PARAMETRIC/NONPARAMETRIC MIXTURE DENSITY ESTIMATION WITH APPLICATION TO FLOOD-FREQUENCY ANALYSIS1

Much attention has been invested in the model choice problem for peak annual flows, in the context of flood frequency analysis. The authors would sidestep this dilemma through non-parametric density estimation methodology, but recognize that the standard nonparametric estimators preclude the use of prior information and related data, and furthermore have virtually no tail at all. Here we offer a remedy for these inadequacies by introducing an estimator which mixes parametric and nonparametric density estimates. We prove that our mixture rule is consistent. By this procedure, we do allow incorporation of prior information, experience, and regional data information, but nevertheless provide a safeguard against incorrect model choice.     

EFFECTS OF IRRIGATION ON WATER QUALITY OF A SHALLOW UNCONFINED AQUIFER1

ABSTRACT: The ground water quality of a shallow unconfined aquifer was monitored before and after implementation of a border strip irrigation scheme, by taking monthly samples from an array of 13 shallow wells. Two 30 m deep wells were sampled to obtain vertical concentration profiles. Marked vertical, temporal, and spatial variabilities were recorded. The monthly data were analyzed for step and linear trends using nonparametric tests that were adjusted for the effects of serial correlation. Average nitrate concentrations increased in the preirrigation period and decreased after irrigation began. This was attributed to wetter years in 1978–1979 than in 1976–1977 which increased leaching, and to disturbance of the topsoil during land contouring before irrigation, followed by excessive drainage after irrigation. Few significant trends were recorded for other determinants, possibly because of shorter data records. Nitrate, sulphate, and potassium concentrations decreased with depth, whereas sodium, calcium, bicarbonate, and chloride concentrations increased. These trends allowed an estimation to be made of the depth of ground water affected by percolating drainage. This depth increased during the irrigation season and after periods of winter recharge. Furthermore, an overall increase in the depth of drainage-affected ground water occurred with time, which paralleled the development of the irrigation scheme.     

The use of computer models in planning upland land use

The British Ecological Society has suggested that computer‐based techniques could be used in the coordination of policies across different land uses in upland planning. This paper looks briefly at one such technique — a linear programming model — and describes its use at a regional scale for the Sedburgh area in north‐west England. The paper concludes with consideration of why such models are not in more common use.     

SENSITIVITY ANALYSIS OF A DRAINAGE SIMULATION MODEL1

A sensitivity analysis of a computer model, simulating major water and nitrogen processes of a soil-water-plant-climatic system on an annual basis, was conducted to determine how the model reacts to the variations in selected hydrologic and nitrogen parameters. Two major output variables (namely, total subsurface drain volume and cumulative nitrate loss with subsurface drain water) were selected for the sensitivity analysis. Model sensitivity analysis shows that the model is most sensitive to hydrologic parameters. The model is very sensitive to variations in the initial water content in the soil profile.     

Assessing Potential for Recovery of Biotic Richness and Indicator Species due to Changes in Acidic Deposition and Lake ph in Five Areas of Southeastern Canada

Biological damage to sensitive aquatic ecosystems is among the most recognisable, deleterious effects of acidic deposition. We compiled a large spatial database of over 2000 waterbodies across southeastern Canada from various federal, provincial and academic sources. Data for zooplankton, fish, macroinvertebrate (benthos) and loon species richness and occurrence were used to construct statistical models for lakes with varying pH, dissolved organic carbon content and lake size. pH changes, as described and predicted using the Integrated Assessment Model (Lam et al., 1998; Jeffries et al., 2000), were based on the range of emission reductions set forth in the Canada/US Air Quality Agreement (AQA). The scenarios tested include 1983, 1990, 1994 and 2010 sulphate deposition levels. Biotic models were developed for five regions in southeastern Canada (Algoma, Muskoka, and Sudbury, Ontario, southcentral Québec, and Kejimkujik, Nova Scotia) using regression tree, multiple linear regression and logistic regression analyses to make predictions about recovery after emission reductions. The analyses produced different indicator species in different regions, although some species showed consistent trends across regions. Generally, the greatest predicted recovery occurred during the final phase of emission reductions between 1994 and 2010 across all taxonomic groups and regions. The Ontario regions, on average, were predicted to recover to a greater extent than either southcentral Québec or the Kejimkujik area of Nova Scotia. Our results reconfirm that pH 5.5–6.0 is an important threshold below which damage to aquatic biota will remain a major local and regional environmental problem. This damage to biodiversity across trophic levels will persist well into the future if no further reductions in sulphate deposition are implemented.     

