EPA/600/A-96/119
Biogenic Emission Estimates for 1995
Thomas E. Pierce01* and Michael P. Dudek®
(1) Atmospheric Modeling Division
Air Resources Laboratory/N OA A
Research Triangle Park, NC 27711
(919) 541-1375
tep@hpcc.epa.gov
(2) DynTel Corporation
PO Box 12804
Research Triangle Park, NC 27709
(919) 558-8782
dop@hpcc.epa.gov
ABSTRACT
Biogenic emissions during 1995 for the contiguous United States have been estimated with
the Biogenic Emissions Inventory System (BEIS2.2). Hourly emissions were computed for each
county using surface observations from the National Weather Service and land use data developed
specifically for biogenic emission calculations. Meteorological data were interpolated to each county
with Barnes analysis. The occurrence of the first and last date of freezing, as interpolated using
Barnes analysis, was used to toggle the occurrence of deciduous leaf biomass. The estimates indicate
annual emissions of 17.2 million metric tons (Tg) of isoprene, 6.1 Tg of monoterpenes, 6.5 Tg of
other volatile organic compounds, and 1.4 Tg of nitric oxides. These estimates are reasonably
consistent with those made by other researchers and are slightly lower than estimates for 1990 made
with an earlier version of BEIS2. The slight decrease in estimated 1995 emissions compared to 1990
can be attributed to the use of freezing dates, better temporal resolution of data (hourly values versus
monthly diurnal averages), and year-to-year variations in meteorology.
INTRODUCTION
Biogenic emissions of volatile organic compounds (VOCs) and nitric oxide (NO) play an
important role in the oxidative capacity of the global troposphere1. In certain locations and times,
these emissions may also influence ozone exceedances and perhaps affect the selection of VOC
versus NOx emission control strategies for alleviating elevated ozone concentrations2,3. VOC
emissions originate from vegetation, and it is thought that isoprene may help protect plants from heat
stress4 while monoterpenes serve a variety of ecological functions such as herbivore protection5.
Other VOCs are known to be emitted, many of which have not yet been quantitatively identified with
gas chromatography/mass spectroscopy6. These other VOCs include sesquiterpenes and oxygenated
hydrocarbons (such as methanol and 2-methyl-3-buten-2-ol). NO emissions appear to originate from
microbial activity in soils7.
As part of ozone reduction planning efforts, many states and localities around the United
States are required to submit periodic reports that include estimates of biogenic emissions. In this
paper, we report on an effort to estimate annual biogenic emissions for the contiguous United States
for 1995. This effort is based on an adaptation of the Biogenic Emissions Inventory System
(BEIS2.2) using meteorological data from National Weather Service reporting stations. The objective
of this work is to provide a foundation for estimates in EPA's Emission Trends Reports8.
TECHNICAL BACKGROUND
Description of BEIS2.2
Version 2.2 of the BEIS was released on the Internet during March 1996 at the following
World Wide Web address: http://www.epa.gov/asmdnerl/biogen.html. It is an extension of earlier
versions of BEIS910 to include updates in emission factors, light and temperature adjustment
algorithms, and land use data. The basic equation for computing biogenic emissions is
— '
On assignment to the National Exposure Research Laboratory, U.S. Environmental Protection Agency
-------
given by the following:
ER = X (EFj x A; x CF)
where ER is the emission rate (g/hr) for each county, EFj is a standardized emission flux ([ig/m2-hr)
for each land use type i, A; is the area (m2) of each land use type i in a county, and CF is a
correction factor that adjusts for the effects of solar radiation and temperature relative to a standard
of 1000 pmol/nr-s of visible solar radiation and 30° C temperature. In BEIS2.2, emissions are
calculated for isoprene, monoterpenes, other VOCs, and nitric oxide.
Emission Factors. Biogenic VOC emission factors for forests were adapted from Geron et
al." and Guenther et al.12. Emission factors in BEIS2.2 have been converted to areal fluxes using
reported leaf biomass densities. Emission factors for other land use types are based on Novak and
Pierce13, except for corn. The emission rate from corn was set nearly to zero based of experimental
results showing negligible VOC emissions14.
Two emission factor tables have been developed for BEIS2.2, a "winter" set and a "summer"
set. The winter set assumes that deciduous vegetation can be mostly ignored and the summer set
assumes full leaf-biomass conditions. The choice of tables for county-level calculation is based on
county freeze dates, with the "summer" table being used after the date of last freeze and before the
date of first freeze.
Soil NO emission factors are based on the work of Williams et al.15. Emission factors have
been extended to agricultural land use types not reported by Williams, by using typical nitrogen
fertilizer application rates to scale between reported emission factors.
Standardized emission fluxes for land use types used in BEIS2.2 are shown in Tables 1 and 2
for summer and winter conditions respectively. These fluxes assume a temperature of 30° C and, for
isoprene, a photosynthetically active radiation flux of 1000 pmol/m2-s.
Environmental corrections. VOC emissions from vegetation and NO emissions from soils
respond quickly to changes in temperature. In addition, isoprene emissions respond to solar radiation
and are negligible when sunlight is not present.
The algorithms for adjusting VOC emissions have been taken from Guenther et al.16 , as
reported by Geron et al". The temperature adjustment equations from Williams et al.15 have been
reformulated so that standard conditions correspond to a soil temperature of 30° C.
Land use. The Biogenic Emissions Landuse Database (BELD) was specifically constructed to
be consistent with the emission factor data base used in BEIS2'6. Special emphasis was given to
estimating the crown cover of high-isoprene-emitting tree species. Because tree species crown
coverage is not routinely available from satellite imagery, the U.S. Forest Service's Forest Inventory
Analysis dataset was extensively used. The hierarchy used for processing the various land use
datasets into the final BELD is summarized in Table 3. The area and percent distribution of the top
20 BELD land use classes are shown in Table 4.
Adapting BEIS2.2 for Annual Estimates
Adapting BEIS2.2 for an annual calculation required minor changes in the FORTRAN source
code and creation of meteorological data for 1995.
