Global Emissions of Trace Gases, Particulate Matter, and Hazardous


Global Emissions of Trace Gases, Particulate Matter, and Hazardous...

0 downloads 86 Views 6MB Size

Article pubs.acs.org/est

Global Emissions of Trace Gases, Particulate Matter, and Hazardous Air Pollutants from Open Burning of Domestic Waste Christine Wiedinmyer,*,† Robert J. Yokelson,‡ and Brian K. Gullett§ †

National Center for Atmospheric Research, Boulder, Colorado 80301, United States Department of Chemistry and Biochemistry, University of Montana, Missoula, Montana 59812, United States § U.S. Environmental Protection Agency, Office of Research and Development, Research Triangle Park, North Carolina 27711, United States ‡

S Supporting Information *

ABSTRACT: The open burning of waste, whether at individual residences, businesses, or dump sites, is a large source of air pollutants. These emissions, however, are not included in many current emission inventories used for chemistry and climate modeling applications. This paper presents the first comprehensive and consistent estimates of the global emissions of greenhouse gases, particulate matter, reactive trace gases, and toxic compounds from open waste burning. Global emissions of CO2 from open waste burning are relatively small compared to total anthropogenic CO2; however, regional CO2 emissions, particularly in many developing countries in Asia and Africa, are substantial. Further, emissions of reactive trace gases and particulate matter from open waste burning are more significant on regional scales. For example, the emissions of PM10 from open domestic waste burning in China is equivalent to 22% of China’s total reported anthropogenic PM10 emissions. The results of the emissions model presented here suggest that emissions of many air pollutants are significantly underestimated in current inventories because open waste burning is not included, consistent with studies that compare model results with available observations.



INTRODUCTION

removed to dump sites, it is not uncommon for the material to be burned by open, uncontrolled fires. The open burning of waste at both the residential level and at dump sites produces many atmospheric pollutants, including greenhouse gases (GHGs), reactive trace gases, particulate matter (PM), and toxic compounds. The emissions from waste combustors are reported in national and global inventories. However, the emissions from open waste burning at homes and dumps are more challenging to characterize and are commonly excluded from inventories. The difficulties associated with developing these emission estimates arise from unknown or unconstrained information associated with the activity factors (when, where, and how much burning occurs) and the emission factors (particularly, the representativeness of these values). Despite these challenges, some efforts have been made to estimate the emissions from open waste burning. The Intergovernmental Panel on Climate Change (IPCC) provides methods to estimate national-scale emissions of GHGs from the open burning of waste.4 Bond et al.5 have developed a global inventory of particulate black carbon (BC) and particulate organic carbon (OC) and include emissions from open waste combustion. Regional efforts also estimate the

Municipal solid waste includes household waste such as food waste, yard waste, containers and packaging, as well as waste produced from other industrial, commercial, and institutional sources.1,2 Overall, 1 to 2 billion metric tons of municipal solid waste are estimated to be produced globally per year.1,3 In the United States in 2011, approximately 2.0 kg of waste are produced per capita daily;2 however, waste production rates and composition vary widely depending on the country and region.1 Waste management practices include collection, recycling, composting, land filling/dumping, and incineration/burning. Burning waste may occur with technologies that promote “clean” efficient controlled combustion, capture energy, and/or mitigate emissions, or through open burning of waste without any emission mitigation efforts. Open burning can happen at the source of the waste production (e.g., at a residence) or at a collection site (e.g., dump). Residential, or domestic, open waste burning is a global occurrence that takes place in both developed and developing countries. In developed regions, domestic waste is typically collected; however, domestic waste burning does occur in rural areas where waste collection services are expensive, unavailable, or infrequent. In developing countries where waste collection services can be sparse, domestic burning of waste is a frequent disposal technique. Further, in many developing countries, even when waste is © XXXX American Chemical Society

Received: May 7, 2014 Revised: July 10, 2014 Accepted: July 14, 2014

A

dx.doi.org/10.1021/es502250z | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

