Competing impact of anthropogenic emissions and ...

J Atmos Chem DOI 10.1007/s10874-016-9331-y

Competing impact of anthropogenic emissions and meteorology on the distribution of trace gases over Indian region Tabish U. Ansari 1 & N. Ojha 2 & R. Chandrasekar 3 & C. Balaji 3 & Narendra Singh 4 & Sachin S. Gunthe 1

Received: 23 September 2015 / Accepted: 15 February 2016 # Springer Science+Business Media Dordrecht 2016

Abstract The spatial distribution of trace gases exhibit large spatial heterogeneity over the Indian region with an elevated pollution loading over densely populated Gangetic Plains (IGP). The contending role and importance of anthropogenic emissions and meteorology in deciding the trace gases level and distribution over Indian region, however, is poorly investigated. In this paper, we use an online regional chemistry transport model (WRF/Chem) to simulate the spatial distribution of trace gases over Indian region during one representative month of only three meteorological seasons namely winter, spring/summer and monsoon. The base simulation, using anthropogenic emissions from SEAC4RS inventory, is used to simulate the general meteorological conditions and the realistic spatial distribution of trace gases. A sensitivity simulation is conducted after removing the spatial heterogeneity in the anthropogenic emissions, i.e., with spatially uniform emissions to decouple the role of anthropogenic emissions and meteorology and their role in controlling the distribution of trace gases over India. The concentration levels of Ozone, CO, SO2 and NO2 were found to be lower over IGP when the emissions are uniform over India. A comparison of the base run with the sensitivity run highlights that meteorology plays a dominant role in controlling the spatial distribution of relatively longer-lived species like CO and secondary species like Ozone while short-lived

Electronic supplementary material The online version of this article (doi:10.1007/s10874-016-9331-y) contains supplementary material, which is available to authorized users.

* Sachin S. Gunthe [email protected]

1

Environmental and Water Resources Engineering Division, Department of Civil Engineering, Indian Institute of Technology Madras, Chennai 600036, India

2

Department of Atmospheric Chemistry, Max Planck Institute for Chemistry, Mainz, Germany

3

Heat Transfer and Thermal Power Laboratory, Department of Mechanical Engineering, Indian Institute of Technology Madras, Chennai 600036, India

4

Aryabhatta Research Institute of Observational Sciences, Manora Peak, Nainital 263002, India

J Atmos Chem

species like NOX and SO2 are predominantly controlled by the spatial variability in anthropogenic emissions over the Indian region. Keywords Trace gases . Air quality . Anthropogenic Emissions . Indo-Gangetic Plain . WRF/ Chem

1 Introduction India is the second most populated country on Earth, with a total population of about 1.25 billion, which is about 18 % of the world’s total population and with an area of around 3.2 million km2. For fulfilling the demands of energy and economy in support of the increasing population and developmental needs, the anthropogenic emissions over this region have increased rapidly (Akimoto et al. 2006). The increased anthropogenic emissions have resulted in poor air quality influencing human health (Lelieveld et al. 2013 and references therein) as well as crop production over India (Ghude et al. 2014; Burney and Ramanathan 2014). For example, in recent years the Indian national capital is experiencing an increase in intensity and frequency of high pollution episodes, owing to the tremendous increase in emission mainly resulting from vehicular emissions. Additionally, tropical convection and stronger winds in the higher altitudes of Indian region lead to significant transport of pollutants from this region to the other parts of the world influencing the global air quality and climate (Brasseur et al. 1999; Seinfeld and Pandis 2006). In order to assess the influences of anthropogenic emissions over the Indian region on the air quality, atmospheric chemistry and global change, the studies of the spatial and temporal distribution of trace gases are essential, especially when as opposed to the Northern America and Europe, the ground-based measurements of air pollutants are sparse and inhibit to study their spatial distribution over the Indian region. Model based studies have indicated that enhancements in the anthropogenic emissions have led to the poor air quality (Pozzer et al. 2012), higher pre-mature mortality (Lelieveld et al. 2013) and reduced crop yields (Ghude et al. 2014) over the Indian region. Additionally, the air masses from highly polluted regions such as Indo-Gangetic Plain (IGP) are found to significantly influence the air quality over the cleaner Himalayan region (Ojha et al. 2012; Sarangi et al. 2014) as well as the marine regions surrounding India (Lelieveld et al. 2001; Lawrence and Lelieveld 2010). Despite such an importance, the ground-based measurements of trace gases, in particular ozone precursors namely carbon monoxide (CO), oxides of nitrogen (NOX) and sulfur dioxide (SO2) are limited (Lal et al. 2000; Naja and Lal 2002; Sahu and Lal 2006a, b; Sarangi et al. 2014) and creating major hindrance to investigate the dominant role of emission and meteorology in currently observed spatial distribution of pollutants. Due to lack of in situ measurements with sufficient spatial and temporal coverage over the Indian region, the satellite retrievals and simulations using chemistry transport models have been used to retrieve the basic information on the distribution of trace gases over this region. The studies utilizing the space-borne observations (Fishman et al. 2003; Jethva et al. 2005; Ghude et al. 2008) and simulations from global chemistry transport models (Beig and Brasseur 2006; Sheel et al. 2010; Ojha et al. 2012) have shown a large heterogeneity in the spatial distribution of the pollutant loading over this region. However, the relative importance of the emissions and meteorology in the observed spatial heterogeneity over this region has not been investigated in detail. In the light of this, the present article aims to decouple the competing influence of spatial distribution of emissions and the meteorology by conducting a sensitivity

