IoT-based Fine-scale Urban Temperature Forecasting

Authors: Jingchao Yang*, George Mason University, Manzhu Yu, Penn State University, Qian Liu, George Mason University, Chaowei Yang, George Mason University
Topics: Geographic Information Science and Systems, Spatial Analysis & Modeling
Keywords: IoT, Machine Learning, Forecasting, Temperature, Weather
Session Type: Virtual Paper
Day: 4/11/2021
Start / End Time: 11:10 AM / 12:25 PM
Room: Virtual 7
Presentation Link: Open in New Window
Presentation File: No File Uploaded

Accurate weather prediction, especially for temperature, is critical for decision-making regarding energy consumption, health risks, and economics. Regional numerical weather prediction (NWP) models produce operational-level temperature forecasts based on local atmospheric circulation conditions. They suffer from data- and computational intensity, resulting in low availability of high-spatiotemporal resolution. A data fusion technique is embedded in the proposed framework to address this shortcoming, integrating measurements from the Internet of Things (IoT) with a high spatiotemporal resolution with observations from weather stations. The framework utilizes a Long Short-Term Memory (LSTM) network to predict surface temperature from the fusion dataset for four major cities in the U.S. (Los Angeles, New York City, Atlanta, and Chicago). The predictive framework achieves an average RMSE of 1.72 ˚C and an average R2 of 0.97 using the past 24 hours to predict the future 12 hours for Los Angeles. A similar but not as robust performance is achieved for the other three cities. The transfer learning is adopted to leverage the pre-trained model from regions with a higher number of observation stations to predict regions with fewer stations. The transferable model improved the predicting MAE for regions with data scarcity up to 26%.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login