Semi-parametric statistical approaches for space-time process prediction

The problem of estimation and prediction of a spatial-temporal stochastic process, observed at regular times and irregularly in space, is considered. A mixed formulation involving a non- parametric component, accounting for a deterministic trend and the effect of exogenous variables, and a parametric component representing the purely spatio-temporal random variation is proposed. Correspondingly, a two-step procedure, first addressing the estimation of the non- parametric component, and then the estimation of the parametric component is developed from the residual series obtained, with spatial-temporal prediction being performed in terms of suitable spatial interpolation of the temporal variation structure. The proposed model formula-tion, together with the estimation and prediction procedure, are applied using a Gaussian ARMA structure for temporal modelling to space-time forecasting from real data of air pollution concentration levels in the region surrounding a power station in northwest Spain.     

Comparison of Point Estimation Techniques in the Spatial Analysis of Radium-226, Radium-228 and Potassium-40 in Soil

Many environmental surveys require the implementation of estimation techniques to determine the spatial distribution of the variable being investigated. Traditional methods of interpolation and estimation, for example, inverse distance squared and triangulation often ignore features of the data set such as anisotropy which may have a significant impact on the quality of the estimates produced. Geostatistical techniques may offer an improved method of estimation by modelling the spatial continuity of the variable using semi-variogram analysis. The theoretical model fitted to the semi-variogram is then used in the assignation of weighting factors to the samples surrounding the location to be estimated. This paper outlines the results of a comparison between three common estimation methods, polygonal, triangulation and inverse distance squared and a geostatistical method, in the estimation of soil radionuclide activities. The geostatistical estimation method known as kriging performed best over a range of parameters used to test the performance of the methods. Kriging exhibited the best correlation between actual and estimated values, the narrowest error distribution and the lowest overall estimation error. Polygonal estimation was best at reproducing the data set distribution. Conditional bias was evident in all the methods, low values being over-estimated and high values being under-estimated.     

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.     

绿色食品产地生态环境质量分析与评价--以庆安县国家级生态农业示范区为例

采用1993年、2000年和2003年庆安县生态环境质量监测数据,通过指数评价和加权综合评价法,分析评价庆安县绿色食品产地的生态环境质量,并利用GIS地理统计等功能,用变异估计法及克里格插值法,以可视化的方式展现绿色食品产地环境质量的空间分布.结果表明,庆安县中部和西南部的环境质量最好,最适宜发展AA级绿色食品基地; 西北部和东南部的绿色食品产地环境质量虽然是清洁、安全的,但有逐渐恶化趋势,需引起进一步关注.     

城市表层土壤重金属污染分析研究

根据某城市土壤表层中的8种重金属Cu、Pb、Cd、Hg、As、Cr、Ni、Zn的浓度数据,利用地统计学原理方法,研究了不同功能区土壤重金属的污染特征。结果表明Cd、Cu、Hg、Pb和Zn在工业区和主干道路区浓度超标,并且这几种元素的富集系数偏高,说明污染来源主要是工业生产、汽车尾气排放及汽车轮胎磨损等。此研究对城市土壤重金属污染治理及修复、重金属污染重点行业污染控制等提供参考。     

土壤连续属性空间插值方法及其精度的研究进展

作为土壤变化的时空定量监测方法,土壤属性空间插值方法及其精度是计量土壤学和"数字土壤"领域的重要研究内容。文章首先介绍了土壤属性空间插值的常用方法,包括克立格插值法(Kriging)、反距离权重法(IDW)、样条插值法(Spline)、贝叶斯最大熵(BME)、地理加权回归(GWR)以及高精度曲面建模方法(HASM);其次阐述了土壤属性空间插值精度验证的方法和指标;再次总结了能够提高土壤属性插值精度的6种途径,包括合理选择插值方法、准确设定插值方法参数、合理设计采样数目和密度、注意空间自相关程度和范围的影响、科学安排实验分析顺序以及结合辅助信息进行插值;最后从插值方法的选择、验证指标的选取以及辅助信息的结合三个方面指出了土壤属性空间插值方法及其精度的未来研究方向。     

Study on Simulating Spatial Distribution and Varying Patterns of Urumqi＇s Population

