Global Gridded Emission Inventoriesof â-HexachlorocyclohexaneY I - F A N L I , * , † M . T R E V O R S C H O L T Z , ‡ A N DB I L L J . V A N H E Y S T §

Meteorological Service of Canada, Environment Canada,4905 Dufferin Street, Downsview, Ontario M3H 5T4, Canada,Canadian ORTECH Environmental Inc., 2395 SpeakmanDrive, Mississauga, Ontario L5K 1B3, Canada, and School ofEngineering, University of Guelph, 50 Stone Road, Guelph,Ontario N1G 2W1, Canada

As the contamination of â-hexachlorocyclohexane(â-HCH) in the Arctic air, seawater, and biota is anenvironmental concern, there is a need for emissioninventories of â-HCH to be used by the modeling communityto predict the fate and transport of â-HCH. This paperpresents such emission inventories for â-HCH. The totalglobal usage of â-HCH between 1945 and 2000 is estimatedat 850 kt, 230 kt of which was emitted to the atmosphereover the same time period. Usage of â-HCH was estimatedto be around 36 kt in 1980 and 7.4 kt in 1990. Totalâ-HCH emissions in 1980 were 9.8 kt with 83% attributedto the application in 1980 and 17% to soil residues due to priorapplications. Total â-HCH emissions in 1990 were 2.4 ktwith 78% attributed to the application in 1990 and 22% tosoil residues. While it assumed that no usage of technicalHCH occurred in 2000, the global â-HCH emissions inthis year due to soil residues were estimated at 66 t.It has shown that the global â-HCH emissions haveundergone a “southward tilt” over the time periodstudied as more northern countries have banned theuse of technical HCH.

IntroductionTechnical HCH, a popular insecticide formulation prior tothe 1990s, contains only 10-12% of the active isomer γ-HCHand is predominantly made up of noninsecticidal R-iso-mer (60-70%), â-isomer (5-12%), δ-isomer (6-10%), andε-isomer (3-4%) (1). About 10 Mt of technical HCH wasapplied between 1948 and 1997 (2).

R-HCH, a prominent contaminant measured in theatmosphere, has been one of the most studied organochlor-ine pesticides (summarized in ref 3). Consequently. thephysical and chemical properties of R-HCH are relativelywell-known. This knowledge, coupled with globally griddedR-HCH usage and emission inventories resolved to 1° × 1°longitude and latitude (4-6), has provided the basis for severalmodels to reproduce the major features of the globaldistribution of this chemical in the different environmentalcompartments and to describe the overall global fate ofR-HCH (7-10).

In recent years, the contamination of the Arctic environ-ment by â-HCH has also attracted scientific attention. For

example, air concentrations of â-HCH in the Alert and Tagishstations in the Arctic have been measured regularly (11, 12).The results indicate that the concentration of â-HCH in theArctic atmosphere is very low in comparison with the morevolatile R-HCH and γ-HCH. The concentration of â-HCH inthe Arctic surface water, however, can be as high as 240 pg/L, thus approaching the concentration of γ-HCH in the samemedia (13). Although residues in marine mammals vary, someseal and whale species show preferential accumulation ofâ-HCH, while others contain predominantly R-HCH withlesser amounts of â- and γ-HCHs (14).

The observations noted above indicate the possibility thatâ-HCH is transported to the Arctic via different transportmechanisms. The presence of the more volatile R- and γ-HCHisomers in the high Arctic air and seawater indicates thatlong-range transport (LRT) is a primary pathway into theArctic. The detection of â-HCH in soils, sediments, humanblood, and human breast milk in the regions of application,however, is more indicative of local technical HCH con-tamination. Recently, Li et al. (13) reported that, in contrastto R-HCH, â-HCH appears to be less subject to directatmospheric loading into the high Arctic as most of â-HCHstays in the source region after application. As with R-HCHand γ-HCH, a portion of the applied â-HCH enters the Arctic,probably by mechanisms involving wet deposition or par-titioning into the North Pacific surface water and subse-quently entering the Arctic in ocean currents passing throughthe Bering Strait. The diverging pathways of â-HCH fromR-HCH can be explained, in part, by differences in theirHenry’s law constant, which is about a factor of 20 lower forâ-HCH (13).

