There are two areas the US National Hurricane Center are watching, one has tropical storm force winds but isn’t really very tropical, the other is tropical but doesn’t have tropical storm force winds …
AL90 is already above tropical storm winds, but isn’t tropical in structure. It will be moving over warmer water as it brushes Bermuda, and may acquire some tropical characteristics and get a name. More on how that works below. AL91, in the Gulf, may barely reach the threshold for being a storm before landfall, but this is only a gusty wind and rain event for Texas.
So how do all these names and codes work? Recall that in the US meteorological community tropical cyclones are typically tracked using an identifier known as the Automated Tropical Cyclone Forecast Identifier (ATCF ID). These identifiers identify unique storms with a code for the format XXNNYYYY, where XX is the basin (ocean) where the storm forms, NN is the storm number for a given year, and YYYY is the year. So the first storm to form in the Atlantic will get the ATCFID AL012021.
What is called a tropical cyclone (TC) has a very specific scientific definition (link goes to Wikipedia definition). The problem is how do you balance the science, the evolving nature of these storms (that start out not being tropical cyclones, become a TC, then become something else) with the need to communicate with the public and issue warnings. That is a complex process. A storm system that does not meet the definition yet (but might or might not) needs to be tracked. These systems are given temporary ID’s. Since no areas of the world get more than 50 storms in a year, storm numbers 90 and above are used as temporary ID’s. So the first “maybe will be something” area of interest, or “INVEST” (short for “investigation” area), gets the temporary ID AL902021. The second, AL91, and so forth. These are recycled, so there will be multiple AL90’s in a year.
What about names? Names are assigned for public use to make it easier (and supposedly names get people’s attention better than a number), but names are only assigned when the storm reaches tropical storm strength (34 knots, or 39mph). Storms weaker than that are called depressions, and are just called by their number. While in the Atlantic the names are consistent, in the Pacific different weather services (especially the Philippines) give storms non-standard names, so using the ATCFID is essential in those parts of the world to keep things straight. But in the Atlantic everybody sticks to the WMO name list.
In theory the progression of the first storm would be from an invest area (AL902021) to a tropical depression (AL012021, called “Tropical Depression One”) to a tropical storm (still AL012021, but with a name; this year the first storm will be called “Ana”). But nature doesn’t always like to follow our rules. The thing near Bermuda already has tropical storm force winds, but does not meet the definition for a tropical cyclone. It’s more like a Nor’easter, a winter storm, because it does not have a warm core and other structures that a tropical system has. But of course your roof doesn’t really care about that – it only sees the wind – and forecasters want to draw your attention to the storm. So the weather service has created a new category they are calling a “subtropical cyclone” – a system that has subtropical (not quite tropical) characteristics but still presents a threat. Expect AL90 to become “subtropical storm Ana” and get the ID AL012021 later today or tomorrow (80% chance according to NHC).
Of course this presents lots of complications. For one thing, it messes up the storm counts and makes people think storms are getting more numerous (and before somebody cries conspiracy, no, it’s not that). In the past nobody bothered with these non-tropical events. At the end of the year Ana will count towards the total even though this same storm in 1992, much less 1982, probably would not have counted. Also, this year NOAA changed the definition of an “average” year from 12 to 14 storms, and their forecast is for “13 to 20” storms. Last year that would be average to above average; this year that is “slightly below to above average.” So whenever you read an article talking about average or above average seasons this year that doesn’t put it in proper context that the definitions changed, be wary …