Verification of ghost storm tracks of August 2018

November 16, 2018

I falsely detected (sub)tropical cyclogenesis in Atlantic 61 times in August 2018, and made tracks in all such situations. Tracks were made of ten systems.

At the lead time of 1 days (24 hours), there were 0 false detections.
At the lead time of 2 days (48 hours), there were 0 false detections.
At the lead time of 3 days (72 hours), there were 3 false detections.
At the lead time of 4 days (96 hours), there were 4 false detections.
At the lead time of 5 days (120 hours), there were 4 false detections.
At the lead time of 6 days (144 hours), there were 4 false detections.
At the lead time of 7 days (168 hours), there were 5 false detections.
At the lead time of 8 days (192 hours), there were 5 false detections.
At the lead time of 9 days (216 hours), there were 5 false detections.
At the lead time of 10 days (240 hours), there were 5 false detections.
At the lead time of 11 days (264 hours), there were 12 false detections.
At the lead time of 12 days (288 hours), there were 22 false detections.
At the lead time of 13 days (312 hours), there were 26 false detections.
At the lead time of 14 days (336 hours), there were 29 false detections.
At the lead time of 15 days (360 hours), there were 31 false detections.
At the lead time of 16 days (384 hours), there were 46 false detections.

Since August 23, 12Z run, I stopped putting a legend to all tracks. Since then, tracks had the following circle coloration scheme.

System #38 was not a ghost storm. Rather, it was Ernesto.

Ghost storm #39

This system varied greatly in timing, but in the end, it became identifiable with a wave, marked here in lower right corner. As such, it was related to Gordon. But it couldn't be counted as Gordon, as it was a duplicate, at some runs coexisting with other "Gordon" tracks at same time.

August 4, 12Z run:

August 4, 18Z run:

August 18, 18Z run:

August 22, 12Z run:

Ghost storm #40

This is not necessarily identifiable with anything. Or perhaps with this tropical wave west of Cape Verde Islands. Lead time was too long to be sure.

August 5, 06Z run:

August 5, 18Z run:

August 6, 06Z run:

August 7, 12Z run:

Ghost storm #41

This isolated incident with an extremely large lead time isn't identifiable at all.

August 5, 18Z run:

Ghost storm #42

This can neither be identified.

August 9, 00Z run:

Ghost storm #43

This is identifiable with this cloudiness west of Senegal.

August 11, 18Z run:

August 12, 00Z run:

August 12, 06Z run:

August 12, 12Z run:

August 13, 06Z run:

August 13, 18Z run:

August 14, 00Z run:

August 14, 06Z run:

August 14, 18Z run:

August 15, 00Z run:

August 15, 06Z run:

August 15, 12Z run:

August 15, 18Z run:

August 16, 00Z run:

August 16, 06Z run:

August 16, 12Z run:

August 16, 18Z run:

August 17, 06Z run:

August 18, 06Z run:

August 18, 12Z run:

August 20, 18Z run:

August 21, 18Z run:

August 24, 06Z run:

System #44 was not a ghost storm. Rather, it was Gordon. Namely, the idea, that the wave, which became Gordon, would have developed in eastern tropical Atlantic.