A practical guide to selecting models for exploration, inference, and prediction in ecology

Selecting among competing statistical models is a core challenge in science. However, the many possible approaches and techniques for model selection, and the conflicting recommendations for their use, can be confusing. We contend that much confusion surrounding statistical model selection results from failing to first clearly specify the purpose of the analysis. We argue that there are three distinct goals for statistical modeling in ecology: data exploration, inference, and prediction. Once the modeling goal is clearly articulated, an appropriate model selection procedure is easier to identify. We review model selection approaches and highlight their strengths and weaknesses relative to each of the three modeling goals. We then present examples of modeling for exploration, inference, and prediction using a time series of butterfly population counts. These show how a model selection approach flows naturally from the modeling goal, leading to different models selected for different purposes, even with exactly the same data set. This review illustrates best practices for ecologists and should serve as a reminder that statistical recipes cannot substitute for critical thinking or for the use of independent data to test hypotheses and validate predictions.

Dispersal and adaptation strategies of the high mountain butterfly Boloria pales in the Romanian Carpathians

Background

Habitat quality is one main trigger for the persistence of butterflies. The effects of the influencing biotic and abiotic factors may be enhanced by the challenging conditions in high-alpine environments. To better our knowledge in this field, we performed a mark-release-recapture study with Boloria pales in the Southern Carpathians.

Methods

We analysed population structure, movement and foraging behaviour to investigate special adaptations to the alpine environment and to reveal differences between sexes. We compared these aspects in one sector with and one sector without grazing to address the effects of grazing intensity on habitat quality.

Results

We observed "soft" protandry, in which only a small number of males appeared before females, and an extended emergence of individuals over the observed flight period, dividing the population's age structure into three phases; both observations are considered adaptations to high mountain environments. Although both sexes were mostly sedentary, movement differences between them were obvious. Males flew larger distances than females and were more flight-active. This might explain the dimorphism in foraging behaviour: males preferred nectar sources of Asteraceae, females Caprifoliaceae. Transition from the grazed to the ungrazed sector was only observed for males and not for females, but the population density was higher and the flight distances of the individuals were significantly longer on the grazed sector compared with the ungrazed one.

Conclusion

Soft protandry, an extended emergence of the individuals and an adapted behavioural dimorphism between sexes render to represent a good adaptation of B. pales to the harsh environmental conditions of high mountain ecosystems. However, land-use intensity apparently has severe influence on population densities and movement behaviour. To protect B. pales and other high-alpine species from the negative consequences of overgrazing, areas without or just light grazing are needed.

The fingerprints of global climate change on insect populations

Synthesizing papers from the last two years, I examined generalizations about the fingerprints of climate change on insects' population dynamics and phenology. Recent work shows that populations can differ in response to changes in climate means and variances. The part of the thermal niche occupied by an insect population, voltinism, plasticity and adaptation to weather perturbations, and interactions with other species can all exacerbate or mitigate responses to climate change. Likewise, land use change or agricultural practices can affect responses to climate change. Nonetheless, our knowledge of effects of climate change is still biased by organism and geographic region, and to some extent by scale of climate parameter.