JSEM: A Framework for Identifying and Evaluating Indicators

There are two issues in indicator development that have not been adequately addressed: (1) how to select an optimal combination of potentially redundant indicators that together best represent an endpoint, given cost constraints; (2) how to identify and evaluate indicators when the endpoint is unmeasured. This paper presents an approach to identifying and evaluating combinations of indicators when the mathematical relationships between the indicators and an endpoint may not be quantified, a limitation common to many ecological assessments. The approach uses the framework of Structural Equation Modeling (SEM), which combines path analysis withmeasurement models, to formalize available informationabout potential indicators and to evaluate their potential adequacy for representing an endpoint. Unlike traditional applications of SEM which require data on all variables, our approach – judgement-based SEM (JSEM) – can utilize expert judgement regarding the strengths and shapes of indicator-endpoint relationships. JSEM is applied in two stages. First, a conceptual model that relates variables in a network of direct and indirect linkages is developed, and is used to identify indicators relevant to an endpoint. Second, an index of indicator strength – i.e., the strength of the relationship between the endpoint and a set of indicators – is calculated from estimates of correlation between the modeled variables, and is used to compare alternative sets of indicators. The second stage is most appropriate for large, long-term assessments. Although JSEM is not a statistical technique, basing JSEM on SEM provides a structure for validating the conceptual model and for refining the index of indicator strength as data become available. Our main objective is to contribute to a rigorous and consistent selection of indicators even when knowledgeabout the ability of indicators to represent an endpoint is limited to expert judgement.     

Limitations of multimedia models for use in environmental decision making

The United States currently is engaged in a complex,multi-billion dollar effort to cleanup a legacy ofboth privately- and federally-owned hazardous wastesites. Decisions regarding the best approach forremediation of these sites often are based on theanalysis of potential risks to human health and theenvironment. A cornerstone of such analysis is thefrequent use of computerized multimedia environmentaltransport models, to evaluate the large quantities ofinformation necessary to understand the present andfuture implications of contamination at a site. Onebarrier to wide-spread use of this analyticalprocedure is the view that results obtained usingcomputer models are highly dependent on user input,and therefore, subject to manipulation. It is widelyrecognized that for decisions to be both credible andimplementable, the public must have confidence in boththe scientific basis for judgments involved and thedecision processes employed (NRC, 1983). Our purposein this article is to overview the difficultiesassociated with application of multimedia models toreal world problems and the contribution these modelscan make to technically sound estimates of exposure and risk.     

国外城市非点源径流水质模型简介

随着人们环保意识的提高。城市中的点源污染已大部分得到了控制，于是非点源污染变得越来越突出,越来越严重．并引起了城市环保者的关注。借此。本文介绍了国外四种最主要的城市非点源污染的管理模型,以供我国城市环保人员在对非点源污染进行管理时作为参考。     

Learning Machines Applied to Potential Forest Distribution

The clearing of forests to obtain land for pasture and agriculture and the replacement of autochthonous species by other faster-growing varieties of trees for timber have both led to the loss of vast areas of forest worldwide. At present, many developed countries are attempting to reverse these effects, establishing policies for the restoration of older woodland systems. Reforestation is a complex matter, planned and carried out by experts who need objective information regarding the type of forest that can be sustained in each area. This information is obtained by drawing up feasibility models constructed using statistical methods that make use of the information provided by morphological and environmental variables (height, gradient, rainfall, etc.) that partially condition the presence or absence of a specific kind of forestation in an area. The aim of this work is to construct a set of feasibility models for woodland located in the basin of the River Liébana (NW Spain), to serve as a support tool for the experts entrusted with carrying out the reforestation project. The techniques used are multilayer perceptron neural networks and support vector machines. Their results will be compared to the results obtained by traditional techniques (such as discriminant analysis and logistic regression) by measuring the degree of fit between each model and the existing distribution of woodlands. The interpretation and problems of the feasibility models are commented on in the Discussion section.