Transcript 2 K -1 - The Heartland Institute`s International Conferences on

Satellite & Model Evidence
for
Global Warming Being Driven
by the
Pacific Decadal Oscillation
Dr. Roy W. Spencer
Principal Research Scientist
The University of Alabama in Huntsville
10 March, 2009
Overview
• Satellite evidence of negative feedback has
been obscured by radiative forcing due to
natural cloud variations
• Negative feedback means that incr. CO2 is too
weak to cause observed warming
• 7.5 Years of Terra satellite data, combined with a
simple climate model, shows that the PDO can
explain most of the temperature variability seen
during the 20th Century
There are TWO RADIATIVE WAYS to Cause Global Warming
1) INCREASE SOLAR RADIATION being absorbed, or
2) DECREASE INFRARED RADIATION (IR) being lost to space.
Observe the Behavior of TODAY’S Climate System
with
NASA’s
Aqua & Terra
Satellites
Traditional Way of Estimating Feedbacks
From Satellite Data
Simple Forcing-Feedback Model of Temperature Variability
dDT/dt = [Forcing – Feedback]/ Cp
Incr. CO2, aerosols,
cloud changes, etc.
(W m-2)
lDT
l = lsw + lLW (W m-2 K-1)
~ Ocean depth
l= 5 W m-2 K-1; 50 m mixed layer
Simple Model
Forced with
Random radiative forcing
True FB
(e.g. cloud variations)
…leads to 100% error
in diagnosed feedback!
(Spencer & Braswell, Nov. 1, 2008 J. Climate)
Diag.
FB
Even the IPCC Models Show Evidence of
“radiative forcing spirals”
GFDL
Simple Model
Forced with
Random non-radiative forcing
(e.g. evap/precip. fluctuations)
0% error in diagnosed feedback
Feedbacks can ONLY be measured
in response to non-radiative forcing,
Not radiative forcing
REAL WORLD:
Both
Radiative & Non-Radiative
forcing
…leads to an underestimate
of diagnosed feedback
depending upon ratio
of radiative to non-radiative forcing
This explains why previous
satellite diagnoses of
Feedback have been so variable…
AND biased toward positive feedback.
True FB
Diag.
FB
Only a FEW IPCC models show “feedback stripes”
Slopes of “feedback stripes”
Corresponds to long-term
Feedbacks diagnosed by
Forster & Taylor (2006)
So, Feedback Is ONLY Indicated
by Linear Striations in the Data
(feedback stripes)
Terra Satellite: CERES vs. AMSU ch. 5 Suggests
STRONGLY Negative Feedback Stripes (8 W m-2 K-1)
Aqua Satellite: CERES vs. AMSU ch. 5 Suggests
STRONG Negative Feedback
…even without
accounting for contamination by radiative forcing
5 W m-2 K-1
Strongly negative feedback
(insensitive climate system) means
radiative forcing from extra CO2
too weak to cause global warming.
IF Manmade CO2 hasn’t Caused Global Warming
over the last 100 years..then what has?
HOW ABOUT MOTHER NATURE..
c. Pacific Decadal Oscillation, 1902-2006
1.2
PDO Index
Natural
Cloud
Variations
2
1.6
0.8
Pacific
Decadal
Oscillation
0.4
0
-0.4
-0.8
-1.2
-1.6
-2
?
