For O3, a large fraction of the observed concentrations is due to non-anthropogenic emission sources, including precursor emissions from vegetation, fires, lightning, and soils, and intrusion from the stratosphere. The complex and varying background sources of ozone precursor emissions are augmented by and interact with varying anthropogenic sources of precursor emissions within a region and outside a region. The contribution of each of the natural and anthropogenic components varies on different time scales from annual to hourly and by location and elevation. Thus, the contribution of and sensitivity to extra-regional emission sources depends on the metric, or policy objective, of interest. The highest O3 concentrations are typically associated with local or regional emissions sources, stagnant meteorology, and high temperatures. However, natural and extra-regional anthropogenic sources alone can produce O3 concentrations that exceed health-protective levels.
The impact of extra-regional anthropogenic emissions is more important for ground-level ozone than for particulate matter or sulfur or nitrogen deposition.
- Extra-regional emissions are more important for column ozone burdens and ozone direct radiative forcing than for surface ozone concentrations.
- For surface ozone, the impact of extra-regional emissions sources is highest in Spring and lowest during Summer.
- For surface ozone, the impact of extra-regional emission sources is larger for longer-term average concentrations than for peak concentrations that tend to be driven by local emissions.
Methane, in combination with nitrogen oxides, is an important ozone precursor on a global basis, and projected increases in methane emissions could, over time, offset the ozone decreases associated with decreases of emissions of nitrogen oxides (NOx) and volatile organic compounds (VOCs).
For Europe, ground-level ozone concentrations are more sensitive to anthropogenic emissions outside of Europe than to anthropogenic emissions inside of Europe. This is the case for all European sub-regions, all seasons, and for a range of ozone metrics including annual and seasonal averages, SOMO35, and POD1. The relative influence of these extra-regional emissions on ozone in Europe varies by location, season, and ozone metric. Extra-regional anthropogenic emissions of NOx and VOCs outside of Europe are estimated to contribute between 2-12 ppb of ozone depending on the season, with an annual average contribution of 4-8 ppb of ozone depending on the model used. The contribution of anthropogenic methane emissions to ozone in Europe is estimated to be about 5-8 ppb (on the basis of 6mDMA1).
For North America, region-wide annual average ozone concentrations appear equally sensitive to anthropogenic emissions outside of North America and emissions inside of North America. Extra-regional anthropogenic emissions of NOx and VOCs outside of North America and changes in global methane, together, are estimated to contribute between 8-13 ppb of ozone in the western United States, 2-12 ppb of ozone in the central United States, and 2-10 ppb of ozone in the eastern United States, depending on the season.
Maritime shipping is well understood to have significant impacts on air quality in coastal regions. In the HTAP2 experiments, all maritime shipping was considered to be extra-regional, leading to an increase in the estimated extra-regional contribution to air pollution levels in all regions compared to HTAP1. Further analysis is needed to understand the source-receptor relationships associated with shipping in different coastal regions subject to ECAs (Emission Control Areas) compared to shipping emissions in the open ocean.
PM2.5 and Deposition
As with O3, observed PM2.5 concentrations are the result of contributions from non-anthropogenic sources, including emissions from vegetation, natural fires, wind-blown dust, sea-salt, and volcanos. Anthropogenic sources from near and far add to these natural sources. The contribution of each of the natural and anthropogenic components varies on different time scales from annual to hourly and by location and elevation. Both natural and anthropogenic sources can affect PM2.5 concentrations on intercontinental scales.
The relative impact of extra-regional anthropogenic sources on PM2.5 concentrations is smaller than on O3 concentrations. The sensitivity of the deposition of sulfur (S), oxidized nitrogen (NOy), and reduced nitrogen (NHx) to changes in extra-regional anthropogenic emissions is similar to the sensitivity of PM2.5 concentrations. In North America, 83% of S deposition, 83 % of NOy deposition, and 93% of NH3 deposition are due to sources within North America. In Europe, the fractions are 64%, 66%, and 88% for S, NOy, and NHX deposition, respectively. However, in the region defined by Russia, Ukraine, and Belarus, only 39%, 41%, and 45% of S, NOy, and NHX deposition, respectively, are due to emissions within the region.
Based on the HTAP1 multi-model experiments, natural and re-emitted Hg account for about 35% to 70% of total Hg deposition on a region-wide, annual average basis depending on the region, whereas intercontinental transport of newly-released anthropogenic Hg emissions accounts for about 10% to 30% of total Hg deposition, on a region-wide, annual average basis. East Asia, which accounted for almost 40% of total global newly-released Hg in 2000, is the most dominant among the four HTAP1 source regions, accounting for 10% to 14% of the annual Hg deposition found in other regions, followed by contributions from Europe, South Asia, and North America. However, where deposition is highest, local and regional anthropogenic emission sources are the dominant sources of Hg deposition.
