key: cord-1017731-fafd957f
authors: Tibrewal, Kushal; Venkataraman, Chandra
title: COVID-19 lockdown closures of emissions sources in India: Lessons for air quality and climate policy
date: 2021-11-09
journal: J Environ Manage
DOI: 10.1016/j.jenvman.2021.114079
sha: 4110d60ae270575c63fbf3535748287005062a14
doc_id: 1017731
cord_uid: fafd957f

Reduced anthropogenic activities during the COVID-19 pandemic caused significant reductions in ambient fine particulate matter (PM(2.5)), SO(2) and NO(x) concentrations across India. However, tropospheric O(3) concentrations spiked over many urban regions. Moreover, reductions in SO(2) and NO(x) (atmospheric cooling agents) emissions unmask heating exerted by warming forcers. Basing governmental guidelines, we model daily emissions reductions in CO(2) and short-lived climate forcers (SLCFs) during different lockdown periods using bottom-up regional emission inventory. The transport sector, with maximum level of closure, followed by power plants and industry reduced nearly −50% to −75% emissions of CO(2), primary PM(2.5), SO(2) and NO(x), while warming SLCFs (black carbon, CH(4), CO and non-methane VOCs) showed insignificant reduction from continuing activity in residential and agricultural sectors. Consequently, the analysis indicates that reduction in the emission ratio of NO(x) to NMVOC coincided spatially with observed increases in O(3), consistent with reduced uptake of O(3) from night-time NO(x) reactions. Also, similar reductions, occurring for longer timescales (say, a year), can potentially increase the annual warming rate over India from the positive regional temperature response, estimated using climate metric. Further, by linking ongoing policies to sectoral reductions during lockdown, this study shows that the relative pacing of implementation among policies is crucial to avoid counter-productive results. A key policy recommendation is introduction and improving efficacy of programs targeting reduction of NMVOC and warming SLCF emissions (shifts away from biomass cooking technologies, household electrification and curbing open burning of crop residues), must precede the strengthening of policies targeting NO(x) and SO(2) dominated sectors.

The outbreak of the new coronavirus has disrupted human lives worldwide, while influencing 38 the environment indirectly. Coronavirus disease (COVID-19, name given to the disease caused by the 39 virus) was first detected around December 2019 in the Wuhan province of China, which later spread 40 throughout the world and was declared as a pandemic by the World Health Organization (WHO, 2020). 41

Restricting human interaction was considered to be the most effective strategy to prevent the spread. 42

Many governments imposed national lockdowns, shutting down major economic activities and 43 mobility. COVID-19 lockdowns provide a unique opportunity to analyse the system's response of such 44

"unprecedented" controls on emission sources. 45

The unprecedented closure of many anthropogenic activities led to a pause in emissions of 46

various pollutantssuch as fine particulate matter (PM2.5), black carbon (BC), organic carbon (OC), 47

oxides of nitrogen (NOx), sulphur dioxide (SO2), carbon monoxide (CO), methane (CH4) and non-48 methane volatile organic compounds (NMVOCs), which affect regional air quality (Venkataraman et 49 al., 2018) and near-term climate (CCAC, 2014; IPCC, 2018) . Besides air quality, these pollutants also 50 exert significant temperature response in the near-term by altering the earth's radiative balance (Collins 51 et al., 2013). Owing to their short atmospheric lifetimes (ranging from few days to months or few years) 52

they are referred to as short-lived climate forcers (SLCFs). SLCFs including BC, CH4, NMVOCs and 53

CO, absorb radiation leading to atmospheric warming, are called warming SLCFs (wSLCFs) while 54

NOx, SO2 and OC, scatter radiation causing a cooling effect, are called cooling SLCFs (cSLCFs). Thus, 55 reduction in cSLCFs without complimentary reductions in wSLCFs can unmask the reduction in 56

warming due to GHG mitigation. 57

The effects of India's COVID-19 related lockdowns on air quality levels have been discussed However, a sequence of lockdowns followed with some activities reopening in a phased manner. 62

