An early start to the tropical season


In a few months when the Atlantic tropical season gets underway, many folks might wonder why the first named storm at that time begins with “B” instead of “A.” Only those who were paying close attention this April will recall that the first tropical cyclone of 2017 occurred unusually early. Tropical Storm Arlene was a very short-lived tropical cyclone that occurred in the central Atlantic, far from any land. This system did not have pure tropical origins, but instead developed from an upper-level low as a subtropical depression, evolved into a tropical depression then became a tropical storm for a very brief period before losing its tropical characteristics and being absorbed into a larger non-tropical low. It is worth taking a look at the situation that gave rise to this system.

Figure 1: Atlantic 500-millibar analysis valid 1200 UTC 18 April 2017.

Looking first at the 500-millibar chart for 1200 GMT on Tuesday, April 18 (Figure 1), we can see that a cut-off low was present in the central Atlantic with the main 500-millibar flow across the Atlantic well to its north. This feature had been present for a couple of days and was associated with a mature surface low in about the same location (Figure 2). This low had fronts associated with it earlier, but by the time of these charts the fronts had washed out and there was little contrast in the air mass around the low.

Figure 2: Atlantic surface analysis valid 1200 UTC 18 April 2017.

One of the characteristics of tropical cyclones is that they occur within a homogenous air mass and they have no fronts. However, another characteristic is that they are associated with a warm core at upper levels, and a 500-millibar low indicates a cold core — not a warm one. Sometimes in these situations, though, a system looks like a tropical system, has no fronts and has precipitation patterns that are symmetric about the center even though it is a cold-core system. These “hybrid” systems are known as subtropical depressions (or storms), meaning they have some, but not all, characteristics of a tropical cyclone. These systems will often transition to a bona fide tropical cyclone, which is what occurred in this case.

Figure 3: Atlantic 500-millibar analysis valid 1200 UTC 19 April 2017.


Figure 4: Atlantic surface analysis valid 1200 UTC 19 April 2017.

Looking at the charts 24 hours later, the 500-millibar chart for 1200 GMT on Wednesday, April 19 (Figure 3), we see the cut-off low was still present but much weaker, meaning that the cold air at upper levels was not as well defined. The surface chart for this same time (Figure 4) shows that the surface low was still quite well defined and had not weakened significantly from the previous day. This indicates that tropical processes had begun to occur with this system, pumping some warmer air aloft and weakening the upper-level low even as the surface low remained strong and well defined. Three hours after the time of these charts, this system was designated a subtropical depression. Also of note, on the 500-millibar chart at this time was a well-defined trough extending from Newfoundland south to subtropical latitudes to the west of the cut-off low, and on the surface chart a low center was developing along a front to the west of the subtropical low in association with the upper trough.

Figure 5: Atlantic surface analysis valid 1200 UTC 20 April 2017.


Figure 6: Atlantic 500-millibar analysis valid 1200 UTC 20 April 2017.

The tropical processes continued to occur through the rest of Wednesday and Wednesday night, and by 1200 GMT on Thursday, April 20, the surface chart (Figure 5) showed the subtropical depression with its center near 35°N, 40°W in the central Atlantic. But, a stronger frontal system had developed to the west and, in fact, the subtropical depression was contained within the circulation of this newly developing non-tropical low. The corresponding 500-millibar chart (Figure 6) shows that the trough had evolved into a large closed low, which was not cut off from the main flow, and the cut-off low that gave rise to the subtropical depression had entirely dissipated. Three hours later at 1500 GMT on Thursday, the system was re-classified as a tropical depression, and it reached tropical storm strength at 2100 GMT on Thursday. The surface chart for 0000 GMT on Friday, April 21 (Figure 7), shows tropical storm Arlene near 39° N, 43° W in the central Atlantic as a very small system embedded within the circulation of the larger low to its west; within another 18 hours, the small system was absorbed into the larger low and dissipated as all tropical characteristics were lost.

Figure 7: Atlantic surface analysis valid 0000 UTC 21 April 2017.

Another chart worth examining is a combined surface chart and satellite image for 0000 GMT on Friday (Figure 8, valid same time as Figure 7). This image clearly shows the very small but distinct cloud pattern associated with the tropical storm that is clearly separate from the larger low to the west.

Figure 8: East Atlantic surface analysis with satellite imagery valid 0000 UTC 21 April 2017.

So, how unusual is a system like this? The process of subtropical depression/storm development is not at all unusual and often happens a couple times each year. What is unusual is the time of year. Typically in April, upper-level winds are too strong for any tropical development to occur in the area that Arlene developed, but in this case conditions were just right to allow the development. Given the strength of the upper-level pattern, and that the system was moving to higher latitudes where sea surface temperatures were colder, no tropical system could last very long, and this explains the rather brief existence of Arlene. Quoting from a forecast discussion from the National Hurricane Center, “Tropical storms in April are rare and Arlene is only the second one observed in this month during the satellite era. It should be noted, however, that this type of storm was practically impossible to detect prior to the weather satellite era.”

By Ocean Navigator