Journal of Geophysical Research Atmospheres - 2001 - Jacobson - Global Direct Radiative Forcing Due To Multicomponent

21562202d, 2001, D2, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.
1029/2000JD900514 by Lucian Blaga Central University Library Of Cluj-Napoca, Wiley Online Library on [14/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 106, NO. D2, PAGES 1551-1568, JANUARY 27, 2001
Global direct radiative forcing due to multicomponent

anthropogenic and natural aerosols
Mark Z. Jacobson
Departmentof Civil andEnvironmental

Engineering,StanfordUniversity,Stanford,California
Abstract. Global simulationsof the compositionof and directforcingdue to aerosolscontaining

naturaland/oranthropogenic sulfate,nitrate,chloride,carbonate,
ammonium,sodium,calcium,
magnesium, potassium, blackcarbon,organicmatter,silica,ferrousoxide,andaluminumoxide
werecarriedout.Chlorideandnaturalsulfatewerefoundto be the mostimportantnaturalaerosol
constituentsin the atmospherein termsof solarplusthermal-infraredforcing.Seaspraywasthe
mostimportantnaturalaerosoltype,indicatingthatit shouldbe accounted for in weatherand
climatecalculations. Ammoniumwasfoundto havea positivedirectforcing,sinceit reduceswater
uptakein sulfate-containingsolutions; thus,anthropogenic ammoniumcontributes to global
warming.The magnitudes of ammoniumandnitrateforcingweresmallerthanthoseof chlorideor
sulfateforcing.Whenorganicsweredividedintothreegroupswith differentassumed UV
absorptioncharacteristics,
totalaerosoldirectforcingat the tropopause
increasedby about+0.03 to
+0.05W m-2(directforcing
byorganics
remained
negative),
suggesting
thatUV absorption
by
organicsis a nontrivialcomponent of the globalenergybalance.Gypsum[CaSO4-2H20],sal
ammoniac[NH4C1],halite [NaC1],halite, andnitrum [KNO3] were estimatedto be the most
commonsulfate-,ammonium-,sodium-,chloride-,andnitrate-containing solid-phaseaerosol
constituents,respectively,in the globalatmosphere. Solidformationin aerosolswasfoundto
increase
total-aerosol
directforcing
by+0.03to+0.05W m-2.Spatial
andvertical
forcing
estimates,sensitivities
of forcingto relativehumidityandconcentration,
andestimatesof global
aerosolliquidwatercontentaregiven.Modeledaerosolopticalpropertiesarecomparedwith
satellite and field measurements.
1. Introduction plus cloudy sky) conditions. Subsequently,Haywood et al.

