Water Quality Monitoring Data for Ensemble Model Validation

Xu Chen's figshare dataset contains water quality, water level, and rainfall data from three monitoring stations in China's South-to-North Water Diversion Project. The 5.5 KB XLS file was last updated in April 2026. It supports research on a knowledge-data coupled framework for anomaly detection and multi-step prediction.

Use Cases

• Validate Stacking ensemble model performance on water quality, water level, and rainfall time-series data.
• Apply rolling-statistics-based detection rules to identify isolated false data points in water quality observations.
• Test difference-threshold rules for detecting rapid-change segments in the high-frequency monitoring data.
• Analyze the separation of pseudo anomalies from true anomalies characterized by sustained structural deviations.
• Use the multi-station data to assess model tracking ability for nonlinear temporal patterns.

Strengths

• Data is sourced from three operational monitoring stations in a major water diversion project.
• Framework validation integrates both knowledge-driven categorization and data-driven ensemble learning.

Limitations

• The dataset is very small at 5.5 KB, indicating extremely limited sample size for model training.
• Specific row counts and temporal coverage of the time-series are not provided.
• Geographic scope is limited to three stations in one specific water diversion project.

Provenance

Source

Monitoring stations in Phase II of the Eastern Route of China's South-to-North Water Diversion Project.

Collection Method

High-frequency monitoring data collection.

Time Range

null

Freshness

null

Geography

China's South-to-North Water Diversion Project.