1902 1912 1922 1932 1942 1952 1962 1972 1982 1992 2002
0.6
YEAR
a. Global Mean Temperature, 1902-2006
Natural
Temperature
Variations
T Anomaly (deg. C)
0.4
0.2
LAND +
OCEAN
0
-0.2
OCEAN
ONLY
-0.4
-0.6
1902 1912 1922 1932 1942 1952 1962 1972 1982 1992 2002
Global
Surface
Temperature
A Simple Model Forced by Cloud Changes
Assumed Associated with the
Pacific Decadal Oscillation:
PDO Can explain 3/4 of 20th Century Warming
0.6
Simple Model of PDO Cloud Changes
Causing Global Warming
Thermometer
data
T Anomaly (deg. C)
0.4
0.2
PDO+CO2
0
PDO-only
-0.2
-0.4
-0.6
-0.8
1900
1910
1920
1930
1940
1950
YEAR
1960
1970
1980
1990
2000
…and Recent Satellite Data Suggests that
there Indeed IS a Connection between
the PDO and Cloud Variations….PDO INDEX
2000
2.5
2001
2002
(monthly)
2003
2004
2005
2.0
1.5
PDO Index
PDO Index
2000-2005
1.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0
Terra Satellite
Observed Cloud
Changes
2000-2005
TOA Flux Anom. (W m-2 K-1)
-1.0
CERES LW+SW YEAR
Flux Anomalies
60N-60S Oceans
-0.5
0.0
0.5
1.0
1.5
2000
2001
2002
2003
2004
2005
…and plotted here as Yearly Averages (updated thru 2007).
1.5
Radiative Forcing (W/m2)
1.0
(Rad. Forcing=
Total Flux
minus
Feedback)
LW+SW
2005
0.5
2003
Model
predicted
0.0
2007
2001
y = 1.39x - 0.23
R2 = 0.67
-0.5
-1.0
SatelliteObserved
-1.5
?
-2.0 2008
?
2000
-2.5
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
PDO Index
Note that PDO forcing is actually measured….
..forcing from incr. CO2 must be computed theoretically
The End
Natural climate variability:
Partly due to Natural Cloud Variations?
0.8
Medieval Warm Period
Temperature Anomaly (deg. C)
0.6
0.4
0.2
0
Vikings
arrive in
Greenland
-0.2
End of Viking
colonization of
Greenland
-0.4
-0.6
Little Ice Age
-0.8
0
200
400
600
800
1000
YEAR (AD)
1200
1400
1600
1800
2000
Conclusions:
Satellite data & simple model suggest…
1. Feedbacks are strongly negative (low climate
sensitivity)
 Previous feedback estimates were contaminated by
radiative forcing due to cloud fluctuations
 CO2 increase too weak to cause warming
2. The PDO causes natural radiative forcing of climate
change
 Enough to explain 75% of global warming
 IPCC assumes cloud cover always remains the same
 Atlantic Multidecadal Oscillation, Arctic Oscillation, El Nino,
La Nina, etc. all need to be investigated as possible climate
change mechanisms
1st SATELLITE TEST OF THE CLIMATE MODELS:
How they gain and lose INFRARED energy
GOOD
AGREEMENT
Between
Models &
satellites
for EMITTED
INFRARED
Number of 5-Year Periods
1400
1200
Positive Feedback
(clouds AMPLIFY
warming)
1000
800
Negative
Feedback
(clouds REDUCE
warming)
18 IPCC
Models
600
400
200
Satellite
0
-4
-3
-2
-1
0
1
2
3
4
5
6
Feedback Parameter Estimate (W m-2 K-1)
7
2nd SATELLITE TEST OF THE CLIMATE MODELS:
How they gain and lose SOLAR energy
NO
AGREEMENT
Between
Models &
satellites
for REFLECTED
SUNLIGHT
Number of 5-Year Periods
1400
1200
Positive Feedback
(clouds AMPLIFY
warming)
Negative Feedback
(clouds REDUCE
warming)
1000
800
Satellite
600
18 IPCC
Models
400
200
0
-4
-3
-2
-1
0
1
2
3
4
5
6
Feedback Parameter Estimate (W m-2 K-1)
7
TOTAL SATELLITE TEST OF THE CLIMATE MODELS:
How they gain and lose SOLAR+INFRARED energy.
NO
AGREEMENT
Between
Models &
Satellites
for REFLECTED
SUNLIGHT+
EMITTED
INFRARED
Number of 5-Year Periods
1400
1200
Positive Feedback
(clouds AMPLIFY
warming)
1000
800
Negative
Feedback
(clouds REDUCE
warming)
18 IPCC
Models
600
Satellite
400
200
0
-4
-3
-2
-1
0
1
2
3
4
5
6
Feedback Parameter Estimate (W m-2 K-1)
7