[1999] discussedthe direct, clear-skysolar forcing, over the
To date, many global studies of direct forcing due to
oceans,of bulk sea-spraydropsrelative to other aerosoltypes.
anthropogenic sulfate, black (elemental) carbon (BC), soil
This paper expands on the study work of Jacobson [1998;
dust, and organic matter (OM) have been carried out [e.g.,
1999a] to examine the importance of sea spray and its
Charlson et al., 1991; Kiehl and Briegleb, 1993; Taylor and
individual components.This study also examines global-scale
Penner, 1994; Boucher and Anderson, 1995; Tegen et al.,
aerosol compositionand liquid water content (LWC), natural
1996; Chuang et al., 1997; Feichter et al., 1997; Hansen et
and anthropogenic aerosol forcing estimates when BC is
al., 1997; Stenchikovet al., 1998; Haywood and Ramaswamy,
treated as a core for particle optics, the effect of ultraviolet
1998; Myhre et al., 1998; Penner et al., 1998; Miller and
(UV) absorptionby particulate OM on forcing, and the effect
Tegen, 1998; Hansen et al., 1998; Jacobson, 2000]. This is a
of solid formation on forcing.
global studyof the direct solar plus thermaMR forcing due to
natural and anthropogenicsulfate, nitrate, chloride, carbonate,
ammonium, sodium, calcium, magnesium, potassium, BC, 2. Model Description
OM, silica, ferrous oxide, aluminum oxide, and other
2.1. Dynamics
components.
A goal of the study is to examine the solar plus thermal-IR The global model used here is GATORG (Gas, Aerosol,
direct forcing due to size-resolvedsea-spraydrops and their Transport, Radiation, and General circulation model), which
individual components. Previously, Tegen et al. [1997] was derived from the limited-area model GATORM [Jacobson
estimatedthe sea-sprayaerosolextinctionoptical depth at 550 1999b] by replacing the dynamics code in GATORM with a
nm with a global transportmodel. Quinn and Coffman [1999] 1994 versionof the University of California at Los Angeles
and otherscalculatedthe extinctionoptical depth at individual General Circulation Model [Arakawa and Lamb, 1981],
locations. Jacobson [1998, 1999a] found that sea spray replacing heating rates in the GCM with those from
constituentsand sea-spraydrops as a whole were among the GATORM's radiationcode,andreplacingthe originalArakawa-
most important aerosol constituents in the atmosphere in Schubert(A-S) cumulusparameterizationwith a modified A-S
termsof global direct forcing underclear-skyand all-sky (clear parameterization[Ding and Randall, 1998]. The gas-aerosol-
radiative portions of GATORG are the same as those in
GATORM.
Copyright
2001bytheAmerican
Geophysical
Union. The model's resolution was set to 4 ø S-N x 5 ø W-E with 22
Papernumber2000JD900514. sigma-pressurelayers from the ground to the top-of-the-
0148-0227/01/2000JD900514509.00 atmosphere (TOA) of 0.425 mbar (=55 km). Initial
1551
21562202d, 2001, D2, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2000JD900514 by Lucian Blaga Central University Library Of Cluj-Napoca, Wiley Online Library on [14/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
1552 JACOBSON:GLOBAL DIRECT RADIATIVE FORCING
Table 1, List of Aerosol Components Carried in Each temperature,watervapor,and wind fields were obtainedfrom
Size Bin in the Model and Information About How Their an October1, 1998 3-D global climatologicaldatabase.Sea-
Concentrations were Determined and Whether Refractive surfacetemperatures were calculatedin time assuminga fixed
Index (RI) Data Were Available for Thema oceanmixed-layerdepth. Surfacealbedoswere interpolated
eachdayfroma monthlyaveraged globaldataset[NASA Data
Species Concentration Initial RI SystemStaff, 1992],exceptthatsnowandseaice albedoswere
Source Burden,Tg Data?
calculatedaccordingto changesin modeled snow cover and
sea-icecover,respectively.The modelwas first integratedin
H20(a q) CM, EQ -- Yes time for one year to generate3-D fields of temperature,water
H2CO3(aq) EQ -- RP
H2SO4(aq) EQ -- Yes vapor, and relative humidity(RH), and cloudliquid and ice
H+ EQ -- PM water contents.These fields were written to files on six days --
NH4+ T2, EQ BD PM the 15th of January, March, May, July, September, and
Na+ T2, EQ. 0.96 PM November, each at 0:00, 06:00, 12:00, and 18:00 Greenwich
Mg2+ T2,EQ 0.14 PM Mean Time (GMT) (24 files were written). The model was then
Ca2+ T2, EQ 0.16 PM
K+ T2, EQ 0.05 PM rerunfor eachof the 24 casesin equilibriummodeto calculate
HSO4' EQ -- PM aerosolcompositionand radiative fields. Although accuracy
8042' SR,T2,EQ BD PM improveswith greatertime resolution,an increasefrom 4 to
NO 3' T2, EQ 0.38 PM 24 simulationschangedtotal-aerosolforcing predictionsby
C1- T2, EQ 1.79 PM
Br- T2 .002 PM <15%, and test casesof an increasefrom 24 to 72 simulations
HCO3- EQ -- PM (6 days, 12 hoursper day) changedforcing predictionsby
CO32- EQ -- PM <6%.
NHnNO3(s) EQ -- Yes
NH4CI(s) EQ -- Yes 2.2. Thermodynamics
NH4HSO4(s) EQ -- Yes
(NH4)2SO4(s) EQ -- Yes The model treated47 aerosolspecies,identified in Table 1,
(NH4)3H(SO4)2(s) EQ -- Yes in each of 17 size bins. Of the 47 species,concentrationsof
(NH4)2CO3(s) EQ -- RP
NaNO3(s) EQ -- Yes 39 (identifiedwith an EQ in Table 1) were determinedfrom
NaCl(s) EQ -- Yes equilibriumcalculations.Of these39, 12 (thoseidentifiedwith
NaHSO4(s) FQ -- Yes an EQ and that have an initial burden in Table 1) were
Na2SO4(s) EQ -- Yes initialized with trimodal distributions over 17 size bins.
Na2CO3(s) SP,EQ 0.39 Yes
KNO3(s) EQ -- Yes
Equilibriumequations
weresolvedin eachbin with EQUISOLV
KCl(s) EQ -- Yes II [Jacobson, 1999c] among the 39 constituentsto distribute
KHSO4(s) EQ -- Yes initial burdensamong ions and solidsand to calculateaerosol
K2SO4(s) EQ -- Yes LWC. EQUISOLV II iterates temperature-dependent
K2CO3(s) SP,EQ 0.45 Yes
equilibrium, solute activity-coefficient, and water activity-
Ca(NO3)2(s) EQ -- RP
CaC12(s) EQ -- Yes
CaSO4(s) EQ -- Yes
CaCO3(s) SP,EQ 1.08 Yes to distribute anthropogenicS(VI). Natural S(VI) concentrations
Mg(NO3)2(s) EQ -- RP were obtained from Table 2b and from SAGE II data, as described
MgC12(s ) EQ -- Yes in the text; SP, the initial concentrations
of soil dustcomponentswere
MgSO4(s) EQ -- Yes partitioned from global near-surfaceand zonally averaged 600
MgCO3(s) SP,EQ 0.83 Yes mbar model predictions from Dentener et al. [1996] with the
SiO2(s) SP 9.18 Yes partitioningfractionslistedin Section2.3 here.The massesfor each
A1203(s) SP 2.16 Yes componentwere further partitionedinto accumulationand coarse-
Fe203(s) SP 1.05 Yes particlemodesfor eachregionwith massfractionsgiven in Tables
BC ER 0.23 Yes
2b and 2c; ER, the initial concentrationsof BC were obtainedby
OM 1 OR 0.51 Yes
OM2 OR 0.36 Yes
partitioningthe near-surfaceJanuaryand July and zonally averaged
uppertroposphericmodelpredictionsof BC from Cooke and Wilson
OM3 OR 0.09 Yes
[1996] into nucleation, accumulation, and coarse modes with the
massfractionsgiven in Table 2c. The lognormalparametersused
a Also shownare initial globallyand annuallyaveragedaerosol for eachmodeare given in Table 2a. Concentrations were reduced
burdens of the 20 key aerosol constituents,before equilibrium by 15% in the baselinesimulation,as describedin the text.; OR, the
calculations.Burdensare natural plus anthropogenicburdensin all initial concentrations of OM were obtainedby partitioningthe near..
cases. Where no initial burden occurs, the speciesconcentration surfaceJanuaryand July and zonally averagedupper tropospheric
was not initialized, but was subsequently calculated with the modelpredictionsof OM from Liousseet al. [ 1996] into nucleation,
equilibrium solver. The total dry soil dust burden was 15.1 Tg, accumulation,and coarsemodeswith the mass fractionsgiven in
calculatedby summingthe burdensdue to SiO2, A1203, Fe203, Table2c. The lognormalparameters
usedfor eachmodeare given
Na2CO 3, K2CO3, MgCO 3, and CaCO3. The total dry sea spray in Table 2a. The total OM in each mode was further divided into
aerosolburdenwas approximately3.3 Tg, calculatedby summing threetypesof OM (OM1, OM2, OM3), each with differentoptical
2+ 2+ + +
the burdensofNa +,Mg ,Ca ,K,Cl-,Br',and25% of Na for characteristics,as describedin the text.; PM, indicates that the ions
ssSO42'.
Concentration
sources
areasfollows:CM, fromcloud were combinedinto hypotheticalaqueouselectrolytes,and partial
modulecalculationsin GATORG; EQ, from equilibriumcalculation molar refractionsof the electrolyteswere used in refractive index
in GATORG; T2, from Table 2b; SR, global near-surfaceand 600 calculations; RP, indicates that refractive index data were not
mbar anthropogenicS(VI) concentrationswere obtained from available, and the partial molar refractionsof the corresponding
Langner arid Rodhe [1991]. AnthropogenicS(VI) was assumedto aqueouselectrolyte was used to calculate the solid's refractive
exist in nucleation-and accumulation-modeparticles,sincethis data indices;BD, tropospheric(below the Jungelayer-- see Figure 2a)
set did not include sources for coarse-mode particles. Thus, burdens
of NH4+andSO42-were0.41and2.23Tg, respectively.
informationfrom only the first two columnsin Table 2c were used Junge-layerburdenswere 0.31 and 2.31 Tg, respectively.
JACOBSON: GLOBAL DIREET RADIATIVE FORCING 1553
coefficientequations.The codeis massand chargeconserving Table 2a. Geometric Mean Volume Radii (rn, r a, rc)and
and unconditionallystable.Aside from equilibrium uptake of GeometricStandardDeviations (On,Oa,Oc) for Nucleation,
liquid water due to hydration, no gas-aerosol transfer Accumulation, and Coarse-ParticleModes, Respectively,
calculationswere performed;thus,all initialized material was for Each of Five Regionsof the Atmosphere
retained in each bin.
The remaining eight (out of 47) speciesin Table 1 were r n, •tm r a, •tm r c, •tm on oa oc
initialized with trimodal distributions, and their
concentrationswere internally mixed with those of the 39 Continental 0.015b 0.175b 3.0b 1.6b 2.1b 2.2b
other constituents
in eachbin (exceptthat BC was treatedas a NH Ocean 0.01c 0.24d 2.96d 1.6b 2.0c 2. l d
SH Ocean 0.01c 0.24d 2.96d 1.6½ 2.0c 2.1d
core for baseline optical calculations). No physical or Polar 0.01 e 0.24 e 2.96 ½ 1.6 e 2.0 e 2.1 e
chemicalprocesses affectedtheseeight species. Uppertrop. 0.01e 0.24e 1.1f 1.6e 2.0e 2.0f
PredictedRHs affect aerosolcompositionin two ways in the
model. First, at high RHs, dissolvedelectrolyteshydratemore aNH is the NorthernHemisphere, SH is the SouthernHemisphere.
liquid water than at low RHs, increasing aerosol LWC, The only variationsto thesevalueswere that rn was0.045 and0.054
increasing particle size, and decreasing solute molality. {.tmfor BC and OM, respectively [Venkataramanet al., 1994|, rc
Dissolved electrolytesin the model that affect water uptake and o c were 1.0 {.tmand 1.85, respectively,for BC [Berner et al.,
1996], and rc andoc were 1.25 {.tmand 1.85, respectively,for OM
include molecular H2SO4(aq) and all combinations of [assumedthe same as ambient aerosolfrom Berner et al., 1996].
cation/anion
pairsin Table1 [e.g.,H+/HSO4-,2H+/SO42-, bWhitby
[1978],Clean-continental
parameters.
Na+/C1-, 2Na+/SO42-,
.... etc.].Sinceseaspray,soildust,and cWhitbv [1978], Nucleation-mode parameters for marine
environme.,<.
sulfateall contain solubleelectrolytes,variationsin RH affect
dKrekov11993,Table1.6,meanaccumulation-andcoarse-mode
sea spray, soil dust, and sulfate forcing throughchangesin
parameters].
LWC. For this study,OM was assumedto be nonhygroscopic eEstimated from ocean values.
due to the dearth of information about the compositionand fKrekov [1993,Table1.9,parameters
forg2distribution
at5 km].
water uptake of OM. Aerosolscan contain hundredsof OM
compounds, 20-67% of which are water soluble[Saxenaand
Hildemann, 1996]. The ability of soluble organicsto hydrate
varies by compound,making an estimate of water uptake 2.3. Initial Aerosol Fields
difficult at this time.
Second, solid electrolyte formation is sensitive to RH. Initial aerosol fields were obtained from several sources.
When the RH is increasing and rises above the deiiquescence Trimodal lognormaldistributioninformationfor some key
RH (DRH) of a solid, the solid, if present, must dissociate. constituents
(Na+, K+, Ca•-
•+, M g2+,natural S(VI), NO 3- CI-,
When the RH is decreasing, dissolved electrolytes cannot and Br-) are given in Tables 2a, 2b, and 3 for several
precipitateuntil their crystallizationRH (CRH) is reached.For atmosphericregions. In the case of other key constituents,
the baseline simulation (DRH case), a solid was permitted to globalsurfaceandeitherelevatedor zonallyaveragedlatitude-
exist when the RH was less than the solid's DRH (although altitude concentrationdistributionswere digitized from model-
whether the solid actually existed depended on whether studymaps.AnthropogenicS(VI) distributions
were obtained
electrolyteswere supersaturated in solution). A sensitivitytest from Langnerand Rodhe[1991, Figures5b and 6b], total BC
was run (CRH case) to examine the effect of this assumption distributions were obtained from Cooke and Wilson [1996,
versus a metastable-stateassumption,where a solid was not Figures 5a, 5b, 11a, and l lb], total OM distributionswere
permittedto exist until the RH droppedbelow the solid'sCRH. obtainedfrom Liousseet al. [1996, Figures 4a, 4b, and 13a],
Total aerosoldirect forcing at the tropopausein the CRH case and total soil dust distributions were obtained from Dentener et
was 0.03 W m-2 morenegativethanin the DRH case.In a al. [1996, Figures4a and 4c]. BC and OM maps,availablefor
second test, where no solids at all were allowed to form, the Januaryand July, were interpolatedin time. Maps for sulfate
forcingwas0.05W m-2 morenegative thanin theDRH case. and soil dustwere annualaveragemaps,and the samemap, in
The effect of solid formation was small for two reasons' (1) both cases,was used every month. Total concentrationsfrom
The most important size regime for radiative forcing is the the digitized maps were partitionedinto trimodal lognormal
accumulation mode, but solids forming in this mode, distributions with the mass fraction information given in
particularlyammoniumbisulfate [NH4HSO4] and mascagnite Table 2c, as described in the footnotes for Table 1. Whereas
[(NH4)2SO4], have DRHs of 40 and 80%, respectively;thus, the useof an annualaveragemap for sulfateleadsto errors,the
when thesesolidsprecipitatefrom solution,the RH is usually globallyintegratederroris not expectedto be large,sincethe
low, and the correspondingloss of liquid water is relatively seasonalvariation in anthropogenicSO2 emissionsis only
small' (2) most aerosols on Earth are in the boundary layer about+/-11% of the annualaverage[E. C. Voldneret al., løx1ø
over the oceans, where the RH is generally high and solid globalSOxandNOx 2-level inventoryresolvedseasonally•nto
formation is somewhat inhibited. emissionsectorsandpoint and areaemissions sources,Global
Once initial equilibrium aerosol compositions were Emissions Inventory Activity (GEIA) web site,
calculated,solutiondensitiesin each size bin were obtainedby www.onesky.umich.edu/geia/, 1999]. Even thoughan annual
combining ions into electrolytes and using the method dry-mass mapwasusedfor sulfateandsoildust,their forcings
describedby Tang [1997]. Solution densitieswere combined varied each of the 24 simulations, since forcings depend on
with solid densities and wet-mass concentrations from particlecomposition
and size,whichvary with predictedRH
EQUISOLV II to estimate wet-volume concentrationsof total and temperaturein the model.
material in each bin. Wet-volume concentrations were divided .... :•-•' the llJbbll-ltlCl-[.JIJltltJll
Cooke et -' [1999] lCVlbCt.l of the
by single-particlevolumesto estimatenumber concentrations, inventory of Cooke and Wilson [1996] and estimated a new
requiredfor optical calculations. fossil fuel plus biomassburningBC burdenof 0.15-0.25 Tg
1554 JACOBSON: GLOBAL DIRF_L-• RADIATIVE FORCING
Table 2b. Estimated Nucleation, Accumulation, and [p. 22,149]. To conform with the revised estimate,the BC
Coarse-ModeMassesfor Each of Five Regions distributions extracted from Cooke and Wilson [1996] were
reduced by 15% so that the baseline yearly-.averagedtotal
burden of BC from fossil fuels and biomassburning in this
Mn' n•4•'.3n•4•'-3
ngm-3 studywas 0.23 Tg, within the range of the revisedestimate.
Whereas, Bond et al. [1998] further suggest that non-
Sa + residential coal-burning-BC emission factors of Cooke and
Continental 0.01a 2.5b 48b
Wilson [1996] may have been overestimated,Bond et al. also
NH Oceans 0.6 a 120 c 2410 c
SH Ocean 0.9 a 170 c 2440 c state that their results were preliminary and may not be
Polar 0.06a 120 210a representative of all coal-burning plants. Due to the
Midtroposphere 0.005a 1e 19e uncertainty,the BC inventorywas not adjustedfurtherhere.
g+ Initial soil dust mass in each grid cell in the model was
Continental 0.02a 6b 64b partitionedwith soil dust compositiondata from Gillette et al.,
NH Oceans 0.08 a 15.7 c 134 c [1993, Table 2.3] in the following manner: 60.0% SiO2(s),
SH Ocean 0.05 a 10 c 115 c
14.1% A1203(s), 6.85% Fe203(s), 7.04% CaCO3(s), 5.45%
Polar 0.03a 6f 62f
Midtroposphere 0.004a 0.7g 7.3g
MgCO3(s), 2.95% K2CO3(s),and 2.55% Na2CO3(s).Soil dust
was internally mixed with other particle components.Since
Ca2+
soil-dustelectrolytesinteractedwith non-soil-dustelectrolytes
Continental 0.07a 17• 153b
NH Oceans 0.3 a 54.5 c 485 c during equilibrium calculations,the compositionand LWC of
SH Ocean 0. la 25.9 c 238 c particlescontainingsoil dustvaried duringsimulations.
Polar 0.1a 20.1f 181f Stratosphericsulfate size distributionswere extracted from
Midtroposphere 0.01a 2e 1.8e 1992 StratosphericAerosol and Gas Experiment (SAGE II)
Mg2+ surface-area data [Thomason et al., 1997], lognormal
Continental 0.03a 7h 78h parameter/sulfate weight-percent estimates versus altitude
NH Oceans 0.2 a 33.4 c 331 c [Krekov, 1993, Tables 1.9 and 1.15], and aerosol number
$H Ocean 0.2 a 32.8 c 399 c
concentrationdata versus altitude [Deshler et al., 1992].
Polar 0.09a 17f 180f
Midtroposphere 0.003a 0.5e 5.5e Initial near-surfacemassconcentrations
for each key species
were distributedwith trimodallognormaldistributionsover 17
Natural S(VI)
Continental 0.07a 18f 450i model size bins, geometricallyspacedfrom 0.01-53 gm in
radius. Near-surface concentrations were assumed to be well-
NH Oceans 0.2 a 31 c 610 c
SH Ocean 0.2 a 42 c 620 c mixed in the air massup to 150 m over the oceanin summer
Polar 0.09a 18f 310f and winter, up to 300 m over continentsin summer,and up to
Midtroposphere 0.2a 40s 120g 500 m over continentsin winter [Krekov, 1993, Table 1.3].
NO3' The mixing heights for spring and autumn were averaged
Continental 0.2a 60b 840b between those of winter and summer over both oceans and
NH Oceans 0.05 a 10.5 c 182 ½
continents.Upper-tropospheric aerosolmasseswere estimated
SH Ocean 0.05 a 9. lC 92 c
Polar 0.03a 4.9f 69f primarilyfrom the data of Talbot et al. [1998], Tabazadehet al.
Midtroposphere 0•03a 6.7g 93g [1998], and Li et al. [1997], as described in Table 2b. Values
for S(VI), total BC, total OM, and total soil dust aloft were
c/-
Continental o.o2a 4b 1lOb obtainedfrom Langner and Rodhe [1991], Cooke and Wilson
NH Oceans 0.7 a 138c 4920 c [1996], Liousse et al. [1996], and Dentener et al. [1996],
SH Ocean 1a 259 c 4270 c respectively.These valueswere also distributedover 17 model
Polar 0.08a 15d 350d
Midtroposphere 0.003a 0.6g 15g
Br'
Continental 0.07a 18J 5J
NH Oceans 0.003 a 0.58 c 3.02 c
SH Ocean 0.0006 a 0.12 c 5.95 c eli et al. [1997] for 5.5 km. Total massdata were partitionedhere
Polar 0.003a 0.590 7.60 between the accumulation and coarse modes with the average
ocean accumulation:coarse mode concentration ratio for the same
Midtroposphere 0.001a 0.2g 2.9g
species,givenhere.
fEstimated
as50%of average
ocean
values.
aNucleation-mode concentrations for continents were scaled with
gAveragedatagatheredby Talbotet al. [1998]for 5.5 km, shown
the nucleation: accumulation mode volume concentration ratio from
in Tabazadeh et al. [1998, Figure 2]. Total mass data were
the continentaldistributionof Whitby[1978]. Thosefor ocean,polar, partitionedherebetweenthe accumulation andcoarsemodeswith
and upper tropospheric environments were scaled with the the averageoceanaccumulation:coarse modeconcentration ratio
nucleation:accumulation mode volume concentration ratio for
for the samespecies,given here.For natural$(VI), the resulting
marine environments from the same source.
upper-troposphericaccumulation-mode masswasmultipliedby 0.3
•Pueschel
[1995,Table5.2].Theaverage
continental
valuewas to estimatethenaturalportionof S(VI).
partitionedbetweenthe accumulationand coarsemodeswith the hEstimated
as50% of totalcontinental
Ca2+,distributed
between
average ocean accumulation:coarse mode concentrationratio for theaccumulation
andcoarse
modes
withtheaverage
Mg2+ocean
the samespecies,givenhere. accumulation:coarse
modeconcentration
ratio, givenhere.
½Datacompiledby Quinnet al. [ 1998],extractedby A. Fridlind. ipueschel
[1995,Table5.2].Theaverage
continental
valuewas
dSturges
and Barrie [1988]. Partitioned
herebetweenthe partitionedherebetweenthe accumulation
and coarsemodeswith
accumulation and coarse modes with the average ocean the averageoceanaccumulation:coarsemode concentrationratio
accumulation:coarse
modeconcentration
ratio for the samespecies, for non-sea-salt sulfate, shown in Table 3.
given here.. JFinlayson-Pitts
andPitts[1986,Table12.14].
,JACOBSON: GLOBAL DIRECT RADIATIVE FORCING 1555
Table 2c. Estimated Nucleation-, Accumulation-, and Table 3. Mean Boundary Layer Composition of Sea-
Coarse-ModeMass Fractionsfor Each of Five Regions Spray Aerosols
Northern Hemisr•here
-
Southern Hemisphere
_
BC Ion < 1 um > 1 um < 1 um > 1 um

