Event 2 — March 7, 2026: Major widespread storm across entire watershed.
This is the largest precipitation event of the study to date, with 1-2.5" falling across virtually all 37 HUC12s. The storm was widespread but showed clear spatial concentration in the Ponca and Pruitt zones.
Heaviest rainfall zones (gauge zone averages): - Pruitt zone: 2.022" (peak 0.803"/hr in Hoskin Creek HUC12 0207) — Cove Creek received 2.43", Hoskin Creek 1.614" - Ponca zone: 1.918" (peak 0.961"/hr in Whiteley Creek HUC12 0205) — Whiteley Creek received 2.545", Beech Creek 1.964", Smith Creek 1.244" - Ungauged: 1.419" — several sub-basins >1.4" - St. Joe zone: 1.358" — highly variable: Shop Creek 1.868", Left Fork Big Creek 1.615", Big Creek-Buffalo 1.661", Lick Creek 1.642", Cane Branch 1.421", Flatrock Creek 1.805", while others were 0.8-1.1" - Richland zone: 1.24" — Falling Water Creek 1.26", Headwaters Richland 1.219" - Bear Creek zone: 1.199" — Headwaters Bear Creek 1.247", Outlet Bear Creek 1.15" - Boxley zone: 1.135" - Harriet zone: 1.071"
Notable HUC12 hotspots (>1.5"): - Whiteley Creek (0205): 2.545" — peak 0.961"/hr ← EXTREME for Ponca zone - Cove Creek (0204): 2.43" — peak 0.74"/hr ← EXTREME for Pruitt zone - Beech Creek (0202): 1.964" — peak 0.768"/hr - Shop Creek (0101): 1.868" — peak 0.632"/hr - Flatrock Creek (0206): 1.805" — peak 0.989"/hr - Big Creek-Buffalo (0303): 1.661" - Lick Creek (0304): 1.642" - Left Fork Big Creek (0301): 1.615" - Hoskin Creek (0207): 1.614" — peak 0.803"/hr
Timing: Rain fell primarily during UTC hours 06:00-18:00 on 3/07 (CST midnight to noon). Peak intensities were concentrated around 05:00-07:00 CST (UTC 11:00-13:00). This is a morning storm.
Antecedent conditions: Critical context — the watershed was already moderately wet from Event 1 (3/05). Seven-day antecedent precipitation ranged from 0.7-1.45" across the watershed before today's rain began. The Boxley/Ponca/Pruitt zones were the wettest (1.0-1.45"). This pre-wetting substantially reduced soil moisture deficits and primed the watershed for an amplified response.
This is a Tier 2-3 event (Recreational to High Water) — the largest response observed in the study.
Boxley (07055646): - Height: 2.47 ft → 3.28 ft, +0.81 ft rise - Rise began ~05:45 CST (from 2.48 to 2.52), accelerated from ~09:00, peaked at 16:15 (3.28 ft) - Currently receding: 3.14 ft at EOD - Notably different behavior from Event 1: Rise was more gradual and sustained (no sharp spike), consistent with pools being already partially filled from Event 1 recession. The peak 1hr rise was only 0.16 ft/hr vs. 0.46 ft/hr in Event 1 — despite similar rainfall totals (1.135" vs 1.038"). This strongly validates the "pool and drop" local knowledge: with pools already full from Event 1, the response was smoother but equally large in total magnitude.
Ponca (07055660): - Discharge: 145 cfs → 2,120 cfs peak at 07:30 CST, +1,975 cfs (+1,362%) - This is an enormous response. Ponca exceeded the Optimal range ceiling (1,600 cfs) and briefly entered Flood territory. - Rise began ~03:00 CST (subtle, 147→153), accelerated dramatically at 05:00 (157→173→200→221→263→406→685→982→1230→1470→1950→2120 in 2.5 hours) - Currently receding: 424 cfs at EOD — still well within Optimal (200-1600 cfs) range - Source attribution: The massive Whiteley Creek rainfall (2.545" with 0.961"/hr peak) plus Beech Creek (1.964") fed directly into Ponca's tributaries. Boxley propagation also contributed but was a smaller component.
