Effects of sample survey design on the accuracy of classification tree models in species distribution models

We evaluated the effects of probabilistic (hereafter DESIGN) and non-probabilistic (PURPOSIVE) sample surveys on resultant classification tree models for predicting the presence of four lichen species in the Pacific Northwest, USA. Models derived from both survey forms were assessed using an independent data set (EVALUATION). Measures of accuracy as gauged by resubstitution rates were similar for each lichen species irrespective of the underlying sample survey form. Cross-validation estimates of prediction accuracies were lower than resubstitution accuracies for all species and both design types, and in all cases were closer to the true prediction accuracies based on the EVALUATION data set. We argue that greater emphasis should be placed on calculating and reporting cross-validation accuracy rates rather than simple resubstitution accuracy rates. Evaluation of the DESIGN and PURPOSIVE tree models on the EVALUATION data set shows significantly lower prediction accuracy for the PURPOSIVE tree models relative to the DESIGN models, indicating that non-probabilistic sample surveys may generate models with limited predictive capability. These differences were consistent across all four lichen species, with 11 of the 12 possible species and sample survey type comparisons having significantly lower accuracy rates. Some differences in accuracy were as large as 50%. The classification tree structures also differed considerably both among and within the modelled species, depending on the sample survey form. Overlap in the predictor variables selected by the DESIGN and PURPOSIVE tree models ranged from only 20% to 38%, indicating the classification trees fit the two evaluated survey forms on different sets of predictor variables. The magnitude of these differences in predictor variables throws doubt on ecological interpretation derived from prediction models based on non-probabilistic sample surveys.     

Toward evaluation of natural objects

According to the suggested approach to the evaluation of natural objects current human needs should not be used in evaluation procedures. The ecological value of natural objects (communities and ecosystems) is proportional to the probabilities of disappearance associated with particular sets of objects. The ecological value is estimated using data on relative fragility, uniqueness, size of regeneration unit for defined types of objects, and intensity of destructive factors. The ecological value of a specific territory is equal to the sum of the values that are attached to the local objects situated at this territory.     

模糊概率事件的故障树分析

许多职业事故的基本致因事件多属于模糊事件,其概率很难确定。本文以煤矿典型危害——内因火灾为例,探讨了模糊概率下的故障树分析问题。煤矿内因火灾故障树虽具有确定的结构形式,其基本致因事件的概率却无法统计。经过长期的生产实践,煤矿中一大批防火专家可以用自然语言较准确地描述基本事件发生的可能性,这就为故障树分析提供了有效信息。文中以模糊数学为基础,引入Delphi方法,征集并整理了25位专家对龙凤矿内因火灾致因的意见,确定了基本事件的模糊概率及其临界重要度排序,提出了行之有效的防治内因火灾措施。     

Flame propagation mechanisms in dust explosions

To reveal the effects of particle characteristics, including particle thermal characteristics and size distributions, on flame propagation mechanisms during dust explosions clearly, the flame structures of dust clouds formed by different materials and particle size distributions were recorded using an approach combining high-speed photography and a band-pass filter. Two obviously different flame propagation mechanisms were observed in the experiments: kinetics-controlled regime and devolatilization-controlled regime. Kinetics-controlled regime was characterized by a regular shape and spatially continuous combustion zone structure, which was similar to the premixed gas explosions. On the contrary, devolatilization-controlled regime was characterized by a complicated structure that exhibited heterogeneous combustion characteristics, discrete blue luminous spots appeared surrounding the yellow luminous zone. It was also demonstrated experimentally that the flame propagation mechanisms transited from kinetics-controlled to devolatilization-controlled while decreasing the volatility of the materials or increasing the size of the particles. Damköhler number was defined as the ratio of the heating and devolatilization characteristic time to the combustion reaction characteristic time, to reflect the transition of flame propagation mechanisms in dust explosions. It was found that the kinetics-controlled regime and devolatilization-controlled regime can be categorized by whether Damköhler number was less than 1 or larger than 1.     

Risk-based inspection (RBI) of a gas pressure reduction station

One of the principles of corrosion management is proper inspection methodology, and in particular the risk-based inspection (RBI). Occasionally a large percentage of the total risk of a unit is concentrated in a small part of the equipment, which can be reduced by using risk management techniques. In this study, a semi-quantitative risk-based inspection method has been performed in a gas station with for its components (pipelines, filters, etc.), to identify the existing damage mechanisms and also to rank its equipment in terms of inspection intervals. In this regard, RBI was performed for the gas station which consisted of phases (Phase A & Phase B) with Class 300 and Class 600 equipment. According to documented conditions and reviewed records the main damage mechanism for the investigated components in the gas station was identified as wall thinning due to erosion corrosion and external corrosion. In terms of inspection, results showed that most of the equipment studied had medium risk level and a few had medium to high level of risk. In this regard, the corrosion rates of Class 300 and Class 600 equipment were also studied and compared, and it was indicated that the corrosion rates were higher at locations for Class 300 equipment where more pressure drop had occurred. Overall, based on the obtained corrosion rates and obtained risk matrixes, it was concluded that wall thinning (erosion-corrosion) was the determining risk factor for equipment inspected in the gas pressure reduction station.     

