- Email: [email protected]
Atmospheric transport of ozone between Southern and Eastern Asia
Science of the Total Environment 523 (2015) 28–39
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Atmospheric transport of ozone between Southern and Eastern Asia T. Chakraborty a, G. Beig a,⁎, F.J. Dentener b, O. Wild c a b c
Indian Institute of Tropical Meteorology, Pune, India European Commission, Joint Research Centre, Institute for Environment and Sustainability, Ispra, Italy Lancaster Environment Centre, Lancaster University, Lancaster, UK
H I G H L I G H T S • • • •
Maximum effect of East Asian pollution over South Asia happens in post-monsoon. Maximum effect of South Asian pollution over East Asia occurs in pre-monsoon. Most densely populated parts of South Asia are affected by East Asian emission. South Asia is largely affected by the East Asian emission change from 2000 to 2010.
a r t i c l e
i n f o
Article history: Received 1 December 2014 Received in revised form 15 March 2015 Accepted 17 March 2015 Available online 7 April 2015 Editor: Xuexi Tie Keywords: Pollution transport HTAP Ozone East Asia South Asia
a b s t r a c t This study describes the effect of pollution transport between East Asia and South Asia on tropospheric ozone (O3) using model results from the Task Force on Hemispheric Transport of Air Pollution (TF HTAP). Ensemble mean O3 concentrations are evaluated against satellite-data and ground observations of surface O3 at four stations in India. Although modeled surface O3 concentrations are 1020 ppb higher than those observed, the relative magnitude of the seasonal cycle of O3 is reproduced well. Using 20% reductions in regional anthropogenic emissions, we quantify the seasonal variations in pollution transport between East Asia and South Asia. While there is only a difference of 0.05 to 0.1 ppb in the magnitudes of the regional contributions from one region to the other, O3 from East Asian sources affects the most densely populated parts of South Asia while Southern Asian sources only partly affect the populated parts of East Asia. We show that emission changes over East Asia between 2000 and 2010 had a larger impact on populated parts of South Asia than vice versa. This study will help inform future decisions on emission control policy over these regions. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Intercontinental transport of air pollution is a global issue as the air quality of a region can be highly influenced by long range transport from remote continents (HTAP, 2010). Pollutants with longer atmospheric life times are important when assessing the impacts of long range transport, but shorter lived air pollutants that produce secondary pollutants are also important. Tropospheric ozone (O3) is a secondary pollutant produced through a sequence of photochemical reactions involving nitrogen oxides (NOx), carbon monoxide (CO), volatile organic compounds (VOCs) and CH4. Due to the long lifetime of CH4, the O3 produced from its oxidation is largely independent of the location of the CH4 emissions (Fiore et al., 2008). The spatial pattern of the O3 formed is controlled by the distribution of OH and NOx, which have much shorter lifetimes and can affect both the location of CH4 oxidation and the amount of O3 production per CH4 molecule oxidized (Fiore et al., ⁎ Corresponding author. E-mail address: [email protected] (G. Beig).
http://dx.doi.org/10.1016/j.scitotenv.2015.03.066 0048-9697/© 2015 Elsevier B.V. All rights reserved.
2008). Increases in atmospheric CH4 contribute to all regions relatively uniformly, averaging 1.5–1.9 ppb O3 since 1960, and this contributes about one third of the O3 increase seen over Europe (EU) and North America (NA) over this period (Wild et al., 2012). From a series of future (2005–2030) transient simulations, it has been demonstrated that costeffective CH4 controls could offset the positive climate forcing from CH4 and O3 and improve air quality (Fiore et al., 2008). Measurements from remote regions consistently indicate that longrange transport exerts a strong influence on observed concentrations of aerosols, O3 and its important precursors (HTAP, 2007). For example, dust of Asian origin has been observed throughout the North Pacific region (Duce et al., 1980; Prospero, 1979). Studies over the west coast of NA identified the influence of Asian emissions on the sulfur budget (Andreae et al., 1988) and on the concentrations of O3, hydrocarbons and peroxyacetyl nitrate (Parrish et al., 1992). Several observation and model-based studies have shown the impact of foreign emissions on countries at Northern mid-latitudes. For example, EU and NA respectively contribute 3.5 ± 1.1 and 2.2 ± 0.5 ppb to the annual mean surface O3 over Japan (Yoshitomi et al., 2011). The annual mean magnitude of
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
NA and EU contributions to surface O3 over East China is 1.2 ppbv and 1.5 ppbv respectively. EU experiences the greatest intercontinental import of O3 due to rapid short-distance transport from NA (Wild and Akimoto, 2001). A widely used technique to study the contribution of upwind emissions to a region downwind is the calculation of source–receptor relationships, which evaluate the impact of a relatively small emission perturbation over a specific source region (Derwent et al., 2001; Fiore et al., 2002; Wild and Akimoto, 2001). Source–receptor relationships depend on various factors including the emission strength, size of the source region, the transport pathway, and the extent of pollutant transformation and loss during transport. This study quantify the source–receptor relationship for surface O3 between East Asia (EA, 15–55°N, 95–160°E) and South Asia (SA, 5°– 35°N, 50°–95°E) using the Task Force on Hemispheric Transport of Air Pollution (HTAP) Phase 1 source–receptor model experiments (www. htap.org). EA and SA are two major developing industrial regions that are also the most heavily populated regions in the world. Development in each of these regions not only affects air quality in the region itself, but also air quality in the other region through the transport of air pollutants from one region to the other. There is widespread scientific agreement that the observed increase in O3 concentrations is the consequence of human activities around the globe. Among anthropogenic factors, the principal one is increasing population (Dietz and Rosa, 1997). A population density map for the year 2000 (Fig. 1) shows that India, Central and Eastern China, Japan, Bangladesh and Korea are the most densely populated regions. Bangladesh and the Indo-Gangetic plain of India are the most densely populated parts of southern Asia. Model simulated surface O3 over this region is quite high, ranging from 35 to 55 ppbv (Fig. 2). Surface O3 over the most populated parts of EA (the coastal region of China, Japan, and Korea) ranges from 45 to 65 ppb. Anthropogenic emissions over both Southern and Eastern Asia have experienced a rapid increase in recent decades (Hilboll et al., 2013; Richter et al., 2005; Zhang et al., 2007). Model based study of biomass burning over SA shows that there is enhancement of surface ozone of about 4 to10 ppb (25–50%) in the Eastern region including Burma (during March and April), 1–3 ppb (10–25%) in the Central India (during
29
Fig. 2. Surface O3 concentration for the months of (a) May, (b) July and (c) October, in the ensemble average of 14 models.
Fig. 1. The population density in Asia (http://sedac.ciesin.columbia.edu).
30
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
March) and 1–7 ppb (4–10%) in the Indo-Gangetic region (Jena et al., 2015). Studies of background O3 concentrations over China for the period 1994 to 2007 suggest that surface O3 has been increasing at an average rate of 0.58 ppbv per year (Wang et al., 2009). This is associated with increased NOx emissions, which have been clearly observed in GOME and SCIAMACHY tropospheric NO2 satellite measurements (Lee et al., 2014; Wang et al., 2009) in China's fast-developing coastal regions. Inter-continental transport of air pollution also plays an important role in the seasonal variation of air quality over a region (Tu et al., 2007). The seasonal outflow of pollutants from EA is maximum in spring as observed both by in situ observation and satellite studies (Duce et al., 1980; Stegmann, 1999). The surface O3 distribution over EA varies dynamically from season to season according to the meteorological conditions and chemical production of O3 from precursor emissions over the region. Yamaji et al. (2006) showed that only 10% (winter) to 54% (summer) O3 concentration over EA could be explained by local emissions. Thus the concentration of O3 in air masses over EA is strongly affected by long-range transport of O3 or its precursors from outside EA, especially in winter and early spring. Global atmospheric modeling studies show that the increase in surface emission of NOx and VOCs over Asia leads to an increase in surface O3 concentrations. These increases are strongest over the polluted regions of northern India, China and Japan due to local sources in these regions (Bethan et al., 1996). The impact of Asian emissions on the global environment has been examined through observational and model studies focusing on key export paths like transpacific transport (Carmichael, 2003; Cooper, 2004; Liu, 2003), seasonal and episodic variability (Stohl et al., 2007; Yienger et al., 2000), and impacts over North America (Fiore et al., 2002; Reidmiller et al., 2009; Singh and Brune, 2009; Zhang et al., 2008) and Europe (Fiedler et al., 2009; Stohl et al., 2007). Future projections based on Representative Concentration Pathways (RCPs) generated for the Climate Model Inter comparison Project (CMIP5) simulations show a large range in emission changes over SA between the different scenarios, and consequently large differences in surface O3 are expected (Wild et al., 2012). Surface O3 over SA under RCP 8.5 shows an increase of more than 5 ppb by 2050 but for RCP 6.0 the changes are close to zero. There is also an increase in surface O3 over EA until 2020 but a decrease for the RCP 2.6 and 4.5 scenarios. This suggests that there is substantial uncertainty in future air quality over the Asian region, but in many cases there is an increase in surface O3. It is therefore important to study how emissions from EA and SA affect air quality over the other region. This will help future science and policy interactions, as well as decision making on emission control policy over these regions. Long range transport of pollutants within Asia has not been studied in detail, with regard to these policy aspects. The seasonality in surface O3 over SA mainly reflects the influence of the Asian Summer Monsoon, hence is quite different from mid-latitude regions. In all seasons, the influence of SA emissions over mid-latitude regions such as NA is much less than that from EA and EU sources (Reidmiller et al., 2009). The reason for this relatively small long-range transport is the meteorological conditions over SA, which combine low, near-surface wind during most of the year with wet scavenging of pollutants due to the abundance of rainfall in the tropical environment (Engardt et al., 2005). The major focus of this study is to quantify the surface O3 concentration change over SA due to changes in emission over the EA region and viceversa.
anthropogenic emission perturbations. In December 2004, the executive body of the UNECE convention on Long Range Trans-boundary Air Pollution (LRTAP) established the Task Force on Hemispheric Transport of Air Pollution (HTAP). During its first phase (2006–2010), the task force coordinated a multi-model assessment to quantify the importance of intercontinental transport of O3 and its precursors and particulate matter. These studies concluded that the O3 response to emission changes in upwind regions is not negligible and that it is particularly strong in spring and fall at northern mid-latitudes. This study uses results from fourteen models contributing to the HTAP intercomparison: ECHAM5, EMEP, FRSGC/UCI, GEMAQ, GEOS-Chem, GISS, LLNL, MOZART, MOZECH, STOC-HadAM3, STOCHEM, TM5, ULAQ, and UM-CAM. The models are driven by meteorological fields for 2001. An average over these 14 models after regriding the results to a common 2.5° × 2.5° resolution is used for the study. Results are for the model's lowest grid box, and since the relevant meteorological information is not available, no attempt is made to extrapolate to near-surface levels corresponding to measurement stations; the analysis considers a base simulation and a perturbation scenario. In the perturbation scenario all anthropogenic emissions including NOx, CO, VOC, SO2 and PM components are reduced by 20% over either SA or EA source regions (identified by HTAP label SR6). The annual mean surface O3 concentration in the base simulation and the standard deviation (relative to the ensemble mean) over the two source regions are shown in Fig. 3. The standard deviation is relatively uniform over the study region, which suggests that the models agree well in the spatial pattern but there is a relatively large absolute difference of 14–20 ppbv between models. The mean annual surface O3 over SA and EA varies from 40–55 ppbv, with high values over the
2. Model evaluation with satellite data and surface observations 2.1. Model simulation This paper addresses the seasonal variation of the source receptor relationship between SA and EA by evaluating the effect of large-scale
Fig. 3. (a) Annual mean surface O3 concentration and (b) the relative standard deviation of 14 models.
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
Himalayan region where the surface altitude is high and where intrusions of ozone-rich stratospheric air may also contribute to high surface ozone (Ma et al., 2014; Moore and Semple, 2005).
