An overview of the Amazonian Aerosol Characterization Experiment 2008

24 Pages·2010·6.95 MB·English

Atmos. Chem. Phys.,10,11415–11438,2010
Atmospheric
www.atmos-chem-phys.net/10/11415/2010/
Chemistry
doi:10.5194/acp-10-11415-2010
©Author(s)2010. CCAttribution3.0License. and Physics
An overview of the Amazonian Aerosol Characterization
Experiment 2008 (AMAZE-08)
S.T.Martin1,M.O.Andreae2,D.Althausen3,P.Artaxo4,H.Baars3,S.Borrmann2,Q.Chen1,D.K.Farmer5,
A.Guenther6,S.S.Gunthe2,J.L.Jimenez5,T.Karl6,K.Longo7,A.Manzi8,T.Mu¨ller3,T.Pauliquevis9,*,
M.D.Petters10,A.J.Prenni11,U.Po¨schl2,L.V.Rizzo4,J.Schneider2,J.N.Smith6,E.Swietlicki12,J.Tota8,
J.Wang13,A.Wiedensohler3,andS.R.Zorn2
1SchoolofEngineeringandAppliedSciencesandDepartmentofEarthandPlanetarySciences,HarvardUniversity,
Cambridge,Massachusetts,USA
2MaxPlanckInstituteforChemistry,Mainz,Germany
3LeibnizInstituteforTroposphericResearch,Leipzig,Germany
4InstituteofPhysics,UniversityofSa˜oPaulo,Brazil
5DepartmentofChemistryandBiochemistryandCooperativeInstituteforResearchintheEnvironmentalSciences,
UniversityofColorado,Boulder,Colorado,USA
6NCAREarthSystemLaboratory,NationalCenterforAtmosphericResearch,Boulder,Colorado,USA
7CenterofWeatherForecastandClimaticStudies(CPTEC-INPE),CachoeiraPaulista,Sa˜oPaulo,Brazil
8NationalInstituteofAmazonianResearch(INPA),Manaus,Brazil
9InstituteofAstronomy,GeophysicsandAtmosphericScience,UniversityofSa˜oPaulo,Brazil
10DepartmentofMarine,Earth,andAtmosphericSciences,NorthCarolinaStateUniversity,Raleigh,NorthCarolina,USA
11DepartmentofAtmosphericScience,ColoradoStateUniversity,FortCollins,Colorado,USA
12DepartmentofPhysics,LundUniversity,Lund,Sweden
13DepartmentofEarthandAtmosphericSciences,UniversityofNebraska,Lincoln,Nebraska,USA
*nowat: EarthandNaturalSciencesDepartment,FederalUniversityofSaoPauloatDiadema,Diadema,Brazil
Received: 4June2010–PublishedinAtmos. Chem. Phys.Discuss.: 30July2010
Revised: 2November2010–Accepted: 17November2010 –Published: 2December2010
Abstract. The Amazon Basin provides an excellent envi- large-scalecontextforthecampaign,especiallyasprovided
ronmentforstudyingthesources,transformations,andprop- bysatelliteobservations.Newfindingspresentedinclude:(i)
erties of natural aerosol particles and the resulting links be- aparticlenumber-diameterdistributionfrom10nmto10µm
tweenbiologicalprocessesandclimate.Withthisframework that is representative of the pristine tropical rain forest and
in mind, the Amazonian Aerosol Characterization Experi- recommended for model use; (ii) the absence of substan-
ment(AMAZE-08),carriedoutfrom7Februaryto14March tial quantities of primary biological particles in the submi-
2008 during the wet season in the central Amazon Basin, cron mode as evidenced by mass spectral characterization;
sought to understand the formation, transformations, and (iii) the large-scale production of secondary organic mate-
cloud-formingpropertiesoffine-andcoarse-modebiogenic rial;(iv)insightsintothechemicalandphysicalpropertiesof
aerosol particles, especially as related to their effects on theparticlesasrevealedbythermodenuder-inducedchanges
cloud activation and regional climate. Special foci included intheparticlenumber-diameterdistributionsandmassspec-
(1) the production mechanisms of secondary organic com- tra; and (v) comparisons of ground-based predictions and
ponents at a pristine continental site, including the factors satellite-basedobservationsofhydrometeorphaseinclouds.
regulating their temporal variability, and (2) predicting and AmainfindingofAMAZE-08isthedominanceofsecondary
understandingthecloud-formingpropertiesofbiogenicpar- organic material as particle components. The results pre-
ticlesatsuchasite. Inthisoverviewpaper,thefieldsiteand sentedhereprovidemechanisticinsightandquantitativepa-
theinstrumentationemployedduringthecampaignareintro- rametersthatcanservetoincreasetheaccuracyofmodelsof
duced. Observationsandfindingsarereported,includingthe theformation,transformations,andcloud-formingproperties
ofbiogenicnaturalaerosolparticles,especiallyasrelatedto
theireffectsoncloudactivationandregionalclimate.
Correspondenceto: S.T.Martin
(scot [email protected])
PublishedbyCopernicusPublicationsonbehalfoftheEuropeanGeosciencesUnion.
