Abstract
Multi-species prairie composed primarily of clonal plants can achieve different ecological functions. They can be a good source of biofuel, a mean of water purification in agriculture or a biodiversity guardian. The goal of the Virtual Prairie project is to study the fundamental mechanisms involved in the population dynamic of a prairie and the optimum engineering design of prairies serving these and other functions. Our study is a modelization and a simulation of the growth of such clonal plants in isolation or competing with each other This study is a very compute intensive enterprise. However it can be eventually managed as an embarrassingly parallel set of jobs. Therefore, we adapted our simulation project to run on BOINC,a platform for volunteer computing, which compute nodes are reached using the commercial Internet connections. To our knowledge, this is the first ecology project that uses volunteer computing and it is the largest simulation ever done on plant ecology. Our preliminary results show that there is no unique optimum design solution but rather different optimum solutions that can co-exist.