Modelling Ecological Presence–absence Data along an Environmental Gradient: Threshold Levels of the Environment

A methodology for estimating environmental thresholds of binary presence–absence data is presented where the level of the threshold is parameterised. Presence–absence data is fitted to three complementary different models: an independent null-model, a monotonically increasing or decreasing model, and an optimum model. The range of the three models is strictly between zero and one and the models are therefore well suited for modelling presence probabilities. The results of the three models may be combined by using Bayesian model selection methodologies. The proposed methodology is exemplified on observed binary presence–absence data of Bauera rubioides along an elevation gradient. Received: May 2005 / Revised: July 2005 An erratum to this article is available at.     

The theoretical value of encounters with parasitized hosts for parasitoids

A female parasitoid searching for hosts in a patch experiences a diminishing encounter rate with unparasitized and thus suitable hosts. To use the available time most efficiently, it constantly has to decide whether to stay in the patch and continue to search for hosts or to search for and travel to another patch in the habitat. Several informational cues can be used to optimize the searching success. Theoretically, encounters with unparasitized hosts should lead to a prolonged search in a given patch if hosts are distributed contagiously. The results of empirical studies strongly support this hypothesis. However, it has, to date, not been investigated theoretically whether encounters with already parasitized hosts (which usually entail time costs) provide a parasitoid with valuable information for the optimization of its search in depletable patches, although the empirical studies concerning this question so far have produced ambiguous results. Building on recent advances in Bayesian foraging strategies, we approached this problem by modeling a priori searching strategies (which differ in the amount of information considered) and then testing them in computer simulations. By comparing the strategies, we were able to determine whether and how encounters with already parasitized hosts can yield information that can be used to enhance a parasitoid’s searching success.

Understanding farmers' influence on land-use change using a participatory Bayesian network approach in a pre-Alpine region in Switzerland

Land-use models can be used to assess the importance of different drivers of land-use change. Local actors make land-use decisions on the basis of both biophysical and policy aspects, but they can also be considered as autonomous drivers as their attitudes and beliefs influence land-use substantially. We use a Bayesian network-based Land-use Modeling Approach (BLUMAP) to analyze influences of local actor characteristics on land-use change in a spatially explicit manner. Our analysis shows that local actor characteristics have a greater influence on land-use change than changes in agricultural policy schemes. Furthermore, focusing on the probabilities of land-use occurrence under different scenarios facilitates the quantification of influences of local actor characteristics on land-use changes and aids in the detection of where land-use changes are more likely to occur. We demonstrate that local actor characteristics could override land-use policy trends; thus, greater consideration should be paid to actors in land-use development processes.     

Toward Integrative Management Advice of Water Quality, Oil Spills, and Fishery in the Gulf of Finland: A Bayesian Approach

Understanding and managing ecosystems affected by several anthropogenic stressors require methods that enable analyzing the joint effects of different factors in one framework. Further, as scientific knowledge about natural systems is loaded with uncertainty, it is essential that analyses are based on a probabilistic approach. We describe in this article about building a Bayesian decision model, which includes three stressors present in the Gulf of Finland. The outcome of the integrative model is a set of probability distributions for future nutrient concentrations, herring stock biomass, and achieving the water quality targets set by HELCOM Baltic Sea Action Plan. These distributions can then be used to derive the probability of reaching the management targets for each alternative combination of management actions.     

Smallholder Farms as Stepping Stone Corridors for Crop-Raiding Elephant in Northern Tanzania: Integration of Bayesian Expert System and Network Simulator

Crop-raiding elephants affect local livelihoods, undermining conservation efforts. Yet, crop-raiding patterns are poorly understood, making prediction and protection difficult. We hypothesized that raiding elephants use corridors between daytime refuges and farmland. Elephant counts, crop-raiding records, household surveys, Bayesian expert system, and least-cost path simulation were used to predict four alternative categories of daily corridors: (1) footpaths, (2) dry river beds, (3) stepping stones along scattered small farms, and (4) trajectories of shortest distance to refuges. The corridor alignments were compared in terms of their minimum cumulative resistance to elephant movement and related to crop-raiding zones quantified by a kernel density function. The “stepping stone” corridors predicted the crop-raiding patterns. Elephant presence was confirmed along these corridors, demonstrating that small farms located between refuges and contiguous farmland increase habitat connectivity for elephant. Our analysis successfully predicted elephant occurrence in farmland where daytime counts failed to detect nocturnal presence. These results have conservation management implications.     