11416 S.T.Martinetal.: AnoverviewoftheAMAZE-08
1 Introduction The worldwide production mechanisms of secondary or-
ganic components from the oxidation of volatile organic
The Amazon Basin is a highly favorable environment for compounds (VOCs) have received considerable attention in
studyingthesources,transformations,andpropertiesofbio- the literature during the past few years (Chung and Sein-
genic aerosol particles (Andreae et al., 2002; Martin et al., feld,2002;TsigaridisandKanakidou,2003;Tsigaridisetal.,
2010). In contrast, advection in the widely polluted north- 2005;HenzeandSeinfeld,2006;TsigaridisandKanakidou,
ernmidlatitudesofNorthAmerica,Europe,andAsiamixes 2007; Healdetal.,2008). Therehasbeensignificantuncer-
together anthropogenic and natural components of parti- tainty regarding the contribution by primary organic emis-
cles. Amazonianstudiescan potentially isolatenaturalpro- sions,comparedtoinsituphotochemicalproductionofsec-
cessesand,indoingso,canprovideabaselineunderstanding ondaryorganicaerosol(SOA),tothemassconcentrationof
againstwhichanthropogeniceffectsonatmosphericaerosol organic particle material in urban areas, ranging from 80%
sources and properties can be understood (Andreae, 2007). primary in the study of Pandis et al. (1992) to 35% pri-
ThiskindofunderstandingisrelevantnotonlytotheNorth- mary in the studies of Zhang et al. (2005) and Volkamer
ern Hemisphere of today but also to the Amazon Basin of et al. (2006, 2007). Attribution between anthropogenic and
thefutureunderseveralpossibledevelopmentscenarios. Fo- biogenicsecondaryorganicmaterialremainsuncertain(Mil-
cused on these possibilities, the Amazonian Aerosol Char- let et al., 2006). Volkamer et al. (2006) and de Gouw and
acterizationExperiment(AMAZE-08)wascarriedoutinthe Jimenez (2009) emphasize that previous conclusions from
wetseasonof2008. global models may vastly underestimate the anthropogenic
A summary of the sources of Amazonian aerosol parti- contribution. There are missing sources of SOA produc-
clesandimportantinfluencesontheiratmosphericlifetimes tioninpollutedanthropogenicregions(deGouwetal.,2005;
and aging processes can be found in the review of Mar- Takegawaetal.,2006;Volkameretal.,2006)andalsointhe
tin et al. (2010). A short introduction is provided herein, free troposphere (Heald et al., 2005), as implied by obser-
as follows. In-Basin natural processes include emissions vations of excess organic mass concentrations compared to
both of primary biological aerosol particles (PBAPs) and best-model predictions. Carbon-14 dating of secondary or-
of gaseous sulfur, nitrogen, and carbon species from the ganicmaterialinareasheavilyinfluencedbyanthropogenic
ecosystem, oxidation of the gaseous species not just in the sourcesimpliesahighbiogenicfractionandthusanimpor-
atmospheric gas phase but also in cloud water to produce tant role of biogenic species (Klinedinst and Currie, 1999;
low-volatility particle-phase products (i.e., secondary par- Szidatetal.,2004). Incontrast,modelsfortheseconditions
ticle components), and removal of gases and particles by predict that the quantities of SOA material produced from
wet and dry deposition. The strongest anthropogenic influ- biogenicsourcesshouldbesmall.Theimplicationofthedis-
enceinsidetheBasinisbiomassburning,whichisespecially crepancybetweenobservationsandmodelsisthatsynergis-
prevalent in the dry season and in the southern part of the ticbiogenic-anthropogenicinteractionsexistthatarenotwell
Basinduringthattime(i.e.,outsideofboththetemporaland understoodatpresent(deGouwetal.,2005). Theseseveral
geographicscopeofAMAZE-08).Influencesoriginatingout observations, among others, have highlighted the need for
of the Basin and advecting into it include marine particles a more thorough understanding of SOA production (Gold-
fromtheAtlanticOcean,dustandbiomassburninginAfrica, stein and Galbally, 2007; Kroll and Seinfeld, 2008; Carlton
andgaseousspeciesfrombothregionsthatcanbeconverted etal.,2009;Hallquistetal.,2009).
withintheBasinintosecondarycomponentsofparticles.The TheAmazoninthewetseasonrepresentsapristineenvi-
particles from African sources arrive in the Basin sporadi- ronment having nearly pure biogenic aerosol particles (Ar-
callyaseventsandaremorecommonintheaustralsummer taxo et al., 1990; Gerab et al., 1998; Andreae et al., 2002;
thanwinter.Thebiomass-burningemissionsfromAfricacan Claeysetal.,2004;Andreae,2007;Martinetal.,2010),mak-
beregardedlargelyasananthropogeniccomponent. ing it an ideal laboratory to isolate natural SOA production
ThescientificobjectiveofAMAZE-08wastounderstand andtherebyprovideabaselineunderstandingagainstwhich
the formation, transformations, and cloud-forming proper- tomeasureanthropogenicinfluences. TheAmazonBasinis
ties of fine- and coarse-mode biogenic natural aerosol par- the region of highest VOC emissions globally (Guenther et
ticles, especially as related to their effects on cloud acti- al.,1995;Kesselmeieretal.,2009;Vrekoussisetal.,2009).
vation and regional climate. Special foci included (1) the Moreover, the high solar flux and large biogenic emissions
productionmechanismsofsecondaryorganiccomponentsat leadtoalargesourceofOHradicals(Lelieveldetal.,2008).
a pristine continental site, including the factors regulating Asaresult,SOAproductioninthegasphaseisinitiateddom-
theirtemporalvariability,and(2)predictingandunderstand- inantlybyOHradicals. Ozoneconcentrations(5to20ppb)
ingthecloud-formingpropertiesofbiogenicparticlesatsuch are relatively low in Amazonia, and O -initiated SOA pro-
3
a site. These foci can be broadly described as an investiga- ductionisthereforelessimportantthanthatoftheOHpath-
tion on the link between biological processes and climate, ways. Nitrate radical concentrations are low in Amazonia
as mediated by atmospheric chemistry (Barth et al., 2005; andarebelievedtocontributenegligiblytoSOAproduction.
