level using county construction employment payrolls.
The allocation does not give a good estimate for the actual county construction emissions
because total emissions for the nine regions are divided among over 3,000 counties. The
dollars-to-acres conversion factors based on 1972 dollars have been changed to, reflect current
dollar value and give a better estimate of acres disturbed. The 1996 NET Inventory uses an
updated emission factor for construction activity and provides a better estimate of total PM10
emissions.
4-6
-------
4.3 Methodology 3: California Emission Inventory Procedure
The methodology used in the Emission Inventory Procedural Manual, Volume HI, Methods
for Assessing Area Source Emissions11 by the California Air Resources Board (CARB) is similar
to the NET methodology, but calculates residential acreage by unit rather than cost and estimates
for cost and number of units are from county sources.
The California manual's Section 7.7, "Building Construction Dust" presents a methodology
for calculating construction emissions from fugitive dust using the same emission factor as used
in the NET method plus a worst-case emission factor for heavy construction areas. The emission
factors used are from a 1996 MRI report9 in which an emission factor was developed using field
test observations from eight construction sites in Las Vegas and California. The factors account
for both exhaust emissions and fugitive dust emissions and do not account for any control
measures, as is standard for all AP-42 construction emission factors.
Because acres under construction are not readily available for a geographic region, it must be
estimated from either the value of construction or the units under construction. The CARB
methodology uses an acreage per dollar conversion factor and an acreage per unit conversion
factor to estimate total acres under construction. Residential construction acres are estimated on
an acres/unit basis with single-unit residential construction having a factor of 1/5 acre/unit in
rural areas and 1/7 acre/unit in urban areas. The factor for multi-unit residential construction is
estimated at 1/20 acre/living unit. Commercial construction is estimated to affect 3.7 acres for
every $1 million valuation. Likewise, industrial construction has a factor of 4.0 acres/$l million
valuation, and institutional construction a factor of 4.4 acres/$ 1 million valuation. The
California methodology assumes that the emission factor includes the effects of typical control
measures11 even though the MRI report lists the factors as uncontrolled.9 The procedure'manual
assumes a 50% control efficiency and recommends doubling the factor for areas in which
watering is not used to control fugitive dust. Table 4-4 provides the estimates for the activity
variables used in the California methodology.
Table 4-4. Estimation of Construction Emissions—California Methodology
Variable
Residential construction acres
Nonresidential construction acres
Construction duration
Data resource
Uses default for acres/residential unit: 1/7 acre for single-unit residences
in urban areas, 1/5 acre for single-unit residences in rural areas, and 1/20
acre/unit for multi-unit residences.
Uses default values for acres/$1 million of construction. The factors for
commercial, industrial, and institutional are 3.7, 4.0, and 4.4 acres/$1
million, respectively.
Uses default value of 6 months for single or multiple residential units and
1 1 months for commercial, industrial, and institutional construction.
Reference: Countess, Richard and Susan. PM10 Fugitive Dust Integration Project. South Coast AQMD Contract
96091, July 1996.
The California emission inventory includes a second section for calculating emissions from
road construction. Section 7.8, "Road Construction Dust," uses the same emission factors from
the BACM Report but uses different activity level indicators to find acreage disturbed. Road
4-7
-------
construction is divided into freeways, state highways, and city and county roads. The area
affected is calculated from the miles of road built and the number of lanes, lane width, and
shoulder width. The number of lanes, width per lane, and shoulder width are estimated for each
type of roadway and from these estimates an area per mile factor is determined. The values
determined in the California procedure are 12.1 acres per mile for freeways, 9.2 acres per mile
for highways, and 7.8 acres per mile for city and county roads. All road construction is assumed
to last 18 months.
The CARB uses a new computerized model, OFFROAD, to develop emission inventories of
PM from construction equipment exhaust activities.
Summary. The CARB methodology uses housing units as an indicator of activity level for
residential construction and dollar value for nonresidential construction. The dividing of the
construction types and the conversion factors used in the California methodology give a higher
level of accuracy to the estimate for the acres of land disturbed by construction. The CARB
methodology indicates that the emissions calculated are for fugitive dust only and the OFFROAD
model is used to estimate the construction equipment exhaust component. However, the
emission factors used in the California methodology were derived from site testing, which
includes both exhaust and fugitive dust. Thus the total PM emissions calculated by CARB for
construction may be too high if both the Area Source Methodology and the OFFROAD model
are used.
4.4 Methodology 4: National Particulate Inventory—Phase I
A national, county-level emission inventory of primary particulate (PM10 and PM2 5) was
prepared by E.H. Pechan and Associates under direction of EPA's Office of Policy, Planning,
and Evaluation (OPPE). The National Particulate Inventory (NPI) projected emissions to the
Year 2005 and utilized a methodology based largely on the methods used to develop the
1990 Interim Inventory, the NET inventory, and the 1985 NAPAP inventory.12 Details of the
methods were documented in a report to OPPE13 and summarized in a paper presented at the
1997 A&WMA annual meeting.12
The methodology to estimate emissions from construction activities used the composite TSP
factor of 1.2 tons/acre/month combined with ratios of PM10/TSP and PM2 5/PM10. The ratios
were stated to be derived from averages measured for three different construction activities at
19 sites.12
The activity level associated with the TSP factor is acres of land affected by the construction
activities. Activity level data for development of the NPI, in acres, were obtained for states in
each EPA Region from construction cost in the regional states. Construction cost date was used
to find acres disturbed by using the same methodology as the NET Inventory.
4-8
-------
State level emissions were allocated to county levels using construction payrolls from the
County Business Patterns database, which provides county, state, and national level business
data for 1977 to 1995. Statistics include number of establishments, payroll (annual and quarter),
number of employees, and number of establishments by size class for 2-digit SIC industry
groupings. The construction payroll data are collected annually by the Bureau of the Census.
Summary. The National Paniculate Inventory follows the same methodology as the NET
inventory and uses interim inventories to make future projections up to the Year 2005 for the
emissions produced by construction activity.
4.5 Methodology 5: Regional Emission Inventories
The AP-42 Section 13.2.3, "Heavy Construction," provides emission factors for estimating
site-specific construction emissions for specific construction phases (demolition, site preparation,
etc.). This effort requires knowledge of the type and duration of construction phases that occur at
each individual site. Examples of regional emission inventories of construction activities are
presented below, as originally prepared for the MRI1993 report, Activity Levels ofPM10 Area
Source Categories Methodology Assessment and Improvement.14 These approaches demonstrate
the use of local sources of construction activity level data.
4.5.1 San Joaquin Valley (SJV)
Activity levels could not be evaluated from the emission inventory report to the San Joaquin
Valley Unified Air Pollution Control District PM10 Nonattainment Area Plan, prepared by
Aerovironment, Inc., Monrovia, California, November 1991. The documentation of activity
levels was not included in the report. Section 3 of the report presents results from a 1990
emission inventory, citing that calculations were performed by the CARB. Appendix A of that
report presents the data from the CARB-developed emission inventory for the San Joaquin
Valley. Appendix C of the report presents the only description of methodology, saying "the
documentation of CARB methodology used for emissions inventory calculations was
inadvertently omitted from the appendices attached to the 1991 PM10 Attainment Plan and
accompanying this document. ARE has determined appropriate procedures for calculating each
emissions inventory category." The SJV activity levels estimates are shown in Table 4-5.
4-9
-------
Table 4-5. Estimation of Construction Emissions—SJV Methodology
Variable
Areas under construction
Data resource
GARB methodology was specified, but data sources not indicated.
Reference: San Joaquin Valley Unified Air Pollution Control District PM10 Nonattainment Area
Plan, Aerovironment, Inc., Monrovia, CA, November 1991.
4.5.2 South Coast Air Quality Management District (SCAQMD)
The SCAQMD used the composite AP-42 TSP emission factor for construction activities in
southern California. Activity data were presented in an MRI document15 that determined total
disturbed acres using the CARB methodology. Section 7-3, "Building Construction," presents
ratios of construction units or valuation to acres of construction for residential, commercial,
industrial, and institutional categories.
