Rate times duration. The ingredients that set the rate, the configurations that stretch the duration, and why the storm that floods you is often the quiet one.

In this lesson you will

  • State the ingredients framework: rainfall rate from ascent, moisture and precipitation efficiency; total from rate and duration.
  • Explain how slow system motion, training echoes and backbuilding maximize rainfall duration over one point.
  • Describe warm rain processes and high precipitation efficiency, and identify where they dominate.
  • Identify the Maddox synoptic flash flood patterns and the role of deep moisture and weak cloud-layer shear.

Rate times duration

Doswell, Brooks and Maddox reduced flash flood forecasting to one paragraph in 1996, and thirty years later it is still the frame everybody uses:

Heavy precipitation is the result of sustained high rainfall rates.
In turn, high rainfall rates involve the rapid ascent of air
containing substantial water vapor and also depend on the
precipitation efficiency. The duration of an event is associated with
its speed of movement and the size of the system causing the event
along the direction of system movement.

Doswell, Brooks & Maddox 1996. Quoted as written.

Unpacked, that is two multiplications. The rate is ascent times moisture times precipitation efficiency. The total at a point is that rate times how long the rain sits over the point. Nothing else enters it. A storm with a spectacular rate that crosses you in eight minutes leaves less water than a mediocre one that will not move.

Which term usually decides it
Duration. Doswell and colleagues, citing Chappell, put it flatly: "it is slow system movement that dominates most cases of flash flood-producing heavy precipitation." Rates vary by a factor of a few between ordinary and extreme storms. Durations vary by a factor of twenty.

The three ways duration gets stretched

Slow system motion
The simplest case. The MCS is moving at ten miles an hour instead of forty, so every point underneath it spends four times as long in the rain. Corfidi vectors are how a forecaster anticipates this from a sounding, and a slow forecast motion in a discussion is a flood signal before any rain has fallen.
Training echoes
Individual cells moving along a line that is itself stationary, so cell after cell crosses the same ground like carriages over a level crossing. Each cell is brief. The point underneath them is in rain for hours. This is the configuration that produces the worst totals in the Southeast.
Backbuilding
New cells forming repeatedly on the upwind flank, so the system regenerates at its rear as fast as it decays at its front and the whole thing stays put over the ground even though every cell in it is moving. Quasi-stationary MCSs usually do this.
Two panels. On the left, five storm cells arranged along a line, each with an arrow pointing along the line in the same direction, and a fixed point on the ground beneath the line receiving all five in turn. On the right, the same five cells spread apart and moving across the line, with the ground point receiving only one. A rain total bar under each panel is tall on the left and short on the right.
Identical cells, identical rainfall rates, identical cell speeds. The only difference is whether the cells travel along the line or across it, and it is worth a factor of five in the total on the ground. WxAlerts

Warm rain, and the storm that floods quietly

Most of what you have learned about severe storms is about ice. Hail needs a freezing level and a deep cold layer above it; lightning needs graupel colliding with ice crystals. Warm rain skips all of that. In a cloud with a deep warm layer below the freezing level and high humidity through it, droplets grow by collision and coalescence, simply by bumping into one another, and they do it fast.

The consequence is high precipitation efficiency: a larger share of the water vapour drawn into the updraft reaches the ground as rain instead of being carried off in the anvil. A warm rain storm can out-rain a much more violent cold-topped one while producing modest lightning, no hail, and nothing that looks alarming.

Why this is on the radar lesson list
MRMS publishes a product called Probability of Warm Rain for one stated reason: "warm rain enhancement of rain rates is underestimated by current Z-R relationships." So the radar runs systematically low in exactly the conditions that produce the heaviest rainfall: deep tropical moisture anywhere it reaches, landfalling tropical systems well inland, and terrain-forced rain in a moist maritime flow. Lesson 6 is about what that does to a flood report, and it is the strongest argument in this course for not auto-rejecting one.

The Maddox patterns

Maddox, Chappell and Hoxit examined more than 150 intense convective precipitation events in 1979 and found that they were not random. The events sorted into a small number of recurring synoptic patterns, which meant a forecaster could recognize a flood setup the day before rather than the hour after.

  • Synoptic, tied to a slow-moving frontal system aloft.
  • Frontal, with convection repeatedly firing along a stalled surface boundary.
  • Mesohigh, where a previous storm complex has left an outflow boundary that the next round trains along.
  • A separate western type, driven by terrain, which does not apply here.

What all of them share is more useful than the categories: convective heavy rain, very high surface dewpoints, a deep moist layer, and weak to moderate cloud-layer shear. That last one is the counter-intuitive item. Strong shear is what module 2 taught you to look for in a severe setup, and strong shear also ventilates a storm and moves it along. The flood setup is the one where the shear is not doing that.

One caveat the paper states itself
Maddox and colleagues wrote that "heavy rains associated with weather systems of tropical origin were not considered". Tropical systems are a large share of the flood climatology on every US coast and well inland of them, so the patterns are a tool and not the whole picture. Lesson 3 covers what the paper left out.

One ingredient the framework treats as given is the ground. Antecedent soil moisture decides how much of the rain becomes runoff instead of soaking in, and impervious surface decides it permanently: NWS material puts urban runoff at roughly two to six times what the same rain would produce on natural terrain. The same storm over the same town floods it or does not, depending on what fell last week.

Check yourself

Two storms drop rain at the same peak rate over the same town. Storm A is a fast, high-topped supercell with large hail. Storm B is a slow line of cells training along a stalled boundary, no hail, modest lightning. Which is the larger flood threat?

  1. Storm B, because duration over a point dominates the total
  2. Storm A, because a stronger updraft lifts more moisture and therefore rains harder
  3. Neither. Equal peak rates mean equal totals
  4. Storm A, because hail melting on the ground adds to the water total
Hint

The two multiplications from the start of the lesson. One of them has been held constant here by the wording of the question.

Answer

Storm B, because duration over a point dominates the total

The rate is given as equal, so the total is decided entirely by duration, and training along a stalled boundary is the configuration that stretches it furthest. Doswell and colleagues, citing Chappell, found slow system movement dominates most flash flood cases. The quiet storm is very often the one that floods.

Sources

  1. Doswell, Brooks & Maddox 1996, Wea. Forecasting 11(4), 560–581: flash flood forecasting, an ingredients-based methodology
  2. Maddox, Chappell & Hoxit 1979, Bull. Amer. Meteor. Soc. 60(2), 115–123: synoptic and meso-alpha scale aspects of flash flood events
  3. NSSL Severe Weather 101, flood basics
  4. Corfidi 2003, Wea. Forecasting 18(6), 997–1017: cold pools and MCS propagation
  5. MRMS Probability of Warm Rain product description