A high-level introduction to the SpaDES framework that can help with reproducible, reusable research.
Here we look at an existing simulation model, LandR-Biomass –> This one is a rewrite of LANDIS-II Biomass Succession
Here we continue with LandR-Biomass –> But we add a module that estimates all parameters and gets all data. The module, Biomass_borealDataPrep, has been tested widely in the boreal forest of Canada. It uses open datasets.
In this section we will present examples of how to bring both static and dynamic R scripts into SpaDES, by creating modules from scratch and linking them
This model, based on Steve Cumming’s work, is a percolation simulation model, that contains all the necessary aparatus to estimate the parameters for a fire regime that is statistically correct, i.e., given an input dataset of historical fires, this module will reproduce the fire size distribution on the same landscape. There are ~8 modules in this model. This module has been used in a variety of places in Canada, but its implementation is still in alpha state, so unexpected behaviour may result.
A SpaDES module is the primary unit of organization for a SpaDES workflow. The key features of the SpaDES anatomy are: 1. Metadata 2.
simList Everything else derives from these two structures
To become more advanced with R, reproducible workflows, profiling etc. we provide some general tips and content to improve skills. These are concentrated on applications with SpaDES, but they are not unique to SpaDES.
We provide some links to useful Cheatsheets
Below are several exercises that may be helpful to work through learning.