Recording Plan: 

Curated Track: 

, 

Call For Participation
Topics relevant to this session include, but are not limited to, the following:

High-performance computing for spatial simulation
Simulation methodologies
Techniques for incorporating heterogenous spatial and social network data
Simulations applications
Simulation of hazards, disasters, and crises
Data- and/or compute- intensive model calibration, sensitivity analysis, and validation
Effective communication of big data generated insights to policy makers
Visualization of simulation processes and output

Description:

As increasingly massive amounts of human and environmental data are generated from sources such as mobile phones, urban sensors, UAV, and satellites, these data resources provide rich opportunities for generating insights on complex modern problems. These rich data sources in turn can be used to support powerful computational simulations that can be applied to representing individual decision making processes and interactions in hypothetical scenarios. However, this intersection of big data and intensive spatial modeling quickly leads to computational challenges, especially for problems that need insights derived from data collected from different sources, in different ranges, and at different scales. This intersection of complex spatially explicit problems and intensive data access provides both challenges and opportunities for spatial modelers such as for handling data processing, integration, and visualization at scale while benefiting from the potential to for example better set model parameters values and better capture ranges in environmental conditions and behaviors. In this session, we aim to explore new approaches, such as high performance computing methods, modeling needs such as visualization, and application-driven examples, such as modeling the COVID-19 pandemic, for applying big data, as expressed using the 5 V’s (Velocity, Volume, Value, Variety, and Veracity), to models of pressing geospatial concerns for more robust coupling between data, models, and knowledge.

Non-Presenting Participants