Evaluation of black carbon estimations in global aerosol models

Atmos Chem Phys 9 9001ndash9026 2009wwwatmos-chem-physnet990012009copy Author(s) 2009 This work is distributed underthe Creative Commons Attribution 30 License

AtmosphericChemistry

and Physics

Evaluation of black carbon estimations in global aerosol models

D Koch12 M Schulz3 S Kinne4 C McNaughton10 J R Spackman9 Y Balkanski3 S Bauer12 T Berntsen13T C Bond6 O Boucher14 M Chin15 A Clarke10 N De Luca24 F Dentener16 T Diehl17 O Dubovik14 R Easter18D W Fahey9 J Feichter4 D Fillmore22 S Freitag10 S Ghan18 P Ginoux19 S Gong20 L Horowitz 19T Iversen1327 A Kirkev ag27 Z Klimont 7 Y Kondo11 M Krol 12 X Liu 2318 R Miller 2 V Montanaro24N Moteki11 G Myhre1328 J E Penner23 J Perlwitz12 G Pitari24 S Reddy14 L Sahu11 H Sakamoto11G Schuster5 J P Schwarz9 Oslash Seland27 P Stier25 N Takegawa11 T Takemura26 C Textor3 J A van Aardenne8and Y Zhao21

1Columbia University New York NY USA2NASA GISS New York NY USA3Laboratoire des Sciences du Climat et de lrsquoEnvironnement Gif-sur-Yvette France4Max-Planck-Institut fur Meteorologie Hamburg Germany5NASA Langley Research Center Hampton Virginia USA6University of Illinois at Urbana-Champaign Urbana IL USA7International Institute for Applied Systems Analysis Laxenburg Austria8European Commission Institute for Environment and Sustainability Joint Research Centre Ispra Italy9NOAA Earth System Research Laboratory Chemical Sciences Division and Cooperative Institute for Research inEnvironmental Sciences University of Colorado Boulder Colorado USA10University of Hawaii at Manoa Honolulu Hawaii USA11RCAST University of Tokyo Japan12Meteorology and Air Quality Wageningen University Wageningen The Netherlands13University of Oslo Oslo Norway14Universite des Sciences et Technologies de Lille CNRS Villeneuve drsquoAscq France15NASA Goddard Space Flight Center Greenbelt MD USA16EC Joint Research Centre Institute for Environment and Sustainability Ispra Italy17University of Maryland Baltimore County Baltimore Maryland USA18Pacific Northwest National Laboratory Richland USA19NOAA Geophysical Fluid Dynamics Laboratory Princeton New Jersey USA20ARQM Meteorological Service Canada Toronto Canada21University of California ndash Davis CA USA22NCAR Boulder CO USA23University of Michigan Ann Arbor MI USA24Universita degli Studi LrsquoAquila Italy25Atmospheric Oceanic and Planetary Physics University of Oxford UK26Kyushu University Fukuoka Japan27Norwegian Meteorological Institute Oslo Norway28Center for International Climate and Environmental Research ndash Oslo (CICERO) Oslo Norway

Received 1 July 2009 ndash Published in Atmos Chem Phys Discuss 24 July 2009Revised 5 November 2009 ndash Accepted 10 November 2009 ndash Published 27 November 2009

Published by Copernicus Publications on behalf of the European Geosciences Union

9002 D Koch et al Evaluation of black carbon estimations in global aerosol models

Abstract We evaluate black carbon (BC) model predictionsfrom the AeroCom model intercomparison project by con-sidering the diversity among year 2000 model simulationsand comparing model predictions with available measure-ments These model-measurement intercomparisons includeBC surface and aircraft concentrations aerosol absorptionoptical depth (AAOD) retrievals from AERONET and OzoneMonitoring Instrument (OMI) and BC column estimationsbased on AERONET In regions other than Asia most mod-els are biased high compared to surface concentration mea-surements However compared with (column) AAOD or BCburden retreivals the models are generally biased low Theaverage ratio of model to retrieved AAOD is less than 07in South American and 06 in African biomass burning re-gions both of these regions lack surface concentration mea-surements In Asia the average model to observed ratio is07 for AAOD and 05 for BC surface concentrations Com-pared with aircraft measurements over the Americas at lati-tudes between 0 and 50N the average model is a factor of8 larger than observed and most models exceed the mea-sured BC standard deviation in the mid to upper troposphereAt higher latitudes the average model to aircraft BC ratio is04 and models underestimate the observed BC loading inthe lower and middle troposphere associated with springtimeArctic haze Low model bias for AAOD but overestimationof surface and upper atmospheric BC concentrations at lowerlatitudes suggests that most models are underestimating BCabsorption and should improve estimates for refractive in-dex particle size and optical effects of BC coating Retrievaluncertainties andor differences with model diagnostic treat-ment may also contribute to the model-measurement dispar-ity Largest AeroCom model diversity occurred in northernEurasia and the remote Arctic regions influenced by anthro-pogenic sources Changing emissions aging removal oroptical properties within a single model generated a smallerchange in model predictions than the range represented bythe full set of AeroCom models Upper tropospheric con-centrations of BC mass from the aircraft measurements aresuggested to provide a unique new benchmark to test scav-enging and vertical dispersion of BC in global models

1 Introduction

Black carbon the strongly light-absorbing portion of car-bonaceous aerosols is thought to contribute to global warm-ing since pre-industrial times It is a product of incom-plete combustion of fossil fuels and biofuels such as coalwood and diesel Black carbon (BC) has several effectson climate primarily warming but potentially also someamount of cooling The ldquodirect effectrdquo is the scattering

Correspondence toD Koch(dkochgissnasagov)

and absorption of incoming solar radiation by the BC sus-pended in the atmosphere The absorption warms the airwhere the BC aerosol is suspended but the extinction ofradiation results in negative forcing at the earthrsquos surface(eg Ramanathan and Carmichael 2008) The ldquoBC-albedoeffectrdquo occurs because black carbon deposited on snow low-ers the snow albedo and may therefore promote snow andice melting (eg Warren and Wiscombe 1980 Hansen andNazarenko 2004) BC may also have important effects onclouds by changing atmospheric stability andor relative hu-midity and thus affect cloud formation this has been termedthe ldquosemi-direct effectrdquo (eg Ackerman et al 2000 Johnsonet al 2004) Finally BC is a primary aerosol particle andinfluences the number of particles available for cloud con-densation (eg Oshima et al 2009) it may thus play an im-portant role for the aerosol cloud ldquoindirect effectrdquo BC mayalso affect the indirect effect by acting as ice nuclei (eg Co-zic et al 2007 Liu et al 2009)

Quantifying the effects of black carbon on climate changeis hindered by several uncertainties Emissions are uncertainbecause of difficulties quantifying sources and emission fac-tors (eg Bond et al 2004) Measurements of BC concentra-tions are uncertain because of instrumental limitations in thepresent measurement techniques (Andreae and Gelencser2006) Optical properties are uncertain since these vary withsource morphology particle age and chemical processingAtmospheric column aerosol absorption comes mostly fromblack carbon in many polluted and biomass burning regionsThis absorption aerosol optical depth (AAOD) has been re-trieved from satellite and an array of sunphotometer mea-surements and these retrievals also help to constrain columnBC However the constraint is limited by uncertainties andassumptions in the retrievals as well as by the fact that otherabsorbing species besides BC are present such as dust andorganic carbon Model simulation of BC is complicated byuncertainties in treatment of initial particle size and shape ap-propriate for initial release in a model gridbox particle up-take in liquid or frozen clouds and precipitation treatment ofmixing state and optical properties Assumptions influencingthe degree of internal vs external mixing with water-solubleparticles in the accumulation mode strongly influence theaerosol absorption (Seland et al 2008) and CCN-activationInternal mixing of BC also affects BC lifetime decreasing itrelative to hydrophobic BC (Ogren and Charlson 1983 Stieret al 2006 2007) Furthermore the BC model predictionsare subject to model uncertainties that apply to any chemicalmodel simulation such as the accuracy of the modelrsquos meteo-rology including transport clouds and precipitation (eg Liuet al 2007)

The aim of this study is to evaluate model-calculated BCin recent state-of-the-art global models with aerosol chem-istry and physics to consider their diversity and comparethem with available observations There has been concernthat some models may greatly underestimate BC absorptionand therefore BC contribution to climate warming (eg Sato

Atmos Chem Phys 9 9001ndash9026 2009 wwwatmos-chem-physnet990012009

D Koch et al Evaluation of black carbon estimations in global aerosol models 9003

et al 2003 Ramanathan and Carmichael 2008 Seland etal 2008) However it is unclear whether this is a problemcommon to all models whether the problem is regional orglobal and the extent to which the bias is due to BC massunderestimation possibly linked to emissions underestima-tion or to model treatment of optical properties leading tounderestimation of BC absorption We examine these issuesby comparing the models to a variety of measurements andworking with a large number of current models We also in-vestigate whether biases in some regions are more problem-atic than in others Finally we make use of one of the modelsthe GISS model (available to the first author of this paper)to consider the effects of changing BC emissions aging re-moval assumptions and optical properties We also use theGISS model to consider the seasonality of model bias andthe spectral dependence of AAOD bias

We compare the models with several types of observa-tions Model surface concentrations are compared with long-term surface concentration measurements Model BC con-centration profiles are compared with aircraft measurementsfor several recent aircraft campaigns spanning the NorthAmerican region from the tropics to the Arctic ColumnBC is assessed by comparing model AAOD with that re-trieved by Dubovik and Kings (2000) inversion algorithmfrom AERONET sunphotometer measurements (Holben etal 1998) as was done in Sato et al (2003) and with OMIsatellite retrievals of AAOD We also compare column bur-den of BC with the AERONET-based estimation as in Schus-ter et al (2005) While the measurements provide constraintsfor the models in the final section we will discuss measure-ment uncertainties and the discrepancies among them that areapparent as we apply them to the models

2 Model descriptions

21 AeroCom models

We evaluate seventeen models from the AeroCom aerosolmodel intercomparison an exercise that has been ongoingfor the past 5 years Model results as well as observa-tion datasets for validation purposes are available at the Ae-roCom website (httpnansenipsljussieufrAEROCOM)The AeroCom intercomparison exercises included an exer-cise ldquoArdquo with each model using its own emissions and anexercise ldquoBrdquo where all models used identical emissions andwere described in detail in Textor et al (2006 2007) Kinneet al (2006) and Schulz et al (2006) Here we work withexercise A unless only B is available for a particular modelin the database The models used year 2000 emissions andin some cases year 2000 meteorological fields Not all di-agnostics were available for all models so we used all thoseavailable for each quantity considered Many aspects of themodels have been evaluated in previous publications and werefer to those for general background information Textor et

al (2006) provided a first comparison of the models in ex-periment A and included basic information on the modelssuch as model resolution chemistry and removal assump-tions Textor et al (2007) described the exercise B modelintercomparison and showed that model diversity was notgreatly reduced by unifying emissions indicating that thegreatest model differences result from features such as me-teorology and aerosol treatments rather than from emissionsKinne et al (2006) discussed the aerosol optical propertiesof the models and Schulz et al (2006) presented the radiativeforcing estimates for the models

Some of the model features most relevant for the BC sim-ulations are provided in Table 1 As shown there nine dif-ferent BC energy emissions inventories and eleven differentbiomass burning emissions inventories were used The mod-els had a variety of schemes to determine black carbon agingfrom a fresh to aged particle where aged particles may beactivated into cloud water Ten models assumed that blackcarbon aged from hydrophobic to hydrophilic after a fixedlifetime five models had microphysical mixing schemes tomake the particles hydrophilic in one model the black andorganic carbon are assumed to be mixed when emitted andone model had fixed solubility In three cases the particlemixing affected opticalradiative properties A variety ofassumptions were made about how frozen clouds removedaerosols compared to liquid clouds ranging from identicaltreatments for frozen and liquid clouds to zero removal byice clouds Black carbon lifetime ranges from 49 to 114days

We note that the model versions evaluated here were sub-mitted to the database in year 2005 and many models haveevolved significantly since (eg Bauer et al 2008 Chin etal 2009 Ghan and Zaveri 2007 Liu et al 2005 2007Myhre et al 2009 Stier et al 2006 2007 Takemura et al2009) Thus this study provides a benchmark at the time ofthe 2005 submission

22 GISS model sensitivity studies

We use the GISS aerosol model to study sensitivity to fac-tors that could impact the BC simulation The GISS aerosolscheme used here includes mass of sulfate sea-salt (Kochet al 2006) carbonaceous aerosols (Koch et al 2007) anddust (Miller et al 2006 Cakmur et al 2006) The sensitiv-ity studies are described below and listed in Table 2 Allsimulations are performed and averaged for 3 years aftera 2-year model spin-up The standard GISS model versionfor these sensitivity studies is slightly different than the ver-sion in the AeroCom database This version does not includedust-nitrate interaction and does not include enhanced re-moval of BC by precipitating convective clouds as was in-cluded in the AeroCom-database GISS model version andtherefore has a somewhat larger BC load

wwwatmos-chem-physnet990012009 Atmos Chem Phys 9 9001ndash9026 2009


Table 1 AeroCom model black carbon characteristics

AeroCom models Energy BB Emis(1) Aging(2) BC Icesnow Mass median BC Refractive MABS References for aerosol moduleEmis(1) lifetime removal(3) diameter of density index at m2 gminus1(56)

days emitted g cmminus3(6) 550 nm(6)

particle(4)

GISS 99 B04 GFED A 72 12 008 16 156ndash05i 84 Koch et al (2006 2007)Miller et al (2006)

ARQM 99 C99 L00 I 67 T 01 15 41 Zhang et al (2001)L96 Gong et al (2003)

CAM C99 L96 A L X X X Collins et al (2006)

DLR CW96 CW96 I 5 008 075 FF X X X Ackermann et al (1998)accum 002strat 037 BB

GOCART C99 GFED A 66 T 0078 10 175ndash045i 100 Chin et al (2000 2002)D03 Ginoux et al (2001)

SPRINTARS NK06 NK06 BCOC L 00695 FF 125 175ndash044i 23 Takemura et al (2000 200201 others 2005)

LOA B B04 GFED A 73 LI 00118 10 175ndash045i 80 Boucher and Anderson (1995)Boucher et al (2002)Reddy and Boucher (2004)Guibert et al (2005)

LSCE G03 G03 A 75 L 014 16 175-044i 35 Claquin et al (1998 1999)(44 ) Guelle et al (1998a b 2000)

Smith and Harrison (1998)Balkanski et al (2003) Bauer etal (2004) Schulz et al (2006)

MATCH L96 L96 A L 01 X X X Barth et al (2000)Rasch et al (2000 2001)

MOZGN C99 M92 A L 01 10 175ndash044i 87 Tie et al (2001 2005)O96

MPIHAM D06 D06 I 49 S 0069 (FF BF) 20 175ndash044i 77 Stier et al (2005)0172 (BB)

MIRAGE C99 CW96 I 61 L 019 0025 17 19-06i 3 aitk Ghan et al (2001) Easter et alL00 6 acc (2004) Ghan and Easter (2006)

TM5 D06 D06 A 57 20 0034 16 175-044i 43 Metzger et al (2002a b)

UIOCTM C99 CW96 A 55 L 01 (FF) 10 155-044i 72 Grini et al (2005) Myhre et al0295 (2003) Berglen et al (2004)0852 (BB) Berntsen et al (2006)

UIOGCM 99 IPCC IPCC I 55 none 00236ndash04 20 20ndash10i 105 Iversen and Seland (2002)Kirkevag and Iversen (2002)Kirkevag et al (2005)

