Which model(s) performed the best in the 2005 season?

This is the general tropical discussion area. Anyone can take their shot at predicting a storms path.

Moderator: S2k Moderators

Forum rules

The posts in this forum are NOT official forecasts and should not be used as such. They are just the opinion of the poster and may or may not be backed by sound meteorological data. They are NOT endorsed by any professional institution or STORM2K. For official information, please refer to products from the National Hurricane Center and National Weather Service.

Help Support Storm2K
Post Reply

Which models(s) performed the best in the 2005 season?

ECMWF
0
No votes
RUC
0
No votes
GFDL
28
88%
GFS
2
6%
UKMET
1
3%
NOGAPS
0
No votes
CMC
1
3%
MM5
0
No votes
The Tropical Models (LBAR,BAMM,BAMD)
0
No votes
 
Total votes: 32
Message
Author
User avatar
cycloneye
Admin
Admin
Posts: 146244
Age: 69
Joined: Thu Oct 10, 2002 10:54 am
Location: San Juan, Puerto Rico

Which model(s) performed the best in the 2005 season?

#1 Postby cycloneye » Mon Oct 31, 2005 8:10 pm

The 2005 season still is not over but as the end nears we can discuss about which model or models performed the best during this season.

LBAR was the best (No JK) :)

:darrow: :darrow: :darrow: :darrow: :darrow: :darrow:

I say GFDL did very well this year in the tracks (The only model that had a SW movement of Katrina thru South Florida) and on intensity although in intensity sometimes the model was extremely bullish but overall it did well.
0 likes   
Top
User avatar
EDR1222
S2K Supporter
S2K Supporter
Posts: 1253
Joined: Mon Nov 10, 2003 12:58 pm
Location: Melbourne, FL

#2 Postby EDR1222 » Mon Oct 31, 2005 8:25 pm

I believe the GFDL did very well this year. A few of the runs during Wilma were kind of strange, but for the most part it performed better than the others IMO.
0 likes   
Top
User avatar
wxmann_91
Category 5
Category 5
Posts: 8013
Age: 34
Joined: Fri Jul 15, 2005 2:49 pm
Location: Southern California
Contact:
Contact wxmann_91

#3 Postby wxmann_91 » Mon Oct 31, 2005 8:27 pm

GFDL and UKMET were the best, but I voted GFDL.
0 likes   
Top
Matt-hurricanewatcher

#4 Postby Matt-hurricanewatcher » Mon Oct 31, 2005 8:30 pm

How doe's the Gfdl beat all the other models? :eek:
0 likes   
Top
User avatar
WindRunner
Category 5
Category 5
Posts: 5806
Age: 34
Joined: Fri Jul 29, 2005 8:07 pm
Location: Warrenton, VA, but Albany, NY for school
Contact:
Contact WindRunner

#5 Postby WindRunner » Mon Oct 31, 2005 8:36 pm

wxmann_91 wrote:GFDL and UKMET were the best, but I voted GFDL.


:uarrow: It was the best more of the time.
0 likes   
Top
f5
Category 5
Category 5
Posts: 1550
Joined: Thu Aug 25, 2005 3:29 pm
Location: Waco,tx

#6 Postby f5 » Mon Oct 31, 2005 9:08 pm

GFDL nailed Katrina's sw movement while the others were yelling FLORDIA PANHANDLE!!!!!!!!!!!!
0 likes   
Top
User avatar
WxGuy1
Professional-Met
Professional-Met
Posts: 538
Joined: Sat Oct 22, 2005 12:55 pm
Location: Oklahoma

#7 Postby WxGuy1 » Mon Oct 31, 2005 9:40 pm

The NHC has some stats up from the 1996-1997 seasons at http://www.nhc.noaa.gov/aboutmodels.shtml . In addition, the NHC has statistics about the for just about every tropical cyclone in the Atlantic in their "after-action" reports. For example, here are some stats for Jeanne: http://www.nhc.noaa.gov/2004jeanne.shtml?#TABLE4 ...

EDIT: Final summaries can be read at http://www.nhc.noaa.gov/pastall.shtml ... Click on the year, then the storm name/number.
Last edited by WxGuy1 on Mon Oct 31, 2005 10:02 pm, edited 2 times in total.
0 likes   
Top
User avatar
wxmann_91
Category 5
Category 5
Posts: 8013
Age: 34
Joined: Fri Jul 15, 2005 2:49 pm
Location: Southern California
Contact:
Contact wxmann_91

#8 Postby wxmann_91 » Mon Oct 31, 2005 9:48 pm

WxGuy1 wrote:The NHC has some stats up from the 1996-1997 seasons at http://www.nhc.noaa.gov/aboutmodels.shtml . In addition, the NHC has statistics about the for just about every tropical cyclone in the Atlantic in their "after-action" reports. For example, here are some stats for Jeanne: http://www.nhc.noaa.gov/2004jeanne.shtml?#TABLE4 ...


Hard to believe it, but during Ivan and Frances, the LBAR, BAM models, and A98E outperformed the official forecast during many instances.
0 likes   
Top
User avatar
WxGuy1
Professional-Met
Professional-Met
Posts: 538
Joined: Sat Oct 22, 2005 12:55 pm
Location: Oklahoma

#9 Postby WxGuy1 » Mon Oct 31, 2005 10:03 pm

wxmann_91 wrote:
WxGuy1 wrote:The NHC has some stats up from the 1996-1997 seasons at http://www.nhc.noaa.gov/aboutmodels.shtml . In addition, the NHC has statistics about the for just about every tropical cyclone in the Atlantic in their "after-action" reports. For example, here are some stats for Jeanne: http://www.nhc.noaa.gov/2004jeanne.shtml?#TABLE4 ...


