Transcript PPT

Near-term climate forcers and climate policy:
black carbon and methane
Daniel J. Jacob
Gorillas and chimpanzees of climate change
CO2: the 800-lbs gorilla
Methane and BC: the chimps
Do we care about the chimps?
Radiative forcing of climate change
Terrestrial flux
Fout ~ T 4
Solar flux
Fin
• Global radiative equilibrium: Fin = Fout
• Perturb greenhouse gases or aerosols
radiative forcing F = Fin - Fout
• Global equilibrium surface temperature responds as To ~ F
Radiative forcing referenced to emissions, 1750-2011
• Radiative forcing from methane
emissions is 0.97 W m-2, compared to
1.68 W m-2 for CO2
• Radiative forcing from black carbon
aerosol (BC) is 0.65 W m-2, highly
uncertain
• Together methane and BC have
radiative forcing comparable to CO2
• But atmospheric lifetimes of methane
(10 years) and BC (~1 week) are
shorter than CO2 (> 100 years)
[IPCC, 2014]
Metrics of climate response to a radiative forcing agent
for 1-kg instantaneous emission at time t = 0
Global Warming Potential (GWP): integrated forcing over time horizon t = H
Atmospheric lifetime:
CO2 13 yrs 1.5 yrs
Global Temperature Potential (GTP): Mean surface temperature change at t = H
Surface T response
from 2008 emissions
taken as pulse
[IPCC, 2014]
Why the ephemeral response from a pulse of methane?
Fin
Fout
To + To
To
To
t<0
t=0
t = 20 years
climate
equilibrium
emission
pulse
climate
response
F = 0
F > 0
F < 0
To
t = 100 years
back to
original
equilibrium
F = 0
Simple calculation of Global Temperature Potential (GTP)
Use impulse response function of surface To to pulse F of 1 W m-2 at time t = 0:
0.63
0.43
 To (t )  (
exp[ t / 8.4] 
exp[ t / 410])
8.4
410
t in years
obtained by fitting results of HadCM3 climate model [Boucher and Reddy, 2008]
GTP is then given by
tH
To (tH )   F (t ) T (tH  t )dt
0
Implication of GTP-based policy for near-term climate forcers
Start controlling methane 40 years before target, BC 10 years before target
IPCC [2014]
Other climate policy metrics (M) have been proposed
I (Cref E )  I (Cref )
M 
W (t )dt
E
0