status, urban versus rural population, and waste collection practices as follows. In developing countries, Pfrac is estimated as the fraction of total population whose waste is not collected plus the fraction of the population whose waste is collected and disposed of in open dumps that are burned. In this case, all rural populations are assumed not to have waste collection. In urban areas, waste that is not collected is assumed to be burnable. Further, collected waste may be available for burning in open dumps. In developed countries, the IPCC guidance4 suggests that Pfrac be assigned the rural population of a nation, unless the developed country’s population is greater than 80% urban, in which case it is then suggested that no open burning of waste occurs. When this guidance is followed, many countries, including the United States where there is known open burning of domestic waste, are assigned no emissions. To correct for this omission, Pfrac in developed countries is assumed to be the rural population, regardless of the overall urban percentage. Urban populations used for these calculations are provided in Table S1 of the Supporting Information and were downloaded from the World Bank (http://data.worldbank.org/indicator/ SP.URB.TOTL, accessed 06 September 2013). Urban areas here are defined by national statistical offices and are calculated using the urban ratios from the United Nations World Urbanization Prospects. Country-level economic status (“developed” versus “developing”) was assigned to each country based on the income levels assigned by the World Bank (http://data.worldbank.org/ country, accessed 18 September 2012). Those countries with income levels assigned as “Low Income” (LI), “Lower Middle Income” (LMI), and “Upper Middle Income” (UMI) are classified as developing countries. All others are assigned as developed. The values applied are available in the Supporting Information (Table S1). The waste burned residentially is estimated differently for developing and developed countries. For each country, reported municipal solid waste collection rates were used to determine the fraction of waste collected and sent to landfills or dump sites (Supporting Information, Table S1). For developing countries, the amount of residential waste burned (WBres) is calculated as

amount of air pollutant emissions from open waste burning. Fiedler6 has developed a Toolkit that provides guidance to develop country-level emission inventories of toxic polychlorinated dibenzo-p-dioxin (PCDD) and polychlorinated dibenzofuran (PCDF) releases from many sources including landfill fires and uncontrolled domestic waste burning. The U.S. Environmental Protection Agency (US EPA) provides guidance to estimate residential garbage burning in rural populations of the United States.7 Each of these efforts focuses on a few pollutants or categories of emissions, and a global, comprehensive, and consistent inventory including all emitted species measured from open waste burning does not exist. Many emissions inventories used for global model applications do not include emissions from open waste burning. The EDGAR v4.2 global emissions inventory,8 for instance, includes emissions estimates of trace gases and particles from waste incineration but not open waste burning. This paper presents the first comprehensive and consistent global inventory of greenhouse gases, reactive trace gases, particulate matter, and toxic emissions from the open combustion of waste. Global and national estimates are calculated, and the emissions are geospatially allocated for more accessible use in chemical and climate models. Many assumptions and inputs to the inventories lead to high uncertainties in the estimates; these uncertainties are discussed and the emission estimates here are compared to other available inventories. The importance of open waste combustion relative to other sources of atmospheric emissions is also considered.



METHODS National emission estimates from the open burning of waste were calculated. The IPCC4 provides guidelines for the estimation of GHG inventories that include open residential and dump waste burning (Chapters 2 and 5, http://www.ipccnggip.iges.or.jp/public/2006gl/vol5.html, accessed 03 January 2011). This general framework is followed to estimate national emission estimates from open burning of waste. Overall, the annual emission of compound i for each country (Ei) is calculated as the product of the amount of waste burned and an emission factor: Ei = WB × EFi

(1)

WBres = [(MSWP × Prural) + (MSWP × Purban × funcollected )]

where WB is the mass of waste burned and EFi is the emission factor, or mass of species i emitted per mass of waste burned. The total amount of waste burned (WB) is equal to the sum of the amount of domestic waste that is burned at individual residences (WBres) and the amount of waste burned at dumps (WBdump). WB does not include waste that is burned in incinerators or modern combustion systems. For each country, the amount of waste burned (WB) is estimated using the general guidelines from section 5.3.2 in the 2006 IPCC Guidelines for National GHG Inventories:4 WB = P × Pfrac × MSWP × Bfrac

× Bfrac

(3a)

where Prural is the rural population, Purban is the urban population, and f uncollected is the fraction of waste that is not collected. In developed countries, the waste collection rate is applied to the entire population; in developing countries, only the urban populations are assumed to have waste collection. In developed countries, the waste burned residentially is calculated as

(2)

WBres = (MSWP × Prural × funcollected ) × Bfrac

where P is the national population, Pfrac is the fraction of the population assumed to burn some of their waste, MSWP is the mass of annual per capita waste production, and Bfrac is the fraction of waste available to be burned that is actually burned. Values and references for P and MSWp used for the calculations here are available in the Supporting Information (Table S1). The fraction of a country’s population that is assumed to burn their waste (Pfrac) is assigned based on national income

(3b)

Only waste that is collected is available for burning in open dumps. No open burning in landfills or dump sites is assumed for developed countries. In developing countries, the amount of waste burned by open burning in dumps (WBdump) is calculated as WBdump = (MSWP × Purban × fcollected ) × Bfrac B

(4)

dx.doi.org/10.1021/es502250z | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

