Gage, S.H., M. Colunga-Garcia, G.R. Safir, J.J. Helly, and A. Momin
Presented at the All Scientist Meeting (2002-10-04 )
Simulation of physical and biological phenomena at multiple time and space scales is playing a pivotal role in areas that range from genetic studies to global change analysis and assessment. Regional simulation models require the use of large data inputs and are evolving in scope and sophistication. A Modeling Applications System Integrative Framework (MASIF) is presented as a method to facilitate utilization of large amounts of spatial-temporal outputs from regional scale simulations. MASIF links an array of existing visualization, analytical, and data management software to manage model inputs and outputs as well as model execution. MASIF can be implemented as a stand-alone or web-based system to accommodate batch and interactive processing scenarios. We have used a maize crop growth model to demonstrate the potential of this approach to handle millions of location-specific input/output records that are processed for each model run. Under MASIF control, model output was processed into a time series of maps that were exported to animation software to create digital videos of model simulation outputs providing a graphical analysis of daily temporal and spatial dynamics of maize growth in the North Central Corn Belt over a 20-year period. In addition, we have included examples that demonstrate MASIF capabilities for statistical analysis. Information from MASIF is shown in visual form, an approach that we believe is preferable for comprehending information contained in large datasets associated with models that simulate processes and patterns at regional scales.
Back to meeting | Show |