Jump to content

2020-21 Global Model QPF/Snowfall Evaluation For PHL. 11/23 Event. New GEFS Gets 1st 1st Place Finish; OP EC Comes In Last.


Recommended Posts

We can't thank everyone enough about all the views we get for this thread. This past season the over 33,700 views was the most ever for any season and the most viewed short term thread of the entire cold season. :clap:

Now onto this season and changes we are making.  Since it is not being supported any longer, we are dropping the NAM.  No one looks at the SREF unless the ARW(s) go ballistic, so they are gone too.  In their place we are adding the CAN GGEM & UKMET.  We are still going to stick within around 84 hours of the end of the event as ensembles sometimes bleed into other events and it is hard to separate them.  Not sure if Pivotal will keep the UKMET free, but we can hope.  This also will standardize snowfall, all of the models will be 10:1.  We know, not perfect, and that should be taken into consideration, especially with fall and late season events.

In case the shock has not worn off, yes the GFS won the last cold season as the Euro took a tumble. Here are the accumulative results for the last three seasons:

 

202021.JPG.ca00d74866f0e3c4c42ac2e38a0d06df.JPG

 

201819.JPG.57af1e66659c68fc6c613afa5128b797.JPG

 

201718.JPG.d8cdf12c8d35fdf94f200db628a29ebc.JPG

 

By design, by resignation to preserve other accuracies, the models all had a wet bias.  Before we start with the Thu/Fri, I will post the brief synopses of the dry events.  But to show how wet they tend to be, last cold season:

The one thing I must emphasize is that the models have an inherent wet bias.  51 total 2019-20 Events, forecasts within 84 hours.

35 events all the 6 evaluated models/ensembles averaged too wet for the event.

13 events none of the model/ensemble runs were drier than the actual precipitation amount.

10 events some models/ensembles averaged too wet, others too dry.

6 events all the evaluated models/ensembles averaged too dry for the event.

1 event (12/9/2019) none of the model/ensemble runs were wetter than the actual precipitation amount.

 

There was an upgrade to the EC and GEFS for this season.  I don't know when V16 of the GFS will be started, but I am not impressed by its even snowier forecasts so far (This was with the Denver event too, not just the 2-3 feet it has had with some runs with this upcoming event only).

Any questions, comments are appreciated.  Enjoy and we hope to have to wear out the snow tab this season.

 

  • Like 3
Link to post
Share on other sites

Here are the six events in which all of the models/ensembles averaged too dry. The highlighted December 9th event was the only event in which all model/ensemble runs were too dry. Contrast that with there were thirteen events last cold season in which all model/ensemble runs were too wet. That is six of fifty-one events and one of them involved upright convection.  A general rule of thumb is if your model is wetter than the gefs, you are in trouble.

dry.JPG.41ea247c8df9c734156f6e139121cb67.JPG

Link to post
Share on other sites

EURO

10/27 0Z: 2.43

10/27 12Z: 2.22

10/28 0z: 1.9

10/28 12z: 2.2

10/29 0z: 2.6

EPS

10/27 0Z: 2.05

10/27 12Z: 2.14

10/28 0z: 2.37

10/28 12z: 2.15

10/29 0z: 2.31

  • Like 1
Link to post
Share on other sites
On 10/27/2020 at 6:42 PM, tombo82685 said:

EURO

10/27 0Z: 2.43

10/27 12Z: 2.22

10/28 0z: 1.9

10/28 12z: 2.2

10/29 0z: 2.6

EPS

10/27 0Z: 2.05

10/27 12Z: 2.14

10/28 0z: 2.37

10/28 12z: 2.15

10/29 0z: 2.31

Maybe the Euro is as wet, but not surprising GFS makes threshold for Sunday CFP. Can we start with the 00z run this morning? (30th).