31
obtained from Surface Radiation Budget Network (SURFRAD) (Wu et al., 2015). 2.3. Surface observations
2.2. Satellite data The Tropospheric Emission Spectrometer (TES), the third of NASA's Earth observing system (EOS) satellite instruments, was launched into sun-synchronous orbit on 15 July 2004. TES is an infrared, high spectral resolution Fourier Transform Spectrometer (FTS). TES nadir-view has a vertical resolution of 6–7 km in cloudless conditions (Bowman, 2002; Worden et al., 2007; Worden, 2004). TES measurements have a positive bias in O3 relative to ozonesondes (Nassar et al., 2008). TES level 3 total tropospheric column ozone (TTCO) monthly mean data at 4°lon × 2°lat resolution for the period 2005 to 2009 have been used to evaluate the model simulations. It is important to identify the proper tropopause height to avoid stratospheric air in the TTCO calculation, because the tropopause height changes with latitude and season. For this calculation, an ozone-based tropopause is defined where O3 mixing ratio exceeds 110 ppbv (Bethan et al., 1996) and consequently O3 concentration in the troposphere–stratosphere region exceeding 110 ppbv is assumed to be in the stratosphere. Many recent studies demonstrate the use of satellite observations in various applications related to terrestrial environment. For example, the result obtained from the MODIS satellite images to estimate daily water requirement using SEBAL algorithm over Tajan catchment is very satisfactory (Rahimi et al., 2014). Land surface temperature obtained from spatio-temporal integrated temperature fusion model (STITFM), which is used to retrieve high temporal and spatial resolution data from multi-scale polar-orbiting and geostationary satellite observation shows an accuracy of 2.5 K with the in-situ land surface temperature
The performance of the models has been validated over four stations in India: Pune (18.5°N, 73.85°E), Udaipur (17.38°N, 78.48°E), Hyderabad (24.5°N, 73.69°E) and Jabalpur (23.16°N, 79.94°E), shown in Fig. 4. Pune is located at 560 m above sea level on the Deccan plateau in western India. Udaipur has an average elevation of 598 m and is located in the southern part of Rajasthan. Hyderabad is located at an average altitude of 542 m, on hilly terrain in Southern India. Udaipur is located in central India at an average elevation of 412 m. All stations are situated in arid and semi-arid regions. All the monitoring stations are designated as semi-urban indicating that stations are away from downtown areas where the influence of local emissions may be high. However, local effects cannot be ruled out under some conditions. Observations over these stations were made with Air Quality Management System (AQMS). Each AQMS comprised of US Environmental Protection Agency approved analyzers housed inside walkway shelters. Ozone was measured with a photometric UV analyzer (Thermo-49i, precision 1 ppbv). Calibration of the O3 analyzer was done on every alternate day using an inbuilt O3 calibrator. The inlets for O3 measurements were located about 3 m above the surface at all of the stations (Beig et al., 2013). 2.4. Comparison between model and surface observations The four stations considered show minimum surface O3 values in the monsoon period and maximum values in the pre-monsoon. The seasonal variation of surface O3 over the four stations is shown in Fig. 5, based
(c) (b) (a) (d)
Fig. 4. Geographical locations of four stations: (a) Pune, (b) Jabalpur, (c) Udaipur and (d) Hyderabad.
32
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
Fig. 5. Comparison between the model ensemble monthly mean O3 and surface observations over (a) Hyderabad, (b) Pune, (c) Jabalpur and (d) Udaipur.
on monthly averages of observations taken each hour and ensemble mean surface O3 from the models over the corresponding station location. The model simulations reproduce the seasonal cycle of surface O3 concentration over the locations, but are substantially higher. The ground-level ozone mixing ratios from the models have been examined collectively as an ensemble and evaluated against a large set of observations from both Europe and East Asia (Fiore et al., 2009). This commonly used multi-model approach can be outperformed by subsets of models optimally selected in terms of bias, error, and correlation, but it does not strictly depend on the skill of individual models, although it may require the inclusion of low-ranking skill-score members (Solazzo et al., 2012). The difference between the models and surface observations may be due to the fact that the global models are not able to capture fine-scale emission patterns at the coarse resolutions used here. A further reason may be that the relatively low sampling height (3 m) does not match the altitude of the middle of each model's grid box. An earlier analysis by Van Dingenen et al. (2009), comparing 30 meter and 10 meter concentrations, showed that in India this may lead to overestimates of the order of 20–30 ppb, and this bias may be even higher when comparing with 3 m. Another reason for this difference may be that over India NOx, one of the most important precursors of O3, is strongly affected by local emissions. Biomass and fossil fuel burning is the main source of NOx over this region. Due to the increasing population and higher economic growth rates, emission of these gases is increasing over SA and EA (Akimoto, 2003; Sheel et al., 2010). Surface observations of NOx over these four stations are much higher (5–17 ppbv) than in the model simulations (0.1–2 ppbv). This suggests that destruction of O3 by direct reaction with emitted NO is much more important than seen in the models, accounting for the overestimation of surface O3 in the models.
2.5. Comparison between model and satellite observation at location of the four stations The seasonal variation of ensemble-mean TTCO over station locations (generated by interpolating the gridded data to the exact station locations) is compared with satellite data in Fig. 6. The model results are consistent with the satellite observations but show systematically higher O3 columns (35–45 DU) in monsoon months (June–September) relative to the satellite observations (27– 40 DU) in these months, whereas in the dry months the bias is much smaller (1–5 DU). The difference in O3 column between satellite observations and model simulations may be due to the fact that in monsoon months the thick clouds in the field of view obscure the infrared emission from the lower troposphere, which greatly reduces TES sensitivity (Nassar et al., 2008). There is also missing satellite data over some parts of Asia in monsoon months. Comparison of ozonesonde data from 32 ground stations around the world suggests that TES overestimates the O3 column by 4 DU (Osterman et al., 2008). An inter-comparison study between satellite TES and OMI (O3 monitoring instrument) shows that monthly mean differences between OMI and TES exceed ±10 DU. The bias and standard deviation which are determined using correlations of mean values of TES O3 and sonde O3 range from 2.9 to 10.6 ppbv in the upper troposphere and 3.7 to 9.2 ppbv in the lower troposphere excluding the Arctic and Antarctic where TES sensitivity is low (Nassar et al., 2008).
2.6. Comparison between model and satellite observation over EA and SA The area averaged TTCO from satellite and model ensemble-mean for SA and EA is shown in Fig. 7.
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
33
Fig. 6. Seasonal variation of the total tropospheric O3 column from model simulations and satellite observation over Hyderabad, Pune, Jabalpur and Udaipur.
The seasonal cycle of TTCO generated by the models over both SA and EA is comparable with satellite observations, but the O3 concentration is much greater in the models over SA, except in the months of April and May. Over EA the satellite shows slightly higher O3 in May, June and
July. The spatial pattern of the difference between model simulated and satellite observed TTCO over EA and SA for May, July and October is shown in Fig. 8. In May the models show lower O3 than the satellite over most parts of EA and SA by up to 3 DU. The models show higher
Fig. 7. Area averaged total tropospheric O3 column from model simulations and satellite observation over (a) SA and (b) EA. Black and green lines show the total tropospheric O3 column from model simulations and satellite observations respectively.
34
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39 Table 1 Dominating climatological conditions in Northern and Southern India during, pre-monsoon, monsoon, and post-monsoon. Pre-monsoon (March to May)
Monsoon (June to September)
Post-monsoon (October to December)
High temperatures over north and central India Coastal region dominated by land/sea breeze In May wind changes from NE to SW.
Heavy rain on west coast and northern India Cloudy conditions and washout of soluble precursors by rain SW wind brings cleaner air from Arabian Sea and Indian Ocean.
Wind changes from SW to NE in October SE India gets rainfall. Temperature falls sharply; decrease in humidity; clear sky over N and central India. Photochemical O3 production increases. NE winds bring polluted continental air from Ganges plains over India
occurs over SA in the monsoon season and the highest responses occur in the post-monsoon. The maximum effect of SA emissions over EA occurs in the pre-monsoon months. The contribution of SA to EA ranges from 0.1 to 0.3 ppbv with a peak in April. The EA contribution to SA ranges from 0.15 to 0.35 ppbv, increasing from June and reaching a maximum in November. The spatial standard deviation of the surface O3 response in SA due to EA emission is large in the post-monsoon months. This indicates that though the maximum response is in the post-monsoon, it is not uniform over the whole SA region. Over EA the spatial standard deviation of the response to SA emissions is greater in the pre-monsoon, indicating that the spread of SA pollutants over EA is also not uniform. It is clear that there is greater spatial variability in the effects of SA sources on EA than in the effects of EA sources on SA. The population weighted O3 change shows a similar seasonal variation in O3 contributions. Fig. 8. Difference between the model simulation and satellite observation of total tropospheric O3 column for the months of (a) May, (b) July and (c) October.
O3 than the satellite observations over most past of EA and SA for July and October of up to 6 DU. In all three months the models show much higher O3 over the Himalayan region. This may explain the higher O3 column over SA in the models than satellite observation in Fig. 7. The model columns over the four stations are more consistent with the satellite observations because over most part of SA the difference between the model simulation and the satellite observation ranges from 0 to 6 DU.
3.1. Spatial pattern of the O3 source–receptor relationship Figs. 10 & 11 illustrate the spatial patterns of O3 concentration change due to emission reductions over SA and EA, and the impact of emissions on the population for the months of May, July and October (characterizing the pre-monsoon, monsoon and post-monsoon periods, respectively) due to emission reductions over the other region. To estimate the impact of emission on the population density we have used a scheme (Ehrlich and Holdren, 1971) which can be expressed by the relation
3. Source–receptor relationships for surface O3
Population Impact ðPIÞ ¼ population density percentage change in ozone:
In how foreign emissions affect regional air quality over the SA and EA regions. The foreign influence on regional air quality is defined by analyzing the difference between the control simulations and those with 20% emission reductions over a given source region. The reduction in surface O3 over the receptor region shows how much emission from the source region influences it. To provide a more relevant assessment of the effects on human health, we also present the populationweighted O3 change over each of the receptor regions. The source contribution varies with season and is sensitive to the prevailing wind pattern. We describe the seasonal variation of this contribution in the context of the three main seasons over the Indian subcontinent: pre-monsoon, monsoon and post-monsoon. The prevailing circulation patterns which control the transport of pollutants change in direction over the seasons in this region, and a brief description is given in Table 1. The seasonal variation of the surface O3 response and population weighted changes in surface O3 over each region for a 20% anthropogenic emission reduction over the other region is shown in Fig. 9. The surface O3 response to foreign emission changes has a strong seasonal cycle over each region. The minimum response to EA emissions
3.1.1. Pre-monsoon (March to May) Surface O3 concentrations over both regions are quite high in this season (Fig. 2). The first panel in Fig. 10(a) shows that most parts of China and the adjacent oceanic region are affected by SA emissions but the most densely populated regions of eastern China, Japan, North and South Korea are not affected much. The impact of SA emissions on the populated parts of EA is not uniform but there are some parts of coastal China where the impact is high as shown in Fig. 10(b). The northeastern parts of India, Nepal, Bhutan, Bangladesh, Pakistan and Afghanistan are more strongly affected by emissions from EA. The most densely populated regions of the Indo Gangetic plain and Bangladesh are highly affected, but in contrast the southern parts of India and Sri Lanka are not influenced much. Fig. 11(b) shows the impact of EA emissions over the densely populated regions of the Indo Gangetic plain and Bangladesh. The area averaged contribution of SA sources to O3 over EA is 0.2 ppbv, and the EA contribution to SA is 0.15 ppbv. Although the effects of SA sources over EA are greater than those of EA over SA in this period, EA emissions affect the most densely populated parts of the SA region. The southwesterly wind pattern in May at 925 hPa transports pollutants from SA to EA.
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
35
Fig. 9. Regional average change in surface O3 over SA and EA due to a 20% emission change over the other region; (a) population weighted and (b) and area average. Error bars show the spatial standard deviation over each region.