UMI L96 P93 N 58 L 01452 (FF) 15 180ndash05i 68 Liu and Penner (2002)0137 (BB)

ULAQ 99 IPCC IPCC A 114 L 002ndash032 10 207ndash06i 75 Pitari et al (1993 2002)

(1) BB = biomass burning B04 = Bond et al (2004) C99 = Cooke et al (1999) L00 = Lavoue et al (2000) L96 = Liouse et al (1996) CW= Cooke and Wilson (1996) GFED = Van der Werf et al (2003) NK06 = Nozawa and Kurokawa (2006) G03 = Generoso et al (2003)R05 = Reddy et al (2005) D03 = Duncan et al (2003) D06 = Dentener et al (2006) M92 = Mueller (1992) O96 = Olivier et al (1996)IPCC = IPCC-TAR (2000) P93 = Penner et al (1993)(2) Aging as it affects particle solubility A= aging with time I= aging by coagulation and condensation particles are internally mixedBCOC = BC assumed mixed with OC N= none indicates that mixingaging also affects particle optical properties(3) T= Temp dependence L= as liquid LI = As liquid for in-cloud removal only S= Stier et al (2005) is relative to water(4) FF = fossil fuel(5) MABS = BC mass absorption coefficient at 550 nm taken from Schulz et al (2006)(6) X models did not simulate optical properties



Table 2 GISS model sensitivity studies

Description Emission Burden Lifetime d AAOD x100Tg yrminus1 mg mminus2 550 nm

Standard run 72 (44 energy 036 92 055see text 28 biomass

burning)EDGAR32 75 037 93 058emissionIIASA emission 81 041 95 060BB 1998 82 038 87 0582x 72 029 76 050(Faster aging)2x 72 051 13 067(Slower aging)2x More ice-out 72 033 85 0522x Less ice-out 72 038 98 057Reff =01microm 72 035 91 047Reff =006microm 72 036 93 070

221 Emissions

The standard GISS model uses carbonaceous aerosol energyproduction emissions from Bond et al (2004) Biomassburning emissions are based on the Global Fire EmissionsDatabase (GFED) v2 model carbon estimates for the years1997ndash2006 (van der Werf et al 2003 2004) with the car-bonaceous aerosol emission factors from Andreae and Mer-let (2001) One sensitivity case had fossil and biofuel emis-sions from the Emission Database for Atmospheric Research(EDGAR V32FT2000 called ldquoEDGAR32rdquo below Olivieret al 2005) combined with emission factors from Bond etal (2004) and in a second those of the International Institutefor Applied Systems Analysis (IIASA) (Cofala et al 2007)In a third we used the largest biomass burning year from theGFED dataset 1998

222 Aging and removal

In the standard GISS model energy-related BC is assumed tobe hydrophobic initially and then ages to become hydrophilicwith an e-fold lifetime of 1 day Biomass burning BC is as-sumed to have 60 solubility so that if a cloud is present60 these aerosols are taken into the cloud water for eachhalf-hour cloud timestep One sensitivity test assigned ashorter lifetime with a halved e-folding time for energy BCand 80 solubility for biomass burning A second test as-sumes a longer lifetime with doubled e-folding time for en-ergy BC and 40 solubility for biomass burning

Treatment of BC solubility is particularly uncertain forfrozen clouds In our standard model BC-cloud uptake forfrozen clouds is 12 of that for liquid clouds A sensitivityrun allowed 24 ice-cloud BC uptake and another case 5

223 Aerosol size

The standard GISS model assumes the BC effective radius(cross section weighted radius over the size distributionHansen and Travis 1974) is 008microm One sensitivity caseincreased this to 01microm and another decreased it to 006micromThe size primarily affects the BC optical properties For BCsizes 01 008 and 006microm the model global mean BC massabsorption efficiencies are 62 84 and 124 and BC singlescattering albedos are 031 027 and 021

3 Model evaluation

31 Surface concentrations

Annual average BC surface concentration measurements areshown in the first panel of Fig 1 The data for the UnitedStates are from the IMPROVE network (1995ndash2001) thosefrom Europe are from the EMEP network (2002ndash2003)some Asian data from 2006 are from Zhang et al (2009)additional data mostly from the late 1990s are referenced inKoch et al (2007) These data are primarily elemental car-bon or refractory carbon which can be somewhat differentthan BC (Andreae and Gelencser 2006) The data were notscreened according to urban rural or remote environmentall available data were used however the IMPROVE sitesare generally in rural or remote locations There are broad re-gional tendencies with largest concentrations in Asia (1000ndash14 000 ng mminus3) then Europe (500ndash5000 ng mminus3) then theUnited States (100ndash500 ng mminus3) then high northern latitudes(10ndash100 ng mminus3) and least at remote locations (lt10 ng mminus3)

Figure 1 also shows BC surface concentrations from theGISS model sensitivity studies The biggest impact for



Observed BC surface concentration standard EDGAR

IIASA BB 1998 lifex2

life2 ice2 icex2

0

10

25

100

200

500

1000

5000

14500

Fig 1 Observed BC surface concentrations (upper left panel) and GISS sensitivity model results (annual mean ng mminus3)

remote regions comes from increasing BC lifetime either bydoubling the aging rate or by reducing the removal by iceDecreasing the BC lifetime has a smaller effect The larger1998 biomass burning emissions mostly increase BC in bo-real Northern Hemisphere and Mexico EDGAR32 emis-sions increase BC in Europe Arabia and northeastern AfricaIIASA emissions increase south Asian BC

Figure 2 show the AeroCom model simulations of BC sur-face concentration using model layer one from each modelFigure 2 also shows the average and standard deviations ofthe models The standard deviation distribution is similar tothe average Regions of especially large model uncertaintyoccur where the standard deviation equals or exceeds the av-erage such as the Arctic Overall the models capture theobserved distribution of BC ldquohot spotsrdquo SPRINTARS is theonly model that successfully captures the large BC concen-trations in Southeast Asia (Table 3) however it overestimatesBC in other regions Unfortunately there are no long-termmeasurements of BC in the Southern Hemisphere biomassburning regions

Table 3 shows the ratio of modeled to observed BC in re-gions where surface concentration observations are availableThe regional ratios are based on the ratio of annual meanmodel to annual mean observed for each site averaged over

each region Thirteen out of seventeen AeroCom modelsover-predict BC in Europe Sixteen of the models underes-timate Southeast Asian BC surface concentrations howevermost of these measurements are from 2004ndash2006 and emis-sions have probably increased significantly since the 1990s(Zhang et al 2009) Nine of the models overestimate re-mote BC in the United States about half the models over-estimate and half underestimate the observations Overallthe models do not underestimate BC relative to surface mea-surements None of the GISS model sensitivity studies showsignificant improvement over the standard case The longerlifetime cases improve the model-measurement agreement inpolluted regions but worsen the agreement in remote regions

32 Aerosol absorption optical depth

The aerosol absorption optical depth (AAOD) or the non-scattering part of the aerosol optical depth provides anothertest of model BC AAOD is an atmospheric column measureof particle absorption and so provides a different perspec-tive from the surface concentration measurements Both BCand dust absorb radiation so AAOD is most useful for test-ing BC in regions where its absorption dominates over dustabsorption Therefore we focus on regions where the dust



AVE ST DEV GISS CAM

GOCART UIO GCM SPRINTARS MPI HAM

MATCH UIO CTM ULAQ LOA

LSCE MOZART TM5 UMI

ARQM MIRAGE DLR

0 10 25 100 200 500 1000 5000 13000

Fig 2 AeroCom models annual mean BC surface concentrations (ngmminus3) First panel shows average second panel shows standard deviationof models

load is relatively small for example Africa south of the Sa-hara Desert However since some sites within these regionsstill have dust we work with model total AAOD includingall species

Figure 3 shows AERONET (1996ndash2006) sunphotometer(eg Dubovik et al 2000 Dubovik and King 2000) andOMI satellite (2005ndash2007 from OMAERUV product Tor-res et al 2007) retrievals of clear sky AAOD A scatterplot compares the AERONET and OMI retrievals at theAERONET sites Table 4 (last 5 rows) provides regional av-erage AAOD for these retrievals The two retrievals broadlyagree with one another However the OMI estimate is larger

than the AERONET value for South America and smaller forEurope and Southeast Asia

The AeroCom model AAOD simulations are in Fig 4 Thestandard deviation relative to the average is similar to the sur-face concentration result it is less than or equal to the aver-age except in parts of the Arctic Table 4 gives the averageratio of model to retrieved AAOD within regions For theratio of model to AERONET we average the model AAODover all AERONET sites within the region and divide by theaverage of the corresponding AERONET values For OMIwe average over each region in the model and divide by theOMI regional average The average model agrees with the



Table 3 Average ratio between model and observed BC surfaceconcentrations within regions for AeroCom models and GISS sen-sitivity studies Number of measurements is given for each regionBottom row is observed average concentration in ng mminus3 Regionsdefined as N Am (130 W to 70 W 20 N to 55 N) Europe (15 W to45 E 30 N to 70 N) Asia (100 E to 160 E 20 N to 70 N)

AeroCom models N Am Europe Asia Rest of26 16 23 World

12

GISS 081 065 043 24ARQM 029 049 012 055CAM 16 22 040 18DLR 30 31 037 14GOCART 12 21 048 12SPRINTARS 77 97 10 44LOA 089 12 023 050LSCE 061 30 043 081MATCH 13 30 025 10MIRAGE 12 17 032 076MOZGN 24 38 076 22MPIHAM 15 073 056 044TM5 18 10 076 12UIOCTM 072 16 037 041UIOGCM 088 29 053 17UMI 081 48 065 10ULAQ 075 30 082 22AeroCom Ave 16 26 050 14AeroCom Median 12 22 043 12GISSsensitivitystd 081 088 042 19r=1 082 090 041 19r=06 082 091 042 20EDGAR32 070 11 034 17IIASA 070 086 050 19BB1998 081 093 042 18Lifex2 088 098 043 29Life2 078 080 038 15Ice2 083 093 041 21Icex2 079 088 041 17Observed(ng mminus3) 290 1170 5880 750

retrievals in eastern North America and with AERONET inEurope (ratios of modeled to AERONET in these regions are086 and 081) it underestimates Asian (ratio is 067) andbiomass burning AAOD (about 05ndash07 for AERONET and04ndash05 for OMI)

AAOD depends not just on aerosol load but also on op-tical properties such as refractive index particle size den-sity and mixing state In Fig 3 we show how the GISSmodel AAOD changes with assumed effective radius Theglobal mean AAOD decreasesincreases 1527 for an in-creasedecrease of 002microm effective radius Note that the

AeroCom model initial particle diameters (Table 1) span be-yond this range (001 to 09microm) and in some cases growas the particles age Increasing particle density from 16to 18 g cmminus3 in the GISS model decreases AAOD about asmuch as increasing particle size from 008 to 01microm (calcu-lated but not shown) Thus the AAOD is highly sensitive tosmall changes in these optical properties

Note that models generally underestimate AAOD but notsurface concentration As we will discuss below this couldresult from inconsistencies in the measurements from modelunder-prediction of BC aloft or from under-prediction of ab-sorption In this connection most models in the 2005-versionof AeroCom did not properly describe internal mixing withscattering particles in the accumulation mode Such mixingincreases the absorption cross section of the aerosols com-pared to external mixtures of nucleation- and Aitken-modeBC particles

33 Wavelength-dependence

Black carbon absorption efficiency decreases less with in-creasing wavelength compared with dust or organic carbon(Bergstrom et al 2007) Therefore comparison of AAODwith retrievals at longer wavelength indicates the extent towhich BC is responsible for biases In Fig 5 we compareAERONET AAOD at 550 and 1000 nm with the GISS modelAAOD for the wavelength intervals 300ndash770 nm and 860ndash1250 nm respectively Table 5 shows the ratio of the GISSmodel to AERONET within source regions for 1000 nm and550 nm for three different BC effective radii In all regionsexcept Europe and Asia the ratio is even lower at the longerwavelength confirming the need for increased simulated BCabsorption rather than other absorbing aerosols that absorbrelatively less at longer wavelengths

34 Seasonality

Our analysis has considered only annual mean observed andmodeled BC Here we present the seasonality of observedAAOD compared with the GISS model to explore how thebias may vary with season Seasonal AAOD are shownfor AERONET OMI and the GISS model in Fig 6 Asin most of the models the GISS model BC energy emis-sions do not include seasonal variation Biomass burningemissions do and dust seasonality is also very pronouncedHowever transport and removal seasonal changes also causefluctuations in model industrial source regions Note thatmore AERONET data satisfy our inclusion criteria for the3 month means compared with annual means (see figure cap-tions) so coverage is better in some regions and seasons thanin the annual dataset in Fig 3 Table 4 (bottom 4 rows)gives regional seasonal retrieved mean AAOD The seasonalmodel-to-measurement ratios are also provided in the middleportion of Table 4



AAOD AER 550 AAOD OMI 500

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

+ North America

+ Europe

+ Southeast Asia

o South America

o South Africa

+ Other

standard r=008 055 r=01 047 r=006 070

00

01

02

05

10

20

30

40

50

60

200

Fig 3 Top Aerosol absorption optical depth AAOD (x100) from AERONET (at 550 nm upper left) OMI (at 500 nm upper right)middle scatter plot comparing OMI and AERONET at AERONET sites and bottom GISS sensitivity studies for effective radius 008 01and 006microm for 300ndash770 nm The AERONET data are for 1996ndash2006 v2 level 2 annual averages for each year were used ifgt8 monthswere present and monthly averages forgt10 days of measurements The values at 550 nm were determined using the 044 and 087micromAngstrom parameters The OMI retrieval is based on OMAERUVd003 daily products from 2005ndash2007 that were obtained through andaveraged using GIOVANNI (Acker and Leptoukh 2007)

Biomass burning seasonality with peaks in JJA for centralAfrica (OMI) and in SON for South America is simulated inthe model without clear change in bias with season In Asiaboth retrievals have maximum AAOD in MAM which themodel underestimates (ie ratio of model to observed is low-est in MAM) The MAM peak may be from agricultural orbiomass burning not underestimed by the model emissionsThe other industrial regions do not have apparent seasonal-ity However the model BC is underestimated most in Eu-rope during fall and winter suggesting excessive loss of BCor missing emissions during those seasons

Summertime pollution outflow from North America seemsto occur in both OMI and the model The large OMI AAODin the southern South Atlantic during MAM-JJA may be aretrieval artifact due to low sun-elevation angle andor sparsesampling however if it is real then the model seasonality inthis region is out of phase

35 Column BC load

Model simulation of column BC mass (Fig 7) in the atmo-sphere should be less diverse than the AAOD since it con-tains no assumptions about optical properties However thereis no direct measurement of BC load Schuster et al (2005)developed an algorithm to derive column BC mass fromAERONET data working with the non-dust AERONET cli-matologies defined by Dubovick et al (2002) The Schusteralgorithm uses the Maxwell Garnett effective medium ap-proximation to infer BC concentration and specific absorp-tion from the AERONET refractive index The MaxwellGarnett approximation assumes homogeneous mixtures ofsmall insoluble particles (BC) suspended in a solution ofscattering material Such mixing enhances the absorptivityof the BC Schuster et al (2005) estimated an average spe-cific absorption of about 10 m2 gminus1 a value larger than most



AVE ST DEV GISS GOCART

UIO GCM ARQM SPRINTARS MPI HAM

MIRAGE UIO CTM ULAQ LOA

LSCE MOZART TM5 UMI

00

01

02

05

10

20

30

40

50

60

200

Fig 4 Annual average AAOD (x100) for AeroCom models at 550 nm First panel is average second panel standard deviation

