txngeo's Blog

Posted by: txngeo, 3:50 PM GMT op 01 oktober 2010 +0
Here are my updated forecast validation numbers through September. Overall, the NHC is hitting the 2/3 forecast radius right at 2/3 of the time. The last three storm have really lowered the overall numbers.  For my methodology, please see my first blog entry.


............. Points... In.... % in
Alex           114      86.... 75.4
TD 2               4        1.... 25.0
Bonnie        20      12.... 60.0
Colin           64      24.... 37.5
TD 5              4        2..... 50.0
Danielle.. 198   120..... 60.6
Earl           233   185..... 79.4
Fiona          60      42..... 70.0
Gaston         6        5...... 83.3
Hermine    16        3...... 18.8
Igor           310   246...... 79.4
Julia         163   121...... 74.2
Karl             43     28...... 65.1
Lisa          100     30...... 30.0
Matthew     33       7...... 21.2
Nichole        6       1...... 16.7
...............1374   913..... 66.4

Here's a couple of images of the last three storms... Lisa which missed a lot of early forecast points as it went unexpectedly east. The pink disks show forecast circles that missed the actual path, blue disks are valid forecasts.


And here's Matthew which was forecast pretty well at 12 hours (disks shown). Yet the longer range forecasts didn't pan out as well.


And the short lived Nichole hit on its first 12 hour forecast but missed the rest.


I am very open on how to improve this or if anyone else is already providing these numbers!  I'm a newbie at this!
Regards,
txngeo
  Permalink | A A A
Posted by: txngeo, 10:41 PM GMT op 25 september 2010 +0
Hi All, my first foray into blogging.... so why not choose something uncontroversial like NHC forecasts! My intent is to provide data and not opinion since I am not a met. However, living in a hurricane area makes me curious as to how good the NHC is doing.There are two ways to look at the NHC forecast cones as you can read on their website. The general way to to see the actual track plotted through the cone to see if the points fall in. This method misses any t...
  Permalink | A A A

« View Older Entries