Kelleretal.,2009). SOAproductionalsooccursincloudwater(Limetal.,2005;
Atmos. Chem. Phys.,10,11415–11438,2010 www.atmos-chem-phys.net/10/11415/2010/
S.T.Martinetal.: AnoverviewoftheAMAZE-08 11417
Carlton et al., 2008, 2009; Ervens et al., 2008). The fac- 2 Measurements
torsinfluencingtherelativeimportanceofgas-andaqueous-
phasepathwaysforSOAproductionremaintobewelldelin- The principal measurement site of AMAZE-08 was tower
eated,especiallyfortheenvironmentalparametersprevalent TT34 (02◦35.6570S, 060◦12.5570W, 110ma.s.l.) (Fig. 1).
intheAmazonBasin. It was located in the central Amazon Basin, 60km NNW
Predicting and understanding the cloud-forming proper- of downtown Manaus and 40km from the metropolis mar-
ties of atmospheric particles, especially for particles having gins. The site, accessed by a 34-km unpaved road,
high organic fractions, is essential for accurate predictions was within a pristine terra firme rainforest in the Reserva
ofconvectiondynamics,precipitation,andenergyfluxesand Biolo´gicadoCuieirasandmanagedbytheInstitutoNacional
theirintegrationwithclimateatallscales. Thepristinecon- de Pesquisas da Amazonia (INPA) and the Large-Scale
ditionsofthewetseasonoftheAmazonBasin,inwhichthe Biosphere-Atmosphere Experiment in Amazonia (LBA).
organicvolumefractionsofparticlesapproach90%(Fuzziet The base of tower TT34 was on a ridge, and the scale of
al.,2007),representanexcellentopportunityforthedevelop- hill-valley relief near the tower was ca. 50m. The forest
ment of this understanding. Both the development of warm canopy height near the tower varied between 30 and 35m.
cloudsbyactivationofcloudcondensationnuclei(CCN)and Housingfortheresearchersanda60kWdieselgeneratorfor
ofcoldcloudsbyactivationoficenuclei(IN)areimportant. power supply were located 0.33km and 0.72km downwind
PriortoAMAZE-08,icenucleihadneverbeforebeenstud- of TT34, respectively. The generator was separated from
iedinAmazonia,andstudiesoftheINpropertiesoforganic TT34bya50-mdeepvalley, withtheresearcherhousingat
particles in other parts of the globe have been hampered by thebottomofthevalley.
thedifficultyofdisentanglingthecontributionoforganicma- Climatological and meteorological information for this
terial from that of sulfate and dust, which have been mixed region of the Amazon Basin is provided by Andreae et
together. For this reason, measurements in the Amazon in al. (2002) and Araujo et al. (2002). The rainfall during
thewetseasonareparticularlyimportantfortheisolationof AMAZE-08 was consistent with the typical monthly values
theINpropertiesoforganicparticles. of the wet season (Fig. S1). The meteorological statistics
Robertsetal.(2001,2002,2003),Rissleretal.(2004),and recordedatthetopofthetowerduringAMAZE-08were23
Mircea et al. (2005) measured CCN number concentrations to27◦C(quartilesofdistribution)and80to99%RHinthe
intheAmazonBasinduringthewetseason. Thosestudies, daytime; thenighttimestatisticswere22to24◦Cand96to
however,didnotinvestigatethedependenceofCCNactivity 100%RH.Thequartilesofpressurewere9.9to10×105Pa.
ontheparticlesize, includingtheirpossibletemporalvaria- TherecommendationwastonormalizeAMAZE-08datasets
tion. The construction of accurate microphysical models of tostandardtemperatureandpressure(0◦C,105 Pa,dryair).
cloudformation,precipitation,andenergyfluxesisnotpos- Normalization facilitated quantitative comparisons among
siblewithoutinformationonthesize-dependentCCNactiv- AMAZE-08datasets,aswellaswithdatasetsofearlierand
ity. Moreadvancedmodelslinkedtovariablechemistryalso futuremeasurementcampaigns.
cannotbedevelopedwithoutmeasurementsonhowparticle In addition to TT34, there were three auxiliary sites.
chemistry potentially changes during the day and with size Tower K34 (2◦36.5450S, 60◦12.5580W, and 130ma.s.l.;
(i.e., from the condensation of secondary organic products 1.6km from TT34; 54m height) housed several instru-
aswellaspossibleheterogeneousagingreactionssuchasby ments directly at its top level (Ahlm et al., 2009). Site
OHradical). K23 (2◦38.3070S, 60◦9.4300W, and 108ma.s.l.; 7.56km
In this overview paper of AMAZE-08 (7 February to 14 from TT34) was used for tethered balloon soundings up to
March2008),thefieldsiteandtheinstrumentationemployed 800m. A light detection and ranging instrument (LIDAR)
during the campaign are introduced (Sect. 2). Observations and one site of the Aerosol Robotic Network (AERONET)
and findings are reported, including the large-scale context were located 19.1km from TT34 at the Silvicultura site
for the campaign (especially as provided by satellite obser- (02◦35.9130S,060◦02.2400W)(Althausenetal.,2009;Ans-
vations) (Sect. 3), a summary of AMAZE-08 publications mannetal.,2009).
available to date (Sect. 3.1), and five sections with new re- At TT34, air was entrained into three inlets fixed at
sultsnotpublishedelsewhere(Sects.3.2.1–3.2.5). Thetop- 38.75m to the top of the tower. The air was brought by
icsofthesesectionsareselectedtoprovideaperspectiveon samplinglinestoaground-levelcontainer(2.2×5.9×2.5m3)
the results from AMAZE-08 regarding the formation, the (Fig.S2). Theinletforgasesconsistedofaninvertedfunnel
transformations, and the cloud-forming properties of bio- havinganinsectscreenacrossit. Theinletforturbulent-flow
genic natural aerosol particles, especially as related to their aerosol sampling was a screen-covered open tube that was
effectsoncloudactivationandregionalclimate. placed very close to a sonic anemometer and positioned in
the direction of predominant wind flow. An inlet with an
aerodynamiccutoffnominallyofPM butactuallyofPM
10 7
for our flow conditions was used for laminar-flow aerosol
sampling. The sampling lines running from these inlets to
www.atmos-chem-phys.net/10/11415/2010/ Atmos. Chem. Phys.,10,11415–11438,2010
11418 S.T.Martinetal.: AnoverviewoftheAMAZE-08
40 m
Inlet TT34
35
Canopy
30
25
Sampling
20
Lines
15
10
Leipzig
Dryer 5
Container Ground
0
Fig.1.DiagramFoifgf.o1u.r-Dseicatgioranmteolefsfcooupri-csetoctwioenr(tTelTes3c4o)p,ischotowwinerg(pToTsi3t4io),nsshoofwininlegt,pcoasnitoipoyn,ssoafmipnllientg,clainneosp,yd,rsyaemr,palnindginlisntreusm, entcontainer.