53

1216

Figure 5 Annual average AAOD (x100) at AERONET stations for 550 nm and 1000 nm (top 1217

left and right) and for the GISS model for 300-770nm (bottom left) and 860-1250nm (bottom 1218

right) 1219

Fig 5 Annual average AAOD (x100) at AERONET stations for 550 nm and 1000 nm (top left and right) and for the GISS model for300ndash770 nm (bottom left) and 860ndash1250 nm (bottom right)



Table 4 Average ratio of model to retrieved AERONET and OMI Aerosol Absorption Optical Depth at 550 nm within regions for AeroCommodels and GISS sensitivity studies Number of measurements is given for AERONET Annual and seasonal measurement values are givenin last 5 rows Regions defined as N Am (130 W to 70 W 20 N to 55 N) Europe (15 W to 45 E 30 N to 70 N) Asia (100 E to 160 E 30 N to70 N) S Am (85 W to 40 W 34 S to 2 S) Afr (20 W to 45 E 34 S to 2 S)

AER AER AER AER AER AER OMI OMI OMI OMI OMI OMIN Am Eur Asia S Am Afr Rest of N Am Eur Asia S Am Afr Rest of

44 41 11 7 5 World 40 World

GISS 10 083 049 059 035 088 073 14 074 029 040 028ARQM 079 036 030 042 025 044 050 061 040 022 023 019GOCART 14 15 14 072 079 096 079 25 14 034 072 046SPRINTARS 14 048 044 18 12 064 076 069 059 083 13 028LOA 057 056 042 044 070 044 032 095 044 025 048 018LSCE 042 055 048 020 018 034 029 11 051 011 021 016MOZGN 15 13 099 060 060 077 082 26 14 032 040 035MPIHAM 039 021 029 043 035 021 021 029 032 022 035 0082MIRAGE 073 055 049 076 078 042 035 091 048 041 058 020TM5 041 032 029 024 020 031 021 048 031 012 022 011UIOCTM 062 067 046 11 061 057 037 11 053 057 054 019UIOGCM 13 11 075 082 054 080 082 18 10 046 042 036UMI 032 029 029 021 021 022 017 044 028 0095 019 0086ULAQ 14 26 21 11 052 11 11 67 15 062 048 071Ave 086 081 067 068 053 055 052 16 071 035 047 026GISSsensitivitystudiesstd 10 083 049 059 035 053 073 14 074 029 040 028r=1 086 066 040 049 028 048 060 12 061 024 032 022r=06 14 11 068 077 047 061 10 18 10 038 053 038EDGAR32 11 081 046 057 035 058 075 14 073 028 041 029IIASA 12 085 057 059 036 055 082 15 090 029 041 032BB1998 11 081 051 067 040 055 080 14 084 031 045 030Lifex2 13 091 054 066 041 058 093 16 088 035 050 039Life2 093 073 046 058 033 052 065 13 067 028 037 023Ice2 11 083 051 062 036 052 081 15 079 029 041 031Icex2 096 074 048 058 034 052 068 13 071 031 039 024Std DJF 085 045 045 029 033 031 040 040 064 022 030 036Std MAM 096 086 051 041 038 046 095 12 060 021 033 038Std JJA 083 097 064 043 030 066 063 17 093 028 036 036Std SON 12 064 056 051 034 040 063 080 071 020 057 038Retrievedx100Annual 069 15 36 18 20 24 085 068 15 22 17 12AverageDJF 057 14 33 14 09 26 10 14 14 15 12 09MAM 079 16 40 10 08 24 072 097 22 14 082 14JJA 088 16 30 24 31 20 10 071 12 27 27 18SON 057 16 30 31 39 22 095 10 14 47 14 11

of the models (see Table 1) however a lower value would in-crease the retrieved burden and worsen the comparison withthe models

An updated version of the AERONET-derived BC col-umn mass is shown in Figs 7 and 8 For this retrieval aBC refractive index of 195ndash079i was assumed within the

range recommended by Bond and Bergstrom (2006) andBC density of 18 g cmminus3 In the retrievals most conti-nental regions have BC loadings between 1 and 5 mg mminus2with mean values for North America (18 mg mminus2) andEurope (21 mg mminus2) being somewhat smaller than Asia(30 mg mminus2) and South America (27 mg mminus2) The current



AAOD 550 DJF MAM JJA SON

OMI

model

00 01 02 05 10 20 30 40 50 60 200

Fig 6 Seasonal average AAOD (x100) for AERONET 550 nm (top) OMI 500 nm (middle) standard GISS model 550 nm (bottom)

Table 5 The average ratio of GISS model to AERONET within regions for 1000 nm and 550 nm

Effective AAOD AAOD AAOD AAOD AAOD AAODRadiusmicrom Nam 44 Eur 41 Asia 11 S Am 7 Afr 5 Rest 21

1000 nmStd r=008 085 087 055 042 028 054r=01 072 073 047 036 023 050r=006 11 11 073 055 036 061550 nmStd r=008 10 083 049 059 035 053r=01 086 066 040 049 028 048r=006 14 11 068 077 047 061

industrial region retrievals are larger than the previous esti-mates of Schuster et al (2005) which were 096 mg mminus2 forNorth America 14 mg mminus2 for Europe and 16 mg mminus2 forAsia The biomass burning estimates are similar to the previ-ous retrievals The differences may be due to the larger spanof years and sites in the current dataset

Figure 7 shows the AeroCom model BC column loadsThe model standard deviation relative to the average is sim-ilar to the surface concentration (Fig 2) and the AAOD(Fig 4) The model column loads are smaller than theSchuster estimate Some models agree quite well in Europe

Southeast Asia or Africa (eg GOCART SPRINTARSMOZGN LSCE UMI) Model to retrieved ratios within re-gions are presented in Table 6 This ratio is generally smallerthan model to retrieved AAOD in North America and Eu-rope The inconsistencies among the retrievals would benefitfrom detailed comparison with a model that includes particlemixing and with model diagnostic treatment harmonized tothe retrievals

Figure 8 has GISS BC column sensitivity study re-sults The load is affected differently than the surfaceconcentrations (Fig 1) The Asian IIASA emissions are



Schuster BC load AVE ST DEV GISS

CAM GOCART UIO GCM ARQM

SPRINTARS MPI HAM MATCH MIRAGE

UIO CTM ULAQ LOA DLR

LSCE MOZART TM5 UMI

00 01 02 05 10 20 50 100 130

mgm2

Fig 7 Annual mean column BC load for 9 AeroCom models mg mminus2 The Schuster BC load is based on AERONET v2 level 15 annualaverages require 12 months of data data include all AERONET up to 2008

larger than Bond (Bond et al 2004) or EDGAR32 so thatthe outflow across the Pacific is greater The large-biomassburning case (1998) also results in greater BC transport toNorthwestern US in the column Increasing BC lifetime in-creases both BC surface and column mass more than theother cases however it has a larger impact on SouthernHemisphere load than surface concentrations The reducedice-out case has somewhat smaller impact on the column thanat the surface especially for some parts of the Arctic Thereduced ice-out thus has an enhanced effect at low levels be-low ice-clouds in the Arctic while having a relatively smallimpact on the column Modest model improvements relative

to the retrieval occur for the case with increased lifetime andfor the IIASA emissions (Table 6)

36 Aircraft campaigns

We consider the BC model profiles in the vicinity of recentaircraft measurements in order to get a qualitative sense ofhow models perform in the mid-upper troposphere and tosee how model diversity changes aloft The measurementswere made with three independent Single Particle Soot ab-sorption Photometers (SP2s) (Schwarz et al 2006 Slowiket al 2007) onboard NASA and NOAA research aircraft attropical and middle latitudes (Fig 9) and at high latitudes



standard 036 EDGAR 037 IIASA 041

BB 1998 038 lifex2 051 life2 029

00

01

02

05

10

20

50

100

130mgm2

ice-out2 038 ice-outx2 033 Schuster BC load

Fig 8 Annual mean column BC load for GISS sensitivity simulations and the Schuster BC retrieval (see Fig 7)

(Fig 10) over North America Details for the campaignsare provided in Table 7 The SP2 instrument uses an intenselaser to heat the refractory component of individual aerosolsin the fine (or accumulation) mode to vaporization The de-tected thermal radiation is used to determine the black carbonmass of each particle (Schwarz et al 2006) The U Tokyoand the NOAA data have been adjusted 5ndash10 (70 dur-ing AVE-Houston) to account for the ldquotailrdquo of the BC massdistribution that extends to sizes smaller than the SP2 lowerlimit of detection This procedure has not been performedon the U Hawaii data however this instrument was config-ured to detect smaller particle sizes so that the unmeasuredmass is estimated to be less than about 13 (3 at smallerand 10 at larger sizes) The aircraft data in each panelof Figs 9 and 10 are averaged into altitude bins along withstandard deviations of the data When available data meanas well as median are shown For cases in which signifi-cant biomass smoke was encountered (eg Figs 9d 10d ande) the median is more indicative of background conditionsthan the mean However for the spring ARCPAC campaign(Fig 10c) four of the five flights encountered heavy smokeconditions so in this case profiles are provided for the meanof the smoky flights and the mean for the remaining flight

which is more representative of background conditions TheARCPAC NOAA WP-3D aircraft thus encountered heavierburning conditions (Fig 10c Warneke et al 2009) than theother two aircraft for the Arctic spring (Fig 10a and b)

Model profiles shown in each panel are constructed byaveraging monthly mean model results at several locationsalong the flight tracks (map symbols in Figs 9 and 10)We tested the accuracy of the model profile-construction ap-proach using the U Tokyo data and the GISS model by com-paring a profile constructed from following the flight trackswithin the model fields with the simpler profile constructionshown in Fig 10a The two approaches agreed very wellexcept in the boundary layer (the lowest 1ndash2 model levels)Potentially more problematic is the comparison of instan-taneous observational snapshots to model monthly meansNevertheless the comparison does suggest some broad ten-dencies

The lower-latitude campaign observations (Fig 9) indicatepolluted boundary layers with BC concentrations decreas-ing 1ndash2 orders of magnitude between the surface and themid-upper troposphere Some of the large data values canbe explained by sampling of especially polluted conditionsFor example the CARB campaign (Fig 9d) encountered



50

100

200

500

1000

a

AVE HoustonNASA WB-57F

November

b

CR-AVENASA WB-57F

February

50

100

200

500

1000

Pre

ssur

e (h

Pa)

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

c

TC4NASA WB-57F

August

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d

CARBNASA DC-8 P3-B

June

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) AVE Houston

(bc) CR-AVE TC4

(d) CARB

ModelsARQMCAMGISSGOCARTSPRINTARS LOALSCEMATCH

MOZARTMPIMIRAGE UIO CTMUIO GCM (dash)ULAQ (dash)UMI (dash)TM5 (dash)DLR (dash)

Fig 9 Model profiles in approximate SP2 BC campaign locations in the tropics and midlatitudes averaged over the points in the map(bottom) Observations (black curves) are average for the respective campaigns with standard deviations where available The Houstoncampaign has two profiles measured two different days Mean (solid) and median (dashed) observed profiles are provided for (d) Themarkers in the map inset denote the location of model profiles in these comparisons with the aircraft measurements that are detailed inTable 7

unusually heavy biomass burning The models used climato-logical biomass burning and would not have included theseparticular fire conditions Nevertheless overall the datasetsshow remarkably consistent mid-tropospheric mean BC lev-els of about 05ndash5 ng kg in the tropics and midlatitudes Withthe exception of the CARB campaign the models generallyexceed the upper limit of the standard deviation of the dataFor CARB most models are within the data standard devi-ations up to about 500 mb (Fig 9d) while about half ex-ceed the upper limit of the observed standard deviation above500 mb

The spring-time Arctic campaigns observed maximum BCabove the surface (Fig 10andashc) which may occur from twomechanisms First background ldquoArctic hazerdquo pollution isthought to originate at lower latitudes and is transported tothe Arctic by meridionally lofting along isentropic surfaces(Iversen 1984 Stohl et al 2006) Most of the observed

profiles and the model results would reflect those conditionsAlternatively BC could be injected into the mid-tropospherenear its source by agricultural or forest fires and then ad-vected into the Arctic This is apparently the case for theARCPAC measurements (Fig 10c) that probed Russian firesmoke (Warneke et al 2009) In both cases the pollutionlevels aloft during springtime are substantial and compara-ble to those levels observed in the polluted boundary layer atmidlatitudes Thus at the lower latitudes BC decreases withaltitude whereas at these higher latitudes it increases towardthe middle troposphere during springtime Model profile di-versity is especially great in the Arctic as discussed in previ-ous sections Many of the models do have profile maximumBC above the surface but most of the springtime peak val-ues are smaller in magnitude than the aircraft measurementsThe three spring campaign measurements have mean BC ofabout 50ndash200 ng kgminus1 at 500 mb 10 of the 17 models are



50

100

200

500

1000

a

Spring ARCTASNASA DC-8

April

b

Spring ARCTASNASA P3-B

April

50

100

200

500

1000

P(h

Pa)

c

ARCPACNOAA WP-3D

April

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d

Summer ARCTASNASA DC-8

Jun-Jul

50

100

200

500

1000

05 1 2 5 10 20 50 100 200 500

e

Summer ARCTASNASA P3B

Jun-Jul

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) Spring ARCTAS

(b) Spring ARCTAS

(c) ARCPAC

(d) Summer ARCTAS

(e) Summer ARCTAS



Fig 10 Like Fig 9 but for high latitude profiles Mean (solid) and median (dashed) observed profiles are provided except for (c) theARCPAC campaign has distinct profiles for the mean of the 4 flights that probed long-range biomass burning plumes (dashed) and mean forthe 1 flight that sampled aged Arctic air (solid)

less than 20 ng kgminus1 yet most of them are within the lowerlimit of the observed standard deviation Overall most mod-els are underestimating poleward transport are removing theBC too efficiently or are not confining pollution sufficientlyto the lowest model levels due to excessive vertical diffusion

The high-latitude summer ARCTAS campaigns encoun-tered heavy smoke plumes for part of their campaign so themean (Fig 10 d-e solid black) values are less characteristicof typical conditions than the median (dashed) Most modelsare within the observed standard deviation for the summer-time data however overestimate relative to median BC above500 mb Many of the models have little change in estimatesbetween spring and summer (eg compare Fig 10b and d)

while the observed background conditions are less pollutedin summer Similar to the lower latitudes the models gener-ally overestimate BC in the upper troposphere (Fig 10a andd) in the Arctic On the other hand the UTLS measurementsin the Arctic region are sparse and may not be statisticallysignificant

The ratio of model to observed BC over the profiles forFig 9 (south) and Fig 10 (north) excluding the bottom 2 lay-ers of each model are given in Table 8 we use medianobserved values for campaigns that encountered significantbiomass burning (Figs 9d 10d and e) and for the ARCPACspring (Fig 10c) we use the background profile The av-erage model ratio is 79 in the south and 041 in the north



Table 6 Average ratio of model to retrieved AERONET BC column load using the Schuster et al (2005) algorithm within regions forAeroCom models and GISS sensitivity studies Last row has average retrieved value in mg mminus2 Number of measurements is given for eachregion

