How to read an ensemble map?
Posted: Thu Dec 07, 2006 3:16 pm
Does anyone know of a good primer on how to accurately read and interpret the ensemble long-range maps on the CDC and NCEP sites?
STORM2K
Welcome to Storm2k! Your Year Round Weather Community since 2002!
http://www.storm2k.org/phpbb2/
How to read an ensemble map?
Page 1 of 1
Posted: Sat Dec 09, 2006 8:03 am
Got a link to the maps you're asking about? Thanks, HPH
Posted: Sat Dec 09, 2006 7:53 pm
Posted: Sat Dec 09, 2006 10:30 pm
Interpret it just like you would interpret any other model.
The shaded region is the anomaly, i.e. the departure from normal. The lines on the map on the left show 500mb geopotential height contours, I think.
The Ensembles are a group of models that are run off a parent model. The CMC and GFS Ensembles are available publicly on the EWall site. http://www.meteo.psu.edu/~gadomski/ewall.html There are also EURO ensembles.
What the map is showing is the ensemble mean. Each individual ensemble member uses its own equations, which yield different results as time passes by (after about 180 hr, you would probably see all types of solutions). The ensemble mean is usually the most accurate for the medium range, as usually, the mean averages out the anomalies of the individual ensemble members, but keeps the areas where they agree on, which are usually more accurate. There are however times where this is not true, mainly where there is greater variablitiy.
The individual ensemble members are very good for determining the variability of the pattern. If they are all over the place, 1) the pattern is going to be hard to pin down, and 2) The ensemble mean could be inaccurate. But if they agree in the medium and long range, then forecasting for medium-range and long-range trends is made a lot easier.
Hope this helps.
The shaded region is the anomaly, i.e. the departure from normal. The lines on the map on the left show 500mb geopotential height contours, I think.
The Ensembles are a group of models that are run off a parent model. The CMC and GFS Ensembles are available publicly on the EWall site. http://www.meteo.psu.edu/~gadomski/ewall.html There are also EURO ensembles.
What the map is showing is the ensemble mean. Each individual ensemble member uses its own equations, which yield different results as time passes by (after about 180 hr, you would probably see all types of solutions). The ensemble mean is usually the most accurate for the medium range, as usually, the mean averages out the anomalies of the individual ensemble members, but keeps the areas where they agree on, which are usually more accurate. There are however times where this is not true, mainly where there is greater variablitiy.
The individual ensemble members are very good for determining the variability of the pattern. If they are all over the place, 1) the pattern is going to be hard to pin down, and 2) The ensemble mean could be inaccurate. But if they agree in the medium and long range, then forecasting for medium-range and long-range trends is made a lot easier.
Hope this helps.
Posted: Sun Dec 10, 2006 9:33 am
It does help ... quite a bit! Thanks so much ... 
Posted: Sun Dec 10, 2006 10:20 am
There's a (sort of long-winded) explanation of all this a couple of clicks back from those specific products, at this address: http://www.cdc.noaa.gov/spotlight/12012001/.
To expand on the previous answer just a bit, paragraph by paragraph:
(1) The two maps on that page are the departures from the mean (on the left) and the normalized departures from the mean, the latter being a measure of how big (relative to other such signals) this set of departures is. That is, if you see a big low-pressure area, the plot on the right will tell you how big that is relative to other such things in the climatology.
(2) If you poke around a bit on that web site and find their "spaghetti plots", you'll see how much variability there is within a particular ensemble -- the more twisted up the spaghetti is, the less confidence you should have in the mean. HPH
To expand on the previous answer just a bit, paragraph by paragraph:
(1) The two maps on that page are the departures from the mean (on the left) and the normalized departures from the mean, the latter being a measure of how big (relative to other such signals) this set of departures is. That is, if you see a big low-pressure area, the plot on the right will tell you how big that is relative to other such things in the climatology.
(2) If you poke around a bit on that web site and find their "spaghetti plots", you'll see how much variability there is within a particular ensemble -- the more twisted up the spaghetti is, the less confidence you should have in the mean. HPH
Posted: Sun Dec 10, 2006 4:56 pm
Thanks DrCloud ... again, appreciate your help and the post from wxmann_91!!