Schaefer: Permafrost - University of Colorado Boulder

advertisement
When is the Permafrost
Carbon Tipping Point?
Me
Lin Liu
Alessio
Gusmereli
Tim
Schaefer
Kevin Schaefer
National Snow and Ice Data Center, University of Colorado
Tingjun
Zhang
Permafrost Primer
Permafrost: Ground at or
below 0°C for at least 2
consecutive years
Active Layer: A layer over
permafrost that freezes and
thaws annually
Permafrost Degradation: A
decrease in permafrost
extent; an increase in active
layer thickness.
Skiklomanov [2007]
Permafrost Classification
Permafrost Classification by Area
Continuous (>90% of area)
Discontinuous (50-90% of area)
Sporadic (10-50% of area)
Isolated (<10% of area)
Brown et al., 1998; Zhang et al., 1999
Permafrost Distribution by Country
Permafrost Profile
Vegetation
Active Layer
Permafrost
Permafrost Profile
Exposed permafrost by river, Siberia
[Davis, 2000]
Thermokarst, Alaska
Cryoturbation
• Movement of soil or rock due to
repeated freezing and thawing
Vegetation
Active
Layer
Permafrost
Pleistocene Cryoturbation, France
Permafrost Features
Frost Heave, Yamal
Stone Circles, Svalbard
Stone Circles, NW Territories
Stripes, Glacier NP
Ice Wedges and Polygons
1st Winter
Active Layer
Frozen
1st Spring
Frozen
Thawed
Permafrost
Soil contracts
& cracks
Polygons, Yena
100th Winter
Crack fills with
water & freezes
100th Spring
Thawed
Ice
Wedge
Polygons, Prudhoe Bay [Zhang, 2009]
Ice Lenses and Layers
Year 1,000
Year 1
Active Layer
Permafrost
Capillary
suction of
water to
permafrost
Ice Lens
Water freezes
& expands
Ice Lenses
Active Layer
Permafrost
Ice Layer
Permafrost is Like Concrete
Wickland
Schaefer
Thermokarst
• Thermokarst: subsidence or collapse of ground
surface due to melting of ground ice
Slope Mountain, Alaska [Schaefer, 2012]
Impacts of Degradation
Foundation Settling
in Chersky
Alaska Road Heaves
Qinghai-Xizang Highway Bridge
Thermokarst in Yakutsk [Skiklomanov, 2005]
Impacts of Degradation
Coastal Erosion, Alaska
Drying lake, Tibet [Zhang, 2007]
Rockfall, Matterhorn [Gruber, 2003]
Ice-wedge thaw, Alaska [Davis, 2000]
Global Carbon Cycle
Atmosphere
750 Gt + 3 Gt yr-1
120
119
1.9
1.7
90
88
6
Vegetation 600 Gt
Soils 1400 Gt
Permafrost 1466 Gt
Ocean
38,000 Gt
Fossil Fuel
4000 Gt
Permafrost Carbon Burial
~1466 Gt C in permafrost [Tarnocai et al., 2009]
Deposition (loess, peat,
erosion, volcanic)
Soil
Depth
Active Layer
Permafrost
Horizon
Permafrost
Permafrost Carbon
Mammoth, Siberia
32,000 year old grass, Alaska
30,000 year old roots, Siberia
[Zimov et al., 2006]
15,000 year old moss, North Slope
[Schaefer , 2012]
Permafrost Carbon Feedback
Amplification of
warming due to release
of CO2 and CH4 from
thawing permafrost
Methane Release from Thawing Permafrost
Methane
emission
Thermokarst
Erosion
Peat
Thaw bulb
Permafrost
Methane
production
Dead
plant &
animal
remains
Burning methane over a
thermokarst lake in
Siberia (K. Walter)
K. Walter ftkmw1@uaf.edu
IPCC A1B Scenario
Atmospheric CO2 (ppm)
800
700
600
500
400
300
200
1960
2000
2040
2080
Date (year)
2120
2160
2200
Current Permafrost
Active Layer Thickness ALT (cm)
Projected Permafrost Degradation
HadCM3 (med)
Active Layer Thickness ALT (cm)
Projected Permafrost Loss
Increase in ALT by 2200 (cm)
CCSM3 (low)
29% loss
HadCM3 (med)
50% loss
MIROC3.2 (high)
59% loss
Cumulative NEE (Gt C)
Permafrost Carbon Tipping Point
PCF Tipping
Point 2023±4
Date (year)
Arctic switches from a sink to a source
Permafrost Carbon Flux (Gt C)
Cumulative Permafrost Carbon Flux
104±37 Gt
190±64 Gt
Date (year)
65±23% of cumulative
landppm
sink (~160 Gt C)
Equivalentglobal
to 87±29
Vostok Ice Core Records
80 ppm
• CO2 lags behind temperature by 600±400 yr
Paleo-Permafrost Carbon Feedback
Palaeocene-Eocene Thermal Maximum (PETM)
Orbit perturbations trigger Antarctic permafrost thaw
[DeConto et al. 2011, in review]
PCF and Fossil Fuel Emissions
• Both inject old carbon into atmosphere
