GPU modelling via the web

Running a simulation model on a GPU can be a daunting task for somebody not acquainted with programming. However, there is a solution, and it is called WebGL. WebGL is a Java-based system that allows you to run a model on the GPU. Below, I have implemented two self-organization models in WebGL, the Klausmeier model for arid vegetation patterns, and my own mussel bed pattern formation models. If you click on the pictures below, an new webpage opens which runs the model, and allows you to set the parameters. If you don’t see anything appearing (for instance on an phone or tablet), then the computer it too light to run the applets.

The first model is the Klausmeier model that described the formation of arid vegetation patterns at the edge of the Sahara (click to run model):


The second model is my model for the formation of banded patterns in mussel beds, as are found for instance in the Wadden Sea, in the north of the Netherlands (click to run model):

Mussel Patterns

Below are a couple of other models that provide interesting patterns and dynamics. The first is the WebGL implementation of the Gray-Scott model, which is very similar to Klausmeier’s model, but is studies for a much broader set of parameter values. The second is the Fitzhugh-Nagumo model for neural excitation. Click on the pictures below to view the web models.

The Grey-Scott model


The Fitzhugh-Nagumo model