In this paper, an approach to simulate the spatial distribution of urban population is proposed using urban land use and population statistical data through the geographical information systems （GIS）. Then, the spatial population distribution of Urumqi as a case is simulated by the approach mentioned above and its varying patterns are analyzed by the raster population surface. As a result, producing raster population surface is more accurate and natural than the traditional choropleth map of population density. Concerning the spatial population distribution of Urumqi, the population density declines from south to north and the population distribution mainly presents＇ ＂T-type＂, the population distribution presents multicentre agglomeration and the population distribution of the districts shows different features. The population density varies significantly with the increase in the distance from central business district （CBD）. Finally, it is found in this paper that the development history of districts, terrain and traffic road are main factors that have an influence Urumqi＇s population distribution. This paper tries to provide more accurate population data for the plan and management of urban land, traffic and public facilities in order to enrich the researches on urban population distribution.     

基于GIS的海洋底栖生物栖息密度空间插值方法

在地理信息系统即GIS的支持下,分别采用反距离加权(inverse distance weighted,IDW)、普通克里格(ordinary krig-ing,OK)、规则样条(regularize spline,RS)和张力样条(tension spline,TS)4种插值方法对2006年7月获得的大连湾底栖生物栖息密度的数据进行空间插值处理,并对插值结果的精确度进行交叉验证,分析和比较不同插值方法获得的分布图。结果表明,插值精确度普通克里格>反距离加权>张力样条>规则样条;4种方法均能较客观的模拟出底栖生物栖息密度的分布趋势,但是在整体趋势和局部趋势两方面的综合考虑下,普通克里格的表现效果更好。文章进一步指出,在确定站位数量及分布前提下,插值结果的精确度可以通过选择空间插值方法得以改善,但其根本还是取决于站位布置的数量和其分布合理性。     

A dual-sensor-based method to recognize pipeline leakage and interference signals

During the detection of pipeline leakages, false alarms of leak detection could be markedly reduced if the interference signals resulting from pressure regulating, pump regulating or valve movements could be accurately distinguished. A digital recognition method for interference signals and leakage signals based on a dual-sensor system is proposed in this paper. It is demonstrated that the direction of the signal can be recognized by a cross-correlation calculation between two signals from the dual-sensor, one of which undergoes forward linear interpolation and backward linear interpolation. Based on this theory, the interference signal and the leak signal can be discriminated exactly, and the distance between the two sensors in the dual-sensor system can be considerably reduced without needing to increase the sampling frequency. The monotonicity of the cross-correlation function is demonstrated, and a fast discrimination algorithm based on a binary extreme search method, which decreases the computational load and maintains global optimization, is also proposed. A pre-processing method of the actual signal is proposed to decrease the identity requirement for the two sensors in a dual-sensor system. In the experiment based on artificial signals, the proposed discrimination algorithm could achieve accurate recognition of the abnormal signal, and as such, the theory and application of pipeline leak detection based on dual-sensor systems are extended.     

AN EVALUATION OF THE AVAILABLE TECHNIQUES FOR ESTIMATING MISSING FECAL COLIFORM DATA1

ABSTRACT: This paper presents the findings of a study aimed at evaluating the available techniques for estimating missing fecal coliform (FC) data on a temporal basis. The techniques investigated include: linear and nonlinear regression analysis and interpolation functions, and the use of artificial neural networks (ANNs). In all, seven interpolation, two regression, and one ANN model structures were investigated. This paper also investigates the validity of a hypothesis that estimating missing FC data by developing different models using different data corresponding to different dynamics associated with different trends in the FC data may result in a better model performance. The FC data (counts/100 ml) derived from the North Fork of the Kentucky River in Kentucky were employed to calibrate and validate various models. The performance of various models was evaluated using a wide variety of standard statistical measures. The results obtained in this study are able to demonstrate that the ANNs can be preferred over the conventional techniques in estimating missing FC data in a watershed. The regression technique was not found suitable in estimating missing FC data on a temporal basis. Further, it has been found that it is possible to achieve a better model performance by first decomposing the whole data set into different categories corresponding to different dynamics and then developing separate models for separate categories rather than developing a single model for the composite data set.