Three-year measurements of nitrous oxide emissions from cotton and wheat-maize rotational cropping systems

- Three-year N2O fluxes are reported for cotton and wheat-maize cropping systems.
- The fertilizer rate is not always an effective indicator of annual N2O emissions.
- Leaching significantly contributes to fertilizer losses in irrigated croplands.
- N2O emission inventory should consider leaching effects.
- Optimized sampling protocols enhance the reliability of discrete flux measurements.

The remarkable expansion of fertilization and irrigation may stimulate nitrous oxide (N2O) emissions from cropping systems in northern China. High-resolution measurements were conducted in irrigated cotton and wheat-maize rotational systems in Shanxi Province, P.R. China, between 2007 and 2010 (three year-round crop cycles, hereinafter referred to as Y1, Y2 and Y3) to investigate the impacts of natural inter-annual variations and agricultural management on annual N2O emissions and direct emission factors (EFs). Overall, N2O emissions fluctuated diurnally, seasonally and inter-annually in the fertilized treatments. The hourly N2O fluxes closely followed the daily air temperature patterns. The daily mean fluxes corresponded to these hourly fluxes, which were observed between 09:00-10:00 and 19:00-20:00. An optimized sampling protocol could improve the reliability of discrete measurements when estimating cumulative emissions. The N2O emissions for the fertilized treatments were 2.7 ± 0.2 (Y1) and 1.6 ± 0.1 kg N ha−1 yr−1 (Y2) from the cotton field and 6.2 ± 0.4 (Y1), 4.5 ± 0.3 (Y2) and 4.5 ± 0.2 kg N ha−1 yr−1 (Y3) from the wheat-maize field. Peak N2O emissions after fertilization and irrigation/rainfall lasted one to three weeks and accounted for 16-55% of the annual emissions. Leaching losses were estimated at 10.4 ± 3.0 (Y1) and 12.5 ± 3.4 kg N ha−1 yr−1 (Y2), which accounted for 16-17% of the fertilizer-N applied to the cotton field. Annual N2O emissions did not increase with increasing fertilization rates or water inputs because significant amounts of fertilizer-N were lost through leaching. Background emissions amounted to one-third to one-half of the total N2O emissions from the fertilized treatments. The direct EFs were 2.2 ± 0.3% (Y1) and 0.9 ± 0.2% (Y2) in the cotton field and 1.3 ± 0.2% (Y1), 0.8 ± 0.1% (Y2) and 0.7 ± 0.1% (Y3) in the wheat-maize field. The large inter-annual variations in N2O emissions and direct EFs emphasize the importance of multiple-year continuous observations.