Workshops, Classes and Symposia
- SAFS 507: Applied Time Series Analysis for Fisheries and Environmental Sciences. Course website includes recordings of the lectures, lecture ppts, and computer code used in the computer labs.[ONLINE]
- R short course developed by Eric Ward and taught by various NWFSC R gurus. [ONLINE]
- ESA 2012: "Analysis of Time-Series Data using State-Space and Hierarchical Modeling" with Eric Ward and Mark Scheuerell [ONLINE]
- ESA 2010: "Analysis of Time-Series Data using State-Space and Hierarchical Modeling" with Eric Ward [ONLINE]
- MM 2009: "Time-series analysis of population monitoring data" [ONLINE]
- ESA 2009: "A Beginner's Course in Typesetting Scientific Manuscripts with LateX" by Yasmin Lucero (Eli helped) [ONLINE]
- ESA 2008: "Analysis of ecological time-series data using state-space models and R" with Eric Ward, Brice Semmens and Yasmin Lucero [ONLINE]
- ESA 2007: "An introduction to the analysis of community time series using Multivariate Autoregressive (MAR) models" with Stephanie Hampden, Mark Scheuerell, and Steve Viscido [ONLINE]
- ESA 2007: "What's the right size ecological model? Views on model complexity and parsimony from different statistical paradigms" with Kevin Gross
- ESA 2006: "Modern paradigms in population ecology: stochastic, statistical, and inferential" with Brian Dennis Abstract and titles
- ESA 2005: "An introduction to state-space models for estimation of population viability analysis" [ONLINE]
- ESA 2004: "Emerging approaches for the analysis of stochastic ecological data: dealing with multiple error sources, hidden states, complex non-linearities, and uncertainty" Abstract and titles
ESA 2004Emerging approaches for the analysis of stochastic ecological data: dealing with multiple error sources, hidden states, complex non-linearities, and uncertainty.
Real population processes are stochastic. Thus any analysis of population data must deal with this characteristic in some fashion. The traditional approach has been to interpret population data via models attributing variability within the data to either measurement error or process error alone. However, population data almost always contain multiple sources of variability: process error, measurement error, non-linear feedbacks, etc. Mis-attributing the sources of variability has multiple consequences ranging from misestimation of the population behavior to misestimation of the level of uncertainty associated with the data analysis. Fitting biologically motivated models to data from stochastic population processes with multiple sources of variability presents difficult challenges. Recently ecologists have made significant in-roads into these problems. This session features talks using these new approaches to analyze population data.
Engen, Steinar*,1, 1 Department of Mathematical Sciences, NTNU, Trondheim, Norway Predictions in age-structured populations in a fluctuating environment.
Lele, Subhash*,1, 1 University of Alberta, Edmonton, Alberta, Canada Statistical inference for Gompertz model with sampling variability: A composite likelihood approach.
Lindley, Steven*,1, 1 NOAA Fisheries, Santa Cruz, CA State-space models for analyzing time series of population abundance.
Hinrichsen, Richard*,1, 1 Hinrichsen Environmental Services, Seattle, WA, USA Multivariate state space approaches to estimating population growth rates.
Rees, Mark *,1, Ellner, Steve2, 1 Imperial College, Silwood Park, Ascot, Berks, UK2 Cornell University, Ithaca, NY 14853-2701 Stop using matrix models: Parsimonious PVA with stochastic integral projection models.
Holmes, Elizabeth*,1, 1 Northwest Fisheries Science Center, Seattle, WA, USA From theory to application: Using diffusion approximations to model complex stochastic populations processes with applications for population viability analyses.
Ives, Anthony*,1, Einarsson, Árni2, Gardarsson, Arnthór2, 3, Jansen, Vincent4, 1 Department of Zoology, Madison, WI, USA2 Myvatn Research Station, IS-660 Myvatn, Iceland3 Institute of Biology, IS-108 Reykjavík, Iceland4 School of Biological Sciences, Egham, Surrey, UK Extreme fluctuations in midge densities: The possibility of resource-consumer dynamics with alternative dynamical states.
Kaplan, Isaac*,1, Kitchell, James 1, 1 University of Wisconsin-Madison, Madison, WI, 53706 State-space models for Yellowfin Tuna Catch-Effort Data.
ESA 2006Modern paradigms in population ecology: stochastic, statistical, and inferential
In the last 10 years, the study of population and community dynamics has shifted towards stochastic models away from the deterministic models so familiar in ecology during the last century. Understanding of the properties of stochastic versions of familiar ecological models is an active area of research, and along the way, the field of theoretical ecology is shifting to new paradigms of thinking about ecological processes. The familiar concept of population state or carrying capacity as a fixed line passing through a series of observations is not particularly meaningful in a stochastic framework, and is replaced by the concept of stationary probability distributions. The concept of equilibria is replaced by the concepts of inflection points in first passage probabilities and modes and antimodes in stationary distributions. At the same time, there has been a fundamental shift away from qualitative visual comparisons of model output with qualitative system behavior, and towards rigorous statistical linking of stochastic ecological models and observations using modern, often numerical, statistical methods, which are suited for non-linear stochastic models which include both process and non-process variability. Concepts such as likelihood surfaces, first passage distributions, conditional probability distributions, prior and posterior distributions, numerical statistical algorithms, and formal model support have joined nonlinear dynamics and stability as permanent parts of the landscape of ecological understanding. This session features some of the contemporary research on stochastic ecological dynamics and estimation that is changing the face of population ecology and that will ultimately fundamentally change the way we think about and make inferences about ecological processes.
1. DENNIS, B. University of Idaho. Estimating density dependence, process noise, and observation error.
2. STAPLES, D Minnesota Department of Natural Resources. Risk-based viable population monitoring.
3. HENSON, S 1., J G Galusha 2., J L Hayward 1., J M Cushing 3. and B Dennis 4. 1. Andrews University, 2. Walla Walla College, 3. University of Arizona, 4. University of Idaho. Identifying demographic and environmental factors in the dynamics of animal behavior in field populations.
4. REUMAN, D C 1., R A Desharnais 2., R F Costantino 3., J E Cohen 1. 1. The Rockefeller University, 2. California State University, 3. University of Arizona. Power spectra reveal the interactions among nonlinear population dynamics, stochasticity, and lattice effects.
5. WIKLE, C 1. University of Missouri. A general framework for spatio-temporal dynamics in hierarchical Bayesian models.
6. TULJAPURKAR, S 1. and C V Haridas 1. 1. Stanford University. Diffusion aproximations with autocorrelation and nonlinearity.
7. PONCIANO, J. University of Idaho. On the use of stochastic population models in experimental evolution.
8. HOLMES, E E 1., W F Fagan 2. and J Sabo 3. 1. National Marine Fisheries Service, 2. University of Maryland, 3. University of Arizona. Parsimonious stochastic models for first-passage and extinction dynamics
9. LELE, S. University of Alberta. Bayesian inference in ecology: Informative, non-informative or disinformative.