Each crossing floods at its own rain signature

1One gauge, every 15 minutes

The JHUB weather station at JKUAT. Its rain column under-reports about 17×, so we read the gauge's running daily total instead; its second gauge rises with the sun and falls at night, so we ignore it.

2Every crossing has a bucket

Rain fills the bucket; it drains by half every so often. That half-life is the crossing’s signature: a campus culvert empties in minutes, a road dip in an hour or two, a river bridge over half a day. A crossing floods when its bucket passes its trigger (red dots).

3Reports set the trigger

A “clear” report at 19 mm says the trigger is above 19; a “flooded” report at 36 mm says it’s below 36. Each report is weighed (riders can be wrong), and the band is where the trigger most likely sits. Try it on a bridge nobody has reported yet:

Honest about the data

• Triggers start from one storm.

  The Kimbo–Matangi band comes from a single dated news report. Wide bands say so on screen, and every tap narrows them.

• Faulty gauges are caught, not trusted.

  Gaps, spikes, flatlines and counters that run backwards are flagged. A crossing with a silent gauge gets NO CALL, never CROSS.

• Calls err towards caution.

  A flooded report in the last 30 minutes forces at least WAIT. A clear report never forces CROSS.

• Locations are approximate.

  Crossings come from OpenStreetMap; the exact flooded spot on Kimbo–Matangi still needs a rider to pin it.

What exists today