基于逻辑树和贝叶斯网络的海洋平台火灾概率分析

火灾概率分析是海洋平台火灾定量风险评估的重要组成部分,考虑到传统事故树和事件树方法存在一定的局限性,提出了基于逻辑树和贝叶斯网络的火灾概率分析模型。首先采用数理统计方法对墨西哥湾地区2 837起火灾事故进行统计分析,依据事故情况构建逻辑树,然后将逻辑树转化为贝叶斯网络,根据历史数据确定贝叶斯网络各节点的先验概率和条件概率。结果表明:海洋平台火灾事故是设备、人因和组织管理多因素耦合作用的结果;基于贝叶斯网络模型得到海洋平台火灾概率约为1.0×10~(-5),为海洋平台火灾定量风险评估提供了基础数据;由贝叶斯网络模型分析得出,人因操作失误与缺乏作业安全分析的后验概率分别达到0.471和0.119,表明人因组织因素对海洋平台火灾事故具有重要影响。     

The multiphase optimization strategy (MOST) and the sequential multiple assignment randomized trial (SMART): two novel evaluation methods for developing optimal training programs

Current methodologies in training evaluation studies largely employ a single method entitled random confirmatory trials, prompting several concerns. First, practitioners and researchers often analyze the effectiveness of their entire omnibus training, rather than the individual elements or identifiable components of the training program. This slows the testing of theory and development of optimal training programs. Second, a common training is typically administered to all employees within an organization or workgroup; however, certain factors may cause individualized training to be more effective. Given these concerns, the current paper presents two training evaluation methodologies to overcome these problems: the multiphase optimization strategy and sequential multiple assignment randomized trials. The multiphase optimization strategy is a method to evaluate a standard training, which emphasizes the importance of a multi‐stage training evaluation process to analyze individual training elements. In contrast, sequential multiple assignment randomized trial is used to evaluate an adaptive training that varies over time and/or trainees. These methodologies jointly overcome the problems noted earlier, and they can be integrated to address several of the key challenges facing training researchers and practitioners. Copyright © 2016 John Wiley & Sons, Ltd.     

Assessment of the Extreme Rainfall Event at Nashville,TN and the Surrounding Region on May 1–3, 2010

This paper analyzes the May 1–3, 2010 rainfall event that affected the south‐central United States, including parts of Mississippi, Tennessee, and Kentucky. The storm is evaluated in terms of its synoptic setting, along with the temporal distributions, and spatial patterns of the rainfall. In addition, the recurrence interval of the storm is assessed and the implications for hydrologic structure designs are discussed. The event was associated with an upper‐level trough and stationary frontal boundary to the west of the rainfall region, which remained quasi‐stationary for a period of 48 h. Heavy rainfall was produced by two slow‐moving mesoscale convective complexes, combined with abundant atmospheric moisture. Storm totals exceeding 330 mm occurred within a large elongated area extending from Memphis to Nashville. Isolated rainfall totals over 480 mm were reported in some areas, with NEXRAD weather radar rainfall estimates up to 501 mm. An extreme value analysis was performed for one‐ and two‐day rainfall totals at Nashville and Brownsville, Tennessee, as well as for gridded rainfall estimates for the entire region using the Storm Precipitation Analysis System. Results suggest maximum rainfall totals for some durations during the May 1–3, 2010 event exceeded the 1,000‐year rainfall values from National Oceanic and Atmospheric Administration Atlas 14 for a large portion of the region and reached up to 80% of the probable maximum precipitation values for some area sizes and durations.     

ArcGIS支持下的样本稀疏山区空间插值模拟探讨

样本稀疏地区空间插值法对区域化变量的精准管理具有重要意义。基于ArcGIS 90，在分析土壤属性空间分布特征的基础上，提出并构建了基于不同土壤类型的土壤特性空间预测模拟模型，对比了传统方法与改进方法空间插值精度，实现了数值插值在复杂地理环境区域的应用，得到以下结论：(1)基于经度、纬度、海拔高度及坡度等地理因子的土壤基础环境因子的空间预测模拟模型，突破以往只能描述土壤属性在水平方向变化的局限，较客观、合理地反映土壤属性随地理位置及海拔高度的立体变化特征；(2)基于不同土壤类型回归模型来增加样本点以推断评价指标在无取样地区的分布状况的处理方式具有一定的数学理论支撑，有效降低了插值误差，提高了评价精度，使评价结果更加接近现实。     

How are Results from Critical Load Calculations Reflected in Lake Water Chemistry?

The Steady State Water Chemistry Model (SSWCM) is a common method for determinations of critical loadof acid and subsequently of critical loadexceedance for lakes. One way to verify the modeloutput, is to compare with chemical indicatorssuch as pH-value, alkalinity or ANC. When themedian chemical status (as ANC) is used 65% ofthe lakes responded according to the exceedancevalue. For these the calculated exceedanceresulted in violation of the critical chemicalvalue while non-exceedance gave no violation.Since biota react on extreme conditions a morecorrect evaluation should be based on minimumvalues for the chemical indicator. This raises thefraction of lakes responding to 78%. Non-exceedance is seldom found inlakes with acid conditions. The evaluationindicates that the calculation of critical load ofacidity by means of SSWCM is very reliable.     