Out of all the isomers of technical HCH, â-HCH is themost persistent with respect to microbial degradation, andit has the lowest volatility. In addition, it may be the mosttoxicologically significant HCH isomer due to its highlypersistent nature in mammalian tissues and its estrogeniceffects in mammalian cells and fish (14). The residues oftotal HCH, and â-HCH in particular, have been found inhuman breast milk in various places around the world (2,15), thus indicating a potential route by which HCHs can betransferred from one generation to the next.

To investigate further the transport of â-HCH into theArctic environment, transport and transformation modelsthat have been proven successful in predicting R-HCHdistribution should be applied to â-HCH to gain furtherinsight into the fate, source contributions, and transportpathways on both regional and global scales. Before this canbe accomplished, the models require global emission in-ventories of â-HCH as initial inputs. This paper presentsthese emission inventories for 1980, 1990, and 2000.

Experimental SectionGridded â-HCH Usage. According to Li (2), global technicalHCH usage from 1948 to 1997 was estimated to be around10 million t, far more than the tonnage of any other pesticideused in history. It is assumed that there was no technicalHCH used in 2000. By using gridded cropland data as asurrogate for the distribution of HCH, gridded technical HCHusage inventories have been developed on a 1° × 1° longitudeand latitude grid system (4). In this paper, â-HCH usageinventories have been compiled from the technical HCHusage inventories by assuming that technical HCH contains9% â-HCH, and the data in Europe were taken from Breiviket al. (16). The total global usage of â-HCH between 1945 and1999 is thus estimated at 850 kt.

* Corresponding author phone: (416)739-4892; fax: (416)739-4288;e-mail: yi-fan.li@ec.gc.ca.

† Environment Canada.‡ Canadian ORTECH Environmental Inc.§ University of Guelph.

Environ. Sci. Technol. 2003, 37, 3493-3498

10.1021/es034157d CCC: $25.00 2003 American Chemical Society VOL. 37, NO. 16, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 3493Published on Web 07/09/2003

Global gridded â-HCH usage data on a 1° × 1° longitudeand latitude grid system for 1980 and 1990 are shown inFigure 1. The usage data for â-HCH, given in Figure 1a,indicates that intensive use of â-HCH in 1980 occurred inthe eastern and central parts of China, most parts of India,the northern part of Nigeria, some areas in Australia, andsome Western European countries. Other countries, such asCanada, the United States, and Japan, did not use thisinsecticide in 1980. Figure 1b shows that the main areas usingthis compound in 1990 were localized to regions of Indiaand Vietnam. By 1990, usage of technical HCH had ceasedin China and many Western European countries. It is assumedthat there was no usage of technical HCH anywhere in theworld in 2000.

Emission Factors. Emission factors for both R-HCH andγ-HCH have been previously presented (6, 17-19). Weeklyemission factors for â-HCH have been generated at a 1° ×1° longitude and latitude resolution using a pesticide emissionmodel (PEM) (20). Details of PEM are provided by Scholtzand co-workers (21, 22). The physical and chemical param-eters being used to calculate the emission factors of â-HCHare listed in Table 1.

In this paper, a simplified approach is taken for estimatingthe annual emission factor using the weekly emission factors(20) and by considering only two different emission factorscenarios: one for a spraying event (Fs) and one for a tillingevent (Ft). The definition of the emission factor is given as(6)

where the subscript k ) s corresponds to a spraying event

and k ) t corresponds to a tilling event. In the case of aspraying event, Vs is the current usage of pesticide as a spray,Es is the amount of pesticide emitted to the atmosphere dueto the spraying event, and the emission factor (Fs) is definedas the ratio of the emission (Es) to the usage (Vs). In the caseof a tilling event, Vt is the current usage of pesticide appliedusing a soil incorporation mode (if any) plus any pesticideresidues remaining in the soil due to applications in previousyears from all application modes. Et is the amount of pesticideemitted to the atmosphere, and the emission factor (Ft) isdefined as the ratio of the emission (Et) to the amount ofcurrent usage of pesticide applied by a soil treatment modeplus the residues in the soil (Vt). This definition is convenientfor considering the emissions due to several consecutive yearsof pesticide application (6).

