FLUXNET: Integrating Global Flux Measurements

1
FLUXNET Integrating Global Flux Measurements

• Dick Olson, Bob Cook,
• Susan Holladay, Les Hook
• ORNL DAAC
• January 2001
• NASA EOS Validation Program
• NASA EOSDIS DAAC

2
FLUXNET Overview

• Data being collected
• carbon, water vapor, and energy fluxes using
  eddy-correlation (Net Ecosystem Exchange - NEE)
• site ancillary data
• Data Access
• ORNL DAAC and registered in Mercury
• MODIS products to be validated
• MOD-17 - 8d PSN/GPP (Gross Primary Productivity)
• MOD-17 - ANPP (Annual Net Primary Productivity)

3
FLUXNET Overview - cont.

• Scaling to MODIS pixels
• towers sample 1 km2 area
• scaling being addressed by the BigFoot project
• Experience in obtaining MODIS data
• need to reproject to relate to field studies
• Plans to communicate results Real-Time
  Validation Exercise
• telecons including Instrument Team scientists
• post flux, model, and MODIS data on Web page
• synthesis papers, workshops planned

4
FLUXNET Community

• Data Component
• Dick Olson, ORNL DAAC
• Science Component
• Dennis Baldocchi, UC-Berkley
• Lianhong Gu, UC-Berkley (post-doc)
• Eva Falge, U Bayreuth (previous post-doc)
• EOS Validation Collaborator
• Steve Running, University of Montana
• Regional Network Representatives
• AmeriFlux, CarboEuroFlux, AsiaFlux, OzNet

5
FLUXNET
Data Users
Global Network
Non-network sites
Regional Networks
CARBOEURO- FLUX
AmeriFlux
AsiaFlux
OzNet
Oceania
Americas
Europe
Asia - Japan
Tower Sites
Architecture of Global / Regional Flux Networks
6
FLUXNET Validation Objectives

• Compile flux data (NEE)
• gap-filled data for core test sites with flux
  towers
• compile site characteristics
• post selected MODIS products
• Support Real-Time Validation exercise
• host telecons to plan validation
• post flux, model, and MODIS data
• Quantify controls on flux patterns
• provide scientific information to validate or
  improve underlying model/MODIS algorithms

7
Recent ResultsQuantifying Flux Patterns

• Variances across hour-to-year time scales
• Annual carbon fluxes by plant functional type
• Controls on net CO2 exchange
• light, temperature, humidity deficits and water
  use, as a function of climate and functional type
  
• length of growing season - temperate broadleaved
  forests
• cloud cover
• landscape heterogeneity
• stand age on CO2 and water vapor flux densities

8
FLUXNET Highlights

• 3 Regional Networks (AmeriFlux, CARBOEUROPE,
  AsiaFlux)
• 148 towers registered
• 62 active towers (submitted data or start year
  defined)
• 306 years of observations for active towers
• 103 site-years of data submitted to FLUXNET
• 11 years of record at a tower (Harvard)
• 12 sites with gap-filled, aggregated data
  available
• 21 site-years of gap-filled, aggregated data
  available
• 3700 ancillary data records (climate,
  vegetation, LAI, NPP, etc)
• 260 parameters in data dictionary
• 300 Flux scientists on FLUXNET email server
• 3000 Flux related bibliographic citations

9
Gap-Filled Flux data registered in Mercury
10
Flux Web Site http//daac.ornl.gov/FLUXNET/data.h
tml
11
FLUXNET core variables

• Net Ecosystem Exchange from eddy correlation, CO2
  flux (NEE)
• Sensible heat from eddy correlation, H
• Latent heat from eddy correlation, LE
• Net radiation, R(net)
• Photosynthetic Active Radiation, PAR

• Air temperature, T(air)
• Precipitation, PPT
• Relative humidity, RH
• Wind speed direction
• Vapor Pressure Deficit, VDP
• Soil temperature, T(soil)
• Carbon dioxide concentration, CO2
• Soil heat flux, G(soil heat)

12
33 sites (EUROFLUX - 14, AmeriFlux - 19) 89
site-years (average of 2.7 years/site)
13
Synthesis Publications

• Overview paper, submitted
• Temporal Dynamics and interannual variability of
  Carbon Dioxide Fluxes
• Source-Sink Controls of CO2 Canopy structure and
  physiological capacity
• 1998 Case Study, Spatial Coherence of Fluxes in
  North America during an El Nino year
• Canopy scale responses to drought
• Canopy Scale Responses of NEE, GPP and Reco to
  Environment
• Using the FLUXNET spectrum of canopy structure
  and topography to examine energy balance closure
• Examining Energy Exchange Relationships with
  FLUXNET data

14
AmeriFlux Sites in Synthesis

• Atqasuk
• Blodgett
• Bondville
• Barrow
• Duke Forest -
• Howland -
• Happy Valley -
• Harvard Forest
• Little Washita
• Metolius -
• Manaus
• Niwot Ridge

• BOREAS Old Aspen
• BOREAS Old Black Spruce
• Ponca City -
• Shilder -
• Sky Oaks - Old Stand
• Sky Oaks - Young Stand
• Upad -
• Walker Branch
• Willow Creek
• Wisconsin Tall Tower
• Wind River Crane
• Core Sites highlighted in yellow
• Ustar correction, - no correction

15
(No Transcript)
16
Real-Time Validation

• Micromet data flowing to AmeriFlux (CDIAC)
• 15 sites sending data to AmeriFlux
• 6 model groups signed up
• Univ. of Montana extracting MODIS LAI, fPAR, and
  NPP products for 11x11 km cutouts
• Univ. of Montana and FLUXNET developing model
  driver data being posted by AmeriFlux
• Flux, model, and MODIS results to be posted on
  AmeriFlux and FLUXNET Web pages
• FLUXNET hosting telecons of flux scientists and
  modelers to review 2000 and plan for 2001

17
Participating AmeriFlux Sites

• Park Falls - deciduous forest
• Willow Creek - deciduous forest
• U. Mich. Bio. Stn - deciduous forest
• Morgan Monroe - deciduous forest
• Blodgett - conifer forest
• Ponca - cropland
• Shidler - grassland
• Barrow - tundra
• Harvard - deciduous forest
• Howland - deciduous forest
• Walker Branch - deciduous forest
• Duke - conifer forest
• Sky Oaks - chaparal Core Sites
  high-
• Le Selva - tropical forest lighted in
  yellow

18
Participating Modeling Groups

• Biome-BGC - Running Nemani, UMT
• provided model initialization data for all sites
• PNET or OptiCal - Aber et al., UNH
• CENTURY - Parton et al., CSU
• GTEC - King and Post, ORNL
• CASA - Field et al., Stanford
• GLO-PEM or CEVAS - Prince Chao, UMD

19
Plans for 2001

• Update FLUXNET database
• Support Real-Time Validation
• periodic telecons
• post flux, model, and MODIS data
• plan for formal flux-model-MODIS intercomparison
• Issues
• minimize lag in acquiring flux data
• more robust gap-filling, night-time corrections
• conversion of NEE to NPP (estimates of soil
  respiration flux)

20
Number of Flux Towers
21
Key site characteristics (NPP, LAI, fPAR,
phenology, climate, hydrology, etc.) compiled
from literature, reports, Web sites, and PIs, as
well as selected remote imagery also documented
methods, processing, uncertainty, and source.
22
Core Test Sites and Networks