BEIS2.2 code changes. The personal computer version of BEIS2.2 was transferred to a
UNIX platform. The first step was to remove all the DOS full-screen menu input options. Loops
were then inserted to calculate hourly emissions for every county for all hours in a month. The code
was also modified to read a freeze date file in order to select between a "summer" or "winter"
emission factor table file. During code execution, this selection is made every day for every county.
Four output files are produced: three quality-assurance files giving detailed calculations for three
counties that can be selected by the user, and one monthly output file giving daily emission fluxes
for each county in the contiguous United States.
-------
Preparation of meteorological data. The meteorological data for running the annual version
of BEIS2.2 consisted of hourly values of temperature and cloud cover for each county in the
contiguous United States. This database was created using Barnes analysis on data from 268 first-
order surface reporting stations operated by the National Weather Service. Barnes analysis is an
interpolation procedure used for spatially distributing meteorological data17 and was used to spatially
interpolate data to all 3111 county/city Federal Information Processing Standard (FIPS) codes. The
latitude and longitude for each FIPS code were obtained from centroids derived from a Geographical
Information System. For the freeze date file, first and last freeze dates for each county were
determined from the interpolated hourly temperatures.
RESULTS
Figure 1 shows daily variations in national isoprene and total biogenic VOCs (BVOC)
estimated with BEIS2.2 for 1995. Emissions are markedly higher during the summer months
(defined here as June, July, and August). The summer months correspond to full-leaf biomass and
higher temperatures and, for isoprene, more solar radiation. Appreciable day-to-day variability can
be seen and is attributable to short-term meteorological fluctuations. Biogenic NO emissions shown
in Figure 2, which respond only to temperature, also peak during the summer and show some day-to-
day variability across the U.S. A seasonal breakdown of emissions is given in Table 5. Isoprene
emissions are negligible during the winter, while 65% of the isoprene emissions occur during the
summer months. The monoterpenes and other VOC categories show a slightly more even
distribution across seasons, with 6-7% of the emissions occurring during the winter months. For
biogenic NO emissions, 41% occur during the summer months, and 13% occur during the winter
months.
Figures 4-6 show the spatial distribution of biogenic emissions estimated for 1995. The
isoprene emissions in Figure 4 are concentrated in areas having high percentages of deciduous
forests, near the Appalachian mountains and west of the Mississippi River from Missouri to the Gulf
Coast. Other areas with high isoprene emission densities occur in areas with relatively extensive
spruce and aspen forests and parts of the western U.S. that are heavily wooded. Total VOC
emissions shown in Figure 5 largely mimic the isoprene pattern, because isoprene comprises such a
large percentage (58%) of the total biogenic VOC inventory. In addition to those areas with high
isoprene emissions, the western U.S. and New England show relatively high concentrations of BVOC
owing to the high percentages of coniferous forests, which tend to emit monoterpenes rather than
isoprene. Biogenic NO emissions shown in Figure 6 are confined mostly to agricultural areas,
especially the corn belt of the Great Plains. Other notable areas of high concentrations of NO
emissions include a few counties in southern Texas, where the U.S. agricultural data indicated large
areas of sorghum and where mean temperatures are high, and in agricultural portions of south-central
Pennsylvania. Forested areas of the U.S. are estimated as having low emissions of biogenic NO.
DISCUSSION
The estimates made for 1995 are comparable with other estimates, although some differences
can be seen in Table 6. Estimates for 1995 are slightly less than those made with BEIS2 for 1990
using a different methodology. It is believed that use of the frost dates in this work caused most of
the reduction in VOCs, as evidenced in a winter-time reduction in BVOC from 1.4 Tg in 1990 to 0.9
Tg in 1995. In addition, minor fixes to the land use data since the 1990 estimates probably affected
the NO emissions. The NO emissions for 1995 are about 10% less than those computed for 1990,
but the corn acreage (which has a relatively high NO emission flux) in the earlier calculations was
too high by as much as a factor of two in many Midwestern counties. A processing error caused
much of the soybean acreage in these counties to be mistakenly coded as corn. Differences in
meteorology between the two years also are likely to affect the calculations, but hourly data from
1990 were not available in time for this paper to investigate its impact on the calculations. Year-to-
year variations due to meteorology will be the subject of future work.
-------
Total VOC emissions (29.8 Tg) are similar to those of Lamb et al,1819» who reported values
ranging from 19.4 Tg - 29.1 Tg. However, isoprene emissions in the BEIS2.2 inventory are much
higher than Lamb et al., 17.2 Tg versus 2.7 Tg - 5.9 Tg. This increase can be attributed to newer
isoprene emission factors in BEIS2.2 that treat each high-emitting tree species and a land use
inventory that tracks each tree genus type. In addition, Lamb et al. used a geometric mean that
resulted in a mean isoprene emission factor about a factor of two lower than if a arithmetic mean had
been used. The emission factors used in BEIS2.2 are based on arithmetic means.
Although the work of" Williams et al.15 serves as the basis of much of the NO emissions
algorithm in BEIS2.2, the two annual estimates vary by a factor of two. In this work, annual
emissions were estimated at 1.4 Tg as compared to Williams et al. estimate of 0.7 Tg. Our higher
estimates may be attributed to the following differences in assumptions: (1) including the
contribution of natural biomes (many of which are considered grass and shrubland in BEIS2 and
have a modest NO flux), which were ignored by Williams et al., (2) including the contributions from
crops other than corn, wheat, soybeans, which were ignored by Williams et al. because of a lack of
emission factor data, (3) assuming the same emission factors for these four crops for the entire
growing season, which were assumed by Williams et al. to be negligible during September - March,
and (4) using hourly temperature data for 1995, which in Williams et al. were based on monthly
climatic averages. Differences arising from these assumptions highlight some of the uncertainty
surrounding calculation of annual emissions of biogenic NO.
