Abstract: Spatially explicit information on the species composition and structure of forest vegetation is needed at broad spatial scales for natural resource policy analysis and ecological research. We present a method for predictive vegetation mapping that applies direct gradient analysis and nearest-neighbor imputation to ascribe detailed ground attributes of vegetation to each pixel in a digital landscape map. The gradient nearest neighbor method integrates vegetation measurements from regional grids of field plots, mapped environmental data, and Landsat Thematic Mapper (TM) imagery. In the Oregon coastal province, species gradients were most strongly associated with regional climate and geographic location, whereas variation in forest structure was best explained by Landsat TM variables. At the regional level, mapped predictions represented the range of variability in the sample data, and predicted area by vegetation type closely matched sample-based estimates. At the site level, mapped predictions maintained the covariance structure among multiple response variables. Prediction accuracy for tree species occurrence and several measures of vegetation structure and composition was good to moderate. Vegetation maps produced with the gradient nearest neighbor method are appropriately used for regional-level planning, policy analysis, and research, not to guide local management decisions. Résumé: Afin d’effectuer l’analyse des politiques touchant les ressources naturelles et appuyer la recherche écologique, il est nécessaire d’obtenir une information spatiale précise sur la structure de la végétation forestière et sur la composition des espèces et ce, à une vaste échelle spatiale. Nous présentons une méthode de cartographie prévisionnelle

1 Canonical community ordination comprises a collection of methods that relate species assemblages to their environment, in both observational studies and designed experiments. Canonical ordination differs from ordination sensu stricto in that species and environment data are analyzed simultaneously. Part I reviews the theory in a non-mathematical way with emphasis on new insights for the interpretation of ordination diagrams. The interpretation depends on the ordination method used to create the diagram. After the basic theory, Part I is focused on the ordination diagrams in linear methods of canonical community ordination, in particular principal components analysis, redundancy analysis and canonical correlation analysis. Special attention is devoted to the display of qualitative environmental variables. Key words: principal components analysis, redundancy analysis, canonical correlation analysis, biplot, ordination diagram, species-environment relations. 2

Classical multidimensional scaling (MDS) is a method for visualizing high-dimensional point clouds by mapping to low-dimensional Euclidean space. This mapping is defined in terms of eigenfunctions of a matrix of interpoint dissimilarities. In this paper we analyze in detail multidimensional scaling applied to a specific dataset: the 2005 United States House of Representatives roll call votes. Certain MDS and kernel projections output “horseshoes” that are characteristic of dimensionality reduction techniques. We show that, in general, a latent ordering of the data gives rise to these patterns when one only has local information. That is, when only the interpoint distances for nearby points are known accurately. Our results provide a rigorous set of results and insight into manifold learning in the special case where the manifold is a curve. 1. Introduction. Classical

Authors Reliable predictions of how changing climate and disturbance regimes will affect forest ecosystems are crucial for effective forest management. Current fire and climate research in forest ecosystem and community ecology offers data and methods that can inform such predictions. However, research in these fields occurs at different scales, with disparate goals, methods, and context. Often results are not readily comparable among studies and defy integration. We discuss the strengths and weaknesses of three modeling paradigms: empirical gradient models, mechanistic ecosystem models, and stochastic landscape disturbance models. We then propose a synthetic approach to multi-scale analysis of the effects of climatic change and disturbance on forest ecosystems. Empirical gradient models provide an anchor and spatial template for stand-level forest ecosystem models by quantifying key parameters for individual species and accounting for broad-scale geographic variation among them. Gradient imputation transfers predictions of fine-scale forest composition and structure across geographic space. Mechanistic ecosystem dynamic models predict the responses of biological variables to specific environmental drivers and facilitate understanding of temporal dynamics and disequilibrium. Stochastic landscape dynamics models predict frequency, extent, and severity of broad-scale disturbance. A robust linkage of these three modeling paradigms will facilitate prediction of the effects of altered fire and other disturbance regimes on forest ecosystems at multiple scales and in the context of climatic variability and change.