Several studies analysed the influence of lockdowns on regional air pollution using data from in-situ air 63 quality monitoring stations across India. Overall, a reduction of ~50% PM2.5 was reported, with 64 reductions varying across sites ( Emissions are resolved into 25 x 25 km grids through sector-specific spatial proxies (Sadavarte 118 and Venkataraman, 2014). Temporally, annual emissions are first distributed by months and then daily 119 emissions are estimated assuming equal emissions in a month. Monthly emissions for residential heating 120

were based on monthly mean-minimum temperatures (Pandey et al., 2014) while the distribution for 121 agricultural residue burning followed the seasonality as present in Global Fire Emissions Database 122 (GFED4) (Giglio et al., 2013) . Brick industry emissions were distributed equally across the operational 123 months (November to June). Emissions for power plants and refineries followed the reported monthly 124 generation and crude throughput values. 125

Under COVD, emissions changes occurred primarily due to reduction in sectoral activity, there 127

were no shifts in technology. Thus, reductions in emissions are proportional to the reductions in activity 128

with respect to BASE ( Figure 1 ). Therefore, the percentage reduction in activity under COVD with 129 respect to BASE for each sector and each lockdown period are multiplied by the BASE emissions to 130 estimate the sectoral emissions during lockdown. Modelling activity reductions requires transferring 131 the closure guidelines ((MoHA, 2020b); Table S1 ) to the inventory source categories (Table 1) . 132

Primarily, sectors such as power plants, refineries, light industry and passenger road transport 133 experienced varying intensities of restrictions across the lockdown periods ( Figure 1 ). 134

We have attempted to include the inherent temporal and spatial variations in our estimates. For 135 power plants and refineries, actual electricity generation (CEA, 2020) and monthly crude throughput 136

(MoPNG, 2020) for 2020 were obtained. Heavy industry (including cement, iron and steel and non-137 ferrous metals) and brick kilns are shut off completely during the first lockdown. During the second 138

lockdown, industries only in the rural areas were allowed to start operation. Since most of these 139 manufacturing plants/units are located in or around rural areas, they are allowed to resume from second 140 lockdown onwards. Light industry is assumed to be completely non-operational during the first 141 lockdown, partially resumed in the second lockdown and completely operational from third lockdown 142

onwards. The ratio of urban population to total population is used as a proxy for activity reduction 143 during this phase. Railways remained shut till 10 th of May and resumed completely henceforth. 144

Estimating activity reductions in passenger road transport sector involved a more complex 145

approach. While all passenger vehicles were completely off until the second lockdown, from third 146 lockdown onwards these were partially resumed based on the district classification. All districts across 147 the country were divided into three zones by the governmentred, orange and green, in descending 148 order of the number of active cases. Zonal distribution of districts at the onset of the third lockdown is 149 used as the base classification ( Figure 2a ). For each zone there were different guidelines for public and 150

private vehicles respectively ( Figure 2b ). Same restrictions are imposed on districts within each zone 151 based on the guidelines. Now, districts got shuffled across different zones on a daily to weekly basis 152 depending on the number of active cases, for which the required information was difficult to collate. 153

Thus, the restriction levels for each zone are simply relaxed gradually from fourth lockdown onwards 154

based on an educated judgement while maintaining the base zonal classification. 155 156 We link estimated emissions changes from scenarios representing the COVID lockdown 157

periods to the observed air quality impacts viz-a viz changes in reported ambient ozone concentrations 158

and estimated potential climate impacts using emissions metric. 159

Tropospheric ozone is formed by photochemical reactions of NOx, NMVOCs, CH4 and CO. to the region's ozone formation regime which dictates the type and relative amounts of precursor(s) that 165 must be controlled. Change in ozone concentrations roughly follows the shifts in "ratio of NMVOCs to 166

