En9'>na8rin9
Unfted States d-on
Environmental r
Anency __-r~~^Z.
joeU.Wa«on'JuUe
Probert, and Stephen D.
the
«**\iirc8»i HW**V
n 0072 metric ton
M Short ton = 0.90^
incre^30 ., .r(-v \ ano u"-
carbon dioxide ^W le.g., nw"---
^srs^^^^ssi ^
fssss^-"
printed on
-------
VOCs emitted from anthropogenic sources.
EPA plans to use the inventory in atmo-
spheric chemistry models developed by
I.S.A. Isaksen at the University of Oslo, in
Norway. His two-dimensional model re-
quires VOC data, speciated in five reac-
tivity classes: (1) paraffins, (2) olefins, (3)
aromatics, (4) formaldehyde, and (5) other
aldehydes. Two other classes - (6) other
aromatics and (7) marginally reactive
compounds - were included in the inven-
tory for comnpleteness and to ensure that
the needs of other atmospheric chemistry
models for geographically gridded and
speciated VOC data could be accommo-
dated. The inventory was geographically
resolved within 10 by 10 degree grids.
Technical Approach
Figure 1 outlines the technical approach
used to develop the global VOC inventory.
The 1985 U.S. NAPAP Emissions Inven-
tory - Version 2 was used to identify these
sources.
Identify Major VOC Source Types in the U.S.
and Group These Sources and Their Emis-
sions by Associating Them With the Produc-
tion/Consumption of a Specific Commodity or
Service.
Develop Emission Factors by Dividing
the Total U.S. Emissions for Each
Source Category by the Total Amount
of the Commodity or Service Produced/
Consumed in the U.S.
Estimate Global Emissions by Multiplying
Global Commodities/Services Statistics by
the U.S. Emission Factors.
Geographically Distribute All Emissions
Based on Available Industrial Activity
and Population Maps.
Figure 1. Technical approach used in develop-
ing global VOC emissions inventory.
Over 3000 different point and area
sources are identified in the NAPAP in-
ventory. Each source is included in the
global inventory using various simplifying
assumptions and source aggregation
techniques. Biomass burning associated
with land clearing (deforestation), for the
creation of agricultural land and pasture,
and savanna burning are not significant
U.S. sources and, therefore, are not spe-
cifically included in the NAPAP inventory.
However, these sources are included in
the global VOC inventory because studies
have shown that they are significant glo-
bal sources of CO2, and are also consid-
ered potentially significant sources of
VOCs. On the other hand, although natu-
ral sources can contribute significantly to
total VOC emissions in areas with sub-
stantial vegetation, developing data on
these VOC emissions was beyond the
scope of this project and, therefore, was
not included.
The major sources of VOC emissions in
the U.S. were grouped according to the
types of commodities with which they are
associated. For example, oil refinery heat-
ers, catalytic cracking, and pipeline leaks
represent a diverse group of VOC emission
sources which are all related to crude oil.
It is possible, therefore, to combine these
U.S. sources and develop a relationship
which describes their emissions as a
function of the amount of crude oil pro-
cessed in the U.S. Hereafter, groups of
similar emission sources are referred to
as source categories.
For each source category, the U.S. VOC
estimates were divided by their associated
commodity values. The resulting com-
modity-related emission factors, when
multiplied by commodities statistics for
other countries, yielded VOC emission
estimates for those countries. For each
category, as many as seven emission
factors were developed to represent the
seven VOC reactivity classes. Many
sources emit all seven reactivity classes
of pollutants, while other sources emit only
one or two.
Concurrent with the assignment of
source categories, many references were
consulted in order to gather the neces-
sary global commodities data. Publications
by the United Nations (UN), the U.S. De-
partment of Commerce, and U.S. trade
organizations contained data for 195
countries.
The final step in developing the inven-
tory was geographic distribution of the
global emissions. Maps of individual
countries were used to locate centers o
population, industry, chemical manufac
ture, and petroleum refining. An atlas was
used to locate the major deserts and for-
ests on a continental basis. In general,
emissions are distributed into grid cells of
10 degrees latitude by 10 degrees longi-
tude.