Continental 0..093a 0.578a 0.329a
NH Oceans 0..093a 0.578a 0.329a Na + 0.12 2.41 0.17 2.44
SH Ocean 0..093a 0.578a 0.329a Ca2+ 0.055 0.49 0.026 0.24
Polar 0..093a 0.578a 0.329a Mg2+ 0.033 0.33 0.033 0.40
Midtroposphere 0..093
a 0.578a 0.329a K+ 0.016 0.13 0.010 0.11
NH4+ 0.12 0.0 0.04 0.001
Anthropogenic
$(VI) CI' 0.14 4.92 0.26 4.27
Continental 0.003ø 0.834c 0.163c
NH Oceans 0.004b 0.846d 0.150d nssSO•42'0.99 0.18 0.31 0.075
SH Ocean 0.004b 0.805a 0.191a ssSO4"- 0.031 0.61 0.042 0.62
Polar 0.003b 0.510½ 0.487½ NO 3- 0.011 0.18 0.009 0.092
Br- 0.006 0.003 0.001 0.006
Midtroposphere 0.003
• 0.510
½ 0.487
½ ,,
A1203(s) Unitsareggm-3.Values
arefromdatacompiled
byQuinnetal.
Continental 0 0.217f 0.783f
998]. Size regimesrefer to particledry diameter.
NH Oceans 0 0.217f 0.783f
SH Ocean 0 0.217f 0.783f
Polar 0 0.217f 0.783f
Midtroposphere 0 0.217
f 0.783
f
OM
Continental 0.004• 0.876c 0.120c The tropospheric NH4+ burdenwas 0.41 Tg-NH4+, closeto that
NH Oceans 0.004• 0.689g 0.307g of Adamset al. (0.39 Tg-NH4+). The BC burden,0.23 Tg, was
SH Ocean 0.004b 0.689• 0.307•
Polar 0.002b 0.678e 0.320e within the range of Cooke et al.'s [1999] estimate (0.15-0.25
b 0.678
½ 0.320
e Tg). The NO 3- burden was 0.38 Tg-NO3•, between those of
Adamset al. [1999] (0.13 Tg-NO3-)andAndreae [1995] (0.5
Si02(s)
Continental 0 0.143f 0.857f Tg-NO3- ). The dry soil loading here was 15.1 Tg, which
NH Oceans 0 0.143f 0.857f compareswith 16.4 Tg from Andreae [1995] and 18.5 Tg from
SH Ocean 0 0.143f 0.857f Tegenet al. [1997]. The dry sea-sprayloadinghere was 3.3 Tg.
Polar 0 0.143f 0.857f The water content associatedwith sea spray was about 12 T g
Midtroposphere 0 0.143
f 0.857
f when an internal mixture was assumed and 15 Tg when an
Fe203(s) external mixture was assumed, giving a global sea-spray
Continental 0 0.332f 0.668f burden of 15-18 Tg. The sea-sprayburden from Tegen et al.
NH Oceans 0 0.332f 0.668f
[1997] was 11.4 Tg. The tropospheric plus stratospheric
SH Ocean 0 0.332f 0.668f
Polar 0 0.332f 0.668f aerosoldry masshere was 25.1 Tg. The massof liquid water •n
f 0.668
f equilibrium with all internally-mixed aerosol componentswas
29.5 Tg.
aAveraged
fromBerneret aI. [1996],Table2.
•Nucleationmode concentrations for continentswere scaledwith
the nucleation: accumulation-modevolume concentrationratio from 2.4. Radiation
thecontinentaldistributionof Whitby[ 1978]. Thosefor ocean,polar,
and upper troposphericenvironmentswere scaled with the Radiative forcing calculations require aerosol, RH,
nucleation:accumulation-mode volume concentration ratio for temperature, and cloud fields. Temperatures,RH, and cloud
marine environments from the same source. fields were obtained from the time simulation. Predicted cloud
cObtainedfrom SouthernCalifornia Air Quality Study data over
land sites liquid and ice contentswere combinedwith modified-gamma
aNss-SO42-
fromTable3 size distribution data from Welch et al. [1980] and power-
½Estimate law/Marshall-Palmer distribution data from Platt [1997] for
fChow
etal. [1994] different cloud types to estimate cloud liquid and ice
gObtainedfrom SouthernCalifornia Air Quality Study data over concentrations versus size.
an island site
Absorption, scattering,and forward scatteringefficiencies
for eachparticle size, wavelength,and core and shell refractive
index were calculated with a Mie code for stratified spheres
bins. Concentrations in each bin were then interpolated [Toon and Ackerman, 1981]. Cores of the stratified spheres
betweenthe boundarylayer and upper troposphere.Once were assumed to consist of BC. Shell refractive indices were
sulfatefields were obtained,ammoniumwas initialized in each obtainedby taking volume-averagesof the refractive indices
sizebin assuming an ammonium:sulfate moleratioof unity. of nonsolutionand non-BC substancesand volume-averaging
'I'able 1 showsthe initial globally and annuallyaveraged these with the solution refractive index. The solution real
burdensof individualaerosolcomponents, beforeequilibrium refractive index was obtained with a partial molal refractic,n
calculations. The burden of natural plus anthropogenic approach [e.g., $telson, 1990; Tang, 1997], which requires
tropospheric
SO42-was2.23Tg (0.73-TgS), closeto thatof partial molar refraction data of either ions or aqueous
Adamset al. [1999] (0.70-Tg S) andbetweenthoseof Chin et electrolytes, available for visible wavelengths from both
al. [1996](0.53-TgS) andLelieveldet al. [1997] (1.05-Tg S). references. Here, ions were combined into hypothetical
1556 JACOBSON:GLOBAL DIRECT RADIATIVE FORCING
aqueous electrolytes and partial molar refractions of O'Brien et al. [1975] and Mylonas et al. [1991] found that
electrolytes were applied. Although the ion and aqueous organicnitratesmade up 12-21% of organic massin Los
electrolyte methods give nearly identical results, the Angeles.The estimatethat 10% of global-scaleOM contains
electrolyte methodwas chosen,since it allows an estimateof strongly-absorbing nitratedorganicsor PAHs is lessthan the
refractive indices at nonvisible wavelengths, where partial percentof nitratedorganics measuredin Los Angeles.The OM
molar refraction data are not available. At such wavelengths, size distributionparametershere (Table 2a) for the nucleation
partial molar refractionsof aqueouselectrolyteswere estimated mode were from tunnel-studyOM data [Venkataramanet al.,
to equal those of the correspondingsolid electrolyte, which 1994]. Those for other modes were taken from ambient data
appears relatively accurate [Stelson, 1990, Figure 1]. The [Berner et al., 1996]. OM was treated as a shell material for
solutionimaginary refractive index was also calculatedwith a optical calculations.
partial molar refraction approachthat simplifies to a mixing Forcingcalculations
of individual
ions(NH4+, Na+, Mg2+,
rule similar to the volume-averagemixing rule, exceptthat, in Ca2+,K+, NO3-,C1-) and familiesof ions/undissociated
the imaginary case, the index of each component in the electrolytes,
including
theS(VI) family[H2SO4(aq)
HSO4-
, and
mixture was scaledby the ratio of the solutiondensityto the SO42-],weredetermined
by running
a simulation
withall
weighted sum of the densities of the individual electrolytes components present,thenrunninga simulationafterremoving
plus water in solutioninsteadof by the volume fractionof the an individual ion or family. Upon removal of the ion or
species in solution. family,chargewasautomatically adjustedin eachsizebin to
Table 1 indicatesthat, of the 47 speciessimulated,all 13 ensureinitial chargeneutrality.For consistency, adjustments
ions were combinedinto hypotheticalelectrolytesfor partial for anions(cations)
weretreatedin the sameway asthosefor
molar refraction calculations. Wavelength-dependent UV, all otheranions(cations).If an anion[e.g.,S(VI), NO3-, CI']
visible, and IR refractive indices were obtained for 30 of the wasremoved, H+ wasfirstreduced.If theinitialanionmolality
remaining 34 species.Real and imaginaryrefractiveindex data exceeded that of H +, H + was reduced to a minimum value and
for SiO2(s),Fe203(s),andBC were obtainedfrom Table 1.11 of CO32-
wasadded
untilinitialcharge
balance
wasobtained.
If a
Krekov [1993]. The midvisible refractive index value used for cation[e.g.,Na+, NH4+,Ca2+,Mg2+,K+] wasremoved,
H+ was
BC was m=1.8-0.74i. The imaginary componentwas between increaseduntil chargebalanceoccurred.In both cases,H+,
that of Bergstrom [1973] (m=2.0-0.66i) and Hiinel [1987] CO32-,
andallother
ionmolalities
changed
during
subsequent
(m=l.9-1.0i). The sensitivity of BC forcing due to the equilibrium calculations.When ions were removed, solution
assumptionof refractiveindex is reportedby Jacobson[ 2000, compositions
andLWCs changed,
affectingforcing.Although
Table 1]. For four species, refractive index data were not othermethodsof balancingchargeare possiblefor individual
available and were estimated by combining partial molar ions,the methodsabovewere chosenso that they couldbe
refractions of ions constituting the species, then applying applied uniformly to all cationsand artions.Following
equations from Tang [1997] to convert partial molar initialization, charge- and mass-conservingequilibrium
refractions to refractive indices.
calculations were performed to predict equilibrium
Once equilibrium compositionand gas, aerosol,and cloud composition.
optical propertieswere calculated,irradianceswere determined The purposeof calculatingforcingdue to naturalaerosol
for 86, 67, and 256 wavelength intervals <0.8, 0.8-4.5, and constituentsis to estimate the relative importanceof such
4.5-1000 gm, respectively (409 intervals total), with the constituentsin the global radiativebalance.If, for example,
radiative transfer algorithm describedby Toon et al. [1989], sea sprayor naturalsoil has a sufficientforcing, it shouldbe
which is integratedinto the GATORG model. includedin thediabaticheatingratetermof globalmodels.The
In the basesimulation,UV absorptionby OM was accounted purposeof calculatingforcing due to natural ions, such as
for. Reductions in UV irradiance of 50-75% have been
naturalchlorideand sulfate,is to maintainconsistency with
observedover large biomass-burningregions [Herman et al., calculationsof anthropogenicsulfate,nitrate, and ammonium
1999; Krotkov et al., 1998]. Such plumes contain OM, certain forcing. An alternative to removing ions is to remove
types of which are strong UV absorbers.Known UVA and electrolytes (e.g., Na+/CI-), but thistechniquecreatescharge-
UVB-absorbing aerosol OM include nitrated aromatics, balanceissues,since,for example,Na+ doesnot balanceCI- in
benzaldehydes,benzoic acids, aromatic polycarboxylicacids, sea spray.
phenols, and polycyclic aromatic hydrocarbons (PAHs)
[Jacobson,1999b]. AbsorbingPAHs, in particular,are present
in biomass-burningaerosols [Reid et al., 1998; Ferek et al., 3. Model ComparisonsWith Observations
1998; Andreae et al., 1998; Fang et al., 1999].
3.1. Global Radiative and Cloud Parameters
To accountfor UV reductionsby OM from biomassburning
and fossil fuel combustion, total OM in each model size bin Table4 compares modeledgloballyandannuallyaveraged
was divided into three types of OM, each with different optical TOA (definedat 0.425 mb) irradiances,
TOA cloudforcings,
characteristics:10% was assumedto consist of compounds, surfaceirradiances,cloud fraction,and precipitationwith
suchas PAHs and nitratedaromatics,that stronglyabsorbUV observations.
The observedcloudforcing,derivedby Kiehl et
radiation (OM3), 40% was assumedto consistof compounds, al. [1998] and Hack et al. [1998] from the Earth Radiation
such as benzaldehdyes,that weakly-to-moderatelyabsorbUV BudgetExperiment,
was about-19.0W m-2. GATORG cloud
(OM2), and the rest was assumedto absorbUV only slightly forcing
wasabout-20.0W m-2.Figure1 compares
December-
(OM1). Wavelength-dependent imaginaryrefractive indicesof January-February
(DJF) andJune-July-August
(JJA)GATORG
OM3 were assumedto be an averageof the refractiveindicesof predictionsof zonally averagedcloud fractionwith Nimbus-7
4-nitrophenoland the 4-nitrophenolanion and those of OM2 satellitedata from Hack et al. [1998, Figure 13]. The figure
were assumedto be an averageof the refractive indicesof 2- suggests that the JJA predictions were consistent with
hydroxybenzaldehydeand 3-nitrophenol. Indices for all these observations,
exceptfor overpredictions
at midlatitudes
in the
speciesare shownin Figure 2 of Jacobson[1999b]. Northern Hemisphere. The DJF predictions were also
JACOBSON: GLOBAL DIRECI' RADIATIVE FORCING 1557
Table 4. Comparisonof GATORG Predictions,Averaged 1 ,,I,,I,,I,,I,,I,,

