Moisture, instability, lift, and the numbers behind them

Three ingredients make a thunderstorm and a fourth decides what kind you get. Each of them has a number attached, and the numbers are on the same forecast pages you already look at.

Three ingredients, and the one that sorts them

Every thunderstorm on earth needs the same three things, and in a warm humid summer none of them is scarce, which is why such places get many ordinary storms and comparatively few organized ones. The fourth item below is not needed to make a thunderstorm at all. It is what decides whether the storm that forms lasts twenty minutes or six hours.

Moisture

Water vapour in the low levels, measured as dewpoint. It is the fuel: condensation releases the latent heat that keeps a rising parcel warmer than its surroundings.

Instability

An atmosphere in which a lifted parcel keeps rising on its own. Measured as CAPE, in joules per kilogram.

Lift

Something to give the parcel its first shove upward, far enough that buoyancy can take over. Fronts, boundaries, terrain, the sea breeze.

Shear

Change of wind with height. Not required for a thunderstorm, and required for almost every organized severe one. Lesson 3 is entirely about this.

Parcel theory in one page

Take an imaginary bubble of air at the surface and lift it. Dry air cools as it rises at the dry adiabatic lapse rate, 9.8 °C per kilometer, which is fixed by physics and does not vary. Once the parcel saturates, condensation starts releasing latent heat into it, so it cools more slowly: the moist adiabatic rate is roughly 4 to 7 °C per kilometer, and it depends on temperature. Warm, moist parcels cool slowest.

The whole of forecasting convection is a comparison between that parcel and the air it is moving through. Warmer than its surroundings means buoyant, which means it keeps going. Colder means it sinks back. Three levels mark where that comparison changes sign.

LCL

Lifted condensation level. Where the rising parcel saturates and cloud forms. This is the height of the cloud base you are looking at.

LFC

Level of free convection. Above this the parcel is warmer than its environment and rises without any further help. Everything below it has to be paid for by lift.

EL

Equilibrium level. Where the parcel finally cools back to the temperature of its surroundings and stops accelerating. The anvil spreads out near here, and momentum carries the strongest updrafts above it as an overshooting top.

CAPE

Convective available potential energy: the area on a sounding between the LFC and the EL where the parcel is warmer than its environment. Energy available to the updraft, in joules per kilogram.

CIN

Convective inhibition: the area below the LFC where the parcel is colder than its environment. Energy the lift has to supply before anything happens at all.

The cap, and why capped days are dangerous

CIN in the field is usually called the cap: a warm layer aloft that a parcel cannot get through without help. A capped day looks like nothing. Blue sky, a few flat cumulus that go nowhere, and huge CAPE underneath doing absolutely nothing because no parcel can reach its LFC.

Cloud base, from a thermometer

You can estimate the height of the cloud base from the surface temperature and dewpoint alone. Espy's approximation puts the LCL about 125 meters above the ground for every degree Celsius of dewpoint depression, which is the gap between temperature and dewpoint.

Rasmussen & Blanchard 1998 and Thompson et al. 2003 both found low LCL heights separating significantly tornadic supercells from the rest. The physical argument is about the near-storm air: a moist boundary layer with a low cloud base gives a downdraft less dry air to evaporate into, so the outflow is less cold and less likely to undercut the updraft.

Translated for somebody standing outside: low, ragged bases on a muggy day are a signal, not noise. You can read the first half of that sentence off a thermometer before you leave the house.

How fast can an updraft possibly go

Parcel theory gives a ceiling. If every joule of CAPE went into vertical motion and nothing was lost, the fastest the updraft could be moving is the square root of twice the CAPE.

Real updrafts never reach those numbers. Entrainment mixes drier environmental air in around the edges and dilutes the buoyancy, and precipitation loading means the updraft is carrying the weight of the water and hail it has made. What the table is good for is sanity: it shows that the Field Guide's figure of a supercell updraft exceeding 100 mph is not rhetoric, it is the low end of what the thermodynamics permits on a big day.

Which CAPE is the CAPE

A number labeled "CAPE" on a model page is meaningless until you know which parcel it was computed for. The same sounding can give values a factor of two apart.

• Surface-based (SBCAPE). Lifted from the surface. The right one when storms are rooted in the boundary layer, and misleading at night when the surface has cooled off.
• Mixed-layer (MLCAPE). Lifted from an average over the lowest 100 hPa or so. Usually the most representative of a well-mixed afternoon, and lower than SBCAPE.
• Most-unstable (MUCAPE). Lifted from whichever level gives the biggest number. This is the one that finds elevated instability above a stable surface layer, which is how nocturnal storms happen at all.

This matters again in lesson 7, where the definition of a high-shear, low-CAPE environment sets separate thresholds for surface-based and most-unstable CAPE and means both of them.

Reading a skew-T, at recognition level only

The intent here is not to make forecasters. It is to make the sounding diagrams on the SPC mesoanalysis page readable, so that the numbers above stop being abstractions. Four things to find:

• The two traces: temperature on the right, dewpoint on the left. The gap between them is the dewpoint depression at every height, so a narrow gap is a moist layer.
• The parcel path, and where it crosses the temperature trace going up (that is the LFC) and coming back down again (the EL).
• The positive area between those two crossings. That is CAPE, drawn.
• The negative area below the LFC. That is CIN, and the fatter it is, the harder the day has to work.

Where the lift comes from here

• Fronts. Cold fronts are the main organized lift almost everywhere outside high summer, and lesson 8 is largely about what forms along them.
• Drylines. A southern Plains feature, and the classic supercell trigger. Half the severe weather literature is written about them, so know what one is even if you will never stand next to one.
• Outflow boundaries. The cold pool left behind by a dead storm, still drifting, still able to lift. Lesson 2 explains where they come from and lesson 5 explains how they build the next storm.
• Terrain. Dominant in the West and the Appalachians, where upslope flow and mountain circulations fire storms on a daily cycle. Negligible on a coastal plain. One of the clearest reasons spotter practice differs by region.
• Sea, lake and bay breezes. The dominant summertime lift on any coast, and around the Great Lakes. Each body of water pushes its own breeze front inland, and where two of them collide is where the afternoon storms go up. Florida runs this pattern from two coasts at once.

Where the moisture comes from is the other half of the story, and it is the half that varies most. Air off warm water carries dewpoints in the mid 70s Fahrenheit with no help from anything, so in the Southeast instability is rarely the missing ingredient in summer and the question is what will lift it. Across the Plains that same moisture has to be transported north, and its arrival or failure is the forecast. In the interior West it is often simply absent, which is why storms there are high-based and why lesson 6 talks about dry microbursts.

Check yourself

It is 31 °C with a dewpoint of 19 °C. Roughly how high is the cloud base, and what does that suggest?

1. About 1,500 m (4,900 ft), which is a relatively high base and not a low-LCL tornado signal
2. About 1,500 m (4,900 ft), which is low enough to be a significant tornado signal
3. About 600 m (2,000 ft), because the air is very humid
4. It cannot be estimated without a sounding

Sources

1. NWS Weather Spotter's Field Guide (YPA-201154, June 2011), section 2
2. NWS JetStream, the thunderstorm and stability sections
3. Rasmussen & Blanchard 1998, Wea. Forecasting 13(4), 1148–1164: a baseline climatology of sounding-derived supercell and tornado forecast parameters
4. Thompson et al. 2003, Wea. Forecasting 18(6), 1243–1261: close proximity soundings within supercell environments from the Rapid Update Cycle