GEMS X006.11 - Spatio-Temporal Accounting of Biotic Threats: Introduction to SDMS & Data Preparation for Modelling
Description
Have your research, studies or work required you to examine the geographic distribution of various species? This could be for crop protection, forestry, environmental protection, environmental impact assessment, urban and natural landscape design and development, or simply an interest in any given species, whether insects, pathogens or weeds. Then you are at the right place, species distribution models allow us to understand the potential and realized distribution of various species across our landscapes at different scales. Understanding of species distribution at different scales provides different insights, for example global maps support decision makers to make high level biosecurity and conservation related decisions. Meso-scale or regional maps can be used for planning of cross-border projects, informing biosecurity related cross-border operations and various environmental projects. Small scale or local maps allow researchers to design custom pest prevention, monitoring and control strategies. Allow conservationists to use evidence-based information in their plans etc. In this 3-hour workshop (first part of the module) participants will have the opportunity to learn uses, types & limitations of species distribution models.
Modelling species distribution with the aim of understanding either current or future habitats of a give species requires some input information about the current location of the species to be studied. Whether a pest or a rare species that need to be conserved, understanding where they are located at the present time gives us important information about where else these species could be found either globally or at any spatial extent. The second part of this workshop will teach you how you can source, clean, verify, validate, standardize, and organize species presence or occurrence points so that you can use them as an input for various species distribution models.
The geographic distribution of any species is affected by a number of underlying factors like climatic variables, environmental attributes or socio-economic drivers. Depending on the scale we want to capture the potential or realized distribution of any give species it is important that we choose the appropriate factors or indicators. This workshop (third part of the module) will cover selection of the appropriate variable for a given species distribution modelling project as well as cover how to pre-process environmental datasets so that they are ready for species distribution modelling.
Outline
- Introduction to species distribution modelling (SDM): concept, rationale, review
- Conceptual discussion on geographic distribution of Species & SDMs
- The rationale for SDMs
- Review of different SDMs
- Sourcing & curating species occurrence datasets for SDMs
- Species occurrence datasets
- Sourcing species presence and absence datasets
- Cleaning species occurrence datasets
- Standardizing, Validation and Curation of species occurrence datasets
- Environmental data, variable selection & processing for SDMs
- Types of predictor datasets (climatic, environmental, and socio-economic datasets) for SDMs
- Sourcing predictors for SDMs
- Pre-processing predictor datasets
- Correlation, collinearity, dimension reduction and variable selection
