RESEARCH LOCATIONS:
Imnavait Creek, Alaska 2003: Measurements were made on 8 flux plots in different vegetation types along the topographic sequence of the west facing slope of Imnavait Creek catchment. A light response curve was measured on each plot several times during early July and several times during late July/early August.

Imnavait Creek, Alaska 2004: Measurements were made on 15 plots along the topographic sequence of the west facing slope of Imnavait Creek catchment, in the same area as 2003. A light response curve was measured on each plot once in late June/early July and once in August.

Toolik Lake, Alaska 2004. Measurements were made on a total of 20 flux plots in treatment (N +P) and control blocks of the LTER fertilization experiment at Toolik Lake, including the moist acidic tussock, non-acidic tussock, non acidic non-tussock, inlet wet sedge and heath sites. Light curves were measured on each plot once in late June/early July and once in August.

Abisko, Sweden 2004. Measurements were made on 11 plots at the "Stepps" site, each measured once in July and once in August. Measurements were made on 13 plots at Paddus and 12 plots at Latnjajaure, once in July and August.

Abisko, Sweden 2005. 15 plots were located at the "Stepps" site, measurements were made on 3 occasions through June/ early July and once in mid August. During the 3rd phase of measurements in June we conducted a diurnal series on five of the plots, measuring a light curve once every 4 hrs for 28 hrs.

Svalbard 2005. Measurements were made on 41 plots at various sites in Adventdalen, one light curve was measured on each plot during July.

Zackenberg 2006. Data were collected on 35 1 x 1m plots and 25 0.3 x 0.3m plots across a range of vegetation types. 13 of the 0.3 x 0.3m plots were located within 1m x 1m plots (identified by plot name with suffix "b1, b2 or b3"), the remainder were independent of the 1 x 1m plots (B14-25). A light curve was measured on each plot once, all measurements were made from 8 July through 1 Aug 2006.

Toolik, Alaska 2009: Data were collected on nine 1m x 1m flux plots, three in each vegetation type. Vegetation types included moist acidic tundra (MAT), dry heath (HTH) and wet sedge (WSG). Each plot was measured at approximately 2 weeks intervals from mid-June to mid-August 2009.

Anaktuvuk River Burn, Alaska 2009: Data was collected on five 0.7m x 0.7m flux plots along a burn severity gradient during mid-June 2009.

Barrow, Alaska 2009: Data was collected on thirteen 0.7m 0.7m flux plots along a moisture gradient during late-July to early-August 2009.

EXPERIMENTAL DESIGN AND METHODS:
CO2 and H2O fluxes were measured using a Licor 6400 photosynthesis system (Li- Inc., Lincoln, Nebraska, USA) connected to a plexiglass chamber. For most sites, the chamber used measured 1m x 1m x 0.25m. Because of the limited accessibility of some sites, a smaller chamber was occasionally utilized: at Zackenberg in 2006 a 0.3 x 0.3 x 0.15m chamber was used on 25 plots and at Barrow and Anaktuvuk River Burn sites in 2009 a 0.7 x 0.7 x 0.25m chamber was used on a total of 18 plots. In Alaska in late season 2004 we measured CO2 and H2O fluxes using a LiCor 6200 photosynthesis system (note: the Licor 6200 does not measure air pressure, estimates from Toolik weather station data are used, with correction for altitude at Imnavait). We also used a LiCor 6200 at Zackenberg, from 25th July onwards - this instrument was borrowed from Susanne Konig in the Copenhagen University group. Measurements made with 6200 system are corrected for effects of water vapor flux using protocols recommended by Hooper et al. 2002. Uncorrected flux values from the 6200 are also given in a separate column.

We fitted the 1m x 1 m chamber over a square aluminum base supported several centimeters above the ground surface by hollow steel legs driven down to the permafrost. An airtight seal was created between base and chamber by lining adjoining surfaces with rubber gasket. We sealed the base to the tundra by weighting an attached plastic skirt with heavy chains; where possible we pushed the chains firmly down into the moss layer to create a good seal. We screwed the LiCor custom chamber head attachment over holes drilled into the plexiglass chamber, again sealing with rubber gasket. The air in the chamber was mixed using four small fans powered by a 12v battery.