Limitations exist with these and other biogenic emission estimates7,19. Emission factors have
changed rapidly during the past few years, and further refinements are likely as additional field study
data become available. Evidence of this change can be found in the factor of five increase of
isoprene that occurred between BEIS1 and BEIS2 for short-term estimates related to ozone modeling
studies. Biogenic NO emissions are particularly uncertain, owing to a lack of knowledge on the
application of nitrogen-based fertilizer, the influence of soil moisture, and uptake by vegetation of
NOx before NO can escape into the free troposphere. Another limitation with biogenic emission
calculations is the land use data. Year-to-year changes in land use distribution have not yet been
accounted for in this analysis. For agricultural data, this can be significant. For example, reports
from the news media (News and Observer, Raleigh, NC, August 12, 1996) indicate that -12% more
corn is being grown in 1996 than in 1995. This increase in corn production will almost cprtainly
result in increased estimates of biogenic NO emissions. Our knowledge of tree cover in urban areas
is also lacking. While this does not affect emissions much on a national scale, it can greatly affect
urban scale calculations. Fortunately, the U.S. Forest Service is undertaking a study in the
northeastern U.S. during the summer of 1996 to improve this knowledge base. The land use data
base should also be viewed as particularly uncertain in the western U.S., where broadly-defined
categories from the U.S. Geological Survey's database are used to infer emission fluxes.
Meteorological data suffer from uncertainties in spatial interpolation. Areas with significant
topographical changes relative to nearby surface observation stations, such as mountainous and
coastal areas, should be viewed as somewhat suspect. Use of the frost dates to estimate leaf biomass
may introduce some uncertainty. Because satellite imagery offers the possibility for estimating
vegetation biomass, we are investigating the use of satellite imagery to temporally model leaf
biomass in future versions of BEIS.
Despite these limitations, this dataset attempts to provide a scientifically-credible estimate of
biogenic emissions for 1995 suitable for inclusion in EPA's emission trend reports. Estimates for
other years between 1985-1995 are expected to be available in the near future.
-------
ACKNOWLEDGMENTS and DATA AVAILABILITY
The authors appreciate development of the Biogenic Emissions Landuse Database (BELD) by
Ellen Kinnee of DynTel and collaboration with Chris Geron of EPA who, although burdened by
numerous responsibilities, continues to provide valuable scientific support for the improvement and
verification of biogenic emission flux estimates.
Data files of the daily estimates are available via anonymous ftp at monsoon.rtpnc.epa.gov.
The directory name is ~/pub/beis2/trend/l995. Questions may be directed to tep@hpcc.epa.gov.
DISCLAIMER
This paper has been reviewed in accordance with the U.S. Environmental Protection Agency's
peer and administrative review policies and approved for presentation and publication. Mention of
trade names or commercial products does not constitute endorsement or recommendation for use.
REFERENCES
1. Fehsensfeld, F., J. Calvert, R. Fall, et al., Global Biogeochem. Cycles 1992 6 389-430.
2. Chameides, W., R. Lindsay, J. Richardson, et al., Science 1988 241 1473-1475.
3. Roselle, S„ Atmos. Environ. 1994 28 1757-1772.
4. Sharkey, T. and E. Sinssaas, Nature 1995 374 796-797.
5. Tingey, D., D. Turner, and J. Weber, Trace Gas Emissions by Plants, T. Sharkey,
E, Holland, and II. Mooney, Eds.; Academic Press, New York, 1991; pp. 93-119.
6. Guenther, A., C. Hewitt, D. Erickson, et al, J. Geophvs. Res. 1995 J00 8873-8892.
7. Williams, E., G. Hutchinson, F. Fehsenfeld, Global Biogeochem. Cycles 1992 6 351-388.
8. Nizich, S., T. Pierce, W. Hohenstein, et al.; National Air Pollutant Emission Trends, 1990-
1994, EPA-454/R-95-011; U.S. Environmental Protection Agency, Research Triangle Park,
1995, 149 pp.
9. Pierce, T., P. Waldruff, J. Air & Waste Management Assoc. 1991 41. 937-941.
10. Birth, T.; User's Guide to the Personal Computer Version of the Biogenic Emissions
Inventory System (PC-BEIS2), EPA-600/R-95-091; U.S. Environmental Protection
Agency, Research Triangle Park, 1995, 31 pp.
11. Geron, C., A. Guenther, T. Pierce, J. Geophvs. Res. 1994 99 12773-12791. .
12. Guenther, A., P. Zimmerman, M. Wildermuth, Atmos. Environ. 1994 28 1197-1210.
13. Novak, J., T. Pierce, Water, Air, and Soil Pollution 1993 67 57-77.
14. Sharkey, T., P. Vanderveer, F. Loreto, "Biogenic hydrocarbons: measurements on corn
and kudzu in 1992," presented at the AWMA Emission Inventory Issues in the 1990s
Conference, Durham, NC, 1992.
15. Williams, E., A. Guenther, F. Fehsenfeld, J. Geophvs. Res. 1992 97 7511-7519.
-------
16. Kinnee, E., C. Geron, T. Pierce, "Creation of a land use inventory for estimating biog
ozone precursor emissions in the United States", Ecological Applications (in press).
17. Barnes, S., J. Appl. Meteorol. 1964 3 396-409.
18. Lamb, B., A. Guenther, D. Gay, et al., Atmos. Environ. 1987 21 1695-1705.
19. Lamb, B„ D. Gay, H. Westberg, et al., Atmos. Environ. 1993 27A 1673-1690.
-------
Table 1, Standardized "summer" emission fluxes used in BEIS2.2. Columns represent the
genus id, isoprene flux, monoterpene flux, other VOC flux, soil NO flux, leaf area index, and
a description. Fluxes are given in units of (ig/m2-hr and are standardized to 30 C. Isoprene
flux is based on a light intensity of 1000 |imoI/m2-s.
