A comparison of chamber-based methods for measuring N2O emissions from arable soils
M. N. Kong, Farhana Ferdous Mitu, Søren O. Petersen, Poul Erik Lærke, Diego Ábalos, Peter Sørensen, Andreas Brændholt, Sander Bruun, Jørgen Eriksen, Christian Dold
Abstract
• N 2 O fluxes measured with non-flow-through chamber (NFT) methods. • N 2 O fluxes measured with flow-through chamber (FT) methods. • Water vapor correction improved agreement between NFT and FT estimations. • NFT measured 7–84 % lower N 2 O emissions than FT in various conditions. • Differences in N 2 O emissions arose from chamber design and headspace mixing. Static chamber-based flux measurements with gas chromatography are commonly used to estimate nitrous oxide (N 2 O) emissions from arable soils. The LI-COR 7820 N 2 O/H 2 O (LI-7820) enables higher-frequency in situ measurements, but side-by-side comparisons with traditional methods are limited. To address this gap, we compared non-steady-state chamber methods including non-flow-through (NFT) and flow-through (FT) chamber methods under field and laboratory conditions with plant cover or bare soil. The LI-7820 was used with the LI-8200S smart chamber (FT-1: ⌀ 20 cm) and a self-built chamber (FT-2: 60 × 60 cm), and compared to differently sized NFTs (1–4: 75 × 75, 27 × 37, 60 × 60, and ⌀ 20 cm) with manual sampling with gas chromatography. Field experiments showed high RMSE for daily N 2 O fluxes within 20 days after fertilizer application between FT-1 and NFTs, particularly for maize and spring barley (183 and 214 μg N 2 O-N m −2 h −1 ), which dropped sharply after 20 days (47 and 54 μg N 2 O-N m −2 h −1 , respectively). FT-2 and NFT-3 for pastures had lower RMSE and MAE, both below 40 μg N 2 O-N m −2 h −1 . In the incubation experiment, bare soil showed smaller error values, remaining below 26 μg N 2 O-N m −2 h −1 . Significant differences were observed between the cumulative N 2 O emissions measured with NFTs and FT-1, while differences were not significant between NFT-3 and FT-2. Several factors may explain these differences. The smaller chamber dimensions of FT-1 may influence water and nitrogen distribution and constrain the capture of spatial heterogeneity, while NFTs could be affected by prolonged deployment times and in-chamber pressure changes. Furthermore, the lack of water-vapor correction in NFTs, unlike the LI-7820, contributed to discrepancies between methods. Understanding these nuances including the impact of the chamber design, is essential for enhancing the comparability of N 2 O emissions and getting closer to achieving unbiased measurements of the true flux.