Modern Spatiotemporal Geostatistics

This scholarly introductory treatment explores the fundamentals of modern geostatistics, viewing them as the product of the advancement of the epistemic status of stochastic data analysis. The book's main focus is the Bayesian maximum entropy approach for studying spatiotemporal distributions of natural variables, an approach that offers readers a deeper understanding of the role of geostatistics in improved mathematical models of scientific mapping. Starting with a overview of the uses of spatiotemporal mapping in the natural sciences, the text explores spatiotemporal geometry, the epistemic paradigm, the mathematical formulation of the Bayesian maximum entropy method, and analytical expressions of the posterior operator. Additional topics include uncertainty assessment, single- and multi-point analytical formulations, and popular methods. An innovative contribution to the field of space and time analysis, this volume offers many potential applications in epidemiology, geography, biology, and other fields.

Originally published: Oxford; New York: Oxford University Press, 2000.

The book focuses on the development of advanced functions for field-based temporal geographical information systems (TGIS). These fields describe natural, epidemiological, economical, and social phenomena distributed across space and time. The book is organized around four main themes: "Concepts, mathematical tools, computer programs, and applications". Chapters I and II review the conceptual framework of the modern TGIS and introduce the fundamental ideas of spatiotemporal modelling. Chapter III discusses issues of knowledge synthesis and integration. Chapter IV presents state-of-the-art mathematical tools of spatiotemporal mapping. Links between existing TGIS techniques and the modern Bayesian maximum entropy (BME) method offer significant improvements in the advanced TGIS functions. Comparisons are made between the proposed functions and various other techniques (e.g., Kriging, and Kalman-Bucy filters). Chapter V analyzes the interpretive features of the advanced TGIS functions, establishing correspondence between the natural system and the formal mathematics which describe it. In Chapters IV and V one can also find interesting extensions of TGIS functions (e.g., non-Bayesian connectives and Fisher information measures). Chapters VI and VII familiarize the reader with the TGIS toolbox and the associated library of comprehensive computer programs. Chapter VIII discusses important applications of TGIS in the context of scientific hypothesis testing, explanation, and decision making.

Statistical Methods for Spatial and Spatio-Temporal Data Analysis provides a complete range of spatio-temporal covariance functions and discusses ways of constructing them. This book is a unified approach to modeling spatial and spatio-temporal data together with significant developments in statistical methodology with applications in R. This book includes: Methods for selecting valid covariance functions from the empirical counterparts that overcome the existing limitations of the traditional methods. The most innovative developments in the different steps of the kriging process. An up-to-date account of strategies for dealing with data evolving in space and time. An accompanying website featuring R code and examples

This volume contains selected contributions from geoENV III - the Third European Conference on Geostatistics for Environmental Sciences, held in Avignon, France in November 2000. This third book of the geoENV series illustrates the new methodological developments in geostatistics, as applied to environmental sciences, which have occurred during the last two years. It also presents a wide variety of practical environmental applications which will be of interest to both researchers and practitioners. The book starts with two keynote papers on hydrogeology and on climatology and atmospheric pollution, followed by forty contributions. The content of this book is foremost practical. The editors have endeavored to compile a set of papers in which the readers could perceive how geostatistics is applied within environmental sciences. A few selected methodological and theoretical contributions are also included. The papers are organised in the following sections: Air Pollution / Climate; Environment; Health / Ecology; Hydrology; Methods; Soil Science / Site Remediation. presenting applications varying from delineation of hazardous areas, monitoring water quality, space-time modeling of sand beaches, areal rainfall estimation, air pollution monitoring, multivariate conditional simulation, soil texture analysis, fish abundance analysis, tree productivity index estimation, radionuclide migration analysis, wombling procedure, tracer tests modeling, direct sequential co-simulation to stochastic modeling of flow and transport. Audience: This publication will be of great interest and practical value to geostatisticians working both in academia and in industry.

Spatio-temporal Analysis of Extreme Hydrological Events offers an extensive view of the experiences and applications of the latest developments and methodologies for analyzing and understanding extreme environmental and hydrological events. The book addresses the topic using spatio-temporal methods, such as space-time geostatistics, machine learning, statistical theory, hydrological modelling, neural network and evolutionary algorithms. This important resource for both hydrologists and statisticians interested in the framework of spatial and temporal analysis of hydrological events will provide users with an enhanced understanding of the relationship between magnitude, dynamics and the probability of extreme hydrological events. Presents spatio-temporal processes, including multivariate dynamic modelling Provides varying methodological approaches, giving the readers multiple hydrological modelling information to use in their work Includes a variety of case studies making the context of the book relatable to everyday working situations

Statistical Methods for Spatio-Temporal Systems presents current statistical research issues on spatio-temporal data modeling and will promote advances in research and a greater understanding between the mechanistic and the statistical modeling communities. Contributed by leading researchers in the field, each self-contained chapter starts with an introduction of the topic and progresses to recent research results. Presenting specific examples of epidemic data of bovine tuberculosis, gastroenteric disease, and the U.K. foot-and-mouth outbreak, the first chapter uses stochastic models, such as point process models, to provide the probabilistic backbone that facilitates statistical inference from data. The next chapter discusses the critical issue of modeling random growth objects in diverse biological systems, such as bacteria colonies, tumors, and plant populations. The subsequent chapter examines data transformation tools using examples from ecology and air quality data, followed by a chapter on space-time covariance functions. The contributors then describe stochastic and statistical models that are used to generate simulated rainfall sequences for hydrological use, such as flood risk assessment. The final chapter explores Gaussian Markov random field specifications and Bayesian computational inference via Gibbs sampling and Markov chain Monte Carlo, illustrating the methods with a variety of data examples, such as temperature surfaces, dioxin concentrations, ozone concentrations, and a well-established deterministic dynamical weather model.

Our living environment continuously changes in space and time. This book explains how to capture and assess these changes through the relevant statistical framework. It is a useful guide to students, teachers and researchers in the fields of biology, ecology and environmental science. Codes on the accompanying CD-ROM aid analyses.