Evaluation of Particulate Pollution and Health Impacts from Planned Expansion of Coal-Fired Thermal Power Plants in India Using WRF-CAMx Modeling System

Power from coal-fired thermal power plants (TPPs) represents a large percentage of the electricity generated in India. As the demand increases, expansion of the coal-fired TPPs is the most likely scenario, which will lead to an array of environmental and health impacts. The proposed projects in India net a generation capacity of 300 GW through 2030. With limited emission control regulations in place, this will increase the number of health impacts—some from direct particulate matter (PM) emissions and some from secondary PM, especially due to the chemical transformation of sulfur emissions. The WRF-CAMx chemical transport modeling system was utilized to study the impact of these emissions from the planned coal-fired TPPs. The additional 300 GW of projects will result in 3-times the coal consumption and at least 2-times the health impacts (premature mortality and asthma attacks), compared to those estimated for the operational TPPs. The technology to control all criteria pollutant emissions, which could reduce the health impacts linked to ambient PM2.5 from the coal-fired TPPs by as much as 50%, is widely available, and the only barrier to implementing these solutions is the lack of a stricter timeline.


INTRODUCTION
India has the 5 th largest electricity generation sector in the world.Over the next ten years, the power sector must expand to meet the needs of a growing economy.And at least through 2030, coal will remain the primary fuel of choice (Chikkatur et al., 2011;WISE, 2012;Prayas, 2013;GBD-MAPS, 2018;Venkatraman et al., 2018).For the coalfired thermal power plants (TPPs) operational in 2011-2012, Guttikunda and Jawahar (2014) highlighted the impacts of lack of environmental regulations, contributions to primary and secondary particulate matter (PM) pollution, and the health impacts associated with PM pollution from them (80,000 to 115,000 premature deaths).In 2014, a standing committee headed by the Ministry of Environment Forests and Climate Change (MoEFCC), updated the total PM emissions standard from 150 mg Nm −3 to 50 mg Nm −3 for all the new coal-fired TPPs, with no change in the regulations for sulfur dioxide (SO 2 ), nitrogen oxides (NO x ), and mercury, nor for the already commissioned TPPs (PIB, 2013).In 2015, MoEFCC ratified the new regulations, with a deadline capacity of the operational coal-fired TPPs is presented in Fig. 1.The database of TPPs includes geographical location in latitude and longitude, number of boiler units, coal characteristics, coal consumption rates, and installed control equipment.These details were documented from their respective environmental impact assessment reports and data published by the state electricity boards (public entities) and private operators (MoEFCC, 2010;MoEFCC, 2018;NPP, 2018).
To counter the future electricity demands, McKinsey (2008) estimated the need for 300 GW, based on the growth of India's manufacturing sector, domestic electricity demands, the need for grid connectivity for 125,000 villages, and to suppress the blackouts and the load shedding.Prayas (2011) listed that 700 GW of power generation from coal is in the pipeline, with environmental clearances and project preparations at various stages and stated that if these come online, the necessary demand for electricity in the industrial and the domestic sector will be met with surplus electricity through 2030, barring the environmental and health impacts of the emissions from the new coal-fired power plants.
The database of TPPs under construction, under advance development, under planning and appraisals, and under consideration, are binned for operations between 2016 and 2030, with their likeliness of being operational in 2017, 2020, 2025, and 2030.The temporal distribution of the TPPs is speculative based on the information available in the project documents and their status, dependent on resource, financial, and environmental viability for each plant.The plants with less probability of securing the required clearances are not included in the assessment.Having excluded these, the estimated installed generation capacity for the years 2017, 2020, 2025, and 2030are 231 GW, 296 GW, 396 GW, and 458 GW, respectively, with new capacity of 300 GW between 2014and 2030.The planned TPPs are presented in Fig. 2 for 2017, 2020, 2025, and 2030.Not all of these power plants are established as new plants.Some of these are extensions at the existing TPPs.The filled circles in Fig. 2 are overlapping with the existing TPPs.
The totals estimated in this study are less than 700 GW of generation capacity anticipated by Prayas (2011), which is a result of cancellations and withdrawals due to lack of either resource, financial, or environmental clearances.We consider these estimates as conservative.With changes in the coal block allocation strategies and coal import regulations in the coming years, it is likely that more plants could be considered for construction or even the planned TPPs could come under operations faster than anticipated (Prayas, 2013).

