Spatial variability and distribution of N2O emissions from a tea field during the dry season in subtropical central China

The recent boom and intensification of the tea industry in subtropical central China, with its large fertiliser inputs, lead us to observe nitrous oxide (N2O) emissions from tea fields to investigate the emission characteristics and mitigation potential. In this study, we examined the spatial variability of N2O emissions observed on 30 October 2010 using 147 static mini chambers gridded almost regularly in a 4.8-ha fertilised tea field. The N2O fluxes for 0–30 min (10:00–10:30 am) ranged from − 6.42 to 79.56 g N ha− 1 d− 1 with an average value of 5.88 g N ha− 1 d− 1 and were positively skewed, thus approximating a log-normal distribution. The geostatistical analysis indicated that the normalised log-transformed N2O emissions exhibited strong spatial autocorrelation, characterised by a Matérn-type semivariogram model and an effective range of 28.5 m. As observed during the dry season, only the elevation was found to be significantly correlated with N2O emissions (r = − 0.42, P < 0.001); none of the other measured soil properties had a significant relationship with N2O emissions (r < 0.10, P > 0.05). Three spatial interpolation methods (ordinary kriging, regression kriging and cokriging) were applied to estimate the spatial distribution of N2O emissions over the study area. Regression kriging with the elevation as an auxiliary predictor and cokriging with the inverse of the normalised elevation and the normalised soil organic carbon as two co-variables slightly outperformed both cokriging with individual normalised environmental factors as the co-variable and ordinary kriging. However, all methods predicted similar total amounts of N2O emissions in the tea field, ranging from 25.6 to 26.8 g N d− 1.

► Tea-field N2O emissions in dry season showed a significant correlation with elevation.
► N2O emissions exhibited strong spatial dependence with an effective range of 28.5 m.
► RK with elevation offered more accurate spatial interpolation of N2O emissions.