In lieu of an abstract, here is a brief excerpt of the content:

Glossary A Abiotic niche—The set of environments in which abiotic conditions are favorable for the species. In practice, largely equivalent to the scenopoetic niche. Contrast with biotically reduced niche. Abiotically suitable area—The geographic region where, in the absence of competitors and other negatively interacting species, and given unlimited dispersal abilities, the abiotic environment is favorable for the species. Contrast with occupied distributional area and potential distributional area. Absence data (see also presence/absence data)—Datasets containing “records” of places where sampling has occurred but the species has not been documented. Contrast with presence-only data. Algorithm—A specific sequence of instructions for solving a problem or developing a task. Examples of algorithms used to model species niches and distributions include BIOCLIM, desktop GARP, Maxent, and so on. Contrast with model. Ancillary data—As used here, additional factors not included in the modeling algorithms and that influence distributions of species beyond the simple effects of scenopoetic variables and bionomic variables. Examples include geographic barriers and historical events. Ancillary data normally are used for postprocessing results of an algorithm. See abiotically suitable area and occupied distributional area; contrast with metadata. Apparent commission error—A kind of commission error that is not real, but rather derives from misinformative evaluation data (see also absence data), inappropriate selection of the study region for evaluation, or both. See nonequilibrium distributions. Area under the curve (ⴝ AUC)—A statistic generated from a receiver operating characteristic plot (ROC), the area under the curve (AUC) represents an overall measure of model performance across all thresholds and strengths of a prediction. AUC is a nonparametric measure that ranges 0 –1 (random expectation of AUC is 0.5), and summarizes the model’s ability to rank presence records higher than absence records (or higher than a sample from the background, in the case of presence–background data); it does not evaluate the model’s goodness-of-fit. See model evaluation. Artifactual absence—A situation when a species is not truly absent, but rather the lack of a record is an artifact of inadequate or nonexistent sampling. See absence data. Asymmetric loss—A loss function that combines the omission error and commission error, but not with equal weight. Contrast with symmetric loss. AUC—See area under the curve. 270 GLOSSARY B Background data—Information on environmental variation across the study area (the “background”), whether or not sampling has occurred or whether or not the species of interest has been found there. See also presence/pseudoabsence data. BAM diagram—A Venn diagram that displays the joint fulfillment in geographic space (G-space) of three sets of conditions that together determine a species’ distribution: B, for biotic conditions; A, for abiotic conditions; and M, for movement of the species. Bias—See sampling bias. Binomial test—A test employed when each independent result is one of two possible outcomes. In the current context, it is often applied to determine whether evaluation data fall into regions of a binary geographic prediction (usually after applying a threshold to a continuous or an ordinal output) more often than expected by chance, constituting a one-tailed test of model significance. See model evaluation. Biogeographic regions—Portions of G-space delimited by patterns of spatial coincidence in the ranges (occupied distributional area) of large numbers of species. Biogeographic regions are usually related to current and/or past dispersal limitations, and to some degree environmental characteristics. Note that this concept differs markedly from that of a biome, which depends almost entirely on environmental characteristics and does not directly include contingent effects of history. Bionomic variables—Variables that are dynamically linked to the occurrence of a species , such as competitors, prey, predators. Contrast with scenopoetic variables, noticing that the distinction is not absolute. Biotic interactions—Interactions between and among species—for example, competition , mutualism, predation. See BAM diagram. Biotically reduced niche—The set of environments in which the abiotic environment is favorable for the species and in which negative interactors are not capable of excluding the populations of the species of interest. See potential distributional area; contrast with scenopoetic niche. Biotope—Equivalent to G-space, defined here as the geographic space composed of cells or pixels covering a particular region. See extent, grain, and study region. C Calibration—See model calibration. Calibration data—The primary occurrence data used to calibrate the model. See model calibration; contrast with evaluation data. Commission error—A measure of model performance based on the confusion matrix. As a rate that ranges from 0 to 1, it...

pdf

Additional Information

ISBN
9781400840670
Related ISBN
9780691136882
MARC Record
OCLC
761318478
Pages
328
Launched on MUSE
2015-01-01
Language
English
Open Access
No
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.