Abie
170.0
5100.0
2775.0
4.5
7
Abies (fir)
Acac
79.3
2380.0
1295.0
4.5
5
Acacia
Acer
42.5
680.0
693.7
4.5
5
Acer (maple)
Aesc
42.5
42 .5
693.7
4.5
5
Aesculus (buckeye)
Aila
42.5
42 .5
693.7
4.5
5
Ailanthus
Aleu
42.5
42.5
693 .7
4.5
5
Aleurites (tung-oil tree)
Alfa
19.0
7.6
11.4
12.8
0
Alfalfa
Ainu
42.5
42.5
693 .7
4.5
5
Alnus (European alder)
Arael
42.5
42.5
693.7
4.5
5
Amelanchier (serviceberry)
As im
42.5
42.5
693.7
4.5
5
Asimina (pawpaw)
Avic
42 .5
42.5
693 .7
4.5
5
Avicennia (black mangrove)
Barl
7.6
19.0
11.4
256.7
0
Barley
Barr
0.0
0.0
0.0
0.0
0
Barren
Betu
42.5
85.0
693.7
4.5
5
Betula (birch)
Borf
910.0
713.0
755.0
4.5
5
Boreal forest (AVHRR/Guen et al 94)
Bume
42.5
42.5
693.7
4.5
5
Bumelia (gum bumelia)
Carp
42.5
680.0
693.7
4.5
5
Carpinus (hornbean)
Gary
42 . 5
680.0
693.7
4.5
5
Carya (hickory)
Casp
42.5
42 .5
693 .7
4.5
5
Castanopsis (chinkapin)
Cast
42 . 5
42.5
693 .7
4.5
5
Castanea (chestnut)
Casu
29750.0
42.5
693.7
4.5
7
Casuarina (Austl pine)
Cata
42,5
42.5
693.7
4.5
5
Catalpa
Cedr
79.3
1269.3
1295.0
4.5
7
Cedrus (Deodar cedar)
Celt
42 .5
85.0
693.7
4.5
5
Celtis (hackberry)
Cere
42.5
42.5
693 .7
4.5
5
Cereis (redbud)
Cham
170.0
340.0
2775.0
4.5
7
Chamaecyparis (prt-orford cedar)
Citr
42.5
680.0
693 .7
4.5
5
Citrus (orange)
Cnif
745.4
1366.6
993.9
4.5
9
BEIS conifer forest
Conf
1550.0
1564.0
1036.0
4.5
6
Conifer forest (AVHRR, Guen)
Corn
0.5
0.0
0.0
577 . 6
0
Corn
Coru
42 .5
680.0
693.7
4.5
5
Cornus (dogwood)
Coti
42 .5
42.5
693 .7
4.5
5
Cotinus (smoke tree)
Cott
7.6
19.0
11.4
256.7
0
Cotton
Crat
42.5
42,5
693.7
4.5
5
Crataegus (hawthorn)
Cswt
1050.0
660.0
770.0
0.2
2
Herbaceous Wetlands (AVHRR, Guen)
Desh
65.0
94.5
56.7
57.8
0
Desert shrub (AVHRR, Guen)
Dios
42.5
42.5
693 .7
4.5
5
Diospyros (persimmon)
Euca
29750.0
1275.0
693 .7
4.5
5
Eucalyptus
Fagu
42 . 5
255.0
693 .7
4.5
5
Fagus (american beech)
Frax
42.5
42.5
693.7
4.5
5
Fraxinus (ash)
Gled
42 .5
42 .5
693.7
4.5
5
Gleditsia (honeylocust)
Gord
42.5
42.5
693.7
4.5
5
Gordonia (loblolly-bay)
Gras
56.2
140.5
84.3
57.8
0
Grass
Gymn
42 . 5
42.5
693.7
4.5
5
Gymnocladus (KY coffeetree)
Hale
42.5
42.5
693.7
4.5
5
Halesia (silverbell)
Harf
8730.0
436.0
882.0
4.5
5
Hardwood forest (AVHRR, Guen)
Hay
37.8
94.5
56.7
12.8
0
Hay
Ilex
42 . 5
85. 0
693.7
4.5
5
Ilex (holly)
Jugl
42.5
1275.0
693.7
4.5
5
Juglans (black walnut)
Juni
79.3
476.0
1295.0
4.5
7
Juniperus (east, red cedar)
Lagu
42 .5
42.5
693.7
4.5
5
Laguncularia (white mangrove)
Lari
42.5
42.5
693 .7
4.5
5
Larix (larch)
Liqu
29750.0
1275.0
693 .7
4.5
5
Liquidambar (sweetgum)
Liri
42 . 5
85.0
693 .7
4.5
5
Liriodendron (yellow poplar)
Macl
42.5
42 . 5
693.7
4.5
5
Madura (osage-orange)
Magn
42.5
1275.0
693 .7
4.5
5
Magnolia
Malu
42 . 5
42 .5
693.7
4.5
5
Malus (apple)
Meli
42.5
42.5
693.7
4.5
5
Melia (chinaberry)
Mixf
11450.0
1134.0
1140.0
4.5
5
Mixed forest (AVHRR, Guen)
Moru
42 . 5
85 . 0
693.7
4.5
5
Morus (mulberry)
Mscp
7.6
19.0
11.4
12.8
0
Misc crops
-------
Nmxf
10150.0
1100.0
850.0
4.5
Nyss
5950 . 0
255.0
693 .7
4.5
Oak
3108.3
255.5
894.2
4.5
Oats
7.6
19.0
11.4
256.7
Odcd
2112.4
368.8
871.8
4.5
Of or
56.2
140.5
84.3
4.5
Oksv
7350.0
100.0
600.0
4.5
Ostr
42.5
42.5
693.7
4.5
Othe
56.2
140.5
84.3
57.8
Oxyd
42.5
255.0
693.7
4.5
Pacp
55.0
79.8
47.9
35.3
Past
56.2
140.5