• Both irreversible
• A1B scenario: 700 ppm by 2100
•
•
•
1345 Gt C total emissions
190 Gt C permafrost carbon flux
1157 Gt C fossil fuel emissions
• Must reduce fossil fuel emissions by additional
15% or overshoot target climate
Conclusions
•
•
•
•
PCF can explain past climate variability
PCF tipping point in mid 2020s
PCF is strong: 190±64 Gt C by 2200
Emission reductions must account for PCF
• Tellus B paper: Schaefer et al. [2011]
Backup Slides
The SiBCASA Model
Input
Weather
Boundary Layer
Canopy
NEE
Latent
Heat
Sensible
Heat
GPP
CO2
Temp
Humidity
R
Temperature
Soil
Moisture
Carbon
Snow
Permafrost Carbon in SiBCASA
Active Layer
Active Layer
Thickness (ALT)
Dmin = max ALT
during spinup
Permafrost
Carbon Pool
Dmax = 3 m
Permafrost
Active Layer
Thawed Carbon
Permafrost
Carbon Pool
Soil
Carbon
Pools
Permafrost
• 313 Gt C in permafrost carbon pool
• 91 Gt C in active layer
• 414 Gt C in top 3 m (575 Gt C estimated*)
*Tarnocai et al. [2009]
Experiment Setup
•
•
•
•
SiBCASA + ERA40 + A1B scenario
Continuous/discontinuous permafrost
1973-2001: “spin up“
2002-2200: random ERA40 + linear trend
•
•
•
MIROC3.2 (high)
HadCM3 (med)
CCSM3 (low)
Estimating Uncertainty
• 18 ensemble members
•
•
•
3 warming rates
3 permafrost carbon densities
2 sub-grid permafrost extents
• Best estimate: ensemble mean
• Uncertainty: ensemble standard deviation
IPCC A1B Arctic Temperatures
Air Temperature (°C)
CCSM3 (low warming)
HadCM3 (medium warming)
MIRC3.2 (high warming)
Average air temperature for permafrost regions
Permafrost Area (%)
Permafrost Area Loss
169±54 Gt C
203±63 Gt C
Date (year)
213±65 Gt C
Frozen Ground Extent
Permafrost
Seasonally Frozen Ground
Intermittently Frozen Ground
Snow Limit
Zhang et al., 2003. EICOP
Permafrost covers 24% of land
surface in Northern Hemisphere
What Drives Permafrost Formation?
Atmosphere
Vegetation
Buffer
Layer
Snow cover
Organic layer
Permafrost
Geothermal
Air Temp (C)
Observed Air
Temperature (C)
2003-4
Snow Depth (cm)
Observed Snow
Depth (cm)
2002-3
Soil Depth (m)
Observed Soil
Temperature (C)
1996-7
Month
Soil Temperature (C)
Barrow,
Alaska
Repeated Soil Freeze/Thaw Cycles Shape
Permafrost Landscape
• Water expands ~9% when it freezes into ice
• Frost Heave: rising of ground surface when
ground water Freezes
• Thaw Settlement: settling of ground surface
when ground ice melts
• Moisture Movement: soil moisture moves from
unfrozen zone to frozen front
Permafrost Features
Frost Heave, Yamal
Stone Circles, Svalbard
Stone Circles, NW Territories
Stripes, Glacier NP
Permafrost Degradation [IPCC, 2007]
4 to 6 C increase in 20th Century
2 to 3 C in last 30 years
1 to 3 °C increase
in past several
decades
>3 °C increase
mid-1950s to 1990
0 to 1 °C
increase since
1970s
Russian Permafrost Temperature Trends
Temperature Anomaly (°C)
4
3
2
0.2 m; Trend = +0.78°C/decade
0.4 m; Trend = +0.79°C/decade
0.8 m; Trend = +0.65°C/decade
1.6 m; Trend = +0.55°C/decade
3.2 m; Trend = +0.66°C/decade
1
0
-1
-2
Frauenfeld et al. [2004]
Zhang et al. [2005]
-3
-4
1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000
Year
Russian Active Layer Trends
Active Layer Depth Anomaly (m)
0.3
1960–1998 Change: +25 cm
0.2
0.1
0
-0.1
-0.2
Frauenfeld et al. [2004]
Zhang et al. [2005]
-0.3
1955 1960 1965 1970 1975 1980 1985 1990 1995 2000
Year
Talik Development
• Talik: Unfrozen soil layer above permafrost, but
below seasonally frozen surface layer
Soil Temperature at 3.2 m in Central Siberia (°C)
1.0
Talik
Permafrost
0.5
Soil Temperature (°C)
Seasonally
frozen
ground
0.0
-0.5
-1.0
-1.5
Talik
Forms
-2.0
-2.5
-3.0
1950
1960
1970
Year
1980
1990
2000
Permafrost Carbon
• 1672 Gt C in permafrost
[Tarnocai et al., 2009]
• 750 Gt C in atmosphere
Roots, Siberia [Zimov et al., 2006]
Mammoth, Siberia
Humus, Siberia [Davis, 2000]
Projections of Permafrost Degradation
Source
Schaefer et al . [2010]
Zhang et al . [2008]
Saito et al . [2007]