1
Globally Over the 24 Simulations,With ObservedGlobally
ß= 0.8- . ,, Model
andAnnuallyAveragedRadiativeandCloudProperties
• 0.6'
Observed Predicted
• 0.4
TOAthermal-IR
irradiance,
W m'2 (- isup) Nimbus-7
-234.8 a -230.1 = 0.2
with gases,with aerosols,with clouds DJF
with gases,with aerosols,no clouds -264.0 a -260.6
TOA solarirradiance,
W m'2 (+ is down) CO 0 '' I'' I'' I''1''1'' 0
with gases,with aerosols,with clouds +238.1 a +240.1 90 60 30 0 -30 -60 -90
with gases,with aerosols,no clouds +286.3 a +290.6
TOAthermal-IRcloudforcing,
W m'2 +29.2 a +30.5 Longitude(degrees)
TOAsolarcloudforcing,
W m'2 -48.2 a -50.5
TOAtotalcloudforcing,
W m'2 -19.0 a -20.0
,,I,,I,,I,,I,,I,,
Surfacethermal-IRwithclouds,W m'2 -66.0b -59.3 1
Surfacesolarirradiance
withclouds,W m-2 + 168.0b + 161.3
Total cloud fraction, % 52.2-62.5 a 61.9 ß= 0.8
Precipitation
rate,mmd-1 2.69 c 2.81
if=0.6'
aKiehlet al. [ 1998]
bKiehl
andTrenberth[
1997]
• 0.4'
CXieand Arkin [1996] Nimbus-7
= 0.2-
• - JJA
CO 0 z
'' I'' I'' I''1''1''
consistentin the Southern Hemisphere but were high in the 90 60 30 0 -30 -60 -90
Northern Hemisphere. Errors in Northern-Hemisphere cloud
Longitude(degrees)
predictionsmay have been due to excessconvectionand water
vaporpredictedin the model, possiblydue to errorsin treating Figure 1. Comparisonof December-January-February(DJF)
convection over heterogeneoustopography with coarse grid and June-July-August(JJA) GATORG predictionsof zonally
spacing.Hack et al. [ 1998, Figure 13] showsthat predictions averaged Nimbus 7 satellite cloud fraction data, obtained from
from the Community Climate Model (CCM3) were similarly
Hack et al. [ 1998, Figure 13].
high in the NorthernHemispherefor DJF.
3.2. Aerosol Optical Parameters aerosol components from TARFOX data over the North
Atlantic
as-0.8to- 1.1W m-2.Base-case
fluxchanges
overthe
Plate l a shows the modeled yearly-averaged spatial North Atlantic from this work under the same conditions were
distribution of the near-surface aerosol optical depth at 550 about-1.1W m-2.TheirJune-July-August
flux change
for all
nm. Plates lb and l c show the correspondingnear-surfacetotal aerosolsover the North Atlantic under clear-skyconditions
aerosolwet mass and aerosolLWC, respectively.Global- and was-5.1 W m-2,whichcompares
with-4.7 W m-2 calculated
hemispheric-average values of optical depth and other here.
parametersare given in Table 5. Modeled midvisible single- Modeledannuallyaveragedmidvisibleopticaldepthsover
scattering albedos (0.86-0.95) and optical depths (0.1-0.4) the globaloceanswerecalculatedhereas 0.13 (Table5), which
over the eastern United States and western Atlantic Ocean are compares well with a mean value of 0.12 obtained from
within the range of measuredvalues from TARFOX, given as advanced very high resolutionradiometer(AVHRR) databy
0.89-0.93 and 0.06-0.55, respectively [Russell et al., 1999a, Husar et al. [1997], althoughsomespatialdifferencesare
1999b; Hignett et al., 1999]. Bergstrom and Russell [1999] evident.The yearly-averaged modeledopticaldepthof the
estimated the annual average aerosol radiative flux change aerosolplume over the oceanto the west of the Saharadesert,
underall-skyconditions
dueto all naturaland anthropogenic observed
in AVHRR data,wasabout0.2-0.33,whichcompares
with an annual value of near 0.4-0.5 estimated from Husar et al.
Table 5. Modeled Annually Averaged Values of Several [1997]. Model errorswere likely due to underestimates
of the
concentration of elevated soil dust.
Parameters
Parameter Globe SH NH Land Sea

Table 6. Comparisonof Base-CasePredictedSingle-
Scattering
AlbedosWith AnnualAverageMeasurements
a
tooat 550 nm 0.91 13.94 0.89 0.85 0.93
goat 550 nm 0.70 0.71 0.68 0.64 0.72
Location Predicted Observed
Zaerat 550 nm 0.14 0.10 0.18 0.16 0.13
0-km LWC 8.1 7.4 8.7 4.1 9.6
5-km LWC 0.58 0.73 0.43 0.28 0.69 Barrow (71.2 øN, 156.3 øW) 0.97 0.96
0-km aerosol wet mass 32 20 44 45 27 AmundsenScott (90 øS) 0.97 0.965
5-km aerosol wet mass 0.78 0.97 0.58 0.38 0.93 Ny Alesund(79 øN, 12 øE) 0.90 0.948
Arctic 0.94 0.96
Mesa Verde (37.1 øN, 108.3 øW) 0.87 0.91
SH is SouthernHemisphere,NH is NorthernHemisphere.
Ahastumani(4!.4 _øN.,
42.5 øE) 0.84. 0.89
Variable too is the near-surface
single-scattering
albedo,gois the Anderson Mesa (35.12 øN, 111.38 øW) 0.89 0.94
near-surface asymmetryparameter,and '•aeris the aerosoloptical
depth.Liquid water content(LWC) and wet massesare in units of
ggm-3. aHeintzenberg
et al. [1997]
1558 JACOBSON: GLOBAL DIRECT RADIATIVE FORCING
0m
-30
-
-90
, ,,
- 180 - 120 -60 0 60 120 180

Longitude(degrees)
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Aerosolopticaldepthat 550 nm (year)
(a)
90
-90
- 180 - 120 -(x) 0 60 120 180
Longitude(degrees)
0 20 40 60 80 100 120 140 160 180 200 220

Near-surfacetotalaerosolmass(ug m-3) (year)
(b)
•)
.)
-9()
- 180 - 120 -60 0 60 120 t 80
la)ngitude(degrees)
20 30 40 50 O0 70 80
Near-surface
LWC tugm-3) (year)
(c)
Plate 1. Modeled global distributions of (a) annually
averaged stratopauseto surface total aerosol optical depth at
550 nm, (b) annually averagednear-surfacetotal aerosol wet
mass,integratedover all sizes,and (c) annually averagednear-
surfaceaerosolliquid water content,integratedover all sizes.
JACOBSON: GLOBAL DIRECT RADIATIVE FORCING 1559
Modeled optical depthsin the Arabian Sea (0.15-0.45) and aerosol LWCs were similar in the Northern and Southern
Indian Ocean (0.05-0.15) at 550 nm, shown in Plate la, were Hemispheres (near 7-9 gg m'3), but higherover the oceans
within the range of values observedduring the Indian Ocean (near10gg m-3)thanoverland(near4 •g m-3).
Experiment (INDOEX) by Jayaraman et al. [1998], given as As shown in Plate lb, the highest concentrationof total
0.2-0.4 and <0.1, respectively. aerosol wet mass was over North Africa, where Sahara dust is
Table 6 compares aerosol single-scattering albedos present.Corresponding
aerosolLWCs werelow in thatregion
calculated from the base-casesimulation (BC treated as core) (Plate l c). Aerosol LWCs over the oceans sometimes reached
with annual average observations at several locations 20-40!.tgm-3,butweremorefrequently
5-25• g m-3.Overland,
[Heintzenberget al., 1997]. The resultsindicateagreementto LWCswerecommonly below10 gg m-3,butwererelatively
within <7% in all cases, less than the error of the high (up to 40 !.tg m-3) in regionsof high sulfate
measurements. Hitzenberger and Puxbaum [1993] found concentrations.
absorptioncoefficientmeasurement
errorsof +15-20% during
high pollution episodes but >+200% for background 5. Baseline Aerosol Forcing Results
atmospheric absorption. All single-scattering albedo
observations, except for one, were at relatively clean Table 7 showspredictionsof near surface,tropopause,and
locations. TOA aerosoldirect forcing, averagedover the 24 simulations
and globally.The table also comparestropopauseforcingover
the Northern versus Southern Hemispheres and over land
4. Aerosol Composition,Mass, and LWC versusocean.Table 8 comparesmodeledtropopauseforcings
from Table 7 with thosefrom previousstudies.Figure 3 shows
Figure 2 shows modeled vertical profiles of globally and
yearly averaged aerosol composition from the simulations. the modeled globally averagedtropopauseforcings in bar-
graph form.
Figure 2a, which shows profiles of sulfate-containing
compoundsand water, indicatesthat gypsum(CaSO4-2H20)
5.1. Sea-Spray Forcing
was the most abundant sulfate-containing solid in the
atmosphere.This is not surprising, since its DRH is 98%, Sea spray covers2/3 of the Earth's surface.Table 7 shows
allowing it to form in sea spray drops, which contain both that, on a global scale, modeled yearly-averagedsea-spray
calcium and sulfate, and in soil dust, which contains calcium forcingwas -0.55 W m-2 at the tropopause
in the presenceof
andbecome
acidified
uponsulfuric-• Despite clouds and -1.1
fid'condensation. W m-2 at the TOA in the absenceof clouds.
its abundance,gypsum plays a rel •tively small role in the These forcings were greater in magnitudethan those due to
global radiative balance, since il forms primarily in the natural sulfate and natural soil under clear and cloudy
coarse-particlemode. When gypsur• was not allowed to form conditions, making sea spray the most important natural
in the model, the direct forcing due to all aerosolswas only aerosol constituent in the atmosphere in terms of direct
0.01W m-2morenegative
thanwhenit wasallowed
to form. forcing.
Thenardite (Na2SO4, DRH=84.2%) and arcanite (K2SO4, The baselinesea-sprayforcing was obtainedwith sea spray
DRH=97.5%) were estimated to be the next most common mixed with soil, sulfate, and other material as a shell
sulfate-salts.Since both are present primarily in the coarse componentin particles. A sensitivitytest was run to calculate
mode (Na+ and K+ are bothpresentin seasprayand soil), they sea spray forcing when it was treated as externally-mixed.
also have relatively little radiativeeffect. Ammonium bisulfate Whenexternally-mixed,sea-spray
forcingwas-0.67 W m-2
(NH4HSO 4, DRH=40%) and mascagnite [(NH4)2SO 4, (versus
-0.55W m-2 for aninternal
mixture).
Whenseaspray
DRH=80%], often assumed to be the main solid forms of was externally-mixed, ammonium and nss-sulfate were
excluded from sea-spray particles, resulting in lower
sulfate, were estimated to be the fourth and fifth most-common
concentrations
sulfate salts. Of the total sulfate initialized in the model, aboutof solids(particularlyCaSO4-2H20 and NHnC1)
14% precipitatedand the rest remainedaqueous. and greaterLWCs. Greater LWCs in the external-mixturecase
The most common ammonium salt (Figure 2b) was sal were also causedby the fact that Na+/C1- hydratesmore liquid
ammoniac (NH4C1, DRH=77.1%), formed from dissolved waterthandoesNH4+/C1- at the sameRH.
ammoniaand sea-spraychloride.Halite (NaC1, DRH=75.3%) Table 7 showsthat chlorideand its associatedliquid water
was the most abundant sodium and chloride-containing salt causeda tropopause forcingof-0.29 W m-2, closeto that of
(Figures2c and2d). Nitram (KNO3, DRH=92.5%),whichforms naturalsulfate(-0.34 W m-2).Chloride'sconcentrationsover
when nitric acid enterssea-sprayor soil-dustparticles,was the noncoastalocean regions (where acidification is small in the
most common nitrate salt (Figure 2d). Calcite (CaCO3, coarsemode)are1-12gg m-3,withanaveragevalueof about
DRH=100%) and magnesite(MgCO3, DRH=100%), presentin 4.8 gg m-3 [Quinnet al., 1998]. Average continental
soil dust,were the most commoncalciumand magnesiumsalts, concentrations
of C1-areonlyabout0.15gg m-3 [Pueschel,
respectively(Figure 2e). 1995]. Near 5.5 km, C1- concentrationsmay range from
Aerosol wet massis the sum of dry massand liquid water. 0.0085to 0.02gg m-3 [Talbotet al., 1998;Li et al., 1997].
The effect of modeled RH on aerosol LWC was discussed in
Section 2.2. Table 5 indicates that near-surface aerosol wet
5.2. BC Forcing
masswashigherin theNorthernHemisphere (near44 !.tgm'3) Table 7 showsthat the modeledall-sky, globally and yearly
thanin theSouthernHemisphere (near20 !.tgm-3)andoverland averagedtropopausebase-caseforcing due to anthropogenic
(near45 gg m'3) thanovertheoceans (near27 gg m-3).Both (total) BC from fossil fuels and biomassburningwas +0.53 W
resultswereexpected,sinceloadingsof soil dust,whichmakes m-2. This estimatewas obtainedby treatingBC as core
up mostaerosolmass,werehigherin the NorthernHemisphere material in an internal mixture for radiative calculations. When
and over land than in the SouthernHemisphereand over the BC was treatedas externally-mixedinsteadof as core material,
oceans,respectively.Table 5 also shows that near-surface itsforcing
wascutin half,to+0.27W m-2.Whentherefractive
(a) 0 i,ii,,ud, ,,,,,hi, ,,.,,d , ,,,,,d , .,,,ml, ß .
'• 400'• Na2SO4 • ! ..L' (NH4)

2SO
-,-. 4
-
-
: 4 ß ..
/
1000-'
KHSO4 (NH4)2SO4 CaSOa
....... 'l ...... '/ ' ...... 'l ...... 'l ........ I ' '
'
NH HSO
4 4
NH C1
4
NH+
10-4 10-3 10-2 10-• 10ø 101 10-3 10-2 10-1 10ø
Aerosol mass
(ggm-3) Aerosol
mass
(ggm-3)
I I I I IIi1• , , , ,,,,,I , , , ,,,,,I
(½) 0 I ........ I ........ I ........ I , ,. o
200 2002
'• 400 400-.' KNO
•• 600
.......... 600-
• 800 NaC1 800NH4NO3,'"/