Pruitt (07055680): - Height: 4.04 ft → 8.55 ft, +4.51 ft rise - Discharge: 174 cfs → 2,610 cfs peak at 13:15 CST - Two-phase response clearly visible: - Phase 1 (06:00-09:00): Gradual rise from 4.05 to 4.28 ft — this is the leading edge of the Ponca propagation pulse and/or direct Pruitt-zone rainfall response - Phase 2 (09:15-13:15): Explosive rise — 4.28 to 8.55 ft in 4 hours. The 1.82 ft/hr peak hourly rise at 10:15 is extraordinary. - Pruitt exceeded the Optimal ceiling (2,000 cfs) and entered Flood range. It was above 2,000 cfs from approximately 12:00 to 15:30 (~3.5 hours). - Currently receding: 6.25 ft / 1,010 cfs — still well within Optimal. - Source attribution: The Cove Creek (0204) received 2.43" with peak 0.74"/hr, and Hoskin Creek (0207) received 1.614" with 0.803"/hr peak. These two HUC12s (123 km² combined) are the direct Pruitt-zone tributaries. The explosive Phase 2 rise is consistent with Ponca's flood pulse (which peaked at 2,120 cfs at 07:30) arriving at Pruitt ~5 hours later, combining with the massive direct Pruitt-zone runoff. - Pruitt gauge precip: 0.91" — consistent with the 2.022" zone average (gauge is at lower elevation in the zone).
St. Joe (07056000): - Height: 3.41 ft → 5.63 ft, +2.22 ft rise — STILL RISING at EOD - Discharge: 153 cfs → 1,150 cfs — STILL RISING at EOD - Rise pattern shows two distinct phases: - Slow rise 05:15-19:00: 3.42 → 4.13 ft (~0.71 ft over 14 hours, ~0.05 ft/hr) — this is from direct rainfall on the St. Joe zone's 14 HUC12s (1,342 km²) plus early Richland Creek contribution - Rapid acceleration 19:00-22:45: 4.13 → 5.63 ft (+1.50 ft in 3.75 hours, ~0.40 ft/hr) — this is the Pruitt propagation pulse arriving - St. Joe entered Optimal (>200 cfs) at approximately 08:15 and is still rising at EOD. - Signal separation — Richland Creek is rising: Richland surged from 1.16 ft to 2.55 ft (+1.39 ft), confirming rain was in the Richland sub-basin (HUC12s 0306, 0307 received 1.22-1.26"). The Richland rise peaked around 16:15-17:15 and is slowly receding. This is consistent with the initial slow St. Joe rise being partly from Richland inflow. - The late rapid acceleration at St. Joe starting ~19:00 is too fast to be only from Richland (which was already receding). This acceleration coincides with the expected arrival time of the Pruitt peak (Pruitt peaked at 13:15, +6 hours = ~19:15 at St. Joe), plus contributions from Big Creek, Little Buffalo, and Cave Creek complexes which received 0.8-1.87" of rain.
Harriet (07056700): - Height: 3.53 ft → 3.78 ft, +0.25 ft rise — slowly rising at EOD - Discharge: 172 cfs → 285 cfs — rising at EOD, now in Optimal range (>200 cfs) - Harriet entered Optimal at approximately 06:45 CST (206 cfs). - The rise has been gradual and steady, reflecting the integration of multiple upstream inputs plus direct Harriet-zone rainfall (1.071"). - Signal separation — Bear Creek surged: Bear Creek rose explosively from 2.37 ft / 9.31 cfs to 2.91 ft / 48.6 cfs starting at ~18:15 CST. This indicates independent rainfall in the Bear Creek watershed. QPE confirms: Headwaters Bear Creek received 1.247" and Outlet Bear Creek 1.15". The Bear Creek surge is arriving at Harriet in the late evening, contributing to the continued rise. - Important: Harriet will likely continue rising significantly overnight as the St. Joe flood pulse (currently at 5.63 ft / 1,150 cfs and still rising) propagates downstream.
Boxley — Event 2 calibration point: - Input: 1.135" in Boxley HUC12 (92 km²), peak 0.434"/hr - Antecedent: 1.443" 7-day (WET — Event 1 recession still ongoing, pools partially filled) - Output: +0.81 ft (2.47→3.28 ft) - Lag: Rain peaked ~06:00 CST (UTC 12:06), gauge peaked ~16:15 CST — ~10 hr lag - Comparison to Event 1: Event 1 had 1.038" with 0.751"/hr peak and produced +0.87 ft with ~9 hr lag. Event 2 had slightly more total rain (1.135") but lower peak intensity (0.434"/hr) and produced a similar rise (+0.81 ft) but with a smoother, more sustained hydrograph. The wet antecedent conditions offset the lower intensity. This is consistent with the "pool and drop" hypothesis — pre-filled pools reduced lag and buffering.