Lawrence and Slater [2005]
Lawrence et al . [2008]
Reduction in
Increase in
Permafrost Area Active Layer
by 2100 (%) Thickness (cm)
4
16-19
60
90
90
19-28
30-80
100-300
500
500
• General Pattern: lose area from the south,
increase active layer thickness everywhere
Feedbacks to Atmosphere
• Energy balance
•
•
•
•
Snow Albedo Feedback
Vegetation Albedo Feedback
Sea Ice Loss and Arctic Amplification
Bowen ratio seasonality
• Trace Gas Feedbacks
•
•
CO2 Fertilization
Permafrost Carbon Feedback
Net Ecosystem Exchange (NEE)
NEE = Respiration - Photosynthesis
Enhanced by
Permafrost
Carbon
Feedback
Enhanced
by CO2
Fertilization
• NEE < 0 means net carbon uptake
58±19 Gt C by 2200 is a lot of carbon
• 3.5% of permafrost carbon
• 26±9 ppm increase comparable
Vostok Ice Core (80 ppm)
• 13-27% of global land sink
• 4±1% of fossil fuel emissions
for 700 ppm target
Walking Points on Permafrost
• Freeze/thaw cycles shape the landscape
• Permafrost degradation has already started
• Permafrost Carbon Feedback will impact
climate and fossil fuel reduction strategies
When is the Permafrost Carbon
Tipping Point?
Kevin Schaefer1, Tingjun Zhang1,
Lori Bruhwiler2, Andrew P. Barrett1
1National
Snow and Ice Data Center, University of Colorado
2NOAA Earth System Research Laboratory
Observed Permafrost
Rock Circle Formation
Expand out in winter when frozen
Drop down in spring when thawed
Permafrost Area by Country
Permafrost Class by Country
Seasonally Frozen Ground
Variations of area extent of seasonally frozen
ground and snow in the Northern Hemisphere
during the winter of 1998/99.
Seasonally Frozen
Ground
Monthly maximum area
extent of seasonally frozen
ground
Seasonally frozen ground
is ~65 x 106 km2 or 68% of
the land area in the
Northern Hemisphere.
A Permafrost Model
Site-specific factors
(albedo, roughness,
slope, aspect, snow,
soil texture, etc.)
Climate/Weather
Soil moisture
conditions
Q*  QH  QLEQG = 0
Soil thermal
properties
Thermal diffusion
equation
Ground
temperature regime
Geothermal
heat flux
Modeling Permafrost
Atmosphere
k = ks()
Snow
hs(t)
Frozen
ground
= (x,t)
C= CFr(x,T)
k = kFr (x, T)
Boundary condition: Prescribed temperature, or
heat flux, or surface energy balance
Moving boundary: heat
conduction in deforming medium
Snow-soil interface: heat conduction with
or without phase change
C= CFr (x,T)
T(Zfr ) = Tf
Moving phase plane
k = kTh (x, T)
Thawed
ground
C = CTh(x,T)
Heat conduction
T(Zth ) = T f
Permafrost or
unfrozen ground
Lower Boundary
Moving phase plane: heat conduction
with or without phase change
Boundary condition: prescribed
temperature or heat flux
Permafrost classification
• By area coverage
•
•
•
•
Continuous (>90% of area)
Discontinuous (50-90% of area)
Sporadic (10-50% of area)
Isolated (<10% of area)
• By Location:
•
•
•
•
Terrestrial
Sub-ice
Sub-sea
Relic
• By Coupling with climate:
•
•
Exposed (terrestrial)
Submerged (sub-ice, sub-sea, and relic)
Frozen Ground Data Products
• http://nsidc.org/fgdc
•
•
•
•
•
•
•
•
•
•
Arctic EASE-Grid Freeze and Thaw Depths, 1901 - 2002
Arctic Soil Freeze/Thaw Status from SMMR and SSM/I,
Version 2
Circumpolar Active-Layer Permafrost System (CAPS)
Global Annual Freezing and Thawing Indices
Modeled Daily Thaw Depth and Frozen Ground Depth
Northern Hemisphere EASE-Grid Annual Freezing and
Thawing Indices, 1901 - 2002
Northern Hemisphere Seasonal and Intermittently Frozen
Ground Areas 1901-2001
Russian Historical Soil Temperature Data
Time Series of Active Layer Thickness in the Russian Arctic,
1915-1990
Circumpolar Active-Layer Permafrost System (CAPS)
Permafrost Monitoring
Permafrost Carbon in SiBCASA
Dactive
Dthreshold
Soil Carbon Pools
Dactive
Dthreshold
Thawed Carbon
Permafrost
Carbon Pool
Permafrost
Carbon Pool
Dactive = active layer depth
Dthreshold = 1973-2001 maximum active layer depth
permafrost carbon density is 2% by mass
Slow (80%)
Metabolic
(5%)
Structural
(15%)
Download