Annotated Water Quality Anomalies from South-to-North Water Diversion

A 5.5 KB Excel file containing manually identified abnormal data points from high-frequency water quality, water level, and rainfall monitoring. The dataset was created by Xu Chen to support research on reducing false alarms in water quality early warning systems and was last updated in April 2026.

Use Cases

• Train anomaly classifiers using labeled abnormal points from water quality, water level, and rainfall time-series.
• Validate rolling-statistics or variance-based detection rules against expert-identified pseudo and true anomalies.
• Benchmark ensemble learning models for multi-step prediction on sequences with known anomaly segments.

Strengths

• Dataset is openly licensed under CC-BY-4.0.
• Anomalies are systematically categorized into pseudo and true types based on defined identification strategies.

Limitations

• The dataset is very small at 5.5 KB, indicating a limited sample of annotated points.
• Specific row counts and column names for the underlying time-series data are unknown.
• Geographic coverage is limited to three monitoring stations on a single water diversion project.

Provenance

Source

figshare, uploaded by author Xu Chen.

Collection Method

Manual annotation of anomalies from high-frequency monitoring data.

Time Range

null

Freshness

Last updated in April 2026.

Geography

Three monitoring stations in Phase II of the Eastern Route of China’s South-to-North Water Diversion Project.