济南北跨地区浅层地下水水化学多元统计分析

根据浅层地下水的水流特征和人类活动影响,将研究区进行了水文地质单元分区,应用因子分析法对各分区浅层地下水水化学特征进行研究。结果表明:黄河沿岸(Ⅰ区)浅层孔隙水形成过程中经历了蒸发浓缩作用、混合作用和溶滤作用,以蒸发浓缩作用为主;径流条件差的引黄灌溉区(Ⅱ区),浅层孔隙水形成过程中经历了混合作用和蒸发浓缩作用,以混合作用为主;径流条件较好的地下水开采区(Ⅲ区),浅层孔隙水形成过程中经历了蒸发浓缩作用和溶滤作用,以蒸发浓缩作用为主。     

淮河中上游流域洪水主要类型及其时空分布特征

洪水过程相似性挖掘对流域雨洪资源利用、水库防洪调度以及河流生态修复等具有重要的现实意义。论文以淮河中上游流域16个水文站2006—2015年125场洪水过程数据为基础,采用洪水量、时间、变化率和形态等特征指标对完整洪水过程进行了全面刻画,以主成分分析和层次聚类等多元统计分析手段辨识了流域内代表性洪水类型,揭示了各洪水类型的时空分布特征。结果表明：① 淮河中上游流域主要有5类代表性洪水类型,分别为长历时且变化剧烈型、多峰长历时型、尖瘦短历时型、矮胖型以及常规型。② 从时间分布来看,2006—2015年间洪水类型数量呈现减少趋势,常规型洪水的比例逐渐增大;丰水年份(如2007年)、平水年份(如2006年)的洪水类型较多,而枯水年份(2011—2013年)洪水类型较少,以常规型和矮胖型洪水出现频率居多。③ 从空间分布来看,源头站点的洪水类型较多,中下游站点的洪水类型比较单一,主要从尖瘦型洪水过程逐渐转变为矮胖型,这与流域内水源涵养能力、工程调蓄能力以及降水多样性等有较大的关系。研究可为流域洪水信息挖掘和特征分析等提供参考借鉴,也可为淮河流域洪水的演变特征分析、水库防洪调控和雨洪资源利用等提供决策依据。     

翡翠贻贝三个野生种群形态性状的差异分析

运用差异系数检验、主成分分析、判别分析和聚类分析等4种多元统计分析方法,采用5个比例性状,比较研究了广西北海、广东湛江和汕尾的3个翡翠贻贝野生种群的形态性状差异。结果表明:(1)3个种群间5个比例性状的差异系数在0.024～1.134之间,种群间尚未达到亚种差异水平;(2)主成分分析构建了3个主成分,PC1、PC2和PC3贡献率分别为40.134%、27.956%和20.614%,累计贡献率为88.704%;(3)利用逐步判别法选择3个贡献率较大的性状进行判别分析,建立了3种群判别函数,判别准确率为P1为73.0%～88.0%,P2为74.6%～86.8%,综合判别率为79.7%;(4)聚类分析显示,北海种群与汕尾种群形态较为接近,湛江种群趋异程度最大。各种分析方法从不同角度反映了群体间的形态学差异,这种差异是环境和遗传共同作用的结果。     

����DEM�Ŀ�˹�������ò���͵�ʶ��

????????ASTER???????RS??GIS???????DEM???????DEM?????TM????CBERS????????????29???????????????????ò?????????5????????????24?????????Spss??Matlab?????????????????????????????????????????ò????????????????????????????????????飬?ó????????Ч????     

Spatial intertidal distribution of bivalves and polychaetes in relation to environmental conditions in the Natori River estuary,Japan

This paper aims to reveal spatial variation in the abundance of infaunal bivalves and polychaetes at different spatial scales (station: 200–800 m intervals; plot: 5–20 m), and to reveal environmental variables affecting the spatial distribution of animals in the Natori River estuary, Japan. We found six bivalve species and eight polychaete species from 52 plots at 12 stations. Nuttallia olivacea and Heteromastus sp. were found to be the most abundant species of bivalves and polychaetes, respectively. Assemblage patterns of bivalves and polychaetes were classified into five distinct groups. Substrata (silt-clay contents), salinity, and relative elevation were the variables found to affect the infaunal assemblage patterns. Chlorophyll a was not a significant variable, but benthic animals were absent at sites with extremely low chlorophyll a conditions. Macrobenthic assemblage patterns were different not only between stations but often differed between plots at the same station, reflecting the complex assemblage structure of benthic invertebrates. Detailing such animal–environment relationship is essential in understanding the potential food supply for estuarine fishes.     

