Nitrous oxide
Mean soil N2O fluxes (µg N2O-N
m-2 h-1) for the reference, buffer,
and no buffer sites for the entire sampling period (June to September)
were 3.6 ± 2.6, 2.1 ± 2.4, and 1.7 ± 1.6, respectively (Figure 4C).
According to the LME model, there was a treatment effect on
N2O fluxes, with significantly (R - NB: p = 0.02, Table
2) higher fluxes at the no buffer sites compared to the reference sites.
Over the sampling period, N2O fluxes were, on average
1.70 times lower at the buffer sites and 2.05 times lower at the no
buffer sites, compared to the reference sites, respectively.
Additionally, mean soil N2O fluxes were the highest in
the spring, and gradually declined throughout the summer to the fall,
with the highest monthly mean measured in June (4.1 ± 2.9 µg
N2O m-2 h-1) and
lowest monthly mean in September (1.6 ± 2.2 µg N2O
m-2 h-1). N2O fluxes
were not significantly different between local groundwater conditions.
On average, over the study period N2O efflux was 2.5 ±
2.2 at the DIS areas and 2.4 ±2.5 µg N2O
m-2 h-1at the ND areas. None of the
environmental variables (i.e. soil moisture, soil temperature, and depth
to the groundwater table) were significant predictors of
N2O fluxes.
Table 2. Pairwise difference between Treatment levels (reference, R; buffer, B; and no buffer, NB) using Tukey’s HSD post hoc test for the linear mixed effects models explaining the dynamics of carbon dioxide, methane, and nitrous oxide fluxes, respectively. All models included the autocorrelation term “AR1(Week + 0 | Site/Chamber)”. Bolded comparison indicates a significant effect at p < 0.05.
Model | Comparison | Est. | SE | p |
CO2 ~ Treatment | R - B | -8.65 | 11.2 | 0.72 |
R - NB | -1.86 | 11.4 | 0.99 |
B - NB | 6.79 | 11.4 | 0.82 |
CH4 ~ Treatment | R - B | -6.65 | 6.82 | 0.59 |
R - NB | -22.35 | 6.96 | < 0.01 |
B - NB | -15.71 | 6.97 | 0.07 |
N2O ~ Treatment | R - B | 0.33 | 0.14 | 0.06 |
R - NB | 0.39 | 0.14 | 0.02 |
B - NB | 0.06 | 0.14 | 0.90 |