Caveats
The inventory described in this docu-
ment was intended to be an initial, limited
effort aimed at producing the needed data
within tight budget and time constraints.
Many of the assumptions used in the de-
velopment of the inventory clearly intro-
duce inaccuracies but were necessary in
order to meet project constraints. For in-
stance, application of emission factors de-
rived from U.S. data to sources in less
fully developed countries is unrealistic but
was necessary because emission factor
data are generally unavailable for such
countries. However, it was felt that rela-
tively early production of a gridded, speci-
ated VOC inventory would meet needs in
the scientific community and would en-
courage the application of additional re-
sources to improve the quality of the in-
ventory for meeting future needs.
Results
Emissions of the seven reactivity classes
are given in tons per year for each of the
648 grid cells of the globe. Results show
total emissions of 120,555,486 tons of
VOC per year.
The inventory is presented by source
category in Table 1. The data in this table
show that a relatively small number of
source categories account for more than
75% of the VOC emissions included in
the inventory. These source groups, listed
in rank order, are:
Percent of
Total VOC
Group Emissions
Fuelwood Utilization (Source
Group 21) 20
Savanna Burning 16
Gasoline Storage, Consumption,
Transportation, and Market-
ing (Source Group 4) 16
Refuse Disposal (Source Group 22) 8
Miscellaneous Emission Sources
(Source Group 27) 7
Rubber, Plastics, and Other
Organic Chemical Manu-
facture (Source Group 11) 7
Solvent Use (Source Group 12) 7
Deforestation for Agriculture
(Source Group 23) 3
Total 84
-------
I
3
o
CO
UJ
.
C3) Q
»- TO
(C <])
^
fc E
£ 2
O <
*
>.
Q) "m
2
O <
I
(0
a.
O
8
|
o
Q.
E
O
o
CM O
O)
CM <*
* CO
00 CM
p T- op
CM CM O
O CO CM
O 00
CM co en
CD *- *-
CM en oo
co co *
m co in
tcoco
8°°
oocoomoo^
i-- co *- *
Sh-
in
00 CO
CMcoo)OO
O^OOCOtOOOtOOCNJOOh-O
to in - i- m o> *- CM
8fc o3883«
CM * T- CO
CM o i^
CO Tf -^
8 CM in
m CM
SCM CM
CO h-
CM i-
O) CO
*- CO
in o>
o co
<* 3
m *-
go
8
in
CM O
CO
3
CO
o o 3
CO
3
CO
3
h-
*r
1^.
CO
°s
3
O T- CO
t h-
O *- * CM
in oo CM
a> co
CO
oo r*» co co o
o o> t- t- co
in ^- o> »- h-
^t m * h-
o> CM
in CM cp CM oo cp CM
!>>. 1^. CM O T- Q CM
r^ co Q co o> o> cp
TT co in T- T- m CM
r>» co co CM CM t^ Q)
CO T- CM CM O
t- CM »- T-
cS
i 2f i^» u>
CO O) ^T _- -_
SCO O> CM CO
T CO O Is*
** ^ »~ o> co
in o co o>
O> CO TT
CM CM CO
CM 00 CO
CO O h"-
CD O) 00
-------
Emissions of the seven reactivity classes
were geographically distributed on the ba-
sis of 10° x 10° global grids. The three
largest source categories and their contri-
bution, in percentage of total emissions,
to each reactivity class are shown below:
Paraffins
Gasoline: Storage, Consumption, Trans-
portation, and Marketing - 20%
Fuelwood Utilization -17%
Solvent Use - 13%
Olefins
Savanna Burning - 43%
Fuelwood Utilization - 18%
Refuse Disposal/Other - 9%
Aromatics
Fuelwood Utilization - 49%
Gasoline: Storage, Consumption,
Transportation, and Marketing - 18%
Miscellaneous Emission Sources - 11%
Formaldehyde
Petroleum Refining - 44%
Gasoline: Storage, Consumption,
Transportation, and Marketing - 12%
Diesel Vehicles - 11%
Other Aldehydes
Rubber, Plastics, and Other Organic
Chemical Manufacture - 24%
Diesel Vehicles - 19%
Miscellaneous Emission Sources - 16%
Other Aromatics
Gasoline: Storage, Consumption,
