A set of associated statistical tests for spatial clustering

The purpose of this paper is to develop a set of associated statistical tests for spatial clustering. In particular, a set of three associated tests will be developed; these will correspond to the three types of tests set out by Besag and Newell (general tests, focused tests, and tests for the detection of clustering). The associated tests draw primarily, though not exclusively, upon existing tests and results. The principal contributions are based upon the score statistic for focused tests, which has been an important approach to testing for clustering around environmental hazards. The first contribution consists of the formulation of a global statistic for general tests that corresponds to focused score statistics, along with an assessment of the distribution of the statistic under the null hypothesis of no raised incidence. The local score statistics used for focused tests will have the property of summing to the global statistic used for the corresponding general test. Attention is also given to the maximum local score statistic for the “test for the detection of clustering”. The critical values of this statistic which are required for testing the null hypothesis are described. Application of the methods is made to leukemia data for central New York State.     

Some statistical and conceptual issues in the detection of low-level environmental pollutants

With increasing concern over chemicals that are potential health hazards at low levels, determination of limits of detection have undergone considerable scrutiny. Most traditional detection limit estimators suffer from extensive statistical and/or conceptual limitations. In this paper, traditional detection limit estimators are described and critically evaluated. Using the terminology of Currie (1968), methods are categorized into decision limits versus detection limits. The methods are further categorized into single concentration design versus calibration design methodologies. While the single concentration design methods are useful for fixing ideas and clarifying definitions, they are shown to be extremely limited in practice since dependence of variability on concentration can neither be estimated or incorporated. Calibration-based detection limit estimators are described, compared and contrasted. Generalizations to non-constant variance, multiple future detection decisions and simultaneous control of Type I and II errors are provided. The various calibration-based methods are illustrated using real data and experimental design issues for detection limit studies discussed.     

Genetic Effects of Habitat Fragmentation on Common Species of Swiss Fen Meadows

Abstract:  The area of Caricion davallianae alliance in Switzerland has been considerably reduced and fragmented during the last 150 years. We assessed the genetic variability, inbreeding level, and among-population differentiation of two common habitat-specific plant species, Carex davalliana SM. and Succisa pratensis Moench, in 18 Caricion davallianae fen meadows subjected to fragmentation. We used a spatial field design of fen systems (six systems total), each consisting of one large habitat island and two small habitat islands. We used allozyme electrophoresis to derive standard genetic parameters ( A, P, H_O, H_E, F_IS, F_ST ). In Carex we identified a consistently lower A in isolated habitat islands; furthermore, H_E was lower in small habitat islands than in large habitat islands. In Succisa we identified a lower H_O in small habitat islands than in larger ones. Small habitat islands were marginally significantly differentiated (  F_ST ) from large islands for Succisa . For both species, no effects were evident for F_IS ; therefore, we argue that genetic drift rather than inbreeding is the main cause of the observed differences. The genetic structure of Carex and Succisa in small habitat islands differed from that in large habitat islands, but differences were small. It appears that the observed differences in genetic variability among fen meadows correspond to observed differences in fitness and demographic traits. We show that habitat fragmentation affects not only the rare species in an ecosystem but also reduces the survival probabilities of common species. One of the main goals of conservation should be to mitigate fragmentation of natural habitats in order to increase population sizes and connectivity.     

Detection and delineation of critical areas using echelons and spatial scan statistics with synoptic cellular data

Geographical surveillance for hotspot detection and delineation has become an important area of investigation both in geospatial ecosystem health and in geospatial public health. In order to find critical areas based on synoptic cellular data, geospatial ecosystem health investigations apply recently discovered echelon tools. In order to find elevated rate areas based on synoptic cellular data, geospatial public health investigations apply recently discovered spatial scan statistic tools. The purpose of this paper is to conceptualize a joint role for these together in the spirit of a cross-disciplinary cross-fertilization to accomplish more effective and efficient geographical surveillance for hotspot detection and delineation, and early warning system.     

Bayesian areal wombling via adjacency modeling

Recently, public health professionals and other geostatistical researchers have shown increasing interest in boundary analysis, the detection or testing of zones or boundaries that reveal sharp changes in the values of spatially oriented variables. For areal data (i.e., data which consist only of sums or averages over geopolitical regions), Lu and Carlin (Geogr Anal 37: 265–285, 2005) suggested a fully model-based framework for areal wombling using Bayesian hierarchical models with posterior summaries computed using Markov chain Monte Carlo (MCMC) methods, and showed the approach to have advantages over existing non-stochastic alternatives. In this paper, we develop Bayesian areal boundary analysis methods that estimate the spatial neighborhood structure using the value of the process in each region and other variables that indicate how similar two regions are. Boundaries may then be determined by the posterior distribution of either this estimated neighborhood structure or the regional mean response differences themselves. Our methods do require several assumptions (including an appropriate prior distribution, a normal spatial random effect distribution, and a Bernoulli distribution for a set of spatial weights), but also deliver more in terms of full posterior inference for the boundary segments (e.g., direct probability statements regarding the probability that a particular border segment is part of the boundary). We illustrate three different remedies for the computing difficulties encountered in implementing our method. We use simulation to compare among existing purely algorithmic approaches, the Lu and Carlin (2005) method, and our new adjacency modeling methods. We also illustrate more practical modeling issues (e.g., covariate selection) in the context of a breast cancer late detection data set collected at the county level in the state of Minnesota.     

