Re: It's a bug in the software that generates that - O/T
Posted by
Chris in Tampa on 10/31/2018, 11:14 pm
I am in the process of trying to fix the issue now.
I updated the image in your post Jim to reflect what the image did show using a saved image. I'm trying to fix the error now and at times the original image may be correct during my testing.
I found the file that comes from HamWeather comes without a category, so the bug can be fixed somehow on our end. I just have to figure it out. The raw file at HamWeather is here:
http://data.hamweather.net/pub/data/tropical/tracks/track_2018_nt_14.txt
In their file it is 134.6 knots. I think they get the mph, which was 155 mph, and then convert that to knots which is 134.691 knots. (which their site interprets as 134.6 knots) They don't get the knots from the forecast advisory for that file I guess. (In real time, which is how I think they get their data, the forecast advisory has data in knots and the public advisory and discussion has data in knots) Since it seems like the data they have only comes from the public advisory, any changes to best track data at any point after the advisory was first released will not be updated. So if there is any change in the best track data, for any storm in the system (since 1998), that will never get reflected in the HamWeather system.
I think in real time HamWeather gets the data for the public advisory and adds it to that file. That's because public advisories include intermediate advisories. If they only got the forecast advisory or discussion, that would miss all the intermediate advisories. For Michael the 134.6 knots is listed for "16A". That was the intermediate advisory in-between advisory 16 and 17. If best track data is ever updated, I doubt HamWeather would update that file, and that line which is marked as an intermediate advisory. Perhaps, but I doubt it.
I did a test where I changed the 134.6 knots value to a little lower after the file is downloaded to HurricaneCity, but before the image is created, and I was able to make the image not show a category 5. However, that was just a temporary test. I need to update the code so that it correctly assigns the category from now on. Still trying to figure that out. Somewhere in the hundreds of files there is a range I still need to update. Or something that rounds too different and is just enough off to show the incorrect range. I might need to slowly increase the testing value I add to that file to see at what value it thinks it is a category 5. That might help. |
58
In this thread:
Cat 5 status for Michael? -
jimw,
10/31/2018, 6:24 pm Post A Reply
This thread has been archived and can no longer receive replies.