Transportation, and Marketing - 56%
Rubber, Plastics, and Other Organic
Chemical Manufacture - 16%
Miscellaneous Emission Sources - 7%
Marginally Reactive Compounds
Solvent Use - 42%
Rubber, Plastics, and Other Organic
Chemical Manufacture - 19%
Miscellaneous Emission Sources -13%
Fuelwood use is one of the largest
source categories for emissions of olefins
and aromatics. This accounts for the heavy
emission rates of these VOCs from Africa
and Asia. Savanna burning is the most
significant source of olefin emissions and
accounts for most of the high olefin emis-
sions occurring in central Africa and South
America. The high rates of emissions of
paraffins, olefins and aromatics in the in-
dustrialized areas of Europe and North
America, as shown in Figures 2 through
4, are due to large contributions from the
gasoline, solvent use, refuse disposal and
miscellaneous source categories. The
"other aldehydes" reaction class is prima-
rily emitted by industrial sources such as
diesel vehicles, chemical manufacture and
gasoline. For this reason, high rates c
other aldehydes emissions were found t<
be limited to the heavily industrialized at
eas of the U.S., Japan and Europe. Th«
distribution of formaldehyde emissions i:
heavily influenced by petroleum refining
Consequently, the areas of heavy formal
dehyde emissions include the Middle Eas
as well as the industrialized areas. The
geographic distributions of other aromat
ics and marginally reactive compounds
generally cover the industrialized countries
The geographic distribution of total VOCs
is shown in Figure 5.
The Netherlands Organization for Ap-
plied Scientific Research (OASR) recently
completed a global inventory of 11 volatile
organic compounds. Their study estimated
that global emissions of the 11 specific
VOCs were 66,700,000 tons per year. This
report also estimated that total non-meth-
ane VOC emissions are roughly twice the
estimate for the 11 specified compounds
or 133,342,000 tons per year. Although
these results are 35% above the estimate
of total VOC emissions in this report, given
all the data uncertainties, they are still in
reasonable agreement.
-1«T -140° -120* -100° -80-
ior 120- 140°
-2cr
-60*
-20-
-140- -Mff -100- -80* -60- -40- -20" (T 20*
Legend (tone/year)
14.999 ffl 100,000 249.999
5,000 49.999 |g 250.000 999.999
50.000 99.999 Q 1.000.000 3.145.499
60* «T 1OT 120" 140" 16CT
Scale 1:160.000.000
Miller Projection
Figure 2. Geographic distribution of emissions of paraffins.
-------
-160° -140° -120" -100" -80* -60° -40° -20° 0° 20° 40° 60° 80° 100° 120" 140° 160°
-60-
-1W -140° -120- -100° -«T -60° -40° -20° 0° 20° 40° 80° 80° 100" 120° 14CT 1«T
Legend (tons/year)
14,999
5,000 24,999
25,00099.999
100.000249,999
250,000499.999
1,000.000 1,598.067
Scale 1:160,000,000
Miller Projection
ure 3. Geographic distribution of emissions of olefins.
-160° -140° -120° -100° -80° -60° 40° -20° 0° 20° 40° 60° 80° 100* 120° 140° 160*
-40°
-160° -140° -120° -100° -80° -60° -40° -20° 0° 20° 40° 60° 80° 100° 120° 140° 160°
Legend (tons/year)
14.999
5,000 49,999
50.00099,999
100.000 149.999
150,000 399,999
400,000 655.934
Scale 1:160.000.000
Miller Projection
jre 4. Geographic distribution of emissions of aromatics.
-------
-------
-------
^SS^^£&«~^^^
The complete report entitled ££? ^flcer (see bel°w). ' C 27515-
Springfield, VA 22161
United States
Penalty for Private Use $300
Cincinnati, OH 45268
BULK RATE
POSTAGE & FEES PA in
EPA PERMIT NOG 35
EPA/600/S8-91/002
------- |