稠油热采井管柱失效动态风险分析研究

为保证稠油热采井筒管柱的安全运行,采用故障树与贝叶斯网络相结合的方法开展稠油热采井管柱失效风险分析,应用故障树分析方法定性识别管柱失效诱因,建立稠油热采井管柱失效场景演化模型,并将故障树转化为贝叶斯网络模型,应用贝叶斯网络对管柱失效风险进行定量分析,得出热采井管柱失效的关键致因和管柱失效的动态失效概率,以新疆油田某区块热采井的管柱失效为对象进行了案例分析。研究结果表明,该方法可为稠油热采井管柱失效风险分析及井筒完整性管理提供理论支撑和工程设计参考。     

基于贝叶斯网络分类的土壤盐渍化遥感监测

贝叶斯网络是一种将贝叶斯概率方法和有向无环图的网络拓扑结构有机结合的概率模型。采用贝叶斯网络分类对具有典型干旱特征的库车县土壤盐渍化情况进行监测，首先应用条件独立性测试原理建立贝叶斯网络结构，把研究区遥感数据进行离散化，然后应用贝叶斯定理作为分类原则，将每个像元分为像元最大概率的类别。研究结果表明该方法分类6种地类的整体分类精度达到96％，为该区盐渍地面积、空间分布等特征监测提供了较好的依据。     

基于PSR与贝叶斯网络的非常规突发事件情景分析

基于情景分析在非常规突发事件应急决策中的重要性,为了提高情景分析的有效性、全面性,在非常规突发事件"情景"界定和演变规律分析的基础上,基于PSR模型构建非常规突发事件的"压力-状态-响应"网络表达方式;利用贝叶斯网络理论,构建非常规突发事件的情景演变分析模型;以简化后的大连输油管道爆炸事件为例,示范基于PSR模型与贝叶斯网络的非常规突发事件情景分析方法的具体流程,并对推理结果进行分析。分析结果与现实情况基本一致,证明方法的可行性和有效性。实际应用中可针对具体的事件进行"多响应"、"多路径"的推理分析,以全面、系统地分析事件情景的演变情况。     

基于贝叶斯网络的尾流事故机理分析

针对事故树分析法的局限性,在尾流事故树的基础上,建立贝叶斯网络(BN)。运用推理运算对BN进行定量分析,得出:空中交通密度太大、空中交通管制(ATC)间隔判断错误和短期冲突告警(STCA)被忽略是事故的关键致因。将针对致因提出的改进措施引入到BN中,评价相关措施的有效性。应用BN进行尾流事故的机理分析,能够以比逻辑门更好的形式表达变量间的不确定性关系,从而更加方便地找到导致事故发生的关键因素。     

A method for simplifying the analysis of traffic accidents injury severity on two-lane highways using Bayesian networks

Introduction

This study describes a method for reducing the number of variables frequently considered in modeling the severity of traffic accidents. The method's efficiency is assessed by constructing Bayesian networks (BN).

Method

It is based on a two stage selection process. Several variable selection algorithms, commonly used in data mining, are applied in order to select subsets of variables. BNs are built using the selected subsets and their performance is compared with the original BN (with all the variables) using five indicators. The BNs that improve the indicators’ values are further analyzed for identifying the most significant variables (accident type, age, atmospheric factors, gender, lighting, number of injured, and occupant involved). A new BN is built using these variables, where the results of the indicators indicate, in most of the cases, a statistically significant improvement with respect to the original BN.

Conclusions

It is possible to reduce the number of variables used to model traffic accidents injury severity through BNs without reducing the performance of the model.

Impact on Industry

The study provides the safety analysts a methodology that could be used to minimize the number of variables used in order to determine efficiently the injury severity of traffic accidents without reducing the performance of the model.