Towards policies for data insertion in dynamic data driven application systems: a case study sudden changes in wildland fire

AbstractWe have applied the Dynamic Data Driven Application System (DDDAS) methodology to predict wildfire propagation. Our goal is to build a system that dynamically adapts to sudden changes in environmental conditions. For this purpose, we are building a parallel wildfire prediction method, which is able to assimilate real-time data to be injected in the prediction process at execution time. This data-injection needs to be intelligent in order noy to disturb the simulation process outputs. In this paper, we propose a policy for data insertion using a statistical approach and we design a set of experiments based on California wildfire where Santa Ana winds generate the ideal conditions for sudden changes in fire behavior