-
Notifications
You must be signed in to change notification settings - Fork 0
About MDiG
The long term aim is for MDiG to be a generic dispersal simulation framework, and an open standard for specifying dispersal models. In other words, the model format (DispersalML) will be readable by clients in multiple GIS platforms. Currently only GRASS is supported, as it is a powerful, open source GIS that is freely available.
Many institutions and researchers are essentially re-implementing the same framework where a series of raster maps are piped through a chain of modules, each representing some kind of process. E.g. A similar raster tool for modelling dispersal is CFS-FBM (Canadian Forest Service - Forest Bioeconomic Model), although it's focus includes more economic factors and dispersal has been added on to it. MDiG stands to be an open, and standardised platform for species dispersal simulation. MDiG has its roots in modelling the spread of invasive species, firstly insects (part of Joel's PhD was modelling the spread of Argentine ants) and now weeds (working under contract).
The main benefits of MDiG are:
- Simulation and replicate management - It will let you run replicates of a model and keep track of all the maps. It will also merge replicate maps into an occupancy envelope (an average map) for each time step.
- Automation of analysis - MDiG will allow you to run an analysis across all your replicates and all their timesteps and collate the results.
- There are a number of default modules that represent different methods species can use to spread. Long distance, shaped neighbourhoods, local constiguous spread.
It is worth noting that the output of this model is a projection rather than a prediction or forecast sensu Keyfitz (1972). The results are the probability distribution of possible future spread scenarios for the species. Rather than making specific conclusions about where the species will have established and at what time, the results indicate a relative likelihood of establishment across the landscape.
Keyfitz N (1972) On future population. Journal of the American Statistical Association 67: 347-363