Methane emissions from stems of Fraxinus mandshurica var. japonica trees in a floodplain forest

We measured methane (CH₄) emissions from the stem surfaces of mature Fraxinus mandshurica var. japonica trees in a floodplain forest. Flux measurements were conducted almost monthly from May to October 2005, and positive CH₄ fluxes were detected throughout the study period, including the leafless season. The mean CH₄ flux was 176 and 97 μg CH₄ m⁻² h⁻¹ at the lower (15 cm above the ground) and upper (70 cm above the ground) stem positions, respectively. The CH₄ concentration was lower in soil gas than in ambient air to a depth of at least 40 cm. One possible source of CH₄ emitted from the stems might be the dissolved CH₄ in groundwater; maximum concentrations were 10,000 times higher than atmospheric CH₄ concentrations. Our results suggest that CH₄ transport from the submerged soil layer to the atmosphere may occur through internal air spaces in tree bodies.     

The effect of growing soybean (Glycine max. L.) on N2O emission from soil

A pot experiment was conducted to investigate the effect of growing soybean on N₂O emission from soil. When soybean was growing in pots, the cumulative N₂O emission during the growing season was 2.26 mg N pot⁻¹, which was 5.9 times greater than that from the identical but unplanted pots (CK). However, the difference in N₂O fluxes between the two treatments was not significant until the grain-filling stage. Of the total N₂O emission, 94% took place during the period from grain-filling to ripening. Premature harvesting of the aerial parts of the plants at various growth stages substantially stimulated N₂O emission from the soil. These results implied that the process of symbiotic N fixation per se does not stimulate N₂O production or emission, but rather senescence and decomposition of the roots and nodules in the late growth stage. Therefore, additional N₂O would be emitted from the soil after harvesting of soybean with roots, litter, and residues left in situ.     

黄土区土壤CO2释放及其影响因素研究

研究表明黄土区土壤CO2释放具有一定特殊性。从当日清晨至次日晨土壤CO2释放量呈由高至低再变高的规律,其变化趋势大体与温度变化一致,但时间上有一定滞后性。土壤CO2释放量有明显季节变化,夏季日释放量最高,秋季次之,冬季最低。不同覆被土壤CO2释放量存在差异,裸地释放量较高。CO2释放量对土质变化敏感,致密土壤则释放量小。     

影响静态箱检测开放式气体排放源N2O排放通量的关键因子

为研究影响静态箱检测开放式气体排放源氧化亚氮(N2O)排放通量的关键因子,以提高静态箱检测气体排放通量的准确性,该文在实验室条件下,探究了箱体配置(有无通气孔、有无风扇)和检测条件(不同密闭时间:30、40、50和60 min;不同排放源表面风速:0、0.5、1.0、1.5和2.0 m/s)对300 mm(直径)×300 mm(高度)(D300 mm×H300 mm)的静态箱检测N2O排放通量准确性的影响规律。结果表明,不同配置的静态箱测量结果偏差率随时间的变化趋势均相同,其中有通气孔和风扇的箱体在不同风速下的检测稳定性较好,检测准确性最高。当排放源表面风速为0~2 m/s时,风扇对静态箱检测准确性无显著性影响,排放源表面的风主要通过通气孔影响静态箱的检测准确性。静态箱检测的气体排放通量与实际排放通量的偏差率随排放源表面风速和箱体密闭时间的增加而显著降低。该试验推荐在排放源表面风速小于2 m/s的无粪便堆积的奶牛运动场以及排放源介质相似的开放式气体排放系统中使用有通气孔和风扇的静态箱对N2O排放通量进行检测,密闭50 min。     

改善农学管理措施减少太湖稻麦轮作NH3和NO排放

以稻麦轮作为对象,采用密闭室-通气法和静态箱-化学发光法对太湖地区空白（不施氮肥）、当地常规（农民习惯管理方式）和保产增效（与当地常规相比,改善作物种植、水分和养分管理方式）3个处理下的氨（NH3）挥发和一氧化氮（NO）排放进行田间原位观测,研究保产增效措施对稻麦轮作NH3 挥发和NO排放的综合影响。结果表明,与当地常规相比,保产增效在施氮量减少25%的情况下,总作物产量没有降低,且农学利用率提高了39%;保产增效NH3挥发量在稻季和麦季分别减少了26%和44%;保产增效在稻季NO排放量较低（0.75±0.03 kg N/hm2）,与空白和当地常规均没有显著差异（p<0.05）,在麦季NO排放量则显著降低了49%。因此,保产增效措施不仅能保障高产和提高氮利用率,还能减少NH3和NO排放,值得在太湖地区推广。     

