Microbial N2O source signatures and associated greenhouse gas fluxes from the Demo trial, Agroscope Reckenholz (Switzerland)

Abstract
This dataset contains N2O isotopic source signatures obtained from Keeling analysis of measured singly substituted isotopologues of N2O (δ15Nα, δ15Nβ, δ18O) along with calculated weekly averaged greenhouse gas fluxes of N2O, CH4 and CO2 at the Demo trial field site, located at the Agroscope Reckenholz Research Station in Zürich, Switzerland. The data was collected in a long-term crop rotation and fertilisation experiment that investigates the effects of nutrient availability on crop development and soil emissions. Fluxes were measured weekly using an automatic time integrating chamber (ATIC) system, with subsequent analysis of gas concentrations and isotopologues performed using high-precision commercial analysers (Picarro G5131-i (2x) and Picarro G2401 (1x)). The dataset provides high-resolution time series of gas fluxes and isotopic source signatures under different fertiliser treatments (NPK and null treatment), with a focus on sugar beet cultivation during the study period. These data are intended to support research on agricultural greenhouse gas emissions, soil microbial processes, and fertilisation management impacts.

keywords
N2O isotope source signatures, greenhouse gas fluxes (N2O), Agricultural soils

Core information: 
Station Metadata (Demo Trial – Agroscope Reckenholz, Zürich)
Name: Demo trial, Agroscope Reckenholz Research Station
Station ID: Demo Site
Country + country code: Switzerland (CH)
Latitude, longitude, elevation (m a.s.l.): 47°25’31” N, 8°30’59” E; 443 m a.s.l.
Climate zone: Köppen Cfb
Ecosystem type: Agricultural cropland (long-term crop rotation + fertilisation trial)
Mean annual air temperature/precipitation:
Temperature: 9.4 °C
Precipitation: 1,031 mm
Responsible institutes: Empa, Agroscope 
Contact persons:
Main scientist: Julius C. Havsteen (Postdoc, Empa): Julius.havsteen@empa.ch
Principal Investigator: Joachim Mohn (Group leader, Emissions and Isotopes, Empa): 
joachim.mohn@empa.ch
Station responsible: Christof Ammann (Group leader, Climate and Agriculture, Agroscope): christof.ammann@agroscope.admin.ch


Description of field site
The field component of this study was conducted at the Demo trial, which is located at the Agroscope-Reckenholz Research Station in Zürich, Switzerland (47°25’31” N, 8°30’59” E; 443 m asl). This long-term fertilisation experiment assesses the effect of nutrient deficiencies on the development of different summer and winter crops. Mean annual temperature at the site is 9.4°C, and mean annual precipitation is 1,031 mm. The topsoil is classified as an Eutric Cambisol (WRB), and it has a loam soil texture (20% clay, 33% silt, 47% sand), and an organic carbon content of 3%. A detailed description of the management practices at the Demo site can be found in (Frei et al., 2024). Briefly, the demonstration trial was established in 1989 on a managed meadow (~ 0.7 ha). The trial consists of a seven-year crop rotation and eight mineral and organic fertiliser treatments, or 56 plots (5 x 8 m) that are arranged in a non-replicated staggered-start design (Loughin, 2006) to allow independent comparisons between same-aged stands that are cultivated with different fertiliser treatments. Specifically in this study, we focused on mineral fertiliser (100% mineral N, P, and K; NPK treatment) and non-fertilized (0% fertilization; Null/Control treatment) treatments. The crop type focused on in this study was sugar beet (Beta vulgaris L.). The entire seven-year crop rotation includes spring wheat (Triticum aestivum L.), sugar beet (Beta vulgaris L.), silage maize (Zea mays L.), potato (Solanum tuberosum L.), winter barley (Hordeum vulgare L.), and two consecutive years of grass-clover-ley (main species: Trifolium pratense L., Trifolium repens L., Dactylis glomerata L., Festuca pratensis Huds., Lolium perenne L., Phleum pratense L.). The crops are grown in parallel rows, and the crop rotation is facilitated by a downward shift in the cultivated crop after each growing season. 

