Wenpan Li

The expansion of China’s modernized eco-environmental monitoring networks has generated vast amounts of data. Consequently, traditional, expertise-reliant analysis is increasingly ill-suited for agile regulatory decision-making. Although large language models (LLMs) present a promising alternative, their practical deployment remains limited by domain-specific knowledge gaps, hallucinations and an inherent difficulty in managing multi-faceted ecological tasks. This study introduces EnvSentry, a novel multi-engine collaborative LLM framework designed for intelligent eco-environmental monitoring and governance. EnvSentry coordinates reasoning, instruction, and multimodal engines, supported by a dynamic, vector-indexed knowledge base and retrieval-augmented generation (RAG) to ensure factual veracity. By transitioning operational workflows from fragmented, latent batch processing to integrated, real-time intelligent agent chains, the system achieves a closed-loop capability of intent recognition, data retrieval, and quality control. The model was evaluated across distinct environmental contexts, specifically water quality anomaly detection and air quality forecasting. Results show that EnvSentry yields higher analytical precision and attribution rates than baseline methods, while compressing decision-making latency from hours to seconds. Relative to baseline models, EnvSentry achieves a 25% improvement in water quality attribution accuracy (50% to 75%), a 90% reduction in decision making latency for anomaly detection, and a 10% absolute gain in data anomaly detection accuracy. In air quality forecasting, it reduces expert judgment time from 60 to 20 min and attains >85% agreement with expert forecasts when used by non-specialist personnel. These improvements suggest a practical shift in eco-environmental monitoring—moving from fragmented, reactive measures toward an integrated and proactive system. Consequently, this approach offers a viable path toward data-driven autonomous ecological management.

To investigate the monitoring history and long-term change trends in surface water quality in China since the reform and opening up, the history of surface water environment monitoring is summarized, including monitoring scope, monitoring methods, and technical requirements. Temporal and spatial patterns of surface water quality in China were analyzed based on the monitoring results. In the past 40 years, the monitoring targets for surface water quality have been continuously improved, the frequency of monitoring has become more science-based, and the monitoring indicators are now comprehensive. Overall, the temporal change trend in surface water quality has followed a “fluctuating changes stage—rapid deterioration stage—fluctuations stalemate stage—rapid improvement stage” pattern. However, the current regional surface water quality is still in a polluted status, and there is a gap between surface water quality status and the goal of building a well-off society. At present, China’s surface water pollution is prone to high numbers of incidents and the treatment of surface water pollution has entered a crucial stage. The potential for the continuous reduction of major pollutant discharges has become more challenging, and the marginal cost for pollution control has increased. It is very difficult to comprehensively solve the outstanding water environment problems. In addition to strengthening the existing work on surface water quality control, it is also necessary to strengthen the work of risk identification, early warning, and regulation implementation of the surface water environment. During the 14th year plan period (2021–2025), the overall planning on water resources, water ecology, and water quality will be implemented, and beautiful rivers and lakes will be created.

Conserving aquatic ecosystems requires efficient tools to accurately assess the biodiversity of aquatic species. However, existing knowledge is insufficient in terms of the reliability and the comparability of methods measuring fish diversity. Environmental DNA (eDNA), as a promising method, was used to detect fish taxa in this study. We optimized the eDNA method in the laboratory, and applied the optimal eDNA method to survey fish diversity in a natural aquatic life reserve. We simulated necessary steps of the eDNA method in the lab to increase the confidence of the field survey. Specifically, we compared different eDNA sampling, extraction, and sequencing strategies for accurately capturing fish species of the target area. We found that 1L water samples were sufficient for sampling eDNA information of the majority taxa. The filtration was more effective than the centrifugal precipitation for the eDNA extraction. The cloning sequencing was better than the high-throughput sequencing. The field survey showed that the Shannon–Wiener diversity index of fish taxa was the highest in Huairou Reservoir. The diversity index also showed seasonal changes. The accuracy rate of detecting fish taxa was positively correlated with the eDNA concentration. This study provides a scientific reference for an application of the eDNA method in terms of surveying and estimating the biodiversity of aquatic species.