Due to high pressure in the western Pacific between 140°E and 160°E shown in Fig. 12, SA sources do not affect the eastern parts of EA (Japan, North and South Korea).
3.1.2. Monsoon (June to September) We consider July as representative of this season and study the continental scale transport processes. During this season the surface O3
Fig. 10. (a) Percentage changes in surface O3 due to 20% emission reductions over SA and (b) the impact on population for May, July and October.
36
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
Fig. 11. Percentage changes in surface O3 due to 20% emission reductions over EA and (b) the impact on population for May, July and October.
concentration is less over EA than over SA (Fig. 2), and the spatial patterns of O3 changes over EA and SA due to emission reductions over the other region are shown in the second panel of Figs. 10 & 11. The area averaged surface O3 contributions over each region from the foreign source region in this month are almost equal, but the effect of SA sources over EA is less than during the Pre-Monsoon. From Fig. 11 we can see that the effect of O3 from EA sources is spread over almost all parts of SA. The spread of O3 from SA towards EA is mainly over the oceanic region and some parts of EA (China, and parts of Mongolia and Japan). The impact of SA emission over the densely populated regions of EA is less that in the pre-monsoon but the effect of EA emissions over Bangladesh and the Indo-Gangetic Plain is the same as in the premonsoon. The spatial standard deviation in the monsoon is higher over EA than over SA, indicating that the EA contribution to SA is more uniform than the SA contribution to EA. The distribution of surface O3 changes can be explained with the help of the dominating wind patterns (Fig. 12). Transport from SA is mainly governed by strong southwesterly winds in these months. Fig. 12(b) shows lower pressure over the Indian subcontinent, and the northeasterly winds north of 30°N transport O3 from EA to SA. The southwesterly winds mainly transport O3 over the oceanic region, and hence EA is not that much affected by O3 from SA. 3.1.3. Post-monsoon (October to December) Surface O3 concentrations over EA and SA in October show an increase in O3 production over SA (Fig. 2). The response of surface O3 concentrations over EA and SA due to 20% emission reductions over the other region is shown in the third panel of Figs. 10 & 11. The effect of SA emissions on EA is much less than in other months, and the EA contributions to SA are maximum in this season. Most parts of SA, especially the southeastern part of India and Sri Lanka, are strongly affected but
the northern parts of SA are less affected (Fig. 11). The relative change in O3 concentration over SA ranges from 1% to 0.2% spatially. The effect of SA emissions spread over the Tibetan plateau, Mongolia and oceanic regions. The effect of SA emission on densely populated parts of EA is less in the post-monsoon. The surface wind pattern in Fig. 12 for October shows northeasterly flow towards the Indian subcontinent, transporting O3 from EA to SA. 3.2. Change in surface O3 contributions in the past decade (2000–2010) In this section we discuss the contribution of foreign emissions to the local surface O3 due to actual emission changes between 2000 and 2010. We consider NOx emission changes over these regions from the MACCity emission database (Granier et al., 2011). NOx emissions show an increase of 34% over SA and 51% increase over EA. The seasonal variation of the regional average and population weighted changes in surface O3 over SA and EA regions to the 2000–2010 anthropogenic emission increases over the other region is shown in Fig. 13. These surface O3 changes account for the nonlinearity of surface O3 changes due to NOx emission change, following the method described in Wild et al. (2012). Fig. 13 shows an increase EA contribution over SA compared with the SA contribution to EA in the past decade. In the post-monsoon months the effect of population weighted EA contribution to SA is twice as large of the SA to EA contribution. It is clear that over the past decade emission changes over East Asia had a greater impact on population exposure to O3 over Southern Asia than vice versa. 4. Conclusions The source–receptor relationship between EA and SA and its seasonal variation have been studied using the results from 14 models
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
37
Fig. 12. (a) The wind patterns at 925 hPa (b) and mean sea level pressure for May, July and October.
contributing to the HTAP model inter-comparison. Comparison between model simulations, satellite observations and surface observations shows some discrepancy in O3 concentration and seasonal variation of surface O3. Possible reasons for the discrepancies are error in satellite measurement, coarse model resolution, which alters chemical timescales and prevents like-for-like comparison with observations, and high temperatures and humidity shorten chemical timescales. Hence biases in O3 formation are larger here than elsewhere. Satellite observations of total tropospheric O3 column over SA show much better agreement, but are generally biased low by 5–10 DU, except in May. Over EA model and satellite derived tropospheric O3 columns are in most seasons in good agreement, except in May, June and July when the model simulations show low values. Surface observations over four Indian stations show much lower values of O3 compared to the model simulations but the TTCO obtained from satellite is relatively consistent with that from the model simulations. The discrepancy in surface O3 may lead to uncertainty in the attribution of foreign contributions to surface O3 over this region, but there is no reason to believe that the foreign fraction is the cause of discrepancies. The transport between eastern and southern Asia is controlled by regional meteorological conditions. In the pre-monsoon O3 contributions from SA to EA are greater
than EA contributions to SA. The surface winds in this season are mainly southwesterly which helps to transport the air pollutants towards EA. During the monsoon season the southwesterly wind strengthens but O3 production over the Indian subcontinent is less due to cloudy conditions. Therefore the SA contribution to EA is not as high as it is earlier in the year. However, the wind north of 30°N is northeasterly and it transports EA pollutants to SA. Hence during the monsoon season the contribution of EA and SA sources to each other is almost equal. In the postmonsoon months there is an increase in the EA contribution to SA. The surface northeasterly wind helps to transport pollutants from EA to SA. In all the above cases EA emissions affect the most populated parts of SA but the effect of SA emissions over densely populated regions of EA is not uniform. Recent NOx emission changes over these regions show substantial increases in the contribution of EA sources to surface O3 over SA. In view of this growing contribution of long-range transport to local pollution, it is imperative to understand all factors involved better. An expansion of high quality ozone measurements at strategic locations, not directly influenced by local pollution, is an important aspect. Continuous and systematic work on Asian emission inventories is needed as well. The coupling of regional models driven by boundary conditions from global models, as is planned in HTAP Phase 2, will
38
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
Fig. 13. Area averaged increases in surface O3 over SA and EA between 2000 and 2010; (a) population weighted and (b) regional average. Error bars show the spatial standard deviation over each region.
provide higher quality evaluations for long range impacts on local air quality.
Acknowledgments Authors acknowledge Indian Institute of Tropical Meteorology (MOES/16/21/12-IITM (MAQ)) for providing research facilities and support, and thank the international research groups who contributed modeling results for the HTAP model inter-comparison. FD acknowledges the FP7 projects PEGASOS and ECLAIRE.
References Akimoto, H., 2003. Global air quality and pollution. Science 302, 1716–1719. http://dx.doi. org/10.1126/science.1092666. Andreae, M.O., Berresheim, H., Andreae, T.W., Kritz, M.A., Bates, T.S., Merrill, J.T., 1988. Vertical distribution of dimethylsulfide, sulfur dioxide, aerosol ions, and radon over the Northeast Pacific Ocean. J. Atmos. Chem. 6, 149–173. Beig, G., Chate, D.M., Ghude, S.D., Mahajan, A.S., Srinivas, R., Ali, K., Sahu, S.K., Parkhi, N., Surendran, D., Trimbake, H.R., 2013. Quantifying the effect of air quality control measures during the 2010 Commonwealth Games at Delhi, India. Atmos. Environ. 80, 455–463. http://dx.doi.org/10.1016/j.atmosenv.2013.08.012. Bethan, S., Vaughan, G., Reid, S.J., 1996. A comparison of ozone and thermal tropopause heights and the impact of tropopause definition on quantifying the ozone content of the troposphere. Q. J. R. Meteorol. Soc. 122, 929–944. http://dx.doi.org/10.1002/ qj.49712253207. Bowman, K.W., 2002. Capturing time and vertical variability of tropospheric ozone: a study using TES nadir retrievals. J. Geophys. Res. 107, 4723. http://dx.doi.org/10. 1029/2002JD002150. Carmichael, G.R., 2003. Regional-scale chemical transport modeling in support of the analysis of observations obtained during the TRACE-P experiment. J. Geophys. Res. 108, 8823. http://dx.doi.org/10.1029/2002JD003117. Cooper, O.R., 2004. A case study of transpacific warm conveyor belt transport: influence of merging airstreams on trace gas import to North America. J. Geophys. Res. 109, D23S08. http://dx.doi.org/10.1029/2003JD003624. Derwent, R., Collins, W., Johnson, C., Stevenson, D., 2001. Transient behaviour of tropospheric ozone precursors in a global 3-D CTM and their indirect greenhouse effects. Clim. Chang. 49, 463–487. Dietz, T., Rosa, E.A., 1997. Effects of population and affluence on CO2 emissions. Proc. Natl. Acad. Sci. U. S. A. 94, 175–179. Duce, R., Unni, C., Ray, B., Prospero, J., Merrill, J., 1980. Long-range atmospheric transport of soil dust from Asia to the tropical North Pacific: temporal variability. Science 209 (80), 1522–1524. Engardt, M., Siniarovina, U., Khairul, N.I., Leong, C.P., 2005. Country to country transport of anthropogenic sulphur in Southeast Asia. Atmos. Environ. 39, 5137–5148. http://dx. doi.org/10.1016/j.atmosenv.2005.05.007.
Fiedler, V., Nau, R., Ludmann, S., 2009. East Asian SO2 pollution plume over Europe—part 1: airborne trace gas measurements and source identification by particle dispersion model simulations. Atmos. Chem. Phys. 9, 4717–4728. Fiore, A.A.M., Jacob, D.D.J., Bey, I., Yantosca, R.M., Field, B.D., Fusco, A.C., Wilkinson, J.G., 2002. Background ozone over the United States in summer: origin, trend, and contribution to pollution episodes. J. Geophys. 107, 4275. Fiore, A.M., West, J.J., Horowitz, L.W., Naik, V., Schwarzkopf, M.D., 2008. Characterizing the tropospheric ozone response to methane emission controls and the benefits to climate and air quality. J. Geophys. Res. 113, D08307. http://dx.doi.org/10.1029/ 2007JD009162. Fiore, A.M., Dentener, F.J., Wild, O., Cuvelier, C., Schultz, M.G., Hess, P., Textor, C., Schulz, M., Doherty, R.M., Horowitz, L.W., MacKenzie, I.A., Sanderson, M.G., Shindell, D.T., Stevenson, D.S., Szopa, S., Van Dingenen, R., Zeng, G., Atherton, C., Bergmann, D., Bey, I., Carmichael, G., Collins, W.J., Duncan, B.N., Faluvegi, G., Folberth, G., Gauss, M., Gong, S., Hauglustaine, D., Holloway, T., Isaksen, I.S.A., Jacob, D.J., Jonson, J.E., Kaminski, J.W., Keating, T.J., Lupu, A., Marmer, E., Montanaro, V., Park, R.J., Pitari, G., Pringle, K.J., Pyle, J.A., Schroeder, S., Vivanco, M.G., Wind, P., Wojcik, G., Wu, S., Zuber, A., Derwent, R.G., Johnson, C.E., 2009. Multimodel estimates of intercontinental source–receptor relationships for ozone pollution. J. Geophys. Res. 114, D04301. http://dx.doi.org/10.1029/2008JD010816. Granier, C., Bessagnet, B., Bond, T., D'Angiola, A., Denier van der Gon, H., Frost, G.J., Heil, A., Kaiser, J.W., Kinne, S., Klimont, Z., Kloster, S., Lamarque, J.-F., Liousse, C., Masui, T., Meleux, F., Mieville, A., Ohara, T., Raut, J.-C., Riahi, K., Schultz, M.G., Smith, S.J., Thompson, A., van Aardenne, J., van der Werf, G.R., van Vuuren, D.P., 2011. Evolution of anthropogenic and biomass burning emissions of air pollutants at global and regional scales during the 1980–2010 period. Clim. Chang. 109, 163–190. http://dx.doi.org/10.1007/s10584-011-0154-1. Hemispheric Transport of Air Pollution 2007. In: Keating, T.J., Zuber, A. (Eds.), Air Pollut. Stud. 16. UNECE, Geneva, Switzerland available at: http://www.htap.org/, 2007. Hemispheric Transport of Air Pollution 2010. In: Dentener, F., Keating, T., Akimoto, H. (Eds.), Air Pollut. Stud. 17. UNECE, Geneva, Switzerland available at: http:// www.htap.org/, 2010. Hilboll, A., Richter, A., Rozanov, A., Hodnebrog, Ø., Heckel, A., Solberg, S., Stordal, F., Burrows, J.P., 2013. Improvements to the retrieval of tropospheric NO2 from satellite — stratospheric correction using SCIAMACHY limb/nadir matching and comparison to Oslo CTM2 simulations. Atmos. Meas. Tech. 6, 565–584. http://dx.doi.org/10. 5194/amt-6-565-2013. Jena, C., Ghude, S.D., Pfister, G.G., Chate, D.M., Kumar, R., Beig, G., Surendran, D.E., Fadnavis, S., Lal, D.M., 2015. Influence of springtime biomass burning in South Asia on regional ozone (O3): a model based case study. Atmos. Environ. 100, 37–47. Ehrlich, P.R., Holdren, J.P., 1971. Impact of population growth. Science 171, 1212–1217. Lee, H.-J., Kim, S.-W., Brioude, J., Cooper, O.R., Frost, G.J., Kim, C.-H., Park, R.J., Trainer, M., Woo, J.-H., 2014. Transport of NOx in East Asia identified by satellite and in situ measurements and Lagrangian particle dispersion model simulations. J. Geophys. Res. 1–23 http://dx.doi.org/10.1002/2013JD021185.Received. Liu, H., 2003. Transport pathways for Asian pollution outflow over the Pacific: interannual and seasonal variations. J. Geophys. Res. 108, 8786. http://dx.doi.org/10.1029/ 2002JD003102. Ma, J., Lin, W.L., Zheng, X.D., Xu, X.B., Li, Z., Yang, L.L., 2014. Influence of air mass downward transport on the variability of surface ozone at Xianggelila Regional Atmosphere Background Station, southwest China. Atmos. Chem. Phys. 14, 5311–5325. http://dx.doi.org/10.5194/acp-14-5311-2014.
