流域水质预警体系研究与应用进展

流域水质预警是水质监控管理的重要组成部分，完善的流域水质预警体系可为流域水污染处理提供充足时间和所需信息，有效削减水污染对水环境造成的影响.对国内外的流域水质预警体系研究进展分别从预警体系目标、预警指标、预警阈值及模拟模型4个方面进行了综述，并且对目前国内外典型流域水质预警体系进行了分析.结果表明:流域水质预警体系目标主要从流域水体功能和水体污染风险2个方面进行考虑；预警指标可分为物化和生物毒性指标2类，生物毒性指标可弥补物化指标的不足，但不能完全替代，二者联合并施是未来预警指标筛选的发展方向；预警阈值的制定依据环境质量标准、排放标准、水质现状及污染源特征等信息，采用单指标多次报警、多指标联合报警或物化与生物毒性指标联合报警等方式，可提高预警体系的稳定和可靠性；污染物迁移转化模型已有一些成熟应用，但大部分模型依然需要大量的参数和历史数据才能保证准确性，因此对于大型流域难以适用.目前我国还未有国家层面完整的流域水质预警体系和相关流程在应用，特别是缺乏可靠的污染物迁移转化模型.未来应从建立典型流域水质预警体系着手，集成地方流域成熟应用的污染物迁移转化模型，逐步完善国家层面的流域水质预警体系，为我国流域水质安全提供保障. 

Advances in Watershed Water Quality Early-Warning System

Watershed water quality early-warning is an important component of water quality monitoring and management. An integral water quality early-warning system can provide sufficient time and necessary information for water pollution treatment, which can effectively reduce the impact of pollution on the water environment. Research on domestic and foreign watershed water quality early-warning systems was reviewed from four aspects:objective, indicator, threshold and simulation model. Furthermore, typical systems were analyzed. Existing research shows that the objectives of water quality early warning systems were mainly considered from only two aspects:water body function and water pollution risk. Early-warning indicators can be divided into two categories:physicochemical and biological toxicity indicators. Biological toxicity indicators can compensate for the shortage of physicochemical indicators but cannot completely replace them. The combination of physicochemical and biological toxicity indicators is the development direction of early-warning indicator. The threshold should be based on environmental quality standards, emission standards, water quality status and pollution source characteristics. The stability and reliability of the system rely on the alarm mode:single indicator multiple alarm, multi-indicator joint alarm, physicochemical and biological toxicity indicators joint alarm, etc. Some mature pollutant transport and transform models have been applied in the system. However, most of the models require large numbers of parameters and historical data to ensure their accuracy. It is difficult to apply to a large watershed. China lacks a complete national early-warning system and related processes in application, especially a reliable simulation model. In the future, establishing a system of typical watershed and integrating local mature model should be the first steps towards enhancing the national water quality early-warning system.