investigating the effects of atmospheric aging on the radiative
TRANSCRIPT
Investigating the effects of atmospheric aging on the direct radiative properties and
climate impacts of black‐ and brown‐ carbon aerosol
Jesse H. Kroll, Colette L. Heald
Department of Civil and Environmental Engineering, MIT
Paul Davidovits, Andrew T. Lambe
Department of Chemistry, Boston College
14 November 2014
BC climate forcing: Large, complex, uncertain
[Bond et al., 2013]
Simple lifecycle of atmospheric BC
optical properties
transport
emission deposition
Simple lifecycle of atmospheric BC
optical properties
transport aging
emission deposition
This project
Black/brown carbon aerosol is chemically dynamic, subject to atmospheric aging reactions; these can lead to dramatic changes in physicochemical properties, and therefore climate forcing effects.
A complete understanding of this aging, and representation of this aging within models, is necessary for the accurate simulation of global direct radiative forcing.
Key BC aging reactions
+1) Coagulation
ox + SO2 H2SO4 (or (NH4)2SO4) coating
(NH3)
ox + NO22) Condensation NH4NO3 coating
NH3
+ VOC ox
SOA coating
ox3) Heterogeneous + OH/O3 surface oxidation oxidation
Effects of aging on BC properties
1) Enhancement of light absorption by coatings (“lensing effect”) [e.g., Schnaitner 2005, Bond et al. 2006, Schwarz et al. 2008, Lack et al. 2009, Cappa et al. 2012]
+ VOC ox
“darker”
2) Increased water‐uptake ability by coated or oxidized BC
+ H2O
Higher hygroscopicity can lead to …
‐more efficient light scattering (due to larger particles from water uptake) ‐ shorter atmospheric lifetimes due to increased wet deposition
(‐more facile activation to form cloud droplets)
Simple lifecycle of atmospheric BC
optical optical properties properties
emission deposition
transport aging
Simple lifecycle of atmospheric BrC
“Brown carbon”: OC that absorbs in the UV/visible (distinct molecules)
? SOA aging
“browner”
formation
?
emission deposition
BrC
“whiter”
This project
Black/brown carbon aerosol is chemically dynamic, subject to atmospheric aging reactions; these can lead to dramatic changes in physicochemical properties, and therefore climate forcing effects.
A complete understanding of this aging, and representation of this aging within models, is necessary for the accurate simulation of global direct radiative forcing.
Major questions: ‐ what are the most important atmospheric aging transformations of BC/BrC?
‐ what effects do aging reactions have on climate‐relevant properties of BC/BrC?
‐ how do these aging reactions impact BC/BrC direct radiative forcing?
Approach: Laboratory + Modeling
Development of global modeling framework for representation of aging, climate effects �
calculation of DRF
Laboratory studies of BC/BrC aging reactions: Changes to chemical + optical properties
IPCC AR5 Estimates that Black Carbon is the 2nd Largest Warming Agent in the Atmosphere.
(but that’s not what models say)
IPCC AR5
Bond et al., 2013 How can these be reconciled?
ACCMIP (Shindell et al., 2013)
IPCC AR4
AeroCom II (Myhre et al., 2013)
BC (FF/BF)
BC (BB)
0 0.2 0.4 0.6 0.8
TOA DRF (Wm‐2)
Observations Also Suggest That Models Overestimate BC
AeroCom means in black, HIPPO obs in colour Obs in black, AeroCom models in colour [Schwarz et al., 2010] [Koch et al., 2009]
AeroCom models overestimate BC over Americas & remote Pacific by factor ~5‐8.
Aging Processes may Reconcile Both Mass and Absorption Constraints
AbsorpƟon ↑ LifeƟme ↓
Aging GEOS‐Chem integrated with RRTMG (GC‐RT) (Heald et al., 2014)
+ Emission Absorption (AERONET) Mass Concentrations (ARCTAS,
EUCAARI, HIPPO + surface)
New Model Aging Processes for BC
Hydro‐phobic
Hydro‐philic
Old Assumptions
1.15 days
Hydro‐phobic
Hydro‐phobic
New Assumptions
Anthropogenic
Biomass burning
Hydro‐philic
Hydro‐philic
Sulfate, etc.