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39 Moore, G.W.K., Semple, J.L., 2005. A Tibetan Taylor Cap and a halo of stratospheric ozone over the Himalaya. Geophys. Res. Lett. 32, L21810. http://dx.doi.org/10.1029/ 2005GL024186. Nassar, R., Logan, J.A., Worden, H.M., Megretskaia, I.A., Bowman, K.W., Osterman, G.B., Thompson, A.M., Tarasick, D.W., Austin, S., Claude, H., Dubey, M.K., Hocking, W.K., Johnson, B.J., Joseph, E., Merrill, J., Morris, G.A., Newchurch, M., Oltmans, S.J., Posny, F., Schmidlin, F.J., Vömel, H., Whiteman, D.N., Witte, J.C., 2008. Validation of Tropospheric Emission Spectrometer (TES) nadir ozone profiles using ozonesonde measurements. J. Geophys. Res. 113, D15S17. http://dx.doi.org/10.1029/2007JD008819. Osterman, G.B., Kulawik, S.S., Worden, H.M., Richards, N.A.D., Fisher, B.M., Eldering, A., Shephard, M.W., Froidevaux, L., Labow, G., Luo, M., Herman, R.L., Bowman, K.W., Thompson, A.M., 2008. Validation of Tropospheric Emission Spectrometer (TES) measurements of the total, stratospheric, and tropospheric column abundance of ozone. J. Geophys. Res. 113, D15S16. http://dx.doi.org/10.1029/2007JD008801. Parrish, D.D., Hahn, C.J., Williams, E.J., Norton, R.B., F.C.F., 1992. Indications of photochemical histories of Pacific air masses from measurements of atmospheric trace species at Point Arena, California. J. Geophys. Res. 97, 15,883–15,901. Prospero, J.M., 1979. Mineral and sea salt aerosol concentrations in various ocean regions. J. Geophys. Res. 84, 725. http://dx.doi.org/10.1029/JC084iC02p00725. Rahimi, S., Gholami Sefidkouhi, M.A., Raeini-Sarjaz, M., Valipour, M., 2014. Estimation of actual evapotranspiration by using MODIS images (a case study: Tajan catchment). Arch. Agron. Soil Sci. 61, 695–709. http://dx.doi.org/10.1080/03650340.2014.944904. Reidmiller, D.R., Fiore, A.M., Jaffe, D.A., Bergmann, D., Cuvelier, C., Dentener, F.J., Duncan, B.N., Folberth, G., Gauss, M., Gong, S., Hess, P., Jonson, J.E., Keating, T., Lupu, A., Marmer, E., Park, R., Schultz, M.G., Shindell, D.T., Szopa, S., Vivanco, M.G., Wild, O., Zuber, A., 2009. The influence of foreign vs. North American emissions on surface ozone in the US. Atmos. Chem. Phys. 9, 5027–5042. Richter, A., Burrows, J.P., Nüss, H., Granier, C., Niemeier, U., 2005. Increase in tropospheric nitrogen dioxide over China observed from space. Nature 437, 129–132. http://dx. doi.org/10.1038/nature04092. Sheel, V., Lal, S., Richter, A., Burrows, J.P., 2010. Comparison of satellite observed tropospheric NO2 over India with model simulations. Atmos. Environ. 44, 3314–3321. http://dx.doi.org/10.1016/j.atmosenv.2010.05.043. Singh, H., Brune, W., 2009. Chemistry and transport of pollution over the Gulf of Mexico and the Pacific: spring 2006 INTEX-B campaign overview and first results. Atmos. Chem. Phys. 9, 2301–2318. Solazzo, E., Bianconi, R., Vautard, R., Appel, K.W., Moran, M.D., Hogrefe, C., Bessagnet, B., Brandt, J., Christensen, J.H., Chemel, C., Coll, I., Denier van der Gon, H., Ferreira, J., Forkel, R., Francis, X.V., Grell, G., Grossi, P., Hansen, A.B., Jeričević, A., Kraljević, L., Miranda, A.I., Nopmongcol, U., Pirovano, G., Prank, M., Riccio, A., Sartelet, K.N., Schaap, M., Silver, J.D., Sokhi, R.S., Vira, J., Werhahn, J., Wolke, R., Yarwood, G., Zhang, J., Rao, S.T., Galmarini, S., 2012. Model evaluation and ensemble modelling of surface-level ozone in Europe and North America in the context of AQMEII. Atmos. Environ. 53, 60–74. http://dx.doi.org/10.1016/j.atmosenv.2012.01.003. Stegmann, P.M., 1999. Global distribution of aerosols over the open ocean as. Global Biogeochem. Cycles 13, 383–397. Stohl, A., Forster, C., Huntrieser, H., Mannstein, H., McMillan, W.W., Petzold, A., Schlager, H., Weinzierl, B., 2007. Aircraft measurements over Europe of an air pollution plume from Southeast Asia — aerosol and chemical characterization. Atmos. Chem. Phys. 7, 913–937. http://dx.doi.org/10.5194/acp-7-913-2007. Tu, J., Xia, Z.-G., Wang, H., Li, W., 2007. Temporal variations in surface ozone and its precursors and meteorological effects at an urban site in China. Atmos. Res. 85, 310–337. http://dx.doi.org/10.1016/j.atmosres.2007.02.003.
39
Van Dingenen, R., Dentener, F.J., Raes, F., Krol, M.C., Emberson, L., Cofala, J., 2009. The global impact of ozone on agricultural crop yields under current and future air quality legislation. Atmos. Environ. 43, 604–618. http://dx.doi.org/10.1016/j.atmosenv.2008.10. 033. Wang, T., Wei, X.L., Ding, A.J., Poon, C.N., Lam, K.S., Li, Y.S., Chan, L.Y., Anson, M., 2009. Increasing surface ozone concentrations in the background atmosphere of Southern China, 1994–2007. Atmos. Chem. Phys. 9, 6217–6227. Wild, O., Akimoto, H., 2001. Intercontinental transport of ozone and its precursors in a three-dimensional global CTM. J. Geophys. Res. 106, 27729. http://dx.doi.org/10. 1029/2000JD000123. Wild, O., Fiore, A.M., Shindell, D.T., Doherty, R.M., Collins, W.J., Dentener, F.J., Schultz, M.G., Gong, S., MacKenzie, I.A., Zeng, G., Hess, P., Duncan, B.N., Bergmann, D.J., Szopa, S., Jonson, J.E., Keating, T.J., Zuber, A., 2012. Modelling future changes in surface ozone: a parameterized approach. Atmos. Chem. Phys. 12, 2037–2054. http://dx.doi.org/10. 5194/acp-12-2037-2012. Worden, J., 2004. Predicted errors of tropospheric emission spectrometer nadir retrievals from spectral window selection. J. Geophys. Res. 109, D09308. http://dx.doi.org/10. 1029/2004JD004522. Worden, H.M., Logan, J.A., Worden, J.R., Beer, R., Bowman, K., Clough, S.A., Eldering, A., Fisher, B.M., Gunson, M.R., Herman, R.L., Kulawik, S.S., Lampel, M.C., Luo, M., Megretskaia, I.A., Osterman, G.B., Shephard, M.W., 2007. Comparisons of Tropospheric Emission Spectrometer (TES) ozone profiles to ozonesondes: methods and initial results. J. Geophys. Res. 112, D03309. http://dx.doi.org/10.1029/2006JD007258. Wu, P., Shen, H., Zhang, L., Göttsche, F., 2015. Remote sensing of environment integrated fusion of multi-scale polar-orbiting and geostationary satellite observations for the mapping of high spatial and temporal resolution land surface temperature. 156, 169–181. Yamaji, K., Ohara, T., Uno, I., Tanimoto, H., Kurokawa, J., Akimoto, H., 2006. Analysis of the seasonal variation of ozone in the boundary layer in East Asia using the Community Multi-scale Air Quality model: what controls surface ozone levels over Japan? Atmos. Environ. 40, 1856–1868. http://dx.doi.org/10.1016/j.atmosenv.2005.10.067. Yienger, J.J., Galanter, M., Holloway, T.A., Phadnis, M.J., Guttikunda, S.K., Carmichael, G.R., Moxim, W.J., Levy II, H., 2000. The episodic nature of air pollution transport from Asia to North America. J. Geophys. Res. 105, 26,931–26,945. Yoshitomi, M., Wild, O., Akimoto, H., 2011. Contributions of regional and intercontinental transport to surface ozone in the Tokyo area. Atmos. Chem. Phys. 11, 7583–7599. http://dx.doi.org/10.5194/acp-11-7583-2011. Zhang, Q., Streets, D.G., He, K., Wang, Y., Richter, A., Burrows, J.P., Uno, I., Jang, C.J., Chen, D., Yao, Z., Lei, Y., 2007. NOx emission trends for China, 1995–2004: the view from the ground and the view from space. J. Geophys. Res. 112, D22306. http://dx.doi.org/10. 1029/2007JD008684. Zhang, L., Jacob, D.J., Boersma, K.F., Jaffe, D.A., Olson, J.R., Bowman, K.W., Worden, J.R., Thompson, A.M., Avery, M.A., Cohen, R.C., Dibb, J.E., Flocke, F.M., Fuelberg, H.E., Huey, L.G., McMillan, W.W., Singh, H.B., Weinheimer, A.J., 2008. Transpacific transport of ozone pollution and the effect of recent Asian emission increases on air quality in North America: an integrated analysis using satellite, aircraft, ozonesonde, and surface observations. Atmos. Chem. Phys. Discuss. 8, 8143–8191. http://dx.doi.org/10. 5194/acpd-8-8143-2008.
