Theory
Week 1
Course Introduction
Understand course expectations and requirements
Discuss research interests and how they align with the course
Describe and discuss current perspectives on ecological modelling & forecasting
Introduction to the course and to the R language using Swirl
Introduction to basic R syntax
Revisting common statistical tests
Week 2
R Time Series data
Learning Objectives:
Discover unique challenges with working with time-date information on computers
Be able to properly format time-date data
Apply time-date knowledge to ecological data
Paleoecological Dynamics
Learning Objectives:
- Describe and discuss past responses of organisms to climate change
- Compare and contrast past responses to change with modern conditions
- Apply knowledge of the past to assess challenges for forecasting
Week 3
Phenology
Learning Objectives
Describe phenology and how it differs from other types of long-term change
Describe different classess of models used in ecological forecasting
Compare and contrast challenges for forecasting phenology vs other types of long-term change
Week 4
Time series decomposition
Learning Objectives:
Describe time series decomposition and how it applies to ecological data
Apply R functions to plot and examine dynamics at different time-scales in data
Community dynamics
Describe contemporary biodiversity dynamics
Critically assess current biodiversity data limitations
Determine what challenges and opportunities both our data and observations of nature imply for ecological forecasting
Week 5
Time Series autocorrelation
Learning Objectives:
Describe autocorrelation in time series data and what generates it
Use R to examine autocorrelation structure in ecological time series data
Introduction to Time Series Data Learning Objectives:
Gain familiarity with the spatial and temporal scales of ecological data collection Explore relationships between forecasting needs and current ecological data collection
Week 6
Time Series modelling in R I
Learning Objectives:
Apply lessons on autocorrelation, seasonal dynamics, and time-date formatting
Describe what an ARIMA model is
Fit ARIMA models to ecological data
Time Series modelling in R II
Learning Objectives:
Apply lessons on autocorrelation, seasonal dynamics, and time-date formatting
Describe what an ARIMA model is
Fit ARIMA models to ecological data
Week 7
Time Series modelling in R III
Learning Objectives:
Describe what a TSLM model is
Fit TSLM models to ecological data
Describe how to integrate linear models and ARIMA
Fit linear models with ARIMA errors
Introduction to Forecasting
Learning Objectives:
Read and communicate using forecasting terminology
Describe general approaches to making forecasts
Describe and discuss current perspectives on ecological forecasting
Describe uncertainty and the implications of uncertainty for ecological forecasting
Week 8
Uncertainty
Learning Objectives
Define the different types and sources of uncertainty and variation associated with ecological forecasts
Describe graphical models of forecast uncertainty in simple population dynamics model
Describe how uncertainty and variation influence ecological forecasts
Week 9
Evaluating Forecasts in R I
Learning Objectives:
- Evaluate point forecast accuracy in R Evaluate forecast uncertainty in R
Evaluating Forecasts in R II
Learning Objectives:
- Evaluate point forecast accuracy in R
Week 10
Species Distribution models I
Learning Objectives:
Describe how species distribution models are built
Define different approaches to species distribution modeling and explain how they relate to explanation vs. prediction
Describe the challenges associated with distribution modeling based forecasts
Species Distribution models II
Learning Objectives:
Describe how species distribution models are built
Define different approaches to species distribution modeling and explain how they relate to explanation vs. prediction
Describe the challenges associated with distribution modeling based forecasts
Week 11
Election Forecasts
Learning Objectives:
Describe data issues and modelling approaches used in election forecasting
Compare and constrast data issues and modelling approaches to those used in ecological forecasting
Assess which aspects of election forecasting may be useful for ecological forecasting
Hurricane Forecasts
Learning Objectives:
Describe different model types used to hurricane forecasting
Compare and contrast issues with hurricane forecasting with issues for ecological forecasting
Assess whether hurricane forecasting is a good model for ecological forecasting
Week 12
Complex Time Series models
Learning Objectives:
Fit generalized dynamic linear models in R using mvgam
Interpret complex time-series models from mvgam Make forecasts using mvgam models in R
State-Space Models
Learning Objectives:
Define state space models
Describe why state space models are useful in ecology
Explain how some non-linear state space models capture population dynamics
Week 13
Complex Time Series Models II
Learning Objectives:
Fit generalized dynamic linear models in R using mvgam
Interpret complex time-series models from mvgam
Make forecasts using mvgam models in R
Week 14
Ethics
Learning Objectives:
- Describe the different types of ethical concerns related to forecasting Scenarios
Learning Objectives:
Define and describe scenario based forecasting methods and their uses
Distinguish between scenario planning, predictions, and forecasting
Describe approaches to multi-group decision making using scenarios
Week 15
Course Wrap-Up
Learning Objectives:
- Synthetic perspective on the strengths and weakness of ecological forecasting