Rooftop Solar Energy Potentials (SEP) in Hong Kong

This repository contains the code and associated datasets for the research article entitled "Rooftop solar enables carbon-neutral transition in Hong Kong’s building sector".

🖥️Machine Learning Model:

The code with pretrained machine learning model and sample datasets for global solar radiation (GSR) estimation are available now.

• Pre-trained machine learning model for cloud-attenuated GSR estimation: Model
• Python Code for daily GSR estimation: Code
• Sample datasets for the model inputs:
  • Daily clear-sky GSR: Day one
  • Daily cloud optical thickness: Day one, Day two

📊Datasets:

The datasets, including high-resolution (0.5m) annual GSR, district-scale annual SEP, building-level annual SEP, building-level analysis on decarbonization target, are available on Zenodo at Link

🌍Web Application:

The web application for visualizing the high-resolution (0.5 m) GSR dataset and district- and building-level rooftop SEP is available at Link

📖Publications:

• Citation for our research articles: [will be added upon acceptance]
• Citation for our generated datasets: Tong, L. Y., & Li, Z. (2025). Rooftop solar enables carbon-neutral transition in Hong Kong's building sector [Data set]. Zenodo. https://doi.org/10.5281/zenodo.17770925