..' "k'
\ %•-
x' '•' •.. NaC1
?',N
1000-• NaNO
3 Na2CO3
- _ Na
+ ,
000- . NaNO
3 NHC1 4 NO- CI'
3
[ ' '''""l ' ....... I ........ [ ' '' ' '' ..... I ........ I ........ I ' ''
10-3 10-2 10-1 10ø 10-3 10-2 10-1 10ø
Aerosol
mass
(ggm-3) Aerosol
mass
(ggm'3)
I I I.,.,.I , , ,,,,,,I , , ,,,,,,I , , ,,,,,,I , , ,,,,,,I , ,,,,,,,I , , ,,,,,,I
(e) 0 I ''' (f) 0 ]
'• 400
200
CaSO
4
•,200
1
'• 400-1 OM2
OM1
• 600 •'--'"'
• -• MgCO
3 600 -'•': '"""-
• 800
1000
• kCa2+
Mg2+ •
800
•.CaCO3,
I '
- '
'_•.t..=..;.
'•
' ''""l
?
i02 -... %•-,,,
........ I ' '
oooq
[ ........ • ........ [ ....... [ , , •.....[
10-3 10-2 10-1 10ø 10-3 10-2 10-1 100 101
Aerosol
mass
(ggm-3) Aerosol
mass
(Bgm-3)
Figure 2. Modeled globally and yearly averagedvertical profiles of (a) sulfate, its salts, and water, (b)
ammoniumand its salts,(c) sodiumand its salts,(d) nitrate,chlorideand their salts,(e) calcium,magnesium,
and their salts,and (f) BC, OM, and insolublesoil compounds.
indices of BC were volume averaged with those of other The baseline anthropogenicsulfate forcing estimate was
componentsin an internal mixture, BC forcing increasedto obtainedwith sulfateinternally-mixed
as a shellcomponent.
A
+0.78W m-2.Thereason
for usingthecoretreatment
overthe test was run to calculate its forcing when it was treated as
external- or well-mixed internal treatment is discussedby externally-mixed. When externally-mixed, anthropogenic
Jacobson [2000]. sulfateforcingwas-0.53 W m-2 (versus-0.32W m-2 when
internally-mixed). In the external-mixture case, sulfate's
5.3. Sulfate, Ammonium, and Nitrate Forcing
chargewas balancedby H+ alone.In the internal-mixturecase,
Table 7 showsthat the modeledall-sky, globally and yearly its chargein submicronparticleswas balancedprimarily by
averagedtropopausebase-caseforcing due to anthropogenic NH4+.At thesame RH,2NH4+/SO42- andNH4+/HSO42- hydrate
sulfate was-0.32 W m-2. This compareswith the lesswaterthandoH+/HSO4
' or2H+/SO42-,
causing
thelesser
Intergovernmental
Panel on Climate Change(IPCC) [1996] negativeforcingwhen su!fatewas internally-mixedthan when
estimate
of-0.4 (-0.2 to -0.8) W m-2. it was externally-mixed.
Here, modelednatural sulfate tropopauseforcing was -0.34 The anthropogenic
and total ammoniumtropopause
forcings
of clouds in Table7 were+0.06and+0.11W m-2, respectively.
W m-2. ModeledTOA sulfateforcingin theabsence The
was about1.8 timesmorenegativethanthat in the presenceof positiveforcing due to ammoniumindicatesthat it contributes
clouds. Thus, clouds decreased the magnitude of negative to climate warming.The reasonfor the positiveforcing is
sulfate forcing. that, in the absenceof ammonium,sulfatechargeis balanced
Table7. ModeledSurface,Tropopause,
andTOA SolarPlusThermal-IRDirectForcing(W m-2),Averaged
Overthe24
Simulations,for All AerosolsExcept SpecifiedComponents(Exceptin the Case of "All aerosols",Where the Forcingis Due to
All Aerosol Constituents)
With clouds No
clouds
Surface Tropopause,
227 mbar, 11 km- TOA, 0.425 mbar
55km
Global Global SH NH Land Ocean Global Global Global Global Global Global
Baseline Baseline Baseline Baseline Baseline Baseline +15% -15% +50% -50% Baseline Baseline
RH Conc. Conc.
Natural PlusAnthropogenic
Aerosols
All aerosols -4.0 -1.37 -1.29 -1.44 -0.73 -1.62 -2.62 -0.51 - 1.99 -0.71 -1.23 -2.48
All soildustb -0.85 -0.13 -0.05 -0.21 -0.11 -0.14 -0.15 -0.09 -0.19 -0.07 -0.14 -0.23
All sea-spraydrops
c -0.54 -0.55 -0.68 -0.41 -0.08 -0.73 -0.99 -0.22 -0.80 -0.29 -0.54 - 1.1
All S(VI) -0.73 -0.66 -0.44 -0.87 -0.86 -0.58 -1.0 -0.37 -0.95 -0.34 -0.52 -0.94
All troposphericS(VI) -O.45 -O.45 -0.20 -0.68 -0.69 -0.35 -0.68 -0.24 -0.64 -0.23 -0.44 -0.84
All stratospheficS(VI) -0.28 -0.21 -0.24 -. 19 -0.17 -0.23 -0.32 -0.13 -0.31 -0.11 -0.08 -0.10
All natural S(VI) -0.40 -0.34 -0.31 -0.36 -0.31 -0.35 -0.51 -0.19 -0.49 -0.17 -0.21 -0.41
All NH4+ +0.12 +0.11 +0.07 +0.15 +0.15 +0.09 +0.15 +0.03 +0.17 +0.06 +0.10 +0.21
All Na+,Na2CO3 +0.04 +0.07 +0.09 +0.05 +0.001 +0.10 +0.26 +0.02 +0.10 +0.04 +0.07 +0.09
All K+,K2CO3 -0.02 -0.002 -0.003 -0.001 +0.002 -0.004 -0.002 -0.002 -0.003 -0.001 -0.003 -0.008
AllCa2+,CaCO
3 -0.03 +0.02 +0.02 +0.01 +0.006 +0.02 +0.02 +0.02 +0.03 +0.01 +0.02 +0.034
AllMg2+,MgCO3 -0.042 +0.005 +0.009 +0.001 +0.007 +0.004 +0.005 +0.005 +0.008 +0.003 +0.005 +0.03
All NO3' -0.07 -0.07 -0.07 -0.08 -0.10 -O.O6 -0.11 -0.03 -0.11 -0.04 -0.07 -0.12
All C1- -0.23 -0.29 -0.35 -0.23 -0.03 -O.39 -O.54 -0.12 -0.42 -0.15 -0.28 -0,51
All OM -0.35 -0.063 -0.06 -0.07 -0.13 -O.O4 -O.O6 -0.06 -0.09 -0.03 -0.07 -0.12
An SiO: -0.27 -0.17 -0.06 -0.27 -0.29 -0.12 -0.17 -0.17 -0.26 -0.08 -0.17 -0.25
All AI203 -0.19 +0,06 +0.02 +0.11 +0.18 +0.02 +0.06 +0.06 +0.09 +0.03 +0.06 +0.062
All Fe203 -0.30 +0.002 -0.001 +0.006 +0.04 -0.01 +0.002 +0.002 +0.003 +0.001 +0.002 -0.02
AnthropogenicAerosols
All anthropogenic
d -2.5 -0.12 -0.08 -0.16 +0.18 -0.24 -0.19 +0.06 -0.20 -0.05 -0.13 -0.89
Anthropogenic
soil duste -0.41 -0.07 -0.03 -0.12 -0.08 -0.07 -0.08 -0.05 -0.10 -0.04 -0.08 -0.15
Anthropogenic
S(VI)f -0.33 -0.32 -0.13 -0.51 -0.55 -0.23 -0.49 -0.18 -0.46 -0.17 -0.31 -0.53
Anthropogenic
BC, fossil - 1.4 +0.53 +0.26 +0.8C, + 1.23 +0.25 +0.55 +0.49 +0.78 +0.27 +0.52 +0.50
fuels and biomass
burning
g
AnthropogenicOM, fossil -0.32 -0.057 -0.05 -0.06 -0.12 -0.03 -0.06 -0.06 -0.08 -0.03 -0.06 -0.11
fuels and biomass
burning
h
Anthropogenic
NH4+.i +0.06 +0•06 +0.04 +0.08 +0.08 +0.05 +0.08 +0.02 +0.08 +0.03 +0.05 +0.12
Anthropogenic
NO3- J -0.04 -0.05 -0.04 -0.05 -0.06 -0.04 -0.07 -0.02 -0.07 -0.03 -0.04 -0.08
aSH is SouthernHemisphere,NH is Northern Hemisphere,RH is relative humidity. Forcing when clouds are present is calculated as
F_ac•'-F.g,.a.-x,c
'•',where
F ,l,isnetdownward
(downward-upward)
irradiance
(Wm'2),
thesubscript
gindicates
with
allgases
present,
the
su•s•ript
a •nchcates
withallaerosols
present,
thesubscript
cindicates
withclouds
present,
thesubscript
a-xindicates
withallaerosols,
except
for
the componentof interest,present,and the subscriptx alone indicateswith no aerosols,exceptthe componentof interest,present.Forcingin the
absence
of clouds
is calculated
as Fg,a $ -F ,a-x $ ß A positive
number
indicates
a netwarming
influence.
Thefirsttwosensitivity
columns
(+15%
RH)account
forthesensitivity
ofthe•ase-case
forcing
tochanges
inRHof+15%
(with
anupper
limitof99%RH),and
thenext
two
columns(+_50%Conc.) accountfor the sensitivityto changesin the concentrations
of all aerosolcomponentssimultaneously
of +50%.
c
bsio2,
+
A1203,
+ 2+
Fe20:t,
•+
CaCO
3,MgCO
3,K2CO
3,Na2CO
3,andassociated
H20(aq).
Na , K , Ca , Mg , CI', S(VI) fromsea-spray
dropsandassociated
H20(aq).
dAllanthropogenic
aerosols
include
70%of S(VI)<1 gmin diameter
[Langner
andRodhe,
1991],63%ofNO3-[Andreae,
1995,Table2], 50%
of NH4+ (estimate),
all BC, 90% of OM [Liousse
et al., 1996,Table1], and50% of soildust[Tegenet al., 1996].
%0% of all soil dust[Tegenet al., 1996].
f70%oftropospheric
S(VI)<1gmindiameter
[Langner
andRodhe,
1991].
g100% of BC.
h90%of all OM [Liousse
etal., 1996,Table1].
i50%ofallNH4+ (estimate).
J63%ofallNO3-[Andreae,
1995,Table2].
primarily
by H+. Since2NH4+/SO42-
andNH4+/HSO42-
hydrate were-0.05and-0.07 W m-2, respectively.
Thesenumbers
lesswaterthandoH+/HSO4 - or2H+/SO42-
atthesameRH, the comparewith-0.04 and-0.06 W m-2, respectivelyfrom
additionof ammoniumto an S(VI) solutiondecreases
liquid Andreae [1995] (scaledfrom NHnNO3 forcingswith the NO3-
water content,causingpositiveammoniumforcing. ßNHnNO3 mole ratio). Like chloride and sulfate, nitrate
Table7 showsthatanthropogenic
andtotalnitrateforcings enhances cooling.
1562 JACOBSON: GLOBAL DIRFL-W RADIATIVE FORCING
Table8. Comparison
of Tropopause
Forcings
(W m-2)in [1996] is about 1.6, an extrapolationof Cooke et al.'s results
the Presence of Clouds From This Work With Those From gives a forcing for biomassburning plus fossil fuel OM of
Other Studies a -J0.06W m-2.
When UV absorptionby OM was ignored here, the total
This Work Other Studies (anthropogenic)OM forcingdecreased
by 0.05(0.045)W m'2
to -0.11(-0.10)W m'2. SincetheUV absorbingproperties
of
Fg,
a,c'•-Fg,a_x,
c,• Fg,
x,c,•-Fg,
c organicsare uncertain,a conservativerange of the incremental
forcing due to UV absorptionby organics is suggestedas
Stratospheric
S(VI) -0.21(tropopause)
-0.1b +0.03 to +0.05 W m'2.
-0.08 (stratopause)
All anthropogenic -0.12 -0.5 (-0.25 to-1.0) c 5.5. Soil Dust Forcing
Anthropogenicsoil -0.07 +0.14d, -0.12e,-0.08f
dust Table 7 showsthat the modeledall-sky, globally and yeafly
Anthropogenic -0.32 -0.40 (-0.2 to -0.8)c averaged tropopausebase-case forcings due to total and
S(VI) - 1.1g,-0.3h,-0.95i,
-0.4J,-0.35k, -0.2e, anthropogenic
soil dust were -0.13and -0.07 W m'2,
-0.6 to-0.81, respectively. The model considered seven soil-dust
-0.55 to-0.81 m, components, SiO2(s), Fe203(s ), A1203(s), Na2CO3(s ),
-0.23 to-0.34 n , K2CO3(s ), CaCO3(s), and MgCO3(s). Much of SiO2(s)'s
-0.32 s
+0.53 from fossil fuels +0.1
negative solar forcing was offset by strongpositive thermal-
Anthropogenic
BC
from fossil fuels (+0.03 to +0.3)% IR forcing,giving SiO2(s) a modestnet forcing.A1203(s)had a
and biomass +0.20 to +0.36ø, net positive tropopause forcing because it absorbs in the
burning +0.20 m, +0.16 to visible and UV spectra. Fe203(s), which had a near-zero
+0.42s,+0.17t tropopauseforcing, is a slightly strongerUV and short-visible
from fossil fuels and
biomassburning
absorberthan is A1203(s), but A1203(s) absorbsmore than
+0.401,+0.27p, does Fe203(s) at long visible wavelengths.Table 8 showsthat
+0.78 q anthropogenicsoil-dustforcing estimatesfrom other studies
-0.06
varyfrom+0.14to -0.12W m-2.Theanthropogenic
soildust
AnthropogenicOM from biomassburning
estimateshere are within the rangesof the other estimates.
from fossil fuels -0.2 (-0.07 to-0.6) c
and biomass from fossil fuels and
burning biomassburning 5.6. Total and Anthropogenic Aerosol Forcing
-0.22½,-0.09r
from fossil fuels Table 7 showsthat the modeledall-sky, globally and yearly
-0.024 t averagedtropopausebase-casesolar plus thermal-IR forcing
due to anthropogenic
aerosols
(-0.12 W m'2) was slightly
aDefinitionsof the aerosolcategoriesare the sameas thosein negative. The magnitudes of all anthropogenic forcing
Table7. UnitsareW m'2. componentswere greater in the Northern Hemispherethan in
briansenet al. [1997]. the SouthernHemisphereand over land than over the oceans.
clPCC [ 1996].
Tropopauseforcing by anthropogenicplus natural aerosols
dTegen
etal. [1996].
CHartsenet al. [ 1998]. was-1.37W m'2. Table8 shows
thattheIPCC [1996]estimate
fMillerandTegen[1998]. of directforcingdue to all anthropogenic
aerosolsrangedfrom
gCharlsonet al. [ 1991]. -0.25to-1.0W m-2,witha suggested
valueof-0.5 W m'2. The
hKiehlandBriegleb[1993]. base-caseestimate
hereof -0.12W m'2 is lowerin magnitude
iTaylorandPenner[1994].
JChuang etal. [1997]. thanthe IPCC estimate,primarilydue to the treatmentof BC as
kFeichteret al. [1997]. core [Jacobson, 2000], but also due to differences in other
1Haywood
andRamaswamy
[1998],BCexternally
mixed. forcing constituents,as indicatedin Table 8.
mpenneret al. [1998], BC externallymixed.
nBoucherand Anderson [ 1995].
øHaywood
et al. [1997],BC externally
to well-mixed
internally. 6. Profiles of AerosolForcing
PThiswork, BC externallymixed.
qThiswork, BC well-mixed internally. Figure4 showsmodeledverticalprofilesof globallyand24-
rThiswork, no UV absorptionby OM. simulationaveragedaerosolforcingsfor the solar,thermal-IR,
SMyhre
et al. [1998],BC externallyto well-mixed
internally. and total spectra.The base-casesolarplus thermal-IRtotal
tCookeet al. [1999], Be externallymixed.
aerosolforcingchanged from-1.23to-4.0 W m-2 betweenthe
TOA andsurface.Total-aerosolsolarforcingwasnegativeand
becamemore negativebetweenthe TOA and surfacedue to
5.4. OM Forcing
decreases in valuesandincreases
in magnitudes of forcingby
Table7 showsthatthe modeledall-sky,globallyandyearly BC, soil dust, and OM, all of which absorb solar radiation.
averaged tropopausebase-case forcing due to total and Althoughless absorbingin the solarspectrumthan BC, soil
anthropogenic OM were -0.063 and-0.057 W m-2, dustandOM alsoabsorb,particularlyin theUV spectrum.
respectively.
The OM forcingestimates
did not accountfor OM Figure4 showsthat solarforcingsdue to sea-spray drops,
uptakeof liquidwater;thus,themagnitude of OM forcingwas chloride,and sulfatewere relativelyuniformwith altitude.If a
probablyunderestimated althoughthe sign shouldbe correct speciesis entirelynonabsorbing, irradiancesare equalat all
(additionof waterenhancesnegativeforcing). altitudes regardlessof the concentrationprofile, and the
For comparison, Cookeet al. [1999]estimated
OM forcing species' forcingprofileis uniformin the verticalandslightly
due to fossil fuels alone of-0.024 W m-2. Since the biomass- negative(due to net backscattering).
Sulfate and chlorideare
burning t• fossil fuel OM emissionsratio from Liousseet al. relatively nonabsorbingin the solar spectrum.Nevertheless,
JACOBSON: GLOBAL DIREET RADIATIVE FORCING 1563
0.5 (a) Naturalplusanthropogenic

