Docstoc

Northern Eurasian Wetlands and t

Document Sample
Northern Eurasian Wetlands and t Powered By Docstoc
					 Overview: Diagnosis and prognosis of effects of
changes in lake and wetland extent on the regional
      carbon balance of northern Eurasia




                    Ted Bohn
      Princeton Workshop, December 4-6, 2006
                           Outline
• Project Details
• Motivation
   – Carbon, Water, and
     Climate
   – High-latitude Wetlands
   – Lakes
• Science Questions
• Strategies
   – Modeling
   – Remote Sensing
   – Validation: In-Situ Data
• Preliminary Results
• Future Work
                                     (Freeman et al., 2001)
                              Project Details
Part of Northern Eurasia Earth Science Partnership Initiative (NEESPI)

Personnel:
• PI:
     –   Dennis Lettenmaier (University of Washington, Seattle, WA, USA)
•   Co-PIs:
     –   Kyle MacDonald (NASA/JPL, Pasadena, CA, USA)
     –   Laura Bowling (Purdue University, Lafayette, IN, USA)
•   Collaborators:
     –   Gianfranco De Grandi (EU Joint Research Centre, Ispra, Italy)
     –   Reiner Schnur (Max Planck Institut fur Meteorologie, Hamburg, Germany)
     –   Nina Speranskaya and Kirill V. Tsytsenko (State Hydrological Institute, Russia)
     –   Daniil Kozlov and Yury N. Bochkarev (Moscow State University)
     –   Martin Heimann (Max Planck Institut fur Biogeochemie, Jena, Germany)
     –   Ted Bohn (University of Washington, Seattle, WA, USA)
     –   Erika Podest (NASA/JPL, Pasadena, CA, USA)
     –   Ronny Schroeder (NASA/JPL, Pasadena, CA, USA)
     –   KrishnaVeni Sathulur (Purdue University, Lafayette, IN, USA)
             NEESPI Region




         Wetlands
Forest




Crops               Grass/Shrub/Tundra
   Carbon, Water, and Climate
                         • Human impact since 1750
                             – Emissions of 460-480 Gt C (as
                               CO2)
                                 • Burning of fossil fuels: 280 Gt C
                                 • Land-use change: 180-200 Gt C
                             – Atmosphere’s C pool has only
                               increased by 190 Gt C (~ 40% of
                               emissions)
                             – Land and ocean have taken up
                               the remainder (roughly 150 Gt C,
                               or 30%, each)
                         • Ability of land/ocean to continue
                           absorbing C is limited and
                           depends on climate
                         • Hydrology plays a major role


(Keeling et al., 1996)
         Terrestrial Carbon Stocks
                                                     (IPCC 2001)




•Wetland soils store the most carbon per unit area
•Wetland extent depends on hydrology
•Wetland behavior depends on hydrology
  High-Latitude Wetlands – Boreal Peatlands
Dual role in terrestrial carbon cycle
• Carbon Sink
   –   cold T & saturated soil for most of year
   –   NPP > Rh and other C losses
   –   70 TgCyr-1 (Clymo et al 1998) - very uncertain
   –   Current storage: 455 Pg C (1/3 of global soil C, ¼ of global terrestrial C) (Gorham
       1991)
• Methane Source
   – Saturated soil → anaerobic
     respiration
   – 46 TgCyr-1 (Gorham 1991;
     Matthews & Fung 1987) very
     uncertain
   – Roughly 10 % of global
     methane emissions
   – Methane is a very strong
     greenhouse gas
                                                          (Wieder, 2003)
• Balance of these effects depends on climate
   – Climate feedback
                Peatland H2O Budget
      Precipitation (P)                        Evaporation (E)
                          Transpiration (Tr)




    Living Biomass
                                                                 Surface
    Acrotelm                                                     Runoff (Qs)
                                                                                To
    Water Table                                                                Ocean
                                                                    Streams


From Upslope                   Subsurface Flow (Qsb)             Groundwater

    Catotelm


          Water Table = f(P, E, Tr, Qs, Qsb)
                               Peatland C Budget*
                     (NPP – Rh ≈ 200-300 Tg C/y)
                    CO2                     CO                CH4 (45 Tg C/y)
                                                       2
                                                                            CO2 (25-40 Tg C/y)
                                                                                  Fire
                   NPP

        Living Biomass
                                 Litter                                         CO2 (25-50 Tg C/y)
        Acrotelm                                                                   Outgassing
                                                 Aerobic Rh
                               Org C                                                    To
        Water Table                                                     DOC      DOC Ocean
                                                                         Streams  (25 Tg C/y)
                               DOC                 Anaerobic Rh

