华东地区地震活动性参数的中短期异常动态变化特征

系统地研究了华东地区中强震前地震活动参数的演化特征 ,尤其是追踪研究了中短期阶段各参数的动态变化特征。结果表明 ,华东地区中强震孕育中短期阶段 ,有不少地震活动性参数图像出现明显的变化 ,如C值、D值、Mf 值和b值在中期阶段存在的大面积异常区开始出现明显的收缩甚至消失 ,这种变化是地震孕育过程进入中短期阶段的重要标志。未来主震往往发生在收缩区边缘或异常消失区 ,若当消失区再次出现小片异常 ,则该异常区是未来最为值得注意的危险区 ,而且时间上很可能进入短临。再如地震算法复杂性AC值由超出 1倍均方差的峰值下降、广义分维Dt由正常值下降并出现超过 1倍均方差的低值异常 ,也是地震孕育过程进入中短期阶段的判据。由于地震孕育是一个十分复杂的过程 ,在实际工作中需应用多种参数扫描结果进行综合判定 ,以便于缩小异常区 ,提高中短期预报的准确率。     

哈尔滨市短临天气预报研究

从短临预报的定义和内容出发，就短临预报业务系统流程、地面要素加密信息收集显示、多普勒雷达预警信息和预报指标、重大灾害性短临预报的气象要素信息／预警信息／预报指标及物理量、不稳定指数预警信息／预报指标及合并显示、卫星云图降水／冰雹估计、3h降水／温度预报方法、预报编辑系统设计等方面，阐述了哈尔滨市短临预报业务的建设过程及其开展的方式、方法。     

基于人体健康风险的水污染事件遗传性致癌物安全浓度研究

确定水污染事件中遗传性致癌物短期暴露安全浓度,是目前我国水环境管理中急需解决的问题.在假设遗传性致癌物暴露剂量与致癌风险呈线性相关的前提下,提出了水体中遗传性致癌物安全浓度计算流程,建立了采用遗传性致癌物终生暴露安全浓度计算短期暴露安全浓度的方法.在我国2000～2010年间发生的60起水污染事故中,特征污染物为砷的水污染事件为6起,占总污染事件比例为10%,这一结果表明砷为我国水污染事件中高发的特征污染物.根据水污染事件遗传性致癌物短期暴露安全浓度计算方法,得出砷短期暴露安全浓度为0.5 mg.L-1,表明该方法在水污染事件应急管理中具有一定的可行性.     

基于CNN_LSTM模型的复杂支路故障电弧检测*

在低压交流配电系统中,当多支路并联的复杂系统的某1支路中出现串联电弧故障时,识别难度大幅提升。为了预防此类情况引发的电气火灾,提出1种卷积神经网络(CNN)与长短时记忆网络(LSTM)结合的串联故障电弧检测方法。首先,搭建实验平台用以采集不同负载在不同支路下发生故障时和正常工作时的干路电流数据;然后,构建CNN_LSTM模型并做出相应改进,将电流数据直接输入到模型中,由模型自主提取波形特征并进行分类。研究结果表明:该方法可以快速、准确地识别出电弧故障,准确率达99.04%以上,且能够较为准确地检测出是哪类负载所在的支路发生电弧故障,准确率达97.90%,可为复杂支路下的电弧故障识别研究提供参考。     

生活垃圾填埋短期好氧预处理工艺优化研究

以模拟有机生活垃圾为底物,通过实验研究了通风和温度协同控制作用下,填埋短期好氧预处理过程垃圾VS、含水率和有机组分的代谢情况.结果表明,好氧预处理过程VS降解主要发生在反应前期(0~6d),占实验全过程降解量的68%~85%;同时垃圾含水量的15%~26%以渗滤液和气体的形式被去除,并且延长通风时间能够增加后者所占比例.反应前期堆体自发升温到高温状态(50~55℃)比控制堆体温度维持在中温状态(42~47℃)更利于碳水化合物的代谢,但对粗蛋白降解的影响不明显.而在中温状态下粗脂肪和木质纤维素的降解率更高.继续延长预处理时间至14d,不同通风和温度的组合对VS和水分总体去除差距不大.控制填埋好氧预处理周期在6d并尽量维持堆体温度在50~55℃,成本-效益较高.     

Returning to normalcy in the short term: a preliminary examination of recovery from Hurricane Harvey among individuals with home damage

This study focuses on coastal counties in Texas, United States, affected by Hurricane Harvey in 2017 to gauge the influence of individual and contextual characteristics on people's ability to return to normalcy in the short term. Data from a survey conducted by the Kaiser Family Foundation and the Episcopal Health Foundation between October and November 2017 were utilised in the analysis. The paper observes, based on the results of an ordered logistic regression, and contrary to previous work, that age, gender, levels of poverty, and social capital are not significant predictors of a return to normalcy. However, indicators such as whether a person evacuated, if he/she identified as Hispanic/Latino, the extent of damage sustained to one's home, and if one's automobile was damaged or destroyed are shown to affect recovery. A discussion of the potential reasons for these findings is provided as a means of informing future research on disaster recovery.     

