Hurricane Andrew Formed on the Day Forecasters Said There Was No Season

The facts, up front

• Storms in the 1992 Atlantic season: seven — six named, plus an unnamed subtropical storm in April

• Hurricanes: four. Major hurricanes: one

• Season energy (ACE): 76, against an average near 92

• Forecast, issued December 1991: eight named storms, four hurricanes, one major — never revised

• Andrew's depression formed: 16 August

• Andrew's landfall: 24 August, near Homestead, 922 millibars

• Damage: $27.3 billion in 1992 dollars, close to $60 billion now

• Homes damaged or destroyed: more than 125,000

Was 1992 a quiet hurricane season?

Yes. One of the quietest on record at the time.

Seven storms formed all year. Six got names. The seventh was a subtropical storm back in April, and subtropical storms didn't get names until 2002.

Four of those seven became hurricanes. Exactly one reached major hurricane strength.

Season energy came in at 76 against an average near 92. Only one year since has produced as few storms.

What did forecasters predict for 1992?

A quiet season. And they were right.

Colorado State put out its first forecast in December 1991: eight named storms, four hurricanes, one major hurricane. They updated in April, June and August and left those three numbers exactly where they were each time.

Here's how it finished.

Colorado State's 1992 forecast against the outcome

Source: Colorado State University seasonal forecast records; NOAA HURDAT2.

On hurricanes and majors, dead on. Colorado State's own write-up of that year puts it plainly: they called one major hurricane, and the one major hurricane was Andrew.

This is the part people get backwards when they tell this story. Nobody blew the forecast. The forecast worked.

When did Hurricane Andrew form?

On 16 August 1992 — the same day the National Hurricane Center was fielding questions about the hurricane season that hadn't arrived.

Max Mayfield, who later ran the Hurricane Center, recalled it years afterward at a hurricane conference. On 16 August, he said, they talked with the media about the lack of a hurricane season.

A tropical wave had come off the west coast of Africa two days before. On the 16th it pulled itself together into a tropical depression.

Same date. Both things.

How fast did Andrew go from nothing to catastrophe?

Eight days. And for the first half of that stretch it looked like it was going to fail.

Andrew got its name on 17 August. Then it struggled. An upper-level low near Bermuda pushed shear across the top of it and stripped the thunderstorms away from the center. On 20 August the storm came close to falling apart entirely.

The same force holding down the whole 1992 season was working on Andrew too. For four days it won.

Then the upper low weakened and slid north. A ridge built in over the southeastern United States. Andrew turned west into clean air, and on 22 August it became a hurricane.

Two days after that it was over Homestead.

What happened at landfall?

It came ashore in the early hours of 24 August. The last radar image out of the Miami office was taken at 4:34 in the morning.

Central pressure was 922 millibars — among the lowest ever recorded for a hurricane hitting the United States. Storm surge near the landfall point ran about 17 feet.

Here's a detail that tells you more than any wind number could. Andrew's peak winds in south Florida were never directly measured, because the instruments that would have measured them were destroyed. The pressure reading came from private barometers that people had in their homes.

The storm crossed the peninsula in about four hours, went into the Gulf, and hit the Louisiana coast again on the 26th.

How much damage did Andrew cause?

$27.3 billion in 1992 dollars. Nearer $60 billion in today's money.

More than 125,000 homes were damaged or destroyed. Roughly 250,000 people in south Miami-Dade had nowhere to live.

Now put that against the rest of the year.

Seven storms formed in 1992. One of them accounted for essentially all of the season's damage and 65 of its 73 deaths.

That is the whole argument in one table. The count told you nothing.

Was Andrew a Category 5?

It is rated Category 5 today. It wasn't in 1992.

At the time, forecasters assessed Andrew as a Category 4 at Florida landfall. Ten years later, researchers went back through the evidence as part of a reanalysis of the hurricane database and upgraded it.

Worth knowing for two reasons. First, if you read a 1992 newspaper it will say Category 4, and that isn't an error. Second, it shows how much of what we know about a storm gets worked out after the fact — which is exactly the problem with using a live forecast to make a decision about your house.

Was 1992 an El Niño year?

Depends who you ask, and that's a real answer rather than a dodge.

A moderate El Niño ran through 1991 and into 1992, then faded. By the heart of hurricane season it had weakened toward neutral. Colorado State's reasoning going into the year cited neutral to weak El Niño conditions and the high wind shear that came with them.

Some classifications count 1992 as an El Niño season. Others call it near-neutral. It comes down to which index you use and which months you average.

For our purposes the label doesn't matter much. What matters is the shear was there, it held the season down to seven storms, and it did not stop Andrew.

What does 1992 tell you about 2026?

That a quiet season and a safe year are not the same thing, and that knowing the first tells you nothing about the second.

2026 has been quieter than 1992 by a wide margin. Six named storms and no hurricanes at all, against 1992's four hurricanes. The suppression this year is stronger and the El Niño driving it is stronger.

But the mechanism is identical. Shear stops storms from getting organized. It does not weaken the ones that get organized anyway.

In 1992 the shear won four times out of five. The fifth time it lost, and the fifth time cost $27.3 billion.

The useful lesson isn't that a monster is coming. Nobody knows that. It's that the season total was never the number worth watching — not in 1992, and not now.

Common questions

How many storms were there in the 1992 hurricane season?

Seven storms formed. Six were named; the seventh was an unnamed subtropical storm in April. Four became hurricanes and one, Andrew, became a major hurricane.

When did Hurricane Andrew form?

The tropical depression that became Andrew formed on 16 August 1992, from a wave that left Africa on 14 August. It was named Andrew on 17 August and made landfall in Florida on 24 August.

Did forecasters predict the 1992 hurricane season correctly?

Yes. Colorado State forecast eight named storms, four hurricanes and one major hurricane in December 1991 and never revised it. The season produced four hurricanes and one major. The forecast was accurate, and the one major hurricane was Andrew.

How strong was Hurricane Andrew at landfall?

Its central pressure at Homestead was 922 millibars, among the lowest ever recorded for a US landfall. Peak winds in south Florida were never directly measured because the instruments were destroyed. The pressure came from private home barometers.

Why was Andrew reclassified as a Category 5?

It was assessed as Category 4 in 1992. A reanalysis of the Atlantic hurricane database in 2002 reviewed the evidence and upgraded it to Category 5.

How much of the 1992 season's damage did Andrew cause?

Almost all of it. The season caused $27.302 billion in damage and Andrew accounted for $27.3 billion. Andrew also caused 65 of the season's 73 deaths.

Where MaxForce™ fits

MaxForce™ Hurricane Track by Fenetex® is tested and certified for the 185mph wind zone, meeting or exceeding all local and International Building Codes, including the High Velocity Hurricane Zone (HVHZ) of Miami-Dade and Broward Counties. It is approved for use in Miami-Dade County and the HVHZ under Florida Product Approval FL8637-R15.

That HVHZ designation didn't exist in August 1992. It exists because of what happened that morning in Homestead.

Next in this series: the building code Andrew wrote, and why South Florida is the only place in the country that regulates the whole outside of your house instead of just the glass.

Storm data from NOAA's National Hurricane Center, National Weather Service and HURDAT2 database, and from Colorado State University's seasonal forecast records. Imagery courtesy of NOAA. MaxForce™ is a product of Fenetex® Corporation.