Annual emission factors of â-HCH for a spraying eventand for a tilling event are given in Figure 2. These annualemission factors represent the potential for a geographicalregion to emit â-HCH if it were applied. In comparison withthe annual emission factors of R-HCH (2), annual emissionfactors of â-HCH for both spaying and tilling events are muchlower. The major reasons are the differences of the chemicalproperties between these two compounds. The Henry’s lawconstant (HLC) of â-HCH is much less than that of R-HCH,and its octanol-air partition coefficient, which governs soil-air exchange, is 20 times lower (13).

Emissions of â-HCH. Since both R- and â-HCH are con-tained only in technical HCH, a number of assumptionsused to calculate emissions of R-HCH (6) are also valid forestimating the emissions of â-HCH and are summarizedbelow.

The global fractional distribution of technical HCH amongdifferent application modes in different earth zones ispresented in Table 2. Only one plant/harvest cycle is assumedin a given year, although it is not uncommon to have two ormore plant/harvest cycles in a year.

The deposition of re-emitted HCHs on soil and watersurfaces after transport in the atmosphere has not beenconsidered in the emission calculations since in this studyit is assumed that the atmospheric concentrations of HCHsare negligible with respect to the soil-surface concentrations.Thus, only first-time (or primary) emissions are estimated inthis paper, and re-emission due to previously wet and drydeposited material is neglected.

The usage data, the annual emission factors, and theapplication modes have been distributed on a 1° × 1° latitudeand longitude grid system and have been combined toestimate â-HCH emissions to the atmosphere. As was donefor R-HCH (6), the resulting emission estimates not onlyaccount for the target year of application but also any residualpesticide in the soil due to application in the previous 15 yrfrom the target year.

The gridded global annual emissions for â-HCH are givenin Figures 3-5 for the target years of 1980, 1990, and 2000,respectively. Total global â-HCH emissions to the atmospherein 1980 were 9.8 kt, of which 82.8% was contributed by thecurrent use of â-HCH in 1980 and 17.2% by the residues dueto use of technical HCH since 1966. Total global â-HCHemissions in 1990 were 2.4 kt, of which 78% was contributedby the current use of â-HCH in 1990 and 22% by the residuesdue to use of technical HCH since 1976. All â-HCH emissionsin 2000 (66 t) were from residues due to use of technicalHCH since 1986, as it was assumed that there was no technicalHCH usage in 2000.

As expected, the distribution of â-HCH emissions due tothe current year’s application shown in Figure 3a for 1980and in Figure 4a for 1990 follows the same general patternas the usage data for 1980 given in Figure 1a and for 1990given in Figure 1b. However, the distribution of â-HCHemissions due to residues of â-HCH used in the previous 15

FIGURE 1. Global gridded â-HCH usage in (a) 1980 and (b) 1990 with1° × 1° longitude and latitude resolution. The total usage of â-HCHwas 36 kt in 1980 and 7.4 kt in 1990.

Fk )Ek

Vk(1)

3494 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 16, 2003

yr shown in Figure 3b for 1980 and in Figure 4b for 1990 issubstantially different. Comparing these two figures withFigure 1, it can be seen that although Japan and the UnitedStates banned the use of technical HCH in 1972 (27) and1976 (28), respectively, â-HCH residues remaining in the soilof these two countries resulted in significant emissions tothe atmosphere in 1980 (Figure 3b). Figure 4b indicates that,7 yr after China banned the use of technical HCH in 1983(29), the residual â-HCH in Chinese soil still producedemissions to the atmosphere in 1990. â-HCH emissions in

2000 shown in Figure 5 were mainly from India and someSouth Asia countries.