InsetphotosareshownlargerinFig.S2.
dryer,andinstrumentcontainer.InsetphotosareshownlargerinFig.S2.
the container included a 6.4-mm (1/400OD) Teflon line for therefore calculated to range from 5 to 7µm for the range
gas sampling, a 12.7-mm (1/200OD; 10.9mm ID) stainless of flow rates employed during AMAZE-08. Therefore, the
steel linefor turbulent-flow aerosol sampling, anda 19-mm lowerandupperlimitsoftransmissionoftheinlet-sampling-
(3/400OD; 17.3mm ID) stainless steel line for laminar-flow lineassemblywerecalculatedas4nmto7µm,respectively,
aerosolsampling. Fromthesamplingheighttothecontainer, fortheconditionoflaminarflow. Comparisonbetweenthis
the three lines were wrapped together and encased by sec- calculatedsize-transmissionwindowandatypicalmeasured
tions of heating tape and insulation. Feedback control was number-diameter distribution suggests that there was mini-
usedtomaintainthetemperatureat30±1◦Ctoavoidwater mal loss of particle number concentration during transit in
condensationinthesamplinglines. the sampling line running from the inlet at the top of the
towertotheinstrumentsinthecontainerbelow. Thisexpec-
Mostoftheparticleinstrumentationinthecontainersam-
tationwasconfirmedforthelaminarlinebythegoodagree-
pledfromthelaminarline(Table1). Forthedimensionsand 33ment between the particle number concentrations recorded
flowofthelaminarsamplingline,theReynoldsnumbervar-
usingcondensationparticlecounters(CPCs)insidetheTT34
ied from 1200 to 2000 during the period of measurements.
container and those recorded by a CPC on the top of tower
Thecalculateddiffusionalandgravitationaldepositionlosses
K34 (Fig. 2). A self-regenerating automatic dryer was
for a particle of dynamic shape factor of 1 and a den-
placed on the top of the container and intercepted the lam-
sity of 1000kgm−3 indicate 50% transmission cutoffs of
inar sampling line prior to entrance into the container. The
4nmand10.5µmthroughthesamplinglines,withincreased
dryerconsistedoftwodiffusiondryersinparallel,onedrying
transmission between those sizes (www.seas.harvard.edu/
the aerosol flow while the other was bypassed and regener-
AerosolCalculator).AReynoldsnumberof2000approaches
ated by dry air (Tuch et al., 2009). The flow was switched
the turbulent regime, and in this case the calculated 50%
from one dryer to the other at a threshold of 40% RH. By
transmissioncutoffsshiftto13nmand3.5µmforlossesby
this sequence, the RH was kept between 20 and 40% when
gravitational settling, diffusional impaction, and inertial de-
measuredbyanin-linesensor(Fig.3). Thecabinethousing
position. The upper limit of the aerodynamic cutoff was
Atmos. Chem. Phys.,10,11415–11438,2010 www.atmos-chem-phys.net/10/11415/2010/
S01.0T2.M,4a2r–ti1n,e0t1a,l..s:yAhPno.mveerhvCiew.soomfttAheAMAZE-08 /0102/1/01/ten.syhp-mehc-som1t1a4.w1w9w
thatTab
ble
OPC OPC SEM UV-APS DMACCNC/C CPCTD HR-ToF-AMS SMPS PAR LIDAR MAAP F-HR-ToF-AM HR-ToF-AMS CFDC2CO3O2NO,NOCO TEOMPM-10 TEOMPM-2.5 eganbefore1.Instrume
P S an
cartridgesgasadsorptionWELASparticlecounterwhite-lightopticalGrimm1.108particlecounterlaser-sourceopticalsamplesmicroscopyoffilterscanningelectron particlesizerultravioletaerosolwithCCNC&CPCCdifferentialmobilityanalyzercountercondensationparticlethermodenuderAerosolMassSpectrometertime-of-flightAerodynehigh-resolutionparticlesizerscanningmobilityonsitemeteorologicaldatapyranometerradiationphotosyntheticallyactiveranginglightdetectionandphotometermultiangleabsorption andsonicanemometermodeoftheHR-ToF-AMSEddycovarianceflux (AMS)AerosolMassSpectrometertime-of-flightAerodynehigh-resolution(icenuclei)diffusionchambercontinuous-flowcarbondioxideozoneoxidesofnitrogencarbonmonoxideunder10µmforparticulatematteroscillatingmicrobalancetaperedelementunder2.5µmforparticulatematteroscillatingmicrobalancetaperedelement Instrument ndstoppedafter,thetimetsmakingmeasurements
NCAR MPI-PC MPI-PC MPI-BG MPI-BG MPI-BG MPI-PCNCAR MPI-PC Lund INPAINPA INPA IfT IfT CU-Harvard Harvard-CU CSUCPTECCPTECCPTECCPTEC CPTEC CPTEC Organization periodofduringAM
28Feb 7Feb 7Feb 15Feb 7Feb 14Feb 7Feb28Feb 7Feb 21Feb ← 18Feb 7Feb 7Feb 7Feb17Feb17Feb17Feb17Feb 17Feb 17Feb start AMAAZE
Z-0
14Mar 14Mar 14Mar 14Mar 14Mar 14Mar 14Mar→ 14Mar → → → 14Mar 14Mar 8Mar14Mar14Mar14Mar14Mar 14Mar 14Mar stopTT34 E-08.8atth
e
n/a laminar laminar laminar laminar laminar laminarlaminar laminar laminar ←→←→ ←→ laminar turbulent turbulent laminargasgasgasgas ground ground linestartstoK34 Location mainsiteTT34
