Livestock Systems Research Department, Teagasc, Moorepark, Fermoy, County Cork, Ireland.
PLoS One. 2011;6(10):e26176. doi: 10.1371/journal.pone.0026176. Epub 2011 Oct 10.
There is uncertainty about the potential reduction of soil nitrous oxide (N(2)O) emission when fertilizer nitrogen (FN) is partially or completely replaced by biological N fixation (BNF) in temperate grassland. The objectives of this study were to 1) investigate the changes in N(2)O emissions when BNF is used to replace FN in permanent grassland, and 2) evaluate the applicability of the process-based model DNDC to simulate N(2)O emissions from Irish grasslands. Three grazing treatments were: (i) ryegrass (Lolium perenne) grasslands receiving 226 kg FN ha(-1) yr(-1) (GG+FN), (ii) ryegrass/white clover (Trifolium repens) grasslands receiving 58 kg FN ha(-1) yr(-1) (GWC+FN) applied in spring, and (iii) ryegrass/white clover grasslands receiving no FN (GWC-FN). Two background treatments, un-grazed swards with ryegrass only (G-B) or ryegrass/white clover (WC-B), did not receive slurry or FN and the herbage was harvested by mowing. There was no significant difference in annual N(2)O emissions between G-B (2.38±0.12 kg N ha(-1) yr(-1) (mean±SE)) and WC-B (2.45±0.85 kg N ha(-1) yr(-1)), indicating that N(2)O emission due to BNF itself and clover residual decomposition from permanent ryegrass/clover grassland was negligible. N(2)O emissions were 7.82±1.67, 6.35±1.14 and 6.54±1.70 kg N ha(-1) yr(-1), respectively, from GG+FN, GWC+FN and GWC-FN. N(2)O fluxes simulated by DNDC agreed well with the measured values with significant correlation between simulated and measured daily fluxes for the three grazing treatments, but the simulation did not agree very well for the background treatments. DNDC overestimated annual emission by 61% for GG+FN, and underestimated by 45% for GWC-FN, but simulated very well for GWC+FN. Both the measured and simulated results supported that there was a clear reduction of N(2)O emissions when FN was replaced by BNF.
当温带草地中的肥料氮（FN）部分或完全被生物固氮（BNF）取代时，土壤一氧化二氮（N2O）排放的潜在减少存在不确定性。本研究的目的是：1）研究 BNF 替代永久草地中的 FN 时 N2O 排放的变化，2）评估基于过程的模型 DNDC 对模拟爱尔兰草地 N2O 排放的适用性。三种放牧处理为：（i）每年接受 226kg FN ha-1 yr-1 的黑麦草（Lolium perenne）草地（GG+FN），（ii）每年接受 58kg FN ha-1 yr-1 的黑麦草/白三叶草（Trifolium repens）草地（GWC+FN），在春季施用，（iii）不接受 FN 的黑麦草/白三叶草草地（GWC-FN）。两个背景处理，仅接受黑麦草的未放牧草地（G-B）或黑麦草/白三叶草（WC-B），没有接受粪肥或 FN，牧草通过刈割收获。G-B（2.38±0.12kg N ha-1 yr-1（平均值±SE））和 WC-B（2.45±0.85kg N ha-1 yr-1）之间的年 N2O 排放量没有显著差异，表明由于 BNF 本身和永久黑麦草/白三叶草草地中白三叶草残留分解而导致的 N2O 排放可以忽略不计。GG+FN、GWC+FN 和 GWC-FN 分别产生 7.82±1.67、6.35±1.14 和 6.54±1.70kg N ha-1 yr-1。DNDC 模拟的 N2O 通量与测量值吻合良好，三种放牧处理的模拟和测量日通量之间存在显著相关性，但背景处理的模拟结果并不非常吻合。DNDC 高估 GG+FN 的年排放量 61%，低估 GWC-FN 的年排放量 45%，但对 GWC+FN 的模拟非常准确。测量和模拟结果均表明，FN 被 BNF 替代时，N2O 排放明显减少。