The IPCC recommends a default value for Bfrac of 0.6,4 representing 60% of the total waste available to be burned that is actually burned. Emissions are calculated following eq 1, where the total amount of waste burned is multiplied by an emission factor. Compiled values of emission factors from the literature are applied to calculate emission estimates of GHGs, reactive trace gases, particulate matter, and toxic compounds. Tables 1 and 2

emitted species (Table 1). When emission factors are reported in mass per carbon burned (e.g., Woodall et al.9), the values have been converted to mass of species emitted per mass of waste burned assuming the carbon content of dry waste is 45% based on reported values.3,10,11 Reported emission factors for polycholorinated and polybrominated dioxins and furans (PCDD/Fs and PBDD/Fs) are provided in Table 2; the emission factors for these hazardous pollutants from both residential and open dump burning are available and hence allow for separate calculations. National emission estimates were spatially allocated to a 0.1° horizontal grid to enable easier input of the emissions to chemistry and climate models. The emissions were allocated to the populations as described by the globally gridded maps of population for 2000 (GPW3) and an urban extents grid (GRUMPv1, 2013).16 In developing countries, the emissions are allocated to the populations; for developed countries, the emissions are spatially allocated only to rural populations. The emissions estimated using the methods described here are highly uncertain. A detailed discussion of the overall uncertainty is included in the Discussion below.

Table 1. Emission Factors (g kg−1 Waste Burned) for Species Emitted from the Burning of Wastea compound carbon dioxide (CO2) carbon monoxide (CO) methane (CH4) acetylene (C2H2) ethylene (C2H4) propylene (C3H6) methanol (CH3OH) formaldehyde (HCHO) acetic acid (CH3COOH) formic acid (HCOOH) hydrogen chloride (HCl) hydrogen cyanide (HCN) benzene (C6H6) ̂ total PAHb c NMOC (identified) NMOC (identified + unidentified) ammonia (NH3) sulfur dioxide (SO2) nitrogen oxides (NOx as NO) PM2.5 PM10 particulate black carbon (BC) particulate organic carbon (OC) mercury (Hg) PCBsd

emission factor

uncertainty

ref

1453 38 3.7 0.40 1.26 1.26 0.94 0.62 2.42 0.18 3.61 0.47 0.9 0.3 7.5 22.6

69 19 4.4 0.28 1.04 1.42 1.25 0.13 3.32 0.12 3.27 n/a 0.21 0.14 7.6 n/a

Akagi et al.12 Akagi et al.12 Akagi et al.12 Akagi et al.12 Akagi et al.12 Akagi et al.12 Akagi et al.12 Akagi et al.12 Akagi et al.12 Akagi et al.12 Akagi et al.12 Akagi et al.12 Woodall et al.9 Woodall et al.9 Akagi et al.12 Akagi et al.12

1.12 0.5 3.74

1.21 n/a 1.48

Akagi et al.12 Akagi et al.12 Akagi et al.12

9.8 11.9 0.65

5.7 n/a 0.27

Akagi et al.12 Woodall et al.9 Akagi et al.12

5.27

4.89

Akagi et al.12

2.1 × 10−04 1.3 × 10−04

n/a n/a

Chen et al.13 Lemieux et al.14



RESULTS Totals of waste production and open waste burning from both residential and dump burning were calculated by country (Figure 1; Supporting Information, Table S2). Globally, this method estimates that 2400 Tg of waste is produced annually, with 620 Tg burned openly at the residential level, and an additional 350 Tg burned openly at dump sites. The countries with the highest total waste production include China, United States, India, Japan, Brazil, and Germany. Those countries with the largest estimates for residential and open dump burning and therefore the largest emissions from open burning of waste are China, India, Brazil, Mexico, Pakistan, and Turkey. National annual emission estimates for all considered air pollutants were calculated (Supporting Information, Tables S3−S4) and spatially allocated to a 0.1° grid resolution (Figure 2 shows the national and gridded emissions for CO). The national and gridded emission estimates are available for download via http://bai.acd.ucar.edu/Data/fire/. Global annual total emissions of various GHGs, reactive trace gases, particulate matter, and toxic compounds are shown in Table 3.

a

Values for uncertainty (variability) are shown when available. Carbon burned is assumed to be 45% of the total mass of waste burned. n/a = not available. b̂ Polycyclic aromatic hydrocarbons. cNonmethane organic compounds. dPolychlorinated biphenyls.



