Interdisciplinary and multidisciplinary approaches to the study of long‐term soil degradation: A case study from Schleswig‐Holstein,Germany

Soil degradation is a serious problem and an important environmental issue in many ecosystems. Without integrative, interdisciplinary and historical approaches, understanding the effects of long‐term soil degradation is difficult. According to this idea it is hypothesized that in order to study long‐term natural and human‐induced soil degradation, it is necessary to use interdisciplinary and multidisciplinary approaches with respect to temporal and spatial landscape changes. The results of the investigation of colluvial sediments and soils in research area in Schleswig‐Holstein (Germany) with a high resolution in space and time—using the four‐dimensional landscape analysis—indicated the temporal and spatial variation of soils and sediments from the Mesolithic until Modern times. Intensive soil degradation occurred as a result of the land clearance and agricultural land use in the investigation areas since the Neolithic time. The general results of this investigation show that the use of an interdisciplinary and multidisciplinary approach with pedological and geomorphological perspectives for different times and places can help to reconstruct the long‐term natural and human‐induced soil degradation. Copyright © 2009 John Wiley & Sons, Ltd.     

竹笋象危害对竹材物理力学性质的影响

对受笋万纹象(Otidognathus rubriceps Chevrolot)、一字竹笋象(O.davidis Fairmaire)、大竹笋象(Cyrtatrachelus longimanus Fabricius)危害后断梢毛竹、竹腔注药防治后毛竹、正常毛竹3种竹材进行竹材抗弯强度、抗弯弹性模量和顺纹抗拉强度测定结果表明,与正常毛竹相比较,受竹笋象危害断梢毛竹差异显著,3个竹材物理力学性质指标分别下降了43.79%、39.01%、29.28%,竹腔注药防治后毛竹3个指标下降了13.87%、10.95%、9.01%,但差异不显著,不影响竹材的使用价值.     

填饲对朗德鹅SCD5、SREBP2和ELOVL6甲基化和mRNA表达的影响

DNA甲基化是表观遗传学的研究热点之一.为了探索填饲是否能够引起朗德鹅Anser anser)肝脏脂类代谢中DNA甲基化水平以及基因表达量的变化和DNA甲基化修饰作用和基因表达之间的关系.本研究运用焦磷酸测序技术分析鹅肝脏硬脂酞辅酶A去饱和酶5基因(stearovl-CoA desaturase 5,SCD5)、固醇调节元件结合蛋白2基因(sterol regulatory element-binding protein 2,SREBP2)和超长链脂肪酸延伸酶6基因(elongase of very long chain fatty acids 6,ELOVL6)的甲基化水平,再运用qRT-PCR技术进行定量验证.结果显示,SCD5、SREBP2和ELOVL6各2个片段的甲基化水平均在20％以上,均属于高甲基化状态;对照组3个基因各2个片段的甲基化水平均高于填饲组,填饲组与对照组的SCD5-1S和SREBP2-1S片段甲基化差异显著(P＜0.05),SREBP2-2S片段甲基化差异极显著(P＜0.01),SCD5-2S、ELOVL6-1S和ELOLV6-2S片段甲基化差异不显著(P＞0.05);填饲组SCD5和ELOVL6的mRNA表达量均高于对照组,而SREBP2在填饲组的mRNA表达量显著低于对照组(P＜0.05).综合各项指标,填饲引起鹅肝脏组织SCD5、SREBP2和ELOVL6甲基化水平降低,SCD5和ELOVL6 mRNA表达量上升,SREBP2的mRNA表达量降低,表明SCD5和ELOVL6的甲基化修饰对mRNA的表达起反向调控的作用.本实验结果为表观遗传学上DNA甲基化研究方法提供参考,为选育鹅肥肝高产高质系提供重要的分子遗传学依据,也为人类肥胖症和脂肪肝等代谢疾病的研究提供理论依据.     

长期施肥下黄土旱塬黑垆土磷平衡及农学阈值

为研究长期施肥条件下土壤磷平衡和土壤有效磷对磷盈亏的响应,明确土壤有效磷(Olsen-P)的农学阈值及合理磷肥施用量,依托甘肃平凉肥料长期定位试验(始于1979),分析了黄土旱塬黑垆土36 a土壤磷盈亏动态、累积磷盈亏与有效磷的响应关系以及土壤磷残余,通过Mitscherlich方程模拟作物相对产量对土壤有效磷的响应关系,计算黄土旱塬黑垆土小麦和玉米的土壤有效磷农学阈值。结果表明:不施肥(CK)和单施氮肥(N)处理土壤磷始终亏缺,N处理每亏缺磷100 kg·hm~(-2),有效磷含量下降1.05 mg·kg~(-1);施磷处理土壤磷当季盈余4.3～207.9 kg·hm~(-2),累积盈余154.9～7 483.6 kg·hm~(-2),其中有机、无机配施(MNP)处理磷盈余最大;土壤累积磷盈余与土壤有效磷增量呈线性正相关,土壤中每盈余100 kg·hm~(-2)磷,秸秆还田配施化肥(SNP,磷肥隔年施)、氮磷配施(NP)、有机肥(M)和有机、无机配施(MNP)处理土壤有效磷分别增加7.55、2.47、0.28 mg·kg~(-1)和0.46 mg·kg~(-1);黄土旱塬黑垆土农田有效磷的小麦和玉米农学阈值分别为22.05 mg·kg~(-1)和13.96 mg·kg~(-1),MNP处理土壤有效磷含量已高于作物农学阈值,NP和SNP处理土壤有效磷含量达到小麦农学阈值分别需要21 a和24 a,达到玉米农学阈值分别需要2 a和8 a,M处理土壤有效磷含量已高于玉米农学阈值,还需要3 a可达到小麦农学阈值。当磷投入量每年平均达22.9 kg·hm~(-2)时,土壤磷呈持平状态;当磷用量达33 kg·hm~(-2)时,不仅作物产量较高,而且磷肥当季利用率也较高;当磷用量增加到233 kg·hm~(-2)时,作物产量对增加磷投入无响应,土壤磷残余超过90%,大量磷素累积在土壤中,增加了土壤磷素的流失风险。黄土旱塬小麦玉米一年一熟轮作黑垆土农田土壤有效磷农学阈值为13.96 mg·kg~(-1)(玉米)和22.05 mg·kg~(-1)(小麦),秸秆还田可促进旱地农田耕层土壤有效磷含量的增加。     