84.3
57.8
Paul
42.5
42.5
693.7
4.5
Pean
102.0
255.0
153 .0
12.8
Pers
42.5
255.0
693 .7
4.5
Pice
23800.0
5100.0
2775.0
4.5
Pinu
79.3
2380.0
1295.0
4.5
Plan
42 .5
42.5
693 .7
4.5
Plat
14875.0
42.5
693.7
4.5
Popu
29750.0
42.5
693 .7
4.5
Pota
9.6
24.0
14.4
192.5
Pros
42.5
42.5
693.7
4.5
Prun
42.5
42.5
693 .7
4.5
Pseu
170.0
2720.0
2775.0
4.5
Quer
29750.0
85.0
693.7
4.5
Rang
37.8
94.5
56.7
57.8
Rhiz
42.5
42.5
693 .7
4.5
Rice
102 . 0
255.0
153.0
0.2
Robi
5950.0
85.0
693.7
4.5
Rye
7 . 6
19.0
11.4
12.8
Sabl
5950.0
42.5
693.7
4.5
Sali
14875.0
42.5
693.7
4.5
Sapi
42.5
42.5
693.7
4.5
Sass
42.5
42.5
693.7
4.5
Scru
37.8
94.5
56.7
57.8
Scwd
2700.0
349.0
651.0
31.2
Sere
14875.0
42.5
693 .7
4.5
Shrf
10750.0
530.0
910.0
4.5
Smxf
17000.0
1500.0
1250.0
4.5
Snow
0 . 0
0.0
0.0
0.0
Sorb
42.5
42.5
693.7
4.5
Sorg
7 . 8
19.5
11.7
577.6
Soyb
22 .0
0.0
0.0
12.8
Spin
1460.0
1983.0
1252.0
4.5
Swie
42 . 5
42.5
693.7
4.5
Taxo
42.5
1275.0
693.7
4.5
Thuj
170.0
1020.0
2775.0
4.5
Tili
42 .5
42 .5
693 .7
4.5
Toba
0 . 0
58.8
235.2
256.7
Tsug
79.3
158.7
1295.0
4.5
Tund
2411.7
120.6
150.7
0,2
Ufor
1988.7
663.7
920.0
4.5
Ugra
56.2
140.5
84.3
57.8
Ulmu
42 . 5
42.5
693.7
4.5
Uoth
11.2
28.1
16.9
11.6
Urba
408.6
161.9
200.5
12.5
Utre
5140.0
1000.0
959.0
4.5
Vacc
42.5
42.5
693 .7
4.5
Wash
5950 . 0
42.5
693 .7
4.5
Wate
0.0
0.0
0.0
0.0
Wcnf
4270 .0
1120.0
1320.0
4.5
Wdcp
2550.0
663.0
2053.0
8.7
Wetf
3820.0
923.0
1232 .0
0.2
Whea
15.0
6.0
9.0
192.5
Wmxf
5720.0
620.0
530.0
4.5
Wwdl
525.0
250.0
360.0
4.5
Northern Mixed Forest (AVHRR, Guen)
Nyssa (blackgum)
BEIS oak forest
Oats
BEIS other deciduous forest
Open forest
Oak Savannah (AVHRR, Guen)
Ostrya (hophornbeam)
Other (unknown, assume grass)
Oxydendrum (sourwood)
Pasture cropland (AVHRR, Guen)
Pasture
Paulownia
Peanuts
Persea (redbay)
Picea (spruce)
Pinus (pine)
Planera (water elm)
Platanus (sycamore)
Populus (aspen)
Potato
Prosopis (mesquite)
Prunus (cherry)
Pseudotsuga (douglas fir)
Quercus (oak)
Range
Rhizophora (red mangrove)
Rice
Robinia (black locust)
Rye
Sabal (cabbage palmetto)
Salix (willow)
Sapium (chinese tallow tree)
Sassafras
Scrub
Scrub woodland {AVHRR, Guen)
Serenoa (saw palmetto)
Southeast/Western Deciduous Forest
Southeast Mixed Forest
Snow
Sorbus (mountain ash)
Sorghum
Soybean
Southern pine (AVHRR, Guen)
Swietenia (W. Indies mahogany)
Taxodium (cypress)
Thuja (W. red cedar)
Tilia (basswood)
Tobacco
Tsuga (Eastern hemlock)
Tundra
BEIS urban forest
BEIS urban grass
Ulmus (American elm)
Urban other (assume 20% grass!
BEIS urban (.2 grass/.2 forest)
Urban trees (.5 Harf/ . 5 Conf)
Vaccinium (blueberry)
Washingtonia (fan palm)
Water
W Coniferous Forest (AVHRR, Guen)
Woodland/cropland (AVHRR, Guen)
Wetland forest (AVHRR, Guen)
Wheat
Western Mixed Forest (AVHRR, Guen)
Western Woodlands (AAVHRR, Guen)
5
5
6
0
6
0
2
5
0
5
0
0
5
0
5
7
3
5
5
5
0
5
5
7
5
0
5
0
5
0
5
5
5
5
0
2
5
5
4
0
5
0
0
3
5
5
7
5
0
7
0
0
0
5
0
0
5
5
5
0
5
3
5
0
4
3
-------
Table 2. Standardized "winter" emission fluxes used in BEIS2.2. Columns represent the land
use id, isoprene flux, monoterpene flux, other VOC flux, soil NO flux, leaf area index, and a
description. Fluxes are given in units of jjg/m2-hr and are standardized to 30 C. Isoprene
flux based on a light intensity of 1000 (imol/m2-s.