BC load AER N Am 39 Eur 43 Asia 10 S Am 7 Afr 4 Rest 47

GISS 036 029 059 036 080 051CAM 032 037 047 050 040 030ARQM 047 045 054 045 040 041DLR 058 087 044 055 11 044SPRINTARS 12 13 091 063 22 065GOCART 053 073 080 048 075 038LOA 028 039 042 031 067 042LSCE 034 058 081 027 032 036MATCH 034 044 039 061 050 033MOZGN 066 080 097 039 053 044MPIHAM 022 019 045 034 038 020MIRAGE 029 036 042 051 067 036TM5 031 027 047 027 033 022UIOCTM 028 044 048 046 082 052UIOGCM 027 022 033 021 027 019UMI 028 064 079 053 038 026ULAQ 038 15 16 031 032 076Ave 042 058 064 042 064 040GISSsensitivitystd 032 039 053 026 024 019EDGAR32 034 041 049 024 023 022IIASA 037 042 064 025 024 023BB1998 034 040 053 027 025 020Lifex2 042 047 060 031 031 026Life2 028 034 047 025 021 016Ice2 035 039 054 027 024 020Icex2 030 036 050 025 023 018Retrieved

18 21 30 27 21 25

In general the ordering of model concentrations in the mid-troposphere is the same across latitudes so the models withsmall upper tropospheric concentrations in the tropics alsoare smaller in the Arctic Typically those that are most suc-cessful compared to the observations at low latitudes do nothave large enough concentrations in the lower and middletroposphere in the Arctic This could result from failure todistinguish between removal of BC by convective and strati-form clouds with convective clouds providing deep-columncleaning of particles primarily at lower latitudes The mod-els may also fail to resolve pollution transport events neededto bring pollution to the Arctic However some models arefairly versatile for example the MIRAGE UMI and GISSmodels attain large lower tropospheric concentrations in theArctic yet relatively low concentrations aloft at low latitudesthese are within a factor of 4 of observed in the south and afactor of 3 in the north (Table 8) Some of the models havea strong minimum at around 300ndash400 hPa probably due to

effective scavenging in a region where condensable watertends to be removed by rain This seems to work well inthe lower latitude regions however it apparently should notapply at the higher latitude springtime where colder cloudsdominate

We also made profiles for the GISS sensitivity simulationsHowever the variability among these cases is much smallerthan for the AeroCom models in Figs 9 and 10 Doublingor halving the GISS BC aging rate generally made the low-est and highest concentrations respectively throughout thecolumn however the difference was less than a factor of twofrom the standard case In the Arctic near the surface thecase with increased ice-out had the lowest concentration butagain the change was not large



Table 7 Single Particle Soot Photometer (SP2) Measurements of Black Carbon Mass from Aircraft

Fig Field Aircraft Investigator Dates Number of Latitude Longitude AltitudeCampaign1 Platform Group2 Flights Range Range Range (km)

9a AVE NASA NOAA 10ndash12 2 29ndash38 N 88ndash98 W 0ndash187Houston WB-57F Nov 2004

9b CR-AVE NASA NOAA 6ndash9 3 1 Sndash10 N 79ndash85W 0ndash192WB-57F Feb 2006

9c TC4 NASA NOAA 3ndash9 5 2ndash12 N 80ndash92 W 0ndash186WB-57F Aug 2007

9d CARB NASA University of 18ndash26 5 33ndash54 N 105ndash127 W 0ndash13DC-8 Tokyo Jun 2008P3-B Hawaii

10a Spring NASA University of 1ndash19 9 55ndash89 N 60ndash168 W 0ndash12ARCTAS DC-8 Tokyo Apr 2008

10b Spring NASA University of 31 Marndash19 8 55ndash81 N 70ndash162 W 0ndash78ARCTAS P3-B Hawaii Apr 2008

10c ARCPAC NOAA NOAA 12ndash21 5 65ndash75 N 126ndash165 W 0ndash74WP-3D Apr 2008

10d Summer NASA University of 29 Junndash 8 45ndash87 N 40ndash135 W 0ndash13ARCTAS DC-8 Tokyo 13 Jul 2008

10e Summer NASA University of 28 Junndash 10 45ndash62 N 90ndash130 W 0ndash83ARCTAS P3-B Hawaii 12 Jul 2008

1AVE Houston NASA Houston Aura Validation Experiment CR-AVE NASA Costa Rica Aura Validation Experiment TC4 NASATropical Composition Cloud and Climate Coupling ARCTAS NASA Arctic Research of the Composition of the Troposphere from Aircraftand Satellites CARB NASA initiative in collaboration with California Air Resources Board ARCPAC NOAA Aerosol Radiation andCloud Processes affecting Arctic Climate2NOAA David Fahey Ru-shan Gao Joshua Schwarz Ryan Spackman Laurel Watts (Schwarz et al 2006) University of Tokyo YutakaKondo Nobuhiro Moteki (Moteki and Kondo 2007 Moteki et al 2007) University of Hawaii Antony Clarke Cameron McNaughtonSteffen Freitag (Clarke et al 2007 Howell et al 2006 McNaughton et al 2009 Shinozuka et al 2007)

37 Summary of model-observation comparison

The average AeroCom model performance compared to eachmeasurement type for each region is given in Table 9 The av-erage model bias tends to be high compared with surface con-centration measurements in all regions except Asia wheremost measurements are relatively recent The average modelbias tends to be low compared with all column retrievals ex-cept for the OMI estimate for Europe The model bias is es-pecially low in biomass burning regions of Africa and SouthAmerica It is also likely that anthropogenic emissions areunderestimated especially in South America (eg Evange-lista et al 2007) The rest-of-world bias is quite low forthe column quantities however the retrievals tend to havegreater difficulty for small aerosol optical depth conditions(eg Dubovik et al 2002) and are therefore biased high Adetailed analysis in which the model diagnostics are screenedwith the same criteria as AERONET would help to resolvethis The remote BC load is sensitive to the BC aging or

mixing rate so resolving the discrepancy is important It ispossible that model aging rate is overestimated in the mod-els resulting in excessive removal and low model bias awayfrom source regions

We have only considered SP2 aircraft measurements overNorth America and generally the models are larger than ob-served both at the surface and in the free troposphere Themodels underpredict AAOD and the Schuster-BC in NorthAmerica but the comparison with aircraft data suggests thatthe models are actually overestimating middle-upper atmo-spheric BC It therefore seems that the optical properties inthe models provide less absorption than they should or thatthe retrievals overestimate AAOD or that the treatment ofthe model diagnostics are not sufficiently harmonized to theretrieval



Table 8 Ratio of model to observed aircraft campaigns for south(Fig 9) and north (Fig 10 using the median values for Figs 9d and10dndashe) The lowest 2 model layers are not used

modelobserved south north

GISS 32 048ARQM 119 11CAM 117 015DLR 36 020GOCART 101 065SPRINTARS 75 072LOA 94 012LSCE 146 033MATCH 95 009MOZART 110 074MPI 41 006MIRAGE 36 041TM5 54 011UIOCTM 51 010UIOGCM 87 069ULAQ 111 055UMI 30 050Ave 79 041

4 Discussion and conclusions

Our comparison of AeroCom models and observations re-veals some large BC discrepancies and diversities To someextent the comparison of AeroCom and GISS sensitivitymodels can be used to infer which parameters might improveperformance

The AeroCom models use a variety of BC emission in-ventories (Table 1) In the GISS sensitivity studies we usedthree recent inventories and did not see dramatic differencesin the model results however the developers of these inven-tories shared similar energy and emission factor informationso it is not surprising that the inventories are not dramaticallydifferent although for specific regions there are some largedifferences Furthermore this is consistent with the Tex-tor et al (2007) comparison of model experiments with andwithout different emissions in which model diversity was notgreatly reduced if the emissions were harmonized It there-fore seems that the lowest order model biases require eitherchanges to BC in most inventories or changes to other modelcharacteristics

The BC inventories continue to improve as information ontechnologies and activities become available especially indeveloping countries In addition it seems that model resultscould derive as much benefit from information about opticalproperties specific for individual emission sources such asparticle size density and mixing state appropriate for modelgrid-box-scale sources Biomass burning emission estima-tions are also improving For example the latest GFED es-timates rely on satellite observations of burned area and fire

Table 9 Summary table ratio of average model to ob-servedretrieved within regions from Tables 3 4 and 6

Average N Am Eur Asia S Am Afr Restmodel biases

Surface 16 26 050 NA NA 14concentrationBC burden 042 058 064 042 064 040AERONET 086 081 067 068 053 055AAODOMI AAOD 052 16 071 035 047 026

counts in deriving the burning history (Giglio et al 2006van der Werf et al 2006) Here we only considered sea-sonal variability in the GISS model and it seemed to agreereasonably well compared with retrieved AAOD seasonalityin the biomass burning regions On the other hand nearlyall models underestimate column BC in these regions espe-cially in South America suggesting that the emission fac-tors (currently based on Andreae and Merlet 2001) or opti-cal properties for the smoke are not generating enough BCandor particle absorption Spackman et al (2008) reportedBC emission factors from fresh biomass burning plumes thatwere 25 to 75 higher than those reported in Andreae andMerlet (2001) consistent with the model underestimationsnoted here Long-term in situ measurements co-located withAERONET sites could help resolve which of these is in error

Many models are developing sophisticated aerosol mi-crophysical schemes including information on nucleationevolving particle size distributions particle coagulation andmixing by condensation of gases onto particles The addedphysical treatment also allows more physical representationof particle hygroscopicity optical properties uptake intoclouds etc However it is challenging to increase physicalsophistication in the schemes while validating the schemesusing field information on how such particles behave in thereal world The assessment here includes some constraint onfinal BC properties While microphysical schemes are es-sential for simulating particle number and size distributionit is not apparent that they improve on BC simulations as ex-amined here Yet the schemes might benefit from increasedsophistication such as including evolution of particle mor-phology effect of internal mixing on particle absorption anddensity (Stier et al 2007)

Bond and Bergstrom (2006) have provided some straight-forward recommended improvements for BC models butmany models presented here had not yet included theseBond and Bergstrom (2006) suggested a typical fresh particlemass absorption cross section (MABS essentially the col-umn BC absorption divided by the load) of about 75 m2 gminus1

and that this should probably increase as particles age Nineof the models have MABS larger than 67 m2 gminus1 Enhance-ment of absorption from BC coating was recommended to



be about a factor of 15 and this had not been included inthe models A recent study with the UIOCTM did includea 15 enhancement of MABS for aged BC and found in-creased radiative forcing of 28 (Myhre et al 2009) Bondand Bergstrom (2006) recommended refractive index valueslarger than the value used in older models ie about 19ndash07i at 550 nm only three of the models have values largerthan 19ndash06i Bond and Bergstrom also pointed out thatmany models have underestimated particle density and rec-ommend a value of about 17ndash19 g cmminus3 Eleven of themodels have densities lower than this range and would haveweaker absorption if the density was increased to the rec-ommended level In summary including particle mixingor core-shell configuration and increasing refractive indexshould increase model particle absorption while increasingdensity will decrease AAOD

Model treatment of BC uptake by clouds is determinedby assuming a fixed uptake rate or by assuming the BCbecomes hydrophilic following some aging time or froma microphysical scheme that includes mixing with solublespecies Relatively little effort has been given to treatmentof BC uptake or removal by frozen clouds and precipitationSome field information is available eg Cozic et al (2007)and although more observations are needed this is a processmodels need to consider more carefully The comparison ofmodels with aircraft data (Figs 9 and 10) suggests that someupper-level removal processes may be missing Alternativelythe model vertical mixing may be excessive It would be use-ful for the models to compare other species with availableaircraft observations to learn whether the bias is primarilyfor BC or occurs also for other species The GISS model isfairly successful at capturing the decrease with altitude forSO2 sulfate DMS and H2O2 (Koch et al 2006) We havehad some success decreasing the BC aloft in the GISS modelby enhancing removal by convective clouds The ECHAM5model has found improved vertical transport results with in-creased vertical resolution Note however that decreasingthe load of BC diminishes the AAOD and worsens that bias

An obvious difficulty in applying the various datasets formodel constraint is the uncertainty in the data Thorough dis-cussion of this topic is beyond the scope of this study but webriefly summarize some issues here There are uncertaintiesin surface measurements and AAOD retrievals failure to ac-curately account for additional absorbing species failure todiagnose the model like the retrievals and mismatch of peri-ods for observations and model

Surface concentration measurements are made by a vari-ety of techniques including various thermal and optical ap-proaches summarized in eg Bond and Bergstrom (2006)This variety contributes to bias scatter in the model evalu-ations In particular the reflectance method used for IM-PROVE is known to measure higher elemental carbon thanthe transmittance method used by EMEP (Chow et al2001) which may explain some of the difference in model-measurement comparisons between these regions

AERONET and OMI retrievals of AAOD use uniformtechniques for their respective retrievals however they havetheir own uncertainties The AERONET retrieval algorithm(Dubovik and King 2000) derives detailed size distributionand spectrally dependent complex refractive index by fittingdirect and transmitted diffuse radiation measured by ground-based sun-photometers (Holben et al 1998) No microphys-ical model is assumed for size distribution or for complexrefractive index Then the values of AAOD are calculatedusing the combination of size distribution and index of re-fraction that provide best fit to the measurements The ma-jor limitation for the retrieval of aerosol absorption is causedby the limited accuracy of the direct Sun radiation measure-ments (Dubovik et al 2000) As shown by Dubovik etal (2000) the retrievals of aerosol absorption and SingleScattering Albedo (SSA = scattering(scattering + absorp-tion)) are unreliable at low aerosol loading conditions withAAOD tending to be biased high but with accuracy of 001

Although no similar limitation has been documented forthe OMI retrieval generally the accuracy of OMI retrievals(as for retrievals by any passive satellite sensors) is also lowerfor smaller aerosol loading conditions since the aerosol sig-nal to radiometric noise ratio decreases The OMI retrievalalso relies on a predetermined limited set of aerosol modelsand the OMI algorithm chooses the model as part of the re-trieval Then the AAOD as well as other aerosol parameters(including Angstrom parameter) are estimated using the re-trieved aerosol optical depth (AOD) and chosen model Ob-viously the incorrect choice of the aerosol model would af-fect the retrievals of both AAOD and angstrom parameterIn contrast the AERONET retrieval uses transmitted radia-tion (not reflected as registered by OMI) and the angstromparameter is derived from direct AOD measurements (not anaerosol model)

Both AERONET and OMI data are for daytime and clear-sky conditions only and the model results used here areall-day and all-sky Ideally model diagnostics should bescreened using similar criteria Within the GISS model wehave found that all-sky and clear-sky AAOD do not differgreatly since the absorbing aerosols are assumed to be un-affected by relative humidity Models that include aerosolmixing would probably have larger differences in AAOD forall-sky and clear-sky conditions

The AAOD measurements include absorption by dust andldquobrownrdquo or absorbing organic carbon We have included allspecies in the model AAOD estimates however we have notattempted to address shortcomings in dust simulations andthe models generally do not yet include significant absorptionfor organic carbon However we have focused on regionswhere carbonaceous aerosols dominate over dust absorptionFurthermore dust and absorbing organics absorb relativelyless at longer wavelengths compared with BC When we usedthe GISS model to consider the spectral dependence of theAAOD bias we found that the bias is generally independentof wavelength suggesting BC is the primary source of bias



A final difficulty is mismatch between dates for measure-ments and model emissions Other than the aircraft mea-surements we used long-term measurements (one year ormore) but the various measurements were taken from a va-riety of years In regions where BC has been changing sig-nificantly we may expect differing biases depending on themeasurement and its date The models generally used emis-sions for the 1990s AERONET measurements are from1996ndash2006 OMI from 2005ndash2007 IMPROVE from 1990sto 2002 EMEP for 2003 many Asian surface concentrationdata are from 2006 and the SP2 measurements are for 2004ndash2008 Over the USA there do not appear to be significanttrends in the IMPROVE data for sites that have long-termsurface concentration measurements (not shown) The otherdatasets are too short to observe significant trends Some ofthe model bias in regions such as southeast Asia where BCmay be increasing during the past 2 decades (Bond et al2007) may be due to a mismatch of emissions and measure-ment dates

