禹城地区土壤铅含量空间分布的指示克里格估值

以鲁西北禹城地区土壤重金属Pb含量为例,针对其存在特异值与偏态分布的特点,利用指示克里格法研究了土壤Pb含量小于特定阈值的条件概率分布,分析了该地区土壤Pb含量的空间分布特征,并对普通克里格和指示克里格的估值效果进行交叉验证。结果表明：研究区土壤Pb含量总体较低,未超过国家标准的限值,但不同部位土壤Pb含量差异较大;普通克里格法和指示克里格法形成的土壤Pb含量空间分布格局相似,但普通克里格法的平滑效果强于指示克里格法;研究区域土壤环境质量总体良好,Pb含量高值区主要集中在各乡镇驻地附近的农田;指示克里格对土壤Pb含量空间估值的精度优于普通克里格,指示克里格法更适用于原始数据呈非正态分布且特异值对估值结果有较大影响的土壤属性。本研究结果为该地区土壤环境质量评价和重金属污染防控提供科学依据。     

基于GIS的上海城市灰尘重金属空间分布特征研究

基于ArcGIS地统计分析模块,以上海市外环以内为研究区域,对上海城市道路和公园灰尘中重金属Pb、Cr、Cu、Ni、Zn含量水平和空间分布特征进行了研究.研究表明:上海城市灰尘重金属平均含量水平普遍较高,Pb、Cr、Cu、Ni、Zn平均含量分别为273.45、144.01、190.01、86.26、708.25 mg·kg-1,分别高出上海土壤背景值10.7、1.9、6.6、2.7、8.2倍;外环区域Pb、Zn、Ni污染严重,尤以工业区、商业区以及交通繁忙区最突出.地统计分析表明,Pb和Cu呈较弱的空间相关性,说明Pb和Cu的空间变异以人为影响为主;Zn、Cr和Ni为中等相关,说明随机性因素和结构因素对这3种元素都有较大影响;基于ArcGIS的地统计学分析工具能够较好地反映重金属污染的空间分布格局.     

利用因子克立格方法研究天津土壤PAHs和土壤理化参数空间多尺度相关性

土壤中的PAHs对人体健康具有潜在的危害 .在天津采样数据的基础上 ,应用多元地统计学中的因子克立格方法 ,探讨了天津地区表层土壤中PAHs含量和一些土壤性质之间的空间相关性 .在天津地区共采集 188个表层土壤样品 ,测定了 16种PAHs的总含量、土壤 pH值、总有机碳含量和土壤粘粒含量 .研究结果表明 ,天津地区表层土壤中PAHs含量和pH、TOC、粘粒含量之间的空间相关性在不同尺度上有很大差异 .     

基于贝叶斯公式的地下水污染源及含水层参数同步反演

针对非均质地下含水层污染源识别及含水层参数反演过程中监测方案优化问题,提出一种基于贝叶斯公式及信息熵最小的累进加井的多井监测方案优化方法.首先,构建假想案例下的二维非均质各向同性潜水含水层水流及溶质运移模型,运用GMS软件进行数值模拟求解.采用最优拉丁超立方抽样方法和Kriging法建立数值模拟模型的替代模型.然后以参数后验分布的信息熵最小为目标函数,采用累进加井的方式进行多井监测方案优化设计.最后根据优化后的监测方案,采用差分进化自适应Metropolis算法进行污染源及含水层参数的同步反演.算例研究表明：在兼顾反演精度及监测成本,并保证每个参数分区内至少有1眼监测井的条件下,5眼井组合监测方案（6,5,1,2,8）为最优监测方案.与信息熵最小的10眼井组合监测方案（1,2,3,4,5,6,7,8,9,10）的参数反演结果相比,5眼井组合监测方案对11个参数α=（X_S,Y_S,T₁,T₂,Q_s,K₁,K₂,K₃,DL₁,DL₂,DL₃）的后验均值偏离率的平均值虽增大1.2%,但监测成本却是10眼井组合监测方案的50%.     

Hydrochemical and multivariate statistical interpretations of spatial controls of nitrate concentrations in a shallow alluvial aquifer around oxbow lakes (Osong area, central Korea)

Hydrochemical and multivariate statistical interpretations of 16 physicochemical parameters of 45 groundwater samples from a riverside alluvial aquifer underneath an agricultural area in Osong, central Korea, were performed in this study to understand the spatial controls of nitrate concentrations in terms of biogeochemical processes occurring near oxbow lakes within a fluvial plain. Nitrate concentrations in groundwater showed a large variability from 0.1 to 190.6 mg/L (mean = 35.0 mg/L) with significantly lower values near oxbow lakes. The evaluation of hydrochemical data indicated that the groundwater chemistry (especially, degree of nitrate contamination) is mainly controlled by two competing processes: 1) agricultural contamination and 2) redox processes. In addition, results of factorial kriging, consisting of two steps (i.e., co-regionalization and factor analysis), reliably showed a spatial control of the concentrations of nitrate and other redox-sensitive species; in particular, significant denitrification was observed restrictedly near oxbow lakes. The results of this study indicate that sub-oxic conditions in an alluvial groundwater system are developed geologically and geochemically in and near oxbow lakes, which can effectively enhance the natural attenuation of nitrate before the groundwater discharges to nearby streams. This study also demonstrates the usefulness of multivariate statistical analysis in groundwater study as a supplementary tool for interpretation of complex hydrochemical data sets.     

