Introduction
An anthropogenic global
emissions inventory with both seasonal
and sector distribution on two height levels (0 - 100m and > 100m) has
been prepared using an emissions data processing system developed for handling
large global data sets. While the inventory presently described is for
emissions of SO2, SO4, NO and NO2, the system is flexible in nature and
can be expanded to many species and sector definitions. The present inventory
is concerned with emissions estimates of SO2, SO4, NO and NO2 for the sectors
listed in Table 1. The two height levels are based
on the physical stack height of point source emission where such data are
available. It is assumed that emissions from any stack with a physical
height greater than 50 metres, will be released in atmospheric layers above
100m after allowing for an expected plume rise. In the absence of stack
information, the type of source is used to infer the initial plume transport
level. Power plants and smelters, for example, are assigned to the upper
inventory level. The inventory has a global resolution of 1°
x 1° on a latitude/longitude grid and is
for the base year circa. 1985. The uncertainty of the global inventory
has been estimated regionally, based on the methodologies used for the
data sets included in the global inventory.
Methodology
Table 2 summarizes the national,
regional and global data sets used in compiling the global inventory with
seasonal and sectoral resolution. The approach used to develop the global
inventories, retains all of the detailed information available in regional/global
emission inventories and uses the inventory system to transform the data
to a common grid projection with sectoral breakdown. The data sets are
then merged to provide a global inventory while keeping the regional data
with the highest resolution as separate quality assured files. Several
options are available for merging data sets such as merging by country,
replacing individual grid cell emission values for any or all of the species/sectors
and overlaying global default emissions data with detailed sector information.
Figure 1 shows hierarchy used in overlaying data
sets for the present inventory leading to a seasonally resolved, two-level
global inventory for the sectors listed in Table 1.
Global Default Emissions: The 1°
x 1° global default inventory for NOx is
by Dignon (1992). This inventory is based on fossil fuel use only and has
no sectoral breakdown. The default for SO2 is the 1°
x 1° global inventory of Spiro et al (1992),
updated to 1985 by Jacobs (personal communication, 1994) which reports
emissions for 13 source categories and provides information on emission
contributions from transportation industry, fuel use and smelters. Country
coding was assigned to the global default inventories based on the a 1985
global population file (Logan, 1993). The country coding allows for replacement
of geographic regions with more detailed regional emission estimates.
North American Emissions: The National Acid
Precipitation Assessment Program (NAPAP) (US EPA, 1989) produced a modelling
emissions inventory for Canada and the Continental United States that has
detailed sector information, temporal distribution, as well as point and
area source information. For the presently described global inventory,
the point sources from the NAPAP inventory were divided into two vertical
emission levels based on whether the physical stack height is above or
below 50 meters. The taller stacks (>50 m) were assigned to the upper level
and further classified into smelters, power plants and other sources based
on the Source Classification Code. The area sources were aggregated from
the finer 1/4° x 1/6°
lat/lon NAPAP grid to the 1°x1°
global grid and assigned country coding based on the centroid of the NAPAP
grid. The area sources were grouped mobile and non-mobile sources (Sector
codes 51000 and 52000 in Table 1). The minor point
sources (those with heights <50m) were aggregated and assigned the Sector
code 53000 to differentiate minor point sources from other area sources
. The data derived from the NAPAP inventory directly replaces the global
default SOx and NOx inventories. Seasonal temporal factors were computed
on an emission weighted basis for each sector. The NAPAP inventory does
not extend beyond 60o N and emissions in this area of North America are
provided by the default inventories.
Asian Emissions: The regional inventory
of SO2 and NO2 by Kato & Akimoto (1992) is for 25 Asian countries,
and was used to replace the global default emissions on a grid cell basis.
Within a sub-domain of the Kato and Akimoto inventory, the more highly
resolved inventory of Tonooka (1992) for SO2 emission in five Asian
countries (China, Japan, North Korea, South Korea and Taiwan). The Tonooka
data resolve the SOx emissions with respect to both sector and temporal
distributions. Major NOx and SO2 point source emissions from smelters and
power plants in the former USSR have been compiled by Pacyna (1991). These
point source emissions were substituted in the default inventory to provide
height and sector resolution for the Asian part of the former USSR.