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Atmospheric transport of ozone between Southern and Eastern Asia T. Chakraborty a, G. Beig a,⁎, F.J. Dentener b, O. Wild c a b c
Indian Institute of Tropical Meteorology, Pune, India European Commission, Joint Research Centre, Institute for Environment and Sustainability, Ispra, Italy Lancaster Environment Centre, Lancaster University, Lancaster, UK
H I G H L I G H T S • • • •
Maximum effect of East Asian pollution over South Asia happens in post-monsoon. Maximum effect of South Asian pollution over East Asia occurs in pre-monsoon. Most densely populated parts of South Asia are affected by East Asian emission. South Asia is largely affected by the East Asian emission change from 2000 to 2010.
a r t i c l e
i n f o
Article history: Received 1 December 2014 Received in revised form 15 March 2015 Accepted 17 March 2015 Available online 7 April 2015 Editor: Xuexi Tie Keywords: Pollution transport HTAP Ozone East Asia South Asia
a b s t r a c t This study describes the effect of pollution transport between East Asia and South Asia on tropospheric ozone (O3) using model results from the Task Force on Hemispheric Transport of Air Pollution (TF HTAP). Ensemble mean O3 concentrations are evaluated against satellite-data and ground observations of surface O3 at four stations in India. Although modeled surface O3 concentrations are 1020 ppb higher than those observed, the relative magnitude of the seasonal cycle of O3 is reproduced well. Using 20% reductions in regional anthropogenic emissions, we quantify the seasonal variations in pollution transport between East Asia and South Asia. While there is only a difference of 0.05 to 0.1 ppb in the magnitudes of the regional contributions from one region to the other, O3 from East Asian sources affects the most densely populated parts of South Asia while Southern Asian sources only partly affect the populated parts of East Asia. We show that emission changes over East Asia between 2000 and 2010 had a larger impact on populated parts of South Asia than vice versa. This study will help inform future decisions on emission control policy over these regions. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Intercontinental transport of air pollution is a global issue as the air quality of a region can be highly influenced by long range transport from remote continents (HTAP, 2010). Pollutants with longer atmospheric life times are important when assessing the impacts of long range transport, but shorter lived air pollutants that produce secondary pollutants are also important. Tropospheric ozone (O3) is a secondary pollutant produced through a sequence of photochemical reactions involving nitrogen oxides (NOx), carbon monoxide (CO), volatile organic compounds (VOCs) and CH4. Due to the long lifetime of CH4, the O3 produced from its oxidation is largely independent of the location of the CH4 emissions (Fiore et al., 2008). The spatial pattern of the O3 formed is controlled by the distribution of OH and NOx, which have much shorter lifetimes and can affect both the location of CH4 oxidation and the amount of O3 production per CH4 molecule oxidized (Fiore et al., ⁎ Corresponding author. E-mail address: [email protected] (G. Beig).
http://dx.doi.org/10.1016/j.scitotenv.2015.03.066 0048-9697/© 2015 Elsevier B.V. All rights reserved.
2008). Increases in atmospheric CH4 contribute to all regions relatively uniformly, averaging 1.5–1.9 ppb O3 since 1960, and this contributes about one third of the O3 increase seen over Europe (EU) and North America (NA) over this period (Wild et al., 2012). From a series of future (2005–2030) transient simulations, it has been demonstrated that costeffective CH4 controls could offset the positive climate forcing from CH4 and O3 and improve air quality (Fiore et al., 2008). Measurements from remote regions consistently indicate that longrange transport exerts a strong influence on observed concentrations of aerosols, O3 and its important precursors (HTAP, 2007). For example, dust of Asian origin has been observed throughout the North Pacific region (Duce et al., 1980; Prospero, 1979). Studies over the west coast of NA identified the influence of Asian emissions on the sulfur budget (Andreae et al., 1988) and on the concentrations of O3, hydrocarbons and peroxyacetyl nitrate (Parrish et al., 1992). Several observation and model-based studies have shown the impact of foreign emissions on countries at Northern mid-latitudes. For example, EU and NA respectively contribute 3.5 ± 1.1 and 2.2 ± 0.5 ppb to the annual mean surface O3 over Japan (Yoshitomi et al., 2011). The annual mean magnitude of
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
NA and EU contributions to surface O3 over East China is 1.2 ppbv and 1.5 ppbv respectively. EU experiences the greatest intercontinental import of O3 due to rapid short-distance transport from NA (Wild and Akimoto, 2001). A widely used technique to study the contribution of upwind emissions to a region downwind is the calculation of source–receptor relationships, which evaluate the impact of a relatively small emission perturbation over a specific source region (Derwent et al., 2001; Fiore et al., 2002; Wild and Akimoto, 2001). Source–receptor relationships depend on various factors including the emission strength, size of the source region, the transport pathway, and the extent of pollutant transformation and loss during transport. This study quantify the source–receptor relationship for surface O3 between East Asia (EA, 15–55°N, 95–160°E) and South Asia (SA, 5°– 35°N, 50°–95°E) using the Task Force on Hemispheric Transport of Air Pollution (HTAP) Phase 1 source–receptor model experiments (www. htap.org). EA and SA are two major developing industrial regions that are also the most heavily populated regions in the world. Development in each of these regions not only affects air quality in the region itself, but also air quality in the other region through the transport of air pollutants from one region to the other. There is widespread scientific agreement that the observed increase in O3 concentrations is the consequence of human activities around the globe. Among anthropogenic factors, the principal one is increasing population (Dietz and Rosa, 1997). A population density map for the year 2000 (Fig. 1) shows that India, Central and Eastern China, Japan, Bangladesh and Korea are the most densely populated regions. Bangladesh and the Indo-Gangetic plain of India are the most densely populated parts of southern Asia. Model simulated surface O3 over this region is quite high, ranging from 35 to 55 ppbv (Fig. 2). Surface O3 over the most populated parts of EA (the coastal region of China, Japan, and Korea) ranges from 45 to 65 ppb. Anthropogenic emissions over both Southern and Eastern Asia have experienced a rapid increase in recent decades (Hilboll et al., 2013; Richter et al., 2005; Zhang et al., 2007). Model based study of biomass burning over SA shows that there is enhancement of surface ozone of about 4 to10 ppb (25–50%) in the Eastern region including Burma (during March and April), 1–3 ppb (10–25%) in the Central India (during
29
Fig. 2. Surface O3 concentration for the months of (a) May, (b) July and (c) October, in the ensemble average of 14 models.
Fig. 1. The population density in Asia (http://sedac.ciesin.columbia.edu).
30
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
March) and 1–7 ppb (4–10%) in the Indo-Gangetic region (Jena et al., 2015). Studies of background O3 concentrations over China for the period 1994 to 2007 suggest that surface O3 has been increasing at an average rate of 0.58 ppbv per year (Wang et al., 2009). This is associated with increased NOx emissions, which have been clearly observed in GOME and SCIAMACHY tropospheric NO2 satellite measurements (Lee et al., 2014; Wang et al., 2009) in China's fast-developing coastal regions. Inter-continental transport of air pollution also plays an important role in the seasonal variation of air quality over a region (Tu et al., 2007). The seasonal outflow of pollutants from EA is maximum in spring as observed both by in situ observation and satellite studies (Duce et al., 1980; Stegmann, 1999). The surface O3 distribution over EA varies dynamically from season to season according to the meteorological conditions and chemical production of O3 from precursor emissions over the region. Yamaji et al. (2006) showed that only 10% (winter) to 54% (summer) O3 concentration over EA could be explained by local emissions. Thus the concentration of O3 in air masses over EA is strongly affected by long-range transport of O3 or its precursors from outside EA, especially in winter and early spring. Global atmospheric modeling studies show that the increase in surface emission of NOx and VOCs over Asia leads to an increase in surface O3 concentrations. These increases are strongest over the polluted regions of northern India, China and Japan due to local sources in these regions (Bethan et al., 1996). The impact of Asian emissions on the global environment has been examined through observational and model studies focusing on key export paths like transpacific transport (Carmichael, 2003; Cooper, 2004; Liu, 2003), seasonal and episodic variability (Stohl et al., 2007; Yienger et al., 2000), and impacts over North America (Fiore et al., 2002; Reidmiller et al., 2009; Singh and Brune, 2009; Zhang et al., 2008) and Europe (Fiedler et al., 2009; Stohl et al., 2007). Future projections based on Representative Concentration Pathways (RCPs) generated for the Climate Model Inter comparison Project (CMIP5) simulations show a large range in emission changes over SA between the different scenarios, and consequently large differences in surface O3 are expected (Wild et al., 2012). Surface O3 over SA under RCP 8.5 shows an increase of more than 5 ppb by 2050 but for RCP 6.0 the changes are close to zero. There is also an increase in surface O3 over EA until 2020 but a decrease for the RCP 2.6 and 4.5 scenarios. This suggests that there is substantial uncertainty in future air quality over the Asian region, but in many cases there is an increase in surface O3. It is therefore important to study how emissions from EA and SA affect air quality over the other region. This will help future science and policy interactions, as well as decision making on emission control policy over these regions. Long range transport of pollutants within Asia has not been studied in detail, with regard to these policy aspects. The seasonality in surface O3 over SA mainly reflects the influence of the Asian Summer Monsoon, hence is quite different from mid-latitude regions. In all seasons, the influence of SA emissions over mid-latitude regions such as NA is much less than that from EA and EU sources (Reidmiller et al., 2009). The reason for this relatively small long-range transport is the meteorological conditions over SA, which combine low, near-surface wind during most of the year with wet scavenging of pollutants due to the abundance of rainfall in the tropical environment (Engardt et al., 2005). The major focus of this study is to quantify the surface O3 concentration change over SA due to changes in emission over the EA region and viceversa.
anthropogenic emission perturbations. In December 2004, the executive body of the UNECE convention on Long Range Trans-boundary Air Pollution (LRTAP) established the Task Force on Hemispheric Transport of Air Pollution (HTAP). During its first phase (2006–2010), the task force coordinated a multi-model assessment to quantify the importance of intercontinental transport of O3 and its precursors and particulate matter. These studies concluded that the O3 response to emission changes in upwind regions is not negligible and that it is particularly strong in spring and fall at northern mid-latitudes. This study uses results from fourteen models contributing to the HTAP intercomparison: ECHAM5, EMEP, FRSGC/UCI, GEMAQ, GEOS-Chem, GISS, LLNL, MOZART, MOZECH, STOC-HadAM3, STOCHEM, TM5, ULAQ, and UM-CAM. The models are driven by meteorological fields for 2001. An average over these 14 models after regriding the results to a common 2.5° × 2.5° resolution is used for the study. Results are for the model's lowest grid box, and since the relevant meteorological information is not available, no attempt is made to extrapolate to near-surface levels corresponding to measurement stations; the analysis considers a base simulation and a perturbation scenario. In the perturbation scenario all anthropogenic emissions including NOx, CO, VOC, SO2 and PM components are reduced by 20% over either SA or EA source regions (identified by HTAP label SR6). The annual mean surface O3 concentration in the base simulation and the standard deviation (relative to the ensemble mean) over the two source regions are shown in Fig. 3. The standard deviation is relatively uniform over the study region, which suggests that the models agree well in the spatial pattern but there is a relatively large absolute difference of 14–20 ppbv between models. The mean annual surface O3 over SA and EA varies from 40–55 ppbv, with high values over the
2. Model evaluation with satellite data and surface observations 2.1. Model simulation This paper addresses the seasonal variation of the source receptor relationship between SA and EA by evaluating the effect of large-scale
Fig. 3. (a) Annual mean surface O3 concentration and (b) the relative standard deviation of 14 models.
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
Himalayan region where the surface altitude is high and where intrusions of ozone-rich stratospheric air may also contribute to high surface ozone (Ma et al., 2014; Moore and Semple, 2005).
