Document

Report
AmeriFlux,
Yesterday, Today and Tomorrow
Dennis Baldocchi, UC Berkeley
Margaret Torn and Deb Agarwal, Lawrence Berkeley
National Lab
Bev Law, Oregon State University
Tom Boden, Oak Ridge National Laboratory
AmeriFlux, circa 2012
Growth in the Network
Data from Bai Yang and Tom Boden
Age of Flux Sites, and the Length of their Data Archive
Pros and Cons of a Sparse Flux Network
• Pros
– Covers Most Climate and Ecological Spaces
– Long-Term Operation Experiences Extreme Events, Gradual Climate
Change, and Disturbance
– Gradients of Sites across Landscapes and Regions Span Range of
Environmental and Ecological Forcing Variables
– Clusters of Sites examine effects of Land Use Change, Management,
and Disturbance (fire, drought, insects, logging, thinning, fertilizer,
flooding, woody encroachment)
– Robust Statistics due to Over-Sampling
• Cons
– Can’t Cover All Physical and Ecological Spaces or Complex Terrain
– Current Record is too Short to Detect Climate or CO2-Induced Trends
– Flux Depends on Vegetation in the Footprint
– Bias Errors at Night, Under Low Winds
The Type of Network Affects the Type of Science
• Sparse Network of Intense Super-Sites and
Clusters of Sites, Producing Mechanistic
Information can Test, Validate and
Parameterize Process and Mechanistic Models
• Denser and More Extensive Network of LessExpensive Sites can Assist in Statistical and
Spatial Up-Scaling of Fluxes with Remote
Sensing
Climate Space of AmeriFlux Sites
Yang et al 2008, JGR Biogeosciences
AmeriFlux Sites, Circa 2003, and
Ecosystem/Climate Representativeness
Hargrove, Hoffman and Law, 2003 Eos
Representativeness of AmeriFlux, Circa 2008
(blue is good!)
Yang et al. 2008 JGR Biogeosciences
Basis of a Successful Flux Network
It Takes People (Scientists, Postdocs, Students and Technicians)
Social Network that Facilitates Meetings, Workshops, Shared
Leadership and a Shared/Central Data Base
This Fosters Getting to Know Each Other, Collaboration,
Communication, Common Vision, Shared Goals,
And Joint Authorship of Synthesis Papers
Past and Current Leadership
Dave Hollinger, Chair
1997-2001
Bev Law, Chair
2001-2011
Margaret Torn
Tom Boden
AmeriFlux PI, 2012- AmeriFlux
Data Archive
Published Use of AmeriFlux Data
184 Papers linked to key word
‘AmeriFlux’
246 Papers linked to key word
‘Fluxnet’
These Papers have been cited over 7000 Times
Issues of standardization, or not?
‘Know Thy Site’
Ray Leuning
Most Flux Instruments are Very Good;
Pick the Instrument System that is Most Appropriate to Your
Weather and Climate
Open-Path CO2 Fluxes were 1.7% Higher than Closed Path Fluxes
Schmidt et al. 2012, JGR Biogeosciences
Site Calibration with Roving Standard
Schmidt et al 2012 JGR Biogeosciences
Extrinsic Contributions
• Data Contribute to Producing Better Models via
Validation, Parameterization, Data-Assimilation &
Defining Functional Responses
– Land-Vegetation-Atmosphere-Climate
• Energy Partitioning, Albedo, Energy Forcing, Land Use
– Remote Sensing, Light Use Efficiency Models
• Regional and Global GPP models
– Ecosystem and Biogeochemical Cycling
• Carbon Cycle, Disturbance, Phenology, Environmental
Change, Plant Functional Types
– Hydrology
• Evaporation , Soil Moisture, Ground-Water, Drought
Lessons Learned
What’s in the Data?
• Magnitudes and Trends in Annual C and H2O Fluxes, by Plant
Functional Type and Climate Space
• Light-Use, Temperature, Rain Response Functions
• Emergent-Scale Properties
– Diffuse Light
– Rain Pulses
– Drought and Ground Water Access
• Disturbance
– Insect Defoliation
– Fire, Logging and Thinning
– Drought and Mortality
• BioPhysical Forcings
– Albedo and Temperature
– Energy Partitioning with Land Use
C Fluxes are a Function of Time Since Disturbances,
as well as Weather, Structure and Function
GPP
NEE
Reco
Harvard Forest
1800
1400
-2
Carbon Flux Density, gC m y
-1
1600
1200
1000
800
0
-200
-400
-600
1990
1992
1994
1996
1998
2000
2002
2004
2006
Year
Urbanski et al. 2007 JGR Biogeosciences
2008
Light Response Curves of CO2 Flux are Quasi-Linear,
Deviating from Monteith’s Classic Paper and
Impacting the Interpretation of C Flux with Remote
Sensing
Gilmanov et al 2010 Range Ecology & Management
Light Use Efficiency INCREASES
with the Fraction of Diffuse Light
Niyogi et al 2004 GRL
Response Functions from Elevation/Climate Gradients
Anderson-Teixeira et al. 2010 GCB
Respiration is a function of
Temperature, Soil Moisture, Growth, Rain Pulses
And Temperature Acclimation
Xu et al. 2004 Global Biogeochemical Cycles
Rain-Induced Pulses in Respiration:
Long –Term Studies Capture More Pulses, Better Statistics
Ma et al. 2012 AgForMet
Disturbance, Fire and Thinning
Dore et al. 2012 GCB
Insect Defoliation, 2007
Clark et al. 2010 GCB
Disturbance Dynamics
C Flux = f(time since disturbance)
Amiro et al. 2010 JGR Biogeosci
Flux Phenology
Gonsamo et al 2012 JGR Biogeosci
Satellite vs Flux Phenology
Gonsamo et al 2012 JGR Biogeosci
It’s Not only CO2!