Organic components
4 hours (also increase fraction emitted as
hydrophillic to 70%)
k = 1/τ = a [SO2] [OH] + b
Based on several lab and field studies, esp Liu et al. (2010); Akage et al. (2012)
Good simulation near source (with or without new aging).Modified aging scheme results in shorter lifetime and better
simulation of low concentrations in remote locations. Vastly better than AeroCom. Generally within a factor of 2.
Impact of New Model Aging Processes on Simulation of BC
HIPPO Continental (Near‐Source)
Good
Better
Still Bad
Considering Absorption Enhancement of BC
Anth BC
BB BC
Smaller size, wider size range
Mie calculation
Larger size, narrower size range
Mie calculation
Absorption Coefficient
Absorption Coefficient
Absorption enhancement from coating (AE=1.1)
Absorption enhancement from coating (AE=1.5)
Also “Most Absorbing” Simulation : Set AE=2 and standard aging mechanism (longer lifetime)
(Akage et al., 2012; Schwarz et al., 2006; 2007; 2008; Lack et al., 2012; Dubovik et al., 2002; Shamjad et al., 2012; Moffet et al., 2009; Knox et al., 2009; Kondo et al., 2011; Lack et al., 2012; Moffet and Prather, 2009; Bond et al., 2006; Cappa et al.,
2012)
Adding Brown Carbon to GEOS‐Chem
Aromatic SOA
Brown Carbon
Get RI from field measurements Absorption
Mie calculation Coefficient 50% of biofuel POA 25% of fire POA
MAE=1 m2/g MAE=0.3 m2/g
Absorption of BrC is highly uncertain ‐ we choose upper‐range estimates
Including Brown Carbon is Critical to Capturing the Spectral Dependence of AERONET AAOD
*AAOD product here using lev2 SSA with lev1.5 AOD
Measurements Still Suggest Absorption is Underestimated
Better able to capture the spectral AAOD with our “best” simulation (including BrC), but still biased low (especially in some biomass burning regions).
Can “scale up” our model to match observations (Bond et al., 2013) – emissions or optics?
*AAOD product here using lev2 SSA with lev1.5 AOD
Our Work Suggests Smaller BC DRF Required to Match All Observational Constraints
Scaled GCRT
BC (FF/BF)
BC (BB)
BrC
"Best" GCRT
Standard GCRT
IPCC AR5
Bond et al., 2013
ACCMIP (Shindell et al., 2013)
IPCC AR4
AeroCom II (Myhre et al., 2013)
0 0.2 0.4 0.6 0.8
TOA DRF (Wm‐2)
Brown Carbon contributes 35% of the warming from carbonaceous aerosols. BC DRF is less than methane and tropospheric ozone.
Suggests that controlling BC is less effective for climate mitigation.