TABLE 1. Physical and Chemical Parameters of â-HCH

parameters units value ref

diffusivity in air m2/d 0.468 23diffusivity in water m2/d 0.474 × 10-4 23Koc (soil sorption coefficient) m3/kg 2.29 24KH (Henry’s law constant) Pa m3/mol 0.12 (20 °C) 25solubility kg/m3 1.0 × 10-4 (20 °C) 25vapor density kg/m3 4.45 × 10-9 (20 °C) 25soil half-life d 730 no data available in literature for â-HCH,

so value quoted is that of γ-HCH (26)atmospheric half-life d 15 18enthalpy of volatilization kJ/mol 57 26

FIGURE 2. Emission factors of â-HCH for (a) a spraying event and(b) a tilling event.

TABLE 2. Global Fractional Distribution of Technical HCHamong Different Application Modes (6)

earth zonesspraymode

seedtreatment

mode

soilincorporated

mode

between 30° S and 30° N 130° S to 45° S or 30° N to 45° N 0.85 0.1 0.05below 45° S or above 45° N 0.2 0.8

FIGURE 3. Gridded global annual emissions for â-HCH in 1980. (a)Annual â-HCH emissions due to current year application in 1980.(b) Annual â-HCH emissions due to residues resulting from use inthe preceding 15 yr. (c) Total global â-HCH emissions in 1980.

VOL. 37, NO. 16, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 3495

DiscussionsFigure 6 summarizes the regional contributions to the globalâ-HCH emissions in the Northern and Southern Hemi-spheres. Emissions in the South Polar region are zero in allthree target years. Figure 6 indicates that in 1980 ap-proximately 79% of the global â-HCH emissions originatedin the Northern Hemisphere mid-latitudes followed byaround 16% in the Northern Hemisphere tropics. The regionalcontributions to the global â-HCH emissions in 1990 aresignificantly different. Figure 6 indicates that in 1990 ap-proximately 48% of the global â-HCH emissions originatedin the Northern Hemisphere tropics followed by around 44%in the Northern Hemisphere mid-latitudes. The greatestfactor in this change can be attributed to China banning theuse of technical HCH in 1983. In 2000, approximately 59%of the global â-HCH emissions originated in the NorthernHemisphere tropics followed by around 33% in the NorthernHemisphere mid-latitudes.

Continental contributions to global â-HCH emissions in1980, 1990, and 2000 are shown in Figure 7. Sources in Asiacontributed approximately 86% to the total global â-HCHemissions in 1980 followed by 6% from Europe, 3.4% fromAmerica, 2.6% from Africa, and 2% from Australia. In 1990,76% of â-HCH emissions originated in Asia, 7.7% fromEurope, 8.3% from Africa, 5% from America, and 3% fromAustralia and Oceania. In 2000, 70% of â-HCH emissionsoriginated in Asia, 10% from Europe, 12% from Africa, 5%from America, and 3% from Australia and Oceania. Dominant

FIGURE 4. Gridded global annual emissions for â-HCH in 1990. (a)Annual â-HCH emissions due to current year application in 1990.(b) Annual â-HCH emissions due to residues resulting from use inthe preceding 15 yr. (c) Total global â-HCH emissions in 1990.

FIGURE 5. Annual â-HCH emissions in 2000 due to residues resultingfrom use in the preceding 15 yr. It is assumed that there was notechnical HCH usage in 2000. The total emission was 66 t.

FIGURE 6. Regional contributions to the global â-HCH emissionsin 1980, 1990, and 2000 in the Northern and Southern Hemispheres.The tropical regions (NT and ST) are defined as the latitude bandfrom the equator to 23.5°; the mid-latitude regions (MN and SM) arebetween 23.5° and 60°; and the polar regions (NP and SP) are fromlatitude 60° to the poles. Emissions in the Southern Hemispherepolar region (SP) are zero and are not shown.

FIGURE 7. Continental contributions to the global â-HCH emissionsin 1980, 1990, and 2000.

3496 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 16, 2003

â-HCH emissions in Asia in both 1980 and 1990 are notsurprising since China and India were among the biggesttechnical HCH users in the world in 1980. India was still thelargest consuming country of technical HCH in 1990.