Ponca — Event 2 calibration point: - Input: 1.918" zone average across 298 km² direct + Boxley propagation; peak intensity 0.961"/hr in Whiteley Creek - Antecedent: ~1.0" 7-day (moderately wet) - Output: +1,975 cfs (145→2,120 cfs) - Lag: Rain peaked ~06:00 CST, gauge peaked ~07:30 CST — only ~1.5 hr lag for the direct-zone response - Transfer ratio: ~1,975 cfs from ~1.9" on 298 km² = ~1,040 cfs per inch per 300 km² = ~3.5 cfs/inch/km² under wet antecedent conditions - Comparison to Event 1: Event 1 produced only +95 cfs from 0.587" on 298 km² = ~162 cfs/inch/300 km² = ~0.54 cfs/inch/km². The ~6.5x amplification factor is due to (a) 3x more rain, (b) much higher intensity, (c) wet antecedent conditions. This is highly nonlinear — validates the expectation that karst-influenced catchments show threshold-type behavior.
Pruitt — Event 2 calibration point (direct zone only): - Input: 2.022" zone average across 123 km² direct, peak 0.803"/hr - Additional: Ponca propagation pulse (2,120 cfs peak arriving ~5 hrs later) - Antecedent: 1.452" 7-day in Cove Creek (very wet) - Output: +4.51 ft height / +2,436 cfs (174→2,610 cfs) - The two-phase response makes it difficult to separate direct vs. propagated contributions. The Phase 2 acceleration starting at 09:15 is likely the combined arrival of Ponca's pulse plus the peak of Cove Creek/Hoskin Creek direct runoff. - Lag (Ponca peak to Pruitt peak): Ponca peaked at 07:30, Pruitt peaked at 13:15 — 5.75 hr Ponca→Pruitt lag at flood flow (~1,500-2,000 cfs). Compare to Event 1: Ponca→Pruitt lag was ~14-15 hours at low flow (~150-180 cfs). Flood waves propagate 2.5x faster than low-flow events. This is a critical finding.
Richland Creek — Event 2 calibration point: - Input: 1.24" zone average across 174 km², peak 0.348"/hr - Antecedent: ~0.9-1.0" 7-day - Output: +1.39 ft (1.16→2.55 ft) — major response - Lag: Rain peaked ~07:00 CST (UTC 13:06), gauge appeared to peak ~16:15 CST — ~9 hr lag - Comparison to Event 1: Event 1 produced only +0.26 ft from 0.677" (moderate rain, drier antecedent). Event 2 produced +1.39 ft from 1.24" (2x rain on wetter ground). The 5.3x response amplification demonstrates significant nonlinearity.
Bear Creek — Event 2 calibration point (FIRST SUSTAINED DETECTION): - Input: 1.199" zone average across 239 km², peak 0.588"/hr - Antecedent: ~0.75" 7-day (relatively dry for this event) - Output: +0.59 ft height / +42 cfs (6.6→48.6 cfs) - Lag: Rain peaked ~07:00 CST (UTC 13:06), gauge surged at ~18:15 CST — ~11 hr lag - This is the first confirmed sustained response at Bear Creek. The detection threshold was >0.5" at >0.3"/hr (Event 1), so 1.2" at 0.588"/hr clearly exceeds it. The long 11-hour lag is notable — Bear Creek appears to have very high initial abstraction capacity (deep karst storage), but once exceeded, the response is relatively fast and sustained.
Event 2 Propagation Chain:
| Gauge | Peak Time (CST) | Peak CFS | Lag from Boxley | Lag from Previous | Est. Velocity |
|---|---|---|---|---|---|
| Boxley | ~16:15 3/07 | — (height only) | 0 hr | — | — |
| Ponca | 07:30 3/07 | 2,120 | -8.75 hr* | — | — |
| Pruitt | 13:15 3/07 | 2,610 | -3 hr* | 5.75 hr from Ponca | ~1.2 m/s |
| St. Joe | Still rising | >1,150 | — | >10 hr from Pruitt (ongoing) | — |
| Harriet | Still rising | >285 | — | — | — |
*Negative lags from Boxley indicate Ponca and Pruitt peaked BEFORE Boxley — because the heaviest rainfall was concentrated in the Ponca and Pruitt zones (Whiteley 2.545", Cove 2.43"), not the headwaters. This is fundamentally different from Event 1 where Boxley led.