European Emissions: A number of emission
inventories have been developed for Europe. For preparing the presently
described global inventory, gridded emissions data provided by the Co-Operative
Program for Monitoring and Evaluation of the Long Range Transmission of
Air Pollutants in Europe (EMEP) (Sandnes and Styve, 1992) and the UN/ECE
CORINAIR project (Bouscaren, 1990) were used. The EMEP inventory contains
temporal, two-level, country coded information on a 150 km polar-stereographic
grid. The emissions were transformed to the 1°x1°
lat/lon grid by area apportionment and used to replace the global default
emissions for Europe. Sector breakdown within a sub-domain of the EMEP
grid is provided by the CORINAIR inventory, which was used in place of
EMEP for some central European countries. Those major NOx and SO2 point
sources in the European part of the former USSR which are not included
in the EMEP domain are from Pacyna (1991).
Australia and Africa: The Australian emission
estimates obtained from Carnovale (Carnovale et. al. 1992) and South African
emissions obtained from Lloyd (1993) both contained sector coded SOx and
NOx emissions for major point sources. These data sets were used to replace
global default emissions.
Uncertainty
The relative uncertainty of the inventory has been rated
regionally as high, medium or low based on the methodology used in preparing
the regional inventories which form part of the global data set. The NAPAP,
CORINAIR, Lloyd (1993), Tonooka (1993) and Carnovale (1992) are estimated
to have low uncertainty; EMEP, Kato and Akimoto (1992) and Pacyna (1991),
medium uncertainty; Dignon (1992) and Spiro et al (1992), high uncertainty.
Results
As is noted in Table 2, not all
of the available data sets used have the required two-level, sectoral and
seasonal breakdown. Table 3 summarizes the extent
of sectoral resolution achieved as a percentage of the total sulphur and
nitrogen emissions. All of the global sulphur emissions have some sectoral
breakdown while 67% of the global NOx emissions have been disaggregated
into the sectors listed in Table 1. The remaining
undistributed NOx emissions are expected to be mainly at low level. The
regional distribution of uncertainty rating is show in Figure
2.
Examples of the global distribution of Level 1 sources (<100m) and Level 2 sources (>100m) of SOx and NOx emissions are shown in Figures 3, 4, 5, and 6.
References
Bouscaren, R. "Inventaire des Emissions de Polluants dans
L'Atmosphere dans La Communate Europeenne en 1985". CITEPA. 1990.
Carnovale, F., P. Alviano, C. Carvalho, G. Dietch, S.
Jiang, D. Macaulay and M. Summers. "Air Emissions Inventory for the Port
Phillip Control Region". Clean Air 26: 134-144. 1992.
Dignon, J. "NOx and SOx Emissions from Fossil Fuels: A
Global Distribution". Atmos. Env. 26a: 1157-1163. 1992.
Kato, N. and H. Akimoto. "Anthropogenic Emissions of SO2
and NOx in Asia: Emissions Inventories (plus errata)". Atmos. Env. 26a:
2997-3017. 1992.
Lloyd, S.M., South African Department of Health and Population
Development, personal communication, 1993.
Logan J. Personnel Communications. Harvard University,
USA. 1993.
Sandnes, H. and H. Styve. "Calculated Budgets for Airborne
Acidifying Components in Europe, 1985, 1986, 1987, 1988, 1989, 1990 and
1991". Meteorological Synthesizing Centre - West. The Norwegian Meteorological
Institute. EMEP/MSC-W 1/92. Oslo, Norway. 1992.
Spiro, P.A., D.J. Jacob and J.A. Logan. "Global Inventory
of Sulfur Emissions with a 1° x 1°
Resolution". J. Geophys. Res. 97: 6023-6036. 1992.
Tonooka , Y., Personnel Communications. 1993.
U.S. EPA, "The 1985 NAPAP Emission Inventory (Version
2): Development of the Annual Data and Modelers Tapes", EPA-600/7-89-012a.
November 1989.
Acknowledgments
The authors would like to acknowledge the co-operation
and assistance of the members of the Global Emissions Inventory Activities
(GEIA) SOx/NOx Working Group in the preparation of this global inventory.
Table 1: Sector List
Table 2: Data Set Details
Table 3: Global SOx,NOx Emission
by Sector
Figure 1: Overlaying procedure
Figure 2: Distribution of Regional
Uncertainty Rating
Figure 3: Global Area Source
SOx
Figure 4: Global Point Source
SOx
Figure 5: Global Area Source
NOx
Figure 6: Global Point Source
NOx
Figure 7: Distribution of Temporal
Resolution
