Introduction to FLUXNET and the FLUXNET Dataset

The FLUXNET datasets consist of over 960 site-years of data from over 253 eddy covariance measurement sites, it supercedes the Marconi FLUXNET dataset that was produced in 2000 (with 97 years of data from 38 European and North American sites). The FLUXNET Synthesis database harmonizes, standardizes and gap-fills the ‘raw’ 30-minute data records of carbon dioxide, water vapour and energy fluxes submitted by members of regional networks from around the world: CarboeuropeIP, AmeriFlux, Fluxnet-Canada, LBA, Asiaflux, Chinaflux, USCCC, Ozflux, Carboafrica, Koflux, NECC, TCOS-Siberia and Afriflux. More information about the FLUXNET project can be found here. These datasets also include value added products like gross primary productivity, ecosystem respiration, climate and site characteristic information. 

 

 

The FLUXNET community has also opened up three versions of the FLUXNET dataset including one that is freely available without requiring any prior approvals. The original dataset is the LaThuile dataset which is only available to synthesis teams including at least one data contributor. The 'Open' dataset is available for use by any synthesis group after acceptance of a proposal submitted to the steering committee. The policy for this 'open' dataset can be found here and the template for submitting a proposal here. This data set consists of site years from PIs who give us permission to designate their datasets with the status of ‘Open'. The third dataset is publicly available free fair-use without requiring a prpoposal. The 'Free Fair-Use' policy' is available. The intent of these datasets is to provide flux data to the broad community of scientists who need flux data to test, parameterize and validate land surface schemes in climate models, dynamic vegetation models, remote sensing algorithms, hydrological models, global carbon cycle models and phenology algorithms. (the meeting kicking off the effort to open the datasets to the larger community was held at Asilomar meeting agenda and pictures)

We hope that this data will enable a new “global flux perspective” where the different eddy covariance networks harmonize and share data with the aim to increase the quality of the research and the collaboration. These datasets will be regularly updated with data from new sites.

 The proposal approval process takes three weeks and is intended to minimize parallel work and maximise synergies between groups and efficiency of the analyses, a “Scientific Moderation Committee” has been created to coordinate the synthesis effort (Dennis Baldocchi, Jiquan Chen, Bev Law, Hank Margolis, Dario Papale, Markus Reichstein and Celso von Randow). At present, there are now more than 110 approved proposals.

 

Finally, a sincere thank you to the institutions that sponsored the FLUXNET-TCO Workshop and are supporting the development and implementation of the database.  These include  Max Planck Institute for Biogeochemistry – Germany, University of Tuscia – Italy, FAO-GTOS,iLEAPs, US Department of Energy, National Science Foundation –USA, Microsoft Research eScience, Berkeley Water Center, University of California - Berkeley, Lawrence Berkeley National Laboratory, University of Virginia, and Oak Ridge National Laboratory.

The FLUXNET Measurement Sites

A nice introduction to the FLUXNET measurement sites, is contained in the review paper: Baldocchi, D.D . 2008. ‘Breathing’ of the Terrestrial Biosphere: Lessons Learned from a Global Network of Carbon Dioxide Flux Measurement Systems. Australian Journal of Botany. 56, 1-26. An excerpt from the paper and links to extensive information about the sites that contributed data to the LaThuile FLUXNET synthesis dataset are below.

Today, carbon dioxide and water vapor fluxes, and many ancillary meteorological, soil and plant variables, are being measured continuously at over four hundred research sites, spread world-wide. Individual research sites, in this dispersed network, are associated with the AmeriFlux and Fluxnet-Canada (Coursolle, Margolis et al. 2006; Margolis, Flanagan et al. 2006) networks in North America, the Large Biosphere Amazon (LBA) project in South America (Keller, Alencar et al. 2004), the EuroFlux and CarboEurope networks in Europe (Ciais, Reichstein et al. 2005; Valentini, Matteucci et al. 2000), OzFlux in Australasia, China Flux (Yu, Wen et al. 2006) and AsiaFlux in Asia and AfriFlux in Africa. There also exists an urban flux network and a regional network across India is in the planning stages (Sundareshwar, Murtugudde et al. 2007).