The number of construction units and value of construction were determined from the
U.S. Census Bureau's annual publication, Privately-Owned Construction Authorized by Building
Permits. It should be noted that U.S. Census Bureau data applies only to private construction.
Public construction works such as a city convention center, airport, or similar public works are.
not included. The SCAQMD methodology is summarized in Table 4-6.
Table 4-6. Estimation of Construction Emissions—SCAQMD Methodology
Variable
Units constructed and Value of
construction
Acres under construction
Construction duration
Data resource
U.S. Census Bureau, Privately Owned Construction
Authorized by Building Permits (an annual publication).
CARB Area Source Methodology, Section 7-3 Building
Construction; ratios of units or valuation to acres under
construction.
Used CARB defaults for months of construction.
Reference: Phil J. Englehart and Gregory E. Muleski. Open Fugitive Dust PM10 Control
Strategies Study, Midwest Research Institute: Kansas City, MO, October 12,1990.
Data are available for SIC 47457-residential; 47365-commercial; 47373-industrial; and 54551-
institutional construction.
A revised and more comprehensive emission inventory of SCAQMD construction sources
was prepared by Richard and Susan Countess in their 1996 report, PM-10 Fugitive Dust
Integration Project.2 This report presented two useful tables for preparation of emission
inventories. Table 4-7 shows a breakdown of construction activities and recommended that
individual AP-42 emission factors be used when the required activity levels are known—rather
than using the composite AP-42 TSP emission factor of 1.2 tons/acre/month. The recommended
4-10
-------
emission factors account for silt and moisture content, average wind speed, average vehicle
speed, the number of vehicles, and climate.
Table 4-7. AP-42 Recommended PM10 Emission Factors for
Construction Operations
Phase Actlvilv
1. Demolition 2. Demolition of buildings
and debris and natural obstacles
removal
Mechanical
dismemberment
Implosion of structure
Drilling/blasting soil
General land clearing
3. Loading and unloading of
debris into trucks
4. Truck transport of debris
3a. Unpaved road travel
3b. Paved road travel
2. Site 1. Bulldozing
Preparation
2. Scrapers unloading topsoil
3. Scrapers in travel
4. Scrapers removing topsoil
5. Loading/unloading trucks
6. Compacting
7. Motor grading
3. General 1a. Travel on unpaved roads
Construction
1b. Travel on paved roads
AP-42 recommended
emission factor reference
NA
NA
Drilling Factor in Table 1 1 .9-4
Dozer Equation (overburden)
in Tables 11. 9-1 and 11. 9-2
Material Handling Factor in
Sec. 13.2.2
Unpaved Road emission
factor in Sec. 13.2.2
Paved Road emission factor
In Sec. 13.2.2
Dozer Equation in Tables
11. 9-1 and 11. 9-2
Scraper unloading factor in
Table 11. 9-4
Scraper (travel mode)
expression in Tables 1 1 .9-1
and 11. 9-2
5.7 kg/vehicle kilometer
traveled (VKT)
Material Handling Factor in
Sec. 13.2.2
Dozer Equation in Tables
11.9-1 and 11. 9-2
Grading Equation In Tables
11. 9-1 and 11. 9-2
Unpaved Road emission
factor in Sec. 13.2.2
Paved Road emission factor
In Sec. 13.2.2
PM10 emission factor1
(uncontrolled omissions)
NA
NA
1.3
0.75 (s)1-5/^)1'4
0.001 1(U/5)1-3/(M/2)1'4
2.1(s/12)(S/30)(W/3)°-7
(w/4)°-5(365-p/365)
0.016(sL/2)°-6S(W/3)1-5
0.75(S)15/(M)1'4
0.04
0.0000037(S)U/(M)2'5
20.2
0.001 1(U/5)1-3/(M/2)1'4
0.75(S)1'5/(M)1-4
0.031 (S)2
2.1(s/12)(S/30)(W/3)°'7(w/4)0-5
(365-P/365)
0.0126(sL/2)°-65(W73)1-6
Units
Ib/hole
Ib/hr
Ib/ton
Ib/VMT
Ib/VMT
Ib/hr
Ib/ton
Ib/VMT
Ib/VMT
Ib/ton
Ib/hr
Ib/VMT
Ib/VMT
Ib/VMT
4-11
-------
Table 4-7 (Continued)
Phase Actlvltv
2a. Portable plants crushing
and screening
2b. Material transfers
3. Other operations
AP-42 recommended
emission factor reference
Factors for similar
material/operations In Section
11 of AP-42
Material Handling Factor In
Sec. 13.2.2
Factors for similar
material/operations In Section
11 of AP-42
PM10 emission factor1
^uncontrolled emissions)
Factors for similar
material/operations In Section 1 1
of AP-42
0.001 1(U/5)1'3/(M/2)1-4
Factors for similar
material/operations In Section 1 1
of AP-42
Units
Ib/ton
Note: s = silt content, %; M = moisture content %; U = mean wind speed, mph; S = mean vehicle speed, mph;
W = mean vehicle weight, tons; w = mean number of wheels/vehicle; sL - silt loading, g/m2, and p = number of days-
with at least 0.01* of precipitation.
Because the composite AP-42 emission factor for TSP can provide only a rough estimate of
PM10 emissions, MRI in their report to SCAQMD recommended alternative emission factors
based on four different levels of construction activity knowledge, as seen in Table 4-8 from the
report.
Table 4-8. Recommended PMin Emission Factors for Construction Operations0
Basis for emission factor Recommended PM10 emission factor
Level 1
Only area and duration known
Level 2
Amount of earth moving known, In addition to total
project area and duration
Level 3
More detailed Information available on duration of
earth moving and other material movement
Level 4
Detailed information on number of units and travel
distances available
0.11 ton/acre/month (average conditions)
0.42 ton/acre/month (worst-case conditions)8
0.011 ton/acre/month for general construction
(for each month of construction activity)
plus
0.059 ton/1,000 cubic yards for on-site cut/fill15
0.22 ton/1,000 cubic yards for off-site cut/fill"
0.13 Ib/acre-work hr for general construction
plus
49 Ib/scraper-hr for on-site haulage0
94 Ib/hr for off-site haulage"
0.13 Ib/acre-work hr for general construction
plus
0.21 Ib/ton-mile for on-site haulage
0.62 Ib/ton-mlle for off-site haulage"
Worst-case refers to construction sites with active large-scale earth moving operations.
These values are based on assumptions that one scraper can move 70,000 cubic yards of earth in one month and one truck
can move 35,000 cubic yards of material In one month. If the on-site/off-site fraction Is not known, assume 100% on-site.
If the number of scrapers in use is not knows, MRI recommends that a default value of 4 be used. In addition, if the actual
capacity of earth moving units Is known, the user is directed to use the following emission rates in units of Ib/scraper-hour for
different capacity scrapers: 19 for 10 yd3 scraper, 45 for 20 yd3 scraper, 49 for 30 yd3 scraper, and 84 for 45 yd3 scraper.
Factor for use with over-the-road trucks. If "off-highway" or "haul" trucks are used, haulage should be considered "on-site".
c Some emission factors were revised by Countess based on median rather than mean values.
4-12
-------
4.5.3 Phoenix
Construction activity levels for the Maricopa planning area were determined from the
document, PM]0 Emissions Inventory Data for the Maricopa and Pima Planning Areas .^ The
Maricopa County Air Pollution Bureau provided information on construction and earth moving
permits, allowing location and area size to be tabulated. Information on permits is variable since
each local governmental entity in the Phoenix metropolitan area establishes the information
needed.
The Maricopa County Bureau of Air Pollution Control prepared a listing with addresses of
approximately 1,500 earthmoving permits issued over a one-year period. Using a street atlas,
each address for an earthmoving permit was manually located on a map of the inventoried area.
