Global Warming - Atmospheric Sciences

Download Report

Transcript Global Warming - Atmospheric Sciences

Global Warming: Potential Effects
on National Parks in the Pacific
Northwest
Cliff Mass,
Department of Atmospheric Sciences
University of Washington
Global Warming and the Northwest
National Parks
Some potential impacts that have been
mentioned:
– More windstorms, with increased damage
to forests and coastal zones
– Heavier rain, with more flooding events.
– Reduced snowpack, and earlier spring
melt.
– Reduction in number and extent of
glaciers.
Questions
• Have we seen such impacts during the past
few decades?
• What should we expect during the
remainder of this century?
The Bottom Line
• Global warming is an extraordinarily serious
global issue, but some areas will have lesser or
greater effects.
• The Northwest may well be a location where
GW effects are delayed and weaker than the
global average.
• At this point in time, we are not able to
demonstrate that global warming has produced
measurable impacts on Northwest weather and
climate.
The Bottom Line
• The impact of global warming over the
Northwest during the next few decades is not
clear.
• Impacts, specifically on snowpack and glacier
extend, should be large by the middle to end of
the century.
Floods
• The November 6, 2006 floods at Mount Rainier and
the December 3, 2007 flooding of Chehalis and
Centralia has sensitized us to this threat.
• The hand-waving argument is that warming
temperatures will result in more water vapor in the
atmosphere, heavier rain, and more floods.
• Reality is much more complicated.
What does the data show?
Coastal Rainfall: The trend in the number of events greater than two
inches over two days from 1950-2000
Increasing Trend
Decreasing Trend
California
Oregon
Washngton
Canada
Trend in the top
60 storms: 24h
rainfall
High Resolution Global Warming
Simulations: Looking to the Future
Its Not So Simple
• Flooding is associated patterns called
“Atmospheric Rivers”….a.k.a. the Pineapple
Express…which are closely associated with the
the jet stream.
Another possibility
• Most most models suggest the jet stream will
weaken and move poleward under global
warming.
• Will the “hose”---the atmospheric rivers move
with it? Is that is what is happening in
Oregon?
• A lot of uncertainty exists.
What about other weather elements
influencing the parks?
• Major Northwest windstorms…like the Columbus Day
Storm or the Chanukah Eve Storm…are closely
associated with the jet stream.
• Unlike hurricanes their energy source is horizontal
temperature contrasts, not warm water.
• Climate models suggest that the jet stream…and
associated temperature contrasts, will weaken and more
northward.
• Implication: weaker and farther north storms in the
midlatitudes
• Interestingly, the number of major windstorms in Oregon
has apparently decreased.
Declining snowpack due to global warming?
• Major issue in the popular threat and among some
politicians—some claiming we have already lost 50%!
From Mote et al
2006
Washington-wide snowpack since 1950,
Relative to 1971-2000 normal
Snowpack in usual maximum month
Washington-wide snowpack period of record,
Relative to 1971-2000 normal
Snowpack in usual maximum month
Cascade Snowpack has INCREASED during the past 30 yrs
1600
1976-2007 Trends in
1 Apr SWE for
Elevation Quartiles (EQs):
EQ4: +7.1%
EQ3: +9.4%
EQ2: +7.8%
EQ1: +6.4%
1400
EQ4
1 Apr SWE (mm)
1200
High
EQ3
1000
EQ2
800
EQ1
600
Low
400
200
0
1980
1985
1990
1995
Water Year
2000
2005
2010
Stoelinga et al 2008. Time series of elevation quartile (EQ)-averaged 1 April SWE (in mm) measured at
snow course during the period 1976-2007. Elevation ranges for the four quartiles are: (1) 792-1119 m; (2)
1119-1288 m; (3) 1288-1542 m; and (4) 1542-1981 m. Dashed lines show best-fit lines for each EQ. Also
listed are the 1976-2007 trends for each EQ (as percent of the 1961-1990 mean).
Pacific Decadal Oscillation (PDO)
PDO is thought to be a natural mode of atmospheric variability
Negative phase of PDO associated with greater snowpack in NW.
A Favored Area?
• The Northwest is downwind of the eastern Pacific and
thus our snowpack is controlled by the Pacific
temperatures.
• The eastern Pacific has not warmed up during the past 30
years.
• Global climate models suggest the eastern Pacific will
warm more slowly than most locations.
-1.4
-1.2
-1.0
-0.8
-0.6
-0.4
-0.2
0.0
+0.2 +0.4 +0.6 +0.8 +1.0 +1.2 +1.4
Air Temperature Trend (1979-2005)
Change in Surface Air Temperature (°C) from 1979-2008
Climate Model Output for 2100
Averaging a collection of the best climate models over the Pacific
Suggests the Same Thing for the Future
(a) SST
SST
(b) Tsfc
0.0
+0.2
+0.4
(c) T850
+0.6
+0.8
Air Temp
+1.0
+2.0
+3.0
°C
850 mb
Stoelinga, Albright and Mass 2008
Predicted linear trend of November-March mean temperature for 2000 to 2025
(°C), as predicted by the ensemble of climate models used in the IPCC AR4 report.
Shown are the ensemble means of (a) sea-surface temperature, (b) surface air
temperature, and (c) 850-hPa temperature.
Summary
• Global warming is certain, the question is its magnitude
and regional effects.
• The magnitudes of the changes will vary geographically,
and the Northwest may see weakened and delayed
effects, because of our location downstream of the
Pacific, and the nature of our storms.
• Global warming is a serious, but complicated issue, and
some of the ideas being thrown around by the popular
press and well-intentioned but misinformed people are
not necessarily correct.
• Little concrete evidence at this time of any major global
warming threats to the region or the parks during the next
several decades