aerosols
-0.5
-1
-1.5
-2
-2.5
o o o m
=gz ¸
(b) Anthropogenic
aerosols
'• 0.5
•: -
•0 0--
.•-•
E -
• -0.5
I I I I I
All All Soil S(VI) BC OM NH4+ NO3-
Figure 3. Modeledbase-casegloballyand yearly averagedall-sky tropopausesolarplus thermal-IRforcings

(W m-2) for (a) anthropogenic
plusnaturalaerosolconstituents
and(b) anthropogenic
aerosolconstituents.
For eachconstituent(exceptin the caseof all-anthropogenic aerosols),two sensitivitybars are given. The
first bar is the variationin forcingdue to RH variationsof +15%. The secondbar is the variationin forcingof
the individualconstituentdue to changesin concentrationof all constituentssimultaneously of +50%. In the
caseof all-anthropogenic aerosols,the third and fourthbarsare variationsin forcingdue to a +50% variation
in the global concentrationof BC and anthropogenicS(VI), respectively,alone. Base-caseforcings and
sensitivitybar numbersare given in Tables 7 and 9.
when they are removed from solution, aerosol composition was lowest. Figure 1 showsthat cloud cover was lowest at
changes, preventing their forcing profiles from being 30øN and S, where synoptic-scale
high-pressure
systems
perfectly straight. persist.
Total aerosol thermal-IR forcing was generally positive,
increasing in value and magnitude between the TOA and
surface. In the case of ammonium, thermal-IR forcing was
7. Effects of Clouds on Aerosol Forcing
negative, since sulfate particles contained less water, were The difference between the last two columns in Table 7
smaller, and absorbed less thermal-IR when ammonium was indicatesthat cloudsdecreasedthe magnitudeof TOA forcingof
presentthan when it was absent.In the case of stratospheric almost all aerosol constituents. Cloud cover blocked the solar
sulfate, positive thermal-IR forcing increasedwith increasing
optical effects of aerosolsover much of the Earth, especially
altitude, since, when sulfate was absent from the stratosphere,
over the tropics, where towering cumulonimbusand cumulus
thermal-IR wavelengthsthat would otherwisebe absorbedby congestuscloudsformed. In suchcases,aerosolswere exposed
sulfate,generallyescapedto space. to less incident solar radiation, which reduced the globally
Near-surfacethermal-IR forcing was greatestfor soil dust,
averagedlevel of aerosol-backscatteredradiation and increased
sea spray, and their components.Soil-dust particlesand sea-
the positivenessof aerosol forcing in comparisonwith when
spray drops are closer in diameter to typical thermal-IR clouds were absent. The effect of clouds on hygroscopic-
wavelengths(10 gm) than are most particles containingBC, aerosol forcing was large, since when thick clouds were
OM, or sulfate. Particles much smaller than a given present,the RH was high, aerosolLWC was large, and clouds
wavelength are inefficient absorbers and scatterers, and prevented hygroscopicaerosol componentsbelow them from
althoughmass absorption/scattering efficienciesdecreaseless
backscatteringradiation. Liao and Seinfeld [1998] similarly
rapidly with decreasing radius than do individual particlefound that cloudsreducedthe magnitudeof ammoniumsulfate-
absorption/scatteringefficiencies, the aggregated mass of particle forcing.
large particles usually exceeds that of small particles, In the case of BC, though, forcing increasedwhen clouds
decreasing the effect of particles much smaller than the were present compared with when they were absent. When
wavelengthon forcing. stratuswere present,BC residingabove or within stratuswere
Plate 2 showsthe modeledyearly-average(24-simu!ation) exposedto more reflected or multiple-scatteredradiation than
spatialdistributionof solarplus thermal-IRall-sky tropopause
when cloudswere absent.The additionalexposureallowed BC
forcingfor severalaerosolconstituents.
Forcingwas generally to absorbre.ore radiation when clouds were presentthan when
strongestwhere concentrationswere highest and cloud cover they were not, enhancingpositiveforcing by BC. Haywood et
1564 JACOBSON: GLOBAL DIREL2T RADIATIVE FORCING
[11111llll
(b) 0 (c) 0.
(a)0.' All I•
.
'- Allsoil IIi .
Al!
200-.aer½sols
[/ - 200-' s•a
Spray
.
' IR
•600- ;:IR
- 800'
800--.
1000--
//
•Solar+I
R .
- //Solar+IR"
.
- 1000'
Solar+IR '
ß
.... [ .... ! .... ! .... ! .... I .... .... I .... I .... I .... I I I I I I I I I I I I I
-5 -4-3 -2-1 0 1 -1.5 -1 -0.5 0 0.5 -1 -0.5 0 0.5
Forcing
(Wm'2) Forcing
(Wm'a) Forcing
(Wm'a)
I .... I,,,,I,,,,I,,
(e)0.'[ (0 e [.... ['i"[ .... [....
•....
,,,!
I'/....
A,,i .
/ Strat..'
s(vi) ' , 200'] II NO
3'
2oo-•
I[s(vI):'
,
400-
II
o
Solar
•6001
•[]riIR 600'
800' -..,
• +IR
,
,
' IR ! Solar[
8001 [ [ Solar -
!IR.'
180•
....
!....
!....
I....
[....
r• ,
1000' :1 t / +• '
"! .... I'"'! .... !'"'!" looo/.... !,.•,l .... •,,i, '
-1.5 -1 -0.5 0 0.5 1 -0.4-0.2 0 0.2 0.4 -0.15 -0.1 -0.05 0 0.05
Forcing
(Wm'2) Forcing
(Wm'a) Forcing
(Wm-a)
,,,,I,,,,I ,,
(g)0 '" [ .... ' .... (h) G (i) 0 ....I....'....!....... .
All All All OM .
200- Cl' - 200- .-
NH4+
400- -
• IR
800-
looO-
r• Solar
+LR ,
i .
'
'
10OO'
••' So,
a/
800-.
]ooo
z .,d'
,
IR
'"'1'"'1'"'1''"1
-0.1 0 0.1 0.2 -0.4 -0.2 0
Forcing
(Wm'2) Forcing
(Wm'2) Forcing
(Wm'a)
.... I .... I .... ! .... I ,,,,I,,,,I,,,,I,,,,I
0)o.F%O•
(k)
0Al203
!!
.
: I :: 200' 2OO- .
:: 400- IlZi
• Solar} ':IR
, 600- Solar
1000 [...•...,...[...[
-0.6 -0.4 -0.2 0 0.2
. 600'
Sol•
100{3 ....
-0.3 -0.2
I .... I ....
-0.1 0
...,
!';"
0.1
800•
_.,• Solar+IR'
1000 .... ! .... • .... ! ....
-0.3 -0.2 -0.1 0
IR
0.1
Forcing
(Wm'2) Forcing
(Wm-2) Forcing
(Wm&)
.u•
,..h,,,h,,,h,,,I,,,,h,,,h,,,h,,,I
m, All ql• (n) 0•,,,
I,,,I,,,I,,,I,,,I --- I,,,,I,,,,I,,,,I,,,,I,,,,h,,,I
Anthro-
• • aerosols•::• 200S(VI)
m"•
4q)iI •400
•1000•
8OO• 1'i •'•
Solar•
IR••.
••Sølar+xR
-3
•4 ::
8
• 1000-[
]Solar
1000'"]
-2
•/Sø'ar
iIR
['"'!'"'l'"'l'"'l""l'"'l""l
-1 0
' ' ' I ' ' ' I ' ' ' I ' ' ' I •' ' 1
-0.8-0.6-0.4-0.2 0 0.2
/ ....I'"'l'"'l'"'l""l'"'/
-2-1.5-1-0.5 0 0.5 1
Forcing
(Wm'2) Forcing
(Wm'2) Forcing
(Wm'2)
Figure 4. Modeledverticalprofilesof globallyandyearlyaveraged all-skysolar,thermal-IR,andsolarplus

thermal-IRdirectforcingfor severalaerosolgroupsor constituents.
JACOBSON:GLOBALDIRECTRADIATIVE FORCING 1565
60- ".•".•, ,..•%.,•'
•-30 J f ;•
-90 i ! • I ' I
-180 - 120 o 60 120 180
Longitude(degrees)
-9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 -4 -3 -2 -1 0 1 2 3 4 5
Total aerosolforcing(W m-2) Anthropogenic-aerosol
forcing(W m-2)
(a) (cl)
90 ..... • ..... I ..... I t..,,,, • ..... I ..... 90 .,,,,,I,,,,..tl, .... I ..... I ..... I .....
- '-'•-',-'•"•"•/"•:""- -• '•' •';' ß-•"'•.,..', ',w.- -
30- -• • '"-• • , -
- %½.,.... , .. -
.•½'-' •....
-90l,•,,, • ..... • ..... r , ,', , , • ..... i ..... J -90 ,'•'",,' • ,' ' ' ' I ..... I ' ' '' I''' ""
- 180 - 120 -(• 0 • 120 180 -180 -120 -• 0 • 120 180
•ngimde (de•s) •ngitude (degr•s)
I
-7 -6 -5 -4 -3 -2 -1 0 I 2 3 -5 -4 -3 -2 -1 0 1
Total-soilforcing(W m-2) Anthropogenic-S(VI)
forcing(W m-2)
(b) (e)
90 90...... • ..... I ..... • ..... • ..... • .....
,,,,, I, ,,,.,I , i,,, I,,,,, i, ,,, I,,, , •t
30-
ß• •.,
• -30- -30-
I ' •
-(•- • • .•
-90 -90 ....,'-i", ,•,•1• • •, , • , , , , , I '' ' ' ' • '"'"•"'"" ITM"i"",'"i'"',
I
-1.80 -120 -60 0 (xO 120 180 -180 -120 -• 0 (• 120 180
Longitude(degrees) •ngitude (degr•s)
-4 -3 -2 -1 0 0 1 2 3 4 5
Sea-sprayforcing(W m-2) BC forcing(W m-2)
(c) (f)
Plate2. Modeledglobaldistributions of yearlyaveraged

