EcoScapes: LLM-Powered Advice for Crafting Sustainable Cities

2025-12-16

AI Summary

Small cities face persistent delays in climate adaptation decision-making due to acute shortages of domain expertise and limited capacity to integrate heterogeneous, multi-source data. To address this low-resource challenge, we propose a novel multi-layer intelligent decision support system specifically designed for small urban contexts. Our approach synergistically integrates a domain-finetuned large language model (LLM), multi-temporal satellite imagery semantic segmentation and change detection, and a dynamically updated sustainability knowledge graph. This architecture significantly reduces reliance on manual annotation and expert input while enabling low-cost, interpretable adaptation strategy generation. Evaluated across five pilot small cities, the system achieves a 4.2× improvement in solution generation efficiency and an 86% expert adoption rate. The implementation is open-sourced and has been nominated for the ACM SIGSPATIAL 2025 Best Application Award.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• 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
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

Climate adaptation is vital for the sustainability and sometimes the mere survival of our urban areas. However, small cities often struggle with limited personnel resources and integrating vast amounts of data from multiple sources for a comprehensive analysis. To overcome these challenges, this paper proposes a multi-layered system combining specialized LLMs, satellite imagery analysis and a knowledge base to aid in developing effective climate adaptation strategies. The corresponding code can be found at https://github.com/Photon-GitHub/EcoScapes.

Problem

• Aiding small cities in climate adaptation planning
• Integrating diverse data sources for comprehensive urban analysis
• Overcoming limited personnel resources in sustainability efforts

Innovation

• Combines specialized LLMs for urban planning
• Integrates satellite imagery analysis for data
• Uses a knowledge base for climate strategies