Individual earthmoving permits listed the areas of disturbed earth and the lineal feet of trenching.
A 20-ft width for each trench was assumed to allow calculation of area (acres for each disturbed
site. All construction projects were assumed to have a 4-month duration so that a
tons/acre/month inventory could be developed. An emission factor of 900 Ib PM jo/acre was
used, and appeared to be derived from the composite AP-42 TSP emission factor. The Phoenix
methodology is summarized in Table 4-9.
Table 4-9. Estimation of Construction Emissions—Phoenix Methodology
Variable
Acres under construction
Construction duration
Data resource
Earthmoving permits from the Maricopa County Bureau of Air Pollution
Control and the Pima County Air Quality Control District. Street addresses
on permits were used to geographically map construction areas;
approximately 1 ,500 permits had to be addressed. The permits listed
acres of disturbed land and lineal feet of trenching; it was assumed that the
disturbed width of trenches was 20 ft.
All construction projects were assumed to have a 4-month duration so that
a tons/acre/month emission rate could be developed.
Reference: Donald R. Holtz. PM10 Emissions Inventory Data for the Maricopa and Pima Planning
Areas, Engineering-Science; Pasadena, CA, January 1987.
4.5.4 Power/Bannock Counties
Construction-related emissions in an Idaho PM-10 nonattainment area were divided into
(1) residential and commercial construction, and (2) road construction by Moore and
Balakrishna.17 They used AP-42 emission factors for construction activities, but devised unique
ways to apportion emissions to smaller county areas (grid cells) for modeling purposes.
Residential and commercial construction activities were allocated to specific cells using
U.S. Census tract data. Households were divided into low-, medium-, and high-growth areas,
excluding urban areas. The numbers of households in each growth area were totaled and then
divided by the total number in all three growth areas to obtain the percentage of households in
each area. It was assumed that this percentage also applied to the number of construction events,
4-13
-------
and subsequently the percentage of emissions from construction. The calculated emissions for
each growth area were divided equally among the total number of cells in each growth area.
Road construction activities were divided" into (1) graveling, (2) rebuilding, (3) paving, and
(4) sealing. Each activity was defined in terms of actual "road miles of construction" and
"width" of the roads under construction. Road miles were multiplied by the road width that
resulted in total acres of road being constructed. For road paving and sealing, the emission rate
[factor] was reduced to half that of road graveling and rebuilding.
4.5.5 Las Vegas (Clark County, Nevada)
4.5.5.1 1991 Methodology
The 1991 emission inventory methodology began with the composite AP-42 TSP emission
factor for construction activity. Activity levels for the Las Vegas, Nevada, nonattainment area
were determined using the methodology presented in the document Air Quality Implementation
Plan for the Las Vegas Valley Particulate Matter PM10.18 The primary piece of information was
the total acres of construction, which was obtained from Topsoil Disturbance Permits from the
Clark County Health District. Clark County requires Topsoil Disturbance Permits for land
development activities affecting areas of 1/4 acre or more in size. These data are entered into the
Clark County Geographic Information System (GIS) to calculate the total number of acres
impacted by construction activities. There was no distinction between types of construction for
the Las Vegas Valley.
PM10 emissions for Clark County were calculated using two components: (a) acres of
construction, and (b) an emission factor of 654 Ib PM10 per acre. A surrogate activity level factor
was 1,000 gal diesel fuel/acre of construction, and resulted in a surrogate emission factor of
21.9 Ib PM10 per 1,000 gal diesel fuel. These factors were "taken from research activities
conducted in Arizona," and were not referenced or discussed further in the reviewed document.
The 1991 Las Vegas methodology is summarized in Table 4-10.
Table 4-10. Estimation of Construction Emissions 1991
Las Vegas Methodology
Variable
Gallons of diesel fuel
Acres under construction
Data resource
Estimate of 1 ,000 gallons of diesel fuel used in construction per acre of
construction impacted land. This estimate was developed from a literature
review that was referenced, but hot discussed in the document.
Clark County Health District Top Soil Disturbance Permits issued. Permits
are required for any land development activity affecting more than one
quarter of an acre. Permit data is entered into Clark County GIS for spatial
distribution to each of 1 6 planning grids.
Reference: Clark County Department of Comprehensive Planning. Air Quality Implementation Plan for the
Las Vegas Valley: Particulate Matter PM10, Las Vegas, Nevada, November 5,1991.
4-14
-------
4.5.5.2 1998 Methodology19
The 1998 methodology to estimate annual PM10 emissions for the year 1995 was improved
by staff from Clark County and considered three different sources of emissions during
construction operations. "Construction activities" included grading, trenching, crushing,
screening, on-site vehicle traffic, blasting, and demolition. A modified BACM9 Level 1
methodology was used to estimate PM10 emissions, and required only the amount of land
involved and the duration of the project, as separated into "large" and "remaining" projects. The
average time to complete construction projects was defined as the number of months from initial
ground breaking to final landscaping and paving.
A recommended BACM emission factor of 0.42 tons/acre/month was used for general
construction sites that included cut and fill areas, large-scale earthmoving operations, or heavy
traffic volumes. The BACM report also recommended an uncontrolled emission factor of
0.11 tons/acre/month for general construction sites that did not include any cut and fill areas,
large-scale earthmoving operations, or heavy traffic volumes. Clark County judged that
"remaining" projects (i.e., commercial, public parks, public buildings, residential homes, and
miscellaneous) sometimes included cut and fill areas, large-scale earthmoving activities, and/or
heavy traffic volumes.19 Consequently, an average emission factor of 0.265 tons/acre/month
[(0.42 + 0.11) / 2] was used for all construction projects other than "large" projects.
A control efficiency of 50 percent was applied because of local watering regulations, and
using the control efficiency described in MRI's 1988 study for U.S. EPA OAQPS, "Control of
Open Fugitive Dust Sources."20 The control efficiency was then decreased by the percentage of
construction sites implementing dust control, as estimated by air quality compliance officers.
"Track-out" dealt with increased paved road dust emissions due to dirt track-out from the
construction site onto the adjacent paved street network. Track-out emissions were estimated for
each type of construction using an estimated number of access points and vehicle traffic volumes
on adjacent paved roadways. The number of access points ranged from 1 per 10 acres to 1 per
30 acres. Traffic that exited the access points was estimated at greater than 25 vehicles per day
and corresponded to the associated emission factor. PM10 emissions from track-out were based
on 13 grams/vehicle times the number of vehicle passes per day on the adjacent paved road, as
recommended in the 1988 MRI report for EPA, "Control of Open Fugitive Dust Sources."20
Traffic on adjacent paved roadways was estimated at 2,157 trips per day and was determined to
match those from collector streets. This resulted in PM10 emissions of 0.0309 ton/day (except
for public parks), from each track-out/access point. A control efficiency of 75 percent was stated
to be determined from compliance rates for street sweeping and watering.
"Wind erosion" emissions from land exposed by construction activities were separately
estimated. The methodology was based on geometric mean hourly emission rates from disturbed
soils within the Las Vegas Valley, as reported in 1996 by David James, "Estimation of PM10
Emissions from Vacant Lands in the Las Vegas Valley." Wind speed dependent emission rates
4-15
-------
in tons/acre/hr were developed for nine wind speed classes (> 15 mph). These rates were
adjusted for vegetative cover and for loss of loose surface material in an initial wind "spike."
The annual number of hours of wind in each wind speed category for the year was then
multiplied by the emission factor in tons/acre/hour of wind. This calculation produced a PM10
emission factor of 0.4472 tons/acre for 1995, and was applied to the permitted acres of
construction in Las Vegas during that same year.
Summary. Regional emission inventories use more detailed information than is normally
available at a national level for estimating county-level construction emissions. The
methodologies do provide estimates that can be compared to estimates found using a composite
emission factor to determine county-level emissions.
