Simple statistical models can be sufficient for testing hypotheses with population time‐series data

Abstract

Time‐series data offer wide‐ranging opportunities to test hypotheses about the physical and biological factors that influence species abundances. Although sophisticated models have been developed and applied to analyze abundance time series, they require information about species detectability that is often unavailable. We propose that in many cases, simpler models are adequate for testing hypotheses. We consider three relatively simple regression models for time series, using simulated and empirical (fish and mammal) datasets. Model A is a conventional generalized linear model of abundance, model B adds a temporal autoregressive term, and model C uses an estimate of population growth rate as a response variable, with the option of including a term for density dependence. All models can be fit using Bayesian and non‐Bayesian methods. Simulation results demonstrated that model C tended to have greater support for long‐lived, lower‐fecundity organisms (K life‐history strategists), while model A, the simplest, tended to be supported for shorter‐lived, high‐fecundity organisms (r life‐history strategists). Analysis of real‐world fish and mammal datasets found that models A, B, and C each enjoyed support for at least some species, but sometimes yielded different insights. In particular, model C indicated effects of predictor variables that were not evident in analyses with models A and B. Bayesian and frequentist models yielded similar parameter estimates and performance. We conclude that relatively simple models are useful for testing hypotheses about the factors that influence abundance in time‐series data, and can be appropriate choices for datasets that lack the information needed to fit more complicated models. When feasible, we advise fitting datasets with multiple models because they can provide complementary information.

The environmental conditions of endemism hotspots shape the functional traits of mammalian assemblages

“…The ability of a species to expand its range through time is not only a function of its size and specialization, but also its ability to reproduce. Low fecundity can result in low abundance [90], which is in turn often correlated with range size [43,91, but see 92]. By contrast, fecund species can rebound quicker and more efficiently after periods of low resource abundance than less fecund species, and are therefore able to outcompete less-fecund species and disperse across landscapes more successfully [32,36].…”

Life‐history connections to long‐term fish population trends in a species‐rich temperate river

“…For comparison, we also analysed species‐specific ‘catch per sample’ in each survey, calculated by dividing species counts by the number of survey‐specific samples, multiplying by 100 and rounding to an integer (i.e. estimating the number of individuals expected per 100 samples in each survey; Wenger et al., 2022).…”

“…Community analyses have benefits over single-species models because they can identify shared responses (Chapman & Purse, 2011), shifts in functional diversity (Laureto et al, 2015) and facilitate comparison of trends among taxa (Cazalis, 2022). Time-series records of species identities and abundances at a consistent set of sites are important for investigating how communities change in response to environmental conditions (Wenger et al, 2022). For instance, decades of data from a French monitoring programme have allowed scientists to assess changing distributions of fishes in response to climate change (Grenouillet & Comte, 2014).…”

“…Time‐series records of species identities and abundances at a consistent set of sites are important for investigating how communities change in response to environmental conditions (Wenger et al., 2022). For instance, decades of data from a French monitoring programme have allowed scientists to assess changing distributions of fishes in response to climate change (Grenouillet & Comte, 2014).…”

Multi-scale spatial models reveal the interplay of weather and habitat effects on cold-water fish

“…We accounted for the complexity and hierarchy of stream habitats by specifying a statistical model with multi-scale random effects. Random effects are commonly used in wildlife population modeling because they enable realistic inference on key drivers by quantifying variation attributable to unobserved ecological mechanisms (Wagner et al, 2006;Wenger et al, 2022). We induced spatial correlation in the random effects using a combination of traditional geostatistical and autoregressive models and a class of recently developed network models that characterize configurations, connectivities, and flow directions of dendritic ecological networks such as stream networks (Ver Hoef and Peterson, 2010;Peterson et al, 2013).…”