p a
s
←→ startstopSilvicultura wellasth
e
0.2 5 1.2 1 5 1 10.6 0.1 1 n/an/a n/a n/a 5 Asabove 0.1 1.5n/an/an/an/a n/a n/a Flow(Lpm auxilia
) ry
1800 60 6 2000 300 10000 1n/a 150 300 n/an/a n/a n/a 60 Asabove 150 btw10and60m300300300300 300 300 CharacteristicTim sitesK34and
in e(s) the
n/a 0.2–10 0.3–20 n/a 0.5–10 0.03–0.3 001>.n/a 0.08–0.8 0.01–0.5 n/an/a n/a n/a Asabove 0.06–0.8 15<.n/an/an/an/a 10< 25<. DiameterRange(µm Silvicultura.L
) ef
t
chromatographymassspectrometryanalyzedbythermaldesorptiongas- number-sizedistributionofparticles number-sizedistributionofparticles particleimagesofthebiologicalfractionparticles,includingidentificationnumber-sizedistributionofcoarse-modeofparticlessize-resolvedCCNactivity particlenumberconcentrationparticlevolatilitystandardversioninletwasmodifiedfromasaboveforotherHR-TOF-AMS; number-sizedistributionofparticlesrelativehumidity;temperaturewindspeedanddirection;rainfallamount;downwardandupwardradiationfluxes coefficientatdifferentwavelengthsverticalprofilesofbackscatterandextinctionofdepositedparticleslightabsorptionat673nmandfluxmassspectraorganic,sulfate,nitrate,chloride,ammonium)non-refractorychemicalcomponents(includingchemicallyresolvedfluxesofsubmicronspectraofparticlessizedistribution;high-resolutionmassammonium,andchloride;aerodynamicparticleparticle-phaseorganic,sulfate,nitrate,size-resolvedmassconcentrationsof icenucleiactivityofparticles particlemassconcentration particlemassconcentration MeasuredQuantity andrightarrowsindicatemeasure
m
e
n
ts
www.atmos-chem-phys.net/10/11415/2010/ Atmos. Chem. Phys.,10,11415–11438,2010
5 80-EZAMAehtfoweivrevonA :.latenitraM.T.S
11420 S.T.Martinetal.: AnoverviewoftheAMAZE-08
aRO T
ndUniversityofSaoPaulo(USP).˜esearch(INPA),LundUniversity(Lund),Mrganizationabbreviations:CenterofWeath sunphotometer(AERONET) nephelometer totalfilterfine&coarsefiltersnucleuscounterCCNCcloudcondensation aethalometermassspectrometerPTR-MSproton-transfer-reactionozoneoxidesofnitrogen Instrument able1.Continued. (cid:3)(cid:2)(cid:2)(cid:3)3Concentrationcm 112505000000000 TKT3344tocwonetrainer
axerF O
PlanckInstituteforCorecastandClimatic USP USP10Feb USP10FebUSP10Feb USP10Feb USP10Feb NCAR8FebNCAR8FebNCAR8Feb rganizationstart 20000 MMara1r28(cid:3)9 TMimare1(cid:2)3UTC(cid:3) Mar14
hemistry–Studies(CP → →→ → → 27Feb27Feb27Feb stopTT34 (cid:3)(cid:2)(cid:2)(cid:3)3cm 1500
PT n
articleCEC),Co 10m10m laminar laminar toptoptop line Lo entratio 1000
hemistry(MPIC-PC),MaxPlanckInstituteforChemistry–BiogeochemistryloradoStateUniversity(CSU),UniversityofColorado(CU),LeibnizInstitute 27Febn/an/a→ 10300 n/adaysn/adays 1300 53600 n/a300n/a300n/a120 startstopstartstopFlow(Lpm)CharacteristicTime(s)DFprttohheonK34Silviculturaipeegrbra.Aoeon2cationmsMte.ehonnPAutitaopesZrwrdltyEeiesca-Farels0isetng8autdtnbhrsirueredie tum2egtnpo.ibotopTetsnrhpoiantowsimFatatceiflinohkittbomoimraswtlepConcegeontoneoiranlhs.vmiedtcwdudslie2eer,eaupaumarive.rnKr5ylrrnbvc5eraPeeee0tmdoeee0n3rrms00aKetrat0tef4batrmassrrteon3to,gasciigoinmaito4acmwnmngnmtt1srgln,isoeehep8t−tnoFsiwudogo:ilnMntf03iciidrifphthgntu0hoaeuetoeart.igmosncohdecwnoSwtoanphefpebohaf4laroreoenuettdosew.ptrrohrlhsseftllmbltTecah,TueeudoatoosnmhhwlteTccsatiniteaaeplhpoon3eenhcostsaidn4nsd0eseeikcuftricp0dorntteadreaasaohi:iwitest0ncvspen1krimeTm0alnteeiid.aedeaeitk6strnmethseiMbrsecnok1rtsgTenreasoseunmud.aewttT(cid:2)6nccamssinpinUTntodt3rakatfreesoTeb4nrTesummrFsfnoeCcttabsp3ahcaroimto(cid:3)p,legin4hieofltcrkowlny0r.dedionTeirao6AeehSneanssnwtK:vTmridtn0o4Micdcera3sa03eete.oihdratT4tsoAn4nKdeeTlt.iustttTuwnZss3rhoiphts(cadua3t4EeglTaaeletaygep4eyirrs-hrbsogptd.0ntebiaeuihonesso,8ecit(pesedet)arerlTvtp.ttpvaesxsochwseahili1oecomekltnieset2hyuerrnnhreu:tae.trpts0teasswoomosnalo0xtipfttTiaedolbonhctttlrsthhhhneouueeaoeteeeets--fr-rtrwhsbtceiheeoorlennnieToiscnvreTatelnhe3bdtt4orataavonanteidndaod5rnisi0n.acs)0temoecnrwpsmtleehi−mcnetng3s-
(MPIC-BG),NforTroposphe n/a all alln/a n/a n/a n/an/an/a iameterRange gicnorcenretesaaisnweeadrrmpbryeiorurtphtoatnosai2mn0sp%ildienignthbtehyeciondnsitstraturiinmbeuer.ntitosTn.hTleihnReesHseitnthspeiodrieenftothoreef