4.6 Methodology 6: Major Construction Project Inventory
A general conformity analysis of construction emissions associated with a major
construction project provides a detailed and systematic procedure for inventorying fugitive dust
PM10 emissions.21 This large project [presumably for enlargement of an Arizona airport]
consisted of seven construction phases: (1) first building; (2) second building; (3) parking lot;
(4) fire station; (5) fuel storage facility; (6) maintenance hangar; and (7) large pavement project.
The inventory team used a spreadsheet to organize input data and calculate emission
estimates. Data that were available to estimate PM10 emissions from the large construction
project included:
• Project timelines and activity schedules
• Area and access points to the construction site
• Types of construction equipment
• Characterization of construction activities
• Quantities of material to be moved, crushed, and screened
• Precipitation and wind data
• Equipment speed and miles traveled
• Soil silt fraction and moisture content
The authors of the general conformity analysis stated that "Exhaust emissions associated
with the construction activities have not been included." While this is true for generators and
other stationary equipment, it is not true for AP-42 emission factors for PM from construction
equipment activity. The emission factors for fugitive dust from construction equipment represent
both exhaust and fugitive dust emissions because of the source profiling test method used by
MRI to develop the AP-42 factors.
Direct PM emissions were estimated from demolition, site preparation, general construction,
truck transport of debris, bulldozing, compacting, etc. Indirect emissions from transport and
4-16
-------
unloading of material to/from the construction site were also estimated. This included VMT
estimates for paved road travel both on-site and off-site. Track-out emissions and wind erosion
emissions from unpaved surfaces were also estimated. Wet suppression of fugitive dust sources
was incorporated into the emission calculations using a control efficiency of 80 percent.
Summary. The detailed inventory done based on "unit-operation" emission factors is useful
in determining the accuracy of emissions calculated for different types of construction activity
using an emission factor for a specific type of construction and in determining which types of
construction activity produce what amounts of emissions.
4.7 Methodology 7: U.S. EPA NONROAD Model
The U.S. EPA Office of Mobile Sources, Assessment and Modeling Division has developed
a model for estimating non-road engine exhaust emissions. A second draft version of the
NONROAD model was announced May 21,1999 with the signing of the Tier 21
Gasoline Sulfur Notice of Proposed Rulemaking. The model is available at
http://www.epa.gov/oms/nonrdmdl.htm.
Construction equipment exhaust emissions are calculated using national or state engine
population for each equipment/engine type. The engine populations are obtained from the
PartsLink database available from Power Systems Research (a commercial source of data), and'
multiplied by the average power, activity, and emission factors to obtain pollutant emissions. The
NONROAD model estimates exhaust emissions under "load" and "no load" conditions. Engine
load is related to soil density, cycle time (distance/speed), and pull required (rolling resistance +
grade resistance.) The following equation shows how NONROAD calculates emissions.
Emissions = (Pop)(Power)(LF)(UL)(EF)
where: Pop = Engine population
Power = Average power of equipment type (hp)
LF = Load Factor (fraction of available power)
UL = Usage level (hrs/yr)
EF = Exhaust emission factor (g/hp-hr)
This equation shows that the NONROAD model uses a multi-parameter activity level
combining engine population number with average power, load factor, and usage level. The
primary element is the number of engines in an area, distributed by age, power, fuel type, and
application. Each equipment/engine type is characterized for usage by horsepower-hours per
year, and adjusted for a power load factor. Nationally-averaged horsepower-hours and the
relative fraction of maximum available power are used.
4-17
-------
The most important data for construction activity levels that are input to the NONROAD
model originate from the 1996 PSR equipment population data (PartsLink), and revised
population allocation data using the F.W. Dodge construction valuation data. Engine populations
are divided into several discreet power levels rather than one average power level for each
equipment application. Equipment populations are adjusted using the F.W. Dodge construction
valuation data. An engine scrappage rate is assumed and the level of activity is a function of
equipment age. The model is flexible and allows a "bottom up" approach with locally-derived
estimates for all variables to estimate and allocate emissions from state to counties and sub-
counties.
NONROAD input files are integral to the model and provide basic data by state and county
that are required to calculate emissions: exhaust emission factors, base year equipment
population, activity levels, load factor, average lifetimes, scrappage rates, growth estimates, and
geographic and temporal allocation algorithms. Default values are provided in these input files,
but the user can replace the default data with better information, either from EPA for national
defaults or from local sources for locality-specific data. The input files can also be modified to
test control strategies.
The NONROAD model can estimate current year emissions for a specified geographic area
as well as project future year emissions and backcast past year emissions. Emissions can also be
calculated for time periods—an entire year, one of the four seasons, or any particular month. The
emissions are then temporally and geographically allocated using appropriate allocation factors.
One of the current shortfalls of the NONROAD model to predict emission estimates for
construction activities is that the model accounts for only exhaust emissions from construction
equipment. A simple correlation of fugitive dust emissions with exhaust emissions is not
possible. For example, construction equipment will be under load at the earth cutting location
and will emit high levels of exhaust emissions, but little fugitive dust will be generated because
of typical sub-grade high moisture content. As the loaded equipment travels to the fill location,
high levels of fugitive dust will be emitted from the exposed ground but the equipment may not
emit high levels of exhaust emissions.
An EEA report of 1997 developed data on construction employees to scale equipment
population as a function of construction employees, but this method did not include all types of
construction activity. Sierra Research, S AI, ENVIRON, and the Texas Transportation Institute
also have examined and used survey methods for obtaining information on construction
equipment usage for input to the NONROAD model. Survey data of current construction
projects were needed to provide location-specific data on a daily level.
The EPA model, PARTS, was developed by the Office of Mobile Sources (QMS) to estimate
PM emissions from only onroad vehicles, and is discussed here for background information and
comparison of vehicle emission estimation methodologies. The name indicates consistency with
the MOBILES model used to calculate emissions of other pollutants from onroad vehicles.
4-18
-------
PARTS uses PM emission factors for direct and indirect sulfate, and carbon (soluble organic
fraction and remaining carbon portion) to calculate exhaust emissions. Road dust, tire wear, and
brake wear emissions are also calculated. The PARTS program uses VMT to calculate PM
emissions in gram/mile. VMT data are obtained using onroad travel statistics available from
local survey information maintained by state and local transportation agencies and assembled by
the FHWA. VMT data are not collected for non-road sources, such as construction equipment.
Summary. The NONROAD model estimates PM emissions only from construction
equipment exhaust. The model is useful to determine the exhaust emission component of the
total emissions calculated using the AP-42 emission factor that includes both suspended dust and
exhaust PM. The PARTS model does not apply to construction activities because it estimates
vehicle exhaust emissions from onroad vehicles only.
Methodologies 1 through 6 are summarized in Table 4-11.
4-19
-------
Table 4-11. Emission Inventory Methodologies
Emission calculation Parameters
Inventory
MRI National Inventory, 1974
National Emission Trends
National Participate Inventory
California Air Resources Board
(CARB)
South Coast Air Quality
Management District
San Joaquin Valley
Las Vegas (Clark Co., NV)
1991
Las Vegas (Clark Co., NV)
1997
Phoenix
Power/ Bannock, 1996
Emission Factor
1.2 tons/acre/month (TSP) AP-42
1 .2 tons/acre/month (TSP) AP-42 adjusted
toPM10andPM25 pw(4^,Z<»
1.2 tons/acre/month(TSP), AP-42; used
PM1(/TSP and PM^5/PM10 ratios derived
from EPA "Gap Filling PM10 Emission
Factors for Selected Area Dust Sources"
1.2 tons/acre/month (TSP) AP-42 adjusted
toPM10andPM25
0.31 tons PM10/acre/month (based on
AP-42 TSP emission factor)
CARB Methodology
654 Ib PM10/acre (activity) plus 21.9 lt>
PMin/1000 gal diesel fuel (equipment)
Heavy Construction — 0.42 tons/acre/mo.;
Other Construction — 0.265 tons/acre/mo.;
Track-Out— 0.0309 ton/day/access pt.
