Detection and Simulation of Urban Heat Islands Using a Fine-Tuned Geospatial Foundation Model for Microclimate Impact Prediction

TL;DR

This paper demonstrates how geospatial foundation models can accurately predict urban heat islands and assist in climate mitigation planning, especially in data-scarce urban areas, by fine-tuning on local data and simulating future scenarios.

Contribution

It introduces a method for fine-tuning geospatial foundation models to predict urban heat patterns and evaluate mitigation strategies in underserved regions.

Findings

Foundation models accurately predict urban heat island effects.

Fine-tuning improves local prediction accuracy.

Model simulations support climate resilience planning.

Abstract

As urbanization and climate change progress, urban heat island effects are becoming more frequent and severe. To formulate effective mitigation plans, cities require detailed air temperature data, yet conventional machine learning models with limited data often produce inaccurate predictions, particularly in underserved areas. Geospatial foundation models trained on global unstructured data offer a promising alternative by demonstrating strong generalization and requiring only minimal fine-tuning. In this study, an empirical ground truth of urban heat patterns is established by quantifying cooling effects from green spaces and benchmarking them against model predictions to evaluate the model's accuracy. The foundation model is subsequently fine-tuned to predict land surface temperatures under future climate scenarios, and its practical value is demonstrated through a simulated inpainting that highlights its role for mitigation support. The results indicate that foundation models offer a powerful way for evaluating urban heat island mitigation strategies in data-scarce regions to support more climate-resilient cities.