Abie
170.0
5100.0
2775.0
4.5
7
Abies (fir)
Acac
0.0
0.0
0.0
4.5
5
Acacia
Acer
0.0
0.0
0.0
4.5
5
Acer (maple)
Aesc
0.0
0.0
0.0
4.5
5
Aesculus (buckeye)
Aila
0.0
0.0
0.0
4.5
5
Ailanthus
Aleu
0.0
0.0
0.0
4.5
5
Aleurites (tung-oil tree)
Alfa
0.0
0.0
0.0
12 . 8
0
Alfalfa
Ainu
0.0
0.0
0.0
4.5
5
Alnus (European alder)
Amel
0.0
0.0
0.0
4.5
5
Amelanchier (serviceberry)
Asim
0.0
0.0
0.0
4.5
5
Asiminia (pawpaw)
Avic
42.5
42 .5
693.7
4.5
5
Avicennia (black mangrove)
Barl
0.0
0.0
0.0
256.7
0
Barley
Barr
0.0
0.0
0.0
0.0
0
Barren
Betu
0.0
0.0
0.0
4.5
5
Betula (birch)
Borf
640.0
706.0
634.0
4.5
5
Boreal forest (AVHRR/Guen et al 94)
Bume
42 .5
42.5
693.7
4.5
5
Bumelia (gum bumelia)
Carp
0.0
0.0
0.0
4.5
5
Carpinus (hornbean)
Cary
0.0
0.0
0.0
4.5
5
Carya (hickory)
Casp
0.0
0.0
0.0
4.5
5
Castanopsis (chinkapin)
Cast
0.0
0.0
0.0
4.5
5
Castanea (chestnut)
Casu
29750.0
42.5
693.7
4.5
7
Casuarina (Austl pine)
Cata
0.0
0.0
0.0
4.5
5
Catalpa
Cedr
79.3
1269.3
1295.0
4.5
7
Cedrus (Deodar cedar)
Celt
0.0
0.0
0.0
4.5
5
Celtis (hackberry)
Cere
0.0
0.0
0.0
4.5
5
Cereis (redbud)
Cham
170.0
340.0
2775.0
4.5
7
Chamaecyparis (prt-orford cedar)
Citr
42.5
680 .0
693 .7
4.5
5
Citrus (orange)
Cnif
0.0
1353.0
835.0
4.5
9
BEIS conifer forest
Conf
1400.0
1548.0
870.0
4.5
6
Conifer forest (AVHRR, Guen)
Corn
0.0
0.0
0.0
577 . 6
0
Corn
Coru
0.0
0.0
0.0
4.5
5
Cornus (dogwood)
Coti
0.0
0.0
0.0
4.5
5
Cotinus (smoke tree)
Cott
0.0
0.0
0.0
256.7
0
Cotton
Crat
0.0
0.0
0.0
4.5
5
Crataegus (hawthorn)
Cswt
1050.0
660.0
770.0
0.2
2
Herbaceous Wetlands (AVHRR, Guen)
Desh
0.0
0.0
0.0
57,8
0
Desert shrub (AVHRR, Guen)
Dios
0.0
0.0
0.0
4.5
5
Diospyros (persimmon)
Euca
29750.0
1275.0
693 .7
4.5
5
Eucalyptus
Fagu
0.0
0.0
0.0
4.5
5
Fagus (american beech)
Frax
0.0
0.0
0.0
4.5
5
Fraxinus (ash)
Gled
0.0
0.0
0.0
4.5
5
Gleditsia (honeylocust)
Gord
0.0
0.0
0.0
4.5
5
Gordonia (loblolly-bay)
Gras
0.0
0.0
0.0
57.8
0
Grass
Gymn
0.0
0.0
0.0
4.5
5
Gymnocladus (KY coffeetree)
Hale
0.0
0.0
0.0
4.5
5
Halesia (silverbell)
Harf
0.0
371.0
185.0
4.5
5
Hardwood forest (AVHRR, Guen)
Hay
0.0
0.0
0.0
12.8
0
Hay
Ilex
42.5
85.0
693 .7
4.5
5
Ilex (holly)
Jugl
0.0
0.0
0.0
4.5
5
Juglans (black walnut)
Juni
79.3
476.0
1295.0
4.5
7
Juniperus (east, red cedar)
Lagu
42 .5
42 .5
693.7
4.5
5
Laguncularia (white mangrove)
Lari
0.0
0.0
0.0
4.5
5
Larix (larch)
Liqu
0.0
0.0
0.0
4.5
5
Liquidambar (sweetgum)
Liri
0.0
0.0
0.0
4.5
5
Liriodendron (yellow poplar)
Macl
0.0
0.0
0.0
4.5
5
Maclura (osage-orange)
Magn
42 . 5
1275.0
693 .7
4.5
5
Magnolia
Malu
0.0
0.0
0.0
4.5
5
Malus (apple)
Meli
0.0
0.0
0.0
4.5
5
Melia (chinaberry)
Mixf
0.0
1077 .0
581.0
4.5
5
Mixed forest (AVHRR, Guenther)
Moru
0.0
0.0
0.0
4.5
5
Morus (mulberry)
Mscp
0.0
0.0
0.0
12.8
0
Misc crops
-------
Nmxf
175.0
1100.0
850.0
4.5
Nyss
0.0
0.0
0.0
4.5
Oak
0.0
217 .0
188.0
4.5
Oats
0.0
0.0
0.0
256.7
Odcd
0.0
313 .0
183.0
4.5
Ofor
0.0
0.0
0.0
4.5
Oksv
0.0
100.0
200.0
4.5
Ostr
0.0
0.0
0.0
4.5
Othe
0.0
0.0
0.0
57.8
Oxyd
0.0
0.0
0.0
4.5
Pacp
0.0
0.0
0.0
35.3
Past
0.0
0.0
0.0
57.8
Paul
0.0
0.0
0.0
4.5
Pean
0.0
0.0
0.0
12.8
Pers
42 .5
255.0
693.7
4.5
Pice
23800.0
5100.0
2775.0
4.5
Pinu
79.3
2380.0
1295.0
4.5
Plan
0.0
0.0
0.0
4.5
Plat
0.0
0.0
0.0
4.5
Popu
0.0
0.0
0.0
4.5
Pota
0.0
0.0
0.0
192.5
Pros
0.0
0.0
0.0
4.5
Prun
0.0
0.0
0.0
4.5
Pseu
170 . 0
2720 . 0
2775.0
4.5
Quer
0.0
0.0
0.0
4.5
Rang
0.0
0.0
0.0
57.8
Rhiz
42 . 5
42 . 5
693.7
4.5
Rice
0.0
0.0
0.0
0.2
Robi
0.0
0.0
0.0
4.5
Rye
0 . 0
0.0
0.0
12 .8
Sabl
5950.0
42 .5
693.7
4.5
Sali
0.0
0.0
0.0
4.5
Sapi
0.0
0.0
0.0
4.5
Sass
0.0
0.0
0.0
4.5
Scru
0.0
0.0
0.0
57 .8
Scwd
0.0
332 .0
332.0
31.2
Sere
14875.0
42.5
693 .7
4.5
Shrf
0.0
0.0
0.0
4.5
Smxf
0.0
1500.0
500.0