The 0.3m x 0.3m chamber was set up identically to the 1m x 1m chamber, using a smaller aluminum frame. This was also sealed to the ground with a plastic skirt and chain. Often it was possible to make the frame level without using legs. We used only one 12V fan to mix air in the chamber, the chamber head was attached exactly in the same way as for the 1m x 1m chamber.

The 0.7m x 0.7m chamber was set up identically to the 1m x 1m chamber, again using a smaller aluminum frame. This was also sealed to the ground with a plastic skirt and chain. We used four 12V fans to mix air in the chamber and the chamber head was attached exactly in the same way as for the 1m x 1m chamber.

At each plot (all chamber sizes) we took measurements to create a light response curve: 2-3 measurements were made at ambient light conditions, followed by 2 flux measurements at each of 3 successive shading levels, with finally 3 measurements under complete darkness. We shaded the chamber by layering 3 fine mesh net cloths, with a tarpaulin to block all light from the chamber. Flux measurements under complete darkness represent ecosystem respiration. At each light level a flux measurement lasted 45 - 60 secs in total, CO2 and H2O concentrations in the chamber being recorded by the Licor every 2-3 secs. After each measurement we lifted the chamber until CO2 and H2O concentrations had stabilized at ambient levels.

After each light curve we determined chamber volume by taking depth measurements from the top of the chamber base to the ground. For the 1m x 1m base 36 depth measurements were made at regular 20cm intervals by setting a 1m x1m plastic frame with a 20cm x 20cm string grid on top of the base. For the 0.3m x 0.3m chamber 9 depth measurements were taken, one at each corner, one at the mid point of each side, and one in the middle of the plot. For the 0.7m x 0.7m chamber, a string grid with smaller intervals was utilized. The volume determined by these depth measurements (chamber surface area*average depth) was added to the volume of the chamber itself. The surface area of the inside of the 1 m x 1 m chamber was 0.89m2. In 2009, the surface area of the inside of the chamber was 0.8836m2 for 1m bases and 0.4096m2 for all 0.7m bases.

For 2003-2006 data, NEP is usually calculated from the first 15-20secs of measurement. Where the data is scattered due to very small net changes in CO2 over time, the entire period of 45-60secs is used. If light levels change during a measurement only periods where the light is constant is used to calculate the flux. Often under shade or darkness the first few seconds of data are discarded as the CO2 concentration change over time is non linear, representing adaptation of the system to new conditions. For 2009 data, NEP is calculated from the entire measurement period. When abnormalities in CO2 slope were observed due to leaks or changes in light levels, certain portions of the measurement were discarded. CO2 slope was always taken from contiguous data points (i.e. points were never removed from the middle of the measurement period). Often under shade or darkness the first few seconds of data are discarded as the CO2 concentration change over time is non linear, representing adaptation of the system to new conditions.

For 2003-2006 data, H2O flux is also usually calculated from the first 15-20secs of the measurement period. However, where CO2 flux was calculated over the entire 60sec period, water flux is also calculated over the entire period. This should be considered as the build up of water vapor in the chamber almost always followed a curve which began to plateau towards the end of the 60secs. For a more reliable measure of evapotranspiration see water fluxes calculated at time t=0secs see 2003-2004waterflux.xls.
For 2009 data, H2O flux is calculated over the same time window as CO2. Further QC of H2O data would be beneficial before use as H20 over time is nonlinear in many cases. Also, H20 fluxes from nighttime diurnal measurements are suspect.