(based on traffic volume of 2,157 trips/day)
Wind Erosion — 0.4472 ton/acre, dependent
on 1995 windspeeds
900 Ib PMln/acre
Activity Level Source
Construction Dollars and dollars to acres
conversion factors(MRI developed factors,1972)
Construction Dollars and dollars to acres
conversion factors(MR) factors, adjusted using
Heisler's method)
Emissions and methods derived from 1993 National
Construction Dollars or Number of Units
Constructed; CARB conversion factors for dollars to
acres and units to acres
CARB Methodology
CARB Methodology
Top Soil Disturbance Permits for acres disturbed
Topsoil Disturbance Permits for acres disturbed; .
other local data from air quality and metropolitan
agencies
Earth Moving Permits for acres disturbed
1.2 tons/acre/month (TSP) AP-42 recommended emission factor
Notes
MRI durations of construction: 6 months
residential, 11 months nonresidential, 18
months nonbuilding
MRI durations: 6 months residential, 1 1
months nonresidential, 18 months
nonbuilding
Emission Trends Inventory
CARB Default Values: 6 months
residential, 1 1 months commercial,
industrial, and institutional
CARB Defaults for Construction
Duration
CARB Methodology
Conversion of 1 acre of construction
impacted land to 1000 gal. of diesel fuel
See text
4 months for all construction projects
4-20
-------
Sections.
Recommended Methodologies and Data Sources
This section presents an improved emission inventory procedure that calculates both
exhaust and fugitive dust emissions from construction activities. The recommended
procedure provides a balance between a "top-down" inventory and "bottom-up" inventory
methodology. PM emissions at the county level are more accurately estimated for different
types of construction operations using improved indicators of activity levels.
5.1 Assumptions and Limitations of Current Methodologies
The NET procedure and the CARB methodology both make assumptions and also use
estimates that may no longer be applicable because of the date of their development. The
NET methodology uses a single, composite emission factor for all types of construction
based only on the dollar amount spent on construction. The first assumption is that all
construction activity produces the same amount of dust on a per acre basis. The amount of
dust produced is not dependent on the type of construction but merely on the area of land
being disturbed by the construction. A second assumption is that land affected by
construction activity is always affected the same amount, i.e., the methodologies do not
account adequately for large scale cut and fill operations. Also, the methodologies are
limited in that the conversion factors used to convert dollars spent on construction to
acreage disturbed, along with the estimates for the duration of construction activity, were
developed by MRI in 1974 and may result in a loss in reliability in calculating emissions.
5.2 Recommended Changes to the NET Methodology
MRI recommends the following changes to the current NET methodology. Following
the California methodology, residential construction acreage should be based on the
number of units constructed rather than the dollar value of construction. Accounting for
the construction of foundations is also seen as a necessary change because of the difference
in the amount of dirt moved when constructing a slab foundation as compared to a
basement. Highway construction with significant cut and fill operations should be based
on the new miles of highway constructed in eVh county. The control efficiency used in the
1996 Trends inventory for PM10 was 62.5% and was 37.5% for PM2 5. MRI recommends
using a control efficiency of 50% for both PM^ and PM2 5 for areas in which dust control
measures are used. The estimates for the duratibn of construction activity levels also need
to be revised for each construction category.
5-1
-------
5.3 General Emission Factor for Construction
Construction emissions can be estimated when two basic construction parameters are
known, the acres of land disturbed by the construction activity and the duration of the
activity. As a general emission factor for all types of construction activity, MRI
recommends using 0.11 tons PM10/acre/month that is based on a 1996 BACM study by
MRI prepared for the California South Coast Air Quality Management District
(SCAQMD).9 However, separate emission factors segregated by type of construction
activity provide better estimates of PM10 emissions and give a more accurate estimate than
could be obtained using a general emission factor. Specific emission factors and activity
levels for residential, nonresidential, and road construction are described below.
5.4 Residential Construction Emissions
Residential construction emissions are calculated for three basic types of residential
construction:
• Single-Family Houses
• Two-Family Houses
• Apartment Buildings
5.4.1 Emission Calculation Procedure
Emissions for housing construction activities are estimated using emission factors
from the MRI BACM report.9 Housing construction emissions are calculated using an
emission factor of 0.032 tons PM10/acre/month, (as recommended by the SCAQMD2), the
number of housing units created, a units-to-acres conversion factor, and the duration of
construction activity. The formula for calculating emissions from residential construction
is:
Emissions = (0.032 tons PMt(/acre/month) x B x f x m
where: B = the number of houses constructed
f = buildings-to-acres conversion factor
m= the duration of construction activity in months
Figure 5-1 illustrates the calculation of residential construction emissions.
5-2
-------
Residential Construction
Number of houses
constructed
Units to acres
conversion factors
Duration of
construction
activity
Emission factor of
0.032 tons/acre-
month
Uncontrolled PM-10
emissions from
residential
construction
Figure 5-1. Residential Construction Emissions Flowchart
Apartment buildings vary in size, number of units, square footage per unit, floors, and
many other characteristics. Since these variations exist and most apartment buildings
occupy a variable amount of space, a dollars-to-acres conversion is recommended for
apartment building construction rather than a building-to-acres factor. The estimate of
2.0 acres/$106 (in 1992 constant dollar value) is recommended to determine the acres of
land disturbed by the construction of apartments. The dollars-to-acres conversion factor
was updated to a 1992 constant dollar value using the Construction Cost Index found in the
annual edition of Statistical Abstract of the United States. A new estimate for the acres
under construction per million dollars was developed using the difference in the 1992
index value and an estimated 1974 value. The approximately 40% difference led to an
updated factor of 2 acres/$106 derived from the original 5 acres/$106 developed by MRI in
1974, The emission factor recommended for the construction of apartment buildings is
0.11 tons PM10/acre/month because apartment construction does not normally involve a
large amount of cut-and-fill operations.
An alternative formula is recommended for residential construction in areas in which
basements are constructed or the amount of dirt moved at a residential construction site is
known. The F.W. Dodge reports give the total square footage of homes for both single-
family and two-family homes. This value can be used to estimate the cubic yards of dirt
moved. Multiplying the total square feet by an average basement depth of 8 ft. and adding
in 10% of the cubic feet calculated for peripheral dirt removed produces an estimate of the
cubic yards of earth moved during residential construction. The added 10% accounts for
5-3
-------
the footings, space around the footings, and other backfilled areas adjacent to the basement.
The cubic yards of earth moved along with the number of houses constructed can be used
with the BACM Level 2 equation (emission factor of 0.011 tons PM10/ acre/month plus
0.059 tons PM10/1000 cubic yards of on-site cut/fill) to calculate emissions for regions in
which basements are constructed or a large amount of dirt is moved during most residential
construction. The Level 2 equation produces a slightly higher estimate of PM10 emissions
than would be estimated using the residential construction emission equation.
5.4.2 Data Sources and Assumptions
The information available to determine activity level of residential construction is the
dollar value of construction put in place and the number of units constructed. Construction
costs vary throughout the United States and residential construction characteristics do not
show as much variance as the cost does, so the number of units constructed is a better
indicator of activity level. The amount of land impacted by residential construction is
determined to be about the same on a per house basis rather than a per dollar basis. The
average 2000 sq. ft. home can vary from the low to upper $ 100,000s depending on where
the home is located in the United States. Incorporating a dollars-to-acres conversion factor
would give a larger estimate for the acreage of land disturbed even though the construction
affects the same amount of land as an area with a lower dollar value for residential
construction and vice versa.
The number of housing units constructed by a county or state are available from the
F.W. Dodge's "Dodge Local Construction Potentials Bulletin." Housing units are
available for the three types of residential construction previously mentioned.
