Current caveats and further directions in the analysis of density-dependent population regulation.

An important issue in population ecology is to disentangle different density-dependent mechanisms that may limit or regulate animal populations. This goal is further complicated when studying long-lived species for which experimental approaches are not feasible, in whose cases density-dependence hypotheses are tested using long-term monitored populations. Here we respond to some criticisms and identify additional problems associated with these kinds of observational studies. Current caveats are related to the temporal and spatial scales covered by population monitoring data, which may question its suitability for density-dependence tests, and to statistical flaws such as the incorrect control for confounding variables, low statistical power, the distribution of demographic variables, the interpretation of spurious correlations, and the often used stepwise series of univariate analyses. Generalised linear mixed models are recommended over other more traditional approaches, since they help to solve the above statistical problems and, more importantly, allow to properly test several hypotheses simultaneously. Finally, several management actions aimed to recover endangered species, such as supplementary feeding, might be considered as field experiments for further testing density-dependence hypotheses in long-lived study models. We expect these opportunities, together with the most adequate statistical tools now available, will help to better our understanding of density-dependent effects in wild populations. [ABSTRACT FROM AUTHOR]