Background nitrous oxide emissions in agricultural and natural lands: a meta-analysis

Aim

This study aimed at better characterising background nitrous oxide (N₂O) emissions (BNE) in agricultural and natural lands.

Methods

We compiled and analysed field-measured data for annual background N₂O emission in agricultural (BNEA) and natural (BNEN) lands from 600 and 307 independent experimental studies, respectively.

Results

There were no significant differences between BNEA (median: 0.70 & mean: 1.52 kg N₂O???N ha^?1 yr^?1) and BNEN (median:0.31 & mean:1.75 kg N₂O???N ha^?1 yr^?1) (P?>?0.05). A simultaneous comparison across all BNEA and BNEN indicated that BNEs from riparian, vegetable crop fields and intentional fallow areas were significantly higher than from boreal forests (P?<?0.05). Correlation and regression analyses supported the underlying associations of soil organic carbon (C), nitrogen (N), pH, bulk density (BD),and/or air temperature (AT) with BNEs to a varying degree as a function of land-use or ecosystem type (Ps?<?0.05).

Conclusions

Although overall BNEN tended to be lower than BNEA on median basis, results in general suggest that land-use shifts between natural and managed production systems would not result in consistent changes in BNE.