Microbial induced nitrous oxide emissions from an arable soil during winter

Nitrous oxide (N(2)O) release rates were measured from an fertilized and unfertilized plot on silty loam (Gleyic Luvisol) cropped with winter wheat. Rates were estimated using a closed soil cover box technique throughout a continuous investigation period of 12 months. The 12 months of investigation were separated into the cropping period (March to November) and the winter period (December to February). Soil management and all N-applications were made during the cropping period. The application of 220 kg N to the soil induced significantly higher N(2)O losses throughout the cropping season compared to the unfertilized soil. No significant differences were found during winter, where 70% of the annual N(2)O emissions were found. The temporal changes of the N(2)O emission rates on both soils were highly correlated (r = 0.96; P less than or equal to 0.001), and could be attributed to temporal changes in soil temperature (r = 0.65; P less than or equal to 0.01) resulting from freezing and thawing cycles. In order to decide whether the N(2)O production can be attributed to microbial or non-microbial processes in soil, the time courses of the N(2)O emissions from a gamma-ray sterilized and a non-sterilized soil were compared in a laboratory experiment, where the freezing and thawing cycles were simulated according to field conditions. The results indicated, that microbial processes were responsible for N(2)O production in thawing and even Frozen soils. (C) 1998 Elsevier Science Ltd. All rights reserved.