NOx emissions (ERNMVOCs/NOx)" before and after any intervention (Liu et industry, freight transport and agricultural sectors activities did not alter, the lockdown affected 192 activities for passenger transport, industry, power plants and refineries very significantly. The transport 193

sector was most strongly influenced with a -100% reduction in activity for the first two lockdown 194 periods followed by a phased renewal of the activity. Activity levels for private and public passenger 195 transport reduced ranging from -85% and -75% in third lockdown recovering to -30% and -25% in the 196 sixth lockdown, respectively. Within the industry sector, activity levels for heavy and light industry 197

were reduced by -100% in the first lockdown. Heavy industry includes fertilizers, cement, non-ferrous 198 metal and steel plants, while the balance such as chemical, pulp and paper, machinery, mining, textile 199

and other industries are included in light industry. While heavy industry had no reduction in activity in 200 the second lockdown, light industry saw a partial recovery with an activity level of -35%. Both sub-201 sectors resumed their activities completely from third lockdown period onwards. Power plants and 202

refineries had a consistent decrease in activity of around -30%, with the reduction for power plants 203

plummeting to just -6% in the sixth lockdown. 204 Figure 3 shows the evolution of different classes of pollutants from January 1st to July 31st of 205 2020, which drive air quality and climate responses respectively. Air quality drivers include the 206 emissions changes related to primary fine particulate matter (primary PM 2.5; Figure 3a ) and ozone 207

precursors -NOx ( Figure 3b ) and NMVOCs (Figure 3c ). Climate drivers are represented by CO2 ( Figure  208 3d) and SLCFs (multiplied by respective GWP20 to obtain CO2-e). SLCFs are further disaggregated as 209

warming wSLCFs (BC, CH4, CO and NMVOCs; Figure 3e ) and cooling cSLCFs (SO2, NOx and OC; 210 Figure 3f ) based on their absorption versus scattering of solar radiation. Overall, we find that the 211 emissions reductions peaked during the first lockdown period of 21 days. However, the reductions 212 varied across the different pollutants. For primary PM2.5, NOx, CO2 and cSLCFs, the reductions in daily 213

emissions during the COVID-19 lockdowns with respect to BASE, were estimated to be as high as -214

75% changing to around -50% within the first lockdown, followed by -25% by the second lockdown 215

and gradually decreased henceforth. However, the reductions for NMVOCs and wSLCFs never 216

diminished below -30%, as these arise primarily from sectors such as residential and agriculture, in 217

which activities continued without change during the COVID-19 lockdowns. While the reductions in 218

NMVOC emissions dropped very gradually during the lockdown periods, for wSLCFs the reductions 219

primarily occurred in the first lockdown (from reductions in activity in industry, brick production and 220 transport sector), with significant recovery (less than -15%) thereafter (from resumption of activity in 221 industry and brick production). Similar ranges in emissions reductions are also reported over major 222

Indian Among the air quality drivers (Figure 3a-c) , primary PM2.5 emissions had an aggregate 225 reduction of 285 Gg for the seven months in COVD as compared to BASE, while NOx and NMVOCs 226

were reduced by 700 and 350 Gg respectively. Primary PM2.5 and NOx are emitted primarily from by 227 industry, transport and followed by power plants. Since the restrictions from industries were lifted from 228 the second lockdown, the reductions were primarily driven by power plants and transport after the first 229

lockdown. Transport had a slightly greater contribution to reduction than power plants. However, for 230

NMVOCs, the reductions were driven only by transport shut down. Among the climate drivers ( Figure  231 3d-f), CO2 had an aggregate reduction of 200 Gg and wSLCFs 35 GgCO2-e for the seven months. 232

Reduction in cSLCFs leads to decrease in the cooling effect, thus analogous to increase in CO2e 233

emissions. Thus, for cSLCFs the aggregated increase in CO2e emissions is estimated to be 135 GgCO2e 234

during the seven months. CO2 and cSLCFs shared a similar sectoral contribution dominated industry, 235 power plants and followed by transport. Power plant closures drove the majority of the reductions from 236 second lockdown onwards. As for reductions in wSLCFs, it had nearly equal contributions from 237 industry, brick production and transport in the first lockdown, while it was driven mostly by transport 238 since the second lockdown. 239