Application of indicator kriging to the complementary use of bioindicators at three trophic levels

The use of biological indicators is widespread in environmental monitoring, although it has long been recognised that each bioindicator is generally associated with a range of potential limitations and shortcomings. To circumvent this problem, this study adopted the complementary use of bioindicators representing different trophic levels and providing different type of information, in an innovative approach to integrate knowledge and to estimate the overall health state of ecosystems. The approach is illustrated using mercury contamination in primary producers (mosses), primary consumers (domestic pigeons and red-legged partridges) and top predators (Bonelli's eagles) in southern Portugal. Indicator kriging geostatistics was used to identify the areas where mercury concentration was higher than the median for each species, and to produce an index that combines mercury contamination across trophic levels. Spatial patterns of mercury contamination were consistent across species. The combined index provided a new level of information useful in incorporating measures of overall environmental contamination into pollution studies.     

Modelling abiotic indicators when obtaining spatial predictions of species richness

When conservation biologists formulate strategy used in decisions concerning the locations of new national parks, refuges and reserves, accurate information about species richness and spatialpatterns of species distributions can be critical. Recent research has demonstrated that spatial models and bioindicator taxa can be quite useful for determining generalized spatial patterns of unrelated taxa on a continental scale. In this research, I incorporate abiotic effects, in this case altitudinal relief, into both the mean and the covariance structures of the spatial prediction model. I use bird species data collected in the Indian subcontinent and cross-validation techniques to illustrate the degree of improvement in prediction accuracy engendered by using theabiotic factor and the modified spatial models.     

Kriging and Bilinear Methods for Estimating Spatial Pattern of Atmospheric Pollutants

Two statistical methods to reconstruct the spatial pattern ofenvironmental data are compared: the Kriging and the Bilinear.The comparison is performed by evaluating two differentcharacteristics of the methods: the interpolation and the estimation. The Kriging method is demonstrated to be better thanthe Bilinear regarding all the statistics studied, although thedifferences are not very significant. Especially, the Kriging ismore efficient than the Bilinear in the reconstruction of peakvalues and it does not extrapolate local trends. The methods areapplied to a set of environmental data measured in an urban area.     

城-郊区域土壤铅(Pb)累积过程时空模拟:以中部某大城市为例

城-郊区域类型众多的高强度人类活动导致土壤Pb累积过程复杂、时空异质性高,使得揭示该类区域土壤Pb累积过程的时空变化特征十分困难.以国内中部某大城市典型城-郊区域为研究区,构建土壤Pb累积过程单元模型,基于土地利用分类和模拟结果,建立土壤Pb累积过程时空模拟方法,模拟研究区2013~2040年土壤Pb累积含量,阐明土壤Pb含量未来时空变化特征.结果表明,研究区2013年Pb含量平均值约为其所在省份表层Pb含量背景值的1.77倍,土壤Pb污染较重.2013~2040年Pb含量将持续增加,增加量较低（0.53~2.25 mg·kg^-1）的区域位于研究区西部、北部和南部,占总面积的25.46 %,增加量较高（3.98~5.70 mg·kg^-1）的区域位于其东部,占总面积的17.14 %.研究区东部林地面积的增加、水域和草地面积的减少导致了该区域土壤Pb含量较大幅度地上升;此外,土壤Pb含量的空间分布还同重要工厂和交通设施的分布高度相关.通过突破以往研究中将土地利用视为静态不变的局限,能够在一定程度上反映区域土地利用变化对重金属累积过程的影响,可为城-郊区域土壤Pb累积过程模拟提供方法,并为该城市城-郊区域土壤Pb污染管控提供依据.     

COMPARATIVE STUDY OF DROUGHT PREDICTION TECHNIQUES FOR RESERVOIR OPERATION1

ABSTRACT: Predicting the likelihood of a drought markedly enhances the efficiency of reservoir operations. This study applies the kriging method and time series analysis to predict inflows to Shihmen Reservoir in northern Taiwan. A subsequent reservoir operation simulation is employed to determine the drought lead time (DLT), the time before the onset of a drought. A more efficient reservoir operational strategy can be established with the aid of DLT and the probability of successful drought prediction (P s). Simulation results of reservoir operation over a period of three decades demonstrate that, at one month DLT, the kriging approach achieves 0.86 of P s for moderate droughts and 0.94 of P s for severe droughts. The kriging approach generally outperformed the time series approach in terms of DLT, P s of drought prediction, and the number of correctly predicted drought events.