The Arctic has no anthropogenic emission sources within the region, so anthropogenic Hg that is deposited there comes from intercontinental transport. A large fraction of transported Hg is deposited in springtime during polar sunrise in atmospheric Hg depletion events (AMDEs), during which Hg0 is rapidly oxidized by photochemical reactants that have built up over the dark winter. Current models show significant deviation in estimates of Hg deposition to the Arctic due to the uncertainties in the model formulation of the processes related to AMDEs and transport to the polar region.
The global Hg models participating in the HTAP1 multi-model experiments provided relatively consistent estimates of the impact of one source region on another despite significant differences in emissions and chemistry in each model. In these global models, the change in Hg deposition in one region is linearly related to the change in emissions in another region, provided that the relative proportion of Hg0 and ionic Hg in the anthropogenic emissions stays the same in both regions. Increases in Hg0 emissions lead to enhanced export, whereas increases in ionic Hg result in higher local deposition in the region itself.
In the Arctic, Hg deposition could be most efficiently controlled by emission reductions in East Asia and Europe due to their proximity to the Arctic, prevailing atmospheric circulation patterns, and the significant contribution of these regions to global anthropogenic emissions.
Within a receptor region, proximity to the intercontinental source region has a small effect on the spatial distribution of Hg deposition. For example, relative Hg deposition decrease in North America caused by emission reduction in East Asia is somewhat higher in the western part of North America, whereas response to the European emission reduction prevails in eastern North America.
The response of Hg deposition to changes in emissions varies with season. Intercontinental transport may be a factor of two more important in summer than in winter due to more active mixing of the boundary layer with air aloft. Thus, seasonal variation of the deposition response to emission reduction in other continents can reach 30% of the annual mean. Distinctly different seasonal cycles characterize the transport between adjacent regions such as East Asia and South Asia or Europe and South Asia, which are driven by changes in monsoonal systems. For example, the most significant effect of South Asian emission reduction on Hg deposition in East Asia takes place during the first half of a year, whereas the opposite is true of the effect of East Asian sources on South Asia, which is most significant in the second half. Similarly, South Asian sources have a somewhat larger influence on Europe in winter; whereas European sources more strongly affect Hg deposition to South Asia in summer.
In the year-long simulations, the large contribution of natural sources and re-emitted legacy Hg to deposition dampens the relative response of Hg deposition to the reduction in new anthropogenic emissions, whether associated with local and regional sources or intercontinental transport. If integrated over much longer time scales, the decrease in deposition will be larger as decreasing new emissions will slowly decrease the amount of legacy Hg that is cycling in the environment.
Persistent Organic Pollutants
As part of the HTAP1 multi-model experiments, the transport of several POPs (Polychlorinated biphenyl-28 (PCB-28), PCB-153, PCB-180, and α-HCH), covering a range of different physical and chemical properties and for which some emissions and observational data are available, was simulated by three models. The three models used different modelling approaches to describe pollutant transport through environmental compartments and different spatial resolutions. Despite their differences, the models provide generally consistent estimates of annual average atmospheric concentrations, the major transport pathways of the selected POPs, and responses to primary emission changes. In most of the cases, the differences between the mean response and the model estimates are within a factor of 2 to 3.
As with the other pollutants, the largest impact on deposition, in most cases, results from changes in local and regional emissions within the region, especially for regions with higher emissions (North America and Europe for PCBs, and South Asia and Europe for a-HCH).
Arctic pollution is mostly sensitive to the changes of emissions in Europe (for all simulated POPs) followed by North America (for PCBs) and South Asia (for α-HCH).
The regional differences in the efficiency of atmospheric POPs transport alone were explored by simulating the transport of selected POPs assuming an equal mass of emissions in each source region. In contrast to the results with regionally different emissions, the uniform emissions produced the same level of response to emission changes for all source-receptor region pairs, about 1.5%-2%. A slightly more significant effect is obtained for the closest pair of regions, South Asia’s influence on East Asia (2-3%). From the equal emission simulations, transport to the Arctic is most efficient from Europe, producing a 5-6% response. A slightly lower response in the Arctic can be seen for emission changes in North America and East Asia (about 4%) and the Arctic response is weakest for South Asian emissions (about 2%).
 SOMO35 = the annual sum of the positive differences between the daily maximum 8-hour average ozone value and the cutoff value set at 35 ppb; POD1 = phytotoxic ozone dose (accumulated stomatal flux above a threshold of 1 nmol/m2/s) accumulated over a growing season
 6mDMA1 = the running mean of the 6-month average of the daily maximum 1-hour ozone concentration