We may infer model underestimation of BC radiative forc-ing from the underestimation of AERONET AAOD Accord-ing to Table 9 the average model underestimates AAODcompared with AERONET by less than a factor of 2 The av-erage AeroCom model BC radiative forcing is +025 Wmminus2

(Schulz et al 2006) If we assume that the radiative forcingis underestimated by the same amount as AAOD then the av-erage of AeroCom models would give a BC radiative forcingcloser to +05 Wmminus2 This enhancement would put the aver-age model estimate close to the +055 Wmminus2 model estimateof Jacobson (2001) who used a BC-core-shell configurationHowever the enhanced estimate is smaller than some otherrecent high estimates such as +08 Wmminus2 of Chung and Se-infeld (2002) for internally mixed BC or the retrieval-basedestimates +10 Wmminus2 of Sato et al (2003) and +09 Wmminus2

of Ramathan and Carmichael (2008)In spite of the uncertainties in models and measurements

our study has revealed some broad tendencies and biases inmodel BC simulations Compared to column estimates ofload and AAOD the models generally underestimate BCThis bias is worst in biomass burning regions where the ratioof average model to retrieved is 04 to 07 remote regions(02 to 05) and southeast Asia (06 to 07) To some extentthe bias can be attributed to differing times for emissions andmeasurements in southeast Asia and to AERONET AAODoverestimation in remote regions On the other hand themodels do not generally underestimate BC surface concen-trations And compared to aircraft measurements at low-mid latitudes of North America the models generally agreenear the surface but overestimate the BC aloft especiallyin the mid-upper troposphere At high latitudes many mod-els underestimate BC in the lower and middle troposphereThe model-aircraft comparison suggests that models allowexcessive vertical transport of BC and may be lacking suf-ficient removal by precipitating clouds they also probablylack sufficient low-level pole-ward transport Unfortunately

enhancing BC rainout especially at middle latitudes is likelyto diminish the BC available to travel pole-ward Further-more it will worsen the underestimate relative to AAOD

This study suggests several future research directionsto help close the gap between models and observationsTo improve treatment of BC optical properties modelsshould include the effect of mixing with other species andincrease refractive index as recommended by Bond andBergstrom (2006) or approximate this effect by enhancingMABS for aged BC by 15 (Bond et al 2006) Developmentof emissions inventories with size information and emissionestimates of absorbing organic aerosols for model simula-tions should also be a priority Models should include diag-nostic simulators that screen in a manner like AERONET andOMI eg only using sufficiently large AOD and clear-skydaytime conditions Although the GISS model AAOD didnot differ much for all-sky and clear-sky conditions modelswith aerosol mixing may have larger impacts from changes inrelative humidity Important additional constraint would beprovided by aircraft measurements over Eurasia the oceansand the biomass burning regions Furthermore it would beuseful to compare models and aircraft measurements usingmodel emissions for specific field seasons and comparingalong flight track particularly for campaigns that samplebiomass burning And long-term surface measurements co-located with AERONET stations especially in remote andbiomass burning regions could help interpretation of modelbiases in these regions

AcknowledgementsWe acknowledge John Ogren and twoanonymous reviewers for helpful comments on the manuscriptand Gao Chen for assistance with interpretation of the aircraftmeasurements D Koch was supported by NASA RadiationSciences Program Support from the Clean Air Task Force isacknowledged for support of SP2 measurements by the Universityof Hawaii Ghan and Easter were funded by the US Depart-ment of Energy Office of Science Scientific Discovery throughAdvanced Computing (SciDAC) program and by the NASAInterdisciplinary Science Program under grant NNX07AI56GThe Pacific Northwest National Laboratory is operated for DOEby Battelle Memorial Institute under contract DE-AC06-76RLO1830 The work with UIO-GCM was supported by the projectsRegClim and AerOzClim and supported by the NorwegianResearch Councilrsquos program for Supercomputing through a grantof computer time We acknowledge datasets for AERONETavailable at httpaeronetgsfcnasagov IMPROVE availablefrom httpvistaciracolostateeduIMPROVE and EMEP fromhttptarantulanilunoprojectsccc Analyses and visualizationsof OMI AAOD were produced with the Giovanni online datasystem developed and maintained by the NASA Goddard EarthSciences (GES) Data and Information Services Center (DISC)We acknowledge the field programs ARCTAS TC4 and ARCPAC(httpwww-airlarcnasagovhttpespoarchivenasagovarchiveand httpwwwesrlnoaagovcsdtropchem2008ARCPACP3DataDownload respectively)

Edited by B Karcher



References

Acker J G and Leptoukh G ldquoOnline Analysis Enhances Useof NASA Earth Science Datardquo Eos Trans AGU 88(2) 14ndash172007

Ackerman A S Toon O B Stevens D E HeymsfieldA J Ramanathan V and Welton E J Reduction oftropical cloudiness by soot Science 288(5468) 1042ndash1047doi101126science28854681042 2000

Ackermann I J Hass H Ebel M M Binkowski F S andShankar U Modal Aerosol Dynamics for Europe Develop-ment and rst applications Atmos Environ 32 2981ndash29991998

Andreae M O and Merlet P Emission of trace gases and aerosolsfrom biomass burning Global Biogeochem Cy 15 955ndash9662001

Andreae M O and Gelencser A Black carbon or brown car-bon The nature of light-absorbing carbonaceous aerosols At-mos Chem Phys 6 3131ndash3148 2006httpwwwatmos-chem-physnet631312006

Balkanski Y Schulz M Claquin T Moulin C and GinouxP Global emissions of mineral aerosol formulation and valida-tion using satellite imagery in Emission of Atmospheric TraceCompounds edited by Granier C Artaxo P and Reeves CE Kluwer 253ndash282 2003

Barth M C Rasch P J Kiehl J T Benkovitz C M andSchwartz S E Sulfur chemistry in the NCAR CCM Descrip-tion evaluation features and sensitivity to aqueous chemistry JGeophys Res 106 20311ndash20322 2000

Baumgardner D Kok G and Raga G Warming of the Arcticlower stratosphere by light absorbing particles Geophys ResLett 31(6) 1ndash4 2004

Bauer S E Wright D L Koch D Lewis E R McGraw RChang L-S Schwartz S E and Ruedy R MATRIX (Mul-ticonfiguration Aerosol TRacker of mIXing state) an aerosolmicrophysical module for global atmospheric models AtmosChem Phys 8 6003ndash6035 2008httpwwwatmos-chem-physnet860032008

Bauer S E Balkanski Y Schulz M Hauglustaine D A andDentener F Global modeling of heterogeneous chemistry onmineral aerosol surfaces Influence on tropospheric ozone chem-istry and comparison to observations J Geophys Res-Atmos109(D2) D02304 doi1010292003JD003868 2004

Berglen T F Berntsen T K Isaksen I S A and Sundet J KA global model of the coupled sulfuroxidant chemistry in thetroposphere The sulfur cycle J Geophys Res 109 D19310doi1010292003JD003948 2004

Bergstrom R W Pilewskie P Russell P B Redemann J BondT C Quinn P K and Sierau B Spectral absorption proper-ties of atmospheric aerosols Atmos Chem Phys 7 5937ndash59432007httpwwwatmos-chem-physnet759372007

Berntsen T K Fuglestvedt J S Myhre G Stordal F andBerglen T F Abatement of greenhouse gases Does locationmatter Climatic Change 74(4) 377ndash411 doi101007s10584-006-0433-4 2006

Bond T C and Bergstrom R W Light absorption by carbona-ceous particles An investigative review Aerosol Sci Tech 4027ndash67 2006

Bond T C Streets D G Yarber K F Nelson S M Woo J-

H and Klimont Z A technology-based global inventory ofblack and organic carbon emissions from combustion J Geo-phys Res 109 D14203 doi1010292003JD003697 2004

Bond T C Habib G and Bergstrom R W Limitations in theenhancement of visible light absorption due to mixing state JGeophys Res 111 D20211 doi1010292006JD007315 2006

Bond T C Bhardwaj E Dong R Jogani R Jung S Ro-den C Streets D G and Trautmann N M Historical emis-sions of black and organic carbon aerosol from energy-relatedcombustion 1850ndash2000 Global Biogeochem Cy 21 GB2018doi1010292006GB002840 2007

Boucher O and Anderson T L GCM assessment of the sensitiv-ity of direct climate forcing by anthropogenic sulfate aerosols toaerosol size and chemistry J Geophys Res 100 26117ndash261341995

Boucher O Pham M and Venkataraman C Simulation of theatmospheric sulfur cycle in the Laboratoire de Meteorologie Dy-namique General Circulation Model Model description modelevaluation and global and European budgets Note scientifiquede lrsquoIPSL 23 2002

Cakmur R V Miller R L Perlwitz J Koch D GeogdzhayevI V Ginoux P Tegen I and Zender C S Constrainingthe global dust emission and load by minimizing the differencebetween the model and observations J Geophys Res 111D06206 doi1010292004JD005550 2006

Chin M Diehl T Dubovik O Eck T F Holben B N SinyukA and Streets D G Light absorption by pollution dust andbiomass burning aerosols a global model study and evaluationwith AERONET measurements Ann Geophys 27 3439ndash34642009httpwwwann-geophysnet2734392009

Chin M Ginoux P Kinne S Torres O Holben B N DuncanB N Martin R V Logan J A Higurashi A and NakajimaT Tropospheric aerosol optical thickness from the GOCARTmodel and comparisons with satellite and Sun photometer mea-surements J Atmos Sci 59(3) 461ndash483 2002

Chin M Rood R B Lin S-J Muller J F and ThompsonA M Atmospheric sulfur cycle in the global model GOCARTModel description and global properties J Geophys Res 10524671ndash24687 2000

Chow J C Watson J G Crow D Lowenthal D H and Mer-rifield T Comparison of IMPROVE and NIOSH carbon mea-surements Aerosol Sci Tech 34 23ndash34 2001

Chung S H and Seinfeld J H Global distribution and climateforcing of carbonaceous aerosols J Geophys Res 107(D19)4407 doi1010292001JD001397 2002

Claquin T Schulz M and Balkanski Y Modeling the mineral-ogy of atmospheric dust J Geophys Res 104 22243ndash222561999

Claquin T Schulz M Balkanski Y and Boucher O The in-fluence of mineral aerosol properties and column distribution onsolar and infrared forcing by dust Tellus B 50 491ndash505 1998

Clarke A D McNaughton C Kapustin V N Shinozuka YHowell S Dibb J Zhou J Anderson B BrekhovskikhV Turner H and Pinkerton M Biomass burning andpollution aerosol over North America Organic compo-nents and their influence on spectral optical properties andhumidification response J Geophys Res 112 D12S18doi1010292006JD007777 2007



Cofala J Amann M Klimont Z Kupiainen K and Hoglund-Isaksson L Scenarios of Global Anthropogenic Emissions ofAir Pollutants and Methane Until 2030 Atmos Environ 418486ndash8499 doi101016jatmosenv200707010 2007

Collins W D Rasch P J Boville B A Hack J J McCaa JR Williamson D L Briegleb B P Bitz C M Lin S-J andZhang M The formulation and atmospheric simulation of theCommunity Atmosphere Model CAM3 J Climate 19 2144ndash2161 2006

Cooke W F Liousse C Cachier H and Feichter J Con-struction of a 1x1 fossil fuel emission data set for carbona-ceous aerosol and implementation and radiative impact in theECHAM4 model J Geophys Res 104 22137ndash22162 1999

Cooke W F and Wilson J J N A global black carbonaerosol model J Geophys Res-Atmos 101(D14) 19395ndash19409 1996

Cozic J Verheggen B Mertes S Connolly P Bower K Pet-zold A Baltensperger U and Weingartner E Scavengingof black carbon in mixed phase clouds at the high alpine siteJungfraujoch Atmos Chem Phys 7 1797ndash1807 2007httpwwwatmos-chem-physnet717972007

Cozic J Mertes S Verheggen B Cziczo D J GallavardinS J Walter S Baltensperger U and Weingartner E Blackcarbon enrichment in atmospheric ice particle residuals observedin lower tropospheric mixed phase clouds J Geophys Res 113D15209 doi1010292007JD009266 2008

Crounse J D DeCarlo P F Blake D R Emmons L K Cam-pos T L Apel E C Clarke A D Weinheimer A J Mc-Cabe D C Yokelson R J Jimenez J L and Wennberg PO Biomass burning and urban air pollution over the CentralMexican Plateau Atmos Chem Phys 9 4929ndash4944 2009httpwwwatmos-chem-physnet949292009

Dentener F Kinne S Bond T Boucher O Cofala J GenerosoS Ginoux P Gong S Hoelzemann J J Ito A Marelli LPenner J E Putaud J-P Textor C Schulz M van der WerfG R and Wilson J Emissions of primary aerosol and pre-cursor gases in the years 2000 and 1750 prescribed data-sets forAeroCom Atmos Chem Phys 6 4321ndash4344 2006httpwwwatmos-chem-physnet643212006

Duncan B N Martin R V Staudt A C Yevich R and LoganJ A Interannual and Seasonal Variability of Biomass BurningEmissions Constrained by Satellite Observations J GeophysRes 108(D2) 4040 doi1010292002JD002378 2003

Dubovik O Smirnov A Holben B N King M D KaufmanY J Eck T F and Slutsker I Accuracy assessment of aerosoloptical properties retrieval from AERONET sun and sky radiancemeasurements J Geophys Res 105 9791ndash9806 2000

Dubovik O and King M D A flexible inversion algorithm forretrieval of aerosol optical properties from Sun and sky radiancemeasurements J Geophys Res 105 20673ndash20696 2000

Dubovik O Holben B Eck T F Smirnov A Kaufman YKing M D Tanre D and Slutsker I Variability of absorptionand optical properties of key aerosol types observed in world-wide locations J Atmos Sci 59 590ndash608 2002

Easter R C Ghan S J Zhang Y Saylor R D Chapman EG Laulainen N S Abdul-Razzak H Leung L R BianX and Zaveri R A MIRAGE Model description and evalua-tion of aerosols and trace gases J Geophys Res 109 D20210doi1010292004JD004571 2004

Evangelista H Maldonado J Godoi R H M Pereira E BKoch D Tanizaki-Fonseca K Van Grieken R Sampaio MSetzer A Alencar A and Goncalves S C Sources andtransport of urban and biomass burning aerosol black carbon atthe south-west Atlantic coast J Atmos Chem 56 225ndash238doi101007s10874-006-9052-8 2007

Generoso S Breon F-M Balkanski Y Boucher O and SchulzM Improving the seasonal cycle and interannual variations ofbiomass burning aerosol sources Atmos Chem Phys 3 1211ndash1222 2003httpwwwatmos-chem-physnet312112003

Ghan S J and Easter R C Impact of cloud-borne aerosol rep-resentation on aerosol direct and indirect effects Atmos ChemPhys 6 4163ndash4174 2006httpwwwatmos-chem-physnet641632006

Ghan S J and Zaveri R A Parameterization of optical proper-ties for hydrated internally-mixed aerosol J Geophys Res 112D10201 doi1010292006JD007927 2007