Emissions changes of such magnitudes (nearly 50 to 75% reduction), even though lasting for a 241 brief period, can provide interesting insights regarding the atmospheric response to such perturbations, 242

in the contexts of air quality and climate. These can serve as heuristics for strategizing future policy 243

actions. In this context, we establish links between the emissions changes and observed as well as 244 potential impacts (Figure 4 and 5) . 245

In regard to regional air quality, sensitivity of ozone concentrations towards precursors has been nearly 80% points shifting within a factor of 2, 10% between 2 to 3 times and 5% more than 3 times 263 (Figure 4c. ). This is because of relatively lesser reductions in NMVOCs than NOx. In India, nearly 63% . We estimated the potential 281 change in regional temperature from reductions in SLCFs over India during the lockdown using 282 ARTP20 as shown in Figure 5 . Combining the change across all SLCFs, it is estimated that any 283 mitigation strategy controlling the sectors in proportions as happened during respective lockdowns will 284 lead to a potential increase in the regional temperature rate, ranging from 0.57 to 0.05 mK/yr across 285 lockdown periods. As we applied ARTP20, this represents the surface temperature response over the 286 tropics band after 20 years, a commonly used time-horizon to evaluate the near-term impacts of policies. 287

Temperature response calculated using this metric (i.e., emissions (Tg/yr) x metric (mK/Tg) = response 288

(mK/yr)) can be interpreted as annual warming rate. Thus, to put things in to perspective, we compared thereby increasing the response gradually just after emissions perturbation, peaking it at timescales 296 shorter than 20 years and declining henceforth. Thus, increase in annual rate of warming may be even 297

higher than this at shorter-timescales. Moreover, since this represents the mean response over the whole 298 latitude band, an estimate within a finer domain (i.e., Indian land mass) may lead to much higher rate 299 of warming. The reduction in NOx and SO2 emissions account for ~80-85% of the estimated temperature 300

response. This reiterates the need to expand policy to include specific sectors targeting wSLCF 301 emissions to counteract unmasking of heating from cSLCF reductions. 302

We analysed the satellite retrieved changes in columnar burdens of NOx (Krotkov et al., 2019) 303

and SO2 , which are found to drive the increase in warming. Mean columnar burden 304 differences between 2020 lockdown periods and corresponding periods from 2019, from daily 305 0.25°×0.25° gridded Level 3 data products from the Ozone Monitoring Instrument (OMI) aboard Aura 306 satellite ( Figure S2 b and c) , showed a majority of the regions experiencing significant reductions, 307

around -50% and above ( Figure S2 ). Overall, we find some consistency in the reduction ranges and 308 spatial coverage over the hotspot regions, between the inventory calculation and satellite detection. A 309

one-to-one comparison of emissions reductions and burdens reductions is not recommended due to 310 nonlinearity in the processes that decide the fate of emissions, such as meteorology and atmospheric 317 We made a link to policy by making an analogy between shutdown in activities and selected 318 ongoing national programs. The aim is to understand the non-intuitive impacts of existing policies 319 toward formulating corrective measures. Since transport, industry (including refineries) and power 320 plants were affected we discuss policies related to these sectors. Policies can be categorised based on 321

their primary agenda such as alleviating regional air quality and actions towards climate change 322 mitigation. 323

Air quality policies primarily promote implementation of emissions control technology to 324

control the degradation of regional air quality. Key policies include stringent emission standards for 325 transport and power plants. For the transport sector, India has already begun the shift to Bharat Stage 326

