LATInsights: Democratising Land Surface Temperature Analytics for Urban Climate Governance through Conversational AI.

Urban climate adaptation planning requires synthesizing complex Earth Observation data, yet traditional Planning Support Systems face an “implementation gap” due to technical barriers preventing non-specialist planners from accessing specialized geospatial workflows.We present LATInsights, a conversational agent system that addresses this gap through natural language interaction, enabling municipal planners to perform sophisticated thermal analytics without remote sensing expertise.The system implements a novel two-tier architecture: a stateful LangGraph-based orchestration engine managing semantic reasoning and multi-turn dialogue, decoupled from a stateless Model Context Protocol Server exposing domain-specific geospatial tools. This separation enables independent scalability while ensuring analytical accountability through structured planning and three-tier error recoverywith Large Language Model-driven re-planning.We demonstrate operational efficacy through Milan’s Via Neera cycling corridor case study, where the system autonomously orchestrates multi-scenario thermal simulations.This work demonstrates how conversational AI can democratize geospatial intelligence for evidence-based urban climate governance.