Link to post
Share on other sites

euro

10/30 0z: .12

10/30 12z: .55

10/31 0z: .46

10/31 12z: .22

11/1 0z: .43

11/1 12z: .37

eps

10/30 0z: .12

10/30 12z:  .43

10/31 0z: .38

10/31 12z: .34

11/1 0z:  .3

11/1 12z: .35

  • Like 1
Link to post
Share on other sites

First one in the books and not an easy feat to overforecast precipitation amounts with the remnants of a tropical system.  But as far as PHL goes, all the models/ensembles did.  At least for one event the world returned the normal.  The EPS came in first, last season's winner GFS came in last.  I didn't see any snow for PHL among these models/ensembles, the Para GFS did.  Now we know the go to model for this winter.

 

2.JPG

past48.png

Link to post
Share on other sites
  • Rainshadow5.8 changed the title to 2020-21 Global Model QPF/Snowfall Evaluation For PHL. 10/29-10/30 Event...EPS Starts Off With A Win; GFS A Too Wet Loss.

Well the GFS has no precipitation until its prediction of the remnants of ETA coming here in mid November.

Glad we started when we did. 

So this Sunday event, it had some similarities to one of the few we are too dry events of last season, a moisture plume advecting in from the Carolinas.  What once looked like a simple CFP started getting more complicated about two days before the onset.  The models that did the best with this (GGEM & EC) caught onto it one sounding cycle faster than the Ukie and two sounding cycles faster than the GFS. 

So the EC came in first as the GGEM went down its too wet bias road (only model to average too wet) while the GEFS came in last by having the wrost of two worlds, too dry when the GFS was too dry (aren't they suppose to show us uncertainty not follow the leader?) and then wetter than the GGEM by go time.

ddd.JPG

g.JPG

past48.png

Link to post
Share on other sites
  • Rainshadow5.8 changed the title to 2020-21 Global Model QPF/Snowfall Evaluation For PHL. 11/1 Event. EC Comes In First; Worst Of Both Worlds GEFS Comes In Last.
On 10/30/2020 at 5:55 AM, tombo82685 said:

euro

10/30 0z: .12

10/30 12z: .55

10/31 0z: .46

10/31 12z: .22

11/1 0z: .43

11/1 12z: .37

eps

10/30 0z: .12

10/30 12z:  .43

10/31 0z: .38

10/31 12z: .34

11/1 0z:  .3

11/1 12z: .35

We can start the Eta tropical connection event with the 00z run this morning (9th).

 

Link to post
Share on other sites

euro

11/9 0z: 4

11/9 12z: 1.8

11/10 0z: 1.6

11/10 12z: 1.8

11/11 0z: 2.03

11/11 12z: 1.78

eps

11/9 0z: 2.15

11/9 12z: 1.85

11/10 0z: 1.9

11/10 12z: 1.93

11/11 0z: 2.11

11/11 12z: 2.34

Friday event

euro

11/10 12z: 0

11/11 0z: .03

11/11 12z: .15

11/12 0z: .15

11/12 12z: .08

11/13 0z: .05

eps

11/10 12z: .05

11/11 0z: .11

11/11 12z: .08

11/12 0z: .12

11/12 12z: .05

11/13 0z: .08

  • Like 2
Link to post
Share on other sites
17 hours ago, tombo82685 said:

euro

11/9 0z: 4

11/9 12z: 1.8

eps

11/9 0z: 2.15

11/9 12z: 1.85

We might need a Friday event separation.  We are trending that way prior to 12z run.  12z GFS held off (.1" southeast of PHL), unless you are already seeing something on the EC/EPS?

Link to post
Share on other sites
22 hours ago, tombo82685 said:

euro

11/9 0z: 4

11/9 12z: 1.8

11/10 0z: 1.6

eps

11/9 0z: 2.15

11/9 12z: 1.85

11/10 0z: 1.9

No GFS too wet at low amounts forecast goes unpunished, so yes there is a Friday event now.

So if you can cut off part 1 at 00z/Friday and include everything else beyond 00z Friday.  The second part can start with the 12z run today (10th).