 From Upslope                             Subsurface Flow (Qsb)

        Catotelm

             Carbon balance = f(NPP, T, water table, fire, DOC export)
* Extremely crude estimates!
 Peatland Distribution in N. Eurasia
                                             West Siberian
Majority of world’s peatlands are in Eurasia
                                              Lowlands
                                           (mostly peatlands)


                                            Belt of major peat
                                            accumulation
                                            overlaps with:

                                            •boreal forest
                                            (taiga)

                                            •permafrost




                                              (Gorham, 1991)
              High-Latitude Lakes
• Accumulate large amounts of carbon
   – Lakes worldwide accumulate 42 Gt C/yr in their sediments (Dean and
     Gorham, 1998)
• Vent terrestrial carbon to the atmosphere
   – Respiration > Productivity in most lakes (Kling et al., 1991, Cole et al.,
     1994)
   – R:P correlates with DOC (del Giorgio et al., 1994)
   – DOC is imported from surrounding terrestrial ecosystems (especially
     true near wetlands)
   – Some Arctic terrestrial ecosystems may become net sources of
     atmospheric carbon when DOC loss is taken into account
• NE Siberian thaw lakes are strong methane sources (Walter et al.,
  2006)
   – Decomposition of “fresh” carbon in newly-thawed soil under lakes
   – Substantial amounts of C could be liberated as methane if all permafrost
     were to thaw
                     Lake H2O Budget
                           Precipitation (P)   Evaporation (E)




Streams, Surface Runoff,
Groundwater
                                                          Streams
                                                                  To
                                                                 Ocean




                    Balance: P + Qin = E + Qout
                           Lake C Budget
                                    CO2                  CO2, CH4 CO2


                                                                       Dis-
                                                                       solution
                                                             Evasion
                                    NPP

                                    Algae          Aerobic
                           DOC                       Rh
Streams, Surface Runoff,
Groundwater                POC                                                DOC
                                    POC
                                       (~30%)                           Streams
                                 Sediment                                      To
                                            42 Gt C/yr
                                 Deposition                                   Ocean

                                                   Anaerobic Rh


        Balance: TOCin + NPP = Rh + TOCout
   High-Latitudes Have Experienced Change
•Increasing precipitation (Serreze et al., 2000)
•Increasing river discharge (Peterson et al., 2002)
•Growing/shrinking lakes (Smith et al., 2005)
                               •Thawing of permafrost
                               (Turetsky et al., 2002)
                               •Increased outgassing of
                               methane (Walter et al., 2006)
                DOC Export
                         DOC export from Arctic
                         land into Arctic Ocean:
                         25.1 Tg C/y (Opsahl et al.
                         1999)

                         Peatlands supply most of
                         this (Pastor et al. 2003)

                         Higher DOC in streams
                         can drive outgassing of
                         CO2 (evasion)

                         Fry and Smith, 2005:
                         •Permafrost zone: DOC
                         export small

                         •Permafrost-free zone:
                         DOC export large
(Opsahl et al., 1999)
        Main Science Issues
• High-latitude lakes and wetlands are
  potentially large sources of CO2 and CH4
• Fluxes and extent are sensitive to climate
  (especially hydrology)
• Lake/wetland extent is underrepresented
  by low-resolution remote sensing
• Long time series of high-resolution remote
  sensing data not available
               Science Questions
• Overarching Science Questions:
   – How have changes in lake and wetland extent in northern Eurasia over
     the last half-century affected the region’s carbon balance?
   – What will the effects be over the next century?

• Sub-Topics:
   – What areas within the region have been/will be affected by changes in
     lake/wetland extent?
   – How are ongoing changes in the tundra region affecting the dynamics of
     wetlands?
       • Changes in permafrost active layer depth
   – How are/will these changes affect the carbon balance of the region?
   – How well can current sensors (MODIS, SAR) detect changes in wetland
     extent?
   – Can high-resolution SAR products be used to provide seasonal and
     interannual variations in lake/wetland extent?
       • Extend the rapid repeat cycle of lower-resolution products like MODIS
           Modeling Strategy
Integrate several models:
• VIC – hydrology (incl. frozen soil, water table,
  explicit lake/wetlands model)
• BETHY – fast ecosystem processes on sub-daily
  timescale (photosynthesis, respiration)
• Walter-Heimann (WHM) methane model –
  methane emissions on daily timescale
  – CH4 flux = f(water table, soil T, NPP)
• LPJ – slow ecosystem processes on yearly
  timescale (change in plant assemblage, fire)
          VIC: Large-scale Hydrology
 • Typically 0.5- to 0.125-degree grid cells
 • Water and energy balance        Inputs
 • Daily or sub-daily timesteps    • Meteorology:
 Mosaic of veg tiles;                            – Gridded ERA-40 reanalysis
 Penman-Monteith ET                       •     Soil parameters:
                                                 – FAO soil properties
                                                 – Calibration parameters
                          Heterogeneous              • Soil layer depth
                          infiltration/runoff        • Infiltration
                                                     • Baseflow
                                          •     Vegetation parameters:
                                                 – Observed veg cover fractions
                                                   (AVHRR)
                                                 – Veg properties from literature
                                       Outputs
                            Non-linear • Moisture and energy fluxes and
                            baseflow      states
Multi-layer soil column                • Hydrograph (after routing)
    VIC Lake/Wetland Algorithm

