Authors: Kostas Alexandridis*, University of the Virgin Islands, Christina Chanes, University of the Virgin Islands, Corporate Extension Service
Topics: Coupled Human and Natural Systems, Cyberinfrastructure, Spatial Analysis & Modeling
Keywords: spatial IoT, microclimate, climate change
Session Type: Paper
Start / End Time: 10:00 AM / 11:40 AM
Room: Maurepas, Sheraton, 3rd Floor
Presentation File: No File Uploaded
While global climate change represents one of key challenges in the 21st century, small coastal and island communities are among the most vulnerable and risk-exposed. The geodemographic data in the US Virgin Islands indicate that proximity to coastal areas is correlated and/or associated with low income and inequality population groups. Developing intelligent environmental sensor and spatial monitoring applications achieve multiple goals: exposing communities and decision makers to the realities of contemporary environmental change; aiding visual and comprehensive understanding of the core science of spatial environmental and climate change, and aiding citizens and communities at large in the collective effort to adapt to change, thus reducing vulnerability and mitigating associated risks.
We developed a truly integrated and intelligent Internet of Things (IoT) spatial-environmental sensor technology application using weather, environmental and locational GPS sensors. The development and deployment of microprocessors/microcontrollers (e.g., Raspberry Pi, Arduino), combined with cloud-based sensor data sensor storage technology (e.g., Azure, Amazon Cloud), along with real-time visualization and data analytics (based on machine learning algorithms), allow us to visualize and analyze high-density and high spatiotemporal resolution data. The applied research study demonstrates how low-cost IoT sensor technology can provide scientific support data that can enhance and increase spatial and temporal accuracy of micro-climatic variation of weather conditions. Using historical hurricane data from the IBTrACS NOAA database, we can demonstrate how we can derive spatial prioritization of areas with high probabilistic estimates of vulnerability for deploying such grid sensor technologies in the US Virgin Islands.