ationalCenterforAtmosphericResearch(NCAricResearch(IfT),NationalInstituteofAmazon wavelengthsscatteringcoefficientatmultipletotalorganiccarbonanalysischromatography;metalsanalysisbyPIXanionandcationanalysisbyion integratedCCNnumberconcentrationsatmultiplewavelengthslightabsorptionofdepositedparticlesorganiccarboncompoundsspeciationandconcentrationofvolatile (µm)MeasuredQuantity tttrtttttahhhhaehheirleeeaervTnaotaaitethuianiceinirvegernsum-erhttjucoreutptumohusrhseacmneorturebdlacmpeppuitornuiturtliiunieiettmdroositdenicen.rptttarhyiwt.nlthesoagflebaosTiomyirennswiwhnnfiapseutmaeiclrdsirraaentepmenr,ugflloctriewohuonber3lfewnu.gii4btndlch)aeyleoTinwntnu3Nthhwetnaee◦aasdasCioafisnepmnNorreaoepndrrpatrdrtrfihlwieuiiRmaeonescalsdHngoesrupdro-2relrtemoei3oexbnfnsa◦cyepset4oChruea0foawer,rrinvexaatasgithpllsodotoeeahmefn(ood7wctsepuwu0oieigobbo%ntarhhnyef---
R),ian E; ated at a Reynolds number of 5000 to 10000, and the
Atmos. Chem. Phys.,10,11415–11438,2010 www.atmos-chem-phys.net/10/11415/2010/
S.T.Martinetal.: AnoverviewoftheAMAZE-08 11421
Feb14
40 30
(cid:37) (cid:76)
35 (cid:72)
y
Humidit 30 atitude 10
e L
v
ati 25 (cid:45)10
el
R
20
(cid:45)30
00:00 06:00 12:00 18:00 00:00 (cid:45)80 (cid:45)60 (cid:45)40 (cid:45)20 0 20
Time UTC Longitude
Fig.F3.igR.el3at.iveRheulmaitdiivtyeinhtuhemlaimdiintayr-flionwtaheerosloalmsaimnpalirn-gfllionewusainegrothseodlrysearmdespcrliibnedgblyinTeuchet 6000
al.(2009).Thedryerconsistsoftwodiffusiondryers(silic(cid:72)agel),a(cid:76)ndtheaerosolflowisdivertedintoonedryer
whileutshiensegcotnhdeonedirsyreegrendereatsecd.riWbheedntbheythTresuhcolhdReHtfaorl.reg(e2n0er0at9io)n.isTreahcehedd,rthyeearerocsoolnfl-owis m 4000(cid:76)
sistsoftwodiffusiondryers(silicagel),andtheaerosolflowisdi- (cid:72)
divertedtothatdryer.Themethodofregenerationisadry-airflowobtainedfromacompressor. ht
verted into one dryer while the second one is regenerated. When g
ei 2000
thethresholdRHforregenerationisreached,theaerosolflowisdi- H
vertedtothatdryer. Themethodofregenerationisadry-airflow 0
obtainedfromacompressor. 0 50 100 150 200 250
BackwardTime hr
correspondingdiffusional,gravitational, andinertialdeposi- 6000
(cid:72) (cid:76)
tion losses suggest 50% transmission cutoffs of 17nm and m 4000(cid:76)
3.1µm, with 100% transmission between these cutpoints. (cid:72)
ht
Figure 3 The line for gas sampling entered the container without in- eig 2000
H
terception. Forsomegasmeasurements,additionalsampling
lines were temporarily place3d5at various heights along the 0
0 50 100 150 200 250
tower(Karletal.,2009).
BackwardTime hr
Theinstrumentshousedinsidethecontainerandthesam-
pling line used by each instrument are listed in Table 1.
Fig. 4. Ten-day HYSPLIT backtrajectories at 200m (red) and
ThesubsetofinstrumentsrelatedtoSeFcti.g3..42.“TNeenw-dRayesHulYts”SPLIT backtrajectories at 200m (red) and (cid:72)30(cid:76)00m (blue) during AMAZE-08. Each point
3000m (blue) during AMAZE-08. Each point along a trajectory
include: a condensation particle counter (CPC, TSI model
alongatrajectoryrepresenrtespraes1e2n-ths sate1p2.-hThsteepg.reeTnhepogrineetnmpaoriknst tmhaerklosctahteiolnocoaftiotonwoefrTT34.
3010); a scanning mobility particle sizer (Lund SMPS de- towerTT34.
scribed by Svenningsson et al., 2008); an optical paFrtigiculree 4
counter (OPC, Grimm model 1.108); an ultraviolet aero-
dynamic particle sizer (UV-APS, TSI model 3314; Huff-
particles and the transformations of particle components.
man et al., 2010); the LIDAR instrument discussed previ-
The backtrajectories arriving at the site at 200 and 3000m
ously; two high-resolution time-of-flight Aerodyne Aerosol
above ground level are shown every 12h from 7 February
MassSpectrometers(HR-ToF-AMS)(DeCarloetal.,2006),
to 14 March 2008 in Fig. 4 (Draxler and Hess, 1998) (see
one belonging to Harvard University Environmental Chem-
Fig.S3for500,1000,2000,and4000m). Thefigureshows
istry Laboratory and equipped with a standard inlet (Liu et
thatthesynoptic-scaletradewindscamepredominantlyand
al., 2007) and the other belonging to the Max Plank In-
consistentlyfromthenortheast,travellingacrosstheAtlantic
stitute – Particle Chemistry (MPI-PC) group and equipped
Ocean and then over 1600km of nearly pristine forest be-
with a novel inlet shifted to favor larger diameter particles;
fore arrival at the research site. The low-altitude flow was
a thermodenuder as described in Wehner et al. (2002); and
from the northeast, changing to easterlies at mid-altitude.
acontinuous-flowdiffusionchamberformeasurementsofice
Local wind measurements at the top of the research tower
nuclei(Rogersetal.,2001).
confirmedthatthedaytimewindsweredominantlyfromthe
north and northeast (Fig. 5). At nighttime, the winds stag-
36
natedattimesandcouldcomefromanydirection.