Abstract not found

OF THE DISSERTATION A Markov Chain Monte Carlo Method for Approximating 2-Way Contingency Tables with Applications in the Stability Analysis of Ecological Ordination by Stanley S. Bentow Doctor of Philosophy in Statistics University of California, Los Angeles, 1999 Professor N. Donald Ylvisaker, Chair This dissertation develops a Markov Chain Monte Carlo method for approximating 2-way contingency tables with an eye toward assessing the stability of ecological ordination. Ecology is a part of biology that deals with the interrelationships between populations, communities and ecosystems and their environment. It draws on knowledge from many other disciplines such as climatology, physical geography, agronomy, and pedology [52]. Odum [75] prefers the de nition \ Ecology is the study of structure and function of nature," and stresses the role of ecosystem research in relation to the use of nature by man. Krebs [58] prefers to think of Ecology as the scienti c study of the interactions t...

Abstract: In this paper, we examine possible sources of hierarchical (nested) structure in vegetation data. We then use the Minimum Message length principle to provide a rational means of comparing hierarchical and non-hierarchical clustering. The results indicate that, with the data used, a hierarchical solution was not as efficient as a nonhierarchical one. However, the hierarchical solution seems to provide a more comprehensible solution, separating first isolated types, probably caused from unusual contingent events, then subdividing the more diverse areas before finally subdividing the less diverse. By presenting this in 3 stages, the complexity of the non-hierarchical result is avoided. The result also suggests that a hierarchical analysis may be useful in determining ‘homogeneous ’ areas. Abbreviatons: MML- Minimum Message Length; MUAP- Modifiable unit area problem.

geodynamic models of continental deformation and inform interpretations of landscape and climate evolution. One widely used, paleobotanical approach reconstructs paleoelevation from the difference in estimated atmospheric enthalpy between a known sea level and a targeted, coeval, elevated fossil floral site. Enthalpy is estimated using Climate‐Leaf Analysis Multivariate Program (CLAMP) on 31 leaf size and shape variables that have been calibrated in living forests. Errors related to CLAMP are significantly greater than often reported, and there are many sources of large potential error related to this method that are either difficult to quantify or unquantifiable and are thus not documented. Here, we quantify one significant bias, toward underestimationofleafareaintheCLAMPdataset(∼50%), that affects all CLAMP climate estimates, including enthalpy. Crucially, errors in paleoelevation when the leaf size bias is included are in the range of ±2 km or more, at least 2 times the previous estimates, and exceeding the plausible paleoelevations of many fossil sites. Previously published paleoelevations derived from this technique are unlikely to be accurate either in magnitude

Despite being conspicuous and influential features of the biosphere, urban ecosystems have been neglected in ecological research. Arthropods are abundant in urban settings, but little is known about how these animals respond to urbanization. We systematically monitored the structure of ground arthropod communities for 12 months at 16 sites representing the four most abundant forms of urban land use (residential, industrial, agricultural, and desert remnant) in a rapidly growing metropolitan area (Phoenix, AZ). Although taxonomic richness was comparable among land-use types, community composition differed, with certain taxa being uniquely associated with each form of land use. Three taxa (springtails, ants, and mites) were extremely widespread and abundant, accounting for over 92 % of captures; when these three taxa were excluded from analysis, however, differences were revealed in arthropod community composition with urban land use. Trophic dynamics also varied with land use: predators, herbivores, and detritivores were most abundant in agricultural sites, while omnivores were equally abundant in all forms of land use. These community-level differences resulted from taxon-specific responses to habitat structure, which varied with land use. Because arthropod community structure is affected by habitat structure and land use, and because arthropods play key roles in nutrient cycling, organic matter decomposition, pollination, and soil aeration, the spatial heterogeneity of urban ecosystems therefore may affect ecosystem functioning. (Q) 2001 Elsevier Science B. V. All rights reserved.

approaches using macroinvertebrates. Doctoral dissertation. ISSN 1401-6230, ISBN 91-576-6716-0. Running water habitats are among the most precious, yet most threatened, ecosystems on earth. Hence, there is a need for reliable methods for detecting the effects of pollution on this valuable commodity. This thesis examined a number of different approaches commonly used in bioassessment of stream ecosystems for their ability to detect ecological change. In particular, focus was on assessing the utility of single metric, multimetric and multivariate approaches using macroinvertebrate communities, with organic pollution and acidity as stress gradients. For detecting the effects of organic pollution, two metrics (the DJ index and the CA scores) representing two different approaches (multimetric and the multivariate) were found to be reliable tools for detecting the effects of stress in streams of southern Sweden. These methods were sensitive (high coefficients of variation) and had high precision (low error) to the stress gradient. Notwithstanding, single metric approaches might also be used when the knowledge of multimetric or multivariate methods is not available. The saprobic index was one such metric that showed promise as a tool for