DISCUSSION The annual global waste generation rate calculated using the methods described here is 2400 Tg yr−1. Although higher, this value compares reasonably well, within 25%, of other estimates. For example, Christian et al.3 estimated the global waste production to be on the order of 2000 Tg yr−1, and the Waste Atlas (http://www.atlas.d-waste.com/, accessed 15 April 2014) gives an estimate of 1900 Tg yr−1 for the current time period (@ 2013). The estimates of waste generation are dependent on the per capita waste generation rate and the population of each country. Each of these values has associated uncertainties, and differences in the reported values can lead to differences in waste production estimates. Overall, 970 Tg or 41% of the total waste generated is estimated to be treated via uncontrolled burning. The estimated global open waste burning emissions of longlived GHGs, such as carbon dioxide (CO2) and methane (CH4), are fairly small when compared to other anthropogenic sources (Table 3). Globally, the CO2 emitted from open waste

Table 2. Emission Factors (g Toxic Equivalents (TEQs) kg−1 Waste Burned) for Polychlorinated Dibenzodioxins/ Dibenzofurans (PCDD/Fs) and Polybrominated Dibenzodioxins/Dibenzofurans (PBDD/Fs)a residential wasteb PCDD/F TEQ (WHO 2005) PBDD/F TEQ (WHO 2005)

−07

1.22 × 10 9.00 × 10−09

open dump burningc 3.70 × 10−07 2.12 × 10−07

a

Emission factors reported have been converted from a per carbon burned basis to per total mass burned. Carbon burned is assumed to be 45% of the total mass of waste burned. bWoodall et al.9 cGullett et al.15

show the emission factors, references, and associated uncertainty (when available) for the emission factors used in this study. Emission factors for both residential and dump burning are available for dioxins (Table 2) but not for other C

dx.doi.org/10.1021/es502250z | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

Figure 1. Total estimated annual waste burned (Gg yr−1) at the residential level (A) and at dumps (B).

Lanka, open waste combustion emissions of CO2 are estimated to be more than each country’s total national CO2 emissions as reported by the United Nations (http://mdgs.un.org/unsd/ mdg/SeriesDetail.aspx?srid=749, downloaded 30 December 2013) (Supporting Information, Table S5). In Sri Lanka, the estimated open waste burning emissions of CO are approximately equal to the total estimated national CO emissions of the EDGARv4.2 inventory, and the open burning PM10 emission estimates are nearly five times greater than the national PM10 anthropogenic emissions (Supporting Information, Table S6). To illustrate the strength of open waste burning relative to total reported anthropogenic emissions, Figure 3 shows the ratio of open waste burning emissions of PM10 and CO to reported total anthropogenic emissions from the EDGARv4.2 inventory. The emissions of trace gases and particulate matter from open waste burning in countries with the highest overall emissions are significant. China has the largest emissions of any country in the EDGARv4.2 emissions inventory, which does not include emissions from open waste burning. The open waste burning emissions of CO, PM10 and NMOC calculated as part of this study are equivalent to 10%, 22%, and 9% of the total anthropogenic emissions for China (Supporting Information, Table S6). The results presented here suggest that current anthropogenic emission inventories of important atmospheric pollutants are underestimated substantially for many countries because open waste combustion emissions are excluded. On the basis of the measurements of waste burning in Mexico, Christian et al.3 estimate that global waste burning could emit between 6 and 9 Tg of HCl per year. The results from this study suggest a smaller amount (3.5 Tg). However,

burning is equivalent to 5% of the 2010 global annual anthropogenic emissions. Global annual CO2 emissions from open biomass burning of forest and grassland range from 5000 to 10000 Tg CO2:21 CO2 emission from open waste burning is equivalent to 12−25% of these biomass burning emissions. The contribution of open waste combustion to other pollutant emissions is rather substantial. When compared to the global emission inventories recently prepared for the Task Force on Hemispheric Transport of Air Pollution (HTAP; http://www.htap.org/, accessed 18 March 2014), the open waste combustion emissions of NMOC and carbon monoxide are each equivalent to 7 and 5% of the global total estimated anthropogenic emissions, respectively (Table 3). The HTAP inventory does not include the emissions of open waste burning. The global emissions of particulate matter (PM) from the open combustion of waste are very significant compared to the global anthropogenic emissions estimated for the HTAP inventory. The open waste combustion PM2.5 emissions are equivalent to 29% of the total global anthropogenic PM2.5 emissions, and the particulate organic carbon (OC) emissions estimated here are equivalent to 43% of the total global anthropogenic OC emissions. The mercury emissions estimated as part of this study are equivalent to 10% of the total emissions reported by the most recent assessment of the United Nations.20 A comparison of published national-level anthropogenic emission estimates to the open waste burning emissions produced by this inventory highlights the importance of open waste burning emissions of PM and gaseous pollutants. In several developing countries, particularly in Africa and Southeast Asia including Lesotho, Burundi, Mali, Somalia, and Sri D

dx.doi.org/10.1021/es502250z | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

Figure 2. Estimated annual emissions of carbon monoxide (CO, Gg yr−1) from the open combustion of waste at residences and dumps. National level emissions are shown on the top panel; emissions distributed to a 0.1° grid are shown on the bottom panel.