The National Resources Inventory: a long-term multi-resource monitoring programme

Interest in natural resources and the environment has led to the development of new federal monitoring efforts, the expansion of existing federal inventory programmes, and discussions of inter-agency collaboration for natural resource assessment data collection. As federal programmes evolve, knowledge gained from existing long-term survey programmes can provide valuable contributions to statistical and operational aspects of survey efforts. This paper describes the National Resources Inventory (NRI), which has been conducted by the US Department of Agriculture's Natural Resources Conservation Service in cooperation with the Iowa State University Statistical Laboratory for several decades. The current NRI is a longitudinal survey of soil, water, and related environmental resources designed to assess conditions and trends every five years on non-federal US lands. An historical overview is provided highlighting the development of the survey programme. Sample design, data collection, and estimation procedures used in the 1992 NRI are described, and statistical issues related to long-term monitoring are discussed.     

Assessing Retrospective National Water Model Streamflow with Respect to Droughts and Low Flows in the Colorado River Basin

Streamflow monitoring in the Colorado River Basin (CRB) is essential to ensure diverse needs are met, especially during periods of drought or low flow. Existing stream gage networks, however, provide a limited record of past and current streamflow. Modeled streamflow products with more complete spatial and temporal coverage (including the National Water Model [NWM]), have primarily focused on flooding, rather than sustained drought or low flow conditions. Objectives of this study are to (1) evaluate historical performance of the NWM streamflow estimates (particularly with respect to droughts and seasonal low flows) and (2) identify characteristics relevant to model inputs and suitability for future applications. Comparisons of retrospective flows from the NWM to observed flows from the United States Geological Survey stream gage network over 22 years in the CRB reveal a tendency for underestimating low flow frequency, locations with low flows, and the number of years with low flows. We found model performance to be more accurate for the Upper CRB and at sites with higher precipitation, snow percent, baseflow index, and elevations. Underestimation of low flows and variable model performance has important implications for future applications: inaccurate evaluations of historical low flows and droughts, and less reliable performance outside of specific watershed/stream conditions. This highlights characteristics on which to focus future model development efforts.     

基于监测统计的垃圾收运车恶臭污染释放特征

为了明确生活垃圾初期收运过程中恶臭物质的释放特征,对典型垃圾收运压缩车和非压缩车展开了为期一年的监测.共检出了6大类别75种恶臭物质,其中以乙醇为主的含氧化合物浓度最高,平均检出浓度超过360μg/m³.垃圾初期收运车辆释放的恶臭物质在秋季污染水平最高,而冬季污染水平较低.通过对全部样品进行统计分析发现,压缩车与非压缩车释放恶臭物质的种类与浓度之间并无明显差异.根据阈稀释倍数法,乙醇、萘、二氯甲烷、二甲二硫醚和丙烯醛被筛选为典型恶臭物质.其中含硫化合物嗅阈值低,即使浓度很低也容易被感受到,在检测过程需要格外关注.本研究可为固体废物收运与管理过程的二次污染控制提供科学依据.     

基于模糊综合评价-集值统计法的大型商场外因火灾风险性分析

大型商场火灾发生的原因多种多样,且易造成严重的后果和恶劣的社会影响。为了提高大型商场火灾风险性等级并且为其评价研究提供一种更加合理的评价方法,采用将2级模糊综合评价与集值统计法相结合的模型对大型商场外因火灾的风险性进行评价。根据相关法律法规并借鉴已有研究成果,充分考虑外因火灾的特点(主要从消防设施及能力因素、消防安全管理因素、人员应急疏散因素3个方面进行分析),建立针对大型商场外因火灾的评价指标体系,其中二级指标包括3个,三级指标包括12个。利用集值统计法将评价指标的描述表示为一段区间值,进而确定各评价指标的权重,并对得出的权重进行可靠性分析,采用加权平均模型确定火灾风险性总的相对隶属度,并求出模糊特征向量;最后根据总得分确定火灾的风险性等级。利用该方法对沈阳某大型商场外因火灾危险性进行了评价分析,并与传统的评价方法(层次分析法)相对比,其评价结果都为"二级(好)",表明该方法的评价结果合理。同时分析得到了大型商场火灾防控的薄弱环节,并提出了相应的有效改进措施。     

A data-driven allocation tool for in-kind resources distributed by a state health department

Objective: The objective of this study was to leverage a state health department's operational data to allocate in-kind resources (children's car seats) to counties, with the proposition that need-based allocation could ultimately improve public health outcomes.

Methods: This study used a retrospective analysis of administrative data on car seats distributed to counties statewide by the Georgia Department of Public Health and development of a need-based allocation tool (presented as interactive supplemental digital content, adaptable to other types of in-kind public health resources) that relies on current county-level injury and sociodemographic data.

Results: Car seat allocation using public health data and a need-based formula resulted in substantially different recommended allocations to individual counties compared to historic distribution.

Conclusions: Results indicate that making an in-kind public health resource like car seats universally available results in a less equitable distribution of that resource compared to deliberate allocation according to public health need. Public health agencies can use local data to allocate in-kind resources consistent with health objectives; that is, in a manner offering the greatest potential health impact. Future analysis can determine whether the change to a more equitable allocation of resources is also more efficient, resulting in measurably improved public health outcomes.