Skills workshops
A. Individual stochasticity: an introduction to demographic models and analysis
8:30-10:30 Wednesday (Newcomb Hall Rm. 389)
Hal Caswell
University of Amsterdam
Variation among individuals in survival and reproduction is central to evolution. It arises from two sources: heterogeneity and individual stochasticity. Heterogeneity refers to genuine differences among individuals in their properties. Individual stochasticity refers to apparent differences that result from probabilistic demographic processes. Their implications are dramatically different. Genetic heterogeneity is subject to selection; this notion is formalized by Crow’s index of the opportunity for selection. Non-genetic heterogeneity may play an important role in demographic dynamics. Individual stochasticity is a property of any life cycle and its probabilities of survival, transition, and reproduction. It is impossible to interpret observed variation without calculating how much variation is implied by individual stochasticity. That calculation is now possible, and this workshop will show how to do it. Methods will be presented for longevity and lifetime reproductive output. The methods are applicable to both age- and stage-classified populations, and to both human and non-human organisms. Participants should be familiar with the basics of age- and stage-classified matrix population models. Participants will need a computer with Matlab or R (and, for R, the packages necessary for matrix manipulation).
B. Comparative demography using COMPADRE and COMADRE
8:30-10:30 Wednesday (Newcomb Hall Rm. 376)
Rob Salguero-Gomez
University of Sheffield, UK
Owen Jones
Max Planck Odense Center, University of Southern Denmark
The participants will be walked through the organization of the COMPADRE Plant Matrix Database and COMADRE Animal Matrix Database, which together contain thousands of matrix population models and useful metadata to unlock global analyses in the realms of plant and animal demography. Participants will get most out of this workshop if they are at least familiarized with matrix models and data manipulation in R. Participants need to bring their own laptops with the latest version of R installed, together with the packages phyloPCA, caper, ape, scales, MASS, matrix, plotrix, devtools, popbio and popdemo.
C. Analyzing transient population dynamics
11:00-1:00 Wednesday (Newcomb Hall Rm. 376)
Iain Stott
Max Planck Odense Center, University of Southern Denmark
Participants will be introduced to the concept of transient population dynamics. These dynamics occur in populations with nonstable structure, which may result from stochastic environmental variation or environmental disturbance events. They are useful measures for exploring new hypotheses about the influence of environmental variation and disturbance on demography and life history evolution. Participants will use the R package popdemo to calculate transient dynamics of population matrix projection models, and perform linear and nonlinear perturbation analyses of transient indices. A basic knowledge of using matrices in R is needed, and participants will need to bring a laptop with the latest version of R and the popdemo package installed.
D. Advanced approaches to population modeling using Integral Projection Models
11:00-1:00 Wednesday (Newcomb Hall Rm. 389)
Rob Salguero-Gomez
University of Sheffield, UK
Jessica Needham
Oxford University
IPMs are a flexible and generalizable tool for analysis of demographic data. Building IPMs, however, is often complicated by limitations of available data, the structure of the data, and computational challenges. In this short course we will present both problems and solutions, moving through a workflow from vital rate models to population-level inference using IPM products. We will provide insights into dealing with missing or biased data, computational bottlenecks, and novel approaches to regressions linking vital rates to traits. Our focus will stem from work on forests, but be relevant for many other organisms. Participants should bring their own laptops with an update version of R installed. An understanding of demographic models, Bayesian analyses, and advanced R coding will be helpful.