Effects of Precipitation and Energy on Evaporation
MI Budyko
Williams et al. 2012 WRR
Long-Term Studies can Assess Links between Drought and Fluxes
Schwalm et al 2012 Nature Geoscience
Net Negative Effects on Carbon and Water Fluxes are Strong:
What about 2012?
Schwalm et al 2012 Nature Geoscience
Land Use and Climate
Forests are warmer
than nearby
Grasslands
Lee et al Nature 2011
Light Use Efficiency Models:
Upscale Fluxes from Towers to Regions
GPP    fPAR( NVDVI )  PAR
Yuan et al. 2007, AgForMet
   0 f (Ts ) f ( )
Heinsch et al 2006 IEEE
   0 f (VPD) f (Tmin )
C and Water fluxes Derived from Satellite-Snap Shots Scale
with Daily Integrated Fluxes from Eddy Covariance
Sims et al 2005 AgForMet
Ryu et al. 2011 AgForMet
Seasonal Maps of NEE, via Regression Tree Analysis,
on AmeriFlux and Modis Data
Xiao et al. 2008 AgForMet
What is the Truth?; How Good is Good-Enough?
Chen et al 2011 Biogeosciences
Regional Estimates of Fire, Drought, Hurricanes on NEE
Xiao et al. 2011 AgForMet
Using Flux Data to Validate Dynamic Vegetation Models-ORCHIDEE
Krinner et al 2005 GBC
Data-Model Fusion/Assimilation
Sacks et al. 2006 GCB
Model Hierarchy Testing: How Much Detail is Needed?
Bonan et al 2012 JGR Biogeosci
Testing Phenology Predictions in Ecosystem-Dynamic Models
The total bias in modeled
annual GEP was
+35 ± 365 g C m-2 yr-1 for
deciduous forests
+70 ± 335 g C m-2 yr-1 for
evergreen forests
across all sites, models, and
years;
Richardson et al, 2012 GCB
It‘s Not Just About CO2:
Significant change in albedo with 3 disturbance types
Hurricane
Beetles
Fire
Albedo change produces radiative forcing of same
magnitude as CO2 forcing in case studies of forest
mortality from hurricane defoliation, pine beetles,
and fire.
Beetle effect occurs mostly after snags fall
O’Halloran et al 2012 GCB
Albedo Scales with Nitrogen
We can Use Albedo to Parameterize N and Ps Capacity in Models!
Hollinger et al 2009 Global Change Biology
The Albedo-N Correlation may be Spurious
Knyazikhin et al 2012 PNAS report that the previously reported
correlation is an artifact—it is a consequence of variations in
canopy structure, rather than of %N.
• an increase in the amount of absorbing foliar
constituents enhances absorption and
correspondingly decreases canopy reflectance
• When the BRF data are corrected for
canopy-structure effects, the residual
reflectance
• variations are negatively related to %N at all
wavelengths in the interval 423–855 nm.
To infer leaf biochemical constituents, e.g., N content, from
remotely sensed data, BRF spectra in the interval 710–790 nm
provide critical information for correction of structural
influences
Validating and Improving Climate Drivers,
like Net Radiation Fields
Jin et al 2011 RSE
Radiation and Evaporation Maps
Jin et al 2011 RSE
Testing Ecohydrology Theories for Soil Moisture
Miller et al 2007 Adv Water Res
Current and Future Collaborations
• COSMOS and Soil Moisture Fields
• Validation of Satellite based estimates of CO2,
LIDAR, Albedo, and Soil Moisture (SMOS, SMAP,
AIRMOSS)
• Priors for CO2-Satellite Inversions (GOSAT, OCO)
• Data-Model Assimilation
• Phenology and Pheno-Camera Networks
• FLUXNET and NEON
Importance of Site Metadata,
A Plea for more LIDAR data to Test New Satellite Products and
Force 3D Ecosystem Dynamic Models
Simard et al 2011 JGR Biogeosciences
Medvigy et al 2009 JGR Biogeoscience
AmeriFlux Plans
• DOE grant to LBL to Manage 10-12 Long Term
Clusters of Flux Towers
– Ensure Cohort of Long Term Sites Extend into the
Future to Address Ecological and Climate Questions on
their Native Time Scales
• Continue Operation of Roving ‘Calibration’ system
to All AmeriFlux Sites
• Central Data Archiving, Processing and Data
Distribution
– Open Access, Prompt Submission, Uniform Processing
• Spare Sensors for Emergencies
Registered AmeriFlux Sites

similar documents