Transcript 03_MATCH_paper2_May17 - Modelling and assessment of

Uncertainties in absolute attribution of
climate change
SB-24
17 May 2006
Joyce Penner
University of Michigan
Overview of paper #2
•
Paper #1 examined the uncertainties associated with
methodological choices in attributing relative
temperature change
•
Here we assess scientific uncertainties in attributing
absolute climate change
•
We use a closure method to evaluate uncertainties:
•
•
•
Emissions to concentrations for long
lived greenhouse gases
Radiative forcing to temperature
change for attribution
Attribution of OECD Annex I countries are used as an
example because these (and their uncertainties) are
available from UNFCCC reporting
Modelling and assessment of contributions to climate change
Overview of paper #2
Contributors:
Prather, Lowe, Raper, Stott, Höhne, Fuglestvedt,
Romstad, Penner, Andronova, Kurosawa, Wagner,
Jain, Pires de Campos, Meinshausen, van Aardenne
Modelling and assessment of contributions to climate change
Method
Emissions
Global inventories of
GHG emissions
based on activities
Emissions derived
from atmospheric
measurements
Concentrations
Radiative forcing
Global average
temperature
change
All sources of
historical radiative
forcing
Observed
temperature increase
Modelling and assessment of contributions to climate change
Total
uncertainty of
OECD Annex
I countries
contribution
Example: N2O
Emissions from inverse
model are well within
the stated uncertainties
of the EDGAR data base
Emissions of OECD
Annex I countries from
EDGAR are within stated
uncertainties from
UNFCCC inventories
Modelling and assessment of contributions to climate change
We estimate a
pdf
for OECD Annex
I N2O emissions
using UNFCCC
uncertainties for
the next step
(RF to T)
Modelling and assessment of contributions to climate change
Example: CH4
Global emissions from
Edgar bottom-up
inventory match well
the emissions required
to fit observations of
CH4.
But the Edgar OECD
Annex I emissions are
significantly higher than
the UNFCCC emissions.
The uncertainties for
UNFCCC emissions must
be increased in RF to T
calculations
Modelling and assessment of contributions to climate change
Uncertainties
in OECD Annex I
countries are
widened for
the next step
(RF to T) to
account for
mis-match between
EDGAR and
UNFCCC reported
emissions
Modelling and assessment of contributions to climate change
CO2
The increase in CO2 concentration can
be explained by the following factors:
• Anthropogenic emissions from
fossil fuels and industrial processes
• Anthropogenic emissions/removals
from land use change and forestry
• Natural removals by the biosphere
• Natural removals by the ocean
Measured
Well known
Unknown
Modelled
Modelled
Modelling and assessment of contributions to climate change
Global LUCF emissions are highly
uncertain due to land use change data
2.5
Land Use (PgC/yr)
2.0
1.5
1.0
HH-Low
HH-Base
HH-High
HYDE-Low
HYDE-Base
HYDE-High
RF-Low
RF-Base
RF-High
0.5
0.0
1900
1920
1940
1960
Year
1980
2000
Even so, OECD Annex I LUCF emissions from
inverse method since 1990 are well known:
0.8
HH-Low
HH-Base
HH-High
HYDE-Low
HYDE-Base
HYDE-High
RF-Low
RF-Base
RF-High
0.7
Land Use (PgC/yr)
0.6
0.5
0.4
0.3
0.2
0.1
0.0
-0.1
-0.2
1900
1920
1940
1960
Year
1980
2000
But UNFCCC LUCF emissions from OECD Annex I countries
are outside the uncertainty range from inverse method: Need
to increase range of uncertainty considered in RF to T
calculation! (not yet included)
0.2
Land Use Emissions (PgC/yr)
0.1
0
1990
1992
1994
1996
-0.1
-0.2
1998
2000
2002
HH-Low
HH-Base
HH-High
HYDE-Low
HYDE-Base
HYDE-High
RF-Low
RF-Base
RF-High
UNFCCC OECD Annex 1
UNFCCC
LUCF
emissions
-0.3
-0.4
Year
Method
Emissions
Global inventories of
GHG emissions
based on activities
Emissions derived
from atmospheric
measurements
Concentrations
Radiative forcing
Global average
temperature
change
All sources of
historical radiative
forcing
Observed
temperature increase
Modelling and assessment of contributions to climate change
Total
uncertainty of
OECD Annex
I countries
contribution
Radiative Forcing and uncertainty was estimated
for all of the important climate factors *
3
3
Solar
Forcing (W/m2)
2
2
Tropospheric ozone
CFCs, HCFCs and
other ODS
SF6
1
1
PFCs
HFCs
0
0
N2O
CH4
-1
-1
CO2
Vulcanic
-2
-2
Areosol forcing
indirect
Carbon aerosol
Sulfate aerosol
-3
1750
1760
