Related Experiment Video
Updated: Dec 21, 2025
Early Detection of Cyanobacterial Blooms and Associated Cyanotoxins using Fast Detection Strategy
Published on: February 25, 2021
Use statistical machine learning to detect nutrient thresholds in Microcystis blooms and microcystin management
Kun Shan1, Xiaoxiao Wang2, Hong Yang3
1Chongqing Key Laboratory of Big Data and Intelligent Computing, Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, Chongqing 400714, China; CAS Key Lab on Reservoir Environment, Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, Chongqing 400714, China.
Reducing total nitrogen (TN) and total phosphorus (TP) is key to controlling harmful cyanobacterial blooms and microcystins (MCs). Specific nutrient targets vary by lake, with temperature also impacting MCs risk.
Area of Science:
- Environmental Science
- Ecotoxicology
- Water Quality Management
Background:
- Cyanobacterial blooms, particularly Microcystis, are increasing due to nutrient enrichment and climate change.
- Microcystins (MCs) pose a significant risk, but understanding nutrient-MC relationships is limited by complex interactions and data constraints.
Purpose of the Study:
- To develop a Bayesian modeling framework to analyze nutrient-MC relationships.
- To identify potential nutrient control targets for reducing cyanotoxin risks in lakes.
Main Methods:
- A Bayesian modeling framework was developed incorporating intracellular and extracellular microcystins.
- The model was applied to data from three Chinese lakes to estimate nutrient thresholds for total nitrogen (TN) and total phosphorus (TP).
- Lake-specific nutrient thresholds were determined using a Bayesian updating process.
Main Results:
- Dual reduction of TN and TP is recommended for managing cyanotoxin risks.
- Achieving specific TP thresholds (0.10 mg/L or 0.05 mg/L) can suppress Microcystis biomass, with stricter targets needed for Dianchi Lake.
- A TN threshold of 1.8 mg/L was estimated to maintain MCs below 1.0 μg/L, though temperature increases can counteract this effect.
Conclusions:
- The study provides an effective method for integrating empirical knowledge into data-driven models for water resource management.
- Identified nutrient thresholds offer practical guidance for water quality managers to mitigate cyanotoxin risks.