J Atmos Chem

study using Weather Research and Forecasting with Chemistry (WRF/Chem), an online regional chemistry transport model. In this study, we focus on the extent to which the spatial distribution of anthropogenic emissions controls the simulated pollutant concentrations in three different seasons (April, May, and July 2012) over the Indian tropical region.

2 Methodology 2.1 Study region: topography, population, and general meteorology Indian region is circumscribed to the southwest by Arabian Sea, to the south east by Bay of Bengal, and to the south by Indian Ocean (Fig. 1 a). The geographical topography of this region is highly complex, which comprises the Himalayas and Tibetan Plateau on the north of Indo-Gangetic Plain (IGP) and the Western Ghats on the east of Kerala (Fig. 1 b). According to the 2011 census the current population of India is around 1.25 billion with urban population of 377 million (31 % of the total population), which is expected to grow further due to rapid industrialization and urbanization. The IGP regions in the north and Kerala region in southwest are the most densely populated regions in India with density reaching as high as 10,000 persons per square km. One of the main reasons for high population density over IGP is the availability of ample water and fertile land as three of the major rivers, the Yamuna, the Ganga, and the Brahmaputra flow through this region. On the other hand Kerala has a locked landscape between Arabian Sea at the west and Western Ghats mountain range on the east, thus making very narrow strip available for the human settlements. Further, the relocation of the rural population to the urban areas has created three megacities: Mumbai (in western

Fig. 1 The Indian domain used in the present study for WRF-Chem simulations: The topography of the Indian region in meters above mean sea level used as the model domain (a) and the distribution of population density (population km−2) over Indian (b). A distinct pattern of high population density over Indo-Gangetic Plain (IGP) and Kerala is evident. The boxes over northern India roughly denote the IGP plain with high population density and box over central India shows relatively lesser population density

J Atmos Chem

India), Kolkata (in eastern India), and New Delhi (in northern India) and a series of city clusters. Owing to rapid urbanization five Indian states (Tamilnadu, Gujrat, Maharashtra, Punjab, and Karnataka) are estimated to have more than 50 % of the population living in cities by 2030 (http://trak.in/tags/business/2010/04/24/india-urban-2030/). There are three distinct meteorological seasons over India: winter (October to January), summer (February to May), and monsoon (June to September) (e. g. Beig et al. 2007; Kumar and Sarin 2009) mainly associated with the movement of ITCZ driven by the differential heating of the Indian Ocean and the Indian subcontinent. Several studies also introduced terms like spring or pre-monsoon and post-monsoon/autumn (Kumar et al. 2010; Kumar et al. 2011) to distinctly emphasize the role of the monsoon in air pollution over Indian region (Ghude et al. 2008; Ramachandran 2007; Bhawar and Devara 2010; Lawrence and Lelieveld 2010). Nevertheless, in the present study we follow the traditional way of defining the seasons over India (three seasons).