Emissions Standards VI (BS-VI) from BS-IV (The Hindu, 2020). Bharat Stage Emissions Standards 327 refer to norms that specify the permissible amounts of air pollutants such as PM2.5, SO2, NOx that can 328 be released from vehicle exhaust. BS-VI are the most stringent emissions standards currently at par with 329 international standards. Shifts to BS-VI can cut the NOx emissions factors by nearly 25% to 85% from On the other hand, climate policies mostly target decarbonization i.e., reduced dependency on 336

fossil fuels through reduction in energy consumption and shifts to alternate cleaner fuel. Key policies 337 include the Perform, Achieve and Trade (PAT) scheme for the industry sector, shifts to electric vehicles 338 in transport and shifts to renewable power generation. Unlike the air quality policies that are based on 339 end-of-pipe control of specific pollutants, these policies influence all the pollutants associated with that 340

technology. Thus, besides mitigating CO2, they have co-benefits in mitigating other pollutants. 341

We need to see the reduction in activity during lockdown through the lens of these policies. By  342 analysing the air quality and climate responses during the lockdown (as discussed above), we can have 343 a rudimentary understanding of the counterproductive effects of implementing these policies. Thus, for 344 each lockdown, reduction in activities from industry, transport and power plants can be thought of as 345

an analogy to implementation of PAT scheme, FGDs, BS-VI and shifts to renewable power generation 346

and EV cars. For the discussed ozone response, BS standards mimic the transport closure partially as it 347

only affects NOx. Since NMVOCs will not be affected at all, thus for a given amount of NOx reduction 348 the ozone increase may be more enhanced as compared to the increase in lockdown conditions with 349

same NOx reduction. Shifts to EV vehicles mimics the transport regulations completely as it creates a 350 similar condition with no emissions from vehicles exhaust. Thus, it will potentially create a similar 351 response as observed during the lockdown for a given reduction. In regard to climate response, 352

implementation of BS-VI and FGDs mimic partially as it only reduces the cSLCFs (SO2 and NOx) with 353 no concurrent reductions in wSLCFs. Thus, the net temperature increase in Figure 5 will be enhanced 354 as there will be no compensation from reduction in wSLCFs. Implementation of the PAT scheme, 355

renewable power generation and shifts to EV mimic the lockdown regulation and may result in similar 356

impacts. 357

Analogies discussed above can help prioritizing the future course of currently ongoing policies. 358

Besides the policies discussed above, there are several ongoing interventions proposed in other sectors 359

as well, such as policies to boost access to clean energy which include shifts to LPG cookstove (Pradhan for fired-clay brick production (Emission standards; (MoEFCC, 2018)). The sectors incorporating these 364 policies did not get regulated during the lockdown, while for the affected sectors the reductions exceed 365 the targets of the policies (Tibrewal and Venkataraman, 2020) within them. Thus, the observed air 366 quality and estimated climate impacts during lockdown closures can be interpreted as a sensitivity 367 scenario where in the interventions in certain sectors are implemented at a much higher potential than 368 those in others. In this case, policies pertaining to sectors such as power plants, industry and transport 369 attain their maximum potential before policies in residential, agricultural and brick industry sectors. 370

This implies, such counter-productive impacts will be felt if, a) policy targets in power plants, industry 371

and transport are strengthened before those in residential, agricultural and brick industry sectors or b) 372

the efficacy in policy implementation in the latter sectors are not monitored. Urgency lies in the fact 373 that both cases (a) and (b) are very likely to happen simultaneously. The former sectors, dominant 374 emitters of CO2 and air pollutants (PM2.5, SO2 and NOx), will continue to experience strengthening of 375 decarbonisation measures and emissions regulations to keep up with global climate agenda and control 376 the mortality attributable to ambient pollution respectively. Additionally, poor policy formulation and 377 unfeasibility in cleaner alternatives (Tibrewal and Venkataraman, 2020) will prevent the policies in the 378 latter sectors realise their true objectives. This relative pacing among sectoral policies is crucial as 379

interventions in the residential, agricultural and brick industry sectors will control emissions of black 380 carbon (a highly potent global warming agent) while simultaneously providing relief to indoor and 381 ambient air pollution. Setting the pace of policies delivering such co-benefits is highly essential for 382 developing nations like India, where the co-benefits approach seems to be the most viable way to drive 383 climate change mitigation plans (Dubash et al., 2013) . 384