Description of data collection

Gas fluxes and microbial N2O isotopic source signatures were determined with weekly resolution using air samples from an automatic time integrating chamber (ATIC) system. In short, after each chamber closure, the ATIC system automatically collects sequential gas samples into impermeable 5 L gas bags (30228-U, Supel-Inert Multi-Layer Foil,  Sigma-Aldrich, USA) according to the approach outlined in Wang et al., 2022. For the presented study, the following timings were applied; each chamber was closed for 15 minutes every 4 hours. During each closure event, four consecutive headspace samples were collected for 15 seconds, with sampling for Bag1 beginning at 3.50 minutes after closure, followed by filling of the subsequent bags (Bag2, Bag3, Bag4) at 7.25, 11.50 and 14.25 minutes, respectively. 
Gas concentrations (N2O, CH4, CO2) and singly substituted isotopic composition of N2O (δ15Nα, δ15Nβ and δ18O) were analysed using commercial analysers (1x, G2401-m: CH4, CO2, CO and 2x, G5131-i N2O, δ15Nα, δ15Nβ and δ18O; Picarro Inc., USA, selected from three available analysers). The raw data were corrected for spectral interferences, instrumental drift, and calibrated against established scales provided by WMO GAW for greenhouse gases and Air-N2 (15N/14N) as well as VSMOW (18O/16O) for N2O isotopologues (Mohn et al., 2022). Full details on the correction and calibration approach are provided in Havsteen et al. 2025. 

Subsequent calculation of weekly mean fluxes of N2O, CH4 and CO2 per chamber using a linear regression approach for gas concentration over time, with an imposed quality criterion of R2 ≥ 0.7 for data acceptance. The concentration-to-flux conversion considered dynamic molar volumes, calculated from the mean ambient temperature over the 7 days measurement period, recorded at the Zürich-Affoltern weather station (REH; 47.4277° N, 8.5180° E; 444 m a.s.l.),  which is part of the automatic monitoring network operated by MeteoSwiss (WIGOS ID: 0-20000-0-06664).

The microbial N2O isotopic source signatures were derived using a two-endmember mixing model (Keeling plot) (Keeling, 1958), where each gas sample collected from a chamber represents a mixture of ambient atmospheric N2O and N2O from soil-derived microbial production. To determine the microbial source endmember for a specific time and chamber, isotopologue values were plotted against the reciprocal N2O concentrations of the respective bags. The intercept of the regression line provides an estimate of the microbial N2O source signatures for δ15Nα, δ15Nβ and δ18O. To assess the data quality, the regression-derived isotopic composition at ambient N2O concentrations was compared against actual values of a well-characterised compressed ambient‑air reference cylinder analysed both at Empa and Tokyo Institute of Technology (δ15Nα = 15.62 ‰, δ15Nβ = -3.07 ‰, δ18O = 43.0 ‰; ID#: 9951224, PanGAS). Three quality classes were defined: Quality 1 (Q1) for agreement within 4 ‰, Q2 (4–6 ‰), Q3 (6–8 ‰) and Q4 (>8 ‰). Regression lines with deviations in ambient isotopic signatures from actual values of more than 8 ‰ were excluded from further analysis. If a sample sequence was measured on two G5131-i analyzers, duplicate analysis provides two Keeling plot intercepts, which were averaged before further interpretation. To avoid false positive results under low flux scenarios, an additional criterion was applied, requiring a minimum N2O flux threshold of 75 µg N2O-N m^-2 h^-1 combined with a regression quality threshold of R2 ≥ 0.7. In the final data processing step, isotopic results from replicate chambers within each treatment type were combined using an uncertainty-weighted mean. This method gives more weight to chambers with higher analytical precision (lower uncertainty), ensuring that these measurements have a stronger influence on the treatment-level δ15Nbulk, and SP values. As a result, the reported δ15Nα and δ15Nβ intercepts, and thus δ15Nbulk and SP (computed from δ15Nα and δ15Nβ), may differ slightly from simple averages because they represent the precision‑weighted contribution of each chamber within a given treatment (NPK or control).