The effect of pesticides and metals on acetylcholinesterase (AChE) in various tissues of blue mussel (Mytilus trossulus L.) in short-term in vivo exposures at different temperatures

The purpose of the study was to assess the impact of short-term exposure to selected toxicants as well as metal accumulation upon acetylcholinesterase (AChE) in the blue mussel, Mytilus trossulus L., in laboratory in vivo experiments. Mussels were exposed for up to 48 hours to a mixture of copper (Cu^{2 +}, 400 μ g L^?1) and cadmium (Cd^{2 +}, 200 μ g L^?1), to dichlorvos (DDVP, 100 μ g L^?1), and to carbaryl (100 μ g L^?1) at two temperatures: 5°C and 20°C. Samples were collected after 0, 6, 12, 24, and 48 hours of exposure, and AChE activity and metal concentration (where applicable) were analysed in gills, digestive gland, mantle+muscles, and the whole soft tissue. Very strong AChE inhibition was observed in response to the dichlorvos treatment, mainly in gills. Carbaryl and the metals caused a short-term inhibition effect. Considerable differences in AChE activity between the two temperatures were noticed. In particular, the metals were accumulated much faster at 20°C than at 5°C, especially in gills. No correlation between AChE activity and metal concentration was found. Gills turned out to be the optimal tissue for AChE activity analysis in short-term studies.     

Differential responses of short-term soil respiration dynamics to the experimental addition of nitrogen and water in the temperate semi-arid steppe of Inner Mongolia, China

We examined the effects of simulated rainfall and increasing N supply of different levels on CO2 pulse emission from typical Inner Mongolian steppe soil using the static opaque chamber technique, respectively in a dry June and a rainy August. The treatments included NH4NO3 additions at rates of 0, 5, 10, and 20 g N/（m2.year） with or without water. Immediately after the experimental simulated rainfall events, the CO2 effluxes in the watering plots without N addition （WCK） increased greatly and reached the maximum value at 2 hr. However, the efflux level reverted to the background level within 48 hr. The cumulative CO2 effluxes in the soil ranged from 5.60 to 6.49 g C/m2 over 48 hr after a single water application, thus showing an increase of approximately 148.64% and 48.36% in the efftuxes during both observation periods. By contrast, the addition of different N levels without water addition did not result in a significant change in soil respiration in the short term. Two-way ANOVA showed that the effects of the interaction between water and N addition were insignificant in short-term soil COz efftuxes in the soil. The cumulative soil CO2 fluxes of different treatments over 48 hr accounted for approximately 5.34% to 6.91% and 2.36% to 2.93% of annual C emission in both experimental periods. These results stress the need for improving the sampling frequency after rainfall in future studies to ensure more accurate evaluation of the grassland C emission contribution.     

基于复合LSTM模型的PM2.5浓度预测

在对淄博市19个空气质量监测站点监测数据进行分析后,提出了一种基于机器学习的复合模型——灰色关联度分析(GRA)-改进的完备总体经验模态分解(ICEEMD)-长短期记忆网络(LSTM)模型。通过分析淄博市2019年大气污染物和气象数据,选用LSTM模型预测PM_2.5浓度。由于传统单一模块机器学习模型具有训练时间较长和预测精度较低的问题,提出了复合LSTM模型。该模型由3部分组成：GRA,用于PM_2.5浓度影响因素变量筛选;ICEEMD,用于PM_2.5分解、分量筛选和原始大气污染物及气象数据处理;LSTM,用于PM_2.5浓度预测。预测结果表明：淄博市中部丘陵地带PM_2.5浓度高于南部山区和北部平原,东部高于西部;淄博市逐月PM_2.5浓度呈“U”形分布,1月最高,8月最低;淄博市PM_2.5浓度受PM₁₀和CO影响较大,受湿度和温度影响较小。对比单一LSTM模型和GRA-LSTM模型,GRA-ICEEMD-LSTM模型...     

济南市大气颗粒物短期暴露对老年人群血清淀粉样蛋白P组分影响的定群研究

为了解济南市不同粒径大气颗粒物暴露对老年人群血清淀粉样蛋白P组分（SAP）的急性影响,评估大气颗粒物暴露对人群神经系统的健康风险,以山东省济南市甸柳社区为研究地点,采用定群研究设计,对社区76名健康老年人进行5次重复测量,结合社区附近（< 2 km）环保监测超级站的PM_2.5和PM₁₀浓度数据,在控制年龄、性别等协变量情况下,利用线性混合效应模型分别分析PM_2.5和PM₁₀暴露对SAP的效应影响.结果表明:大气颗粒物短期暴露与老年人群SAP的升高呈正相关;PM_2.5累积滞后0~6 h,每升高1个四分位数间距（IQR）,SAP水平升高18.73%（95% CI为9.20%~29.08%）（FDR_H-P < 0.05）（其中,CI为可信区间,FDR_H-P为经多重校正后的P值）;PM₁₀累积滞后0~6 h,每升高1个IQR,SAP水平亦升高,但变化不显著（FDR_H-P>0.05）.研究显示,大气颗粒物暴露可以引起人群反应神经退行性病变的SAP指标升高,提示大气颗粒物对人群神经系统具有潜在威胁;PM_2.5对SAP的效应高于PM₁₀,提示小粒径颗粒物具有较高的健康危害,应加强对小粒径颗粒物的健康影响和干预防护研究.