The conversion for single-family housing is estimated to be 1/4 acre per house. The
conversion factor was determined by finding the area of the base of a home and estimating
the area of land affected by grading and other construction activities beyond the "footprint"
of the house. The average home is around 2000 sq. ft. Using a conversion factor of
1/4 acre/house indicates that five times the base of the house is affected by the construction
of the home. This estimate is reasonable when considering the amount of grading, cut and
fill, and transportion of materials on the property that occurs during residential
construction.
The conversion for two-family housing was found to be 1/3 acre per building. The
1/3 acre was derived from the average square footage of a two-family home, around
3500 sq. ft., and the land affected beyond the base of the house, about 4 times the base for
two-family residences.
5-4
-------
5.4.3 Example Emission Calculation
Table 5-1 presents an example calculation of county-level emissions for residential
construction.
Table 5-1. Example Annual PM10 Emissions from Residential Construction in a
Hypothetical County
Residential
type
Single-
family
Two-family
Apartment
Total
No. of
buildings
2422
48
59
Acreage
per
building
1/4
1/3
1/2
Total
Acres
disturbed
606
16
30
Duration of
construction
6
6
12
Emission
factor
(tonsPM10/
acre/month)
0.032
0.032
~OQ92-
D'll
Uncontrolled
PM10 (tons)
116
3.1
us
130
PM10
control
efficiency
(%)
0
0
0
Controlled
PM10 (tons)
116
3.1
113
130
A comparison of emission calculations using unit-operation emission factors, the
residential construction emission equation, and the BACM Level 2 calculation shows that
the Level 2 equation provides a higher estimate of emissions than using the general
residential emission factor. The unit-operation emission calculation for bulldozing and
grading produces an estimate similar to that fronVthe Level 2 equation. The general
residential emission factor calculates PM10 emissions from the construction of one single-
family home to be 96 Ibs/house. The Level 2 equation for a single-family home with a
basement produces emissions of 109 Ib PM10/house. The emission calculation for
bulldozing and grading estimates emissions to 112 Ib/house PM10 (assuming 10 days of
operation, 8%'silt content, and 6% moisture content).
The comparison of residential construction emission methods for one single-family home
were based on typical parameters for a single-family home:
area of land disturbed
area of home
duration
basement depth
moisture level
silt content
1/4 acre
2000 sq.ft.
6 months
8 ft.
6%
8%
Residential construction emission factor calculations are shown below. The general
residential calculation is:
0.032 tons PM10/acre/month x 1/4 acre x 6 months = 0.048 tons or 96 Ib PM10
5-5
-------
The BACM Level 2 emission calculation is:
Cubic yards of dirt moved: 2000 ft2 x 8 ft. x 110% = 17600 ft3 = 652 yd3
(0.011 tons PM10/acre/month x 1/4 acre x 6 months)-*-
(0.059 tons PM jo/1000 yd3 dirt x 652 yd3 dirt) =
0.016 +0.038 = 0.0545 tons or 109 Ib PM10
33 *1<°
The Unit Operation Emissions (Bulldozing) calculation from AP-42 is:
PM10 = 0.75 (s)L5/(M)u = 0.75 (8)U/(6)L4
= 1.4 Ib PM10/hr x 10 days x 8 hours = 112 Ib PM10
5.5 Nonresidential Construction Emissions
Nonresidential construction includes building construction (commercial, industrial,
institutional, governmental) and also public works.
5.5.1 Emission Calculation Procedure
The emissions produced from the construction of nonresidential buildings are
calculated using the value of the construction put in place. The formula for calculating the
emissions from nonresidential construction is:
Emissions = (0.19 tons PMt(/acre/month) x $ x f x m
where: $ = dollars spent on nonresidential construction in millions
f = dollars-to-acres conversion factor
m = duration of construction activity in months
Figure .5-2 illustrates the calculation of PM10 emissions from non residential
construction.
The emission factor of 0.19 tons PM10/acre/month was developed using a method
similar to a procedure originated by Clark County, NV (Las Vegas) and the emission
factors recommended in the MRI BACM Report.9 A quarter of all nonresidential
construction is assumed to involve active earthmoving in which the recommended emission
factor is 0.42 tons PM10/acre/month. The 0.19 tons PM10/acre/month was calculated by
taking 1/4 of the heavy construction emission factor, 0.42, plus 3/4 of the general emission
factor 0.11 tons/acre/month. The 1/4:3/4 apportionment is based on a detailed analysis of a
Phoenix airport construction where specific unit operations had been investigated for PM10
emissions21.
5-6
-------
Nonresidential Construction
Dollars spent on
nonresidential
construction
Dollars to acres
conversion factor
Duration of construction
activity
Emission factor of
0.19 tons/acre-month
)\
Uncontrolled PM-10
emissions from
nonresidential construction
activity
Figure 5-2. Nonresidential Construction Emissions Flowchart
Regions known to have extensive earthmoving activities will produce higher amounts
of PM10 emissions. Since this larger amount would not be accounted for in building
construction, the B ACM "heavy construction emission factor" of 0.42 tons
PMjQ/acre/month may provide a better estimate for areas in which a significant amount of
earth is disturbed.
An emission inventory for a 114-acre airport project21 provides a comparison of
detailed PM10 emissions as contrasted with the new recommended PMj0 emission factor of
0.19 tons/acre/month. The results show total uncontrolled PM10 emissions using the
detailed unit operation emission inventory methodology is 210 tons PM10 for the duration
of the construction. The proposed emission factor results in total uncontrolled PM10
emissions of 260 tons PMI0. The new factor along with the acres under construction as an
indicator of activity level provides an estimate of PM10 emissions from nonresidential
construction within 25% of the emissions calculated using detailed engineering plans and
"unit-operation" emission factors.
5.5.2 Data Sources and Assumptions
The dollar amount spent on nonresidential construction is available from the U.S.
Census Bureau, Census of Construction Industries and the Dodge Construction Potentials
Bulletin. Census data are divided by SIC Code whereas the Potentials Bulletin divides
activity by the types of building being constructed rather than by SIC Code.
5-7
-------
MRI has determined that the previous 1974 dollars-to-acres conversion factors can be
updated to a single factor for nonresidential, nonroad construction. It is estimated that for
every million dollars spent on construction, in 1992 constant dollars, 2 acres of land are
impacted. The conversion factor reflects the current dollar value using the Price and Cost
Indices for Construction that are available from the Statistical Abstract of the United States,
published yearly. For example, the 1997 dollars-to-acres conversion factor would be
2/(l 18.7%) or 1.7 acres/ $ 106. The estimate for the duration of nonresidential construction
is 11 months.
5.5.3 Example Emission Calculation
Table 5-2 presents an example calculation of 1992 PM10 emissions from
nonresidential, nonroad construction for a hypothetical county.
Table 5-2. Example 1992 PM10 Emissions for Nonresidential Construction in a
Hypothetical County
Construction
put in place
($106)
57.7
1992($to
acres)
2acres/$106
Acres
. disturbed
115
Duration of
activities
11
PM10 emissions factor
(tons/acre/month)
0.19
Uncontrolled
PM10(tons)
240
5.6 Roadway Construction Emissions
Roadway construction emissions are highly correlated with the amount of eaithmoving
that occurs at a site. Almost all roadway construction involves extensive earthmoving and
equipment travel, causing emissions to be higher than found for other construction types.
5.6.1 Emission Calculation Procedure
The PM10 emissions produced by road construction are calculated using the BACM
recommended emission factor for heavy construction and the miles of new roadway
constructed. The formula used for calculating roadway construction emissions is:
Emissions = (0.42 tons PMj (/acre/month) x M x f x d
where: M = miles of new roadway constructed
f = miles-to-acres conversion factors
d = duration of roadway construction activity in months
5-8
-------
The emission factor of 0.42 tons/acre/month is used to account for the large amount of
dirt moved during the construction of roadways. Since most road construction consists of
grading and leveling the land, the higher emission factor more accurately reflects the high
level of cut and fill activity that occurs at road construction sites. Figure 5-3 illustrates the
calculation of road construction emissions of PM10.