Output parameters in CSV file:

Date: the date chamber bags were collected in the field. Measurements represent the prior 7 days.

ISODate: the same as Date but in ISO format.

Round: sampling round, with Round 1 corresponding to the first measuring week.

Treatment: crop type and fertilization applied, e.g., sugar_beet_NPK.

Possible values include: N2O, CH4, CO2 (gas fluxes in µg/m2/h), d15Na, d15Nb, d18O (intercepts of the Keeling plot, proxies for N2O source signatures in ‰), d15Nbulk = (d15Na + d15Nb)/2 in ‰, SP = d15Na − d15Nb in ‰, rain (7-day accumulated rainfall in mm), and temperature (7-day average temperature in °C).

Analyzers: the instrument used. N2O and isotopic data were obtained using three Picarro G5131-i analyzers from KIT (K), Thünen Institute (T), and Empa (E). If a sample was measured on multiple analyzers, the value is averaged. CH4 and CO2 were obtained on a Picarro G2401-m analyzer. Agroscope provides rain and temperature data.

AvgValue: averaged measured value for a given parameter across analyzers if multiple measurements exist.

Uncertainty: uncertainty of the measurement, if available.

AvgQuality: quality of the averaged data, graded from Q1 (best) to Q4 (worst). Only Q1 to Q3 are included. Q1 indicates the Keeling line intersects the ambient isotopic end-member within 2‰, Q2 within 4‰, Q3 within 6‰, and Q4 above 8‰ (excluded).

AvgNumDataPoints: number of data points used to calculate the averaged value. This may be missing for some measurements.


Funding: 
This work is part of the project 21GRD10 quantiAGREMI, which has received funding from the European Partnership on Metrology, co-financed from the European Union’s Horizon Europe Research and Innovation Programme and by the Participating States. In addition, the project was financed by the European Union’s Horizon Europe Research and Innovation programme under HORIZON-CL5-2022-D1-02 Grant Agreement No 101081430 – PARIS. The Empa contribution has received funding from the Swiss State Secretariat for Education, Research and Innovation (SERI). 


References

Frei, J., Wiesenberg, G. L., & Hirte, J. (2024). The impact of climate and potassium nutrition on crop yields: Insights from a 30-year Swiss long-term fertilization experiment. Agriculture, Ecosystems & Environment, 372, 109100. https://doi.org/10.1016/j.agee.2024.109100

Havsteen, J. C., & Fatima, M., with Brunamonti, S., Pogány, A., Hausmaninger, T., Wolf, B., Well, R., & Mohn, J. (2025). Correction and calibration protocol for isotope data via CRDS: A study case for N₂O and other isotope systems. Atmospheric Measurement Techniques, https://doi.org/10.5194/egusphere-2025-4954 

Keeling, C. D. (1958). The concentration and isotopic abundances of atmospheric carbon dioxide in rural areas. Geochimica et cosmochimica acta, 13(4), 322-334. 

Loughin, T. M. (2006). Improved experimental design and analysis for long‐term experiments. Crop Science, 46(6), 2492-2502. https://doi.org/10.2135/cropsci2006.04.0271


Mohn, J., Biasi, C., Bodé, S., Boeckx, P., Brewer, P. J., Eggleston, S., Geilmann, H., Guillevic, M., Kaiser, J., & Kantnerová, K. (2022). Isotopically characterised N2O reference materials for use as community standards. Rapid Communications in Mass Spectrometry, 36(13), e9296. https://doi.org/10.1002/rcm.9296
 
Wang, Y., Paul, S. M., Jocher, M., Alewell, C., & Leifeld, J. (2022). Reduced Nitrous Oxide Emissions From Drained Temperate Agricultural Peatland After Coverage With Mineral Soil. Frontiers in Environmental Science, 10. https://doi.org/10.3389/fenvs.2022.856599 