In the early phases of the global flux network, the data records were too short to study inter-annual variability, but now many long datasets exist. At present, the dataset acquired at Harvard Forest site is over 15 years long (Urbanski, Barford et al. 2007), the Walker Branch, Tennessee and Takayama, Japan datasets and many Euroflux sites have been operating for a decade and more. With these extended datasets, scientists are starting to ask how climate fluctuations (temperature, precipitation, solar radiation), antecedent conditions (drought, freezes, extreme weather events) and length of growing season affect net carbon exchange and its component fluxes (gross canopy photosynthesis and ecosystem respiration).

The Data

At individual nodes (sites) of the network, the eddy covariance method is used to measure mass and energy exchange across a horizontal plane between vegetation and the free atmosphere. Fluxes of carbon dioxide, water vapor and heat are determined by measuring the covariance between fluctuations in vertical velocity (w) and the mixing ratio of trace gases of interest (c) (Aubinet, Grelle et al. 2000; Baldocchi 2003); negative covariance values of net ecosystem CO2 exchange (FN) represent a loss of CO2 from the atmosphere and a gain by the surface. The temporal information being produced at each network node ranges across the scales of seconds, hours, days, weeks, seasons and years. The spatial scales of observations at each tower extends through the flux footprint around the tower (ranging between 100 and 1000 m) (Gockede, Rebmann et al. 2004). However, the information produced at each node reaches far beyond its proximate geographical region due to its wider scale representativeness (Granier, Pilegaard et al. 2002; Hargrove and Hoffman 2005). Today, carbon dioxide and water vapor fluxes, and many ancillary meteorological, soil and plant variables, are being measured continuously at FLUXNET sites, spread world-wide.(Baldocchi 2007) 

Information collected at a site encompasses sensor measurements, field observations, laboratory analysis of field samples, as well as anecdotal descriptions. Each network site deploys a number of sensors measuring micro-meteorology such as precipitation or wind speed and carbon flux. The field data are processed to generate 1/2 hourly flux-met data files. The other, ancillary data include relatively infrequent measurements of variables such as soil carbon, leaf out dates, vegetation species, canopy height, soil characteristics, and site disturbances such as planting or wildfire. An important aspect of the ancillary data is that each measurement is often annotated. For more information about ancillary data and related reporting protocol see (Law etal. 2008).

The 1/2 hourly flux-met data is passed from the sites to their regional networks and then on to FLUXNET. When the data reaches FLUXNET, it is quality assessed and gap-filled using techniques described in (Papale etal. 2006, Reichstein etal. 2005, Moffat etal. 2007, and Papale and Valentini 2003). The goal of gap filling is the reproduction of the NEE time series. A diagram showing the data processing work flow to generate the LaThuile FLUXNET synthesis dataset can be found here.

FLUXNET Data Usage

There are three versions of the FLUXNET synthesis dataset: LaThuile, Opened, and Free Fair-Use. The complete set of all site years in the FLUXNET synthesis activity is referred to as the LaThuile Dataset.  This version of the dataset is available to data contributors. In order to gain access to the LaThuile Dataset, the synthesis team must include at least one data contributor and have an approved proposal. The Opened Dataset is the subset of the LaThuile Dataset that can also be shared with synthesis teams that do not include any data contributors.

To access the LaThuile or the Opened dataset a proposal must be submitted (template available here - Submission instructions available here - choose cancel if asked for authentication). All interested people can submit one or more proposals to the Opened Dataset and teams involving data contributors can submit proposals to the LaThuile Dataset. Proposals typically take three weeks for approval. The authors and contributors listed in the proposal are approved for access to the dataset once the proposal is approved by the Steering Committee.

The data can NOT be shared except with approved collaborators.  To add a collaborator to a proposal, the coordinator of the proposal must send the CV of the collaborator to the steering committee for approval.  Approval typically takes about two weeks.  If a collaborator has been approved, they can be given access to the data. This procedure applies to all contributors including data providers joining proposal teams and students in the laboratories of primary authors.

