AN EVALUATION OF THE AVAILABLE TECHNIQUES FOR ESTIMATING MISSING FECAL COLIFORM DATA1

ABSTRACT: This paper presents the findings of a study aimed at evaluating the available techniques for estimating missing fecal coliform (FC) data on a temporal basis. The techniques investigated include: linear and nonlinear regression analysis and interpolation functions, and the use of artificial neural networks (ANNs). In all, seven interpolation, two regression, and one ANN model structures were investigated. This paper also investigates the validity of a hypothesis that estimating missing FC data by developing different models using different data corresponding to different dynamics associated with different trends in the FC data may result in a better model performance. The FC data (counts/100 ml) derived from the North Fork of the Kentucky River in Kentucky were employed to calibrate and validate various models. The performance of various models was evaluated using a wide variety of standard statistical measures. The results obtained in this study are able to demonstrate that the ANNs can be preferred over the conventional techniques in estimating missing FC data in a watershed. The regression technique was not found suitable in estimating missing FC data on a temporal basis. Further, it has been found that it is possible to achieve a better model performance by first decomposing the whole data set into different categories corresponding to different dynamics and then developing separate models for separate categories rather than developing a single model for the composite data set.     

洪涝灾害评价的威布尔模型

首次将威布尔分布用于淮河流域水灾成灾面积研究,揭示了淮河流域洪涝灾害成灾面积形成的 内在规律,进而利用河南、安徽、江苏和山东4省的各自灾度对淮河流域的灾害风险建立了线性回归模型.实证分析表明,本方法切实可行,特别适用于大样本计算.     

中国水灾社会脆弱性评估方法的改进与应用--以长沙地区为例

中国是洪水灾害发生频繁的国家之一,研究水灾脆弱性对中国的灾害风险管理有重要的意义.但是,合理评估脆弱性尤其是社会脆弱性却面临着极大的挑战.论文对评估社会脆弱性指数的研究现状进行了分析,指出了传统评估方法存在的问题,并尝试改进Hoovering评估模式.选择湘江流域的长沙地区为研究区,应用改进模型对研究区进行了社会脆弱性指数的评估.结果表明,长沙地区社会脆弱性从1980年至2000年基本处于下降趋势,但是在2002年和2003年增长迅速.2003年,长沙5区4县中社会脆弱性指数最大的是开福区,其次是长沙县,而宁乡县的社会脆弱性指数最低.     

地下水水质评价的多元线性回归分析模型研究

运用回归分析理论和方法,建立了一个基于多元线性回归分析法的地下水水质评价模型,并将该模型用于遵义市海龙坝地下水水质评价.结果表明,建立的模型较符合本研究区的实际情况.     

基于卫星夜间灯光数据的中国分省碳排放时空模拟

中国能源统计数据"横向不可比,纵向不可加"现象依然突出,尤其是分省能源消费统计千差万别,给分省碳排放评估带来了较大困难,如何利用卫星遥感数据科学合理地估算中国分省碳排放是当前亟须研究的问题。本文运用DMSP/OLS全球稳定夜间灯光数据,在通过相互校正、年内融合和年际间校正等系列处理得到中国分省稳定夜间灯光数据的基础上,首先分别构建中国分省稳定夜间灯光亮度DN值与人均碳排放和单位面积碳排放之间的时空地理加权回归模型,两个模型整体效果均较好,拟合优度分别高达96.74%和99.24%;其次运用稳定夜间灯光亮度DN值对分省人均碳排放和单位面积碳排放进行时空模拟;最后运用人口规模和土地面积对分省碳排放进行估算。估算结果显示:(1)整体来看,2000—2013年年均碳排放模拟值与实际值6.3349×109t较为接近,两个模型的相对误差均在0.5%以内。(2)分年度来看,所有年份的相对误差均在5%以内,2006年分省加总碳排放模拟值与实际碳排放6.2036×109t最为接近,绝对误差和相对误差均较小,两个模型模拟值的相对误差均为0.04%。(3)分省域来看,2000—2013年年均碳排放模拟值与实际碳排放均非常接近,除海南和宁夏外,其余28个省区市的相对误差均在1%以内。(4)分年度分省域来看,以2013年为例,40%省份的相对误差在2%以内,70%省份的相对误差在5%以内。从整体、分年度、分省域、分年度分省域的估算结果来看,基于稳定夜间灯光数据的中国分省碳排放时空模拟效果良好。因此,运用卫星夜间灯光数据可以较为准确地对中国分省碳排放进行估算和预测,为卫星遥感影像数据服务分省碳排放监测和评估提供一种补充性参考。     

Boosted Regression Tree Models to Explain Watershed Nutrient Concentrations and Biological Condition

Boosted regression tree (BRT) models were developed to quantify the nonlinear relationships between landscape variables and nutrient concentrations in a mesoscale mixed land cover watershed during base‐flow conditions. Factors that affect instream biological components, based on the Index of Biotic Integrity (IBI), were also analyzed. Seasonal BRT models at two spatial scales (watershed and riparian buffered area [RBA]) for nitrite‐nitrate (NO₂‐NO₃), total Kjeldahl nitrogen, and total phosphorus (TP) and annual models for the IBI score were developed. Two primary factors — location within the watershed (i.e., geographic position, stream order, and distance to a downstream confluence) and percentage of urban land cover (both scales) — emerged as important predictor variables. Latitude and longitude interacted with other factors to explain the variability in summer NO₂‐NO₃ concentrations and IBI scores. BRT results also suggested that location might be associated with indicators of sources (e.g., land cover), runoff potential (e.g., soil and topographic factors), and processes not easily represented by spatial data indicators. Runoff indicators (e.g., Hydrological Soil Group D and Topographic Wetness Indices) explained a substantial portion of the variability in nutrient concentrations as did point sources for TP in the summer months. The results from our BRT approach can help prioritize areas for nutrient management in mixed‐use and heavily impacted watersheds.     