et al.5 assumed that no waste burning occurs in rural areas of developing countries. The differences in both magnitude and spatial allocation of these emissions could have significant consequences to modeled climate and chemical outcomes using these emissions. Residential household waste open burning is included in the 2011 National Emissions Inventory (NEI2011) compiled by the U.S. EPA (R. Huntley, personal communications, ftp://ftp. epa.gov/EmisInventory/2011nei/doc/, 30 January 2014). Table 4 compares the 2011 NEI emission factors and emission estimates for selected pollutants with the results for the US from this study. Differences between the two estimates result from the estimates of waste burned and the emission factors. The population and total waste generation estimates of the two methods are within 1%; however, the US EPA estimates 50% more waste is openly burned than the method described in this manuscript primarily due to assumptions made to predict per capita waste production, rural population distributions, and fraction of waste burned. Differences in the emission factors are the other primary reason for the discrepancies between the two inventories, particularly with the emissions of PM, NMOCs, and HCl (Table 4). The disparities between these two national inventories here emphasize the large uncertainty in the

the estimate presented here is nearly twice the estimate of 2 Tg yr−1 of Keene et al.,22 suggesting that HCl emitted from open waste burning is underpredicted in past inventories and is a more important source of HCl to the atmosphere than previously predicted. Likewise, Zhang et al.19 present a global polycyclic aromatic hydrocarbons (PAHs) inventory for 2004 that includes emissions from waste incineration and biomass burning but not open waste burning. The global total PAH emission in that inventory is 520 Gg yr−1. We calculate that the global PAH emissions from open waste burning is 334 Gg yr−1, suggesting that this current global PAH inventory is underestimated by as much as ∼40% due to the exclusion of open waste burning emissions. Comparing our estimates of emissions from waste combustion to other results is challenging, since other inventories are rare. Bond et al.5 estimate that, globally, 33 Tg of waste is burned openly per year, resulting in the emissions of 44 Gg BC yr−1 and 58 Gg OC yr−1. In this study, 972 Tg of waste is estimated to burn annually, resulting in annual emissions of 631 Gg BC and 5.1 Tg OC, a factor of 10 to 100 greater. In addition to using different assumed values for waste production, fraction burned, and emission factors, dissimilarities in the estimates may be present because Bond E

dx.doi.org/10.1021/es502250z | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

ambiguous, and in some regions, are rapidly changing as areas develop, populations move, and economies change. Only a few measurements have been made of emission factors and only for a few species and regions; those factors that are available are highly variable and are dependent on the composition of the waste burned as well as the burning conditions. For this study, we chose defensible inputs while acknowledging that large uncertainties are associated with the estimates. Values of uncertainty are not available for many of the inputs, and assumed values were assigned to each input to test the sensitivity of the results to each (Supporting Information, Table S7). In an effort to quantify the combined impacts of the input uncertainties, a Monte Carlo simulation of global CO emission estimates was completed to determine an overall estimate of uncertainty on the emission estimates from open waste burning. The emissions of CO were chosen for this example, although emissions of other species with more highly variable emissions factors such as PM2.5 or NMOC, will result in higher uncertainties. For this procedure, the uncertainty of the various model inputs (Supporting Information, Table S7) was assigned randomly for each country, and the CO emissions estimated. Overall, this analysis determined that the global annual estimated CO emissions can range from −57% to +67% from the average value calculated using the best guess inputs. From this analysis, the uncertainty in the estimated emission estimates in this paper is on the order of a factor of 2. Of the uncertainties associated with the estimates, many have to do with the waste composition and the way in which the waste was burned. Many emissions factors are reported as a function of carbon in the waste burned; therefore, an estimate of the carbon content from the waste composition must be considered. This value can be highly variable and vary depending on the locality. A globally or even regionally consistent value for waste carbon content is unavailable at this time, and values reported from other studies were used here. Further, measured emission factors are dependent on the combustion efficiency, or the extent to which the waste treated (by fire) completely combusts or oxidizes to CO2. The combustion efficiency is highly variable throughout the course of a burn and can vary significantly with waste composition, resulting in an emission factor that also varies, challenging determination of a representative value. Therefore, we are applying the average value of reported emission factors and provide the uncertainty range when available. Despite the uncertainties, the results here suggest that global estimates of emissions from open waste burning are substantial and current estimates of emissions are underestimating total emissions by omitting this source. This observation is consistent with recent modeling studies where simulations underpredicted concentrations of atmospheric pollutants. For example, Kopacz et al.23 applied satellite and aircraft observations with an adjoint of a global chemical transport model to constrain CO emissions. Their estimates of CO emissions from combustion were much higher than current inventories and suggested that vehicle cold starts and residential heating were underpredicted. Open waste burning could also be a missing source. Chemical transport models have been shown to underestimate the observed mixing ratios of CO in the lower troposphere in the Asian outflow.24 Inverse modeling of CO over the Beijing region resulted in regional estimates of CO emissions 50−150% higher than those emissions in the Regional Emission Inventory in Asia version 1.1,25 an inventory

