Micro-scale Urban Heat Island Analytics with Anthropogenic Heat Releases and A Deep Learning Based Spatiotemporal Prediction Framework
Abstract
As one of the adverse effects of urbanization and climate change, Urban Heat Island (UHI) can affect human health by contributing to general discomfort, respiratory difficulties, heat cramps and exhaustion, non-fatal heat stroke, and heat-related mortality. The global warming trend is further contributing to the UHI by increasing the already higher temperatures in heat island areas. Current researches have been relying on remote sensing imagery and focusing on the broad understanding of the UHI phenomenon and mitigation support, but problems remain due to the complex human-weather interactions in the urban area. The emergence of advanced Internet of Things (IoT) technologies has broadened the opportunities for urban informatics studies. Advances in IoT technology can provide higher spatial and temporal resolution while reducing the impact of weather on observed data, as well as continuous growth and better spatial coverage. This research proposes a state-of-art framework in the field for short-term UHI prediction based on a Convolutional Neural Network (CNN) and Long Short-term Memory (LSTM) build upon the temperature values collected from the IoT sensor network in Los Angeles. The correlation between micro-scale UHI (microclimate) and the surrounding anthropogenic heat sources (i.e., human mobility, building energy consumption, and traffic volume) is studied to establish a human-weather relationship. Together with the prediction framework, the study improves the short-term prediction with detailed human activity impacts and high-resolution environmental measurement. The prediction result is expected to serve as an alarm mechanism that assists the decision-making process for vulnerable populations with heat-related health concerns.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2019
- Bibcode:
- 2019AGUFMGC21I1360Y
- Keywords:
-
- 0493 Urban systems;
- BIOGEOSCIENCES;
- 0231 Impacts of climate change: agricultural health;
- GEOHEALTH;
- 1622 Earth system modeling;
- GLOBAL CHANGE;
- 1630 Impacts of global change;
- GLOBAL CHANGE