Experimental study on sediment-copper distributions in hyper-concentrated turbulent solid-liquid system

This study presents a special problem on vertical distribution for sediment and copper in hyper-concentrated turbulent solid-liquid system that is essentially different from the ordinary low-concentrated turbulent system. A resonance type turbulent simulation equipment is used for the experimental study in which a vertically uniform turbulent field of the mixture of loess and water is produced in a testing cylinder with a grille stirrer that moves up and down harmoniously with varying vibration frequencies, in order to compare the variations of the vertical profiles of sediment and copper in low- and hyper-concentratod solid-liquid system, different scenarios for input sediment content ranging from 5 to 800 kg/m^3 was considered in the experimental studies. It was found that solids copper content increases with input sediment content, So, and reaches its peak as So goes to 10 kg/m^3 and then decreases rapidly with increasing input sediment content. Such a behavior is possibly resulted from the joint effect of the specific adsorption of copper on loess, precipitation of carbonate and hydroxide of copper due to high carbonate content in the loess and the so-called ＂particulate concentration effect＂ due to the present of the sediment variation in water. The vertical sediment concentration distribution resulted from the uniform turbulence is generally uniform, but slight non-uniformity does occur as sediment concentration exceeds certain value. However, the vertical concentration distributions of soluble copper seem not to be affected much by the variation of sediment concentrations.     

Discrepancies in the acute versus chronic toxicity of compounds with a designated narcotic mechanism

In this study, it was illustrated that even for certain simple organic compounds with a designated mode of action (MOA) (i.e. narcotic toxicity) unexpected differences in acute and chronic toxicity can be observed. In a first part of the study, species sensitivity distributions (SSDs) based on either acute or chronic toxicity data of three narcotic test compounds (methanol, ethanol and 2-propanol) were constructed. The results of the acute SSDs were as expected for narcotic compounds: rather similar sensitivity and small differences in toxicity were observed among different species. On the contrary, the chronic SSDs of methanol and ethanol indicated larger interspecies variation in sensitivity. Furthermore, the chronic toxicity trend (ethanol > methanol > 2-propanol) was unexpectedly different from the acute toxicity trend (2-propanol > ethanol > methanol) and acute versus chronic extrapolation could not be successfully described for methanol and ethanol using an ACR of 10 (as suggested for narcotic compounds). In contrast to the interspecies approach in the first part of this study, the second part of the study was focused on the assessment of acute and chronic toxicity of the three test compounds in Daphnia magna, which was identified as one of the most sensitive organisms to methanol and ethanol. Here, the differences in acute and chronic toxicity trend were in accordance to the results of the SSDs. The enhancement of membrane penetration due to the small molecular size of methanol and ethanol, in combination with the higher toxicity of their respective biotransformation products were suggested as potential causes of the increased chronic toxicity. Furthermore, it was stressed that larger awareness of these irregularities in acute to chronic extrapolations of narcotic compounds is required and should receive additional attention in further environmental risk assessment procedure.     

Statistical Issues in Assessing Anthropogenic Background for Arsenic

Conceptual and statistical issues surrounding the estimation of a background concentration distribution for arsenic are reviewed. How background area is defined and samples collected are shown to impact the shape and location of the probability density function that in turn affects the estimation and precision of associated distributional parameters. The overall background concentration distribution is conceptualized as a mixture of a natural background distribution, an anthropogenic background distribution and a distribution designed to accommodate the potential for contamination site samples being included into the background sample set. This concept is extended to a discussion of issues surrounding estimation of natural and anthropogenic background distributions for larger geographic areas. Finally, the mixture model is formally defined and statistical approaches to estimating its parameters discussed.     

Dobbins概率边界模型及应用

在W.J.Padgett等对随机水质模型研究的基础上,本文以样本资料为重要信息,直接推求BOD和DO的联合分布函数。以此为基础,结合地面水水质标准,推出BOD-DO的耦合达标率公式,同时结合Dobbins模型导出了上游断面的BOD-DO概率边界值计算式。以沱江流域顺河场至中瓷厂水质资料作为实例进行计算,结果表明模型有广阔应用前景。      

Fundamental insights on when social network data are most critical for conservation planning

As declines in biodiversity accelerate, there is an urgent imperative to ensure that every dollar spent on conservation counts toward species protection. Systematic conservation planning is a widely used approach to achieve this, but there is growing concern that it must better integrate the human social dimensions of conservation to be effective. Yet, fundamental insights about when social data are most critical to inform conservation planning decisions are lacking. To address this problem, we derived novel principles to guide strategic investment in social network information for systematic conservation planning. We considered the common conservation problem of identifying which social actors, in a social network, to engage with to incentivize conservation behavior that maximizes the number of species protected. We used simulations of social networks and species distributed across network nodes to identify the optimal state-dependent strategies and the value of social network information. We did this for a range of motif network structures and species distributions and applied the approach to a small-scale fishery in Kenya. The value of social network information depended strongly on both the distribution of species and social network structure. When species distributions were highly nested (i.e., when species-poor sites are subsets of species-rich sites), the value of social network information was almost always low. This suggests that information on how species are distributed across a network is critical for determining whether to invest in collecting social network data. In contrast, the value of social network information was greatest when social networks were highly centralized. Results for the small-scale fishery were consistent with the simulations. Our results suggest that strategic collection of social network data should be prioritized when species distributions are un-nested and when social networks are likely to be centralized.