Login to AccuWeather.com Premium Login to AccuWeather.com Professional Login to AccuWeather.com RadarPlus AccuWeather.com

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
> Weather Models
asd123
post Dec 14 2014, 07:47 PM
Post #1




Rank: Whirlwind
*

Group: Member
Posts: 36
Joined: 11-December 14
From: Orlando, FL
Member No.: 30,092





I know that models have, for example ECMWF have an operational, control, and ensembles (ECMWF ensembles comprise of ECMWF Control+op+ens members). But I have noticed models such as ECMWF also have ensemble control and ensemble mean.

Bottom line, what is a model's ensemble control and ensemble mean?


P.S. I did read a similar question on the next page, but it didn't answer mine.

This post has been edited by asd123: Dec 14 2014, 11:22 PM
Go to the top of the page
 
+Quote Post
bdgwx
post Dec 15 2014, 02:40 PM
Post #2




Rank: Whirlwind
*

Group: Member
Posts: 44
Joined: 15-November 14
From: St. Louis, MO
Member No.: 30,021





Totally not an expert...but here is my understanding.

op - The operational run of the model. It is the one everyone sees and is familiar with.

control - This is part of the ensemble system. It is the run that has the best guess at the initial state of the atmosphere and does not perturb the initial conditions or inject perturbations at certain intervals in the forecast phase. It is usually run at a lower resolution than the operational run.

member - One of many different perturbed runs within the ensemble system. The initialization may be tweaked slightly relative to the control run to produce another plausible set of initialize conditions that could represent the current state of the atmosphere. It is also possible that perturbations are injected at some interval in the forecast phase as well. Parameterization schemes could be tweaked between members as well.

mean - This is basically the average of all of the individual members. I'm not sure what all is involved here, but it's possible that it is more complex than just a trivial average. What happens if the trivial average produces an unrealistic representation of the atmosphere or an atmosphere that is not in hydrostatic balance? My point is that when I say "average" be careful not to take it too literally.

The specifics on how any one ensemble system (GEFS, ECMWF, etc.) operates is unknown to me since I haven't bothered to research them...yet. But, I hope this gives you a very rough overview of the terminology at least.

And of course, if I am wrong on any point then someone please chime in and correct any errors. It would be an opportunity for me to learn as well.
Go to the top of the page
 
+Quote Post
asd123
post Dec 15 2014, 03:31 PM
Post #3




Rank: Whirlwind
*

Group: Member
Posts: 36
Joined: 11-December 14
From: Orlando, FL
Member No.: 30,092





QUOTE(bdgwx @ Dec 15 2014, 02:40 PM) *
Totally not an expert...but here is my understanding.

op - The operational run of the model. It is the one everyone sees and is familiar with.

control - This is part of the ensemble system. It is the run that has the best guess at the initial state of the atmosphere and does not perturb the initial conditions or inject perturbations at certain intervals in the forecast phase. It is usually run at a lower resolution than the operational run.

member - One of many different perturbed runs within the ensemble system. The initialization may be tweaked slightly relative to the control run to produce another plausible set of initialize conditions that could represent the current state of the atmosphere. It is also possible that perturbations are injected at some interval in the forecast phase as well. Parameterization schemes could be tweaked between members as well.

mean - This is basically the average of all of the individual members. I'm not sure what all is involved here, but it's possible that it is more complex than just a trivial average. What happens if the trivial average produces an unrealistic representation of the atmosphere or an atmosphere that is not in hydrostatic balance? My point is that when I say "average" be careful not to take it too literally.

The specifics on how any one ensemble system (GEFS, ECMWF, etc.) operates is unknown to me since I haven't bothered to research them...yet. But, I hope this gives you a very rough overview of the terminology at least.

And of course, if I am wrong on any point then someone please chime in and correct any errors. It would be an opportunity for me to learn as well.


Like you said, a model has a op, a control run (no perturbations), and then ensembles members. I know the ensemble is comprised of the control, the operational, and the ensemble members. This average of all this is the ensemble mean like you said.

But what I don't know is what is the ensemble control?
Go to the top of the page
 
+Quote Post
The Snowman
post Feb 21 2015, 01:06 PM
Post #4




Rank: F5 Superstorm
***

Group: Member
Posts: 7,173
Joined: 7-December 10
From: Norman, OK
Member No.: 24,567





QUOTE(asd123 @ Dec 15 2014, 04:31 PM) *
Like you said, a model has a op, a control run (no perturbations), and then ensembles members. I know the ensemble is comprised of the control, the operational, and the ensemble members. This average of all this is the ensemble mean like you said.

But what I don't know is what is the ensemble control?

The ensemble control is the 'base' for the ensembles; think of it as the pristine forecast member. All other ensembles are based off of that control member, and those ensembles are tinkered with a bit to make them work right. Here's the description from the ECMWF agency.

QUOTE
For the medium-range forecasts an ensemble of 52 individual ensemble members are created twice a day. One member is at a higher spatial resolution than the other members (called the HRES at ECMWF), its initial state is the most accurate estimate of the current conditions and it uses the currently best description of the model physics. Another member of the ensemble is at a lower spatial resolution than the HRES but at that lower resolution it utilises the most accurate estimate of the current conditions and the currently best description of the model physics (called the control, CNTL, at ECMWF). The rest of the ensemble members (50 members) are similar to the CNTL but their initial states and model physics have been perturbed to explore the currently understood range of uncertainty in the observations and the model.
The CNTL and perturbed members are continued beyond ten days at a reduced horizontal resolution.


--------------------
Annual Snowfall

2014-2015: 46"
2013-2014: 69.5"
2012-2013: 37''


Groundhog Day Blizzard 2011: 24"
Super Bowl Sunday Blizzard 2015: 18"
Go to the top of the page
 
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 

RSS Lo-Fi Version Time is now: 21st February 2018 - 06:18 PM