Atmospheric Emissions
The total annual emissions were calculated for individual plants, using information on total coal consumed and an emission factor dependent on coal characteristics and the emission control technology for each of the pollutants.The total emissions were established for PM, SO 2 , NO x , carbon monoxide (CO), non-methane volatile organic compounds (NMVOCs), and carbon dioxide (CO 2 ).
The regional emission factors for the coal-fired TPPs is summarized in Table 2 in both tons PJ −1 and tons hr −1 (Guttikunda and Jawahar, 2014).The latter is for comparisons with any data available from online monitoring.Previous studies estimated emissions for all of India and for all the power plants in Asia, and most are estimated for the base year 2000-2005.A serious lack of emissions monitoring data at the stack results in large uncertainty.Under the national ambient monitoring program (NAMP) of India, the large point sources, such as the coal-fired TPPs, are required to conduct continuous emissions monitoring from all the operational stacks for all the criteria pollutants (including other heavy industries like iron and steel, fertilizers, mineral processing, and refineries) (CPCB, 2018).However, the emission rates reported in power plant performance reports are based on intermittent audits and presented as averages or minimum/maximum of the measurements (CEA, 2013(CEA, , 2017)).This makes it harder to compare and/or scrutinize the stack emissions in real time.This is a major barrier linked to lack of protocols on how the monitoring procedures are practiced and how the data is reported to the regulatory agencies.
To understand the uncertainty in emissions rates and total emissions, Monte-Carlo simulations were conducted at the plant level, with normal distribution applied to all the parameters in the emissions calculator.By changing plant size, coal consumption rates, coal characteristics, and   1) and the second color in each map represents all the new plants and expansions expected after 2014 and likely to be operational in the representative year.The largest circle is 4620 MW.Note that many of these circles are overlapping due their close proximity to other TPPs.efficiency of emission controls, this exercise led to an evaluation of likely uncertainty in the overall emissions.For example, coal usage at the new power plants operating at super-critical conditions is lower than those observed at the old power plants.For the plants known to operate FGD systems, controls were applied and evaluated for uncertainty.With normal distribution applied to all the parameters, the overall uncertainty gets compounded.The uncertainty was ±34% for total PM 2.5 emissions, ±18% for total SO 2 emissions, and ±11% for total CO 2 emissions.The lower uncertainty rate for CO 2 emissions is primarily because of lower uncertainty in coal's carbon content than that observed for coal's sulfur content, which originates from both local and foreign mines.These uncertainty rates are similar to those reported for the coal-fired TPPs in China (Chen et al., 2014;Wang et al., 2016).With the availability of stack measurements for individual plants, Chen et al. (2014) reported some lower uncertainties, which is also a likely scenario when better data is available from all the stack monitors in India (CPCB, 2018).d the emission factor segregation was for China, Japan, and Others in Asia.e calculated as ratios of total emissions and coal consumption corresponding to the power sector, PM factor is for total suspended particulates.f the range corresponds to coal fired boilers with and without low NO x burner technology, by boiler size.g range corresponds to the estimated average emission rate per plant in each state.h PM factor is for total suspended particulates; based on measurements at one station in Delhi per stack.
The sulfur content in Indian coal varies from as low as 0.1% from the mines in Jharkhand and 0.4% from the mines in Chhattisgarh to a national average of 0.55%.Coal from the mines of Ohio (USA) are known to contain more than 1.5% sulfur and those from the mines of South Africa is up to 4.0% (Cropper et al., 2012;Sadavarte and Venkataraman, 2014).While the sulfur content is lower in Indian coal, the ash content is higher than 40%.The higher uncertainty rate for the PM emissions also stems from uncertainty in control efficiencies.The older power plants with smaller boilers tend to emit more than the modern power plants equipped with electrostatic precipitators (ESPs) with higher control efficiency, partly due to the segregated standard based on the boiler size (Table 1)-as high as 350 mg Nm −3 for boiler size < 210 MW and 150 mg Nm −3 for boiler size > 210 MW.In 2011-2012, 72% of the operational boilers were of the size ≤ 210 MW (Guttikunda and Jawahar, 2014).