Ghan S Laulainen N Easter R Wagener R Nemesure SChapman E Zhang Y and Leung R Evaluation of aerosol di-rect radiative forcing in MIRAGE J Geophys Res 106 5295ndash5316 2001

Giglio L van der Werf G R Randerson J T Collatz G J andKasibhatla P Global estimation of burned area using MODISactive fire observations Atmos Chem Phys 6 957ndash974 2006httpwwwatmos-chem-physnet69572006

Ginoux P Chin M Tegen I Prospero J Holben B DubovikO and Lin S-J Sources and global distributions of dustaerosols simulated with the GOCART model J Geophys Res106 20255ndash20273 2001

Gong S L Barrie L A Blanchet J-P Salzen K V LohmannU Lesins G Spacek L Zhang L M Girard E LinH Leaitch R Leighton H Chylek P and Huang PCanadian Aerosol Module A size-segregated simulation of at-mospheric aerosol processes for climate and air quality mod-els 1 Module development J Geophys Res 108(D1) 4007doi1010292001JD002002 2003

Grini A Myhre G Zender C S and Isaksen I S A Modelsimulations of dust sources and transport in the global atmo-sphere Effects of soil erodibility and wind speed variability JGeophys Res 110 D02205 doi1010292004JD005037 2005

Guelle W Balkanski Y J Dibb J E Schulz M and DulacF Wet deposition in a global size-dependent aerosol transportmodel 2 Influence of the scavenging scheme on 210 Pb verti-cal profiles surface concentrations and deposition J GeophysRes 103 28875ndash28891 1998a

Guelle W Balkanski Y J Schulz M Dulac F and MonfrayP Wet deposition in a global size-dependent aerosol transportmodel 1 Comparison of a 1 year 210 Pb simulation with groundmeasurements J Geophys Res 103 11429ndash11445 1998b

Guelle W Balkanski Y J Schulz M Marticorena B Berga-metti G Moulin C Arimoto R and Perry K D Model-ing the atmospheric distribution of mineral aerosol Comparisonwith ground measurements and satellite observations for yearlyand synoptic timescales over the North Atlantic J GeophysRes 105 1997ndash2012 2000

Guibert S Matthias V Schulz M Bsenberg J Eixmann RMattis I Pappalardo G Perrone M R Spinelli N andVaughan G The vertical distribution of aerosol over Europe ndash



Synthesis of one year of EARLINET aerosol lidar measurementsand aerosol transport modeling with LMDzT-INCA Atmos En-viron 39 2933ndash2943 2005

Hansen J E and Travis L D Light scattering in planetary atmo-spheres Space Sci Rev 16 527ndash610 1974

Hansen J and Nazarenko L Soot climate forcing via snow andice albedos P Natl Acad Sci 101(2) 423ndash428 2004

Holben B N Eck T F Slutsker I Tanre D Buis J P Set-zer A Vermote E Reagan J A Kaufman Y J NakjimaT Lavenu F Jankowiak I and Smirnov A AERONE ndash Afederated instrument network and data archive for aerosol char-acterization Remote Sens Environ 66 1ndash16 1998

Howell S G Clarke A D Shinozuka Y Kapustin V NMcNaughton C S Huebert B J Doherty S and An-derson T The Influence of relative humidity upon pollu-tion and dust during ACE-Asia size distributions and impli-cations for optical properties J Geophys Res 111 D06205doi1010292004JD005759 2006

Iversen T On the atmospheric transport of pollution to the ArcticGeophys Res Lett 11 457ndash460 1984

Iversen T and Seland O A scheme for process-tagged SO4and BC aerosols in NCAR CCM3 Validation and sensitivityto cloud processes J Geophys Res-Atmos 107(D24) 4751doi1010292001JD000885 2002

Jacobson M Z Strong radiative heating due to the mixing stateof black carbon in atmospheric aerosols Nature 409 695ndash6972001

Johnson B T Shine K P and Forster P M The semi-directaerosol effect Impact of absorbing aerosols on marine stra-tocumulus Q J Roy Meteorol Soc 130(599) 1407ndash1422doi101256qj0361 2004

Kinne S Schulz M Textor C Guibert S Balkanski Y BauerS E Berntsen T Berglen T F Boucher O Chin M CollinsW Dentener F Diehl T Easter R Feichter J Fillmore DGhan S Ginoux P Gong S Grini A Hendricks J HerzogM Horowitz L Isaksen I Iversen T Kirkevag A KlosterS Koch D Kristjansson J E Krol M Lauer A LamarqueJ F Lesins G Liu X Lohmann U Montanaro V MyhreG Penner J Pitari G Reddy S Seland O Stier P Take-mura T and Tie X An AeroCom initial assessment - opticalproperties in aerosol component modules of global models At-mos Chem Phys 6 1815ndash1834 2006httpwwwatmos-chem-physnet618152006

Kirkevag A and Iversen T Global direct radiative forcing byprocess-parameterized aerosol optical properties J GeophysRes 107(D20) 4433 2002

Kirkevag A Iversen T Seland Oslash and Kristjansson J E Re-vised schemes for aerosol optical parameters and cloud conden-sation nuclei in CCM-Oslo in Institute Report Series No 28Department of Geosciences University of Oslo Oslo Norway2005

Koch D Schmidt G A and Field C V Sulfur sea salt andradionuclide aerosols in GISS ModelE J Geophys Res 111D06206 doi1010292004JD005550 2006

Koch D Bond T Streets D Unger N and van derWerf G R Global impacts of aerosols from particularsource regions and sectors J Geophys Res 112 D02205doi1010292005JD007024 2007

Lavoue D Liousse C Cachie R H Stocks B J and

Goldammer J G Modeling of carbonaceous particles emittedby boreal and temperate wildfires at northern latitudes J Geo-phys Res-Atmos 105(D22) 26871ndash26890 2000

Liu X and Penner J E Effect of Mt Pinatubo H2SO4H2Oaerosol on ice nucleation in the upper troposphere using a globalchemistry and transport model (IMPACT) J Geophys Res107(D12) 4141 doi1010292001JD000455 2002

Liu X Penner J E and Herzog M Global simulation of aerosoldynamics Model description evaluation and interactions be-tween sulfate and nonsulfate aerosols Journal of Geophysi-cal Research 110(D18) D18206 doi1010292004JD0056742005

Liu X Penner J E Das B Bergmann D RodriguezJ M Strahan S Wang M and Feng Y Uncertain-ties in global aerosol simulations Assessment using threemeteorological datasets J Geophys Res 112 D11212doi1010292006JD008216 2007

Liu X Penner J E and Wang M Influence of anthropogenicsulfate and soot on upper tropospheric clouds using CAM3 cou-pled with an aerosol model J Geophys Res 114 D03204doi1010292008JD010492 2009

McNaughton C S Clarke A D Kapustin V Shinozuka YHowell S G Anderson B E Winstead E Dibb J ScheuerE Cohen R C Wooldridge P Perring A Huey L G KimS Jimenez J L Dunlea E J DeCarlo P F Wennberg PO Crounse J D Weinheimer A J and Flocke F Obser-vations of heterogeneous reactions between Asian pollution andmineral dust over the Eastern North Pacific during INTEX-B At-mos Chem Phys 9 8283ndash8308 2009httpwwwatmos-chem-physnet982832009

Metzger S Dentener F Krol M Jeuken A and Lelieveld JGasaerosol partitioning II global modeling results J GeophysRes-Atmos 107(D16) 4313 doi1010292001JD0011032002a

Metzger S M Dentener F J Lelieveld J and Pandis S NGasaerosol partitioning I a computationally efficient modelJ Geophys Res 107(D16) 4312 doi1010292001JD0011022002b

Miller R L Cakmur R V Perlwitz J Geogdzhayev I VGinoux P Kohfeld K E Koch D Prigent C RuedyR Schmidt G A and Tegen I Mineral dust aerosols inthe NASA Goddard Institute for Space Sciences ModelE at-mospheric general circulation model J Geophys Res 111D06208 doi1010292005JD005796 2006

Moteki N and Kondo Y Effects of mixing state on black car-bon measurement by Laser-Induced Incandescence Aerosol SciTech 41 398ndash417 2007

Moteki N Kondo Y Miyazaki Y Takegawa N Komazaki YKurata G Shirai T Blake D R Miyakawa T and KoikeM Evolution of mixing state of black carbon particles Aircraftmeasurements over the western Pacific in March 2004 GeophysRes Lett 34 L11803 doi1010292006GL028943 2007

Mueller J-F Geographical distribution and seasonal variation ofsurface emissions and deposition velocities of atmospheric tracegases J Geophys Res 97 3787ndash3804 1992

Myhre G Berglen T F Johnsrud M Hoyle C R Berntsen TK Christopher S A Fahey D W Isaksen I S A Jones TA Kahn R A Loeb N Quinn P Remer L Schwarz J Pand Yttri K E Modelled radiative forcing of the direct aerosol



effect with multi-observation evaluation Atmos Chem Phys 91365ndash1392 2009httpwwwatmos-chem-physnet913652009

Myhre G Berntsen T K Haywood J M Sundet J K HolbenB N Johnsrud M and Stordal F Modelling the solar radia-tive impact of aerosols from biomass burning during the SouthernAfrican Regional Science Initiative (SAFARI-2000) experimentJ Geophys Res 108 8501 doi1010292002JD002313 2003

Nozawa T and Kurokawa J Historical and future emissions ofsulfur dioxide and black carbon for global and regional climatechange studies CGER-Report CGERNIES Tsukuba Japan2006

Ogren J A and Charlson R J Elemental Carbon in theAtmosphere Cycle and Lifetime Tellus 35B 241ndash254 1983

Olivier J G J Bouwman A F Maas C W M V D BerdowskiJ J M Veldt C Bloss J P J Vesschedijk A J H ZandveldtP Y J and Haverlag J L Description of EDGAR Version 20A set of global emission inventories of greenhouse gases andozone-depleting substances for all anthropogenic and most natu-ral sources on a per country basis and on a 1times1 grid Rijkinstituutvoor Volksgezondheid en Milieu Bilthoven The NetherlandsRIVM report 771060002TNO-MEP report R96119 1996

Olivier J G J Van Aardenne J A Dentener F Ganzeveld Land Peters J A H W Recent trends in global greenhouse gasemissions regional trends and spatial distribution of key sourcesin Non-CO2 Greenhouse Gases (NCGG-4) edited by van Am-stel A Millpress Rotterdam ISBN 905966 043 9 325ndash3302005

Oshima N Koike M Zhang Y Kondo Y Moteki NTakegawa N and Miyazaki Y Aging of black carbon in out-flow from anthropogenic sources using a mixing state resolvedmodel Model development and evaluation J Geophys Res114 D06210 doi1010292008JD010680 2009

Penner J E Eddleman H and Novakov T Towards the devel-opment of a global inventory of black carbon emissions AtmosEnviron 27A 1277ndash1295 1993

Pitari G Mancini E Rizi V and Shindell D T Impact offuture climate and emissions changes on stratospheric aerosolsand ozone J Atmos Sci 59 414ndash440 2002

Pitari G Rizi V Ricciardulli L and Visconti G High-speedcivil transport impact Role of sulfate nitric acid trihydrateand ice aerosol studied with a two-dimensional model includingaerosol physics J Geophys Res 98 23141ndash23164 1993

Ramanathan V and Carmichael G Global and regional climatechanges due to black carbon Nat Geosci 1 221ndash227 2008

Rasch P J Collins W D and Eaton B E Understanding theINDOEX aerosol distributions with an aerosol assimilation JGeophys Res 106 7337ndash7355 2001

Rasch P J Feichter J Law K Mahowald N Penner JBenkovitz C Genthon C Giannakopoulos C KasibhatlaP Koch D Levy H Maki T Prather M Roberts D LRoelofs G-J Stevenson D Stockwell Z Taguchi S MK Chipperfield M Baldocchi D McMurry P Barrie LBalkanski Y Chatfield R Kjellstrom E Lawrence M LeeH N Lelieveld J Noone K J Seinfeld J Stenchikov GSchwartz S Walcek C and Williamson D L A comparisonof scavenging and deposition processes in global models resultsfrom the WCRP Cambridge Workshop of 1995 Tellus B 521025ndash1056 2000

Reddy M S and Boucher O Global carbonaceous aerosol trans-port and assessment of radiative effects in the LMDZ GCM JGeophys Res 109(D14) D14202 doi1010292003JD0040482004

Reddy M S Boucher O Bellouin N Schulz M Balkan-ski Y Dufresne J-L and Pham M Estimates of globalmulticomponent aerosol optical depth and radiative forcingperturbation in the Laboratoire de Meteorologie Dynamiquegeneral circulation model J Geophys Res 110 D10S16doi1010292004JD004757 2005

Sato M Hansen J Koch D Lacis A Ruedy R DubovikO Holben B Chin M and Novakov T Global atmosphericblack carbon inferred from AERONET P Natl Acad Sci 1006319ndash6324 doi101073pnas0731897100 2003

Schulz M Textor C Kinne S Balkanski Y Bauer SBerntsen T Berglen T Boucher O Dentener F GuibertS Isaksen I S A Iversen T Koch D Kirkevag A LiuX Montanaro V Myhre G Penner J E Pitari G ReddyS Seland Oslash Stier P and Takemura T Radiative forc-ing by aerosols as derived from the AeroCom present-day andpre-industrial simulations Atmos Chem Phys 6 5225ndash52462006httpwwwatmos-chem-physnet652252006

Schuster G L Dubovik O Holben B N and ClothiauxE E Inferring black carbon content and specific absorptionfrom Aerosol Robotic Network (AERONET) aerosol retrievalsJ Geophys Res 110 D10S17 doi1010292004JD0045482005

Schwarz J P Gao R S Fahey D W et al Single-particlemeasurements of midlatitude black carbon and lightscatteringaerosols from the boundary layer to the lower stratosphere JGeophys Res 111 D16207 doi1010292006JD007076 2006

Schwarz J P Spackman J R Fahey D W et al Coat-ings and their enhancement of black carbon light absorptionin the tropical atmosphere J Geophys Res 113 D03203doi1010292007JD009042 2008

Seland Oslash Iversen T Kirkevag A and Storelvmo T Onbasic shortcomings of aerosol-climate interactions in atmo-spheric GCMs Tellus 60A 459ndash491 doi101111j1600-0870200800318x 2008

Shinozuka Y Clarke A D Howell S G Kapustin V NMcNaughton C S Zhou J and Anderson B E Air-craft profils of aerosol microphysics and optical properties overNorth America Aerosol optical depth and its association withPM25 and water uptake J Geophys Res 112 D12S20doi1010292006JD007918 2007

Slowik J G Cross E S Han J-H et al An intercomparisonof instruments measuring black carbon content of soot particlesAerosol Sci Tech 41 295 2007

Smith M H and Harrison N M The sea spray generation func-tion J Aerosol Sci 29 S189ndashS190 1998

Spackman J R Schwarz J P Gao R S Watts L A ThomsonD S Fahey D W Holloway J S de Gouw J A Trainer Mand Ryerson T B Empirical correlations between black car-bon and carbon monoxide in the lower and middle troposphereGeophys Res Lett 35 L19816 doi1010292008GL0352372008

Sprung D Jost C Reiner T Hansel A and Wisthaler A Ace-tone and acetonitrile in the tropical Indian Ocean boundary layer



and free troposphere Aircraft-based intercomparison of AP-CIMS and PTR-MS measurements J Geophys Res 106(D22)28511ndash28527 2001

Stier P Seinfeld J H Kinne S and Boucher O Aerosol ab-sorption and radiative forcing Atmos Chem Phys 7 5237ndash5261 2007httpwwwatmos-chem-physnet752372007