Figure 8 presents the latitudinal distribution of annualâ-HCH emissions. Emissions of â-HCH in 1980 are concen-trated between 20° and 50° N, while those in 1990 and 2000are concentrated between 0° and 45° N. The curves indicatethat, like R-HCH (6), the peak â-HCH emissions have movedfrom higher latitudes toward lower ones, thus undergoing a“southward tilt” (30) from 1980 to 2000 due to the more rapiddecrease of technical HCH usage in the more northerncountries.

Figure 9 depicts the long-term temporal trends of globalemissions of â-HCH along with R-HCH from 1945 to 2000.The trends of these two curves are similar with the exceptionof their scales. Although â-HCH contained in technical HCHis around 1/7th of R-HCH, the amount of emissions of â-HCHis only 1/20th of R-HCH. This is because of much lower valuesof emission factors for â-HCH than those for R-HCH. Theglobal emissions of R- and â-HCH increased steadily sincethe 1940s until the emissions reached a peak in the early1970s. The ensuing decrease in emissions is most likely dueto the ban of use of technical HCH in the United States (1976),Canada (1971), Japan (1972), and many European countries(2). Another peak happened in the early 1980 followed by asharp decrease due to the ban of technical HCH in China in1983.

As discussed in Li et al. (6), the major uncertainty of theemission data comes from the usage data of technical HCH.A qualitative uncertainty estimate has been assigned byregions according to the level of regional detail of the usagedata, and a scale of low, medium, and high has been used.A detailed discussion of this issue can be found in Li et al.(6).

AcknowledgmentsThis work was financially supported by Environment Canadaand Indian and Northern Affairs Canada through NorthernContaminants Program (NCP). Thanks go to S. Venkateshand K. Puckett of Environment Canada for their encourage-ments. This research is a contribution to the Global EmissionsInventory Activity (GEIA), a component of the InternationalGlobal Atmospheric Chemistry core project of the Interna-tional Geosphere-Biosphere Program. The global griddedemissions data sets of â-HCH on a 1° × 1° longitude andlatitude grid system for 1980, 1990, and 2000 are freelyavailable at http://www.msc.ec.gc.ca/data/gloperd/.

Literature Cited(1) Iwata, H.; Tanabe, S.; Sakai, N.; Tatsukawa, R. Environ. Sci.

Technol. 1993, 27, 1080.(2) Li, Y. F. Sci. Total Environ. 1999, 232, 123.(3) Macdonald, R. W.; Barrie, L. A.; Bidleman, T. F.; Diamond, M.

L.; Gregor, D. J.; Semkin, R. G.; Strachan, W. M. J.; Li, Y. F.;Wania, F.; Alaee, M.; Backus, S.; Bewers, M.; Gobeil, C.; Halsall,C.; Hoff, J.; Lockhart, L.; Mackay, D.; Muir, D.; Pudykiewicz, J.;Reimer, K.; Smith, J.; Stern, G.; Schroeder, W.; Wagemann, R.;Yunker, M. Sci. Total Environ. 2000, 254, 93.

(4) Li, Y. F. J. Geophys. Res. 1999, 104 (D19), 23,785.(5) Li, Y. F.; McMillan, A.; Scholtz, M. T. Environ. Sci. Technol. 1996,

30, 3525.(6) Li, Y. F.; Scholtz, M. T.; van Heyst, B. J. J. Geophys. Res. 2000,

105 (D5), 6621.(7) Pudykiewicz, J. Atmos. Environ. 1998, 32, 3039.(8) Koziol, A. S.; Pudykiewicz, J. A. Chemosphere 2001, 45, 1181.(9) Wania, F.; Mackay, D. Environ. Toxicol. Chem. 1999, 18, 1400.

(10) Wania, F.; Mackay, D.; Li, Y. F.; Bidleman, T. F.; Strand, A.Environ. Toxicol. Chem. 1999, 18, 1390.

(11) Halsall, C. J.; Bailey, R.; Stern, G. A.; Barrie, L. A.; Fellin, P.; Muir,D. C. G.; Rosenberg, B.; Rovinsky, F. Ya.; Kononov, E. Ya.;Pastukhov, B. Environ. Pollut. 1998, 102, 51.

(12) Hung, H.; Halsall, C. J.; Blanchard, P.; Li, H. H.; Fellin, P.; Stern,G.; Rosenberg, B. Environ. Sci. Technol. 2002, 36, 862.