Table 3. Annual Global Emissions of Various Species from the Burning of Waste and Total Global Anthropogenic Emissions of Various Pollutants species

open waste burning

carbon dioxide (CO2)

1413 Tg

31350 Tg

methane (CH4)

3.6 Tg

364 Tg

carbon monoxide (CO) NMOC (identified) NMOC (identified + unidentified) acetylene (C2H2) ethylene (C2H4) propylene (C3H6) methanol (CH3OH) formaldehyde (HCHO) acetic acid (CH3COOH) formic acid (HCOOH) hydrogen chloride (HCl) hydrogen cyanide (HCN) benzene total polycyclic aromatic hydrocarbons (PAHs) total polychlorinated biphenyls (PCBs) PCDD/F TEQ (WHO 2005) PBDD/F TEQ (WHO 2005) mecury (Hg)

37 Tg

554 Tg

UN MDG Indicators Dataa EDGAR v2.4 for 2008b HTAP v2 for 2008c

7.3 Tg 22 Tg

145 Tg

HTAP v2 for 2008c

389 Gg 1.2 Tg 1.2 Tg 914 Gg 603 Gg

2.5 Tg 6.5 Tg 2.7 Tg 4 Tg 1.2 Tg

RETRO RETRO RETRO Jacob et RETRO

2.4 Tg

9.7 Tgd

RETRO 200017

6−9 Tg

Christian et al.3

3.5 Tg 520 Gg

RETRO 200017 Zhang et al.19

204 Mg

1.0−4.1 Gg

1.1 Tg 486 Gg 3.6 Tg

47 Tg 109 Tg 113 Tg

UN Mercury Assessment20 HTAP v2 for 2008c HTAP v2 for 2008c HTAP v2 for 2008c

10 Tg 12 Tg 632 Gg 5.1 Tg

34 Tg 51 Tg 5.5 Tg 12 Tg

HTAP HTAP HTAP HTAP

ammonia (NH3) sulfur dioxide (SO2) nitrogen oxides (NOx as NO) PM2.5 PM10 BC OC

total reported anthropogenic

ref

200017 200017 200017 al.18 200017

175 Gg 3.5 Tg 457 Gg 875 Gg 334 Gg

123 Mg 206 kg 80 kg

v2 v2 v2 v2

for for for for

2008c 2008c 2008c 2008c

a

Downloaded from http://mdgs.un.org/unsd/mdg/SeriesDetail. aspx?srid=749, December 30, 2013. bDownloaded from http:// edgar.jrc.ec.europa.eu/overview.php?v=42, December 30, 2013. c Downloaded from http://edgar.jrc.ec.europa.eu/htap_v2/index. php?SECURE=123, December 30, 2013. dSum of all acids in the inventory.

emission estimates and, due to the size and therefore importance of this source, highlight the need to better constrain the estimates in future studies, with particular attention to emission factors and estimates of waste burned. The emission estimates described herein contain relatively high uncertainty compared to many other source categories. The inputs, including the socioeconomic drivers such as population totals, assignment of developed/developing countries, and urban and rural populations, have associated uncertainties. Further, rates of waste production and the fraction of population that burns waste are also highly F

dx.doi.org/10.1021/es502250z | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

Figure 3. Ratio of the emissions of PM10 (top) and CO (bottom) from open waste burning to total reported anthropogenic emissions from the EDGARv4.2 inventory.

Although many sources may be underestimated in current model applications, a large emissions source from open waste burning currently not included in inventories would be consistent with these results. In the future, chemical and climate models ought to include these emissions as part of the inventories. The emissions presented here are also very important when considering the health outcomes of populations near the source. Many of the emissions from open burning are toxic air pollutants and can cause detrimental health effects. The results here suggest a much stronger source strength of many of these hazardous air pollutants than previously considered. The relative strength of these open burning sources should encourage more structured waste management strategies, particularly as developing countries continue to progress.