1770
1780
1790
1800
1810
1820
1830
1840
1850
1860
1870
1880
1890
1900
1910
1920
1930
1940
1950
1960
1970
1980
1990
2000
-3
Stratospheric ozone
Total
* Refers to preliminary assessment
Modelling and assessment of contributions to climate change
Comparison of D and A (inverse model) aerosol
forcing with bottom-up aerosol forcing
0
1700
1750
1800
1850
1900
1950
2000
2050
-0.5
-1
5Stott
Forcing
-1.5
50Stott
95Stott
-2
5Penner
50Penner
-2.5
95Penner
-3
Bottom up method
yields wider uncertainty
range, but encompasses
inverse method
-3.5
-4
Year
Modelling and assessment of contributions to climate change
Uncertainty in natural forcing
deduced using different reconstructions
Volcanic
Forcing (w/m2)
sato
0.5
0
-0.51850
-1
-1.5
-2
-2.5
-3
-3.5
Ammann03
Solar
Ammann06
HOYT&SCHATTEN
LEAN
LEAN_II
0.4
1870
1890
1910
1930
1950
1970
1990
0.2
0
1850
1870
1890
1910
1930
1950
1970
-0.2
-0.4
year
year
Modelling and assessment of contributions to climate change
1990
Additional contributions from land use
albedo change and dust – based on TAR
estimates
0.6
0.4
forcing
0.2
0
1700
-0.2
1750
1800
1850
1900
1950
2000
2050
case 1
case 2
-0.4
case 3
-0.6
-0.8
-1
-1.2
year
Modelling and assessment of contributions to climate change
What will alter median and
spread of bottom up forcing?
Uncertainty range
Median magnitude
3
3
2.5
2.5
2
2
1.5
1.5
1
1
0.5
0.5
0
0
GHG
Trop O3
Bottom-up
AP
D&A AP
Other
Uncertainty range in bottom up
forcing
GHG
Trop O3
Bottom-up
AP
D&A AP
Other
Median magnitude of bottom
up forcing
Forcing calculated to year 2000
Modelling and assessment of contributions to climate change
Use uncertainties in individual components
to define uncertainty in total forcing
Modelling and assessment of contributions to climate change
Results space of modeled temperatures
compared to observed warming since 1880
Observations
are shown in
black
Modelling and assessment of contributions to climate change
Compare forcing from bottom up and forcing from
inverse: not all forcing scenarios are consistent with the
observed temperature change*
Forcing from bottom up estimates*
Forcing from inverse calculation
*Preliminary values
Modelling and assessment of contributions to climate change
Method
Emissions
Global inventories of
GHG emissions
based on activities
Emissions derived
from atmospheric
measurements
Concentrations
Radiative forcing
Global average
temperature
change
All sources of
historical radiative
forcing
Observed
temperature increase
Modelling and assessment of contributions to climate change
Total
uncertainty of
OECD Annex
I countries
contribution
Effect of group’s emissions
CH4
Undisturbed forcing
2.5th percentile
5th percentile
25th percentile
50th percentile
75th percentile
95th percentile
97.5th percentile
W/m2
1.7
1.65
1.6
1.55
1.5
1.45
1.4
1.35
1.3
1.25
1.2
1990
0.5
0.48
0.46
0.44
0.42
0.4
0.38
0.36
0.34
0.32
0.3
1990
Undisturbed forcing
2.5th percentile
5th percentile
25th percentile
50th percentile
75th percentile
95th percentile
97.5th percentile
1995
2000
N2O
0.17
0.16
0.15
0.14
W/m2
W/m2
CO2
1992 1994
1996 1998
2000 2002
0.13
0.12
Undist urbed f orcing
2.5t h percent ile
0.11
5t h percent ile
25t h percent ile
0.1
50t h percent ile
75t h percent ile
0.09
95t h percent ile
97.5t h percent ile
0.08
1990
1992
1994
1996
1998
Modelling and assessment of contributions to climate change
2000
2002
Combined effect of uncertainties on warming
from OECD Annex 1 countries due to CO2
Combined effect of uncertainty in global mean forcing, climate sensitivity, ocean
diffusivity and OECD Annex 1 forcing uncertainty on warming from OECD Annex 1
countries due to CO2
A likelihood was estimated for the unperturbed case using agreement with
observed warming. The prior probability for the OECD Annex 1 perturbations was
also included.
The fraction of warming attributable to OECD Annex I countries is 0.23 with a
95% confidence interval of 0.08 to 0.38.
Combined effect of uncertainty on warming from OECD
Annex 1 countries due to CO2, CH4 and N2O*
Combined effect of uncertainty in global mean forcing, climate sensitivity, ocean diffusivity and
OECD Annex 1 forcing uncertainty on warming from OECD Annex 1 countries due to CO2, CH4,
and N2O
A likelihood was estimated for the unperturbed case using agreement with observed warming.
The prior probability for the annex 1 perturbations was also included.