31
obtained from Surface Radiation Budget Network (SURFRAD) (Wu et al., 2015). 2.3. Surface observations
2.2. Satellite data The Tropospheric Emission Spectrometer (TES), the third of NASA's Earth observing system (EOS) satellite instruments, was launched into sun-synchronous orbit on 15 July 2004. TES is an infrared, high spectral resolution Fourier Transform Spectrometer (FTS). TES nadir-view has a vertical resolution of 6–7 km in cloudless conditions (Bowman, 2002; Worden et al., 2007; Worden, 2004). TES measurements have a positive bias in O3 relative to ozonesondes (Nassar et al., 2008). TES level 3 total tropospheric column ozone (TTCO) monthly mean data at 4°lon × 2°lat resolution for the period 2005 to 2009 have been used to evaluate the model simulations. It is important to identify the proper tropopause height to avoid stratospheric air in the TTCO calculation, because the tropopause height changes with latitude and season. For this calculation, an ozone-based tropopause is defined where O3 mixing ratio exceeds 110 ppbv (Bethan et al., 1996) and consequently O3 concentration in the troposphere–stratosphere region exceeding 110 ppbv is assumed to be in the stratosphere. Many recent studies demonstrate the use of satellite observations in various applications related to terrestrial environment. For example, the result obtained from the MODIS satellite images to estimate daily water requirement using SEBAL algorithm over Tajan catchment is very satisfactory (Rahimi et al., 2014). Land surface temperature obtained from spatio-temporal integrated temperature fusion model (STITFM), which is used to retrieve high temporal and spatial resolution data from multi-scale polar-orbiting and geostationary satellite observation shows an accuracy of 2.5 K with the in-situ land surface temperature
The performance of the models has been validated over four stations in India: Pune (18.5°N, 73.85°E), Udaipur (17.38°N, 78.48°E), Hyderabad (24.5°N, 73.69°E) and Jabalpur (23.16°N, 79.94°E), shown in Fig. 4. Pune is located at 560 m above sea level on the Deccan plateau in western India. Udaipur has an average elevation of 598 m and is located in the southern part of Rajasthan. Hyderabad is located at an average altitude of 542 m, on hilly terrain in Southern India. Udaipur is located in central India at an average elevation of 412 m. All stations are situated in arid and semi-arid regions. All the monitoring stations are designated as semi-urban indicating that stations are away from downtown areas where the influence of local emissions may be high. However, local effects cannot be ruled out under some conditions. Observations over these stations were made with Air Quality Management System (AQMS). Each AQMS comprised of US Environmental Protection Agency approved analyzers housed inside walkway shelters. Ozone was measured with a photometric UV analyzer (Thermo-49i, precision 1 ppbv). Calibration of the O3 analyzer was done on every alternate day using an inbuilt O3 calibrator. The inlets for O3 measurements were located about 3 m above the surface at all of the stations (Beig et al., 2013). 2.4. Comparison between model and surface observations The four stations considered show minimum surface O3 values in the monsoon period and maximum values in the pre-monsoon. The seasonal variation of surface O3 over the four stations is shown in Fig. 5, based
(c) (b) (a) (d)
Fig. 4. Geographical locations of four stations: (a) Pune, (b) Jabalpur, (c) Udaipur and (d) Hyderabad.
32
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
Fig. 5. Comparison between the model ensemble monthly mean O3 and surface observations over (a) Hyderabad, (b) Pune, (c) Jabalpur and (d) Udaipur.
on monthly averages of observations taken each hour and ensemble mean surface O3 from the models over the corresponding station location. The model simulations reproduce the seasonal cycle of surface O3 concentration over the locations, but are substantially higher. The ground-level ozone mixing ratios from the models have been examined collectively as an ensemble and evaluated against a large set of observations from both Europe and East Asia (Fiore et al., 2009). This commonly used multi-model approach can be outperformed by subsets of models optimally selected in terms of bias, error, and correlation, but it does not strictly depend on the skill of individual models, although it may require the inclusion of low-ranking skill-score members (Solazzo et al., 2012). The difference between the models and surface observations may be due to the fact that the global models are not able to capture fine-scale emission patterns at the coarse resolutions used here. A further reason may be that the relatively low sampling height (3 m) does not match the altitude of the middle of each model's grid box. An earlier analysis by Van Dingenen et al. (2009), comparing 30 meter and 10 meter concentrations, showed that in India this may lead to overestimates of the order of 20–30 ppb, and this bias may be even higher when comparing with 3 m. Another reason for this difference may be that over India NOx, one of the most important precursors of O3, is strongly affected by local emissions. Biomass and fossil fuel burning is the main source of NOx over this region. Due to the increasing population and higher economic growth rates, emission of these gases is increasing over SA and EA (Akimoto, 2003; Sheel et al., 2010). Surface observations of NOx over these four stations are much higher (5–17 ppbv) than in the model simulations (0.1–2 ppbv). This suggests that destruction of O3 by direct reaction with emitted NO is much more important than seen in the models, accounting for the overestimation of surface O3 in the models.
2.5. Comparison between model and satellite observation at location of the four stations The seasonal variation of ensemble-mean TTCO over station locations (generated by interpolating the gridded data to the exact station locations) is compared with satellite data in Fig. 6. The model results are consistent with the satellite observations but show systematically higher O3 columns (35–45 DU) in monsoon months (June–September) relative to the satellite observations (27– 40 DU) in these months, whereas in the dry months the bias is much smaller (1–5 DU). The difference in O3 column between satellite observations and model simulations may be due to the fact that in monsoon months the thick clouds in the field of view obscure the infrared emission from the lower troposphere, which greatly reduces TES sensitivity (Nassar et al., 2008). There is also missing satellite data over some parts of Asia in monsoon months. Comparison of ozonesonde data from 32 ground stations around the world suggests that TES overestimates the O3 column by 4 DU (Osterman et al., 2008). An inter-comparison study between satellite TES and OMI (O3 monitoring instrument) shows that monthly mean differences between OMI and TES exceed ±10 DU. The bias and standard deviation which are determined using correlations of mean values of TES O3 and sonde O3 range from 2.9 to 10.6 ppbv in the upper troposphere and 3.7 to 9.2 ppbv in the lower troposphere excluding the Arctic and Antarctic where TES sensitivity is low (Nassar et al., 2008).
2.6. Comparison between model and satellite observation over EA and SA The area averaged TTCO from satellite and model ensemble-mean for SA and EA is shown in Fig. 7.
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
33
Fig. 6. Seasonal variation of the total tropospheric O3 column from model simulations and satellite observation over Hyderabad, Pune, Jabalpur and Udaipur.
The seasonal cycle of TTCO generated by the models over both SA and EA is comparable with satellite observations, but the O3 concentration is much greater in the models over SA, except in the months of April and May. Over EA the satellite shows slightly higher O3 in May, June and
July. The spatial pattern of the difference between model simulated and satellite observed TTCO over EA and SA for May, July and October is shown in Fig. 8. In May the models show lower O3 than the satellite over most parts of EA and SA by up to 3 DU. The models show higher
Fig. 7. Area averaged total tropospheric O3 column from model simulations and satellite observation over (a) SA and (b) EA. Black and green lines show the total tropospheric O3 column from model simulations and satellite observations respectively.
34
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39 Table 1 Dominating climatological conditions in Northern and Southern India during, pre-monsoon, monsoon, and post-monsoon. Pre-monsoon (March to May)
Monsoon (June to September)
Post-monsoon (October to December)
High temperatures over north and central India Coastal region dominated by land/sea breeze In May wind changes from NE to SW.
Heavy rain on west coast and northern India Cloudy conditions and washout of soluble precursors by rain SW wind brings cleaner air from Arabian Sea and Indian Ocean.
Wind changes from SW to NE in October SE India gets rainfall. Temperature falls sharply; decrease in humidity; clear sky over N and central India. Photochemical O3 production increases. NE winds bring polluted continental air from Ganges plains over India
occurs over SA in the monsoon season and the highest responses occur in the post-monsoon. The maximum effect of SA emissions over EA occurs in the pre-monsoon months. The contribution of SA to EA ranges from 0.1 to 0.3 ppbv with a peak in April. The EA contribution to SA ranges from 0.15 to 0.35 ppbv, increasing from June and reaching a maximum in November. The spatial standard deviation of the surface O3 response in SA due to EA emission is large in the post-monsoon months. This indicates that though the maximum response is in the post-monsoon, it is not uniform over the whole SA region. Over EA the spatial standard deviation of the response to SA emissions is greater in the pre-monsoon, indicating that the spread of SA pollutants over EA is also not uniform. It is clear that there is greater spatial variability in the effects of SA sources on EA than in the effects of EA sources on SA. The population weighted O3 change shows a similar seasonal variation in O3 contributions. Fig. 8. Difference between the model simulation and satellite observation of total tropospheric O3 column for the months of (a) May, (b) July and (c) October.
O3 than the satellite observations over most past of EA and SA for July and October of up to 6 DU. In all three months the models show much higher O3 over the Himalayan region. This may explain the higher O3 column over SA in the models than satellite observation in Fig. 7. The model columns over the four stations are more consistent with the satellite observations because over most part of SA the difference between the model simulation and the satellite observation ranges from 0 to 6 DU.
3.1. Spatial pattern of the O3 source–receptor relationship Figs. 10 & 11 illustrate the spatial patterns of O3 concentration change due to emission reductions over SA and EA, and the impact of emissions on the population for the months of May, July and October (characterizing the pre-monsoon, monsoon and post-monsoon periods, respectively) due to emission reductions over the other region. To estimate the impact of emission on the population density we have used a scheme (Ehrlich and Holdren, 1971) which can be expressed by the relation
3. Source–receptor relationships for surface O3
Population Impact ðPIÞ ¼ population density percentage change in ozone:
In how foreign emissions affect regional air quality over the SA and EA regions. The foreign influence on regional air quality is defined by analyzing the difference between the control simulations and those with 20% emission reductions over a given source region. The reduction in surface O3 over the receptor region shows how much emission from the source region influences it. To provide a more relevant assessment of the effects on human health, we also present the populationweighted O3 change over each of the receptor regions. The source contribution varies with season and is sensitive to the prevailing wind pattern. We describe the seasonal variation of this contribution in the context of the three main seasons over the Indian subcontinent: pre-monsoon, monsoon and post-monsoon. The prevailing circulation patterns which control the transport of pollutants change in direction over the seasons in this region, and a brief description is given in Table 1. The seasonal variation of the surface O3 response and population weighted changes in surface O3 over each region for a 20% anthropogenic emission reduction over the other region is shown in Fig. 9. The surface O3 response to foreign emission changes has a strong seasonal cycle over each region. The minimum response to EA emissions
3.1.1. Pre-monsoon (March to May) Surface O3 concentrations over both regions are quite high in this season (Fig. 2). The first panel in Fig. 10(a) shows that most parts of China and the adjacent oceanic region are affected by SA emissions but the most densely populated regions of eastern China, Japan, North and South Korea are not affected much. The impact of SA emissions on the populated parts of EA is not uniform but there are some parts of coastal China where the impact is high as shown in Fig. 10(b). The northeastern parts of India, Nepal, Bhutan, Bangladesh, Pakistan and Afghanistan are more strongly affected by emissions from EA. The most densely populated regions of the Indo Gangetic plain and Bangladesh are highly affected, but in contrast the southern parts of India and Sri Lanka are not influenced much. Fig. 11(b) shows the impact of EA emissions over the densely populated regions of the Indo Gangetic plain and Bangladesh. The area averaged contribution of SA sources to O3 over EA is 0.2 ppbv, and the EA contribution to SA is 0.15 ppbv. Although the effects of SA sources over EA are greater than those of EA over SA in this period, EA emissions affect the most densely populated parts of the SA region. The southwesterly wind pattern in May at 925 hPa transports pollutants from SA to EA.
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
35
Fig. 9. Regional average change in surface O3 over SA and EA due to a 20% emission change over the other region; (a) population weighted and (b) and area average. Error bars show the spatial standard deviation over each region.