Stier P Seinfeld J H Kinne S Feichter J and BoucherO Impact of nonabsorbing anthropogenic aerosols on clear-sky atmospheric absorption J Geophys Res 111 D18201doi1010292006JD007147 2006

Stier P Feichter J Kinne S Kloster S Vignati E Wilson JGanzeveld L Tegen I Werner M Balkanski Y Schulz MBoucher O Minikin A and Petzold A The aerosol-climatemodel ECHAM5-HAM Atmos Chem Phys 5 1125ndash11562005

Stohl A Characteristics of atmospheric transport intothe Arctic troposphere J Geophys Res 111 D11306doi1010292005JD006888 2006

Takemura T Egashira M Matsuzawa K Ichijo H Orsquoishi Rand Abe-Ouchi A A simulation of the global distribution andradiative forcing of soil dust aerosols at the Last Glacial Maxi-mum Atmos Chem Phys 9 3061ndash3073 2009httpwwwatmos-chem-physnet930612009

Takemura T Nozawa T T Emori S Nakajima T Y and Naka-jima T Simulation of climate response to aerosol direct and in-direct effects with aerosol transport-radiation model J GeophysRes 110 D02202 doi1010292004JD005029 2005

Takemura T Nakajima T Dubovik O Holben B N andKinne S Single-scattering albedo and radiative forcing of var-ious aerosol species with a global three-dimensional model JClimate 15(4) 333ndash352 2002

Takemura T Okamoto H Maruyama Y Numaguti A Hig-urashi A and Nakajima T Global three-dimensional simula-tion of aerosoloptical thickness distribution of various origins JGeophys Res 105 17853ndash17873 2000

Textor C Schulz M Guibert S Kinne S Balkanski Y BauerS Berntsen T Berglen T Boucher O Chin M DentenerF Diehl T Easter R Feichter H Fillmore D Ghan S Gi-noux P Gong S Grini A Hendricks J Horowitz L HuangP Isaksen I Iversen I Kloster S Koch D Kirkevag AKristjansson J E Krol M Lauer A Lamarque J F LiuX Montanaro V Myhre G Penner J Pitari G Reddy SSeland Oslash Stier P Takemura T and Tie X Analysis andquantification of the diversities of aerosol life cycles within Ae-roCom Atmos Chem Phys 6 1777ndash1813 2006httpwwwatmos-chem-physnet617772006

Textor C Schulz M Guibert S Kinne S Balkanski Y BauerS Berntsen T Berglen T Boucher O Chin M DentenerF Diehl T Feichter J Fillmore D Ginoux P Gong SGrini A Hendricks J Horowitz L Huang P Isaksen I SA Iversen T Kloster S Koch D Kirkevag A KristjanssonJ E Krol M Lauer A Lamarque J F Liu X MontanaroV Myhre G Penner J E Pitari G Reddy M S Seland OslashStier P Takemura T and Tie X The effect of harmonizedemissions on aerosol properties in global models - an AeroComexperiment Atmos Chem Phys 7 4489ndash4501 2007httpwwwatmos-chem-physnet744892007

Tie X Brasseur G Emmons L Horowitz L and KinnisonD Effects of aerosols on tropospheric oxidants A global modelstudy J Geophys Res 106 22931ndash22964 2001

Tie X X Madronich S Walters S Edwards D P Gi-noux P Mahowald N Zhang R Y Lou C and BrasseurG Assessment of the global impact of aerosols on tro-pospheric oxidants J Geophys Res-Atmos 110 D03204doi1010292004JD005359 2005

Torres O Tanskanen A Veihelmann B Ahn C Braak RBhartia P K Veefkind P and Levelt P Aerosols andsurface UV products from Ozone Monitoring Instrument ob-servations An overview J Geophys Res 112 D24S47doi1010292007JD008809 2007

van der Werf G R Randerson J T Collatz G J and Giglio LCarbon emissions from fires in tropical and subtropical ecosys-tems Global Change Biol 9 547ndash562 2003

van der Werf G R Randerson J T Collatz G J Giglio L Ka-sibhatla P S Arellano A F Olsen S C and Kasischke E SContinental-scale partitioning of fire emissions during the 1997to 2001 El NinoLa Nina period Science 303 73ndash76 2004

van der Werf G R Randerson J T Giglio L Collatz G JKasibhatla P S and Arellano Jr A F Interannual variabilityin global biomass burning emissions from 1997 to 2004 AtmosChem Phys 6 3423ndash3441 2006httpwwwatmos-chem-physnet634232006

Warneke C Bahreini R Brioude J Brock C A de Gouw JA Fahey D W Froyd K D Holloway J S MiddlebrookA Miller L Montzka S Murphy D M Peischl J RyersonT B Schwarz J P Spackman J R and Veres P Biomassburning in Siberia and Kazakhstan as an important source forhaze over the Alaskan Arctic in April 2008 Geophys Res Lett36 L02813 doi1010292008GL036194 2009

Warren S and Wiscombe W A model for the spectral albedo ofsnow II Snow containing atmospheric aerosols J Atmos Sci37 2734ndash2745 1980

Zhang L Gong S-L Padro J and Barrie L A Size-segregatedParticle Dry Deposition Scheme for an Atmospheric AerosolModule Atmos Environ 35(3) 549ndash560 2001

Zhang X Y Wang Y Q Zhang X C Guo W and GongS L Carbonaceous aerosol composition over various re-gions of China during 2006 J Geophys Res 113 D14111doi1010292007JD009525 2009



Abstract We evaluate black carbon (BC) model predictionsfrom the AeroCom model intercomparison project by con-sidering the diversity among year 2000 model simulationsand comparing model predictions with available measure-ments These model-measurement intercomparisons includeBC surface and aircraft concentrations aerosol absorptionoptical depth (AAOD) retrievals from AERONET and OzoneMonitoring Instrument (OMI) and BC column estimationsbased on AERONET In regions other than Asia most mod-els are biased high compared to surface concentration mea-surements However compared with (column) AAOD or BCburden retreivals the models are generally biased low Theaverage ratio of model to retrieved AAOD is less than 07in South American and 06 in African biomass burning re-gions both of these regions lack surface concentration mea-surements In Asia the average model to observed ratio is07 for AAOD and 05 for BC surface concentrations Com-pared with aircraft measurements over the Americas at lati-tudes between 0 and 50N the average model is a factor of8 larger than observed and most models exceed the mea-sured BC standard deviation in the mid to upper troposphereAt higher latitudes the average model to aircraft BC ratio is04 and models underestimate the observed BC loading inthe lower and middle troposphere associated with springtimeArctic haze Low model bias for AAOD but overestimationof surface and upper atmospheric BC concentrations at lowerlatitudes suggests that most models are underestimating BCabsorption and should improve estimates for refractive in-dex particle size and optical effects of BC coating Retrievaluncertainties andor differences with model diagnostic treat-ment may also contribute to the model-measurement dispar-ity Largest AeroCom model diversity occurred in northernEurasia and the remote Arctic regions influenced by anthro-pogenic sources Changing emissions aging removal oroptical properties within a single model generated a smallerchange in model predictions than the range represented bythe full set of AeroCom models Upper tropospheric con-centrations of BC mass from the aircraft measurements aresuggested to provide a unique new benchmark to test scav-enging and vertical dispersion of BC in global models

1 Introduction

Black carbon the strongly light-absorbing portion of car-bonaceous aerosols is thought to contribute to global warm-ing since pre-industrial times It is a product of incom-plete combustion of fossil fuels and biofuels such as coalwood and diesel Black carbon (BC) has several effectson climate primarily warming but potentially also someamount of cooling The ldquodirect effectrdquo is the scattering

Correspondence toD Koch(dkochgissnasagov)

and absorption of incoming solar radiation by the BC sus-pended in the atmosphere The absorption warms the airwhere the BC aerosol is suspended but the extinction ofradiation results in negative forcing at the earthrsquos surface(eg Ramanathan and Carmichael 2008) The ldquoBC-albedoeffectrdquo occurs because black carbon deposited on snow low-ers the snow albedo and may therefore promote snow andice melting (eg Warren and Wiscombe 1980 Hansen andNazarenko 2004) BC may also have important effects onclouds by changing atmospheric stability andor relative hu-midity and thus affect cloud formation this has been termedthe ldquosemi-direct effectrdquo (eg Ackerman et al 2000 Johnsonet al 2004) Finally BC is a primary aerosol particle andinfluences the number of particles available for cloud con-densation (eg Oshima et al 2009) it may thus play an im-portant role for the aerosol cloud ldquoindirect effectrdquo BC mayalso affect the indirect effect by acting as ice nuclei (eg Co-zic et al 2007 Liu et al 2009)

Quantifying the effects of black carbon on climate changeis hindered by several uncertainties Emissions are uncertainbecause of difficulties quantifying sources and emission fac-tors (eg Bond et al 2004) Measurements of BC concentra-tions are uncertain because of instrumental limitations in thepresent measurement techniques (Andreae and Gelencser2006) Optical properties are uncertain since these vary withsource morphology particle age and chemical processingAtmospheric column aerosol absorption comes mostly fromblack carbon in many polluted and biomass burning regionsThis absorption aerosol optical depth (AAOD) has been re-trieved from satellite and an array of sunphotometer mea-surements and these retrievals also help to constrain columnBC However the constraint is limited by uncertainties andassumptions in the retrievals as well as by the fact that otherabsorbing species besides BC are present such as dust andorganic carbon Model simulation of BC is complicated byuncertainties in treatment of initial particle size and shape ap-propriate for initial release in a model gridbox particle up-take in liquid or frozen clouds and precipitation treatment ofmixing state and optical properties Assumptions influencingthe degree of internal vs external mixing with water-solubleparticles in the accumulation mode strongly influence theaerosol absorption (Seland et al 2008) and CCN-activationInternal mixing of BC also affects BC lifetime decreasing itrelative to hydrophobic BC (Ogren and Charlson 1983 Stieret al 2006 2007) Furthermore the BC model predictionsare subject to model uncertainties that apply to any chemicalmodel simulation such as the accuracy of the modelrsquos meteo-rology including transport clouds and precipitation (eg Liuet al 2007)

The aim of this study is to evaluate model-calculated BCin recent state-of-the-art global models with aerosol chem-istry and physics to consider their diversity and comparethem with available observations There has been concernthat some models may greatly underestimate BC absorptionand therefore BC contribution to climate warming (eg Sato






21 AeroCom models












particle(4)


























221 Emissions





223 Aerosol size


3 Model evaluation








life2 ice2 icex2

0

10

25

100

200

500

1000

5000

14500










AVE ST DEV GISS CAM



LSCE MOZART TM5 UMI

ARQM MIRAGE DLR

0 10 25 100 200 500 1000 5000 13000










12








34 Seasonality





01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

+ North America

+ Europe

+ Southeast Asia

o South America

o South Africa

+ Other

standard r=008 055 r=01 047 r=006 070

00

01

02

05

10

20

30

40

50

60

200




35 Column BC load







LSCE MOZART TM5 UMI

00

01

02

05

10

20

30

40

50

60

200


53

1216



right) 1219






44 41 11 7 5 World 40 World








OMI

model

00 01 02 05 10 20 30 40 50 60 200















LSCE MOZART TM5 UMI

00 01 02 05 10 20 50 100 130

mgm2









BB 1998 038 lifex2 051 life2 029

00

01

02

05

10

20

50

100

130mgm2









50

100

200

500

1000

a


November

b

CR-AVENASA WB-57F

February

50

100

200

500

1000

Pre

ssur

e (h

Pa)

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

c

TC4NASA WB-57F

August

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d

CARBNASA DC-8 P3-B

June

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) AVE Houston

(bc) CR-AVE TC4

(d) CARB









50

100

200

500

1000

a


April

b


April

50

100

200

500

1000

P(h

Pa)

c

ARCPACNOAA WP-3D

April

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d


Jun-Jul

50

100

200

500

1000

05 1 2 5 10 20 50 100 200 500

e


Jun-Jul

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) Spring ARCTAS

(b) Spring ARCTAS

(c) ARCPAC

(d) Summer ARCTAS

(e) Summer ARCTAS













18 21 30 27 21 25


























































Edited by B Karcher



References
















































































































































21 AeroCom models












particle(4)


























221 Emissions





223 Aerosol size


3 Model evaluation








life2 ice2 icex2

0

10

25

100

200

500

1000

5000

14500










AVE ST DEV GISS CAM



LSCE MOZART TM5 UMI

ARQM MIRAGE DLR

0 10 25 100 200 500 1000 5000 13000










12








34 Seasonality





01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

+ North America

+ Europe

+ Southeast Asia

o South America

o South Africa

+ Other

standard r=008 055 r=01 047 r=006 070

00

01

02

05

10

20

30

40

50

60

200




35 Column BC load







LSCE MOZART TM5 UMI

00

01

02

05

10

20

30

40

50

60

200


53

1216



right) 1219






44 41 11 7 5 World 40 World








OMI

model

00 01 02 05 10 20 30 40 50 60 200















LSCE MOZART TM5 UMI

00 01 02 05 10 20 50 100 130

mgm2









BB 1998 038 lifex2 051 life2 029

00

01

02

05

10

20

50

100

130mgm2









50

100

200

500

1000

a


November

b

CR-AVENASA WB-57F

February

50

100

200

500

1000

Pre

ssur

e (h

Pa)

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

c

TC4NASA WB-57F

August

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d

CARBNASA DC-8 P3-B

June

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) AVE Houston

(bc) CR-AVE TC4

(d) CARB









50

100

200

500

1000

a


April

b


April

50

100

200

500

1000

P(h

Pa)

c

ARCPACNOAA WP-3D

April

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d


Jun-Jul

50

100

200

500

1000

05 1 2 5 10 20 50 100 200 500

e


Jun-Jul

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) Spring ARCTAS

(b) Spring ARCTAS

(c) ARCPAC

(d) Summer ARCTAS

(e) Summer ARCTAS













18 21 30 27 21 25


























































Edited by B Karcher



References
















































































































































particle(4)


























221 Emissions





223 Aerosol size


3 Model evaluation








life2 ice2 icex2

0

10

25

100

200

500

1000

5000

14500










AVE ST DEV GISS CAM



LSCE MOZART TM5 UMI

ARQM MIRAGE DLR

0 10 25 100 200 500 1000 5000 13000










12








34 Seasonality





01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

+ North America

+ Europe

+ Southeast Asia

o South America

o South Africa

+ Other

standard r=008 055 r=01 047 r=006 070

00

01

02

05

10

20

30

40

50

60

200




35 Column BC load







LSCE MOZART TM5 UMI

00

01

02

05

10

20

30

40

50

60

200


53

1216



right) 1219






44 41 11 7 5 World 40 World








OMI

model

00 01 02 05 10 20 30 40 50 60 200















LSCE MOZART TM5 UMI

00 01 02 05 10 20 50 100 130

mgm2









BB 1998 038 lifex2 051 life2 029

00

01

02

05

10

20

50

100

130mgm2









50

100

200

500

1000

a


November

b

CR-AVENASA WB-57F

February

50

100

200

500

1000

Pre

ssur

e (h

Pa)