4.5
Snow
0.0
0.0
0.0
0.0
Sorb
0.0
0.0
0.0
4.5
Sorg
0.0
0.0
0.0
577 , 6
Soyb
0 . 0
0.0
0.0
12.8
Spin
0.0
1963.0
1052.0
4.5
Swie
42 .5
42 .5
693.7
4.5
Taxo
42.5
1275.0
693.7
4.5
Thu j
170.0
1020.0
2775.0
4.5
Tili
0.0
0.0
0.0
4.5
Toba
0 . 0
0.0
0.0
256.7
Tsug
79.3
158.7
1295.0
4.5
Tund
0.0
0.0
0.0
0.2
Uf or
0.0
631.0
469.0
4.5
Ugra
0,0
0.0
0.0
57.8
Ulmu
0.0
0.0
0.0
4.5
Uoth
0.0
0.0
0.0
11.6
Urba
0.0
154.0
102.0
12.5
litre
700.0
960.0
528.0
4.5
Vacc
0.0
0.0
0.0
4.5
Wash
5950.0
42.5
693.7
4.5
Wate
0,0
0.0
0.0
0.0
Wcnf
3500 . 0
1120.0
1200 .0
4.5
Wdcp
0.0
630 . 0
1047.0
8.7
Wetf
0.0
877 .0
628.0
0.2
Whea
0.0
0.0
0.0
192 .5
Wmxf
0 . 0
620.0
330.0
4.5
Wwdl
0.0
250.0
360.0
4.5
Northern Mixed Forest (AVHRR, Guen)
Nyssa (blackgum)
BEIS oak forest
Oats
BEIS other deciduous forest
Open forest
Oak Savannah (AVHRR, Guen)
Ostrya (hophornbeam)
Other (unknown, assume grass)
Oxydendrum (sourwood)
Pasture cropland (AVHRR, Guen)
Pasture
Paulownia
Peanuts
Persea (redbay)
Picea (spruce)
Pinus (pine)
Planera (water elm)
Platanus {sycamore)
Populus (aspen)
Potato
Prosopis (mesquite)
Prunus (cherry)
Pseudotsuga (douglas fir)
Quercus (oak)
Range
Rhizophora (red mangrove)
Rice
Robinia (black locust)
Rye
Sabal (cabbage palmetto)
Salix (willow)
Sapium (Chinese tallow tree)
Sassafras
Scrub
Scrub woodland (AVHRR, Guen)
Serenoa (saw palmetto)
SE/W Deciduous Forest (AVHRR, Guen)
SE Mixed Forest (AVHRR, Guen)
Snow
Sorbus (mountain ash)
Sorghum
Soybean
Southern pine (AVHRR, Guen)
Swietenia (W. Indies mahogany)
Taxodium (cypress)
Thuja (W. red cedar)
Tilia (basswood)
Tobacco
Tsuga (Eastern hemlock)
Tundra
BEIS urban forest
BEIS urban grass
Ulmus (American elm)
Urban other (assume 20% grass)
BEIS urban (.2 grass/.2 forest)
Urban tree (.5 Harf/.5 Conf)
Vaccinium (blueberry)
Washingtonia (fan palm)
Water
W Coniferous Forest (AVHRR, Guen)
Woodland/cropland (AVHRR, Guen)
Wetland forest (AVHRR, Guen)
Wheat
Western Mixed Forest (AVHRR, Guen)
Western Woodlands (AVHRR, Guen)
5
5
6
0
6
0
2
5
0
5
0
0
5
0
5
7
3
5
5
5
0
5
5
7
5
0
5
0
5
0
5
5
5
5
0
2
5
5
4
0
5
0
0
3
5
5
7
5
0
7
0
0
0
5
0
0
5
5
5
0
5
3
5
0
4
3
-------
Table 3. Hierarchial rules used for processing the land use data in the Biogenic Emissions
Landuse Database (BELD) for each county in the contiguous United States.
Priority
Description of raw data
Resulting BELD data
I
U.S. Forest Service's Forest Inventory database
for the eastern U.S. (circa 1990), containing
information from ~97,000 1 -acre ground survey
plots resolved to the county level. Tree species
and diameter measurements are used to estimate
crown cover by genus.
Total forest area, crown cover by tree
genus for counties in the eastern U.S.
II
U.S. Geological Survey's Land Cover
Characteristics Dataset, based on classification of
imagery from the AVHRR satellite; resolved into
1-km pixels.
Inland water
Ilia
U.S. Census Bureau, urbanized areas from 1990.
Total urban area
Illb
U.S. Forest Service (Dave Nowak, personal
communication) fraction of urban area assumed
to be forested, based on potential natural
vegetation and relative percentage of native tree
species in areas surrounding urban regions.
Total urban forest area and tree genus
composition
IV
U.S. Department of Agriculture, 1987 crop
statistics by county.
Area of specific crop types
V
U.S. Department of Agriculture, 1987 farm areas
by county.
Area assumed as miscellaneous crop
VI
U.S. Geological Survey's Land Cover
Characteristics Dataset, land use classes assigned
to Guenther et al.12 land use types; used
extensively for the western U.S.
Areas of generalized land use types
VII
Undesignated, area in a county lacking
classification.