   soil                                                       land surface
saturated                                                     runoff enters
                                                                  lake

                                                              evaporation
   lake                                                       depletes soil
recharges                                                       moisture
    soil
 moisture

            Lake drainage = f(water depth, calibration parameter)
             Model Integration
         Precipitation,    Obs Met Data or
                                                        Obs or
        Air temperature,   Climate Model
                                                        Projected
        Wind, Radiation                                 [CO2]


  VIC                        BETHY                      LPJ

  •Hydrology                •Photosynthesis             •Species
                            •Respiration                distribution
                            •C storage                  •Fire
                 Soil moisture,
                                              C fluxes
               evapotranspiration
                                       Plant functional types

Water table,                         NPP
Soil temperature
                                                              (Completed)
                    Walter-Heimann Methane Model
                    •Methane emissions
   Validation: Remote Sensing
JERS: 100m SAR imagery
1 mosaic, acquired 1997/1998
      Validation: In-Situ Data
• Landcover classifications:
  – 5-yearly landcover summaries (SHI) 1950s-
    1990s
• Hydrological observations:
  – Soil moisture (SHI) 1960s-1980s
  – Evaporation (pan & actual) (SHI) 1960s-
    1990s
• Carbon fluxes:
  – TCOS towers (hourly, 1998-2002)
soil moisture

soil moisture
    and T

evap

flux tower
         Preliminary Results




Valdai/Fyodorovskoye Sites   Ob Site
Hydrology at Valdai
Estimated Methane Emissions at Valdai
Carbon Fluxes at Fyodorovskoye Tower
            Productivity and Respiration
g C/m2d




                 Net CO2 Emissions
g C/m2d




                       Date
                     Future Work
• Remote Sensing:
  – Validation of remote sensing classifications
      • In-situ data
      • Other remote-sensing products
  – Extension of classifications back in time via relationships with
    other remote sensing or in-situ products
• Models:
  – Finish integration of models
      • Add photosynthesis, respiration, etc. to VIC
      • Take into account decomposition of carbon formerly locked up in
        permafrost (specifically: yedoma)?
      • DOC leaching from terrestrial systems
      • Take into account C cycling in lakes
      • Add long-term vegetation dynamics
                          Future Work
– Validate models against historical observations
    • Landcover timeseries (from remote sensing/in situ data)
         – Lake extent (seasonal)
         – Wetland extent
         – Vegetation cover
    • Hydrological fluxes and storage
         –   soil moisture and temperature
         –   evaporation
         –   runoff
         –   water table
         –   snow depth and cover
    • Carbon fluxes and storage
         –   CO2
         –   CH4
         –   standing biomass
         –   soil carbon profiles
         –   DOC in soil, streams, lakes
         –   C accumulation rates in soils, lake sediments
– Expand from point estimates to regional estimates
– Use climate models to predict changes over next century
Thank You




 (Corradi et al., 2005)
Peatlands: Long-term C Sink but
  Initial Greenhouse Source
Adding 1 m2 of peatland produces                 Methane Greenhouse
the equivalent CO2 emissions:                    Warming Potential (GWP):
                                                 •62 (20 years)
  6 g CO2/m2day over next 20 years               •23 (100 years)
                                                 •7 (500 years)

   1 g CO2/m2day over next 100 years             Compared to CO2, CH4 is
                                                 a stronger, but shorter-lived,
           0 net greenhouse effect               greenhouse gas
           over next 149 years


                                   Net greenhouse
                                   sink thereafter

                                                     Removing 1 m2 of peatland
                                                     is initially a greenhouse sink,
                                                     then a source
            Friborg et al., 2003
          Modeling Strategy
• Previous Studies:
  – Coarse statistical relationships between soil
    moisture and methane emissions
  – Some used explicit ecosystem C-cycling
  – Some handled frozen soils
  – None used explicit lake/wetland formulations
  – Large disagreement on magnitude of future
    emissions

				
DOCUMENT INFO