Naiad: Novel Agentic Intelligent Autonomous System for Inland Water Monitoring

TL;DR

NAIAD is an innovative agentic AI system that uses LLMs and external tools to provide comprehensive, natural-language-driven inland water monitoring insights from Earth Observation data, suitable for both experts and non-experts.

Contribution

It introduces a holistic AI assistant leveraging LLMs, external analytical tools, and retrieval-augmented generation for integrated inland water quality monitoring.

Findings

Achieved over 77% correctness and 85% relevancy on a dedicated benchmark.

Demonstrated strong adaptability and robustness across diverse query types.

Gemma 3 and Qwen 2.5 models offer optimal balance between efficiency and reasoning.

Abstract

Inland water monitoring is vital for safeguarding public health and ecosystems, enabling timely interventions to mitigate risks. Existing methods often address isolated sub-problems such as cyanobacteria, chlorophyll, or other quality indicators separately. NAIAD introduces an agentic AI assistant that leverages Large Language Models (LLMs) and external analytical tools to deliver a holistic solution for inland water monitoring using Earth Observation (EO) data. Designed for both experts and non-experts, NAIAD provides a single-prompt interface that translates natural-language queries into actionable insights. Through Retrieval-Augmented Generation (RAG), LLM reasoning, external tool orchestration, computational graph execution, and agentic reflection, it retrieves and synthesizes knowledge from curated sources to produce tailored reports. The system integrates diverse tools for weather data, Sentinel-2 imagery, remote-sensing index computation (e.g., NDCI), chlorophyll-a estimation, and established platforms such as CyFi. Performance is evaluated using correctness and relevancy metrics, achieving over 77% and 85% respectively on a dedicated benchmark covering multiple user-expertise levels. Preliminary results show strong adaptability and robustness across query types. An ablation study on LLM backbones further highlights Gemma 3 (27B) and Qwen 2.5 (14B) as offering the best balance between computational efficiency and reasoning performance.