A new study published on arXiv explores the use of AlphaEarth and TESSERA embeddings for fine-scale Local Climate Zone (LCZ) mapping in Switzerland. Researchers compared these embeddings with traditional Sentinel-1/2 composites, finding that both embedding types showed promising potential for upscaling coarse LCZ maps to a 10-m resolution using a U-Net architecture. The TESSERA embeddings consistently outperformed both Sentinel-1/2 and AlphaEarth in the conducted experiments, though transferring models across different years remains a challenge. AI
IMPACT Demonstrates potential for AI embeddings to streamline urban climate modeling and mapping processes.
RANK_REASON The cluster contains an academic paper detailing a new methodology for climate zone mapping using AI embeddings.
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