Table 4. Emission Factors and Annual Estimates of Various Emissions from Open Waste Burning (Gg yr−1) in the United States from the U.S. EPA 2011 National Emissions Inventory (NEI2011) (SCC Code 2610030000) and This Study emission factors (g kg−1 waste burned)

annual emissions (Gg)

pollutant

NEI2011

this study

NEI2011

this study

carbon monoxide (CO) PM10 PM2.5 NMOC (identified) nitrogen oxides (NOx as NO) benzene hydrogen cyanide (HCN) sulfur dioxide (SO2) hydrogen chloride (HCl)

42 19 17 4 3 1.2 0.5 0.5 0.28

38 11.9 9.8 7.5 3.7 0.9 0.47 0.5 3.6

187 84 77 19 13 5.5 2.1 2 1.3

110 35 28 22 11 2.6 1.4 1.5 10



ASSOCIATED CONTENT

S Supporting Information *

This material is available free of charge via the Internet at http://pubs.acs.org.



that does not include open waste burning.26 Seethala et al.27 found that a regional chemical transport model consistently underpredicted particle concentrations in tropical India and suggested that emissions could be underestimated in the region. These and other studies suggest an underprediction in emissions of gases and particles in current model applications.

AUTHOR INFORMATION

Corresponding Author

*Phone: 303-497-1414; fax: 303-497-1400; e-mail: christin@ ucar.edu. G

dx.doi.org/10.1021/es502250z | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

Notes

(16) Balk, D. L.; Deichmann, U.; Yetman, G.; Pozzi, F.; Hay, S. I.; Nelson, A. Determining global population distribution: Methods, applications and data. Adv. Parasitol. 2006, 62, 119−156. (17) Schultz, M.; Rast, S.; van het Bolscher, M.; Pulles, T.; Brand, R.; Pereira, J.; Mota, B.; Spessa, A.; Dalsøren, S.; van Nojie, T.; Szopa, S. Emission Data Sets and Methodologies for Estimating Emissions; RETRO project report D1-6; European Union: Brussels, Belgium, 2007. (18) Jacob, D. J.; Field, B. D.; Li, Q. B.; Blake, D. R.; de Gouw, J.; Warneke, C.; Hansel, A.; Wisthaler, A.; Singh, H. B.; Guenther, A., Global budget of methanol: Constraints from atmospheric observations. J. Geophys. Res.-Atmos. 2005, 110, (D8). (19) Zhang, Y. X.; Tao, S. Global atmospheric emission inventory of polycyclic aromatic hydrocarbons (PAHs) for 2004. Atmos. Environ. 2009, 43 (4), 812−819. (20) UNEP Global Mercury Assessment 2013: Sources, Emissions, Releases and Environmental Transport; UNEP Chemicals Branch: Geneva, Switzerland, 2013. (21) Langmann, B.; Duncan, B.; Textor, C.; Trentmann, J.; van der Werf, G. R. Vegetation fire emissions and their impact on air pollution and climate. Atmos. Environ. 2009, 43 (1), 107−116. (22) Keene, W. C.; Khalil, M. A. K.; Erickson, D. J.; McCulloch, A.; Graedel, T. E.; Lobert, J. M.; Aucott, M. L.; Gong, S. L.; Harper, D. B.; Kleiman, G.; Midgley, P.; Moore, R. M.; Seuzaret, C.; Sturges, W. T.; Benkovitz, C. M.; Koropalov, V.; Barrie, L. A.; Li, Y. F. Composite global emissions of reactive chlorine from anthropogenic and natural sources: Reactive Chlorine Emissions Inventory. J. Geophys. Res.-Atmos. 1999, 104 (D7), 8429−8440. (23) Kopacz, M.; Jacob, D. J.; Fisher, J. A.; Logan, J. A.; Zhang, L.; Megretskaia, I. A.; Yantosca, R. M.; Singh, K.; Henze, D. K.; Burrows, J. P.; Buchwitz, M.; Khlystova, I.; McMillan, W. W.; Gille, J. C.; Edwards, D. P.; Eldering, A.; Thouret, V.; Nedelec, P. Global estimates of CO sources with high resolution by adjoint inversion of multiple satellite datasets (MOPITT, AIRS, SCIAMACHY, TES). Atmos. Chem. Phys. 2010, 10 (3), 855−876. (24) Allen, D.; Pickering, K.; Fox-Rabinovitz, M., Evaluation of pollutant outflow and CO sources during TRACE-P using modelcalculated, aircraft-based, and Measurements of Pollution in the Troposphere (MOPITT)-derived CO concentrations. J. Geophys. Res.Atmos. 2004, 109, (D15). (25) Ohara, T.; Akimoto, H.; Kurokawa, J.; Horii, N.; Yamaji, K.; Yan, X.; Hayasaka, T. An Asian emission inventory of anthropogenic emission sources for the period 1980−2020. Atmos. Chem. Phys. 2007, 7 (16), 4419−4444. (26) Tang, X.; Zhu, J.; Wang, Z. F.; Wang, M.; Gbaguidi, A.; Li, J.; Shao, M.; Tang, G. Q.; Ji, D. S. Inversion of CO emissions over Beijing and its surrounding areas with ensemble Kalman filter. Atmos. Environ. 2013, 81, 676−686. (27) Seethala, C.; Pandithurai, G.; Fast, J. D.; Polade, S. D.; Reddy, M. S.; Peckham, S. E. Evaluating WRF-chem multi-scale model in simulating aerosol radiative properties over the TropicsA case study over India. MapanJ. Metrol. Soc. India 2011, 26 (4), 269−284.

