Isotopic Composition of Net Ecosystem CO2 Exchange at Harvard Forest EMS Tower since 2011

This archive features long-term measurements of the eddy and storage fluxes of 16O12C16O, 16O13C16O, and 18O12C16O at the Harvard Forest EMS flux tower. Provided are the individual isotopologue fluxes, the total CO2 flux, the δ13C and δ18O isofluxes, and various ancillary flux and environmental data. The data are described in Wehr et al (2013), Long-term eddy covariance measurements of the isotopic composition of the ecosystem–atmosphere exchange of CO2 in a temperate forest, Agricultural and Forest Meteorology 181, 69–84. They are also analyzed in Wehr and Saleska (2015), An improved isotopic method for partitioning net ecosystem–atmosphere CO2 exchange, Agricultural and Forest Meteorology 214-215, 515–531, as well as in Wehr et al 2016, Seasonality of Temperate Forest Photosynthesis and Daytime Respiration, Nature (in press).
The eddy (iso)fluxes were measured by eddy covariance (EC), with a 30- or 35-minute integration period on a 40- or 45-minute duty cycle (the precise duty cycle was changed during the record to accommodate various synergistic measurement campaigns). The storage fluxes were measured as the increase in storage below 29 m during the EC integration period, based on vertical integrations over 7 air sampling heights on the tower (0.2, 1.0, 7.5, 12.7, 18.1, 24.1, 29.0 m, prior to July 3, 2012), or over 6 air sampling heights on the tower (0.2, 1.0, 7.5, 12.7, 18.1, 29.0 m, after July 3, 2012). Some periods are missing at regular intervals because the system was being used for other measurements, not reported here.
Corrected and uncorrected versions of the eddy (iso)fluxes are provided; the corrections account for high-frequency signal attenuation, and were made by comparing w-CO2 and w-T cospectra. The precise method is novel and complex and is described, along with all further details of the measurements, in Wehr et al (2013), Long-term eddy covariance measurements of the isotopic composition of the ecosystem–atmosphere exchange of CO2 in a temperate forest, Agricultural and Forest Meteorology 181, 69–84. The original, uncorrected cospectra are provided as well, for those wishing to employ an alternate correction method. These are not Fourier cospectra but rather multiresolution cospectra, as described in the following reference: Howell and Mahrt (1997), Multiresolution flux decomposition, Boundary-Layer Meteorology 83, 117-137. The covariance is the simple sum (i.e. not the discretized integral) of all 15 points in the corresponding multiresolution cospectrum.
This archive also includes measurements of the total molar mixing ratio of CO2 to dry air and of the isotope ratios δ13C and δ18O, at each of the 7 heights. For the lower six heights (or the lower 5 heights after July 3, 2012), these measurements are provided only as 40-second averages of the original 4 Hz data (which are too copious to archive here), made once every 40 minutes. For the 29 m inlet (the main inlet, used for EC), these measurements are provided as 2-minute averages of the 4 Hz data, made continuously during the 30-minute EC integration periods. The last 40-seconds of each EC integration period is included also as the 7th height (6th height after July 3, 2012) in the vertical profile record.
The file hf209-10-qc-filtered.csv contains a subset of the data in the other files, after quality control filtering that removed data points for which the sonic anemometer data were likely erroneous.