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

c

TC4NASA WB-57F

August

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d

CARBNASA DC-8 P3-B

June

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) AVE Houston

(bc) CR-AVE TC4

(d) CARB









50

100

200

500

1000

a


April

b


April

50

100

200

500

1000

P(h

Pa)

c

ARCPACNOAA WP-3D

April

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d


Jun-Jul

50

100

200

500

1000

05 1 2 5 10 20 50 100 200 500

e


Jun-Jul

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) Spring ARCTAS

(b) Spring ARCTAS

(c) ARCPAC

(d) Summer ARCTAS

(e) Summer ARCTAS













18 21 30 27 21 25


























































Edited by B Karcher



References

















































































































































221 Emissions





223 Aerosol size


3 Model evaluation








life2 ice2 icex2

0

10

25

100

200

500

1000

5000

14500










AVE ST DEV GISS CAM



LSCE MOZART TM5 UMI

ARQM MIRAGE DLR

0 10 25 100 200 500 1000 5000 13000










12








34 Seasonality





01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

+ North America

+ Europe

+ Southeast Asia

o South America

o South Africa

+ Other

standard r=008 055 r=01 047 r=006 070

00

01

02

05

10

20

30

40

50

60

200




35 Column BC load







LSCE MOZART TM5 UMI

00

01

02

05

10

20

30

40

50

60

200


53

1216



right) 1219






44 41 11 7 5 World 40 World








OMI

model

00 01 02 05 10 20 30 40 50 60 200















LSCE MOZART TM5 UMI

00 01 02 05 10 20 50 100 130

mgm2









BB 1998 038 lifex2 051 life2 029

00

01

02

05

10

20

50

100

130mgm2









50

100

200

500

1000

a


November

b

CR-AVENASA WB-57F

February

50

100

200

500

1000

Pre

ssur

e (h

Pa)

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

c

TC4NASA WB-57F

August

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d

CARBNASA DC-8 P3-B

June

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) AVE Houston

(bc) CR-AVE TC4

(d) CARB









50

100

200

500

1000

a


April

b


April

50

100

200

500

1000

P(h

Pa)

c

ARCPACNOAA WP-3D

April

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d


Jun-Jul

50

100

200

500

1000

05 1 2 5 10 20 50 100 200 500

e


Jun-Jul

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) Spring ARCTAS

(b) Spring ARCTAS

(c) ARCPAC

(d) Summer ARCTAS

(e) Summer ARCTAS













18 21 30 27 21 25


























































Edited by B Karcher



References















































































































































life2 ice2 icex2

0

10

25

100

200

500

1000

5000

14500










AVE ST DEV GISS CAM



LSCE MOZART TM5 UMI

ARQM MIRAGE DLR

0 10 25 100 200 500 1000 5000 13000










12








34 Seasonality





01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

+ North America

+ Europe

+ Southeast Asia

o South America

o South Africa

+ Other

standard r=008 055 r=01 047 r=006 070

00

01

02

05

10

20

30

40

50

60

200




35 Column BC load







LSCE MOZART TM5 UMI

00

01

02

05

10

20

30

40

50

60

200


53

1216



right) 1219






44 41 11 7 5 World 40 World








OMI

model

00 01 02 05 10 20 30 40 50 60 200















LSCE MOZART TM5 UMI

00 01 02 05 10 20 50 100 130

mgm2









BB 1998 038 lifex2 051 life2 029

00

01

02

05

10

20

50

100

130mgm2









50

100

200

500

1000

a


November

b

CR-AVENASA WB-57F

February

50

100

200

500

1000

Pre

ssur

e (h

Pa)

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

c

TC4NASA WB-57F

August

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d

CARBNASA DC-8 P3-B

June

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) AVE Houston

(bc) CR-AVE TC4

(d) CARB









50

100

200

500

1000

a


April

b


April

50

100

200

500

1000

P(h

Pa)

c

ARCPACNOAA WP-3D

April

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d


Jun-Jul

50

100

200

500

1000

05 1 2 5 10 20 50 100 200 500

e


Jun-Jul

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) Spring ARCTAS

(b) Spring ARCTAS

(c) ARCPAC

(d) Summer ARCTAS

(e) Summer ARCTAS













18 21 30 27 21 25


























































Edited by B Karcher



References













































































































































AVE ST DEV GISS CAM



LSCE MOZART TM5 UMI

ARQM MIRAGE DLR

0 10 25 100 200 500 1000 5000 13000










12








34 Seasonality





01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

+ North America

+ Europe

+ Southeast Asia

o South America

o South Africa

+ Other

standard r=008 055 r=01 047 r=006 070

00

01

02

05

10

20

30

40

50

60

200




35 Column BC load







LSCE MOZART TM5 UMI

00

01

02

05

10

20

30

40

50

60

200


53

1216



right) 1219






44 41 11 7 5 World 40 World








OMI

model

00 01 02 05 10 20 30 40 50 60 200















LSCE MOZART TM5 UMI

00 01 02 05 10 20 50 100 130

mgm2









BB 1998 038 lifex2 051 life2 029

00

01

02

05

10

20

50

100

130mgm2









50

100

200

500

1000

a


November

b

CR-AVENASA WB-57F

February

50

100

200

500

1000

Pre

ssur

e (h

Pa)

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

c

TC4NASA WB-57F

August

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d

CARBNASA DC-8 P3-B

June

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) AVE Houston

(bc) CR-AVE TC4

(d) CARB









50

100

200

500

1000

a


April

b


April

50

100

200

500

1000

P(h

Pa)

c

ARCPACNOAA WP-3D

April

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d


Jun-Jul

50

100

200

500

1000

05 1 2 5 10 20 50 100 200 500

e


Jun-Jul

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) Spring ARCTAS

(b) Spring ARCTAS

(c) ARCPAC

(d) Summer ARCTAS

(e) Summer ARCTAS













18 21 30 27 21 25


























































Edited by B Karcher



References















































































































































12








34 Seasonality





01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

+ North America

+ Europe

+ Southeast Asia

o South America

o South Africa

+ Other

standard r=008 055 r=01 047 r=006 070

00

01

02

05

10

20

30

40

50

60

200




35 Column BC load







LSCE MOZART TM5 UMI

00

01

02

05

10

20

30

40

50

60

200


53

1216



right) 1219






44 41 11 7 5 World 40 World








OMI

model

00 01 02 05 10 20 30 40 50 60 200















LSCE MOZART TM5 UMI

00 01 02 05 10 20 50 100 130

mgm2









BB 1998 038 lifex2 051 life2 029

00

01

02

05

10

20

50

100

130mgm2









50

100

200

500

1000

a


November

b

CR-AVENASA WB-57F

February

50

100

200

500

1000

Pre

ssur

e (h

Pa)

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

c

TC4NASA WB-57F

August

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d

CARBNASA DC-8 P3-B

June

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) AVE Houston

(bc) CR-AVE TC4

(d) CARB









50

100

200

500

1000

a


April

b


April

50

100

200

500

1000

P(h

Pa)

c

ARCPACNOAA WP-3D

April

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d


Jun-Jul

50

100

200

500

1000

05 1 2 5 10 20 50 100 200 500

e


Jun-Jul

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) Spring ARCTAS

(b) Spring ARCTAS

(c) ARCPAC

(d) Summer ARCTAS

(e) Summer ARCTAS













18 21 30 27 21 25


























































Edited by B Karcher



References














































































































































01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

01

02

05

1

2

5

10

OM

I

01 02 05 1 2 5 10

AERONET

+ North America

+ Europe

+ Southeast Asia

o South America

o South Africa

+ Other

standard r=008 055 r=01 047 r=006 070

00

01

02

05

10

20

30

40

50

60

200




35 Column BC load







LSCE MOZART TM5 UMI

00

01

02

05

10

20

30

40

50

60

200


53

1216



right) 1219






44 41 11 7 5 World 40 World








OMI

model

00 01 02 05 10 20 30 40 50 60 200















LSCE MOZART TM5 UMI

00 01 02 05 10 20 50 100 130

mgm2









BB 1998 038 lifex2 051 life2 029

00

01

02

05

10

20

50

100

130mgm2









50

100

200

500

1000

a


November

b

CR-AVENASA WB-57F

February

50

100

200

500

1000

Pre

ssur

e (h

Pa)

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

c

TC4NASA WB-57F

August

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d

CARBNASA DC-8 P3-B

June

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) AVE Houston

(bc) CR-AVE TC4

(d) CARB









50

100

200

500

1000

a


April

b


April

50

100

200

500

1000

P(h

Pa)

c

ARCPACNOAA WP-3D

April

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d


Jun-Jul

50

100

200

500

1000

05 1 2 5 10 20 50 100 200 500

e


Jun-Jul

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) Spring ARCTAS

(b) Spring ARCTAS

(c) ARCPAC

(d) Summer ARCTAS

(e) Summer ARCTAS













18 21 30 27 21 25


























































Edited by B Karcher



References
















































































































































LSCE MOZART TM5 UMI

00

01

02

05

10

20

30

40

50

60

200


53

1216



right) 1219






44 41 11 7 5 World 40 World








OMI

model

00 01 02 05 10 20 30 40 50 60 200















LSCE MOZART TM5 UMI

00 01 02 05 10 20 50 100 130

mgm2









BB 1998 038 lifex2 051 life2 029

00

01

02

05

10

20

50

100

130mgm2









50

100

200

500

1000

a


November

b

CR-AVENASA WB-57F

February

50

100

200

500

1000

Pre

ssur

e (h

Pa)

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

c

TC4NASA WB-57F

August

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d

CARBNASA DC-8 P3-B

June

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) AVE Houston

(bc) CR-AVE TC4

(d) CARB









50

100

200

500

1000

a


April

b


April

50

100

200

500

1000

P(h

Pa)

c

ARCPACNOAA WP-3D

April

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d


Jun-Jul

50

100

200

500

1000

05 1 2 5 10 20 50 100 200 500

e


Jun-Jul

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) Spring ARCTAS

(b) Spring ARCTAS

(c) ARCPAC

(d) Summer ARCTAS

(e) Summer ARCTAS













18 21 30 27 21 25


























































Edited by B Karcher



References















































































































































44 41 11 7 5 World 40 World








OMI

model

00 01 02 05 10 20 30 40 50 60 200















LSCE MOZART TM5 UMI

00 01 02 05 10 20 50 100 130

mgm2









BB 1998 038 lifex2 051 life2 029

00

01

02

05

10

20

50

100

130mgm2









50

100

200

500

1000

a


November

b

CR-AVENASA WB-57F

February

50

100

200

500

1000

Pre

ssur

e (h

Pa)

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

c

TC4NASA WB-57F

August

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d

CARBNASA DC-8 P3-B

June

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) AVE Houston

(bc) CR-AVE TC4

(d) CARB









50

100

200

500

1000

a


April

b


April

50

100

200

500

1000

P(h

Pa)

c

ARCPACNOAA WP-3D

April

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d


Jun-Jul

50

100

200

500

1000

05 1 2 5 10 20 50 100 200 500

e


Jun-Jul

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) Spring ARCTAS

(b) Spring ARCTAS

(c) ARCPAC

(d) Summer ARCTAS

(e) Summer ARCTAS













18 21 30 27 21 25


























































Edited by B Karcher



References














































































































































OMI

model

00 01 02 05 10 20 30 40 50 60 200















LSCE MOZART TM5 UMI

00 01 02 05 10 20 50 100 130

mgm2









BB 1998 038 lifex2 051 life2 029

00

01

02

05

10

20

50

100

130mgm2









50

100

200

500

1000

a


November

b

CR-AVENASA WB-57F

February

50

100

200

500

1000

Pre

ssur

e (h

Pa)

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

c

TC4NASA WB-57F

August

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d

CARBNASA DC-8 P3-B

June

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) AVE Houston

(bc) CR-AVE TC4

(d) CARB









50

100

200

500

1000

a


April

b


April

50

100

200

500

1000

P(h

Pa)

c

ARCPACNOAA WP-3D

April

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d


Jun-Jul

50

100

200

500

1000

05 1 2 5 10 20 50 100 200 500

e


Jun-Jul

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) Spring ARCTAS

(b) Spring ARCTAS

(c) ARCPAC

(d) Summer ARCTAS

(e) Summer ARCTAS













18 21 30 27 21 25


























































Edited by B Karcher



References

















































































































































LSCE MOZART TM5 UMI

00 01 02 05 10 20 50 100 130

mgm2









BB 1998 038 lifex2 051 life2 029

00

01

02

05

10

20

50

100

130mgm2









50

100

200

500

1000

a


November

b

CR-AVENASA WB-57F

February

50

100

200

500

1000

Pre

ssur

e (h

Pa)

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

c

TC4NASA WB-57F

August

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d

CARBNASA DC-8 P3-B

June

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) AVE Houston

(bc) CR-AVE TC4

(d) CARB









50

100

200

500

1000

a


April

b


April

50

100

200

500

1000

P(h

Pa)

c

ARCPACNOAA WP-3D

April

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d


Jun-Jul

50

100

200

500

1000

05 1 2 5 10 20 50 100 200 500

e


Jun-Jul

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) Spring ARCTAS

(b) Spring ARCTAS

(c) ARCPAC

(d) Summer ARCTAS

(e) Summer ARCTAS













18 21 30 27 21 25


























































Edited by B Karcher



References














































































































































BB 1998 038 lifex2 051 life2 029

00

01

02

05

10

20

50

100

130mgm2









50

100

200

500

1000

a


November

b

CR-AVENASA WB-57F

February

50

100

200

500

1000

Pre

ssur

e (h

Pa)

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

c

TC4NASA WB-57F

August

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d

CARBNASA DC-8 P3-B

June

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) AVE Houston

(bc) CR-AVE TC4

(d) CARB









50

100

200

500

1000

a


April

b


April

50

100

200

500

1000

P(h

Pa)

c

ARCPACNOAA WP-3D

April

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d


Jun-Jul

50

100

200

500

1000

05 1 2 5 10 20 50 100 200 500

e


Jun-Jul

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) Spring ARCTAS

(b) Spring ARCTAS

(c) ARCPAC

(d) Summer ARCTAS

(e) Summer ARCTAS













18 21 30 27 21 25


























































Edited by B Karcher



References













































































































































50

100

200

500

1000

a


November

b

CR-AVENASA WB-57F

February

50

100

200

500

1000

Pre

ssur

e (h

Pa)

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

c

TC4NASA WB-57F

August

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d

CARBNASA DC-8 P3-B

June

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) AVE Houston

(bc) CR-AVE TC4

(d) CARB









50

100

200

500

1000

a


April

b


April

50

100

200

500

1000

P(h

Pa)

c

ARCPACNOAA WP-3D

April

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d


Jun-Jul

50

100

200

500

1000

05 1 2 5 10 20 50 100 200 500

e


Jun-Jul

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) Spring ARCTAS

(b) Spring ARCTAS

(c) ARCPAC

(d) Summer ARCTAS

(e) Summer ARCTAS













18 21 30 27 21 25


























































Edited by B Karcher



References













































































































































50

100

200

500

1000

a


April

b


April

50

100

200

500

1000

P(h

Pa)

c

ARCPACNOAA WP-3D

April

05 1 2 5 10 20 50 100 200 500

BC(ngkg)

d


Jun-Jul

50

100

200

500

1000

05 1 2 5 10 20 50 100 200 500

e


Jun-Jul

0

20

40

60

80

-180 -120 -60 0

Campaigns

(a) Spring ARCTAS

(b) Spring ARCTAS

(c) ARCPAC

(d) Summer ARCTAS

(e) Summer ARCTAS













18 21 30 27 21 25


























































Edited by B Karcher



References
















































































































































18 21 30 27 21 25


























































Edited by B Karcher



References

































































































































































































Edited by B Karcher



References















































































































































































Edited by B Karcher



References































































































































































Edited by B Karcher



References





















































































































































Edited by B Karcher



References













































































































































References




















































































































































































































































































































































































































Documents

Evaluation of black carbon estimations in global aerosol models