Can models of presence‐absence be used to scale abundance? Two case studies considering extremes in life history

• Published in: Ecography
• Main Authors: Nielsen, Scott E., Johnson, Chris J., Heard, Douglas C., Boyce, Mark S.
• Format: Article in Journal/Newspaper
• Language: English
• Published: Wiley 2005
• Subjects: Alces alces
• Online Access: http://dx.doi.org/10.1111/j.0906-7590.2005.04002.x; https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1111%2Fj.0906-7590.2005.04002.x; https://onlinelibrary.wiley.com/doi/pdf/10.1111/j.0906-7590.2005.04002.x

Understanding patterns of species occurrence and abundance is a central theme of ecology, natural resource management, and conservation. Although occurrence models have been widely used for describing species distribution, particularly for rare species, abundance models are less common, despite greater information for conservation and management. Because presence‐absence data are easier and less expensive to collect, predictions of abundance from patterns of occurrence could prove useful. We examined the relationship between occurrence and abundance for two species with very different life histories: bracken fern Pteridium aquilinum and moose Alces alces . We predicted that if occurrence and abundance were functionally related we should observe: 1) correlation between predicted probability of occurrence and observed abundance; 2) similar environmental covariates and estimated coefficients for occurrence models developed separately for low‐density, high‐density, and global data sites; and 3) parallel coefficients for the occurrence and abundance components of zero‐inflated count models. Probability of occurrence was not correlated with abundance‐when‐present for bracken fern, while evidence for a relationship for moose was apparent at densities of animals below 7 individuals per cutblock. Coefficients for models at different levels of density did not vary significantly. However, once occurrence was accounted for, measured environmental data appeared less important in describing abundance. For bracken, covariates of zero‐inflated count models differed in their expression of occurrence and abundance. Differences were less extreme for moose; however, results from the two‐process models suggest that distribution and abundance may be a function of different processes. Environmental factors influencing abundance may differ from those limiting distribution. Life history, scale, site history, and socio‐competitive processes further help shape patterns of abundance. Two‐stage modeling provides a powerful tool for ...