3 Observationsandfindings
The trajectories represented in Fig. 4, suggesting smooth
Back-trajectory analyses and satellite observations covering trajectoriesofairparcels, areaHYSPLITproductthatuses
the trajectory paths during the period of AMAZE-08 pro- a Lagrangian framework and omits treatment of vertical
vide large-scale information concerning both the sources of mixing.Treatingatmospherictransportoveraten-dayperiod
www.atmos-chem-phys.net/10/11415/2010/ Atmos. Chem. Phys.,10,11415–11438,2010
11422 S.T.Martinetal.: AnoverviewoftheAMAZE-08
from day to day and the weak vertical shear in the lower
a Daytime 3–5 km of trade winds both suggest that the representation
in Fig. 4 is a reasonable first-order description of airmass
NNNN
(cid:72) (cid:76) transport. IntheparticularsituationofAMAZE-08,thereis
strongobservationalsupportforthevalidityofthisapproach.
NNNNWWWW NNNNEEEE Ben-Amietal.(2010)documentedthetransportofdustand
smoke across the Atlantic Ocean along the path suggested
bythetrajectoriesinFig.4byusingacombinationofremote
sensingandsurfaceobservations.
An implication of the backtrajectories shown in Fig. 4 is
that,inadditiontoin-Basinsourcesandtransformations,the
WWWW EEEE
particlesoftheAmazonBasinduringAMAZE-08werealso
affectedattimesbysoildusttransportfromnorthernAfrica
0000....1111
and biomass burning in equatorial Africa. Satellite obser-
0000....2222
vations support this possibility and further indicate that the
0000....3333 long-rangetransportofmaterialfromAfricatotheresearch
sitewasepisodicduringAMAZE-08(Ben-Amietal.,2010).
SSSSWWWW SSSSEEEE
Comparison of the ten-day backtrajectories and satellite-
based MODIS fire counts (Figs. 4 and 6) suggests that ad-
SSSS vection of biomass-burning products from Africa may have
occurred. LIDAR measurements record a large plume of
b Nighttime
dustandsmokeoverCapeVerdeon3February2008thatar-
NNNN rivesoneweeklaterattheresearchsiteintheAmazonBasin
(cid:72) (cid:76)
(Ansmann et al., 2009). Particle optical depth recorded by
MODISontheAQUAsatelliteduringtheperiodofAMAZE-
NNNNWWWW NNNNEEEE
08isalsosupportiveofthepossibilityoflong-rangetransport
of biomass-burning products (Fig. 7). In contrast, an influ-
ence of regional biomass burning appears less plausible be-
cause of the absence of fires along the in-Basin trajectories
(Fig. 8). The absence of markers of fresh biomass burning
WWWW EEEE bothintheparticle-phasemassspectra(e.g.,elevatedsignal
intensity at m/z 60 and 73) and in the gas-phase mass spec-
0000....1111 tra(e.g.,thebenzene:acetonitrileratio)furtherconfirmsthis
0000....2222 conclusion(Chenetal.,2009;Karletal.,2009).
Elevated African dust emissions occurred several times
0000....3333
duringtheperiodofAMAZE-08,asapparentintheMODIS
SSSSWWWW SSSSEEEE observationsofFig.7. Basedonthebacktrajectoriesduring
these times, a portion of these emissions may have reached
theresearchsiteapproximatelyoneweeklater. Simulations
SSSS
using the global chemical transport model “GEOS-Chem”
0(cid:45)1 1(cid:45)2 2(cid:45)3 (cid:62)3ms(cid:45)1 for the AMAZE-08 time period, as well as measurements
of fine-mode dust concentrations at the research site, sug-
gestseveralsignificantevents(>0.5µgm−3)ofAfricandust
Fig.5. DaytimeandnighttimeFiwg.i5n.dDraoystiemsemanedansiughrtetidmeatwtihndertoosepsomfeatsouwreedratTtThe3t4opdoufringtrAanMspAorZt tEo-t0h8e.reTsheaercchosnicteenco-mpared to background con-
tower TT34 during AMAZE-08. The concentric circles represent centrations (<0.2µgm−3) (Prenni et al., 2009a), in gen-
tric circles repFriegsueren 5t the fractitohenafrlacftrioenqaulefrneqcuyenocfywofiwnidnsdsffrroommddififfefreenrtesnetctsoercdtiroerctidoinrse.ctions. Within each sector, the
eralagreementwithearlierreportsofepisodicdustintrusion
Withineachsector,thecolorcodingofradialdistanceindicatesthe
color coding of radial distance indicates the relative frequencies of the wind speeds.(Prospero et al., 1981; Andreae et al., 1990; Artaxo et al.,
relativefrequenciesofthewindspeeds.
1990;Swapetal.,1992;Formentietal.,2001).
In addition to the influence of synoptic-scale upwind
sources, the vertical structure of the local atmosphere must
with a Lagrangian parcel model is a strong simplification be considered for the interpretation of nighttime compared
of the complexities of atmospheric movement. Vertical ex- to daytime measurements at TT34. Nighttime and day-
change and cloud processing take place along the path of time differences in the height and the stability of the plan-
transport and modify concentrations and properties of air- etary boundary layer largely bifurcate the scientific context
masstracers.Nevertheless,theconsistencyofthetrajectories of the measurements at TT34 during these two periods. At
Atmos. Chem. Phys.,10,11415–11438,2010 www.atmos-chem-phys.net/10/11415/2010/
37
S.T.Martinetal.: AnoverviewoftheAMAZE-08 11423
Fig.6. GFriidgd.e6d.stGatriisdtidcaeldssutmatmisatriiceaslosfufimremsaorvieersAoffrificareasnodvSeoruAthfrAicmaeraincad.SPoauntehlsAarmeeprriecpaa.rePdaunseilnsgatrheeperigehpta-rdeadycuosminpgosthiteesofthe
ClimateModelingGrid(CMG)FireProductsoftheMODISCollection-5ActiveFireProduct(Giglioetal.,2006). Redisscaledfrom0to
eight-daycompositesoftheClimateModelingGrid(CMG)FireProductsoftheMODISCollection-5Active
25pixelfirecounts,withcorrectionsforcloudcover. Somepixelshavemorethan25firepixels(e.g.,maximumvalueforscenesshownis
1000pixeFlifirerePcrooudnutsc)t.(Giglioetal.,2006). Redisscaledfrom0to25pixelfirecounts,withcorrectionsforcloudcover.