[Wang, et al., ACP, 2014]
Lab Study 1: Heteogeneous oxidation of BC
with Kevin Wilson (LBNL), Manjula Canagartna and Paola Massoli (ARI)
AtomizerN2
Pump
T
O2
Drier
SP-AMS
BC source
[Browne, et al., submitted]
SP‐AMS with VUV ionization
VUV (12 eV photons)
Soot‐Particle Aerosol Mass Spectrometer: Real‐time measurement of mass, composition of particles (OC, BC, inorganics)
Ionization efficiencies PAHs: 7‐8 eV Most organics: 9‐10 eV BC fragments (e.g. C3): ~11.5 eV
[Onasch et al. 2012]
Heterogeneous aging: BC “cores” unaffected
300 C x/20 (Hz)3 SMPS Integrated volume (ug/m )250
Cx accounts for the majority200
of the flame soot mass 150
100
12:00 AM 2:00 AM 4:00 AM2/2/2013
0.5
Frac
tion
of C
x spectrum remainsCCx
0.4 C 2 constant with aging: C 3 BC “core” is inert on the0.3 C 4 timescale of atmospheric agingC5 - C100.2
0.1
120.0 0.5 1.0 1.5 2.0x10-3OH exposure (molecules cm s)
Changes to adsorbed organic species N
orm
aliz
ed to
Cx
0.16
0.12
0.08
0.04
0.00
80x10
CH CHO CHO>1
no oxidation
20 40 60 80 100 120
-3
60
40
20
0
~2 days oxidation
20 40 60 80 100 120
Organics associated with BC (representing small fraction of total particles) undergo dramatic chemical changes
Continual changes with oxidant exposure
Approximate days0.0 0.5 1.0 1.5 2.0 2.5
1.2
Sig
nal n
orm
aliz
ed to
Cx
1.0
0.8
0.6
0.4
0.2
0.0
250x109200150100500 OH exposure (molecules s cm
-3 )
Reactant CxHy +
CxHyO+
CxHyOz +
Rapid loss of organic species
[Org
]/[O
rg] 0
1.0
0.8
0.6
0.4
0.2
0.0
0 5 10 15 20 Approximate days of OH exposure
Lifetime of adsorbed organics is short
Follow‐up experiments: Water uptake (hydrophobic hydrophilic)
m/z 202 m/z 216
Loss of adsorbed PAHs
[PAH
]/[PA
H] 0
1.0
0.8
0.6
0.4
0.2
0.0
m/z 202: pyrene
m/z 216: benzo[a]fluorene
0 5 10 15 20 Approximate days of OH exposure
Long‐timescale “survival” of PAHs (morphology)
Next: Coating + heterogeneous ox. experiments
Flow tube reactor
soot particle generation
reagent/oxidant preparation
analytical instrumentation
SPMS (size) CPMA (mass)
SP‐AMS (composition)
EAD (surface area) CAPS‐SSA, CRD‐PAS
(absorption, scattering)
March 2015
)
Experimental matrix
BC source ‐ fractal soot from McKenna burner (denuded at 300oC) ‐ also atomized black carbon spheres
Particle size ‐monodisperse, 30‐300 nm
Aging type ‐ OH + SO2 (sulfuric acid coating, het. ox) ‐ OH + NH3 + SO2 (ammonium sulfate coating, het. ox) ‐ OH/O3 + VOCs (SOA coating, het. ox ‐mixed coatings
Parameterization
Changes to composition/hygroscopicity + optics as a function of atmospheric exposure (to match representation in Wang et al. 2014)
Key optical parameters determined, included in a “lookup table” (or interpolated function) based on experimental results
particle size mass extinction efficiency
relative humidity single scattering albedo
wavelength asymmetry parameter
mixing state (BC:org:SO4)
Summary/conclusions
‐Modeling vs measurements of BC: models overestimate loadings, underestimate aerosol absorption
‐ Aging processes can affect both concentrations (via changes to deposition) and optical properties (via changes to coatings); need for an improved understanding, description of such processes
‐ Global modeling results: Improved agreement between predicted, measured BC loadings and properties (but AAOD still underestimated!)
‐ Laboratory results: Heterogeneous oxidation an efficient way to change organic components of soot; oxidation can dramatically decrease“brown”‐ness of brown carbon
‐ Next step: Laboratory results implementation in models
Acknowledgements
Manjula Canagaratna Kevin Wilson Eleanor Browne Andrew Lambe Paola Massoli Thomas Kirchstetter Xuan Wang Paul Davidovits Timothy Onasch Jonathan Franklin Douglas Worsnop Kelsey Boulanger
David Ridley
Xiaolu Zhang Christopher Cappa
Joshua Schwarz Anne Perring Ryan Spackman
Dantong Liu Hugh Coe
Anthony Clarke
Publications from this Project: Heald et al., ACP, 2014 Wang et al., ACP, 2014 Browne et al., JPCA, submitted Lambe et al., ES&T, 2013
RD83503301