Area of other
-------
Table 4. Abundance of the top 20 land use types found in the Biogenic Emissions Landuse
Database (BELD) for the contiguous United States.
Land use type
Area (million hectares)
Percent of total
Grass
144.8
18.3
Miscellaneous crops
97.4
12.3
Other (assumed grass)
66.3
8.4
Western coniferous forest
59.4
7.5
Scrub
54.7
6.9
Quercus (oaks)
34.4
4.3
Barren
32.1
4.1
Corn
25.0
3.2
Open forest (assumed grass)
24.0
3.0
Hay
23.3
2.9
Wheat
21.5
2.7
Soybeans
21.5
2.7
Water
20.3
2.6
Western woodlands
18.6
2.4
Pinus (pines)
19.3
2.4
Urban other (assumed 20%
12.5
1.6
grass)
Acer (maples)
12.2
1.5
Woodland/cropland
9.5
1.2
Western mixed forest
8.1
1.0
Carya (hickory)
6.6
0.8
Cumulative total
711.5
89.8
-------
Table 5. Seasonal breakout of biogenic emissions estimated for the contiguous United States
for 1995. Emissions reported in units of million metric tons. Numbers may not add up
exactly because of rounding.
Chemical
Winter (DJF)
Spring (MAM)
Summer (JJA)
Fall (SON)
Isoprene
0.10
2.98
11.13
2.95
Monoterpenes
0.42
1.21
3.09
1.36
Other VOCs
0.38
1.26
3.36
1.46
Total BVOC
0.90
5.46
17.58
5.76
NO
0.18
0.32
0.59
0.35
Table 6. Comparison of various estimates of annual biogenic emissions (million metric tons)
for the contiguous United States.
Source
Isoprene
Total VOC
Biogenic NO
This work
17.2
29.8
1.4
BEIS2 for 1990s
17.8
30.5
1.5
Lamb et al. (1987)18
2.7
19.4
—
Lamb et al. (1993)19
5.9
29.1
—
Williams et al.15
—
—
0.7
LIST OF FIGURES
Figure 1. Daily emissions of isoprene and biogenic VOCs for the contiguous United States
estimated with BEIS2.2 for 1995.
Figure 2. Daily emissions of biogenic NO for the contiguous United States estimated with
BEIS2.2 for 1995.
Figure 3. Spatial distribution of isoprene emissions estimated with BEIS2.2 for 1995.
Figure 4. Spatial distribution of biogenic VOC emissions estimated with BEIS2.2 for 1995.
Figure 5. Spatial distribution of biogenic NO emissions estimated with BEIS2.2 for 1995.
-------
300,000
»
§ 250,000
1—
O
0
w
c
o
w
0
111
O
o
>
200,000
150,000
100,000
50,000
0
0
A yi
I
A A/It
>\ fr\r*
. 1
V w
)VUu
j/\l\
\
laj /5
Ml ^
11 /'•
f* 1 * 1 *
n* t 4 v* i«
>« • « Si t<
h
I v «
t* * ^ •*
1 « « t « i
ki If*
*n • * t
'
\
i
y
•
#1/ V
I
1
i
'* \ A A
A.a/* ^|Vv
hi |A I i ;«
Pi Jr\l ; *•
/ ii if .i ^
K 5f
t
/lsU,.wil
J U Af"
r ./fw i
f 4V v.;1
Isop
rene
~ • *5 A IAi
« < ! « i jfi
i • ! «1 »1
i i !#« • «
• * V** •**
w* J-/' K f?" •••
^V*r a * i
" ¥v
50
100
150 200
Day of Year
250
300
350
-------
8,000
*0 7,000
c
£ 6,000
0 '
© 5,000
2
« 4,000
c
1 3,000
m '
| 2,000
z 1,000
0
0 50 100 150 200 250 300 350
Day of Year
/I JIM .
/m
llli
r\
L
u/
n
||V
i
-------
Biogenic Isoprene Emission Rux for 1995
(metric tons per square kilometer)
-------
Biogenic VOC Emission Flux for 1995
(metric tons per square kilometer)
-------
Biogenic NO Emission Flux for 1995
(metric tons per square kilometer)
-------
TECHNICAL REPORT DATA
1. REPORT NO.
EP A/600/A-96/11 9
2.
3.
4. TITLE AND SUBTITLE
Biogenic emissions of 1995
5.REPORT DATE
6.PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
T.E, Pierce1* and M.P. Dudek2
8.PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
'U.S. EPA, Atmospheric Modeling Division, National Exposure Research
Laboratory, Research Triangle Park, NC
JDyntel Corporation, Research Triangle Park, NC
*NOAA personnel assigned to EPA
10 PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
NATIONAL EXPOSURE RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NC 27711
13.TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/600/9
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Biogenic emissions during 1995 for the contiguous United States have been estimated with the Biogenic Emissions
Inventory System (BEIS2.2). Hourly emissions were computed for each county using surface observations from the National
Weather Service and land use data developed specifically for biogenic emission calculations. Meteorological data were
interpolated to each county with Barnes analysis. The occurrence of the first and last date of freezing, as interpolated using
Barnes analysis, was used to toggle the occurrence of deciduous leaf biomass. The estimates indicate annual emissions of
17.2 million metric tons (Tg) of isoprene, 6.1 Tg of monoterpenes, 6.5 Tgof other volatile organic compounds, and 1.4 Tg of
nitric oxides. These estimates are reasonably consistent with those made by other researchers and are slightly lower than
estimates for 1990 made with an earlier version of BEIS2. The slight decrease in estimated 1995 emissions compared to
1990 can be attributed to the use of freezing dates, better temporal resolution of data (hourly values versus monthly diurnal
averages), and year-to-year variations in meteorology.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b IDENTIFIERS/ OPEN ENDED TERMS
c.COSATI
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21.NO. OF PAGES
20, SECURITY CLASS (This Page)
UNCLASSIFIED
22. PRICE
------- |