all-skytropopausedirectforcingdueto (a) total
aerosols;
(b) totalsoil;(c) sea-spray;
(d) anthropogenic
aerosols;
(e) anthropogenic-S(VI);
(f) BC.
al. [1998] also foundthat cloudsenhancedabsorptionby BC Table 9. Mean Tropopause Anthropogenic Aerosol
for a similar reason. Forcing(W m'2) in thePresence
of CloudsDueto +/-50%
Variationsin Concentrations
of IndividualAnthropogenic
8. Sensitivities to RH Aerosol Components
a
To examine the sensitivityof forcing to RH, 3-D RH fields
+50% Concentration -50% Concentration
predictedby the modelwereadjusted+15% of the RH valuefor
equilibriumcalculations.The adjustedRHs werenot allowedto
exceed 99%; thus, if the original RH was 90%, the +15% Anthropogenic
soildust -0.17 -0.11
Anthropogenic
S(VI) -0.21 -0.02
sensitivity RHs were 0.85x90% and min[1.15x90%, 99%], Anthropogenic
BC +0.06 -0.30
respectively.The purposeof the sensitivitytest was not to from fossil fuels and
suggestthat the uncertaintyin RH is +15%, but to providea biomass
burning
means of scalingforcing estimatesto higher or lower RHs. Anthropogenic
OM -0.15 -0.09
from fossil fuels and
Table 7 indicatesthat direct forcing was sensitiveto changes
biomass
burning
in RH, as suggestedpreviously [e.g., Pilinis et al., 1995;
Nernesure et al., 1995; Pilinis and Li, 1998; West et al.,
aCompare
withbaseline
value
of-0.12W m'2fromTable
7 (the
1998]. Hygroscopic aerosol components,such as sulfate, values
givenareforcings,
notdifferences
in forcing).
nitrate, and chloride, and sea-spray drops as a whole, were
more sensitive than were others, such as BC, SiO2(s),
halite, halite, and nitrum were the most common ammonium-,
A1203(s), and Fe203(s). Since soil dust consistsprimarily of
sodium-, chloride-, and nitrate-containingsalts, respectively.
nonhygroscopiccomponents,such as SiO2(s), A1203(s), and
(2) Sea spray and BC were estimated to be the most
Fe203(s), soil dust'swater uptakeand sensitivityto RH were
important aerosol types in the atmospherein terms of direct
less than were those of hygroscopicmaterial, as similarly
forcing and should be accounted for in climate-model
suggested by Li-Joneset al. [1998].
predictions.Chloride was the most important natural aerosol
componentin termsof direct forcing.
9. Sensitivitiesto Aerosol Loadings
(3) Ammonium forcing was positive, since it reduces
To examine the sensitivityof results to aerosolloadings, hydration in sulfate-containing particles. As such,
the concentrationsof all aerosolcomponentswere first varied anthropogenic ammonium appears to contribute to global
simultaneously by +50% for a forcing calculation(when all warming.
aerosols were present and when all aerosols minus the (4) The globally averagedbase-caseanthropogenicaerosol
component(s)of interestwere present).The purposeof this tropopausedirectforcingin the presenceof cloudsand with BC
test was to provide a means of scaling forcing estimatesto treated asa corecomponent was -0.12W m'2, indicating that
higheror lower loadings.The 50% variationsin concentration anthropogenicaerosolsmay not lead to so much cooling as
are sensitivityboundsrather than error bounds.Error bounds previouslythought. Total natural plus anthropogenicaerosol
in the time-dependent global distribution of aerosol- forcingin the presenceof cloudswasestimatedto be -1.4 W m'
componentconcentration are not possibleto determinetoday, 2. Resultsaresensitive to RH, aerosolburdens, andtheaerosol
sincesucherror estimatesrequireknowledgeof time-dependent mixing state.
observations over the globe, between the surface and (5) The treatmentof UV absorptionby organicsincreased
tropopause, of eachaerosolcomponent.As such,no previous the tropopauseforcing of organics(which remainednegative)
studynor compilationof studies(e.g., IPCC) of aerosolglobal on a globalscaleby an estimated 0.03-0.05W m-2 underthe
direct radiative forcing has reportedan actualerror boundin assumption that 10% of OM stronglyabsorbed UV radiation.
global concentrationsor direct forcing. Instead, they have (6) Solidformationincreasedtotal-aerosol directforcingby
reportedsensitivityboundsor rangesof valuesfrom different 0.03-0.05W m'2 ona globalscale.
simulations. (7) Total solar plus thermal-IR anthropogenicplus natural
In a third sensitivitytest, the concentrationsof individual aerosolforcingbecamemorenegativebetweenthe tropopause
,
anthropogenicaerosolcomponentswere changedby +50%, andsurface.Solarforcingdueto nonabsorbing components did

and the resultingeffect on anthropogenic aerosolforcing was not changesignificantlywith altitude.Thermal-IR forcingsof
examined. Table 9 shows results from this test. The table soil and sea-sprayconstituents were greaterthan thoseof all
shows that the base-casetropopauseanthropogenicaerosol other constituents.
forcingvaluehereof-0.12 W m'2 wassensitive to BC and (8) Cloudsdecreased themagnitude of TOA forcingof almost
sulfate concentration variations but less-so to soil dust and all aerosolconstituents.An exceptionwas BC. Positive BC
OM concentration variations. A +50% change in BC forcing increasedwhen cloudswere presentcomparedwith
concentration alone resulted in anthropogenicforcings of when they were absent.
+0.06and-0.3Wm'2, respectively. A +50%change in sulfate (9) SiO2(s) was estimatedto be the dominantforcing
concentrationresultsin a anthropogenic forcingsof-0.21 and componentin soil dust. A1203(s) tropopauseforcing was
-0.02W m'2, respectively. positive.
(10) Modeled near-surfaceaerosolLWCs were higher over
10. Summary the oceans than over land.
A globalstudyof the solarplusthermal-IRdirectforcingdue Acknowledgments. This work was supportedby grantsfrom the
Environmental ProtectionAgencyunderassistance
agreement
823186-
to multiple anthropogenicand natural aerosolsconstituents 01-0, theNationalScienceFoundation
underagreementsATM-9504481
was discussed. Conclusions are as follows. andATM-9614118, the NationalAeronauticsand SpaceAdministration
(1) Gypsum was calculatedto be the most commonsulfate- underthe New Investigator
Program(NIP) in EarthSciences,
and the
David and Lucile Packard Foundation and the Hewlett-Packard
containing salt in the global atmosphere. Sal ammoniac, companythrougha StanfordUniversityTermanFellowship.I greatly
appreciate helpfulcomments and suggestions

by AzadehTabazadeh, Finlayson-Pitts,B. J., and J. N. Pitts Jr. Atmospheric Chemistry:
AndyAckerman, PatHamill,PhilRussell,
andBobBergstrom. I would Fundamentals and ExperimentalTechniques, JohnWiley, New York,
alsoliketo thankPatriciaQuinnfor supplying aerosolcompositiondata 1986.
overtheoceanandAnnFridlindfor compiling statistics
fromthedataset Gillette, D. A., E. M. PattersonJr., J. M. Prospero,and M. L. Jackson,
for use in Table 3.
Soil aerosols.in AerosolEffectson Climate,editedby S. G. Jennings,
pp. 73-109, Univ. of Ariz. Press,Tucson,1993.
References Gong,S. L., L. A. Barfled,J. M. Prospero,D. L. Savoie,G. P. Ayers,J.-
P. Blanchet, and L. Spacek, Modeling sea-salt aerosols in the
Adams,P., J. H. Seinfeld,and D. M. Koch, Global concentrations
of atmosphere,2, Atmosphericconcentrations and fluxes,J. Geophys.
troposphericsulfate,nitrate,andammonium aerosolsimulatedin a Res., 102, 3819-3830, 1997.
generalcirculationmodel,J. Geophys.
Res.,104, 13,791-13,823, Hack, J. J., J. T. Kiehl, and J. W. Hurrell, The hydrologic and
1999. thermodynamiccharacteristicsof the NCAR CCM3, J. Clim., 11,
Andreae,M. O., Climatic effects of changingatmosphericaerosol 1179-1206, 1998.
levels,in WorldSurveyof Climatology,
vol. 16,FutureClimatesof the H•inel, G., Radiationbudgetof the boundarylayer, II, Simultaneous
WorM,editedby A. Henderson-Sellers, pp. 341-392,ElsevierSci., measurementof the mean solar volume absorptionand extinction
New York, 1995. coefficientsof particles,Contrib.Atmos.Phys.,60, 241-247, 1987.
Andreae,M. O., et ai., Airborne studiesof aerosolemissionsfrom Hansen,J., et al., Forcingsand chaosin interannualto decadalclimate
savannafires in southernAfrica, 2, Aerosolchemicalcomposition,J. change,J. Geophys.Res., 102, 25,679-25,720, 1997.
Geophys. Res.,103, 32,119-32,128,1998. Hansen,J. E., M. Sato,A. Lacis, R. Ruedy,I. Tegen, and E. Matthews,
Arakawa, A., and V. R. Lamb, A potential enstrophyand energy Climate forcingsin the industrialera, Proc. Natl. Acad. Sci., U.S.A.,
conserving
scheme
for the shallowwaterequations,
Mon. Weather 95, 12,753-12,758, 1998.
Rev., 109, 18 - 36, 1981. Haywood, J. M., D. L. Roberts,A. Slingo, J. M. Edwards,and K. P.
Bergstrom,
R. W., Extinctionand absorption coefficientsof the Shine, General circulation model calculations of the direct radiative
atmospheric
aerosolas a functionof particlesize,Contrib.Atmos. forcingby anthropogenic
sulfateandfossil-fuelsootaerosol,J. Clim.,
Phys.,46, 223-234, 1973. 10, 1562-1577, 1997.
Bergstrom,
R. W., and P. B. Russell,Estimationof aerosoldirect Haywood,J. M. and V. Ramaswamy,Global sensitivitystudiesof the
radiative effects over the mid-latitude North Atlantic from satellite
direct radiative forcing due to anthropogenicsulfate and black
andin situmeasurements, Geophys. Res.Lett.,26, 1731-1734,1999. carbonaerosols,J. Geophys.Res., 103, 6043-6058, 1998.
Berner,A., S. Sidla,Z. Galambos,C. Kruisz,R. Hitzenberger,H. M. ten Haywood,J. M., V. Ramaswamy,andB. J. Soden,Tropospheric aerosol
Brink, and G. P. A. Kos, Modal characterof atmosphericblack climate forcing in clear-sky satellite observationsover the oceans,
carbonsizedistributions,J. Geophys. Res.,101, 19,559-19,565,1996. Science, 283, 1299-1303, 1999.
Bond, T. C., R. J. Charlson,and J. Heintzenberg,Quantifyingthe He•ntzenberg, J., R. J. Charlson, A.D. Clarke, C. Liousse, V.
emissionof light-absorbing particles:Measurements tailored to Ramaswamy,K. P. Shine,M. Wendisch,and G. Helas, Meosurements
climatestudies,Geophys. Res.Lett.,25, 337-340,1998. andmodellingof aerosolsingle-scattering albedo:Progress,problems
Boucher,O., and T. L. Anderson,Generalcirculationmodel assessment and prospects, Beitr. Phys.Atmos.,70, 249-263, 1997.
of thesensitivity
of directclimateforcingby anthropogenic
sulfate Herman, J. R., N. Krotkov, E. Celarier, D. Larko, and G. Labow,
aerosols
to aerosolsizeandchemistry,J. Geophys.Res.,100, 26,117- Distribution of UV radiation at the Earth's surface from TOMS-
26,134, 1995.
measured UV-backscattered radiances, J. Geophys. Res., 104,
Charlson,R. J., J. Langner,H. Rodhe,C. B. Leovy,and S. G. Warren, 12,059-12,076, 1999.
Ferturbationof the northern hemisphereradiative balance by Hignett,P., J.P. Taylor, P. N. Francis,andM.D. Glew, Comparisonof
backscatteringfrom anthropogenicsulfate aerosols, Tellus, 43AB, observedand modeleddirect aerosolforcing duringTARFOX, J.
152-163, 1991. Geophys.Res.,104, 2279-2287, 1999.
Chin, M., D. J. Jacob, G. M. Gardner, M. S. Foreman-Fowler, and P. A. Hitzenberger,R., and H. Puxbaum,Comparisons of the measuredand
Spiro, A global three-dimensionalmodel of troposphericsulfate,J. calculatedspecificabsorption coefficientsfor urbanaerosolsamples
Geophys.Res.,101, 18,667-18,690,1996. in Vienna, Aerosol Sci. Technol., 18, 323-345, 1993.
Chow, J. C., J. G. Watson,E. M. Fugita,Z. Lu, D. R. Lawson,and L. L. Husar, R. B., J. M. Prospero,and L. L. Stowe, Characterizationof
Ashbaugh,Temporal and spatial variations of PM2.5 and PM10 tropospheric aerosolsover the oceanswith the NOAA advancedvery
aerosolin the SouthernCalifornia Air Quality Study,Atmos.Environ., high resolutionradiometeroptical thicknessoperationalproduct,J.
28, 2061-2080, 1994. Geop•,s. Res., 102, 16,889-16,909, 1997.
Chuang,C. C., J. E. Penner, K. E. Taylor, A. S. Grossman,and J. J. Intergovernmental Panel on Climate Change(IPCC), Climate Change
Walton, An assessmentof the radiative effects of anthropogenic 1995,in TheScienceof ClimateChange,editedby J. T. Houghtonet
sulfate,J. Geophys.Res.,102, 3761-3778, 1997. al., CambridgeUniv. Press,New York, 572 pp., 1996.
Cooke,W. F., and J. J. N. Wilson, A global black carbonaerosolmodel, Jacobson, M. Z., Global directradiativeforcingdueto multicomponent
J. Geophys.Res.,101, 19,395-19,409,1996. anthropogenicand natural aerosols(abstract),Eos Trans.,AGU, 79,
Cooke, W. F., C. Liousse, H. Cachier, and J. Feichter, Construction of a F157, 1998.
1øxl ø fossil fuel emission data set for carbonaceous aerosol and
Jacobson,M. Z., A studyof the global direct radiativeforcing due to
implementationand radiative impact in the ECHAM4 model, J. multicomponentanthropogenicand natural aerosols,in American
Geophys.Res., 104, 22,137-22,162, 1999. Meteorological Society Symposiumon InterdisciplinaryIssues in
Dentenet, F. J., G. R. Carmichael, Y. Zhang, J. Lelieveld, and P. J. Atmospheric Chemistry, pp. 39-41, Am. Meteorol. Soc., Boston,
Crutzen, Role of mineral aerosolas a reactive surfacein the global Mass., 1999a.
troposphere, J. Geophys.Res.,101, 22,869-22,889,1996. Jacobson,M. Z., Isolatingnitratedand aromaticaerosolsand nitrated
Deshler, T., D. J. Hofmann, B. J. John,son, and W. R. Rozier, aromaticgasesas sourcesof ultravioletlight absorption,J. Geophys.
Balloonborne measurements of the Pinatube aerosol size distribution Res., 104, 3527-3542, 1999b.
and volatility at Laramie, Wyoming during the summer of 1991, Jacobson,M. Z., Studyingthe effe•ts of calciumand magnesiumon
Geophys.Res.Lett., 19, 199-202, 1992. size-distributednitrate and ammc.
v.ium with EQUISOLV II, Atmos.
Ding, P., and D. A. Randall, A cumulusparameterizationwith multiple Environ., 33, 3635-3649, 1999c.
cloud-baselevels.J. Geophys.Res.,103, 11,341-11,353,1998. Jacobson,M. Z., A physically-basedtreatment of elemental carbon
Fang, M., M. Zheng,F. Wang, K. L. To, A. B. Jaafar,and S. L. Tong, optics:Implicationsfor global directforcing of aerosols,Geophys.
The solvent-extractableorganiccompoundsin ,•heIndonesiabiomass Res. Lett., 27, 217-220, 2000.
burningaerosols:- Characterizationstudies,A•.,•,,os•
Environ.,33, 783- Jayaraman, A., D. Lubin, S. Ramachandran, V. Ramanathan, E.
795, 1999. Woodbridge,W. D. Collins, andK. S. Zalpuri, Direct observations of
Feichter,J., U. Lobmann,and I. Schult,The atmosphericsulfurcycle in aerosolradiative forcing over the tropical Indian Ocean during the
ECHAM-4 and its impacton the shortwaveradiation,Clim. Dyn., 13, January-February1996 pre-INDOEX cruise, J. Geophys.Res. 103,
235-246, !997. !3,827-13,836, 1998.
Ferek, R. J., J. S. Reid, P. V. Hobbs, D. R. Blake, and C. Liousse, Kiehl, J. T., and B. P. Briegleb,The relativerolesof sulfateaerosolsand
Emission factors of hydrocarbons,halocarbons,trace gases, and greenhouse gasesin climateforcing,Science,260, 311-314, 1993.
particlesfrom biomassburning in Brazil, J. Geophys.Res., 103, Kiehl, J. T., and K. E. Trenberth•Earth'sannual global mean energy
32,107-32,118, 1998. budget,Bull. Am. Meteorol. Soc.,78, 197-208, 1997.
1568 JACOBSON: GLOBAL DIRECT RADIATIVE FORCliNG
Overview of the Tropospheric Aerosol Radiative Forcing