WRF-CAMx Dispersion Model
Atmospheric dispersion modeling was conducted to study the movement of emissions on a regional scale, formation of the secondary sulfate particles (part of PM 2.5 ), and contribution of power plant pollution to health impacts.Ni et al. (2018) utilized the WRF-Chem model to evaluate the impact of coal-fired TPP emissions in China and Dodla et al. (2017) utilized a combination of WRF and HYSPLIT to evaluate the impact of two coal fired TPPs in India.For this study, the ENVIRON Comprehensive Air Quality Model with Extensions (CAMx), an Eulerian photochemical dispersion model, was selected for its modular nature in characterizing and treating the plumes from point sources.The model formulation, advection and scavenging schematics (dry and wet deposition), chemical solvers, and chemical mechanisms are detailed in the model manual.The model includes gas-to-aerosol conversions for SO 2 to sulfates, NO x to nitrates, and VOCs to secondary organic aerosols and supports plume rise calculations for each power plant using 3-dimentional (3D) meteorological data.
The meteorological data (3D wind, temperature, pressure, relative humidity, precipitation fields, and other parameters necessary for chemical transport modeling) is derived from the National Center for Environmental Prediction (NCEP, 2016) global reanalysis database and processed through WRF meteorological model at 1-hour temporal resolution.To further localize the initial conditions, the model was looped for the first 10 days of January before starting the full-year model calculations.After initializing the model, the emissions from the power plants were utilized to isolate the impacts of these emissions on the ambient PM 2.5 concentrations.The combination of WRF-CAMx modeling system was also utilized to evaluate the VOC speciation procedures applied to SAPRC chemical mechanism over the Indian subcontinent (Sarkar et al., 2016) and evaluation of the coal-fired TPPs in India (Guttikunda and Jawahar, 2014).
The modeling domain extends from 7°N to 39°N in latitudes and 37°E to 99°E in longitudes at 0.25° grid resolution (Fig. 1).The vertical resolution of the model extends to 12 km stretched over 28 layers, with the lowest layer designated at 30 m and 12 layers within 1 km to advance vertical advection closer to the ground level.

Health Impacts
In India, the estimated morbidity and mortality burden of outdoor air pollution, in terms of work days lost, lost productivity, and loss in terms of gross domestic product, is approximately USD 23.4 billion and 1.7% of national GDP (World Bank, 2012).The direct link between outdoor air pollution and human health has been extensively documented under the Global Burden of Disease (GBD) study for overall impact between 1990 and 2016 (GBD, 2017) and for major polluting sources (GBD-MAPS, 2018), with more than 695,000 and 1,000,000 premature deaths attributed to outdoor air pollution from PM 2.5 and ozone in 2010 and 2015, respectively.Guttikunda and Jawahar (2014) established that the emissions from the coal-fired TPPs in 2011-2012 could account for 80,000 to 115,000 annual premature deaths, approximately 11.5 to 16.5% of the total estimated premature deaths under the GBD analysis.An updated assessment under GBD-MAPS (2018) estimated 83,000 premature deaths, approximately 7.6% of the total estimated for 2015.
Using the GBD methodology, the total health risk, as mortality, can be quantified to assess the impact of India's planned coal expansion.The GBD methodology accounts for the cases of ischemic heart disease (which can lead to heart attacks), cerebrovascular disease (which can lead to strokes), chronic obstructive pulmonary diseases, lower respiratory infections, and cancers (in trachea, lungs, and bronchitis) (Burnett et al., 2014).Each of these endpoints have established integrated exposure risk (IER) coefficients, sub-divided into age groups, with detailed descriptions at the country level presented in Cohen et al. (2017).A combination of all these risks can be summarized as an allcause mortality supra-linear function for relative risk (RR) of PM 2.5 pollution (Pope et al., 2015) as the following: where δC is the PM 2.5 concentration at the grid level (including both the primary and secondary contributions), IR is the death incidence rate established as part of the GBD study, and δPOP is the population exposed in each grid.The total concentrations at the grid level are taken from the CAMx simulations for each corresponding year.Pope et al. (2015) utilized the all-cause mortality function to assess how the supra-linear shape of the IER curve influences the health benefits of air pollution abatement policies in the United States.Given the uncertainties in the study calculations from fuel usage, emission factors, and control efficiencies, and to total emissions, which are propagated through a dispersion model (with chemistry module to include secondary contributions) to estimate final PM 2.5 concentration shares, it was decided to use the allcause mortality risk function for the end discussion instead of evaluating 5 distinct mortality risk functions.
The total population per grid is extracted from gridded databases (Landscan, 2013;GRUMP, 2015).Most national censuses count populations by measuring where people live rather than where they work or travel.Landscan integrates daytime movements and collective travel habits into a single measure to produce a better representation of where people are located during an average day.This database is available at 30-second resolution, which is aggregated to 0.25° CAMx model resolution.

