How Hurricane Names and ID numbers (sort of) Work

Hurricane naming and numbering is simple yet remarkably confusing, and the rules change through the life of a storm. The basic simplified full life cycle of a storm is a disturbance becomes an Investigation Area or invest, then becomes a tropical depression, tropical storm, hurricane (or typhoon in the West Pacific), then as it degrades back through storm and depression phases, or becomes post-tropical in some form. Here is a try at explaining it, using the Atlantic but noting differences in other basins as appropriate.

Let’s start when the system first shows up on the National Hurricane Center’s Tropical Weather Outlook. At that point, it’s often just some clouds with no organization at all (and, if it’s based on model runs, maybe not even clouds!). This is the “blob of doom” phase, and it is just a number from One to ? depending on how many blobs there are … here is an outlook from mid September 2025 …

You will see here this system is simply “Atlantic Disturbance #1.” Notice on the left edge is “East Pacific Disturbance #1” … so that’s the first potentially confusing thing.

At this point there wasn’t anything for the models to “lock on” to, so as of yet the blob hasn’t been assigned a temporary ID in the US hurricane tracking system (known as the Automated Tropical Cyclone Forecasting system or ATCF). But soon our system got strong enough for dedicated model tracking. Now it becomes an Investigation area or INVEST, and it was assigned a temporary ATCFID, AL922025.

Full ATCFID’s consist of the region, a two digit ID number, followed by the four digit year. Numbers between 90 and 99 are “temporary” ID’s that are recycled through the year, since not every INVEST spins up to become a real tropical cyclone. (Numbers in the 80’s are reserved for tests and exercises, so you guys should never see them.) Here are the region codes:

AL - North Atlantic
WP - Western North Pacific
CP - Central North Pacific
EP - Eastern North Pacific
IO - North Indian Ocean
SH - Southern Hemisphere

According to the official US Navy documentation, “The only tricky part is that the southern hemisphere storm years start on July 1 and end on June 30.” So a southern hemisphere storm forming on July 1st 2025 would get the ID SH012026. But you might find the Navy has a lax definition of “tricky” … there are more sources of potential confusion. Sometimes the year is left off – that’s not so bad. But there is a “shorthand” version of the ATCFID that consists of the ID number and a single letter for the subbasin. So you will see AL922025 as 92L in some reports and maps. Why not 92A? You would ask that. Because “A” is used for the Arabian Sea. Here are the single letter codes:

L = North Atlantic
E = East Pacific
C = Central Pacific
W = West Pacific
U = Australian
S = South Indian
P = South Pacific
A = North Arabian Sea
B = Bay Of Bengal
O = South China Sea
T = East China Sea

As noted above for our new system, the next available temporary ID in the Atlantic was 92, so the “name” at that point was AL922025 or AL92. Or 92L. But on the NHC Tropical Weather Outlook it’s still called “Disturbance #1.” On my maps it shows as “INVEST” but I’ll probably add the short ATCFID at some point.

Eventually our storm (might) become a real tropical cyclone. At that point it get’s a permanent ATCFID. The next permanent ID in the Atlantic was 7, so our storm now gets the permanent ID of AL072025 (or in shorthand either AL07 or 07L). The good news is that ID will stick with it for it’s entire lifetime, and all the data that was associated with it before it became a storm is transferred in to that ID. The bad news is the “common name” may or may not be assigned yet. Only tropical storms (winds 34 knots or higher) get common names, so if it is structurally a tropical cyclone but the winds are not high enough, it is called a tropical depression, and the given name is “Seven” …

If our storm continues to intensify and becomes a tropical storm, it gets a permanent name. In the Atlantic, there are six “lists” of names that are recycled every six years. So for this year, the seventh named storm is given the name “Gabrielle.” Be careful, there is a gotcha – a storm can get a number and never get strong enough to get a name. So the seventh numbered storm of the year doesn’t always start with the letter “G” if there were weaker storms earlier in the year that got permanent ID’s, but never got over 34 knots. In 1992, for example, there were three weaker storms before the first named storm, Andrew, so Andrew’s ATCFID is AL041992! If a storm is bad enough (like Andrew was), the name is retired and a new “A” name is put on that list for next time. So while there was a tropical storm Andrew in 1986, there will never again be an Andrew after 1992. The first name name on List Two, “Andrew,” was replaced in 1998 with “Alex” (and will come up again in 2028). So it’s good practice to always add the year in addition to the name to be sure folks know which year and storm you mean (eg Andrew 1992).

This year all the previous storms were named. For our example storm (live as of September 17th, 2025), we’re set: it is now AL072025, or Gabrielle (2025). Or AL07. Or 07L.

While the Atlantic, East, and Central Pacific regions are consistent, others have a bit of peculiarity between agencies. For example, The Philippines has their own name and number convention, so the name and number used by PAGASA (their weather agency) doesn’t match the Joint Typhoon Warning Center, or Japan Meteorological Agency, or China’s weather agency. The Aussies use a custom number scheme as well. But otherwise the ATCFID system is globally consistent for US sources and used by pretty much everybody else except those who don’t 😛 …

As a storm weakens it keeps the name and ID, with a couple of exceptions. Storms moving from the Atlantic to the Pacific across Central America get new names and ID’s. Likewise, crossing from the Central to West Pacific gets a new ID. But moving from East Pacific to Central Pacific does not cause a name change. If a storm degrades to the point NHC/JTWC stops tracking, but later regenerates, as long as it kept a more or less a coherent structure it returns with the same name and ID. Mario (EP132025) did this in September 2025, degenerating then coming back.

Fortunately storms can’t cross the equator.

Hopefully that helps clear up some confusion without adding more to this business of names and numbers – because it’s just a job to do.

---

If you like this commentary you can subscribe to the emails, which is the best and only really reliable way to get them, and/or donate to keep it going, at this this link. You can also follow us on X/Twitter (@EnkiResearch), Telegram, BlueSky, or even Facebook if that’s your preferred social media mayhem.