The fraction of warming attributable to OECD Annex I countries is 0.34 with a 95%
confidence interval of 0.23 to 0.53. (*preliminary analysis)
Conclusions
• We examined uncertainties in emissions inventories
for both global mean values and OECD Annex I GHG
emissions
• We examined the consistency between the emissions
and observed concentrations
• We estimated forcing and forcing uncertainty from all
other known climate factors
• We examined the implications of this uncertainty for
predicted global average temperature change and
the change associated with 1990 - 2002 OECD
Annex I emissions
GWP weighted emissions
Radiative forcing
Temperature increase
Contribution
Low
14160 MtCO2eq.
0.32 W/m2
0.10 °C
High
-6%
-28%
-49%
16%
53%
139%
Modelling and assessment of contributions to climate change
Backup slides
Closure for long-lived
greenhouse gases
• Compare bottom-up inventories to those
determined from inverse models to determine
uncertainty in global emissions
• Define OECD Annex I emissions using
UNFCCC reported emissions and reported
uncertainties
• Compare OECD Annex I emissions from
inverse model and adjust uncertainty in
UNFCCC emissions if needed
Modelling and assessment of contributions to climate change
Comparison of OECD Annex I
emissions with global emissions
Modelling and assessment of contributions to climate change
OECD Annex 1 warming due to CO2, and
effect of uncertainty in climate sensitivity
Modelling and assessment of contributions to climate change
OECD Annex 1 warming due to CO2, and
effect of uncertainty in ocean diffusivity
Modelling and assessment of contributions to climate change
Uncertainties in global mean forcing
Modelling and assessment of contributions to climate change
OECD Annex 1 warming due to CO2, and
effect of uncertainty in global mean forcing
Modelling and assessment of contributions to climate change
Uncertainty in OECD Annex 1
forcing from N2O
N2O
0.17
0.16
0.15
W/m2
0.14
0.13
0.12
Undist urbed f orcing
2.5t h percent ile
0.11
5t h percent ile
25t h percent ile
0.1
50t h percent ile
75t h percent ile
0.09
95t h percent ile
97.5t h percent ile
0.08
1990
1992
1994
1996
1998
2000
2002
Modelling and assessment of contributions to climate change
Combined effect of uncertainty on warming from
OECD Annex 1 countries due to N2O
Combined effect of uncertainty in global mean forcing, climate sensitivity, ocean
diffusivity and OECD Annex 1 forcing uncertainty on warming from OECD Annex 1
countries due to N2O
A likelihood was estimated for the unperturbed case using agreement with
observed warming. The prior probability for the OECD Annex 1 perturbations was
also included.
The fraction of warming attributable to OECD Annex I countries is 0.015 with
a 95% confidence interval of 0.0075 to 0.045.
Uncertainty in OECD Annex 1
forcing from CH4
W/m2
CH4
0.5
0.48
0.46
0.44
0.42
0.4
0.38
0.36
0.34
0.32
0.3
1990
Undisturbed forcing
2.5th percentile
5th percentile
25th percentile
50th percentile
75th percentile
95th percentile
97.5th percentile
1995
2000
Modelling and assessment of contributions to climate change
Combined effect of uncertainty on warming from
OECD Annex 1 countries due to CH4
Combined effect of uncertainty in global mean forcing, climate sensitivity, ocean
diffusivity and OECD Annex 1 forcing uncertainty on warming from annex 1
countries due to CH4
A likelihood was estimated for the unperturbed case using agreement with
observed warming. The prior probability for the OECD Annex 1 perturbations was
also included.
The fraction of warming attributable to OECD Annex I countries is 0.085 with
a 95% confidence interval of 0.06 to 0.12.
Comparison of aerosol forcing to year 2000 from
bottom-up with D and A reconstruction
D and A reconstruction
100
100
90
90
80
80
70
70
Cumulative Prob
cumulative prob
Bottom-up reconstruction – used in
subsequent analysis
60
50
40
30
60
50
40
30
20
20
10
10
0
-3.5
-3
-2.5
-2
-1.5
Forcing
-1
-0.5
0
0
-3
-2.5
-2
-1.5
-1
-0.5
Forcing
Modelling and assessment of contributions to climate change
0
Inverse model used to estimate forcing for
2 different values of climate sensitivities
Time filtered forcing values
smooth1.7
Annual values
smooth4.2
2.5
1.5
Forcing
Forcing
2
1
0.5
0
-0.51800
1850
1900
1950
2000
2050
4
3
2
1
0
-11800
-2
1850
ann1.7
ann4.2
1900
1950
2000
Year
Year
Inverse calculation showing plume of forcing curves for different
climate sensitivity based on TAR GCM models.
Modelling and assessment of contributions to climate change
2050
Would need to add other forcings to make all scenarios
from bottom up estimates consistent with observed T
pdf from inverse
pdf from bottom up
Minus
pdf of extra forcing
that needs to be
added to bottom up
to achieve
consistency with
temperature record
Sample all combinations
Modelling and assessment of contributions to climate change