Road Construction
Miles of roadway
constructed
Slate to counly
apportionment using bousing
stall data if necessary
Duration of road
construction activity
Miles to acres conversions by
type of roadway constructed
Emission factor of
0.42 tons/acre-month
Uncontrolled PM-10 emissions from
roadVvay construction activities
Figure 5-3. Road Construction Emissions Flowchart
5.6.2 Data Sources and Assumptions
The miles of new roadway constructed are available at the state level from the
Highway Statistics book published yearly by the Federal Highway Administration and the
Bureau of Census' Statistical Abstract of the United States. The miles of new roadway
constructed can be found by determining the change in the miles of roadway from the
previous year to the current year. The amount of roadway constructed is apportioned from
the state to the county level using housing start data that is a good indicator of the need for
new roads.
The conversion of miles of roadway constructed to the acres of land disturbed is based
on a method developed by the California Air Resources Board. This calculation is done by
estimating the roadway width, then multiplying by a mile to determine the acres affected by
one mile of roadway construction. The California conversion factors are for freeway,
highway and city/county roads. In the Highway Statistics book, roadways are divided into
separate functional classes. MRI developed the miles-to-acres conversion according to the
roadway types found in the "Public Road Length, Miles by Functional System" table of the
5-9
-------
annual Highway Statistics. The functional classes are divided into four groups. Group 1
includes Interstates and Other Principal Arterial roads and is estimated to have a conversion
factor of 15.2 acres/mile. Group 2 includes Other Freeways and Expressways (Urban) and
Minor Arterial Roads and is estimated at 12.7 acres/mile. Group 3 has Major Collectors
(Rural) and Collectors (Urban) and a conversion factor of 9.8 acres/mile. Minor Collectors
(Rural) and Local roads are included in Group 4 and converted at 7.9 acres/mile. Table 5-3
shows the data used to calculate the acres per mile of road constructed.
Table 5-3. Road Miles-to-Acres Conversion Calculation
_ane Width (feet)
dumber of Lanes
Average Shoulder Width (feet)
Number of Shoulders
Roadway Width* (feet)
Area affected beyond road width
Width Affected (feet)
Acres Affected per Mile of New Roadway
Group 1
12
5
10
4
100
25
125.0
15.2
Group 2 Group 3
12
5
10
2
80
25
105.0
12.7
12
3
10
2
56
25
81.0
9.8
Group 4
12
2
8
2
40
25
65.0
7.9
'Roadway Width= (Lane Width x # of Lanes) + (Shoulder Width x # of Shoulders)
Since the amount of new roadway constructed is available on a yearly basis, the
duration of the construction activity is determined to be 12 months. The duration accounts
for the amount of land affected during that time period and also reflects that construction of
roads normally lasts longer than a year. The estimate for the duration of construction to
find the total emissions produced by the construction over the length of the activity is
18 months.
5.6.3 Example Emission Calculation
Table 5-4 presents an example calculation of PM emissions from road construction.
State miles are obtained from Table HM-50 in the annual report of the FHWA Report,
Highway Statistics. State emissions are apportioned to the county level based on new
housing statistics that are believed to be a good indicator for the construction of new road
mileage.
5-10
-------
Table 5-4 Example PM10 Emissions from Road Construction in a
Hypothetical County
State
road
mileage
Road
Type
1
2
3
4
1996
2980
3470
4200
11100
1997
3030
3530
4400
11500
New
1997
state
road
mileage'
50
60
200
300
Miles to
Acre factor
15.2
12.7
9.8
7.9
Affected
state
acres
760
762
1960
2370
Duration of
construction
(mo)
12
12
12
12
Emission
factor (tons
PM10/acre/
month)
0.42
0.42
0.42
0.42
State
uncontrolled
PM10emisslons
(tons)
3830
3840
9878
11945
County X
uncontrolled
PM10 emissions
(tons)*
192
192
494
597
'Based on 0.05 fraction of state housing constructed in County "X".
5.7 Correction Parameters
The regional variances in construction activity, as previously mentioned, cause PM
emissions to vary even though the same level of activity may occur at construction sites.
These differences are accounted for using correction parameters.
5.7.1 Control Efficiency
The first correction parameter accounts for the emission reductions afforded by dust
control measures used at construction sites. At most large construction sites watering is
used to control dust suspended by construction equipment activity and vehicle travel on
unpaved roads. The recommended emission factors are representative of uncontrolled sites
which is consistent with the AP-42 manual. The recommended control efficiency for PM
emissions, including PM-10 and PM-2.5, is 50% based on data presented in Reference 20
and recent MRI unpaved road tests.
5.7.2 Soil Moisture Level and Silt Content
The emission factors developed in the BACM report were developed from test sites in
the southwestern United States which have different moisture levels and silt contents than
other areas in the country. To account for the differences in moisture level and silt content,
adjustments are applied to the controlled PM emissions.
Soil moistures for the areas from which the emission factors were developed are
typically much lower than other regions. Thornthwaite's Precipitation-Evaporation Index
ranges from 7 to 41 and is shown in Figure 5-4. The average value for the test sites is 24.
The adjustment for moisture is:
5-11
-------
7»
Figure 5-4. Map of PE Values for State Climatic Divisions
5-12
-------
Moisture Level Corrected Emissions = Base Emissions x (24/PE)
where PE = the Precipitation-Evaporation value for the county being inventoried
The average dry silt content found for the test sites in the BACM report was 9%. To
adjust for the level of silt content of surface soil in a particular county, a proportionality is
used along with the base emissions. The equation to adjust for silt content is:
Silt Content Corrected Emissions = Base Emissions x (s / 9%)
where s = % dry silt content in soil for area being inventoried
The silt content of soil for a county can be found using the same procedure as in the
NET Inventory. Section 4.8.2.2.1.1 in Reference 6 gives the methodology for determining
the silt percentage of soils. The silt percentage is corrected using information from the
California ARB which gives the conversion from a wet silt value to a dry silt value23. The
dry silt percentage is used as a correction parameter for construction emissions. Typical silt
contents for the various soil types are listed in Table 5-5, as reported in Reference 6.
Table 5-5. Dry Silt Content by Soil Type
Soil type
Silt Loam
Sandy Loam
Sand
Loamy Sand
Clay
Clay Loam
Organic Material
Loam
Silt content (%)
52
33
12
12
29
29
10-82
40
5.7.3 Emissions Adjustments
County level emissions of PM10 should be adjusted for dust control measures,
precipitation/evaporation, and dry silt content of the soil. PM10 emissions can also be used
to estimate PM2 5 emissions using a PM2 5/PM10 ratio.
PM2 5 Emissions = Uncontrolled PM10 Emissions x 50% x (24 / PE) x (s / 9%) x
PM2.5/PM10
5-13
-------
where: PE =PE value
s = % dry silt content
50% = 50% Control efficiency from periodic watering
PM25/PM10 = 0.15
Table 5-6 presents the data sources, emission factors, and correction parameters for
all three types of construction.
Table 5-6. Recommended Methodology
Construction
activity type
Residential
Nonresidentlal
Road
Activity level data source
Houses:
Number of housing units
Apartments:
Value of apartment
construction
(Statistical Abstract of the
United States, published
annually by the U.S. Census
Bureau, or the F.W. Dodge
Reports)
Dollar Value of New
Construction (Statistical
Abstract of the United
States or the F.W. Dodge
Reports)
New highway miles
(Highway Statistics, FHWA
annual publication)
Emission factor
Houses:
0.032 tons
PM1(/acre/month
(Source: South Coast
Air Quality Managment
District PM/10 Fugitive
Dust Integration
Project 1996)
Apartments:
0.11 tons
. PM10/acre/month
0.1 9 tons
PM10/acre/month
(Source: SCAQMD,
BACM Report No. 1.
