← 2026-07-11

Hailstone (Upper Buffalo) — Daily Analysis: 2026-07-15

Classification: correct Confidence: Model confidence remains high. The physics engine successfully predicted the magnitude of a small rise in wet conditions, reinforcing the validity of the current calibration after previous adjustments.

Event Summary

The physics model accurately predicted a minimal peak rise (20.3 CFS vs 20.9 CFS actual) following an evening rainfall event in wet antecedent conditions, with only minor magnitude error.

Prediction vs Actual

Metric Predicted Actual Error
Peak CFS 20 CFS 21 CFS -1 CFS
Total rise 2.6 CFS

Band Contributions

Band Zone Precip Predicted Rise Intensity

Analysis

The physics prediction was highly accurate in terms of magnitude, underestimating the peak by only 0.6 CFS (-2.9%). Given the small basin response and the 'Wet' antecedent moisture condition, this level of precision confirms that the current response coefficient (234.6 CFS/inch) and moisture multiplier (1.5) are well-calibrated for low-flow events. The model correctly identified that the rainfall would result in only a negligible rise, avoiding any false positive alarm.

The timing error of 10.4 hours is notable but less critical for magnitude calibration. The rainfall occurred primarily in the evening (18:05-20:05 UTC), while the predicted peak was placed the following morning. This discrepancy likely stems from the fixed 1.5-hour lag parameter not fully capturing the subsurface or slower channel response times during wet soil conditions, or simply the smoothing effect of the dispersion model. However, since the magnitude was spot-on, the primary hydraulic response coefficient does not require adjustment.

Coefficient Adjustments

No changes made.

← 2026-07-11