Shifting computer models on Irma 

Posted: 11:24 am Wednesday, September 6th, 2017

By Kirk Mellish

This does not by any means any kind of stand down with regard to the threats posed by Irma.

We start with a look at the latest models which continue to vacillate, especially in the longer range beyond 3 days, which is to be expected. Remember the error at 5 days and beyond is measured in hundreds of miles! This is the normal and expected variance even when models are performing better than they have been with this system.

Screen Shot 2017-09-06 at 10.57.35 AMScreen Shot 2017-09-06 at 10.54.06 AMScreen Shot 2017-09-06 at 10.56.23 AMScreen Shot 2017-09-06 at 10.55.16 AMScreen Shot 2017-09-06 at 10.58.54 AM

Screen Shot 2017-09-06 at 10.56.59 AM

There is now basic agreement between the NHC and the numerical guidance:

Screen Shot 2017-09-06 at 11.13.38 AM

small

OFFICIAL NATIONAL HURRICANE CENTER FORECAST PRODUCTS:

145453_5day_cone_no_line_and_wind

145453_earliest_reasonable_toa_no_wsp_34

Screen Shot 2017-09-06 at 11.53.48 AM

large

Screen Shot 2017-09-06 at 11.17.52 AM

5-DAY RAINFALL:

p120i

Screen Shot 2017-09-06 at 5.36.55 AM

My previous Irma posts here.

FOLLOW ME on TWITTER @MellishMeterWSB

 

0 comments