We measured NDVI on each flux plot using either a portable 2 channel field sensor (Skye Instruments) or Unispec spectral analyzer (PP Systems) or both:

Toolik 2003: Skye sensor
Toolik 2004: Unispec and Skye sensor
Abisko 2004: Unispec and Skye sensor (except Stepps site phase 2)
Abisko and Svalbard 2005: Unispec and Skye sensor (except June at Abisko)
Zackenberg 2006: Unispec
Toolik 2009: Unispec
Barrow 2009: Unispec

The unispec spectral analyzer measures reflected light intensity in 256 portions of the visible spectrum from ~300nm to ~1100nm. A foreoptic cable transmits light reflected from the target to the instrument, a measurement scan lasts for ~10ms. 9 scans were measured in a regular grid for each of the flux plots. In 2003-2006 measurements, the end of the fiber optic was kept 80 cm vertically above the ground surface resulting in a view of the vegetation of approx. 20cm diameter. In 2009 measurements, the end of the fiber optic was kept 1 m vertically above the ground surface resulting in a view of the vegetation of approx. 20cm diameter. For the 0.3m x 0.3m plots, 3 scans were taken from directly above the plot with a field of view of approximately 30cm diameter. The average of these three scans was then taken. Incident radiation was measured using a reference standard in order that reflectance from the vegetation could be calculated as a percentage of incoming solar radiation

The program Multispec5.1.5.exe was used to compile Unispec reflectance spectra from the raw target spectra.

The skye NDVI sensor measures total radiation reflected from the vegetation (without measuring incident radiation) in the wavebands 570nm-680nm and 725nm - 1100nm. NDVI is then calculated as below.

CALCULATIONS:
Fluxes are calculated from the slope of chamber CO2 (umol mol-1) [or H2O (mmol mol-1)] concentration against time.

NEP = (rho * vol * dC/dt)/ SA rho = (P_av * 1000)/(R* T)

where NEP = net CO2 flux [umol m-2 s-1]
rho = air density [mol/m3]
P_av = pressure [kPa]
R = ideal gas constant 8.314 [J mol K-1]
T = temperature [K] = Temp_av [0c] + 273.
vol = chamber volume [m3]
dC/dt = slope of chamber CO2 conc against time [umol mol-1 s-1]
SA = chamber surface area [m2] = 1

RE = NEP during dark measurement
GEP = RE - NEP

NDVI = (RII-RI)/ (RI + RII)

where RI = average reflectance from 570nm to 680nm
RII = average reflectance from 725nm to 1000nm.

COMMENTS: All times are given as local times. Shade level was not recorded in 2003. A negative net flux indicates uptake of carbon by the vegetation, a positive net flux indicates loss of carbon from vegetation to the atmosphere. Re and GEP are both given as positive numbers. Air temperature values from the 6400 were measured using a thermistor located inside the chamber head air space. Air temperature using the 6200 was measured using a thermocouple attached to the chamber head, and inserted inside the flux chamber.

FOR MORE INFORMATION CONTACT: Gus Shaver, The Ecosystems Center, Woods Hole, MA, 02543, USA

REFERENCE CITATIONS:
Douma, J.C., van Wijk, M.T., Lang, S.I., Shaver, G.R. (2007) The contribution of mosses to the carbon and water exchange of artic ecosystems: quantification and relationships with system properties. Plant, Cell and Environment 30: 1205-1215.

Hooper D.U., Cardon Z. G., Chapin III F. S. Durant, M. (2002) Corrected calculations for soil and ecosystem measurements of CO2 flux using the LI-COR 6200 portable photosynthesis system. Oecologia 132:1–11.

Shaver, G.R., Street, L.E., Rastetter, E.B., van Wijk, M.T., Williams, M. (2007) Functional convergence in regulation of net CO2 flux in heterogeneous tundra landscapes in Alaska and Sweden. Journal of Ecology 95:802-817.

Street, L.E., Shaver, G.R., Williams, M., van Wijk, M.T. (2007) What is the relationship between changes in canopy leaf area and changes in photosynthetic CO2 flux in arctic ecosystems? Journal of Ecology 95: 139-150.

Williams, M., Street, L.E., van Wijk, M.T., and Shaver, G.R. (2006) Identifying Differences in Carbon Exchange among Arctic Ecosystem Types. Ecosystems 9:288-304.

FORMAT OF DATA FILE: ASCII

NUMBER OF RECORDS: 5001