Atmospheric Emissions
Summary of annual coal consumption and emissions in 2014 and from the planned TPPs in 2030 is presented in Table 3 • Total coal consumption is expected to increase 3-times from 660 million tons year −1 in 2014 to 1800 million tons year −1 in 2030, and accordingly, the CO 2 emissions from 1,590 to 4,320 million tons year −1 .This is after taking into consideration that a majority of the newer TPPs are of minimum 660 MW capacity and will operate under super-critical conditions with better performance rates than the existing TPPs.Four TPPs in Odisha are expected to utilize ultra-critical technology, with further improvement to the performance ratios.• The PM, SO 2 , and NO x emissions will at least double in the same period.This is less than the 3-time increase in generation capacity, primarily due to the introduction of newer boiler technology, utilizing less coal per MWh of electricity generated.With no emission regulations in place for SO 2 and NO x , these are assumed uncontrolled and allowed to release at the elevated stacks for dispersion.The scenario analysis includes application of FGD systems to further reduce the overall emission rates.An important impetus in this analysis is that the emissions and the impacts are studied only from the supply side of the electricity generation and not from the demand side.It is arguable if the proposed list of TPPs through 2030 can meet the demand or not meet the demand or is this surplus.The subjective judgment is based on available information on the supply side for those TPPs which have secured some form of environmental clearance and showed resource and financial arrangement for operations.We expect this to be different in the coming years.

Ambient PM 2.5 and SO 2 Pollution
The dispersion modeling results for each month were averaged to obtain monthly, seasonal, and annual  Streets, 2012;Prasad et al., 2012).
The collective impact of the TPPs over each state is presented in Table 4, as population weighted concentrations, which is indicative of the pollution load observed in each state, irrespective of the size of the installed capacity in that state.The model grid size is 0.25° (~25 km × 25 km).Due to this coarse spatial coverage, these numbers cannot be directly compared to the data from the ambient monitoring stations, which only represent their immediate vicinity.These concentrations highlight the role of the long-range transport of high-stack emissions.For example, the northeastern states of Assam, Nagaland, Mizoram, Manipur, Meghalaya, Arunachal Pradesh, Tripura, and Sikkim; the smaller states, Goa, Puducherry, and Delhi; and the hilly states, Uttarakhand and Himachal Pradesh, do not operate any coal-fired TPPs within their administrative boundaries and yet experience a significant amount of pollution originating from them.The cities with no TPPs or the cities not located near a TPP also experience impact of TPP emissions.For example, major cities in the Korba region (of Chhattisgarh), Ranchi, Jamshedpur, Rourkela, Jabalpur, Nagpur, and Raipur (capital of Chhattisgarh); major cities in the Mundra region, Jamnagar (major industrial port), Rajkot, and Ahmedabad (300 km away, with two TPPs of 1000 MW in the city); and the city of Delhi, with large TPPs within 100 km of radius, experience the impact of these emissions.

