Improved management practice and process hazard analysis techniques for minimizing likelihood of process safety incidents in Taiwan

In Taiwan, process safety accidents often occur despite the prior implementation of process hazard analysis (PHA). One of the main reasons for this is the poor quality of the PHA process; with the main hazards not being properly identified, or properly controlled. Accordingly, based on the findings of 86 process safety management (PSM) audits, dozens of post-accident site resumption review meetings, and hundreds of PSM review sessions, this study examines the main deficiencies of management practice and PHA implementation in Taiwan, and presents several recommendations for improved PHA assessment techniques and procedures. The study additionally examines the feasibility for using PSM-related information, such as process safety information and process incident information, as a tool for further enhancing the PHA quality. Overall, the study suggests that, in addition to following the basic rules of PHA and requirements of OSHA (1992),management in Taiwan should also provide training in the enhanced assessment techniques proposed herein and take active steps to incorporate PSM information into the PHA framework in order to improve the general quality of PHA and reduce the likelihood of process safety accidents accordingly.     

Mechanisms and impacts of an incentive-based conservation program with evidence from a randomized control trial

Conservation science needs more high-quality impact evaluations, especially ones that explore mechanisms of success or failure. Randomized control trials (RCTs) provide particularly robust evidence of the effectiveness of interventions (although they have been criticized as reductionist and unable to provide insights into mechanisms), but there have been few such experiments investigating conservation at the landscape scale. We explored the impact of Watershared, an incentive-based conservation program in the Bolivian Andes, with one of the few RCTs of landscape-scale conservation in existence. There is strong interest in such incentive-based conservation approaches as some argue they can avoid negative social impacts sometimes associated with protected areas. We focused on social and environmental outcomes based on responses from a household survey in 129 communities randomly allocated to control or treatment (conducted both at the baseline in 2010 and repeated in 2015–2016). We controlled for incomplete program uptake by combining standard RCT analysis with matching methods and investigated mechanisms by exploring intermediate and ultimate outcomes according to the underlying theory of change. Previous analyses, focused on single biophysical outcomes, showed that over its first 5 years Watershared did not slow deforestation or improve water quality at the landscape scale. We found that Watershared influenced some outcomes measured using the survey, but the effects were complex, and some were unexpected. We thus demonstrated how RCTs can provide insights into the pathways of impact, as well as whether an intervention has impact. This paper, one of the first registered reports in conservation science, demonstrates how preregistration can help make complex research designs more transparent, avoid cherry picking, and reduce publication bias.     

Hierarchical spatial point process analysis for a plant community with high biodiversity

A complex multivariate spatial point pattern of a plant community with high biodiversity is modelled using a hierarchical multivariate point process model. In the model, interactions between plants with different post-fire regeneration strategies are of key interest. We consider initially a maximum likelihood approach to inference where problems arise due to unknown interaction radii for the plants. We next demonstrate that a Bayesian approach provides a flexible framework for incorporating prior information concerning the interaction radii. From an ecological perspective, we are able both to confirm existing knowledge on species’ interactions and to generate new biological questions and hypotheses on species’ interactions.

External drift kriging of NOx concentrations with dispersion model output in a reduced air quality monitoring network

In the mid nineteen eighties the Dutch NO_x air quality monitoring network was reduced from 73 to 32 rural and city background stations, leading to higher spatial uncertainties. In this study, several other sources of information are being used to help reduce uncertainties in parameter estimation and spatial mapping. For parameter estimation, we used Bayesian inference. For mapping, we used kriging with external drift (KED) including secondary information from a dispersion model. The methods were applied to atmospheric NO_x concentrations on rural and urban scales. We compared Bayesian estimation with restricted maximum likelihood estimation and KED with universal kriging. As a reference we also included ordinary least squares (OLS). Comparison of several parameter estimation and spatial interpolation methods was done by cross-validation. Bayesian analysis resulted in an error reduction of 10 to 20% as compared to restricted maximum likelihood, whereas KED resulted in an error reduction of 50% as compared to universal kriging. Where observations were sparse, the predictions were substantially improved by inclusion of the dispersion model output and by using available prior information. No major improvement was observed as compared to OLS, the cause presumably being that much good information is contained in the dispersion model output, so that no additional spatial residual random field is required to explain the data. In all, we conclude that reduction in the monitoring network could be compensated by modern geostatistical methods, and that a traditional simple statistical model is of an almost equal quality.

Univariate Bayesian nonparametric binary regression with application in environmental management

In environmental management, we often have to deal with binary response variables whose outcome dictates the course of action. This paper introduces a nonparametric Bayesian binary regression model with a single predictor variable that is more flexible than the commonly used logistic or probit models. Due to the Bayesian feature, the model can be easily used to combine observed data with our knowledge of the subject to produce site-specific results. By using three examples, this paper shows the potential application of the model in the environmental management, and its advantages in terms of flexibility in model specification, robustness to outliers, and realistic interpretation of data.     