Key propagation finding: Ponca→Pruitt propagation velocity at flood flow (~1,500-2,000 cfs average transit flow) is ~1.2 m/s, versus ~0.46 m/s at low flow (Event 1). This is a 2.6x increase in wave speed with high flows. For a ~25 km Ponca→Pruitt channel distance, this means: - Low flow: ~15 hr transit - Flood flow: ~5.75 hr transit
Pruitt→St. Joe propagation estimate: The rapid acceleration at St. Joe starting ~19:00 CST, roughly 5.75 hours after Pruitt's peak at 13:15, gives a similar propagation velocity for the Pruitt→St. Joe reach (~50 km) of approximately ~50,000m / 20,700s ≈ 2.4 m/s. However, St. Joe's rise also includes massive direct rainfall contributions (1,342 km² × 1.358" average), so this is an upper bound on propagation velocity. The true propagation component is mixed with direct runoff.
Event 2 occurred while the watershed was still in recession from Event 1. Key implications: - All gauges started today at elevated baseflows (Pruitt at 4.05 ft/174 cfs vs pre-Event-1 baseflow of 3.44 ft/64 cfs) - The wet antecedent conditions from Event 1 dramatically amplified Event 2 responses - The Ponca-Pruitt discharge relationship has normalized during high flow: Pruitt (2,610 cfs peak) vastly exceeds Ponca (2,120 cfs peak), as physically expected for a downstream gauge. The low-flow inversion is entirely absent at these flow levels, consistent with the hypothesis that the inversion is a low-flow phenomenon (karst losses and/or rating curve issues at low stage).
Events 1+2 combined watershed impact: - Total 7-day precipitation now exceeds 2" across most of the watershed and >3" in the Ponca/Pruitt zones - The watershed is now deeply saturated — any additional rainfall in the next 48 hours would produce extremely fast, amplified responses
Boxley smooth response validates "pool and drop" local knowledge. Event 1 (dry antecedent) showed a sharp biphasic response with 0.46 ft/hr max rise. Event 2 (wet, pools filled) showed a smooth sustained rise with only 0.16 ft/hr max rise despite similar rainfall. This is textbook pool-filling dynamics. Confidence in this local knowledge calibration target: Very High.
Nonlinear amplification is extreme. Ponca saw ~20x the discharge response for ~3.3x the rainfall compared to Event 1. Richland saw ~5x the height response for ~1.8x the rainfall. This confirms that the watershed has strong threshold behavior — below a critical saturation level, most rain is absorbed; above it, runoff efficiency increases dramatically.
Bear Creek 11-hour lag is surprisingly long for a 239 km² watershed receiving 1.2" with 0.588"/hr peak intensity. This suggests very high initial abstraction capacity in the Bear Creek drainage, likely deep karst storage. Once the threshold was exceeded, the rise was rapid (2.36→2.91 ft in ~1 hour).
St. Joe is still rising rapidly at EOD — the 0.7 ft/hr acceleration from 19:00-22:45 is the fastest rate observed at St. Joe in this study. With 1,150 cfs and still climbing, St. Joe may peak significantly higher overnight. This will be the first real test of the St. Joe detection/response system. The Richland Creek signal separation is working well — Richland peaked hours before the St. Joe acceleration, confirming the acceleration is from mainstem propagation + Big Creek/Little Buffalo/Cave Creek inputs, not Richland.
The Pruitt rating curve appears to be performing well at high stage. The height-discharge relationship traced smoothly from 4.04 ft/174 cfs through 8.55 ft/2,610 cfs, with no obvious discontinuities. The mid-event rating curve shift observed on 3/05 appears to have been corrected or the current curve handles the higher range well.
Ponca's rise was extraordinarily rapid — 157 cfs to 2,120 cfs in 2.5 hours (05:00-07:30). The Whiteley Creek HUC12 (160 km², 2.545" total, 0.961"/hr peak) is clearly a flashy contributor. This sub-watershed warrants particular attention for flood warning purposes — it can produce extreme flows at Ponca with very short lead time.