Assessment of heavy metal pollution in estuarine surface sediments of Tangxi River in Chaohu Lake Basin

A total of 30 surface sediments samples from the estuary wetland of the Tangxi River, Chaohu Lake Basin were obtained and tested. Enrichment factor (EF) and geoaccumulation index (I_(geo)) as well as multivariate statistical analysis methods including Factor Analysis (FA) and Hierarchical Cluster Analysis (HCA) were applied for the assessment of heavy metal pollution in surface sediments. The results of EF values show that the pollution of copper (Cu) and cadmium (Cd) occurs in the estuarine sediments, and that zinc (Zn), lead (Pb) and chrome (Cr) may originate from crustal materials or natural weathering process. The mean EF values of the five heavy metals are in the decreasing order: Cu>Cd>Zn>Pb>Cr. Based on the I_(geo) of target heavy metals, the surface sediments collected from the study area can be approximately categorized as unpolluted with Zn, Pb and Cr, and moderately polluted with Cu and Cd. The degree of heavy metal pollution decreases in the order of Cu>Cd>Zn>Pb>Cr. Three groups of pollution factors are presented from FA: Zn-TOC, Cu-Cd and Cr-Pb, which respectively accounte for 27.22%, 25.20% and 21.05% of variance. By means of HCA, a total number of seven groups are distinguished from 30 sampling sites. Results indicate that Cu and Cd are the prior controlled pollutants in the estuarine sediments of the Tangxi River.     

A New Model of Peaks over Threshold for Multivariate Extremes

The peaks over threshold（POT） methods are used for the univariate and multivariate extreme value analyses of the wave and wind records collected from a hydrometric station in the South China Sea. A new multivariate POT method： Multivariate GPD（MGPD） model is proposed, which can be built easily according to developed parametric models and is a natural distribution of multivariate POT methods. A joint threshold selection approach is used in the MGPD model well. Finally, sensitivity analyses are carried out to calculate the return values of the base shear, and two declustering schemes are compared in this study.     

Comparison of event‐specific rainfall–runoff responses and their controls in contrasting geographic areas

Numerous studies have examined the event‐specific hydrologic response of hillslopes and catchments to rainfall. Knowledge gaps, however, remain regarding the relative influence of different meteorological factors on hydrologic response, the predictability of hydrologic response from site characteristics, or even the best metrics to use to effectively capture the temporal variability of hydrologic response. This study aimed to address those knowledge gaps by focusing on 21 sites with contrasting climate, topography, geology, soil properties, and land cover. High‐frequency rainfall and discharge records were analysed, resulting in the delineation of over 1,600 rainfall–runoff events, which were described using a suite of hydrologic response metrics and meteorological factors. Univariate and multivariate statistical techniques were then applied to synthesize the information conveyed by the computed metrics and factors, notably measures of central tendency and variability, variation partitioning, partial correlations, and principal component analysis. Results showed that some response magnitude metrics generally reported in the literature (e.g., runoff ratio and area‐normalized peak discharge) did not vary significantly among sites. The temporal variability in site‐specific hydrologic response was often attributable to the joint influence of storage‐driven (e.g., total event rainfall and antecedent precipitation) and intensity‐driven (e.g., rainfall intensity and antecedent potential evapotranspiration) meteorological factors. Mean annual temperature and potential evapotranspiration at a given site appeared to be good predictors of hydrologic response timing (e.g., response lag and lag to peak). Response timing metrics, particularly those associated with response initiation, were also identified as the metrics most critical for capturing intrasite response variability. This study therefore contributes to the growing knowledge on event‐specific hydrologic response by highlighting the importance of response timing metrics and intensity‐driven meteorological factors, which are infrequently discussed in the literature. As few correlations were found between physiographic variables and response metrics, more data‐driven studies are recommended to further our understanding of landscape–hydrology interactions.