This eddy platform on Toolik Lake was installed in 2010 as part of an NSF-NEON grant to Dr. Gus Shaver, and has been maintained recently by an NSF AON grant to Dr. George Kling.  This station measures air temperature and relative humidity (Campbell Scientific CS215-L), air pressure (Campbell Scientific CS106), wind speed and direction (with Met One 034b anemometer), and upwelling and downwelling shortwave and longwave radiation (Kipp and Zonen CNR1 Radiometer).  Water temperature is measured with a PME thermistor chain.  Measurements of all variables are taken at a 10 second interval, averaged every 30 minutes, and logged into a Campbell Scientific CR3000 datalogger.  The eddy platform is anchored in place as soon as possible in June (when ice is off lake) and remains in place until mid to late August. The meteorological data are available in a separate dataset. This one contains all the flux (turbulence) data measured during the 2009 to 2015 summer seasons. 2009 was not used in publications since it was only a short time series with the first valid test measurements, but are included for sake of completeness.

The measurement set-up consisted of a Campbell CSAT-3 ultrasonic anemometer-thermometer and two infrared gas analyzers (IRGA) with high throughflow to measure covariances (fluxes) between the vertical wind speed component and the high-frequency (20 Hz) fluctuations of the gas concentrations of CO2, CH4, and H2O. The two IRGAs were (1) a Licor 7000 closed-path non-dispersive infrared gas analyser measuring CO2 and H2O in mole fractions in the sample cell relative to the reference cell that we kept CO2 and H2O free using a scrubber with soda-lime and magnesium perchlorate; and (2) a Los Gatos Research off-axis integrated cavity output spectrometer. The first instrument was a fast methane analyzer (FMA) that only measured CH4 mole fractions in moist atmosphere. Later we modified the instrument and used components of a fast greenhouse gas analyser (FGGA) that has two near-infrared lasers, one for the measurement of CH4 and H2O fluctuations, the other for CO2 fluctuation measurements, both in moist air.

Data processing for each 30-minute averaging interval included (1) rotating coordinates of the 3-d wind vector to align the U component with the mean wind streamlines (that is, two rotations are made so that V=0 m/s and W=0 m/s on average); (2) finding the time lag between the W' wind component and the scalar fluctuations of the IRGAs; this was done individually for each scalar and instrument, using a realistic search window based on pump speed and sample cell volume; (3) correcting the raw fluxes for concurrent water vapor density fluctations (WPL correction) for CH4 fluxes, because these mole ratios were measured in moist atmosphere and H2O fluctuations are nonzero in the sample cell. See Hiller et al. (2012) for a detailed description. Their set-up was almost identical to the one used on Toolik Lake, although with a different brand ultrasonic anemometer and a different brand vacuum pump, but same tube lengths and data acquisition. The data acquisition system was described in Eugster and Plüss (2010). The dataset referenced here was used in the publication by Eugster et al. (2020), where additional details can be found.

References

Eugster W, DelSontro T, Shaver G R, Kling G W (2020) Interannual, summer, and diel variability of CH4 and CO2 effluxes from Toolik Lake, Alaska, during the ice-free periods 2010–2015.  Environmental Science: Processes & Impacts, DOI: 10.1039/D0EM00125B

Eugster W, Plüss P. (2010) A fault-tolerant eddy covariance system for measuring CH4 fluxes. Agricultural and Forest Meteorology 150: 841–851, DOI: 10.1016/j.agrformet.2009.12.008

Hiller R V, Zellweger C, Knohl A, Eugster W (2012) Flux correction for closed-path laser spectrometers without internal water vapor measurements. Atmospheric Measurement Techniques Discussion, 5: 351–384, DOI: 10.5194/amtd-5-351-2012