Health Impacts
A summary of the health impacts associated with the planned coal-fired TPP expansion in India is presented in Table 5.The total premature mortality due to the emissions from coal-fired TPPs is expected to at least double, reaching 186,500 to 229,500 annually, and the associated asthma cases reaching 42.7 million in 2030.For the future years, we assumed 1.1% net annual growth rate in the population (starting with the latest census projections), with the total population reaching 1.5 billion.For a higher growth rate of 1.3%, the total premature mortality due to the emissions from coal-fired TPPs could be as high as 238,000 annually in 2030.
These results are further disaggregated at the state level in Table 6.The most populated states of Maharashtra, Uttar Pradesh, Bihar, Andhra Pradesh (including Telangana), Odisha, Madhya Pradesh, and West Bengal, which harbor the largest clusters of the power plants, are listed with the most number of premature deaths associated with the emissions from coal-fired TPPs.Gujarat and Chhattisgarh, despite large clusters, register lower health impacts for two reasons-lower population density in the state and longrange transport of the emissions.In case of Gujarat and other coastal states, land-sea breeze is an advantage during some months.A large amount of the emissions from the states of Bihar, Chhattisgarh, Madhya Pradesh, Maharashtra, and Andhra Pradesh get deposited within the states or in the neighboring states.Of the northeastern states, Assam is the most populated and immediate to the mainland, with significant health impacts.These estimates do not include the health impacts of the TPP emissions in the neighboring countries-Nepal, Pakistan, Bangladesh, and Sri Lanka.

EMISSION REGULATIONS AND CONTROLS
It is ironic that SO 2 is the only pollutant reported with concentrations under the national ambient air quality standard in India (Guttikunda et al., 2014).This is misleading since all the monitors under NAMP are in the cities and this under-represents pollution unmonitored outside urban areas, where the main sources of SO 2 are located.Two of  concentrations (Lu and Streets, 2012;Prasad et al., 2012).

Application of FGD Systems in India
Table 1 presents the new stack emission standards for coal-fired TPPs (MoEFCC, 2015).While the new PM emission standard in India is comparable to other countries, there is a conspicuous lack of operations to control SO 2 and NO x emissions.The secondary sulfate aerosols are a large part of the PM 2.5 concentrations (Guttikunda and Jawahar, 2014) which can be avoided by mandating some form of control for SO 2 emissions (Chikkatur et al., 2011).In India, only three coal-fired TPPs in Maharashtra and one in Karnataka operate FGD systems.According to MoEFCC, installation of FGD is in process at NTPC Bongaigaon (Assam), NTPC Vindhyachal Stage V (Uttar Pradesh), and Adani Power Mundra Phase III (Gujarat) (PIB, 2012).
The Trombay thermal power plant (TTPP) (near Mumbai) uses sea water's natural alkalinity to scrub SO 2 from the flue gas.After neutralization, the sea water passes through the cooling water and heat exchanger, and the effluent is discharged back into the sea.The removal efficiency is estimated at 85-90%.Because of the use of sea water for scrubbing (and no additional sorbent), the designing and operations are performed at the lowest cost.A disadvantage, however, is that the pollution is discharged into the sea, which in the long run, could lead to other contamination.
The Dahanu thermal power plant (DTPP) started its commercial operation in 1996.However, only after an order from Bombay High Court dated 12 May 1999 were they required to install an FGD system for environmental safety and protection and for the well-being of the people of Dahanu.This 2 × 250 MW plant also uses 80% local coal from the Korba mining areas (in Chhattisgarh) and remaining 20% imported from Indonesia and South Africa.DTPP also utilizes sea water for scrubbing and cooling in its FGD plant.
The 2 × 600 MW Udupi thermal power plant (UTPP) (near Mangalore) started its commercial operations in 2010, is also a coastal power plant, and operates limestone injection and gypsum production system to control SO 2 emissions.This FGD technology is a zero-discharge system utilizing all wastewater in the system, thus reducing the need for fresh water and eliminating waste disposal costs.
According to Prayas (2011) only 7 plants, or just 3.2% of the total coal-fired TPP capacity that has been granted environmental clearance, have provision for installing and operating an FGD (Table 7).With no mandatory requirements, in all the remaining future power plants, only a space provision is mentioned in the environmental clearance process, in case an FGD is to be installed in the future.MoEFCC (2015) issued December 2017 as the deadline for implementation of FGD at all operational TPPs.However, this is now delayed by at least 4 years.This method includes application of wet limestone scrubbing or a spray dryer process on the flue gas, after the combustion, to form gypsum as a by-product.A wet FGD flue gas treatment system is usually located after removal of PM via an electrostatic precipitator (ESP) and the cleaned gas is discharged to the stack for further dispersion.Gypsum can be used for producing building material.• The high efficiency regenerative desulfurization process is relatively expensive compared to the other two processes and produces SO 2 rich gas (~97%) which can be used as raw input in chemical industry to produce sulfuric acid or even elementary sulfur.Caustic soda (sodium hydroxide) is used as sorbent, which is regenerated to keep the sorbent losses to the minimum.Typical sulfur removal rate of more than 98% is possible, along with tons of commercial by-products.