Link to post
Share on other sites
On 11/9/2020 at 6:12 PM, tombo82685 said:

euro

11/9 0z: 4

11/9 12z: 1.8

11/10 0z: 1.6

11/10 12z: 1.8

11/11 0z: 2.03

11/11 12z: 1.78

eps

11/9 0z: 2.15

11/9 12z: 1.85

11/10 0z: 1.9

11/10 12z: 1.93

11/11 0z: 2.11

Friday event

euro

11/10 12z: 0

11/11 0z: .03

11/11 12z: .15

eps

11/10 12z: .05

11/11 0z: .11

11/11 12z: 

At least we can stop one of them with the 12z run today.

 

Link to post
Share on other sites

@tombo82685, Tom the 00z run tonight can be the last one for the early Friday morning event and the first one for the CFP on Sunday afternoon/night.  That 4" first forecast by the Euro will crush it for this soon to end event, I think the EPS is going to come in first with this one.

 

 

Link to post
Share on other sites

Well the EPS does remnants of tropical systems quite well.  Second similar event it kicked the buttocks of all other global op models and ensembles.  The OP EC might have had a chance if not for that 4" start.  The (too dry) GEFS came in last.

1111.JPG.1fe6f724bca69a5bf6b1782ceef8c2f7.JPG

This was about 90% of the entire event.

24hr_1.png.7e91690a358f2a4747c857dc04e24d27.png

Link to post
Share on other sites
  • Rainshadow5.8 changed the title to 2020-21 Global Model QPF/Snowfall Evaluation For PHL. 11/11-12 Event. EPS Does Tropical Connections Well Came In 1st, GEFS Last.

Euro

11/13 0z:  .25

11/13 12z: .38

11/14 0z: .42

11/14 12z: .37

11/15 0z: .44

11/15 12z: .33

eps

11/13 0z: .38

11/13 12z: .44

11/14 0z:  .39

11/14 12z: .38

11/15 0z:  .41

11/15 12z: .32

Link to post
Share on other sites
17 hours ago, Rainshadow5.8 said:

This parting shot today (13th) it is going to be either the GEFS or EPS.  Barring something really weird by the OP EC and even weirder for the EPS, the too wet GGEM will be in last.

Yup, make that 3 out of the first 4 events with the EPS in first.  Much better start than last cold season.

11.JPG.b351144c1a3b2b40b5e0ff903d05216e.JPG

 

Link to post
Share on other sites
  • Rainshadow5.8 changed the title to 2020-21 Global Model QPF/Snowfall Evaluation For PHL. 11/13 Finisher. EPS Off to Hot Start, 1st Again, Wet GGEM Last.
On 11/13/2020 at 7:28 PM, tombo82685 said:

Euro

11/13 0z:  .25

11/13 12z: .38

11/14 0z: .42

11/14 12z: .37

eps

11/13 0z: .38

11/13 12z: .44

11/14 0z:  .39

11/14 12z: .38

Let's make the 12z today (15th) the last, too close to 00z/16th to be useful to include that one.

Link to post
Share on other sites

This is almost nostalgic as the EPS continues on fire.  While PHL was in one of the heavier bands, it's amount was not outrageously (fast movement did not hurt) high and given that all of the models were too high, the EPS/EC would probably come in first with a slightly wider basin average.  The GFS that always does poorly with summer convective amounts came in last.  Hope the EPS saves some of these for the winter.

 

1.JPG

past24.png

  • Like 1
Link to post
Share on other sites
  • Rainshadow5.8 changed the title to 2020-21 Global Model QPF/Snowfall Evaluation For PHL. 11/15 Event. EPS On Fire, 1st Again, GFS Is Not (Last).
On 11/13/2020 at 7:28 PM, tombo82685 said:

Euro

11/13 0z:  .25

11/13 12z: .38

11/14 0z: .42

11/14 12z: .37

11/15 0z: .44

11/15 12z: .33

eps

11/13 0z: .38

11/13 12z: .44

11/14 0z:  .39

11/14 12z: .38

11/15 0z:  .41

11/15 12z: .32

Monday event we can start with the 00z run this morning (20th).

Link to post
Share on other sites
  • Rainshadow5.8 changed the title to 2020-21 Global Model QPF/Snowfall Evaluation For PHL. 11/23 Event. New GEFS Gets 1st 1st Place Finish; OP EC Comes In Last.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...