Due to high pressure in the western Pacific between 140°E and 160°E shown in Fig. 12, SA sources do not affect the eastern parts of EA (Japan, North and South Korea).
3.1.2. Monsoon (June to September) We consider July as representative of this season and study the continental scale transport processes. During this season the surface O3
Fig. 10. (a) Percentage changes in surface O3 due to 20% emission reductions over SA and (b) the impact on population for May, July and October.
36
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
Fig. 11. Percentage changes in surface O3 due to 20% emission reductions over EA and (b) the impact on population for May, July and October.
concentration is less over EA than over SA (Fig. 2), and the spatial patterns of O3 changes over EA and SA due to emission reductions over the other region are shown in the second panel of Figs. 10 & 11. The area averaged surface O3 contributions over each region from the foreign source region in this month are almost equal, but the effect of SA sources over EA is less than during the Pre-Monsoon. From Fig. 11 we can see that the effect of O3 from EA sources is spread over almost all parts of SA. The spread of O3 from SA towards EA is mainly over the oceanic region and some parts of EA (China, and parts of Mongolia and Japan). The impact of SA emission over the densely populated regions of EA is less that in the pre-monsoon but the effect of EA emissions over Bangladesh and the Indo-Gangetic Plain is the same as in the premonsoon. The spatial standard deviation in the monsoon is higher over EA than over SA, indicating that the EA contribution to SA is more uniform than the SA contribution to EA. The distribution of surface O3 changes can be explained with the help of the dominating wind patterns (Fig. 12). Transport from SA is mainly governed by strong southwesterly winds in these months. Fig. 12(b) shows lower pressure over the Indian subcontinent, and the northeasterly winds north of 30°N transport O3 from EA to SA. The southwesterly winds mainly transport O3 over the oceanic region, and hence EA is not that much affected by O3 from SA. 3.1.3. Post-monsoon (October to December) Surface O3 concentrations over EA and SA in October show an increase in O3 production over SA (Fig. 2). The response of surface O3 concentrations over EA and SA due to 20% emission reductions over the other region is shown in the third panel of Figs. 10 & 11. The effect of SA emissions on EA is much less than in other months, and the EA contributions to SA are maximum in this season. Most parts of SA, especially the southeastern part of India and Sri Lanka, are strongly affected but
the northern parts of SA are less affected (Fig. 11). The relative change in O3 concentration over SA ranges from 1% to 0.2% spatially. The effect of SA emissions spread over the Tibetan plateau, Mongolia and oceanic regions. The effect of SA emission on densely populated parts of EA is less in the post-monsoon. The surface wind pattern in Fig. 12 for October shows northeasterly flow towards the Indian subcontinent, transporting O3 from EA to SA. 3.2. Change in surface O3 contributions in the past decade (2000–2010) In this section we discuss the contribution of foreign emissions to the local surface O3 due to actual emission changes between 2000 and 2010. We consider NOx emission changes over these regions from the MACCity emission database (Granier et al., 2011). NOx emissions show an increase of 34% over SA and 51% increase over EA. The seasonal variation of the regional average and population weighted changes in surface O3 over SA and EA regions to the 2000–2010 anthropogenic emission increases over the other region is shown in Fig. 13. These surface O3 changes account for the nonlinearity of surface O3 changes due to NOx emission change, following the method described in Wild et al. (2012). Fig. 13 shows an increase EA contribution over SA compared with the SA contribution to EA in the past decade. In the post-monsoon months the effect of population weighted EA contribution to SA is twice as large of the SA to EA contribution. It is clear that over the past decade emission changes over East Asia had a greater impact on population exposure to O3 over Southern Asia than vice versa. 4. Conclusions The source–receptor relationship between EA and SA and its seasonal variation have been studied using the results from 14 models
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
37
Fig. 12. (a) The wind patterns at 925 hPa (b) and mean sea level pressure for May, July and October.
contributing to the HTAP model inter-comparison. Comparison between model simulations, satellite observations and surface observations shows some discrepancy in O3 concentration and seasonal variation of surface O3. Possible reasons for the discrepancies are error in satellite measurement, coarse model resolution, which alters chemical timescales and prevents like-for-like comparison with observations, and high temperatures and humidity shorten chemical timescales. Hence biases in O3 formation are larger here than elsewhere. Satellite observations of total tropospheric O3 column over SA show much better agreement, but are generally biased low by 5–10 DU, except in May. Over EA model and satellite derived tropospheric O3 columns are in most seasons in good agreement, except in May, June and July when the model simulations show low values. Surface observations over four Indian stations show much lower values of O3 compared to the model simulations but the TTCO obtained from satellite is relatively consistent with that from the model simulations. The discrepancy in surface O3 may lead to uncertainty in the attribution of foreign contributions to surface O3 over this region, but there is no reason to believe that the foreign fraction is the cause of discrepancies. The transport between eastern and southern Asia is controlled by regional meteorological conditions. In the pre-monsoon O3 contributions from SA to EA are greater
than EA contributions to SA. The surface winds in this season are mainly southwesterly which helps to transport the air pollutants towards EA. During the monsoon season the southwesterly wind strengthens but O3 production over the Indian subcontinent is less due to cloudy conditions. Therefore the SA contribution to EA is not as high as it is earlier in the year. However, the wind north of 30°N is northeasterly and it transports EA pollutants to SA. Hence during the monsoon season the contribution of EA and SA sources to each other is almost equal. In the postmonsoon months there is an increase in the EA contribution to SA. The surface northeasterly wind helps to transport pollutants from EA to SA. In all the above cases EA emissions affect the most populated parts of SA but the effect of SA emissions over densely populated regions of EA is not uniform. Recent NOx emission changes over these regions show substantial increases in the contribution of EA sources to surface O3 over SA. In view of this growing contribution of long-range transport to local pollution, it is imperative to understand all factors involved better. An expansion of high quality ozone measurements at strategic locations, not directly influenced by local pollution, is an important aspect. Continuous and systematic work on Asian emission inventories is needed as well. The coupling of regional models driven by boundary conditions from global models, as is planned in HTAP Phase 2, will
38
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39
Fig. 13. Area averaged increases in surface O3 over SA and EA between 2000 and 2010; (a) population weighted and (b) regional average. Error bars show the spatial standard deviation over each region.
provide higher quality evaluations for long range impacts on local air quality.
Acknowledgments Authors acknowledge Indian Institute of Tropical Meteorology (MOES/16/21/12-IITM (MAQ)) for providing research facilities and support, and thank the international research groups who contributed modeling results for the HTAP model inter-comparison. FD acknowledges the FP7 projects PEGASOS and ECLAIRE.
References Akimoto, H., 2003. Global air quality and pollution. Science 302, 1716–1719. http://dx.doi. org/10.1126/science.1092666. Andreae, M.O., Berresheim, H., Andreae, T.W., Kritz, M.A., Bates, T.S., Merrill, J.T., 1988. Vertical distribution of dimethylsulfide, sulfur dioxide, aerosol ions, and radon over the Northeast Pacific Ocean. J. Atmos. Chem. 6, 149–173. Beig, G., Chate, D.M., Ghude, S.D., Mahajan, A.S., Srinivas, R., Ali, K., Sahu, S.K., Parkhi, N., Surendran, D., Trimbake, H.R., 2013. Quantifying the effect of air quality control measures during the 2010 Commonwealth Games at Delhi, India. Atmos. Environ. 80, 455–463. http://dx.doi.org/10.1016/j.atmosenv.2013.08.012. Bethan, S., Vaughan, G., Reid, S.J., 1996. A comparison of ozone and thermal tropopause heights and the impact of tropopause definition on quantifying the ozone content of the troposphere. Q. J. R. Meteorol. Soc. 122, 929–944. http://dx.doi.org/10.1002/ qj.49712253207. Bowman, K.W., 2002. Capturing time and vertical variability of tropospheric ozone: a study using TES nadir retrievals. J. Geophys. Res. 107, 4723. http://dx.doi.org/10. 1029/2002JD002150. Carmichael, G.R., 2003. Regional-scale chemical transport modeling in support of the analysis of observations obtained during the TRACE-P experiment. J. Geophys. Res. 108, 8823. http://dx.doi.org/10.1029/2002JD003117. Cooper, O.R., 2004. A case study of transpacific warm conveyor belt transport: influence of merging airstreams on trace gas import to North America. J. Geophys. Res. 109, D23S08. http://dx.doi.org/10.1029/2003JD003624. Derwent, R., Collins, W., Johnson, C., Stevenson, D., 2001. Transient behaviour of tropospheric ozone precursors in a global 3-D CTM and their indirect greenhouse effects. Clim. Chang. 49, 463–487. Dietz, T., Rosa, E.A., 1997. Effects of population and affluence on CO2 emissions. Proc. Natl. Acad. Sci. U. S. A. 94, 175–179. Duce, R., Unni, C., Ray, B., Prospero, J., Merrill, J., 1980. Long-range atmospheric transport of soil dust from Asia to the tropical North Pacific: temporal variability. Science 209 (80), 1522–1524. Engardt, M., Siniarovina, U., Khairul, N.I., Leong, C.P., 2005. Country to country transport of anthropogenic sulphur in Southeast Asia. Atmos. Environ. 39, 5137–5148. http://dx. doi.org/10.1016/j.atmosenv.2005.05.007.
Fiedler, V., Nau, R., Ludmann, S., 2009. East Asian SO2 pollution plume over Europe—part 1: airborne trace gas measurements and source identification by particle dispersion model simulations. Atmos. Chem. Phys. 9, 4717–4728. Fiore, A.A.M., Jacob, D.D.J., Bey, I., Yantosca, R.M., Field, B.D., Fusco, A.C., Wilkinson, J.G., 2002. Background ozone over the United States in summer: origin, trend, and contribution to pollution episodes. J. Geophys. 107, 4275. Fiore, A.M., West, J.J., Horowitz, L.W., Naik, V., Schwarzkopf, M.D., 2008. Characterizing the tropospheric ozone response to methane emission controls and the benefits to climate and air quality. J. Geophys. Res. 113, D08307. http://dx.doi.org/10.1029/ 2007JD009162. Fiore, A.M., Dentener, F.J., Wild, O., Cuvelier, C., Schultz, M.G., Hess, P., Textor, C., Schulz, M., Doherty, R.M., Horowitz, L.W., MacKenzie, I.A., Sanderson, M.G., Shindell, D.T., Stevenson, D.S., Szopa, S., Van Dingenen, R., Zeng, G., Atherton, C., Bergmann, D., Bey, I., Carmichael, G., Collins, W.J., Duncan, B.N., Faluvegi, G., Folberth, G., Gauss, M., Gong, S., Hauglustaine, D., Holloway, T., Isaksen, I.S.A., Jacob, D.J., Jonson, J.E., Kaminski, J.W., Keating, T.J., Lupu, A., Marmer, E., Montanaro, V., Park, R.J., Pitari, G., Pringle, K.J., Pyle, J.A., Schroeder, S., Vivanco, M.G., Wind, P., Wojcik, G., Wu, S., Zuber, A., Derwent, R.G., Johnson, C.E., 2009. Multimodel estimates of intercontinental source–receptor relationships for ozone pollution. J. Geophys. Res. 114, D04301. http://dx.doi.org/10.1029/2008JD010816. Granier, C., Bessagnet, B., Bond, T., D'Angiola, A., Denier van der Gon, H., Frost, G.J., Heil, A., Kaiser, J.W., Kinne, S., Klimont, Z., Kloster, S., Lamarque, J.-F., Liousse, C., Masui, T., Meleux, F., Mieville, A., Ohara, T., Raut, J.-C., Riahi, K., Schultz, M.G., Smith, S.J., Thompson, A., van Aardenne, J., van der Werf, G.R., van Vuuren, D.P., 2011. Evolution of anthropogenic and biomass burning emissions of air pollutants at global and regional scales during the 1980–2010 period. Clim. Chang. 109, 163–190. http://dx.doi.org/10.1007/s10584-011-0154-1. Hemispheric Transport of Air Pollution 2007. In: Keating, T.J., Zuber, A. (Eds.), Air Pollut. Stud. 16. UNECE, Geneva, Switzerland available at: http://www.htap.org/, 2007. Hemispheric Transport of Air Pollution 2010. In: Dentener, F., Keating, T., Akimoto, H. (Eds.), Air Pollut. Stud. 17. UNECE, Geneva, Switzerland available at: http:// www.htap.org/, 2010. Hilboll, A., Richter, A., Rozanov, A., Hodnebrog, Ø., Heckel, A., Solberg, S., Stordal, F., Burrows, J.P., 2013. Improvements to the retrieval of tropospheric NO2 from satellite — stratospheric correction using SCIAMACHY limb/nadir matching and comparison to Oslo CTM2 simulations. Atmos. Meas. Tech. 6, 565–584. http://dx.doi.org/10. 5194/amt-6-565-2013. Jena, C., Ghude, S.D., Pfister, G.G., Chate, D.M., Kumar, R., Beig, G., Surendran, D.E., Fadnavis, S., Lal, D.M., 2015. Influence of springtime biomass burning in South Asia on regional ozone (O3): a model based case study. Atmos. Environ. 100, 37–47. Ehrlich, P.R., Holdren, J.P., 1971. Impact of population growth. Science 171, 1212–1217. Lee, H.-J., Kim, S.-W., Brioude, J., Cooper, O.R., Frost, G.J., Kim, C.-H., Park, R.J., Trainer, M., Woo, J.-H., 2014. Transport of NOx in East Asia identified by satellite and in situ measurements and Lagrangian particle dispersion model simulations. J. Geophys. Res. 1–23 http://dx.doi.org/10.1002/2013JD021185.Received. Liu, H., 2003. Transport pathways for Asian pollution outflow over the Pacific: interannual and seasonal variations. J. Geophys. Res. 108, 8786. http://dx.doi.org/10.1029/ 2002JD003102. Ma, J., Lin, W.L., Zheng, X.D., Xu, X.B., Li, Z., Yang, L.L., 2014. Influence of air mass downward transport on the variability of surface ozone at Xianggelila Regional Atmosphere Background Station, southwest China. Atmos. Chem. Phys. 14, 5311–5325. http://dx.doi.org/10.5194/acp-14-5311-2014.