Need for Mandating FGD at India's TPPs
The only justification towards not mandating the FGD systems in India is the availability of the low sulfur fuel (compared to sulfur content of coal from other countries).This scenario will change as soon as more coal is imported from countries like Indonesia, South Africa, and Australia to meet the demand.While the sulfur content is low in the coal, more tonnage of coal consumed means more tonnage of sulfur emitted at the same clusters.This was also evident from the analysis of satellite observations, suggesting an increase of 60% in the regional sulfur concentrations, especially in and around the TPP clusters, between 2005and 2012(Lu et al., 2013).
An immediate benefit of installing and operating FGD systems is for the human health.The share of secondary sulfates contributing to the ambient PM 2.5 ranges from 30-40% and can be as high as 60% for the denser clusters (Guttikunda and Jawahar, 2014).By controlling sulfur emissions either during the combustion, which can achieve up to 60% removal, or post-combustion, which can achieve up to 98% removal, the overall health impacts can be reduced accordingly.For example, for 2011-2012, the health impacts calculated for the modeled PM 2.5 pollution from the coal-fired TPPs ranged between 80,000 and 115,000 per year.With application of FGD systems for all these TPPs, this could have been reduced by at least 26,000 (for 60% removal efficiency) or 38,000 (for 95% removal efficiency).Even with a conservative value of INR 2,000,000 (approximately USD 40,000) per life lost, based on the average life insurance policies issued in India, the estimated benefits could have ranged from INR 5,100 to 7,600 crores (approximately USD 0.9 to 1.3 billion).
For the planned coal expansion, these benefits at least double.A summary of the anticipated health benefits with FGDs operational at all the TPPs is summarized in Table 9.By mandating some form of FGD system operational at all the plants, the benefits of lives saved alone could range from INR 12,200 to 20,300 crores (approximately USD 2.0 to 3.4 billion) in 2030.Assuming the systems are operational in 2017, as required when the new emission standards were passed (MoEFCC, 2015), cumulative benefits through 2030 will be an estimated USD 23.8 to 39.2 billion, enough to justify the costs of implementing and operating an FGD at every existing coal-fired TPP.This does not include the morbidity costs of hospital visits or hospitalization in case of asthma attacks, bronchitis, or other cardio-vascular ailments, which could at least double these estimated health benefits.
The co-benefits of an FGD system extend to other pollutants.For example, during wet FGD process, total PM is also trapped in the sorbents, resulting in further removal of the PM emissions in the flue gas.Given the volume of the coal consumed and the ash content, even a fraction of improvement in the PM removal efficiency will result in large benefits for ambient PM concentrations and health impacts.This is included in the calculations discussed above.For the operational and the proposed coal-fired TPPs, the benefits of operating an FGD are open for interpretation and only require a stricter timeline on implementation of FGDs for overall improvements in air quality and health.