• C is the atmospheric variable perturbed by emission E
• I is the impact function of interest (T, sea level, precip, GNP, health…)
• W(t) is the temporal weighting factor
 W(t) = 1 for t < tH , = 0 for t > tH (as for GWP)
 W(t) = (t – tH) Dirac function (as for GTP)
 W(t) = exp[-t/tH] exponential discount rate
As societal relevance of the metrics increase,
so does uncertainty
Flugestvedt et al. [2003]
Controlling methane and BC should be part of climate policy
… but for reasons totally different than CO2
•
•
•
•
•
•
•
It addresses climate change on time scales of decades – which we care about
It offers decadal-scale results for accountability of climate policy
It is less sensitive to arguments over what discount rates should be used
It is an alternative to geoengineering by aerosols
It has important air quality co-benefits
BC has additional regional, hydrological impacts
Measures to reduce emissions can have lasting effects over long time horizons
Trend in Arctic sea ice volume
Geoengineering: cloud seeeding
Black carbon in the atmosphere
diesel engines
residential fuel
open fires
freshly emitted
BC particle
Global BC emission [Wang et al., 2014]
Loss of BC is by
wet deposition
(lifetime ~ 1 week)
BC exported to upper troposphere is major component of forcing
…because it’s above white clouds instead of dark surface
Integral contribution
To BC forcing
•
• Export to upper
•
deep
troposphere
convection
•
Global mean
BC profile
(chemical
transport model)
•
•
50% from
BC > 5 km
scavenging
•
• •• •
• • • • ••• •
• ••• • • • •
• • •• •
frontal
lifting
BC source
region
(combustion)
BC forcing
efficiency
Ocean
Samset and Myhre [2011]
Multimodel intercomparisons and comparisons to observations
AeroCom
chemical transport models (CTMs) used by IPCC
overestimate BC by order of magnitude in upper troposphere
Pressure, hPa
TC4 aircraft campaign (Costa Rica)
Observed
Models
Such large overestimate must be due
to model errors in scavenging
BC, ng kg-1
Pressure, hPa
HIPPO aircraft campaign over Pacific
obs
models
60-80N
BC, ng kg-1
obs
models
20S-20N
BC, ng kg-1
Koch et al. [2009], Schwarz et al. [2010]
Previous
application
to Arctic spring
(ARCTAS)
BC/aerosol
scavenging
in GEOS-Chem
CTM used
at Harvard
Cloud updraft
scavenging
Anvil
precipitation
Large scale precipitation
IN+CCN
CCN+IN,
impaction
entrainment
detrainment
CCN
• Meteorological data including convective mass fluxes
from NASA GEOS assimilation system
• Aerosols are scavenged in cloud by similarity with
condensed water
• Additional scavenging below cloud by rain/snow
• In-cloud scavenging efficiency from freezing/frozen
clouds is highly uncertain
• Additional uncertainty for BC is its efficiency as
cloud condensation nucleus (CCN) and ice nucleus (IN)
BC lifetime in GEOS-Chem is 4 days (vs. 7±2 days in AeroCom models)
GEOS-Chem BC simulation: source regions and outflow
Tests sources, export
Observations (circles) and model (background)
Wang et al., 2014
Normalized mean bias (NMB) in range of -30% to +10%
NMB= -27%
surface
networks
NMB= 6.6%
AERONET BC optical depth NMB= -32%
Aircraft profiles in continental/outflow regions
Asian outflow
HIPPO
US
observed
(A-FORCE)
(HIPPO)
model
(US)
Arctic
(ARCTAS)
NMB= -12%
Comparison to HIPPO BC observations across the Pacific
Model
PDF
PDF, (mg m-3 STP)-1
Observed
• Model doesn’t capture
low tail, is too high at N
mid-latitudes
• Mean column bias is
+48%
• Still much better than
the AeroCom models
Wang et al., 2014
BC top-of-atmosphere direct radiative forcing (DRF)
Absorbing aerosol optical depth (AAOD)
DRF = Emissions X Lifetime X
Mass absorption
Forcing
X
coefficient
efficiency
Global atmospheric load
Emission Global load
Tg C a-1
(mg m-2)
This work 6.5
AeroCom
[2006]
7.8 ±0.4
Bond et al. 17
[2013]
[% above 5 km]
BC
AAOD
x100
Forcing
efficiency
(W m-2/AAOD)
Direct radiative
forcing (W m-2)
fuel+fires
0.15 [8.7%]
0.17
88
0.19 (0.17-0.31)
0.28 ± 0.08
[21±11%]
0.22±0.10 168 ± 53
0.34 ± 0.07
0.55
0.60
0.88
147
• Our best estimate of 0.19 W m-2 is much lower than IPCC recommendation of
0.65 (0.25-1.1) W m-2
• IPCC value is from models that greatly overestimate BC in upper troposphere
BC is much less important for climate forcing than stated in IPCC
Wang et al., 2014
Atmospheric methane: long-term trends are not understood
the last 30 years
the last 1000 years
E. Dlugokencky, NOAA
Source attribution is difficult due to diversity, complexity of sources
Global sources,
Tg a-1
Wetlands
180
Other natural
40
Fires
50
Livestock
90
Landfills
70
Gas
60
Rice
40
Coal
40
Individual sources uncertain
by at least factor of 2; emission
factors are highly variable,
poorly constrained
Satellite data as constraints on methane emissions
“Bottom-up” emissions (EDGAR):
best understanding of processes
Satellite data for methane columns
2009-2011
537 Tg a-1
Optimal estimate inversion
using GEOS-Chem model adjoint
Ratio of optimal estimate
to bottom-up emissions
Turner et al.,
submitted
Basics of inverse modeling
Optimize state vector
x (emissions) using obs vector y (atm. concentrations)
Observations
y + εI
atmospheric concentrations
from satellite, aircraft
Prior estimate
Minimize cost function
xA + εA
J ( x ) ~ y - F(x)
Bottom-up
inventory
with error weighting,
xA regularization
Forward model
yM = F(x) + εM
GEOS-Chem
chemical transport model
Posterior estimate
2
Analytical
or numerical
(variational)
method
xˆ + εˆ
Using satellite data for high-resolution inversion
of methane emissions in North America
EDGAR emission
Inventory for methane
Bottom-up methane emissions for N. America (2009-2011)
total: 63 Tg a-1
livestock: 14
waste: 10
wetlands: 20
CONUS anthropogenic
emissions:
25 Tg a-1 (EDGAR)
27 Tg a-1 (EPA)
8 oil/gas
9 livestock
6 waste
3 coal
oil/gas: 11
coal: 4
Turner et al., submitted
Global inversion of GOSAT data
feeds boundary conditions for North American inversion
GOSAT observations, 2009-2011
Dynamic
boundary
conditions
Adjoint-based inversion
at 4ox5o resolution
Analytical inversion
with 369 Gaussians
correction factors to EDGAR v4.2 + LPJ prior
Turner et al., submitted
Correction factors to bottom-up inventory
• CONUS anthropogenic emission of 40-43 Tg a-1 vs. EPA value of 27 Tg a-1
• Livestock source is underestimated by EPA; What about oil/gas?
Turner et al., submitted
Methane emissions in CONUS:
comparison to previous studies, attribution to source types
Ranges from
prior error
assumptions
2004
satellite
2007
2009-2011
surface, satellite
aircraft
• EPA national inventory underestimates anthropogenic emissions by 30%
• Livestock is a contributor: oil/gas production probably also
Turner et al., submitted
Future of satellite observations for methane monitoring
Methane is readily observable over land by solar backscatter at 1.6/2.3 µm

Scattering by
Earth surface
Backscattered
intensity IB
absorption
Methane column
 ln[ I B (l2 ) / I B (l1 )]

1  1 / cos 
l1 l2
wavelength
• GOSAT (2009-): high-quality 5x5 km2 pixels but sparse
• TROPOMI (2016 launch): global daily coverage with 7x7 km2 pixels
• Geostationary (proposed): hourly coverage over N America with 2x2 km2 pixels