Freshwater Fishpond Multisensor Water Quality Dataset for Multiclass Degradation Classification (2153 Samples)

This dataset contains multisensor water quality measurements collected from a freshwater earthen fishpond environment. The dataset consists of 2153 samples and includes seven physicochemical parameters commonly used in aquaculture water quality monitoring:

• Temperature (°C)
• pH
• Dissolved Oxygen (mg/L)
• Turbidity (NTU)
• Electrical Conductivity (µS/cm)
• Total Dissolved Solids (mg/L)
• Oxidation–Reduction Potential (mV)

Each sample is labeled into one of four water quality degradation levels:

• Normal
• Caution
• Warning
• Severe

The class labels were generated using a multi-parameter scoring protocol based on cumulative deviations from recommended aquaculture water quality ranges. The dataset was used for multiclass machine learning classification experiments in water quality degradation assessment.

Number of Samples: 2153
Number of Features: 7
Number of Classes: 4