基于BP人工神经网络的城市PM2.5浓度空间预测

针对PM2.5日均质量浓度,采用BP人工神经网络模型,预测研究区空气中PM2.5浓度的空间变异,通过与普通克里格(Ordinary Kriging)插值方法对比验证BP人工神经网络预测模型的精度.结果表明:BP人工神经网络预测模型下研究区检验样本点位置的PM2.5仿真浓度与观测浓度之间的均方差、平均绝对误差、平均相对偏差和相关系数分别为0.296 μg2/m6、0.412 μg/m3、1.650％和0.851;而与此同时,普通克里格插值方法下的对应结果分别为1.041 μg2/m6、0.689 μg/m3、11.910％、0.638.研究成果在肯定BP人工神经网络预测模型可用于揭示PM2.5浓度空间变异特征的同时,也证实了其相对于普通克里格插值方法在固定空间点位准确预测PM2.5浓度方面的优势.     

AA3流动注射仪快速测定地表水中的总氰化物

在反复实验的基础上建立了运用Bran+Luebbe AutoAnalyzer 3流动注射分析仪快速测定地表水中总氰化物的分析方法。结果表明,该方法与传统的分光光度法相比,具有自动进样、分析速度快、测定成本低等优点。与经典的分析方法相对照,结果无显著性差异,能够满足大批量地表水监测的要求。在测试系统中引入了恒温槽,进一步提高了测试的精密度和准确度。     

The Dual Model for Tourism Resource Productivity Improvement

For natural resource science, resource productivity studying is an important subject. But researches on tourism resource productivity are limited. The most significant influencing factor on tourism resource productivity is the potential market scale of tourist locations, and second most important influencing factor is the resource endowment. Regional urban population is significantly correlated with region tourist numbers, being the decisive factor of region potential tour market scale. In tourism development, the dual model should be adopted: on one hand to enhance tour spot attractiveness, on the other hand to cultivate the potential market by improving urbanization level and other means. In the situation of tourism development fever spreading, the dual model for improving tourism productivity helps to avoid the "Great Leap Forward" which means that too rapid tourism construction divorces from actual market demand.     

一种灾害等级的区间数评估模型

由于灾害损失本身的复杂性及灾情统计方面存在的问题,利用区间数替代单一实数值作为灾害等级评估指标的输入更加符合实际。基于上述思路,提出了一种基于区间数的灾害等级评估新模型。首先,给出区间数距离的新定义,该定义能够充分利用区间数所携带的信息;然后,利用线性加权评价函数计算评估对象与评价标准之间的综合相对距离测度,评估对象隶属于综合距离测度最小值所对应的灾害等级。该模型建模过程思路清晰,物理意义明确,计算过程简捷,评估结论符合实际。最后,用实例说明模型的应用方法。     

The fragmentation patterns of maternal plasma cell-free DNA and its applications in non-invasive prenatal testing

The discovery of cell-free DNA (cfDNA) in maternal plasma has opened up new promises for the development of non-invasive prenatal testing (NIPT). Application of cfDNA in NIPT of fetus diseases and abnormalities is restricted by the low amount of fetal DNA molecules in maternal plasma. Fetus-derived cfDNA in maternal plasma are shorter than maternal DNA, thus leveraging the maternal and fetus-derived cfDNA molecules size difference has become a novel and more accurate method for NIPT. However, multiple biological properties such as size distribution of plasma DNA, proportion of fetal-derived DNA and methylation levels in maternal plasma across different gestational ages still remain largely unknown. Further insights into the size distribution and fragmentation pattern of circulating plasma cfDNA will shed light on the origin and fragmentation mechanisms of cfDNA during physiological and pathological processes in prenatal diseases and enhance our ability to take the advantage of plasma cfDNA as a molecular diagnostic tool. In the review, we start by summarizing the research techniques for the determination of the fragmentation profiles of cfDNA in maternal plasma. We then summarize the main progress and findings in size profiles of maternal plasma cfDNA and cffDNA. Finally, we discuss the potential diagnostic applications of plasma cfDNA size profiling.     

江苏省及南京市工业危险废物现状研究与分析

通过对全国、江苏省以及南京市工业危险废物的产生量和排放量进行调查、总结和对比分析,指出江苏省作为一个工业大省、强省,每年工业危险废物的产生量接近全国的10%,且呈逐年下降趋势.而排放量所占比例大于10%,且出现了先降低、再增加、又降低到零的现象,最高时达到16.67%.造成这一现象的原因可能跟工业危险废物的管理有关.同时指出了目前江苏省乃至南京市目前存在的问题以及相应的对策措施.     

基于污泥堆肥产物与余泥渣土制备的人工土壤的理化性质分析与评价

为实现污水厂污泥及余泥渣土的协同处理和资源化利用,以“两步法”制备人工土壤,考察不同配比条件对人工土壤理化性质的影响。结果表明：污水污泥与3种不同辅料 (玉米秸秆、菌菇渣、木屑) 配比后进行好氧堆肥,堆肥结束后腐熟度符合国家相关标准要求,将污泥堆肥产物和余泥渣土制备得到人工土壤,其理化性质与天然土壤类似。堆肥添加量能显著影响人工土壤水稳性团聚体、pH值、EC、有机质、总氮及有效磷含量,在人工土壤种植应用时,污泥+玉米秸秆的堆肥种植效果最好,采用改进的内罗梅综合指数法进行综合评价,综合考虑,污泥+玉米秸秆堆肥产物添加量为30%、余泥渣土添加量为70%和污泥+玉米秸秆堆肥产物添加量为50%、余泥渣土添加量为50%是最优质的人工土壤,污泥+木屑堆肥次之,污泥+菌菇渣堆肥效果最差。     

Variations of bacterial community structure in flooded paddy soil contaminated with herbicide quinclorac

The denaturing gradient gel electrophoresis (DGGE) method was applied to determine the relative genetic complexity of microbial communities in flooded paddy soil treated with herbicide quinclorac (3,7-dichloro-8-quinoline-carboylic acid). The results obtained showed a significant effect of quinclorac on the development of bacterial populations in soils contaminated with different concentrations of the herbicide at the early time after application. In general, however, the number of populations of the same soil sample treated with the same concentration of the quinclorac differed obviously with increasing incubation time within the early 8 weeks. The scale of differences in banding patterns-showed that the microbial community structures of the quinclorac-treated and non-quinclorac-treated soils were not significantly different after 21 weeks of incubation. Quantification, as demonstrated in this paper, was studied by establishing dose-response relationships. Significant pattern variations were quantified. Prominent DGGE bands were excised, cloned and sequenced to gain insight into the identities of predominant bacterial populations. The majority of DGGE band sequences were related to bacterial genera Clostridium, Sphingobacterium, Xanthomonas and Rhodococcus.