The "time to floatable / too low" countdowns come straight from the river's own record: every time a gauge sat at this level in this season, how long did it really take to fall to the threshold? (Rebuilt September 2026.)
For each gauge, each starting flow band and each threshold we keep the distribution of hours it actually took to first reach that threshold, measured from every falling-limb hour of 2014–2026 (paths interrupted by a rise of more than 10% are excluded, exactly the rule the scorecard grades by). The countdown is the middle of that history for your current flow, interpolated between bands; the small "typically X–Y" line under it is the 25th–75th percentile — about half of past recessions from this level finished inside it. Boxley and Ponca use season-specific tables; the others use the whole record. No live-event anchor: a three-hour slope at these flows is mostly gauge noise, and every version with one tested worse.
Why the rebuild: the previous curve (below, still used on the other creeks) learned its decay rates only from hours when the river was actively dropping and skipped every flat or bumpy hour — so on the Buffalo mainstem it ran 1.3–2.3× fast, worst at St. Joe and Harriet. On held-out years the table lands within a few percent of the real wait at all five gauges and its range covers 47–53% of cases, as designed. The countdowns assume no significant new rain: small showers are inside the numbers; a real storm restarts the clock.
A falling river doesn't drop at one rate — it falls fast when high and slower as it approaches baseflow. For each gauge we store a table of decay rates k by flow band (fraction lost per hour). The countdown integrates those bands from the current flow down to each threshold, and anchors to the actual event: if this recession is running faster or slower than the historical median, the whole forecast scales to match (clamped to 0.5–2×).
| Gauge | Falling from high water | Falling near floatable |
|---|---|---|
| Boxley | 89% per day near 1,500 cfs | 40% per day near 350 cfs |
| Ponca | 64% per day near 1,600 cfs | 21% per day near 200 cfs |
| Pruitt | 55% per day near 2,000 cfs | 21% per day near 200 cfs |
| St. Joe · settles to a ~76 cfs baseflow | 49% per day near 8,000 cfs | 18% per day near 200 cfs |
| Harriet · settles to a ~103 cfs baseflow | 50% per day near 9,370 cfs | 21% per day near 200 cfs |
Median rates shown; the forecast carries the measured 25th–75th percentile spread, and the confidence label reflects it (see the confidence page).
Paddler folklore says the water drops faster in summer than winter. We measured it: in identical flow bands, rain-free summer recessions run 1.2× faster at Boxley and Pruitt, ~1.35× at St. Joe and Harriet, and 2.3× at Ponca than winter ones — the trees really are drinking the river (evapotranspiration). We then tested adding a calendar multiplier to the countdowns — and it made them worse, because the live anchor above already measures each recession's actual speed and adapts, which beats any calendar rule. We publish that null result here on purpose. (September 2026 postscript: with the anchor gone, the table above carries season-specific entries at Boxley and Ponca, where the winter/summer difference held up on held-out years; at Pruitt, St. Joe and Harriet winter recessions still run somewhat slower than the all-season table says.)
What DID ship: seasonal floors. The level a river settles toward swings enormously — St. Joe typically holds above ~600 cfs in March but falls to ~50 by September. A spring countdown to a summer-only level is a promise the river won't keep: in backtest, targets below the month's typical floor were reached within 10 days only 5–16% of the time. Countdowns to such targets now say "typically settles near ~X this time of year" instead of a number.
Each countdown carries high / medium / low, set by the weakest of three tests. Spread: the forecast carries the measured 25th–75th percentile decay range; if the resulting time window is tight relative to its middle (within about ±40%), that's high — a window wider than the estimate itself is low. Horizon: under 3 days out is high, under a week medium, beyond a week low — past that, weather is the dominant unknown, not the curve. Data: the label is capped by how many recent readings anchor the fit (8+ for high, 4+ for medium). A 6-day-out countdown can never claim high confidence, no matter how clean the curve looks — that's deliberate.
Only when a rise is genuinely imminent at this gauge: local rain building there, or a tracked wave arriving within about 3 hours. A wave that's still half a day out no longer hides the countdown — you'll see both together ("floatable in ~6h · wave arriving ~9 PM"), which is the honest picture. In replay, the old any-upstream-prediction rule blacked out countdowns wrongly 97–99% of the time it fired on a falling river.
Every number on this page is generated from the deployed model files and an 11½-year hourly replay (2014-12 → 2026-06, baseline v6). Generated 2026-09-03T17:47:59+00:00. Experimental — never rely on a single number for safety decisions.