Hard to believe it, but during Ivan and Frances, the LBAR, BAM models, and A98E outperformed the official forecast during many instances.


In the case of Ivan, the official forecast was particularly bad. The FSSE (Florida State Super Ensemble) did an absolutely remarkable job forecasting Ivan! http://www.nhc.noaa.gov/2004ivan.shtml?#TABLE6

Code: Select all

Forecast Technique   Period (hours)
           12         24      36          48          72           96         120
CLP5   25 (62)    51 (60)    86 (58)    121 (55)    175 (51)    229 (47)    296 (44)
GFNI   27 (57)    57 (55)    95 (55)    134 (53)    219 (49)    325 (45)    443 (41)
GFDI   28 (60)    52 (58)    77 (57)    94 (55)     147 (51)   204 (47)   261 (43)
GFDL   29 (61)    52 (59)    76 (57)    92 (56)     136 (52)   195 (48)   252 (44)
GFDN   29 (56)    51 (54)    79 (53)    118 (53)    198 (49)    298 (45)    412 (41)
GFSI   29 (60)    55 (58)    79 (57)    100 (55)    154 (51)   212 (47)   285 (43)
GFSO   32 (63)    56 (61)    80 (58)    99 (56)     147 (52)   198 (48)   268 (43)
AEMI   27 (60)    49 (58)    73 (57)    95 (55)     151 (51)   203 (47)   252 (43)
NGPI   26 (59)    54 (57)    86 (56)    119 (54)    200 (50)    304 (46)    449 (42)
NGPS   29 (61)    52 (59)    82 (56)    109 (54)    186 (50)    278 (46)    413 (42)
UKMI   24 (58)    47 (56)    72 (55)    92 (53)     147 (49)   214 (45)   286 (41)
UKM    26 (30)    44 (29)    66 (28)    87 (28)      133 (26)   189 (24)   279 (22)
A98E   27 (61)    43 (59)    72 (57)    103 (55)     179 (51)   248 (47)    349 (44)
A9UK   27 (28)    43 (27)    71 (26)    95 (25)      149 (23)        
BAMD   31 (61)    54 (59)    78 (57)    96 (55)      147 (51)   219 (47)   300 (44)
BAMM   32 (61)    53 (59)    86 (57)    116 (55)     172 (51)    214 (47)   283 (44)
BAMS   48 (61)    93 (59)    142 (57)    179 (55)    236 (51)    251 (47)    286 (44)
CONU   22 (60)    47 (58)    74 (57)    97 (55)     154 (51)   224 (47)    307 (43)
GUNA   22 (58)    45 (56)    71 (55)    92 (53)     147 (49)   213 (45)   289 (41)
FSSE   21 (53)    38 (51)    58 (51)    81 (51)     126 (47)   171 (43)   199 (38)
OFCL   24 (63)    47 (61)    79 (59)    108 (56)    161 (52)    222 (48)    289 (44)
NHC Official, 1994 - 2003 mean (number of cases)   44 (3172)    78 (2894)    112 (2636)    146 (2368)    217 (1929)    248 (421)    319 (341)
0 likes   
Top
User avatar
canetracker
S2K Supporter
S2K Supporter
Posts: 751
Age: 62
Joined: Wed Jul 27, 2005 8:49 pm
Location: Suburbia New Orleans...Harahan, LA

#10 Postby canetracker » Mon Oct 31, 2005 10:15 pm

GFDL has been right on. Just to mention two storms, in Katrina it predicted the SW movement and in Beta, when we all thought the GFDL was on drugs for being the southern-most model, it was correct again.
0 likes   
Top
WeatherEmperor
S2K Supporter
S2K Supporter
Posts: 4806
Age: 41
Joined: Thu Sep 04, 2003 2:54 pm
Location: South Florida

#11 Postby WeatherEmperor » Mon Oct 31, 2005 10:19 pm

GFDL hands down.

<RICKY>
0 likes   
Top
Derecho
Category 5
Category 5
Posts: 1011
Joined: Sat Mar 15, 2003 3:15 pm

#12 Postby Derecho » Mon Oct 31, 2005 10:34 pm

This isn't a subjective issue as model error can be quite precisely calculated, so I don't see why it's appropriate for a poll; what is interesting is people's vague impressions of model accuracy often turn out to be completely wrong.

I can calculate it for each storm with Stormtrakker but there are odd runs missing here and there for various reasons, and putting it all together for the season so far would be a massive pain; I'd just as soon wait for the report for this season on model accuracy that will be out next year.

From looking at the reports of individual storms the interesting thing seems to be a comeback in accuracy by the UKMET.

Full reports on the accuracy of all the models for each basin are avaliable through 2004 in PDF format here:

http://www.nhc.noaa.gov/verification/verify3.shtml?
0 likes   
Top
User avatar
deltadog03
Professional-Met
Professional-Met
Posts: 3580
Joined: Tue Jul 05, 2005 6:16 pm
Location: Macon, GA

#13 Postby deltadog03 » Mon Oct 31, 2005 11:49 pm

none of them....the closest one would have to be the euro.....
0 likes   
Top
User avatar
wxmann_91
Category 5
Category 5
Posts: 8013
Age: 34
Joined: Fri Jul 15, 2005 2:49 pm
Location: Southern California
Contact:
Contact wxmann_91

#14 Postby wxmann_91 » Tue Nov 01, 2005 12:05 am

deltadog03 wrote:none of them....the closest one would have to be the euro.....


Aren't the Euro and the ECMWF the same model?
0 likes   
Top
Post Reply

Return to “Talkin' Tropics”

Who is online

Users browsing this forum: fllawyer, Lizzytiz1 and 91 guests