This dataset includes the CO2 flux and the FTICR-MS data reported in https://doi.org/10.1007/s10533-021-00790-y.

CO2 flux data
CO2 fluxes were measured in a laboratory incubation. Approximately 20 g (oven dry weight equivalent, ODE) of soil was weighed into each of 44 glass microcosms (487 ml). Twenty ‘trace gas’ microcosms (5 per treatment) were repeatedly monitored for CO2 production and destructively harvested at the end of the experiment. For the initial 16 days, all microcosms were exposed to a 4-day oxic (compressed medical grade air)/4-day anoxic (flushing with N2) pre-incubation period to allow soil respiration to stabilize. After the pre-incubation period, all microcosms were amended with 180 mg 13C-labeled ground ryegrass litter (97 atom%, Isolife, Wageningen, Netherlands; ~6% of the soil’s native C content) and incubated for 44 days. Microcosms were split into four redox treatments, managed via headspace manipulation: (1) static anoxic (N¬2 gas), (2) static oxic (medical air), (3) 4 days oxic/4 days anoxic (high frequency), (4) 8 days oxic/4 days anoxic (low frequency). Both high and low frequency treatments started and ended with oxic phases. Headspace samples were collected approximately every 4 days from the 20 trace gas microcosms (4 redox regimes × 5 replicates) to assess fluxes of CO2 and 13C-CO2 concentrations. For fluctuating treatments, microcosms were sampled immediately before the redox conditions were altered. To measure CO2 fluxes, microcosms were temporarily sealed for 2 hours, and gas samples were collected at the beginning and end of the sealed period by removing 30 ml of the headspace via septa into pre-evacuated 20 ml glass vials. Microcosms were sealed for 3 hours when CO2 production rates decreased towards the end of the experiment. CO2 concentrations were measured on a gas chromatograph (GC-14A, Shimadzu, Columbia, MD), equipped with a thermal conductivity detector. CO2 fluxes were determined by calculating the concentration difference during the sealed period, assuming a linear flux rate. An extra set of gas samples collected from each microcosm after the sealed period was analyzed for the 13C/12C CO2 isotope ratio with an isotope ratio mass spectrometer (IRMS; IsoPrime 100, Elementar, Hanau, Germany).

FTICR-MS data
Soil microcosms were destructively harvested at three timepoints during the experiment, on days 20 and 36 (n = 3 per treatment each day) and day 44 (n = 5 per treatment, the trace gas microcosms).
Samples exposed to an oxic headspace preceding the harvest were processed on the benchtop; those finishing an anoxic period were processed in an anoxic glove box (Coy Laboratory Products, Grass Lake, MI). 100 mg of soil (ODE) was shaken with water (2 ml) for 2 hours on an Eppendorf Thermomixer and centrifuged before collecting the supernatant. Water extracts were then desalted by solid phase extraction using Varian PPL cartridges according to Dittmar et al. (2008). A 12 T Bruker SolariX FTICR mass spectrometer was used to collect high resolution mass spectra of the organic compounds in the extracts. Refer to the GRE_13C_FTICR.R for instructions on processing the data in R.