The final dataset contains site years that have been designated as accessible to Free Fair-Use Users. In order to access this dataset request a login on the fluxdata.org site. All registered users have unrestricted access to the Free Fair-Use Dataset but they must abide by the fair-use policy.

References

1. Baldocchi, D.D . 2008. ‘Breathing’ of the Terrestrial Biosphere: Lessons Learned from a Global Network of Carbon Dioxide Flux Measurement Systems. Australian Journal of Botany. 56, 1-26.
2. Coursolle C, Margolis HA, et al. (2006) Late-summer carbon fluxes from Canadian forests and peatlands along an east-west continental transect. Canadian Journal of Forest Research-Revue Canadienne De Recherche Forestiere 36, 783-800.
3. Margolis HA, Flanagan LB, Amiro BD (2006) The Fluxnet-Canada Research Network: Influence of climate and disturbance on carbon cycling in forests and peatlands. Agricultural and Forest Meteorology 140, 1-5.
4. Keller M, Alencar A, et al. (2004) Ecological research in the large-scale biosphere-atmosphere experiment in Amazonia: Early results. Ecological Applications 14, S3-S16.
5. Ciais P, Reichstein M, et al. (2005) Europe-wide reduction in primary productivity caused by the heat and drought in 2003. Nature 437, 529-533.
6. Valentini R, Matteucci G, et al. (2000) Respiration as the main determinant of carbon balance in European forests. Nature 404, 861 - 865.
7. Yu G-R, Wen X-F, Sun X-M, Tanner BD, Lee X, Chen J-Y (2006) Overview of ChinaFLUX and evaluation of its eddy covariance measurement. Agricultural and Forest Meteorology Carbon Exchange Research in ChinaFLUX 137, 125-137.
8. Sundareshwar PV, Murtugudde R, et al. (2007) ENVIRONMENT: Environmental Monitoring Network for India. 10.1126/science.1137417. Science 316, 204-205.
9. Urbanski S, Barford C, et al. (2007) Factors controlling CO2 exchange on timescales from hourly to decadal at Harvard Forest. J. Geophys. Res. 112, 10.1029/2006JG000293. 
10. Aubinet M, Grelle A, et al. (2000) Estimates of the annual net carbon and water exchange of Europeran forests: the EUROFLUX methodology. Advances in Ecological Research 30, 113-175.
11. Gockede M, Rebmann C, Foken T (2004) A combination of quality assessment tools for eddy covariance measurements with footprint modelling for the characterisation of complex sites. Agricultural and Forest Meteorology 127, 175-188.
12. Granier A, Pilegaard K, Jensen NO (2002) Similar net ecosystem exchange of beech stands located in France and Denmark. Agricultural and Forest Meteorology 114, 75-82.
13. Baldocchi, D.D . 2008. ‘Breathing’ of the Terrestrial Biosphere: Lessons Learned from a Global Network of Carbon Dioxide Flux Measurement Systems. Australian Journal of Botany. 56, 1-26.
14. Law, B. E., Arkebauer, T., Campbell, J. L., Chen, J., Sun, O., Schwartz , M., van Ingen, C., Verma, S. Terrestrial Carbon Observing Protocols for Vegetation Sampling and Data Submission. 2008. http://www.fao.org/gtos/doc/pub55.pdf
15. Papale D., et al. (2006) Towards a standardized processing of Net Ecosystem Exchange measured with eddy covariance technique: algorithms and uncertainty estimation, Biogeosciences, (3), 571-583.
16. Reichstein M., et al. (2005) On the separation of net ecosystem exchange into assimilation and ecosystem respiration: review and improved algorithm. Global Change Biology (11) 1424-1439
17. Moffat A., et al. (2007) Comprehensive comparison of gap-filling techniques for eddy covariance net carbon fluxes. Agricultural and Forest Meteorology (147) 209-232
18. Papale D., Valentini R. (2003) A new assessment of European forests carbon exchanges by eddy fluxes and artificial neural network spatialization. Global Change Biology (9) 525-535

    FLUXNET Participating Institutions