Dynamic Modeling of Ground‐Water Quality Using Bayesian Techniques1

Abstract: Water industry experts have been arguing that the traditional techniques are not an accurate means of measuring water contamination. This is mainly because these techniques emphasize neither the stochastic nature of the water contamination process nor the precision and the accuracy of the tested methods used by environmental laboratories. In this work, we describe the development and application of prototype Dynamic Bayesian Networks (DBNs) that model ground‐water quality to determine the impact of chemical contaminants on ground‐water quality in the Salalah area, which is allocated to the south of Oman. We also present a new technique for data pre‐processing because it is needed for the treatment of ground‐water datasets that are used as the data source to learn the probabilities for dynamic decision models. Among more than 20 wells in area, only four wells were selected to be analyzed and the results show that we achieved an acceptable level of efficiency.     

Lessons learned from recent incidents: Facility siting, atmospheric venting, and operator information systems

Recent incidents have focused attention on a number of technical and management systems that need to be addressed by industry. A multiple layer of protection approach is essential for the prevention of incidents and/or reduction of consequences. Safety culture and operational discipline are the overall embracing factors that influence the safety performance of a facility. However, as recent events have indicated, there are a number of technical and engineering issues that must also be developed and implemented appropriately. Some of these issues that could lead to incidents with catastrophic consequences include facility siting and atmospheric relief venting. Impact of operator information systems on the prevention of releases of hazardous materials from their containment is also another significant factor that should be given appropriate attention.

This paper describes these three topics based on the findings from recent incidents and historical data. Engineering standards, regulatory requirements, and industry practices are discussed for facility siting, atmospheric relief venting, and operator information systems. Finally, a summary of gaps and needs in technology, standards, and practices is presented.     

Leading indicators of safety in virtual organizations

A primary purpose in measuring safety is to develop intervention strategies to avoid future accidents. Recognizing signals before an accident occurs offers the potential for improving safety, and many organizations have sought to develop programs to identify and benefit from alerts, signals and prior indicators. In this paper, we address the challenge of identifying and evaluating leading indicators of safety in virtual organizations–organizations comprised of multiple, distributed members, temporarily linked together for competitive advantage, that share common value chains and business processes supported by distributed information technology. We begin by discussing risk propensity in virtual organizations and leading indicators of safety. We then describe a pilot study to identify leading indicators for one safety-critical system, and use the results of that study and the literature just described to propose an approach to developing leading indicators in virtual organizations.     

INTEGRATING GEOGRAPHIC INFORMATION SYSTEMS IN GROUND-WATER APPLICATIONS USING NUMERICAL MODELING TECHNIQUES1

ABSTRACT: The Sand and Gravel Aquifer is the sole source of potable water in Escambia County, Florida. In order to better understand the hydraulics of the aquifer, a numerical computer model of the aquifer was developed. The model applied a finite element technique which allowed for density-dependent transport and flow in three dimensions. The modeling technique was integrated with GIS to develop a system for optimal management of the resource. The GIS was the primary tool in the development of the model grid, as well as being the integral component in the modeling procedure. Multiple model grids were developed for simulating regional flow and local flow/transport phenomena. The model grids were generated by the GIS where nodal and element sequencing were recorded. The grid topology was stored in the GIS with the element numbers, node numbers, and the related hydrogeologic attributes. The Triangulated Irregular Network (TIN) module was used for transferring interpolated value between GIS coverages. TIN allowed a fit of the model grid to the physical dimensions of the aquifer and for interpolating boundary values for telescopically refined grids. Calculations between TIN surfaces provided the residuals of the dependent variable from observed TIN surfaces. Model calibration was conducted within the GIS environment through a combination of visual and relational querying. The GIS provided an integrated environment which facilitated model analyses and data storage and retrieval.     

Integration of different data bodies for humanitarian decision support: an example from mine action

Geographic information systems (GIS) are increasingly used for integrating data from different sources and substantive areas, including in humanitarian action. The challenges of integration are particularly well illustrated by humanitarian mine action. The informational requirements of mine action are expensive, with socio-economic impact surveys costing over US$1.5 million per country, and are feeding a continuous debate on the merits of considering more factors or 'keeping it simple'. National census offices could, in theory, contribute relevant data, but in practice surveys have rarely overcome institutional obstacles to external data acquisition. A positive exception occurred in Lebanon, where the landmine impact survey had access to agricultural census data. The challenges, costs and benefits of this data integration exercise are analysed in a detailed case study. The benefits are considerable, but so are the costs, particularly the hidden ones. The Lebanon experience prompts some wider reflections. In the humanitarian community, data integration has been fostered not only by the diffusion of GIS technology, but also by institutional changes such as the creation of UN-led Humanitarian Information Centres. There is a question whether the analytic capacity is in step with aggressive data acquisition. Humanitarian action may yet have to build the kind of strong analytic tradition that public health and poverty alleviation have accomplished.     

