Title: Harnessing an Open Data Cube and Geospatial Hub for a Coastal Digital Twin: Earth Observations, IoT, Models, and Dynamic Coastal Hazards
Presenting Author: Tom Allen
Organization: Old Dominion University
Co-Author(s): Brian Terry, Yin-Hsuen Chen, Blake Steiner, Sridhar Katragadda, Navid Tahvildari, Soenke Dangendorf, George McLeod, Sunghoon Han, Brandon Feldhaus, Joshua Baptist, Oguz Yetkin, and Heather Richter

Abstract:
The Virginia Open Data Cube is developed as a hub for digital twin prototype analyses in Hampton Roads, Virginia. The Data Cube combines a growing archive of Earth observing satellite data, IoT flood sensors, static geospatial infrastructure, and operational, near real-time forecast hydrodynamic flood models. The data cube is the central resource for cloud data processing and integration for this nascent digital twin, ingesting regional satellite application-ready datasets, a local UAS product catalog, and fine-scale population and community infrastructure data. Three use case scenarios are presented for the current prototype: 1) coastal flood impact assessment (with or without sea level rise); 2) potential vector-borne disease transmission; and 3) mapping of medically-fragile populations for fine-scale vulnerability assessment and intervention. Flood impact assessment combines the operational TideWatch, DELFT3D, and National Water Model forecast models with regional parcel, building, and population data. Vector-borne disease transmission risk is estimated with dynamic hydrometeorological observations and mosquito habitat suitability indices (modeled from land cover, digital terrain, and mosquito vector surveillance.) Land use/land cover and Census data are used for fine-scale equity and environmental justice (EEJ) mapping using Intelligent Dasymetric Mappping (IDM) and open-source Python Spatial Analysis Library (PySAL.) A parallel ArcGIS hub architecture is deployed for geovisual communication and decision-support products from the data cube (e.g., flood visualizations, sea level scenarios, spatio-temporal change, and what-if management and planning actions.) Results highlight improvements in the architecture of the Data Cube to handle diverse data streams and as well as challenges such as error and uncertainty assessment and local, proprietary, or non-federated data. The project summarizes advances and near-term opportunities for the Open Data Cube architecture to replicate and enable more complex digital twin analyses.