(13) Li, Y. F.; Macdonald, R. W.; Jantunen, L. M. M.; Harner, T.;Bidleman T. F.; Strachan, W. M. J. Sci. Total Environ. 2002, 291,229.

(14) Willett, K. L.; Ulrich, E. M.; Hites, R. A. Environ. Sci. Technol.1998, 32, 2197.

(15) Jensen, A. A. In Chemical Contaminants in Human Milk; Jensen,A. A., Slorach, S. A., Eds.; CRC Press: Boston, 1991; Chapter 5.

(16) Breivik, K.; Pacyna, J. M.; Munch, J. Sci. Total Environ. 1999,239, 151.

(17) Berdowski, J. J. M.; Baas, J.; Bloos, J. P. J.; Visschedijk, A. J. H.;Zandveld, P. Y. J. The European emission inventory of heavymetals and persistent organic pollutants for 1990; TechnicalReport UBA-FB; TNO Institute of Environmental Sciences:Apeldoom, The Netherlands, 1997.

(18) Scholtz, M. T.; McMillan, A. C.; Slama, C.; Li, Y. F.; Ting, N.;Davidson, K. Pesticide emissions modelling: development of aNorth American pesticide emissions inventory; Technical ReportCGEIC-1997-1; Canadian Global Emissions Interpretation Cen-tre: Toronto, Canada, 1997.

(19) Scholtz, M. T.; McMillan, A. C.; Li, Y. F.; Ting, N.; Taylor, A.; VanHeyst, B. J. Global Pesticide emissions: development of pesticideemission factors on a global 1° × 1° longitude-latitude grid andglobal emissions of hexachlorocyclohexane; Technical ReportCGEIC-1998-2; Canadian Global Emissions Interpretation Cen-tre: Toronto, Canada, 1998.

(20) Scholtz, M. T.; Van Heyst, B. J. Global â-HCH emission factorswith 1° × 1° longitude/latitude resolution; Technical Report toEnvironment Canada: Toronto, Canada, 2001.

(21) Scholtz, M. T.; Voldner, E.; McMillan, A. C.; Van Heyst, B. J.Atmos. Environ. 2002, 36, 5005.

(22) Scholtz, M. T.; Voldner, E.; Van Heyst, B. J.; McMillan A. C.;Pattey, E. Atmos. Environ. 2002, 36, 5014.

(23) Sherwood, T. K.; Pigford, R. L.; Wilke, C. R. Mass Transfer;McGraw-Hill Inc.: New York, 1975; p 78.

(24) Mackay, D.; Shiu, W.-Y.; Ma, K.-C. Illustrated handbook ofphysical-chemical properties and environmental fate for organicchemicals, Vol. V, Pesticide Chemicals; Lewis Publishers: BocaRaton, FL, 1997; p 1223.

FIGURE 8. Latitudinal distribution of annual â-HCH emissions in1980, 1990, and 2000.

FIGURE 9. Annual global emissions of â-HCH and r-HCH between1945 and 2000.

VOL. 37, NO. 16, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 3497

(25) Suntio, L. R.; Shiu, W. Y.; Mackay, D.; Seiber, J. N.; Glotfelty, D.Rev. Environ. Contamin. Toxicol. 1988, 103, 1.

(26) Harner, T.; Bidleman, T. F.; Jantunen, L. M. M.; Mackay, D.Environ. Toxicol. Chem. 2001, 20, 1612.

(27) Loganathan, B. G.; Tanabe, S.; Goto, M.; Tatsukawa, R. Environ.Pollut. 1989, 62, 237.

(28) U.S. Environmental Protection Agency. Environ. News 1976,November 11.

(29) Chinese Ministry of Agriculture. A new pesticide manual;Agriculture Publisher: Beijing, 1989; p 688 (in Chinese).

(30) Goldberg, E. D. Proc. R. Soc. London, Ser. B 1975, 189, 277.

Received for review February 20, 2003. Revised manuscriptreceived May 9, 2003. Accepted May 23, 2003.

ES034157D

3498 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 16, 2003