The present paper models the evolution of emissions of CO2 and SLCFs during the COVID-19 386 lockdown in India. By comparing against a baseline with no lockdown, we report the reduction in 387 emissions for these pollutants in response to closure of major economic activities pertaining to transport, 388 power plants and industry sectors. To the best of our knowledge, this is the first exclusive study to report 389 emissions evolution and changes during COVID-19 lockdown in India. The lockdown imposed an 390

unprecedented scale of activity closure in the transport sector followed by industry, power plants and 391 refineries, while residential and agricultural activities continued to operate. This proved to be a highly 392

unbalanced emissions control scenario where in certain pollutants (CO2, SO2, NOx and CO and primary 393 PM2.5) saw reductions as high as -75% while the others (BC, CH4, NMVOCs and OC) were limited 394 below -30%. Therefore, while prima facie it appears promising, there are counterproductive 395 implications on regional air quality and climate. We link the disproportionate reductions in NOx and 396

NMVOCs to the observed increases in tropospheric O3 concentrations at various urban regions across 397

India. Further, we estimate that such disparate controls in cSLCFs and wSLCFs will lead to a net 398 increase in future temperature, adding to the existing rate of warming over the country. While it has 399 been established that policies mitigating an appropriate "basket of pollutants" across multiple sectors 400

would be more effective in delivering simultaneous benefits in air quality and climate, our study goes 401

beyond to provide evidence that the "relative pacing" among these policies in attaining their maximum 402 mitigation potential is also very crucial to prevent counter-productive impacts.

Further, we believe the framework and analyses discussed in this paper hold international 405 relevance. Our study provides a high resolution spatio-temporally varying emissions of all major global 406

warming pollutants (i.e., GHGs, constituents of primary fine particulate matter and precursor gases) 407

during the COVID-19 lockdown over India based on detailed bottom-up modelling, which can be 408 incorporated to improve global assessments. Secondly, a simple metric-based approach (as illustrated 409

here, for India) can provide a good first-level range of potential temperature response until extensive 410 climate model simulations or observations are available. Finally, we illustrated a framework to utilize 411 the lockdown closures as a tool to gauge the counter-productive impacts from differential pacing of 412 sectoral policies. 413 

Current and future global climate impacts resulting from COVID-19

Climate Impacts 449 of COVID-19 Induced Emission Changes

Analysis of daily, monthly, and annual burned 452 area using the fourth-generation global fire emissions database (GFED4)

Enhanced secondary pollution 457 offset reduction of primary emissions during COVID-19 lockdown in China

India BHARAT STAGE VI emission standards

Global warming of 1.5°C: An IPCC special report on the impacts of global warming of context of strengthening the global response to the threat of climate change

Social and Travel Lockdown Impact Considering Coronavirus Disease 467 (COVID-19) on Air Quality in Megacities of India: Present Benefits, Future Challenges and 468 Way Forward

The Climate Response to Emissions Reductions Due to 477 COVID-19: Initial Results From CovidMIP

Global anthropogenic emissions of particulate matter including black carbon

OMI/Aura NO2 Cloud-Screened Total and Tropospheric Column 484 L3 Global Gridded 0.25 degree x 0.25 degree V3

Impact of lockdown measures during COVID-19 on air quality-A 488 case study of India

OMI/Aura Sulfur Dioxide (SO2) Total Column L3 1 day 495 Best Pixel in 0.25 degree x 0.25 degree V3

Feasibility and difficulties of China's new air 498 quality standard compliance: PRD case of PM2.5 and ozone from 2010 to 2025

Effect of lockdown amid COVID-19 pandemic on air quality 508 of the megacity Delhi

A global anthropogenic emission inventory of 512 atmospheric pollutants from sector-and fuel-specific sources (1970-2017): an application of the 513 Community Emissions Data System (CEDS)