EDUCATING RURAL COMMUNITY LEADERS IN WASTEWATER TREATMENT ALTERNATIVES1

ABSTRACT: Providing adequate wastewater treatment facilities for the citizens in small rural communities in the United States can cost millions of dollars. An educational program has been developed and conducted in Ohio to teach local officials about wastewater treatment alternatives. The five-session workshop covers wastewater treatment principles, needs assessment, system management, and resource gathering. The knowledge and skills gained by the 96 community leaders who participated in six workshops were measured using a pretest/pest test and surveys conducted during the workshops. From 25 percent to 100 percent of the participants increased their knowledge of treatment systems, alternatives, and relative costs. From 23 percent to 84 percent applied new skills, such as conducting a sanitary survey or examining treatment systems. The pretest also revealed previous knowledge of the participants in the areas of system maintenance and consultant qualifications.     

COMPARATIVE EVALUATION OF LAND COVER DATA SOURCES FOR EROSION PREDICTION1

ABSTRACT: A fundamental problem in protecting surface drinking water supplies is the identification of sites highly susceptible to soil erosion and other forms of nonpoint source (NPS) pollution. The New York City Department of Environmental Protection is trying to identify erodible sites as part of a program aimed at avoiding costly filtration. New York City's 2,000 square mile watershed system is well suited for analysis with geographic information systems (GIS); an increasingly important tool to determine the spatial distribution of sensitive NPS pollution areas. This study used a GIS to compare three land cover sources for input into the Modified Universal Soil Loss Equation (MUSLE), a model estimating soil loss from rangeland and forests, for a tributary watershed within New York City's water supply system. Sources included both conventional data (aerial photography) and Landsat data (MSS and TM images). Although land cover classifications varied significantly across these sources, location-specific and aggregate watershed predictions of the MUSLE were very similar. We conclude that using Landsat TM imagery with a hybrid classification algorithm provides a rapid, objective means of developing large area land cover databases for use in the MUSLE, thus presenting an attractive alternative to photo interpretation.     

The value content of agricultural technologies and their effect on rural regions and farmers

The premise of this article is that technologies are not neutral in terms of their design objectives, their scale, and the fact that they reflect the physical and social environments in which they have developed. Specific agricultural technologies, the midwestern plow, the California tomato harvester, and various biotechnologies, are evaluated in these terms and shown to have generally predictable impacts upon rural regions and farmers. Finally, the article examines a series of major threats such as climate change that require the rethinking and restructuring of agriculture along more sustainable lines, something the can be accomplished only with a conscious recognition of the non-neutrality of technologies.This article was prepared for the conference Ethical Choice in an Age of Pervasive Technology, University of Guelph, October 25–29, 1989.     

GEOSTATISTICAL ESTIMATION OF HORIZONTAL HYDRAULIC CONDUCTIVITY FOR THE KIRKWOOD‐COHANSEY AQUIFER1

ABSTRACT: The Kirkwood‐Cohansey aquifer has been identified as a critical source for meeting existing and expected water supply needs for southern New Jersey. Several contaminated sites exist in the region; their impact on the aquifer has to be evaluated using ground water flow and transport models. Ground water modeling depends on availability of measured hydrogeologic data (e.g., hydraulic conductivity, for parameterization of the modeling runs). However, field measurements of such critical data have inadequate spatial density, and their locations are often clustered. The goal of this study was to research, compile, and geocode existing data, then use geostatistics and advanced mapping methods to develop a map of horizontal hydraulic conductivity for the Kirkwood‐Cohansey aquifer. Spatial interpolation of horizontal hydraulic conductivity measurements was performed using the Bayesian Maximum Entropy (BME) Method implemented in the BMELib code library. This involved the integration of actual measurements with soft information on likely ranges of hydraulic conductivity at a given location to obtain estimate maps. The estimation error variance maps provide an insight into the uncertainty associated with the estimates, and indicate areas where more information on hydraulic conductivity is required.     

MODELING DRINKING WATER QUALITY VIOLATIONS WITH BAYESIAN NETWORKS1

ABSTRACT: Compliance violations at community water systems are rare but represent significant human health risks. These risks are mediated by the decision schema of human operators at water treatment facilities. However, causal uncertainty among physical and human factors involved in water quality problems complicates assessment of their probability and severity. This study uses a probabilistic Bayesian network modeling approach to explore the causes of compliance violations in a sample of water treatment systems in Pennsylvania. The model presented here is one of several created by treatment system operators during an expert elicitation process. The expert model alone predicts violations poorly, suggesting that experts make inaccurate quantitative estimates. However, Bayesian networks are capable of combining the subjective expertise of treatment system operators with the objective compliance histories of the facilities they manage, and the expert model accurately predicts violations when trained with historical compliance data. Analysis of the trained network reveals those components of the treatment process, including environmental and system characteristics as well as operator decisions, that play the greatest role in determining the likelihood of major violation types. Among operator decisions, coagulant dosing and filter backwash frequency are the most important determinants of violation likelihood.