1996, assumes 1/4 of
all nonresidential
construction activity Is
heavy construction
0.42 tons
PM10/acre/month
(Source: SCAQMD,
BACM Report No. 1,
1996)
Control
efficiency
a
o
z
s«
Climatic
factor
X
CD
T3
_C
c •
o
1
o
a.
§
i
i
"o
£
OL
Soil factor
~
'
i
E
£
T3
€
Q>
>
5
C:
»
£•
8
35
£<
Q
5.8 PM10 Emissions from Combustion of Cleared Materials
Construction operations begin with general site preparation. This involves the
clearing of trees, shrubs, and other vegetation that are usually burned. PM emissions are
produced during the combustion of cleared materials.
The PM emissions from the combustion of cleared materials can be calculated using
the emission factors from AP-42 Section 13.1, Wildfires and Prescribed Burning. The
information needed to find PM emissions from burning are the acres affected by the
construction activity and the tons of fuel per acre (available from Table 13.1-1 of AP-42
5-14
-------
by region). The total acres affected by construction can be found by using the conversion
factors for units to acres, dollars to acres, and miles to acres for the three types of
construction.
The emission factors used for the combustion of cleared materials come from
Table 13.1-4 of AP-42 and are by region. Piled slash best represents vegetative residue
cleared at a construction site and is typically 1/2 of the regional average emission factor
for prescribed burning. The PM10 emission factor used for each region is 5 g PM10/kg
fuel for the Pacific Northwest, 6.5 g PM10/kg fuel for the Pacific Southwest, 9.4 g
PM10/kg fuel for the Southeast, 6 g PM10/kg fuel for the Rocky Mountain regio'n, and 7 g
PM10/kg fuel for the North Central and Eastern Regions.
The equation for calculating PM10 emissions from the combustion of cleared
materials is:
PM10 Emissions = EF x t x a
where: EF = Regional emission factor for combustion in g/kg
t = conversion from acres to tons of available fuel
(AP-42 Table 13.1-1)
a = total acres affected by construction
Table 5-7 gives the PMi0 emission factors by region for the combustion of materials
cleared from construction activities by region.
Table 5-7. Combustion of Cleared Materials Emission Factors by Region
Region
Pacific Northwest
Pacific Southwest
Southeast
Rocky Mountain
North Central and Eastern
PM10 emission factor (g/kg of fuel)
5.0
6.5
9.4
6.0
7.0
An example calculation of PM^emissions from the burning of vegitative residues for
a hypothetical county in the Rocky Mountain Region is shown in Table 5-8.
5-15
-------
Table 5-8. Example Calculation of PM10 Emissions from the Burning of Vegetative
Residues
Construction type
Residential
Non-residential
Roads
Total
Acres affected
652
115
293
emission factor
(g/kg)
6.0
6.0
6.0
Fuel
loading per
acre
(ton/acre)
60
60
60
PM10
Emissions
(tons)
234
41
105
380
5-16
-------
Section 6.
References
1. U.S. EPA, Compilation of Air Pollutant Emission Factors. AP-42. Fifth Edition,
Research Triangle Park, NC, September 1995.
2. Countess, Richard and Susan. PM10 Fugitive Dust Integration Project. South Coast
AQMD Contract 96091, July 1996
3. U.S. EPA, Air Quality Criteria for Paniculate Matter. EPA-600/P-45/001aF. 1996.
4. Cowherd, Chatten, and William Kuykendal. Fine Particle Components ofPM-10
from Fugitive Dust Sources. Paper 97-WP96.04 presented at the Annual Meeting of
the Air & Waste Management Association, June 1997.
5. Cowherd, Chatten, Christine Guenther, and Dennis Wallace. Emissions Inventory of
Agricultural Tilling, Unpaved Roads and Airstrips, and Construction Sites. EPA-
450/3-74-085, U.S. Environmental Protection Agency, Research Triangle Park, NC,
November 1974.
6. National Air Pollution Emission Trends Procedures Document for 1900-1996.
Electronically published at www.epa.gov/ttn/chief/eLdata.html/ttETDP
7. Heisler, S.L. Interim Emissions Inventory for Regional Air Quality Studies. Electric
Power Research Institute Report EPRIEA-6070, November 1988.
8. Midwest Research Institute, Gap Filling PM10 Emission Factors for Selected Open
Area Dust Sources. EPA-450/4-88-OQ3, February 1988.
9. Muleski, Greg. Improvement of Specific Emission Factors (BACM Project No. 1),
Final Report. Midwest Research Institute, March 1996.
10. Carlson, Patricia M. and Janna Hummel. Incorporation of State Emission Inventory
Data into the U.S. Environmental Protection Agency's National Emission Trends
Inventory. Paper 98-WPB.16P (Al 10) presented at the Annual Meeting of the Air &
Waste Management Association, June 1998.
11. California Environmental Protection Agency, Emission Inventory Procedural Manual,
Volume HI, Methods for Assessing Area Source Emissions, Electronically published
atarbis.arb.ca.gov/emisinv/areasrc/index7.htm
6-1
-------
12. Barnard, William R. Development of A National Emission Inventory to Support
Revision of the Paniculate National Ambient Air Quality Standard. Paper 97-
WP96.05 presented at the Annual Meeting of the Air & Waste Management
Association, June 1997.
13. E.H. Pechan and Associates, Inc. Development of the OPPE Paniculate Programs
Implementation Evaluation System. September 1994.
14. Midwest Research Institute. Activity Levels for PM10 Area Source Categories:
Methodology Assessment and Improvement. Kansas City, Missouri, May 1993.
15. Englehart, Phil J., and Muleski, Gregory E. Open Fugitive Dust PM10 Control
Strategies Study, Midwest Research Institute, Kansas City, Missouri, October 1990.
16. Holtz, Donald R. PM}0 Emissions Inventory Data for the Maricopa and Pima
Planning Areas. Engineering-Science, Pasadena, California. July 1987.
17. Moore, Teresa M., and Rashmi Balakrishna. Emission Inventory of the Power-
Bannock Counties PM-10 Nonattainment Area, Part I, Area Sources. Presented at the
Annual Conference of the Air & Waste Management Association, June 1996.
18. Clark County Department of Comprehensive Planning, Air Quality Implementation
Plan for the Las Vegas Valley Paniculate Matter PM10, Las Vegas, Nevada.
November 1991.
19. Clark County Board of Commissioners, Paniculate Matter (PM]0) Attainment
Demonstration Plan for Las Vegas Valley Non-attainment Area, Clark County, NV.
Aug 1997.
20. Cowherd, C., G. E. Muleski, and J. S. Kinsey, Control of Open Fugitive Dust
Sources, EPA 450/3-88-008, Prepared for the Office of Air Quality Planning and
Standards, U.S. EPA, Research Triangle Park, NC, 1988.
21. Anderson, Cari L., and Maria J. Brady. General Conformity Analysis for Major
Construction Projects: An Example Analysis of Fugitive PM-10 Emissions. Paper
98-MP4B.06 presented at the Annual Meeting of the Air & Waste Management
Association, June 1998.
22. U.S. EPA, Air Quality Criteria for Paniculate Matter, Electronically published at
www.epa.gov/ncea/partmatt.htm
6-2
-------
23. Campbell, S.G., D.R. Shrimp, and S.R. Francis. Spatial Distribution ofPM10
Emissions from Agricultural Tilling in the San Joequin Valley. Pp. 119-127 in
Geographic Information Systems in Environmental Resources Management, Air and
Waste Management Association, Reno NV, 1996.
Internet Web Pages
1. U.S. Census Annual Data on Construction
www. census, gov/prod/www/abs/cons-hou. htm/
2. FHWA Highway Statistics
www.fhwa.dot.gov/ohirn/hs97/hm50.pdf
3. F.W. Dodge Report
www. fwdodge.com/newdodgenews.asp
6-3
------- |