FuseTen: A Generative Model for Daily 10 m Land Surface Temperature Estimation from Spatio-Temporal Satellite Observations

2025-07-30

AI Summary

To address the longstanding trade-off between spatial and temporal resolution in remote sensing land surface temperature (LST) products, this paper proposes a nonlinear generative fusion framework that integrates multi-source observations—Sentinel-2 (10 m), Landsat 8 (30 m), and Terra MODIS (1 km)—to generate daily 10 m LST maps. The method innovatively incorporates a physics-guided mean supervision strategy, jointly leveraging channel attention, instance normalization, and a PatchGAN discriminator to enhance reconstruction accuracy and preserve fine-scale spatial details. Quantitative evaluation across multi-temporal test cases demonstrates an average improvement of 32.06% over linear baselines in standard metrics (e.g., RMSE, SSIM), with a 31.42% gain in visual fidelity. The proposed approach consistently outperforms state-of-the-art fusion methods, enabling high spatiotemporal resolution LST monitoring critical for fine-grained climate applications such as urban heatwave and drought analysis.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multimodal Learning

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Urban heatwaves, droughts, and land degradation are pressing and growing challenges in the context of climate change. A valuable approach to studying them requires accurate spatio-temporal information on land surface conditions. One of the most important variables for assessing and understanding these phenomena is Land Surface Temperature (LST), which is derived from satellites and provides essential information about the thermal state of the Earth's surface. However, satellite platforms inherently face a trade-off between spatial and temporal resolutions. To bridge this gap, we propose FuseTen, a novel generative framework that produces daily LST observations at a fine 10 m spatial resolution by fusing spatio-temporal observations derived from Sentinel-2, Landsat 8, and Terra MODIS. FuseTen employs a generative architecture trained using an averaging-based supervision strategy grounded in physical principles. It incorporates attention and normalization modules within the fusion process and uses a PatchGAN discriminator to enforce realism. Experiments across multiple dates show that FuseTen outperforms linear baselines, with an average 32.06% improvement in quantitative metrics and 31.42% in visual fidelity. To the best of our knowledge, this is the first non-linear method to generate daily LST estimates at such fine spatial resolution.

Problem

• Estimating high-resolution daily land surface temperature
• Bridging satellite spatial-temporal resolution trade-offs
• Improving accuracy in climate-related surface condition analysis

Innovation

• Generative model fuses multi-satellite data
• Uses attention and normalization modules
• PatchGAN discriminator ensures realistic outputs