T. Chakraborty et al. / Science of the Total Environment 523 (2015) 28–39 Moore, G.W.K., Semple, J.L., 2005. A Tibetan Taylor Cap and a halo of stratospheric ozone over the Himalaya. Geophys. Res. Lett. 32, L21810. http://dx.doi.org/10.1029/ 2005GL024186. Nassar, R., Logan, J.A., Worden, H.M., Megretskaia, I.A., Bowman, K.W., Osterman, G.B., Thompson, A.M., Tarasick, D.W., Austin, S., Claude, H., Dubey, M.K., Hocking, W.K., Johnson, B.J., Joseph, E., Merrill, J., Morris, G.A., Newchurch, M., Oltmans, S.J., Posny, F., Schmidlin, F.J., Vömel, H., Whiteman, D.N., Witte, J.C., 2008. Validation of Tropospheric Emission Spectrometer (TES) nadir ozone profiles using ozonesonde measurements. J. Geophys. Res. 113, D15S17. http://dx.doi.org/10.1029/2007JD008819. Osterman, G.B., Kulawik, S.S., Worden, H.M., Richards, N.A.D., Fisher, B.M., Eldering, A., Shephard, M.W., Froidevaux, L., Labow, G., Luo, M., Herman, R.L., Bowman, K.W., Thompson, A.M., 2008. Validation of Tropospheric Emission Spectrometer (TES) measurements of the total, stratospheric, and tropospheric column abundance of ozone. J. Geophys. Res. 113, D15S16. http://dx.doi.org/10.1029/2007JD008801. Parrish, D.D., Hahn, C.J., Williams, E.J., Norton, R.B., F.C.F., 1992. Indications of photochemical histories of Pacific air masses from measurements of atmospheric trace species at Point Arena, California. J. Geophys. Res. 97, 15,883–15,901. Prospero, J.M., 1979. Mineral and sea salt aerosol concentrations in various ocean regions. J. Geophys. Res. 84, 725. http://dx.doi.org/10.1029/JC084iC02p00725. Rahimi, S., Gholami Sefidkouhi, M.A., Raeini-Sarjaz, M., Valipour, M., 2014. Estimation of actual evapotranspiration by using MODIS images (a case study: Tajan catchment). Arch. Agron. Soil Sci. 61, 695–709. http://dx.doi.org/10.1080/03650340.2014.944904. Reidmiller, D.R., Fiore, A.M., Jaffe, D.A., Bergmann, D., Cuvelier, C., Dentener, F.J., Duncan, B.N., Folberth, G., Gauss, M., Gong, S., Hess, P., Jonson, J.E., Keating, T., Lupu, A., Marmer, E., Park, R., Schultz, M.G., Shindell, D.T., Szopa, S., Vivanco, M.G., Wild, O., Zuber, A., 2009. The influence of foreign vs. North American emissions on surface ozone in the US. Atmos. Chem. Phys. 9, 5027–5042. Richter, A., Burrows, J.P., Nüss, H., Granier, C., Niemeier, U., 2005. Increase in tropospheric nitrogen dioxide over China observed from space. Nature 437, 129–132. http://dx. doi.org/10.1038/nature04092. Sheel, V., Lal, S., Richter, A., Burrows, J.P., 2010. Comparison of satellite observed tropospheric NO2 over India with model simulations. Atmos. Environ. 44, 3314–3321. http://dx.doi.org/10.1016/j.atmosenv.2010.05.043. Singh, H., Brune, W., 2009. Chemistry and transport of pollution over the Gulf of Mexico and the Pacific: spring 2006 INTEX-B campaign overview and first results. Atmos. Chem. Phys. 9, 2301–2318. Solazzo, E., Bianconi, R., Vautard, R., Appel, K.W., Moran, M.D., Hogrefe, C., Bessagnet, B., Brandt, J., Christensen, J.H., Chemel, C., Coll, I., Denier van der Gon, H., Ferreira, J., Forkel, R., Francis, X.V., Grell, G., Grossi, P., Hansen, A.B., Jeričević, A., Kraljević, L., Miranda, A.I., Nopmongcol, U., Pirovano, G., Prank, M., Riccio, A., Sartelet, K.N., Schaap, M., Silver, J.D., Sokhi, R.S., Vira, J., Werhahn, J., Wolke, R., Yarwood, G., Zhang, J., Rao, S.T., Galmarini, S., 2012. Model evaluation and ensemble modelling of surface-level ozone in Europe and North America in the context of AQMEII. Atmos. Environ. 53, 60–74. http://dx.doi.org/10.1016/j.atmosenv.2012.01.003. Stegmann, P.M., 1999. Global distribution of aerosols over the open ocean as. Global Biogeochem. Cycles 13, 383–397. Stohl, A., Forster, C., Huntrieser, H., Mannstein, H., McMillan, W.W., Petzold, A., Schlager, H., Weinzierl, B., 2007. Aircraft measurements over Europe of an air pollution plume from Southeast Asia — aerosol and chemical characterization. Atmos. Chem. Phys. 7, 913–937. http://dx.doi.org/10.5194/acp-7-913-2007. Tu, J., Xia, Z.-G., Wang, H., Li, W., 2007. Temporal variations in surface ozone and its precursors and meteorological effects at an urban site in China. Atmos. Res. 85, 310–337. http://dx.doi.org/10.1016/j.atmosres.2007.02.003.
39
Van Dingenen, R., Dentener, F.J., Raes, F., Krol, M.C., Emberson, L., Cofala, J., 2009. The global impact of ozone on agricultural crop yields under current and future air quality legislation. Atmos. Environ. 43, 604–618. http://dx.doi.org/10.1016/j.atmosenv.2008.10. 033. Wang, T., Wei, X.L., Ding, A.J., Poon, C.N., Lam, K.S., Li, Y.S., Chan, L.Y., Anson, M., 2009. Increasing surface ozone concentrations in the background atmosphere of Southern China, 1994–2007. Atmos. Chem. Phys. 9, 6217–6227. Wild, O., Akimoto, H., 2001. Intercontinental transport of ozone and its precursors in a three-dimensional global CTM. J. Geophys. Res. 106, 27729. http://dx.doi.org/10. 1029/2000JD000123. Wild, O., Fiore, A.M., Shindell, D.T., Doherty, R.M., Collins, W.J., Dentener, F.J., Schultz, M.G., Gong, S., MacKenzie, I.A., Zeng, G., Hess, P., Duncan, B.N., Bergmann, D.J., Szopa, S., Jonson, J.E., Keating, T.J., Zuber, A., 2012. Modelling future changes in surface ozone: a parameterized approach. Atmos. Chem. Phys. 12, 2037–2054. http://dx.doi.org/10. 5194/acp-12-2037-2012. Worden, J., 2004. Predicted errors of tropospheric emission spectrometer nadir retrievals from spectral window selection. J. Geophys. Res. 109, D09308. http://dx.doi.org/10. 1029/2004JD004522. Worden, H.M., Logan, J.A., Worden, J.R., Beer, R., Bowman, K., Clough, S.A., Eldering, A., Fisher, B.M., Gunson, M.R., Herman, R.L., Kulawik, S.S., Lampel, M.C., Luo, M., Megretskaia, I.A., Osterman, G.B., Shephard, M.W., 2007. Comparisons of Tropospheric Emission Spectrometer (TES) ozone profiles to ozonesondes: methods and initial results. J. Geophys. Res. 112, D03309. http://dx.doi.org/10.1029/2006JD007258. Wu, P., Shen, H., Zhang, L., Göttsche, F., 2015. Remote sensing of environment integrated fusion of multi-scale polar-orbiting and geostationary satellite observations for the mapping of high spatial and temporal resolution land surface temperature. 156, 169–181. Yamaji, K., Ohara, T., Uno, I., Tanimoto, H., Kurokawa, J., Akimoto, H., 2006. Analysis of the seasonal variation of ozone in the boundary layer in East Asia using the Community Multi-scale Air Quality model: what controls surface ozone levels over Japan? Atmos. Environ. 40, 1856–1868. http://dx.doi.org/10.1016/j.atmosenv.2005.10.067. Yienger, J.J., Galanter, M., Holloway, T.A., Phadnis, M.J., Guttikunda, S.K., Carmichael, G.R., Moxim, W.J., Levy II, H., 2000. The episodic nature of air pollution transport from Asia to North America. J. Geophys. Res. 105, 26,931–26,945. Yoshitomi, M., Wild, O., Akimoto, H., 2011. Contributions of regional and intercontinental transport to surface ozone in the Tokyo area. Atmos. Chem. Phys. 11, 7583–7599. http://dx.doi.org/10.5194/acp-11-7583-2011. Zhang, Q., Streets, D.G., He, K., Wang, Y., Richter, A., Burrows, J.P., Uno, I., Jang, C.J., Chen, D., Yao, Z., Lei, Y., 2007. NOx emission trends for China, 1995–2004: the view from the ground and the view from space. J. Geophys. Res. 112, D22306. http://dx.doi.org/10. 1029/2007JD008684. Zhang, L., Jacob, D.J., Boersma, K.F., Jaffe, D.A., Olson, J.R., Bowman, K.W., Worden, J.R., Thompson, A.M., Avery, M.A., Cohen, R.C., Dibb, J.E., Flocke, F.M., Fuelberg, H.E., Huey, L.G., McMillan, W.W., Singh, H.B., Weinheimer, A.J., 2008. Transpacific transport of ozone pollution and the effect of recent Asian emission increases on air quality in North America: an integrated analysis using satellite, aircraft, ozonesonde, and surface observations. Atmos. Chem. Phys. Discuss. 8, 8143–8191. http://dx.doi.org/10. 5194/acpd-8-8143-2008.