RADIUM - 226 IN GROUND WATER OF WEST CENTRAL FLORIDA1

ABSTRACT: Principal U.S. phosphate production is from central Florida where mining, processing, and waste disposal practices intimately associate the industry with water resouces. Available radium-226 data from 1966 and from 1973–1976 in mined and unmined mincralized areas and nonmineralized areas in the primary study area in Polk, Hardee, Hillsborough, Manatee, and De Soto counties were studied using log-normal probability plots and nonparametric statistical tests for significant difference as functions of time, depth, and location. Plots of radium in the water table and Floridan aquifers for mineralized and nonmineralized areas indicate that neither phosphate mineralization nor the industry is a probable factor. For the Lower Floridan aquifer, three separate radium populations are indicated with geometric means of 0.7, 3, and 10 pCi/1. Geometric mean radium-226 content of the water table aquifer is 0.17 pCi/1. Radium in the Floridan aquifer in Manatee and Sarasota Counties is elevated relative to that in the water table and in other areas of Florida. For Sarasota County, geometric mean radium content of the water table is 15 pCi/l versus 7.5 pCi/l in the Floridan. Potential sources include shallow phosphate sediments and monazite sands and possibly crystalline basement rocks or other strata unrelated to phosphatic zones of current economic interest. The existing radium-226 data base is rather marginal in terms of number and spatial distribution of analyses, particularly for the water table and Upper Floridan aquifer. Existing radium data do not substantiate widespread contamination of ground water as a result of the phosphate industry. However, local contamination associated with specific operations has occurred.     

Do statistical inferences allowing three alternative decisions give better feedback for environmentally precautionary decision-making?

Environmental policies and guidelines often specify standards for environmental indicators. The first part of this paper argues that, where compliance with these standards is assessed with the help of statistical inference, an inference employing a three-alternatives decision rule can provide more sensible feedback to environmental managers for precautionary decision-making. The second part of the paper shows how a three-alternatives statistical inference about compliance with a percentile standard might be applied to a small number of observations using a non-parametric binomial interval. This interval expression of uncertainty results in the sample size requirements for various percentile ranks becoming explicit.     

Application of Likelihood Ratio and Logistic Regression Models to Landslide Susceptibility Mapping Using GIS

For landslide susceptibility mapping, this study applied and verified a Bayesian probability model, a likelihood ratio and statistical model, and logistic regression to Janghung, Korea, using a Geographic Information System (GIS). Landslide locations were identified in the study area from interpretation of IRS satellite imagery and field surveys; and a spatial database was constructed from topographic maps, soil type, forest cover, geology and land cover. The factors that influence landslide occurrence, such as slope gradient, slope aspect, and curvature of topography, were calculated from the topographic database. Soil texture, material, drainage, and effective depth were extracted from the soil database, while forest type, diameter, and density were extracted from the forest database. Land cover was classified from Landsat TM satellite imagery using unsupervised classification. The likelihood ratio and logistic regression coefficient were overlaid to determine each factors rating for landslide susceptibility mapping. Then the landslide susceptibility map was verified and compared with known landslide locations. The logistic regression model had higher prediction accuracy than the likelihood ratio model. The method can be used to reduce hazards associated with landslides and to land cover planning. Note: This version was published online in June 2005 with the cover date of August 2004.     

A fuzzy expert system for soil characterization

As soil is a natural resource not always renewable, the risk characterization of contaminated soils is an issue of great interest. Artificial Intelligence (AI), based on Decision Support Systems (DSSs), has been developed for a wide range of applications in contaminated soil management. Decision trees have already shown to be easy to interpret and able to treat large scale applications. Fuzzy logic gives an improvement in the perturbations and the variance of the training data, due to the elasticity of fuzzy set formalism. In this study, we have developed a classificatory tool applied to characterize contaminated soil in function of human and environmental risks. Knowledge engineering for constructing the Soil Risk Characterization Decision Support System (SRC-DSS) involves three stages: knowledge acquisition, conceptual design and system implementation. A total of 26 parameters were divided into three groups to facilitate the configuration of the expert system: source attributes, transfer vector attributes, and local properties. Sixteen case studies were evaluated with the SRC-DSS. In comparison with other techniques, the results of the current study have shown that SRC-DDS is an excellent tool to classify and characterize soils according to the associated risk.     

“Decision-making” in conflicts of wild chimpanzees (<Emphasis Type="Italic">Pan troglodytes</Emphasis>): an extension of the Relational Model

>We examined the "decision-making" process of aggressive interactions within a community of wild chimpanzees (Pan troglodytes verus) in the Taï National Park, Côte d'Ivoire (West Africa). Costs and benefits were investigated for 876 dyadic aggressive interactions among 18 adults (including 4 independent adolescents) of either sex. An extended version of the Relational Model was developed to describe the dynamics of the "decision-making" process in Taï chimpanzees, which suggests that the net benefit determines the occurrence of conflicts. Both sexes fought more frequently for the resources that were most important to them, food for females and social contexts for males. Individuals used two different strategies according to their likelihood of winning the aggressive interaction, determined by the dominance relationship of the conflict partners. Dominant initiators had longer and more intense aggressive interactions, but they limited their social disadvantages by fighting non-cooperative partners. Subordinate initiators had shorter and less intense aggressive interactions, but risked more social costs, which they could reduce afterwards by reconciliation. Both strategies included a positive overall net benefit. The extended Relational Model fits the complexity of wild chimpanzee conflicts and allows for more flexibility in the "decision-making" compared to the original version.Communicated by E. Sterck