Kristine Bonnevies hus (map)
UiO, Campus Blindern Blindernveien 31 Entr. Moltke Moes vei
This Journal Club deals with the vast and growing body of research concerning structured population models such as classical matrix models, trait-based and individual based models. We discuss techniques for estimating demographic rates, building and analysing structured models, as well as issues and problems associated with such models. On the one hand, our meetings are a place to pluck apart scientific literature, but on the other hand, they also represent a platform for discussing your individual problems and challenges with structured population models in a compact group of people.
Sex differences in vital rates and mate availability can have important effects on population- and evolutionary dynamics. These effects and how they vary depending on mating strategies can be explored with extensions to traditional matrix models.
Integral projection models (IPMs) are population models structured by continuous traits such as body size, and have risen in popularity over the last decade. While most perturbation analyses developed for matrix models can be applied, additional considerations are necessary when working with IPMs.
Recent work has highlighted the importance of including individual heterogeneity into population models. This includes both traits that are fixed over the lifespan of an individual (e.g. morphology, genotype) and characteristics that change over time (e.g. age, body conditions). How influential such traits are for individual fitness (and population dynamics), may however depend on sex.
Stage structure is fundamental in quantitative population models, but there are different approaches to deal with stage duration and individual-/cohort variation therein.
Differences between individuals can be large and have profound consequences for the dynamics of populations. Even if such differences have unknown causes and/or are unobservable, they can be incorporated into population models, allowing to assess their impacts on population-level patterns.
Species' responses to changes in the environment can be either genetic or plastic, and adaptive or non-adaptive. Identifying the exact nature of such responses requires integration of population models and quantitative genetics theory.
Climate can affect populations not only directly and through resource availability, but also by altering predation risk. Such interactions can lead to unexpected or counter-intuitive responses, and considering them can be important for predicting population responses to climate change.
The evolution and spread of human culture are intriguing topics by themselves, but who knew cultural dynamics could be included into demographic population models?
Body size is often linked intricately with survival and reproductive rates, and therefore affects population dynamics. It is not unlikely for population collapses to be preceded by a change in body size distributions. If those changes happen long enough in advance, they may serve as early warning signals to predict population collapses.
Experts have repeatedly predicted that human life expectancy soon will reach a ceiling, but they have been proven wrong every time. Annual increase in life expectancy has not slowed down, and it continues to increase by 3 months every year.
Body mass is an important indicator of general condition as it reflects energy accessible for survival and reproduction. Recent evidence show that several species have experienced shifts in their body mass due to climate change. In the monogamous wandering albatross, average body mass and breeding success has increased over the last years. Surprisingly, the increase in breeding success seems to be due to heavier fathers investing more in their sons.
Most demographic population models ignore males, but empirical evidence suggest that they should be included when vital rates are sex-specific. Assumptions about adult sex ratio, social structure, and mating system have been shown to affect estimates of extinction risk and projections of population dynamics. We discuss about when and how to apply two-sex models.
Our first statistics course warns us about making predictions beyond the observed range of data. What that means exactly is difficult to say though when we use more complex models with link functions, higher order effects and interactive terms. We discuss a quantitative method for assessing bias when extrapolating.
The evolution of reaction norms such as thermal performance is tightly linked to ecological processes, and eco-evolutionary models can provide important insights especially in varying environments.
The demographic buffering hypothesis states that the temporal variation in vital rates is smaller the more these vital rates contribute to fitness. The pattern is appears well supported, but recent evidence indicates that life histories may be buffered or labile, and that there is a phylogenetic signal in this.
There is increasing support for body size as an important factor determining population dynamics in both animal and plant species. Integral Projection Models (IPMs) make it possible to assess the interplay of body size and population dynamics, but does body size always matter?
Effective population size is an important concept in evolutionary biology, providing information about genetic variability, inbreeding and the efficiency of directional selection. Despite its obvious relevance, it has rarely been discussed in the context of fishery-induced selection.
Integral projection models (IPMs) have become a popular tool to assess questions relating to eco-evolutionary dynamics. Within IPMs, change in a continuous trait of interest (body length, leaf area, horn size etc.) are modelled both within generations (growth) and across generations (inheritance). However, current methods of estimating growth and inheritance inherently fail to properly estimate phenotypic evolution.
Estimates of fishing mortality commonly used in stock assessment models are often conditional on restrictive assumptions about natural mortality. However, integrating data from various sources in bayesian state-space models can allow to independently estimate mortalities of different sources.
In many harvested ecosystems, laws and regulations protect animals below a certain size from being killed. However, in species such as fish, it is often the large, old animals that represent the reproductive capital of a population, and that might need protection even more.
Traditionally, population models are often built using only the female half of a population and males are considered nothing but "ecological noise". However, males do matter, and particularly so when there is sex-selective harvest going on.
Population matrix models have come a long way and perturbation analyses developed for them are among the most powerful tools of population ecologists. Most population projections are unthinkable without sensitivity analyses and LTREs (Life Table Response Experiments). Population projections are also most needed when climates are changing and habitats are altered, and that is when classic perturbation analyses for equilibrium systems fail.
In a recent paper, Koons et al. explore how to do LTREs in a transient world:
Trait-based demographic models such as IPMs relate not only future adult size but also offspring size to the current size of an individual. This is in stark contrast to the quantitative genetic approach, and may have consequences for the predictions of evolutionary dynamics.
We look into this issue with a recent paper from Chevin:
In a recent session, discussion rose about reproductive value and fitness. In fact, there is much controversy concerning definitions, and while the definition of reproductive value is relatively straightforward, its interpretation, in particular as a fitness measure, is more confusing.
For this session, we want to look a bit more into that topic, starting off with Jacob Moorad's recent paper:
Population models such as IPM's allow the inclusion of continuous traits for demographic analysis, and therefore the tracking of phenotypic change over time. Especially in the case of rapid phenotypic change, it was therefore only a matter of time until methods were developed to identify the sources of such change as evolution, plasticity or demography. One such method is the age-structured Price equation, developed by Coulson & Tuljapurkar in the paper we are discussing this session:
"The Dynamics of a Quantitative Trait in an Age-structured Population Living in a Variable Environment" (Coulson & Tuljapurkar 2008)