“We cannot let this happen again”: reversing UK flood policy in response to the Somerset Levels floods, 2014

The Multiple Streams Framework offers a theoretical account of how policy proposals move from latent possibilities to becoming favored for implementation. We apply this framework in the context of the policy response to the 2013–2014 flooding of the Somerset Levels and Moors. Stakeholder interviews and analysis of news media coverage evidence the way in which a specific policy option that had fallen out of favor with the national Environment Agency – dredging – came to the fore and was eventually adopted during the period in which the conjunction of problem, policy, and political pressures came to a head. Local political activists mobilized a wider campaign with the help of social media and capitalized on national political sensitivities to successfully promote dredging. What is less clear is the longevity of the policy reversal, given funding constraints.     

Using multivariate regression modeling for sampling and predicting chemical characteristics of mixed waste in old landfills

Municipal solid waste landfills pose a threat on environment and human health, especially old landfills which lack facilities for collection and treatment of landfill gas and leachate. Consequently, missing information about emission flows prevent site-specific environmental risk assessments. To overcome this gap, the combination of waste sampling and analysis with statistical modeling is one option for estimating present and future emission potentials. Optimizing the tradeoff between investigation costs and reliable results requires knowledge about both: the number of samples to be taken and variables to be analyzed.This article aims to identify the optimized number of waste samples and variables in order to predict a larger set of variables. Therefore, we introduce a multivariate linear regression model and tested the applicability by usage of two case studies. Landfill A was used to set up and calibrate the model based on 50 waste samples and twelve variables. The calibrated model was applied to Landfill B including 36 waste samples and twelve variables with four predictor variables.The case study results are twofold: first, the reliable and accurate prediction of the twelve variables can be achieved with the knowledge of four predictor variables (Loi, EC, pH and Cl). For the second Landfill B, only ten full measurements would be needed for a reliable prediction of most response variables. The four predictor variables would exhibit comparably low analytical costs in comparison to the full set of measurements. This cost reduction could be used to increase the number of samples yielding an improved understanding of the spatial waste heterogeneity in landfills.Concluding, the future application of the developed model potentially improves the reliability of predicted emission potentials. The model could become a standard screening tool for old landfills if its applicability and reliability would be tested in additional case studies.     

Predictive Model of Avian Electrocution Risk on Overhead Power Lines

Electrocution on overhead power structures negatively affects avian populations in diverse ecosystems worldwide, contributes to the endangerment of raptor populations in Europe and Africa, and is a major driver of legal action against electric utilities in North America. We investigated factors associated with avian electrocutions so poles that are likely to electrocute a bird can be identified and retrofitted prior to causing avian mortality. We used historical data from southern California to identify patterns of avian electrocution by voltage, month, and year to identify species most often killed by electrocution in our study area and to develop a predictive model that compared poles where an avian electrocution was known to have occurred (electrocution poles) with poles where no known electrocution occurred (comparison poles). We chose variables that could be quantified by personnel with little training in ornithology or electric systems. Electrocutions were more common at distribution voltages (≤33 kV) and during breeding seasons and were more commonly reported after a retrofitting program began. Red‐tailed Hawks (Buteo jamaicensis) (n = 265) and American Crows (Corvus brachyrhynchos) (n = 258) were the most commonly electrocuted species. In the predictive model, 4 of 14 candidate variables were required to distinguish electrocution poles from comparison poles: number of jumpers (short wires connecting energized equipment), number of primary conductors, presence of grounding, and presence of unforested unpaved areas as the dominant nearby land cover. When tested against a sample of poles not used to build the model, our model distributed poles relatively normally across electrocution‐risk values and identified the average risk as higher for electrocution poles relative to comparison poles. Our model can be used to reduce avian electrocutions through proactive identification and targeting of high‐risk poles for retrofitting. Modelo Predictivo del Riesgo de Electrocución de Aves en Líneas Eléctricas Elevadas     

基于GM-SVR的小样本条件下化工设备可靠性预测

为了准确预测化工设备可靠性趋势，针对化工设备失效寿命数据为小样本的情形，基于灰色估计法与支持向量回归机在小样本数据处理中的优势，建立了失效寿命时间服从三参数威布尔分布的化工设备可靠性模型；结合GM（1，1）和SVR对模型进行参数估计，在压缩机可靠性分析中进行了实例应用，对比分析了最小二乘法、灰色估计法和GM-SVR的估计效果。研究结果表明:GM-SVR对威布尔分布参数的估计精度明显优于最小二乘法和灰色估计法，可以有效地应用于化工设备失效数据为小样本时的可靠性预测。