IDENTIFICATION OF CHANGES IN STREAMFLOW CHARACTERISTICS1

ABSTRACT: A set of procedures for identifying changes in selected streamflow characteristics at sites having long‐term continuous streamflow records is illustrated by using streamflow data from the Waccamaw River at Freeland, North Carolina for the 55‐year period of 1940–1994. Data were evaluated and compared to streamflow in the adjacent Lumber River Basin to determine if changes in streamflow characteristics in the Waccamaw River were localized and possibly the result of some human activity, or consistent with regional variations. Following 1963, droughts in the Waccamaw Basin seem to have been less severe than in the Lumber Basin, and the annual one‐, seven‐, and 30‐day low flows exhibited a slightly increasing trend in the Waccamaw River. Mean daily flows in the Waccamaw River at the 90 percent exceedance level (low flows) during 1985–194, a relatively dry period, were very nearly equal to flows at the same exceedance level for 1970–1979, which represents the 10‐year period between 1940 and 1994 with the highest flows. Prior to the 1980s, flows per unit drainage area in the Waccamaw Basin were generally less than those in the Lumber Basin, but after 1980, the opposite was true. The ratio of base flow to runoff in the Waccamaw River may have changed relative to that in the Lumber River in the late 1970s. There was greater variability in Waccamaw River streamflow than in Lumber River flow, and flow variability in the Waccamaw River may have increased slightly during 1985–1994.     

COMPARISON OF PARAMETRIC TAIL ESTIMATORS FOR LOW‐FLOW FREQUENCY ANALYSIS1

ABSTRACT: In recent years, several approaches to hydrologic frequency analysis have been proposed that enable one to direct attention to that portion of an overall probability distribution that is of greatest interest. The majority of the studies have focused on the upper tail of a distribution for flood analyses, though the same ideas can be applied to low flows. This paper presents an evaluation of the performances of five different estimation methods that place an emphasis on fitting the lower tail of the lognormal distribution for estimation of the ten‐year low‐flow quantile. The methods compared include distributional truncation, MLE treatment of censored data, partial probability weighted moments, LL‐moments, and expected moments. It is concluded that while there are some differences among the alternative methods in terms of their biases and root mean square errors, no one method consistently performs better than the others, particularly with recognition that the underlying population distribution is unknown. Therefore, it seems perfectly legitimate to make a selection of a method on the basis other criteria, such as ease of use. It is also shown in this paper that the five alternative methods can perform about as well as, if not better than, an estimation strategy involving fitting the complete lognormal distribution using L‐moments.     

2006年5-6月国内环境事件

简要统计了2006年5-6月国内发生的各种环境事件197起,包括沙尘天气18起,污染事件48起,地震13起,山体滑坡和泥石流22起,虫害12起,旱灾15起,以及其他自然灾害69起.对自然灾害进行了讨论.     

The Merseyside Accident Information Model (MAIM) can reveal components of accidents that lead to attendance at fracture clinics and cause disability: a new approach to accident prevention

The objectives of this study were to identify components of accidents that cause the most disability and to discover the principal sources of injuries treated in the fracture clinics. Patients attending fracture clinics of the Royal Liverpool University Hospital were interviewed using a portable computer-based questionnaire, the Merseyside Accident Information Model (MAIM). Patients were followed up by telephone interview or letter to enquire about disability continuing after discharge. Disability was measured by the pre-accident to post-discharge changes in scores for 11 normal functions. Of the 1326 patients interviewed, 900 (68%) were successfully followed up and 37% reported disability after discharge. First events ‘tripping’, ‘slipping’ and ‘other underfoot events’ accounted for 433 patients (194 reporting disability), and ‘collapsed/fainted — no other event' for 66 patients (27 reporting disability). Activities at the time of accident most frequently associated with disability involved moving about on foot. Among first event objects, ground surfaces and underfoot hazards were reported in 35%. Sources of injuries included underfoot accidents (48%), sport (13%), and transport accidents (12%.). Underfoot accidents contributed to 58% of patients reporting disability, sport 6% and transport accidents 11%. Underfoot accidents together with ‘collapsed/fainted — no other event’ accounted for 79% of female patients reporting disability and 50% of men. Such data could be used for cost-effective targeting of preventative measures, and to study the effectiveness of accident prevention initiatives.     

Using hidden Markov chains and empirical Bayes change-point estimation for transect data

Consider a lattice of locations in one dimension at which data are observed. We model the data as a random hierarchical process. The hidden process is assumed to have a (prior) distribution that is derived from a two-state Markov chain. The states correspond to the mean values (high and low) of the observed data. Conditional on the states, the observations are modelled, for example, as independent Gaussian random variables with identical variances. In this model, there are four free parameters: the Gaussian variance, the high and low mean values, and the transition probability in the Markov chain. A parametric empirical Bayes approach requires estimation of these four parameters from the marginal (unconditional) distribution of the data and we use the EM-algorithm to do this. From the posterior of the hidden process, we use simulated annealing to find the maximum a posteriori (MAP) estimate. Using a Gibbs sampler, we also obtain the maximum marginal posterior probability (MMPP) estimate of the hidden process. We use these methods to determine where change-points occur in spatial transects through grassland vegetation, a problem of considerable interest to plant ecologists.