Computational science

Theorists have always developed mathematical models to attempt to gain insight into how physical systems operate. However, rapid advances in computational capabilities over the last several decades have enabled the creation of high-fidelity simulation software based on those models, which are then exercised as a proxy for learning about the real world. This new approach, which serves as a complement to pure theory and experiment, has come to be known as computational science.

Effective computational science research requires expertise in mathematical modeling, numerical analysis, software engineering, high-performance computing, and statistics, as well as a deep understanding of the technical application area under study. As a result, it is a deeply interdisciplinary endeavor, requiring the combined efforts of computer scientists, mathematicians, statisticians, and application scientists.

At NIST, computational scientists work to predict properties of atomic, chemical, biological, and material systems from first principles, as well as for engineered systems, such as buildings and communication networks. Others use computation to study how fires and their contaminants spread within buildings and at the wildland-urban interface. NIST mathematicians work to develop more efficient and accurate numerical methods that enable higher fidelity simulations, computer scientists develop techniques and tools to map such computations onto modern parallel and distributed computing systems, and to visualize the often complex data that emerges. More mature research efforts can result in the distribution of well-engineered software enabling members of the broader scientific community to perform simulation studies of their own.