基于深层神经网络的猪声音分类

猪的声音能够反映生猪的应激状态以及健康状况,同时声音信号也是最容易通过非接触方式采集到的生物特征之一。深层神经网络在图像分类研究中显示了巨大优势。谱图作为一种可视化声音时频特征显示方式,结合深层神经网络分类模型,可以提高声音信号分类的精度。现场采集不同状态的猪只声音,研究适用于深层神经网络结构的最优谱图生成方法,构建了猪只声音谱图的数据集,利用Mobile Net V2网络对3种状态猪只声音进行分类识别。通过分析对比不同谱图参数以及网络宽度因子和分辨率因子,得出适用于猪只声音分类的最优模型。识别精度方面,通过与支持向量机,随机森林,梯度提升决策树、极端随机树4种模型进行对比,验证了算法的有效性,异常声音分类识别精度达到97.3%。该研究表明,猪只的异常发声与其异常行为相关,因此,对猪只的声音进行识别有助于对其进行行为监测,对建设现代化猪场具有重要意思。     

土壤呼吸对生物炭添加的短期响应：发酵后处理和无机氮的影响

The biodegradability of chars derived from pyrolysis and hydrothermal carbonisation (HTC) was studied in short-term dynamic incubation experiments under controlled conditions. Carbon dioxide C (CO$_{2}$) emissions from soil-char mixtures in combination with solid digestate or mineral nitrogen (N) fertiliser were measured in dynamic chambers for 10 d.~Compared to the original material (maize straw), pyrolysis and HTC chars showed significantly lower CO$_{2}$ emissions and slower decay dynamics; and compared to the soil control, HTC char increased soil respiration to a significant extent, while pyrolysis char did not. The addition of mineral N resulted in a delayed respiration dynamics for HTC char, while the addition of digestate resulted in an increase in the respired CO$_{2}$ for pyrolysis char and a decrease for HTC char. For the first time, a peculiar two-stage decay kinetics was observed for HTC char, indicating a highly inhomogeneous substrate consisting at least of two C pools.     

Nitrification controls N2O production rates in a frozen boreal forest soil

The winter season has been identified as a significant contributor to N₂O emissions from boreal soils, but our understanding of the processes regulating these emissions is fragmentary. We investigated potential N-sources and pathways involved in N₂O formation in a frozen boreal forest soil by labeling soil samples with ¹⁵N-containing substrates, and measured rates of ¹⁵N₂O/¹⁵N₂ formation under both oxic and anoxic conditions. Our results showed that all N₂O produced in the frozen samples originate from denitrification, but the rate-limiting factor is NO₃⁻ availability, which is largely governed by nitrification. This suggests that N₂O formation in frozen boreal soils may be sustained for a prolonged period of time, but is governed by a delicate balance of the O₂ regime.     

Dinitrogen and N2O emissions in arable soils: Effect of tillage, N source and soil moisture

A laboratory investigation was performed to compare the fluxes of dinitrogen (N₂), N₂O and carbon dioxide (CO₂) from no-till (NT) and conventional till (CT) soils under the same water, mineral nitrogen and temperature status. Intact soil cores (0-10 cm) were incubated for 2 weeks at 25 °C at either 75% or 60% water-filled pore space (WFPS) with ¹⁵N-labeled fertilizers (100 mg N kg⁻¹ soil). Gas and soil samples were collected at 1-4 day intervals during the incubation period. The N₂O and CO₂ fluxes were measured by a gas chromatography (GC) system while total N₂ and N₂O losses and their ¹⁵N mole fractions in the soil mineral N pool were determined by a mass spectrometer. The daily accumulative fluxes of N₂ and N₂O were significantly affected by tillage, N source and soil moisture. We observed higher (P<0.05) fluxes of N₂+N₂O, N₂O and CO₂ from the NT soils than from the CT soils. Compared with the addition of nitrate (NO₃⁻), the addition of ammonium (NH₄⁺) enhanced the emissions of these N and C gases in the CT and NT soils, but the effect of NH₄⁺ on the N₂ and/or N₂O fluxes was evident only at 60% WFPS, indicating that nitrification and subsequent denitrification contributed largely to the gaseous N losses and N₂O emission under the lower moisture condition. Total and fertilizer-induced emissions of N₂ and/or N₂O were higher (P<0.05) at 75% WFPS than with 60% WFPS, while CO₂ fluxes were not influenced by the two moisture levels. These laboratory results indicate that there is greater potential for N₂O loss from NT soils than CT soils. Avoiding wet soil conditions (>60% WFPS) and applying a NO₃⁻ form of N fertilizer would reduce potential N₂O emissions from arable soils.