Nitrogen oxides concentration and 517 emission change detection during COVID-19 restrictions in North India

Global tropospheric ozone responses to reduced NO x emissions linked to the COVID-19 521 worldwide lockdowns

National Policy for Management of Crop Residues. Ministry of Agriculture, Department 523 of Agriculture & Cooperation, Krishi Bhawan

The Gazette of India, Ministry of Environment Forests 526 and Climate change, Government of India

The Gazette of India, Ministry of Environment Forests 529 and Climate change, Government of India

Government of India issues Orders prescribing lockdown for containment of COVID-533 19 Epidemic in the country. Ministry of Home Affairs

Ministry of Home Affairs

Absorbing aerosols and high-538 temperature extremes in India: A general circulation modelling study

Pradhan Mantri Sahaj Bijli Har Ghar Yojana -Saubhagya

Monthly production. Minstry of Petroleum and Natural Gas, Government of India 543

Examining Effects of the COVID-19 National 548 Lockdown on Ambient Air Quality across Urban India

Trends in multi-pollutant emissions 551 from a technology-linked inventory for India: II. Residential, agricultural and informal industry 552 sectors

A short-term decline in 554 anthropogenic emission of CO 2 in India due to COVID-19 confinement

Trends in multi-pollutant emissions from a technology-linked 557 inventory for India: I. Industry and transport sectors

Temperature changes in India

Assessment of Climate 562 Change over the Indian Region: A Report of the Ministry of Earth Sciences (MoES), 563 Government of India

Modeling the relationship between carbon 566 emissions and environmental sustainability during COVID-19: a new evidence from asymmetric 567 ARDL cointegration approach

Effect of restricted emissions 570 during COVID-19 on air quality in India

Climate response to regional radiative forcing during the twentieth 573 century

Diurnal and temporal 575 changes in air pollution during COVID-19 strict lockdown over different regions of India

Evaluating the climate and air quality impacts 583 of short-lived pollutants

Amid lockdown, India switches to BS-VI emission norms

Climate co-benefits of air quality and clean energy policy in

Air Quality in India During COVID-19

pollution over India

Aggravating O3 pollution due to 595 NOx emission control in eastern China

speeches/detail/who-director-general-s-opening-remarks-at-the-media-briefing-on-600 covid-19

Quantifying the 602 emission changes and associated air quality impacts during the COVID-19 pandemic on the 603 North China Plain: a response modeling study

Fast Climate 606 Responses to Aerosol Emission Reductions During the COVID-19 Pandemic

609 Impact of reduced anthropogenic emissions during COVID-19 on air quality in India

Satellite-613 based estimates of decline and rebound in China's CO 2 emissions during COVID-19 pandemic

Sectoral emissions evolution and percentage reduction during COVID-19 lockdown periods 648 in India for (a) Primary PM2.5; Ozone precursors: (b) NOx and (c) NMVOCs

CO2, (e) warming SLCFs (wSLCFs) and (f) cooling SLCFs (cSLCFs)

Spatial 658 distribution of ratio of mean daily emissions of NMVOCs to NOx (ERNMVOCs/NOx) from 25 th March 659 2020 to 15 th April 2020 (period of first lockdown) under (a) BASE scenario and (b) COVD scenario. 660 (c) Correlation between ERNMVOCs/NOx in BASE and COVD. (d) Spatial distribution of

ERNMVOCs/NOx in BASE to COVD overlayed with sites where there was recorded increase in ambient 662 O3 concentrations (red circles). Spatial distribution in reductions and sectoral contribution of (e) NOx 663 and (f) NMVOCs emissions over India

Establishing links of COVID-19 closure to potential climate response. The stacked bars 667 represent the change in temperature from individual changes in SLCFs emissions using regional 668 temperature potential

White circles represent the net change in temperature by combining the effects of all SLCFs

 Tables:  672  673  Table 1 J o u r n a l P r e -p r o o f

☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:J o u r n a l P r e -p r o o f