C_(18)键合磁珠固相萃取-液质联用检测蜂王浆中磺胺二甲嘧啶残留

[目的]建立一种应用C18键合磁珠固相萃取-液质联用检测蜂王浆中磺胺二甲嘧啶残留的方法。[方法]MCX固相萃取小柱联合C18磁珠对蜂王浆进行前处理,用HPLC-串联质谱仪对净化后样本进行检测。[结果]该方法的线性范围是5～80μg/kg,相关系数0.993 1。添加水平为5、10、20μg/kg时各种化合物的平均回收率范围为93%～104%,RSD值均不大于11.3%。[结论]该方法简单、快速,结合自动化设备,可实现蜂王浆样品中磺胺二甲嘧啶测定前处理的自动化。     

矮牵牛长柱头性状遗传分析

以闭花瓣型短柱头矮牵牛为母本,正常开放纯合型长柱头的矮牵牛作为父本,采用杂交、自交、回交方法,对不同世代的植株进行柱头长短调查统计分析,初步探讨了矮牵牛长柱头是由两对显性核基因控制的遗传,提出了转育长柱头闭花瓣型矮牵牛的遗传模式,以期获得长柱头闭花瓣型矮牵牛材料。     

Trends in grain yields and soil organic C in a long‐term fertilization experiment in the China Loess Plateau

Changes in grain yields and soil organic carbon (SOC) from a 26 y dryland fertilization trial in Pingliang, Gansu, China, were recorded. Cumulative C inputs from straw and root and manure for fertilizer treatments were estimated. Mean wheat (Triticum aestivum L.) yields for the 18 y ranged from 1.72 t ha^–1 for the unfertilized plots (CK) to 4.65 t ha^–1 for the plots that received manure (M) annually with inorganic N and P fertilizers (MNP). Corn (Zea mays L.) yields for the 6 y averaged 2.43 and 5.35 t ha^–1 in the same treatments. Yields declined with year except in the CK for wheat. Wheat yields for N only declined with time by 117.8 kg ha^–1 y^–1 that was the highest decrease among all treatments, and that for NP declined by 84.7 kg ha^–1 y^–1, similar to the declines of 77.4 kg ha^–1 y^–1 for the treatment receiving straw and N annually and P every second year (SNP). Likewise, the corn yields declined highly for all treatments, and the declined amounts ranged from 108 to 258 kg ha^–1 y^–1 which was much higher than in wheat. These declined yields were mostly linked to both gradual dry weather and nutrients depletion of the soil. The N only resulted in both P and K deficiency in the soil, and soil N and K negative balances in the NP and MNP were obvious. Soil organic carbon (SOC) in the 0–20 cm soil layer increased with time except in the CK and N treatments, in which SOC remained almost stable. In the MNP and M treatments, 24.7% and 24.0% of the amount of cumulative C input from organic sources remained in the soil as SOC, but 13.7% of the C input from straw and root in the SNP, suggesting manure is more effective in building soil C than straw. Across the 26 y cropping and fertilization, annual soil‐C sequestration rates ranged from 0.014 t C ha^–1 y^–1 for the CK to 0.372 t C ha^–1 y^–1 for the MNP. We found a strong linear relationship (R² = 0.74, p = 0.025) between SOC sequestration and cumulative C input, with C conversion–to–SOC rate of 16.9%, suggesting these dryland soils have not reached an upper limit of C sequestration.     

基于布谷鸟搜索算法优化LSTM的大坝变形预测

对大坝变形进行合理分析和准确预测是确保大坝安全运行的重要手段.大坝变形监测数据具有趋势性、周期性、随机性和非线性等特性,现有的机器学习模型大都基于大坝变形监测数据的非线性特点进行构建,而忽略了监测数据还具有趋势性和周期性的线性特征.提出了一种大坝变形预测模型,通过采用布谷鸟搜索算法(CS)对长短期记忆人工神经网络(LSTM)进行优化,再基于物联网传感器的实时监测数据,使用局部加权回归的周期趋势分解方法(STL)将数据分解成趋势分量、周期分量和余项分量,采用优化后的LSTM模型对趋势分量和余项分量分别进行预测,并通过简单周期估计方法进行计算,将3个分量的预测结果求和后得到最终变形预测结果.试验选取浙江利山水库开展变形预测研究,结果表明：STL-CS-LSTM模型的水平和沉降变形预测精度都高于其他模型,水平位移预测精度由高到低依次为LSTM模型、支持向量回归模型SVR和人工神经网络模型ANN,沉降预测精度由高到低依次为ANN模型、LSTM模型、SVR模型.