Kiehl, J. T., J. J. Hack, G. B. Bonan, B. A. Boville, D. L. Williamson, and
P. J. Rasch, The National Center for Atmospheric Research ObservationalExperiment(TARFOX), J. Geophys.Res., 104, 2213-
communityclimatemodel:CCM3, J. Clim., 11, 1131-1149,1998. 2222, 1999a.
Krekov, G. M., Models of atmosphericaerosols,in Aerosol Effectson Russell,P. B., J. M. Livingston,P. Hignett,S. Kinne,J. Wong,A. Chien,
Climate, edited by S. G. Jennings,pp. 9-72, Univ. of Ariz. Press, R. Bergstrom,P. Durkee, and P. V. Hobbs,Aerosol-inducedradiative
Tucson, 1993. flux changesoff the UnitedStatesmid-Atlanticcoast:Comparison of
Krotkov, N. A., P. K. Bhartia, J. R. Herman, V. Fioletov, and J. Kerr, values calculated from sunphotometerand in situ data with those
Satelliteestimationof spectralsurfaceUV irradiancein the presence measuredby airborne pyranometer,J. Geophys.Res., 104, 2289-
of troposphericaerosols,1, Cloud-freecase,J. Geophys.Res.,103, 2307, 1999b.
8779-8793, 1998. Saxena, P., and L. M. Hildemann, Water-soluble organics in
Langner,J., and H. Rodhe, A global three-dimensional model of the atmospheric particles: A critical review of the literature and
tropospheric sulfurcycle,J. Atmos.Chem.,13, 225-263, 1991. applicationof thermodynamics to identifycandidatecompounds, J.
Lelieveld, J., G.-J. Roelofs, L. Ganzeveld, J. Feichter, and H. Rodhe, Atrnos. Chem., 24, 57-109, 1996.
Terrestrial sourcesand distributionof atmosphericsulphur,Philos. Stelson,A. W., Urban aerosolrefractiveindex predictionby partial
Trans. R. Soc. London, 352, 149-158, 1997. molar refraction approach,Environ. Sci. Technol.,24, 1676-1679,
Li, S.-M., A.M. Macdonald,J. W. Strapp,Y.-N. Lee, and X.-L. Zhou, 1990.
Chemical and physical characterizationsof atmosphericaerosols Stenchikov, G. L., I. Kirchner, A. Robock, H.-F. Graf, J. C. Antuna, R.
over southernCalifornia, J. Geophys.Res., 102, 21,341-21,353, 1997. G. Grainger,A. Lambert,and L. Thomason,Radiativeforcingfrom
Liao, H., and J. H. Seinfeld, Effect of clouds on direct aerosol radiative the 1991 Mount Pinatubovolcaniceruption,J. Geophys.Res.,103,
forcingandclimate,J. Geophys.Res.,103, 3781-3788, 1998. 13,837-13,857, 1998.
Li-Jones, X., H. B. Maring, and J. M. Prospero,Effect of relative Sturges,W. T., and L. A. Barrie, Chlorine, bromine,and iodine in Arctic
humidityon light scattering by mineraldustaerosolasmeasured in the aerosols,Annos. Environ., 22, 1179-1194, 1988.
marineboundarylayer, J. Geophys.Res.,103, 31,113-31,121, 1998. Tabazadeh,A., M. Z. Jacobson, H. B. Singh,O. B. Toon,J. S. Lin, R. B.
Liousse,C., J. E. Penner,C. Chuang,J. J. Walton, H. Eddieman,andH. Chatfield, A. N. Thakur, R. W. Talbot, and J. E. Dibb, Nitric acid
Cachier, A global three-dimensionalmodel study of carbonaceous scavengingby mineraland biomassburningaerosols,Geophys.Res.
aerosols,J. Geophys.Res.,101, 19,411-19,432, 1996. Lett., 25, 4185-4188, 1998.
Miller, R. L., and I. Tegen, Climate responseto soil dust aerosols,J. Talbot, R. W., J. E. Dibb, and M. B. Loomis, Influence of vertical
Clim., 11, 3247-3267, 1998. transport on free tropospheric aerosols over central USA in
Myhre, G., F. Stordal,K. Restad,and I. S. A. Isaksen,Estimationof the springtime, Geophys. Res.Lett.,25, 1367-1370,1998.
direct radiativeforcing due to sulfateand sootaerosols,Tellus,50B, Tang, I., Thermodynamicand optical propertiesof mixed-saltaerosols
463-477, 1998. of atmospheric importance,J. Geophys.Res.,102, 1883-1893,1997.
Mylonas, D. T., D. T. Allen, S. H. Ehrman, and S. E. Pratsinis,The Taylor, K. E., and J. E. Penner, Responseof the climate systemto
sourceand size distributionsof organonitrates in Los Angelesaerosol, atmosphericaerosolsand greenhousegases,Nature, 369, 734-737,
Atmos. Environ., 25A, 2855-2681, 1991. 1994.
NASA Climate Data System Staff, Greenhouseeffect detection Tegen,I., A. A. Lacis,and I. Fung,The influenceon climateforcingof
experiment (GEDEX) selected data sets, NASA Goddard Space mineral aerosolsfrom disturbedsoils,Nature, 380, 419-422, 1996.
Flight Cent.,Greenbelt,Md., 1992. Tegen, I., P. Hol!rig, M. Chin, I. Fung, D. Jacob, and J. Penner,
Nemesure, S., R. Wagenet, and S. E. Schwartz, Direct shortwave Contributionof different aerosol speciesto the global aerosol
forcing of climate by anthropogenicsulfate aerosol:Sensitivityto extinctionopticalthickness:
Estimatesfrom modelresults.J. Geophys.
particle size, composition,and relative humidity, J. Geophys.Res., Res., 102, 23,895-23,915, 1997.
100, 26,105-26,116, 1995. Thomason,L. W., L. R. Poole,and T. Deshler,A globalclimatologyof
O'Brien, R. J., J. H. Crabtree,J. R. Holmes, M. C. Hoggan,and A. H. stratospheric aerosol surface area density deduced from
Bockian, Formation of photochemicalaerosol from hydrocarbons, StratosphericAerosol and Gas ExperimentII measurements:1984-
Environ. Sci. Technol., 9, 577-582, 1975. 1994,J. Geophys.Res.,102, 8967-8976, 1997.
Penner, J. E., R. J. Chadson, J. M. Hales, N. S. Laulainen, R. Leifer, T. Toon, O. B., and T. P. Ackerman, Algorithms for the calculationof
Novakov, J. Ogren, L. F. Radke, S. E. Schwartz, and L. Travis, scatteringby stratifiedspheres, Appl. Opt.,20, 3657-3660, 1981.
Quantify and minimizing uncertainty of climate forcing by Toon, O. B., C. P. McKay, T. P. Ackerman,and K. Santhanam,Rapid
anthropogenic aerosols,Bull. Am. Meteorol.Soc.,75, 375-400, 1994. calculationof radiativeheatingratesand photodissociation ratesin
Penner, J. E., C. C. Chuang, and K. Grant, Climate forcing by inhomogeneous multiplescatteringatmospheres, J. Geophys.Res.,94,
carbonaceousand sulfateaerosols,Clim. Dyn., 14, 839-851, 1998. 16,287-16,301, 1989.
Pilinis, C., andX. Li, Particleshapeandinternalinhomogeneity effects Venkataraman,C., J. M. Lyons, and S. K. Friedlander,Size distributions
on the optical propertiesof troposphericaerosolsof relevanceto of polycyclic aromatic hydrocarbonsand elemental carbon, 1,
climateforcing,J. Geophys.Res., 103, 3789-3800, 1998. Sampling, measurement methods, and source characterization,
Pilinis, C., S. N. Pandis,and J. H. Seinfeld,Sensitivityof directclimate Environ. Sci. Technol., 28, 555-562, 1994.
forcing by atmosphericaerosolsto aerosolsize and composition,J. Wang, W.-C., M.P. Dudek, and X.-Z. Liang, The greenhouse effect of
Geophys.Res.,100, 18,739-18,754,1995. tracegases,in World Surveyof Climatology,vol. 16, Future Climates
Platt,C. M. R., A parameterization of the visibleextinctioncoefficientof of the World,editedby A. Henderson-Sellers, pp. 341-392,Elsevier
ice clouds in terms of the ice/water content, J. Atmos. Sci., 54, 2083- Sci., New York, 1995.
2098, 1997. Welch, R. M., S. K. Cox, and J. M. Davis, Solar Radiation and Clouds,
Pueschel,R., Atmosphericaerosols,in Composition,Chemistry,and Meteorol.Monog. 17, Am. Meteorol. Soc.,Boston,Mass., 1980.
Climate of the Atmosphere,edited by H. B. Singh, p. 145, Van West,J. J., C. Pilinis,A. Nenes,andS. N. Pandis,Marginaldirectclimate
Nostrand Reinhold, New York, 1995. forcing by atmosphericaerosols,Atmos. Environ., 32, 2531-2542,
1998.
Quinn, P. K. and D. J. Coffman, Comment on "Contributionof different
aerosol speciesto the global aerosol extinction optical thickness: Whitby, K. T., The physicalcharacteristics
of sulfuraerosols,Atmos.
Estimatesfrom modelresults"by Tegenet al., J. Geophys.Res.,104, Environ., 12, 135-159, 1978.
4241-4248, 1999. Xie, P., andP. A. Arkin, Analysesof globalmonthlyprecipitation
using
Quinn, P. K., D. J. Coffman, V. N. Kapustin,T. S. Bates, and D. S. gauge observations,satellite estimates,and numerical model
Covert, Aerosol optical propertiesin the marine boundarylayer predictions,J. Clim.,9, 840-858,1996.
during the First Aerosol CharacterizationExperiment(ACE 1) and
the underlyingchemicaland physicalaerosolproperties,J. Geophys.
Res., 103, 16,547-16,563, 1998.
Reid, J. S., P. V. Hobbs, R. J. Ferek, D. R. Blake, J. V. Martins, M. R. M. Z. Jacobson,
Departmentof Civil andEnvironmental
Engineering,
Dunlap,andC. Liousse,Physical,chemical,andopticalpropertiesof TermanEngineering
Center,RoomM-13, Stanford University,
Stanford,
regionalhazesdominatedby smokein Brazil, J. Geophys.Res.,103, CA 94305-4020.(e-mail:jacobson@ce.stanford.edu)
32,059-32,080, 1998.
Russell, P. B., P. V. Hobbs, and L. L. Stowe, Aerosol propertiesand (Received
April 7, 2000;revisedJuly 14,2000;
radiative effects in the United States East Coast haze plume: An accepted
August28, 2000.)

Journal of Geophysical Research Atmospheres - 2001 - Jacobson - Global Direct Radiative Forcing Due To Multicomponent

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Journal of Geophysical Research Atmospheres - 2001 - Jacobson - Global Direct Radiative Forcing Due To Multicomponent

Uploaded by

Copyright:

Available Formats

21562202d, 2001, D2, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.

Global direct radiative forcing due to multicomponent

Departmentof Civil andEnvironmental

Abstract. Global simulationsof the compositionof and directforcingdue to aerosolscontaining

1. Introduction plus cloudy sky) conditions. Subsequently,Haywood et al.

BC Ion < 1 um > 1 um < 1 um > 1 um

Table 4. Comparisonof GATORG Predictions,Averaged 1 ,,I,,I,,I,,I,,I,,

Parameter Globe SH NH Land Sea

- 180 - 120 -60 0 60 120 180

0 20 40 60 80 100 120 140 160 180 200 220

(a) 0 i,ii,,ud, ,,,,,hi, ,,.,,d , ,,,,,d , .,,,ml, ß .

'• 400'• Na2SO4 • ! ..L' (NH4)

'• 400 400-.' KNO

• 800 NaC1 800NH4NO3,'"/

0.5 (a) Naturalplusanthropogenic

Figure 3. Modeledbase-casegloballyand yearly averagedall-sky tropopausesolarplus thermal-IRforcings

Figure 4. Modeledverticalprofilesof globallyandyearlyaveraged all-skysolar,thermal-IR,andsolarplus

60- ".•".•, ,..•%.,•'

Plate2. Modeledglobaldistributions of yearlyaveraged

anthropogenicaerosolcomponentswere changedby +50%, andsurface.Solarforcingdueto nonabsorbing components did

appreciate helpfulcomments and suggestions

Overview of the Tropospheric Aerosol Radiative Forcing

You might also like