2.2 Model description We use the version 3.4.1 of the Weather Research and Forecasting (WRF) model coupled Bonline^ with chemistry (Grell et al. 2005), which can effectively be applied on synoptic scale to microscale. The model does online calculation of dynamical inputs (winds, temperature, boundary layer, clouds etc.), transport (advective, convective, and diffusive), dry deposition (Wesely 1989), gas-phase chemistry, radiation, and photolysis rates (Tie et al. 2003). The chemical mechanism used in this study is the Regional Acid Deposition Model- version 2 (RADM2) (Stockwell et al. 1990) which includes 158 reactions among 36 species. The aerosol module used is based on the Modal Aerosol Model for Europe (MADE-SORGAM) (Ackermann et al. 1998) with Secondary Organic Aerosol Module (Schell et al. 2001). The horizontal resolution of the model is kept as 30 km and time-step of simulation are 3 min. The entire domain is constituted by 100 × 100 grid points with 51 vertical levels with top level of the model was set at 5 hPa. In this study, the physical schemes used to parameterize different processes include Purdue Lin microphysics scheme (Lin et al. 1983), the Rapid Radiative Transfer Model (RRTM) for long wave radiation (Mlawer et al. 1997), Goddard shortwave scheme (Kim and Wang 2011), the Monin-Obukhov surface layer scheme (Monin and Obukhov 1954) with the Yonsei University (YSU) Planetary Boundary Layer (PBL) scheme (Hong et al. 2006), and the Noah Land Surface Model (Chen and Dudhia 2001). National Center for Environmental Prediction (NCEP) reanalysis data (rda.ucar.edu/ datasets/ds083.2) at the spatial resolution of 1o × 1o available every 6 h has been used as meteorological initial and boundary conditions. Initial and boundary conditions for chemical fields have been used by re-speciation mapping from MOZART-4 GEOS5 (http://www.acd. ucar.edu/wrf-chem/mozart.shtml). A detailed description of the MOZART model and its evaluation can be seen in Emmons et al. 2010. The PREP-CHEM-SRC program version 1.4 (Freitas et al. 2011) has been used to map the SEAC4RS anthropogenic emissions on the WRF/ Chem domain (see Fig. 2 a-d). Biogenic emissions in the model are calculated online based on the land use using Guenther scheme (Guenther et al. 1993). The biomass-burning emissions are provided from the NCAR Fire INventory (FINN) (Wiedinmyer et al. 2011). Figure 3 (a-c) shows the exemplary images of CO emissions resulting from the biomass burning according to FINN emissions. It is evident from the figure that the Indian region is strongly affected by the intense biomass burning during the month of April, especially the Northern and Central part of

J Atmos Chem

a

b

c

d

Fig. 2 The spatial distribution of various anthropogenically emitted primary pollutants: Carbon monoxide, which resembles the distribution strongly associated with population density (a), oxides of nitrogen (NOX) associated with city clusters, power plants, and industrial corridors (b), sulfur dioxide (SO2) has the strongest association with spatial distribution of power plants over India and industrial areas (c), and non-methane volatile organic compounds (NMVOCs) coinciding with population density and city clusters (d). All the emissions are in mol km−2 h−1 over the model domain obtained from SEAC4RS inventory

India Wet scavenging and lightning NOX parameterizations were turned off in all the simulations.

2.2.1 Anthropogenic emissions The anthropogenic emissions from SEAC4RS (Studies of Emissions, Atmospheric Composition, Clouds and Climate Coupling by Regional Surveys) inventory of the base year 2012 for CO, NO2, SO2 and volatile organic compounds (VOCs) were used in this study and their

J Atmos Chem

a

b

c

Fig. 3 Biomass burning emissions of CO (in mol km−2 h−1) obtained from Fire Inventory of NCAR (FINN) for the months of a April b July and c December. These FINN fire emissions were used for the base and sensitivity runs using WRF-Chem. A very high amount of biomass burning is evident over domain region during April causing high level of CO emissions, which extremely high emissions over Myanmar

distribution over Indian region is shown in Fig. 2(a-d) respectively. The anthropogenic emissions in this study are kept temporally uniform and do not vary diurnally or seasonally. The emissions represented here are inclusive of four major sectors industry, power, transport, and domestic. The percentage contribution in emissions of different species from four major sectors to the total annual anthropogenic emissions is shown in Table 1. In general, the CO and VOC emissions follow the distribution similar to that in the population density (Fig. 1 b). SO2 and NOX emissions may be mainly attributed to the coal plants and few of the much-localized sources in agglomerated cities. The CO emissions peak up to 1424 mol km−2 h−1, SO2 emissions peak up to 863 mol km−2 h−1 while NOx peaks to 530 mol km−2 h−1 and total NMVOCs show the maximum emission rate of 153 mol km−2 h−1. The CO emissions are in the range from 0 to 1400 mol km−2 h−1 with spatial distribution strongly correlated with population density, location of megacities/metros, and industrial clusters. The high emission rate of CO around 300–700 mol km−2 h−1 was prominently visible over Mumbai – Gujrat industrial corridor and other similar regions. The additional few hotspots in anthropogenic emissions of CO clearly resemble the major city clusters of industrial areas all over India. According to SEAC4RS emissions the total anthropogenic emission of CO including four major sectors in year 2012 was 47.99 Tg. The major contribution (~70 %) was from residential sector, while the lowest contribution (