Somepixelshavemorethan25firepixels(e.g.,maximumvalueforscenesshownis1000pixelfirecounts).
nighttime, katabatic (i.e., hill-valley) flows are important et al. (2008). In the morning (12:00UTC is 08:00LT), the
(Araujoetal.,2002). Thenighttimemeasurementstherefore PBL-topheightwas200m. From13:00to14:00, thePBL-
hadamicroscalefetch,incontrasttothemesoscalereachof topheightincreasedfrom400to800m.By14:30,cloudfor-
the daytime measurements. As a result, the nighttime mea- mation occurred. The figure also shows that residual layers
surements at TT34 were influenced by local activities, such presentat500and800minthemorningbecameentrainedby
asnighttimeecosystememissions. Althoughthereisconsid- localnooninthePBL.Anotherlayerofparticlesisapparent
erable interest in the nighttime activities of tropical forests at approximately 2000m, possibly the result of long-range
(Graham et al., 2003a; Guyon et al., 2003a; Gilbert, 2005), transportoracloudresiduallayer,amongotherpossibilities.
thefocusofAMAZE-08wasonmesoscaleprocesses. The height-resolved backscatter coefficients averaged
Thedevelopmentoftheboundary-layerstructureatnight acrosstwodifferenttimeperiodsareshownintherightpanel
wasdrivenbytheformationofanocturnalplanetarybound- of Fig. 11. In the first period (12:30–13:10UTC), a shal-
ary layer (PBL) that had a height comparable to the local low convective boundary layer and an overlying residual
topographical relief, which was 50 to 100m around TT34. layer from the day before are revealed in the profile of the
InthisregionoftheAmazonBasin,thethicknessofthecon- backscatter coefficient. A sharp, steep gradient occurs at
vectivePBLtypicallyvariesfrom100–200matnightto1to the top of the PBL. A lofted particle layer is located be-
2km when fully developed in the mid afternoon (Martin et 38tween1300mand2400matthattime.Anhourlater(14:07–
al., 1988; Garstangetal., 1990; Fischetal., 2004). Several 14:20UTC), the vertical structure of the backscatter coeffi-
balloonsoundingsmadeduringAMAZE-08wereconsistent cient evolves significantly, showing the growth of the PBL
with these values (Figs. 9 and 10), with a PBL-top height up to 800m and the mixing of the lofted residual layer into
near700to800mbylocalnoon. the PBL. The second lofted layer has descended, having its
The development of the planetary boundary layer, which lowerlimitatabout1100m.
can be observed by LIDAR measurements, is shown in the Some of the measurement periods at TT34 were affected
leftpanelofFig.11for13March2008.Theblacklineshows by local pollution, especially at night. The nocturnal hill-
thePBL-topheightthatwasderivedbythemethodofBaars valleyflow,incombinationwiththeshallowboundarylayer,
www.atmos-chem-phys.net/10/11415/2010/ Atmos. Chem. Phys.,10,11415–11438,2010
11424 S.T.Martinetal.: AnoverviewoftheAMAZE-08
25Jan2008to01Feb2008 18Feb2008to25Feb2008
02Feb2008to09Feb2008 26Feb2008to04Mar2008
10Feb2008to17Feb2008 05Mar2008to12Mar2008
0 0.2 0.4 0.6 (cid:62)0.8
FFigi.g7..7P.artPicalertoicpltiecaolpdetipcthalobdseeprvtehdobybstheervMeOdDbISyitnhsteruMmeOntDonIStheinAsQtrUuAmseantetlloitne(tLheevyAeQtaUl.,A20s0a7;teRlelimteer(eLtealv.,y20e0t8a).l.S,h2ow00n7ar;e
thelevel-3eight-daycompositesofCollection5.
Remeretal.,2008). Shownarethelevel-3eight-daycompositesofCollection5.
led at times to a circulation of generator exhaust to the by their respective horizontal scales. Unless stated specifi-
research station, and these events (which were evident in callyotherwise,thedataassociatedwiththeperiodsoflocal
the data setsFiogfumre u7ltiple instruments; Fig. S4) were tagged pollution(constituting24%ofthemeasurementperiod)were
and logged into a community file accessed by AMAZE-08 excluded from AMAZE-08 analyses. Besides these tagged
researchers. Intersection of the generator plume with the periods,thesitewasfreeoflocalanthropogenicinfluences.
research station was abrupt (i.e., total particle concentra-
tionscouldincreaseabove1000cm−3 for15min,dropback 3.1 Publishedresults
to 300–400cm−3 for 15min, and then jump again above
1000cm−3; Fig.S4). Theabruptbehaviorreflectedthenar- SeveralrecentpublicationsdescribefindingsfromAMAZE-
row horizontal extent and the shifting nature of the gener- 08. Karl et al. (2009) employed measurements of isoprene
ator plume. The duration of one of these events was typi- photo-oxidationproductstoinfermissingchemistryinstan-
cally several hours. Other local pollution (both at day and dardmodelsofatmosphericchemistry,complementingother
night)arrivedattimesfrommetropolitanManaus. Thispol- recent theoretical (Peeters et al., 2009), laboratory (Paulot
lutionwasapparentinmultipledatasets(Fig.S4),wascon- et al., 2009), and field (Kuhn et al., 2007; Lelieveld et al.,
tinuous, lasted for time periods of up to a day or more (the 2008) results on this topic. These results suggested more
longest period was 38h), and corresponded to local winds rapid oxidation of biogenic emissions than presently imple-
from the direction of Manaus. The greater duration of the 39 mented in chemical transport models and, correspondingly,
Manausplumecomparedtothatofthegeneratorisexplained enhanced formation of oxygenated products that might fa-
vorably partition to the particle phase (Guyon et al., 2003a,
Atmos. Chem. Phys.,10,11415–11438,2010 www.atmos-chem-phys.net/10/11415/2010/