This document has been reviewed in accordance with U.S. Environmental Protection Agency policy and approved for publication. The views expressed in this article are those of the author[s] and do not necessarily represent the views or policies of the U.S. Environmental Protection Agency. The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors thank L. Emmons and R. Kumar for their thoughtful comments on the manuscript. The National Center for Atmospheric Research is operated by the University Corporation for Atmospheric Research under the sponsorship of the National Science Foundation.



REFERENCES

(1) Hoornweg, D.; Bhada-Tata, P.-T. What a Waste: A Global Review of Solid Waste Management; The Worldbank: Washington DC, 2012. (2) Facts and Figures.Municipal Solid Waste in the United States; EPA: Washington, DC, 2011; http://www.epa.gov/epawaste/nonhaz/ municipal/msw99.htm. (3) Christian, T. J.; Yokelson, R. J.; Cardenas, B.; Molina, L. T.; Engling, G.; Hsu, S. C. Trace gas and particle emissions from domestic and industrial biofuel use and garbage burning in central Mexico. Atmos. Chem. Phys. 2010, 10 (2), 565−584. (4) IPCC 2006 IPCC Guidelines for National Greenhouse Gas Inventories; National Greenhouse Gas Inventories Programme Japan; IPCC: Geneva, Switzerland, 2006. (5) Bond, T. C.; Streets, D. G.; Yarber, K. F.; Nelson, S. M.; Woo, J. H.; Klimont, Z., A technology-based global inventory of black and organic carbon emissions from combustion. J. Geophys. Res.-Atmos. 2004, 109, (D14). (6) Fiedler, H. National PCDD/PCDF release inventories under the Stockholm convention on persistent organic pollutants. Chemosphere 2007, 67 (9), S96−S108. (7) AP 42, 5th ed.; EPA: Washington, DC, 1995; Vol. I, Chapter 2. (8) Emission Database for Global Atmospheric Research (EDGAR), release version 4.2; European Commission, Joint Research Centre (JRC)/Netherlands Environmental Assessment Agency (PBL): The Hague, Netherlands, 2011. (9) Woodall, B. D.; Yamamoto, D. P.; Gullett, B. K.; Touati, A. Emissions from small-scale burns of simulated deployed u.s. military waste. Environ. Sci. Technol. 2012, 46 (20), 10997−11003. (10) Solorzano-Ochoa, G.; de la Rosa, D. A.; Maiz-Larralde, P.; Gullett, B. K.; Tabor, D. G.; Touati, A.; Wyrzykowska-Ceradini, B.; Fiedler, H.; Abel, T.; Carroll, W. F. Open burning of household waste: Effect of experimental condition on combustion quality and emission of PCDD, PCDF and PCB. Chemosphere 2012, 87 (9), 1003−1008. (11) Lundin, L.; Gullett, B.; Carroll, W. F., Jr.; Touati, A.; Marklund, S.; Fiedler, H. The effect of developing nations’ municipal waste composition on PCDD/PCDF emissions from open burning. Atmos. Environ. 2013, 79, 433−441. (12) Akagi, S. K.; Yokelson, R. J.; Wiedinmyer, C.; Alvarado, M. J.; Reid, J. S.; Karl, T.; Crounse, J. D.; Wennberg, P. O. Emission factors for open and domestic biomass burning for use in atmospheric models. Atmos. Chem. Phys. 2011, 11 (9), 4039−4072. (13) Chen, L. G.; Liu, M.; Fan, R. F.; Ma, S. X.; Xu, Z. C.; Ren, M. Z.; He, Q. S. Mercury speciation and emission from municipal solid waste incinerators in the Pearl River Delta, South China. Sci. Total Environ. 2013, 447, 396−402. (14) Lemieux, P. M.; Lutes, C. C.; Santoianni, D. A. Emissions of organic air toxics from open burning: A comprehensive review. Prog. Energy Combust. Sci. 2004, 30 (1), 1−32. (15) Gullett, B. K.; Tabor, D.; Bertrand, A.; Touati, A. Quality control for sampling of PCDD/PCDF emissions from open combustion sources. Chemosphere 2013, 93 (3), 494−498. H

dx.doi.org/10.1021/es502250z | Environ. Sci. Technol. XXXX, XXX, XXX−XXX