The Global Environmental Stratification uses statistical clustering of bioclimatic variables to classify the world into 125 strata, and has been shown to successfully partition environmental regions at a fine spatial scale.
I've been helping [http://www.sruc.ac.uk/asoteriades Andreas Soteriades] and [http://www.ed.ac.uk/schools-departments/geosciences/people?indv=1578&cw_xml=person.html Marc Metzger] to look at how these might shift in the future due to climate change. This involves:
* building a classifier which can use the limited set of variables available from future GCMs, and training it on current data.
* taking General Circulation Model (GCM) output, and deriving values for future bioclimatic variables.
* using the classifier to produce a map of future strata.
* analysing and communicating the results.
As well as the machine learning, I've been involved in visualising the data - see the blog posts below, and more pictures to come.
== Blog Posts ==
In the [http://www.mo-seph.com/blog/datavis1 last post], talking about my work with Andreas and [http://www.ed.ac.uk/schools-departments/geosciences/people?indv=1578&cw_xml=person.html Marc], we saw how a bit of data visualisation helped to understand why some output was looking funny, and how the choice of classifier contributed to some strange behaviour. In that case, we were really lucky to spot it.