INCONSISTENcy

[commonchtgpt]

I'm working on a SWAT model for the Gadarwara catchment (~2,184 km²) in the Shakkar sub-basin (Madhya Pradesh), and I’m facing a persistent issue with extremely high and flat discharge outputs (~5,000 m³/s) — even when rainfall is very low or zero.

I've tried two different simulations, and in both cases, the results were unrealistic:

🔹 Simulation 1:

• Stream threshold: 7,500 ha → resulted in 13 subbasins

• HRU thresholds: Land use = 12%, Soil = 15%, Slope = 15%

• Output: Discharge was ~1800 m³/s continuously, even without rainfall

🔹 Simulation 2:

• Stream threshold: Reduced to 2,000 ha → 53 subbasins

• HRU thresholds updated to 10% for land use, soil, and slope

• Slope classification: 0–3%, 3–8%, >8%

• Land use raster reclassified properly (no NoData)

• Still got flat, high discharge (~5,000 m³/s)

Rainfall Input:

• From 19 stations