First eddy covariance flux measurements by PTR-TOF

Abstract

The recently developed PTR-TOF instrument was evaluated to measure methanol fluxes emitted from grass land using the eddy covariance method. The high time resolution of the PTR-TOF allowed storing full mass spectra up to m/z 315 with a frequency of 10 Hz. Three isobaric ions were found at a nominal mass of m/z 33 due to the high mass resolving power of the PTR-TOF. Only one of the three peaks contributed to eddy covariance fluxes. The exact mass of this peak agrees well with the exact mass of protonated methanol (m/z 33.0335). The eddy covariance methanol fluxes measured with PTR-TOF were compared to virtual disjunct eddy covariance methanol fluxes simultaneously measured with a conventional PTR-MS. The methanol fluxes from both instruments show excellent agreement.

Fig. 1

Six minutes sum spectrum from m/z 20 to m/z 315 containing 150 000 data acquisition bins (0.2 ns time intervals).

Fig. 2

Cutout of a 4 h (26 July 2009; 10:00–14:00) integrated PTR-TOF mass spectral signal at m/z 33 (red squares) and the corresponding bin-wise calculated covariances $F_{\text{cps}}^{\text{bin}}$ (black dots) with corresponding standard deviation intervals (calculated based on 30 min resolved data) in arbitrary units (au).

Fig. 3

(a) two raw PTR-TOF sum spectra (26 July 2009, at 10:00 (black squares) and 23:00 (circles)) at m/z 33, (b) same sum spectra after performing a mass shift correction.

Fig. 4

Peak fitted 6 min sum spectrum 1 (26 July 2009, at 10:00) together with a single 10 Hz spectrum within that 6 min and corresponding summation intervals (hatched areas).

Fig. 5

Calculated covariances $F_{\text{ncps}}^{\text{stick}}$ for m/z=32.9957 Th, m/z=33.0204 Th and m/z=33.0335 Th (methanol) in arbitrary units (au).

Fig. 6

Comparison of eddy covariance fluxes $F_c^{\text{PTR-TOF}}$ (red circles) and virtual disjunct eddy covariance fluxes $F_c^{\text{PTR-MS}}$ (black squares) using 10 Hz PTR-TOF and disjunct PTR-MS data; the inset depicts the corresponding 1:1 relationship and an orthogonal distance linear regression.