This project focuses on predicting the Urban Heat Island (UHI) effect in urban centers using satellite-derived data, specifically Land Surface Temperature (LST) from Landsat imagery. The goal is to provide data-driven insights into how urbanization and surface characteristics influence localized temperature increases, which are becoming increasingly severe due to climate change.
Climate change is intensifying extreme heat events, particularly in urban areas where concrete and asphalt dominate the landscape. These environments absorb and retain more heat than natural ecosystems, creating the Urban Heat Island effect—a major environmental and public health concern.
Understanding and predicting UHI dynamics is critical to:
- Mitigating health risks for vulnerable urban populations,
- Informing sustainable urban planning,
- Supporting climate-resilient infrastructure development.
This project directly supports the following United Nations SDGs:
-
🛡️ SDG 3 – Good Health and Well-being:
UHI exacerbates heat-related illnesses. Accurate prediction can inform early warnings and public health interventions. -
🌆 SDG 11 – Sustainable Cities and Communities:
By analyzing spatial heat patterns, city planners can design greener, more resilient urban environments. -
🌡️ SDG 13 – Climate Action:
Modeling urban heat dynamics contributes to broader efforts in understanding and adapting to climate change. -
🌳 SDG 15 – Life on Land:
Understanding land surface changes and their thermal impacts helps guide urban greening efforts that support biodiversity and ecosystem health.
The project leverages multi-temporal satellite data, particularly thermal infrared bands from Landsat, to estimate and analyze land surface temperatures in urban areas. It integrates geospatial processing and machine learning techniques to detect spatial patterns and predict UHI intensities.
- Predict spatial variation of LST in urban environments.
- Identify areas most affected by UHI for targeted intervention.
- Provide a reproducible workflow for researchers and planners.
This repository includes:
- Jupyter notebooks for data preprocessing and modeling,
- Documentation of methodology and results,
- Scripts for extracting, cleaning, and transforming satellite-based LST data.
The codebase and methodology are open for use and extension in other regions or applications. This is part of a broader effort to make climate and urban planning data more accessible and actionable.