CONCLUSIONS
Supporting the 3 rd largest economy in the world, the supply of power in India can scarcely keep up with the demand.Across the country, households and industry suffer from regular power cuts, with more than 400 million still lacking access.The need to expand power generation capacity and deliver more electricity to India's growing population and economy is urgent.
The current pipeline of projects will result in establishing a new generation capacity of 300 GW through 2030; some of these facilities are expected to perform better than the TPPs operational in 2014 and 2015.While the sulfur content of Indian coal is lower than that observed in other countries,  164,000-197,500 54,500-90,500 10,900-18,100 Year 2030 186,500-229,500 61,000-101,500 12,200-20,300 with an increase in the overall coal consumption (at least tripling through 2030), these emissions will be large enough to significantly affect the ambient PM 2.5 concentrations and the resulting health impacts.The technology to control SO 2 emissions, is widely available, and the only barrier to implementing these solutions at all coal-fired TPPs is the lack of a timeline mandating that the TPPs operate FGD systems.
In India, the mixture of emissions is complex, and the sources of these emissions are many.The coexistence of high concentrations of primary and secondary gaseous and aerosol pollutants results in numerous heterogeneous reactions.These reactions change the oxidizing capacity of the atmosphere, and the chemical compositions and optical properties of PM, resulting in accelerated haze formation and other air pollution impacts.Thus, it is imperative that the chemical mechanisms leading up to these effects are studied and controlled, even if it means controlling one pollutant and one source at a time.In this paper, the dispersion modeling results and health benefits analysis of controlling SO 2 and PM emissions are presented since these two pollutants have proven direct linkages to human health and their fractional contributions to the ambient PM 2.5 concentrations are higher than those of other pollutants.This study will be extended to include the impacts of TPP emissions on the overall regional photochemistry, involving the full chemical mechanisms of the NO x -VOC-O 3 cycle.

Fig. 1 .
Fig. 1.Installed capacity of the coal-fired TPPs in India.The largest circle is 4620 MW.Note that many of these circles are overlapping due their close proximity.The inlays present details for the four largest clusters.

Fig. 2 .
Fig. 2. Proposed locations of the coal-fired TPPs in India through 2030.The brown circles represent the TPPs operational in 2014 (details in Fig. 1) and the second color in each map represents all the new plants and expansions expected after 2014 and likely to be operational in the representative year.The largest circle is 4620 MW.Note that many of these circles are overlapping due their close proximity to other TPPs.

Fig. 5 .
Fig. 5. 48 hour forward trajectories drawn over the Korba (Chhattisgarh), Jhajjar (Haryana), Mundra (Gujarat), and Mumbai (Maharashtra) power plant clusters to illustrate the movement of the emissions for three months, using the NOAA HYSPLIT trajectory model.

Table 2 .
Regional emission factors database.the range corresponds to the averages over the states.b base line factors for various technologies without or limited controls, global program.c base line factors with best available control technology for each pollutant, global program. a Telangana is the new state carved out of Andhra Pradesh in 2014, which also harbors the largest coal mines, bordering Odisha and Maharashtra, and plans to establish more than 10 GW of coal-fired TPPs to support the planned industrial and domestic electrification plans.
• Coal-fired TPPs in the northeastern states will be commissioned in Assam before 2018 (which is under construction) and in Meghalaya before 2025 (which is under appraisal).

Table 3 .
Total power generation capacity (in GW), annual coal consumption (in million tons) and annual emissions (in ktons for PM 2.5 , SO 2 , NO x , and CO; and in million tons for CO 2 ) from the coal-fired TPPs in 2014 and 2030 in India at the state level.

Table 4 .
Modelled state average PM 2.5 concentrations (indicative of the pollution load) due to the emissions from the planned expansion of coal-fired TPPs in India.The concentrations are in µg m -3 and the data represents -population weighted state average concentration ± standard deviation of concentrations for all grids covering the state and (in the brackets -maximum concentration among the grids covering the state).

Table 5 .
Anticipated health impacts due to ambient PM 2.5 pollution from the planned coal-fired TPPs in India.

Table 6 .
Anticipated health impacts by state due to ambient PM 2.5 pollution from the planned coal-fired TPPs and the total population from Census-2011 in India.

Table 9 .
Anticipated health impacts of emissions from planned coal-fired TPPs and likely number of lives saved by operating a FGD units in India.