More on NOAA's Summer Outlook
Last week, I wrote about how the National Oceanic and Atmospheric Administration (NOAA) Climate Prediction Center's (CPC) summer outlook could be more transparent. Specifically, I criticized CPC's outlook map, which paints the metro area in orange -- suggesting HOT -- whereas its own probabilities suggest a 60% or better chance it will be average or cooler than normal. I stated that a more meaningful way to present this information would be to use pie charts to clearly break down the chances of different outcomes. Well, I got a tip that CPC does just this, so let's take a look at its product...
Probability of it being near normal (white), above normal (red), and below normal (blue) at DCA in June-July-August, according to CPC. Courtesy NOAA CPC.
The CPC generated pie-chart to the right shows the chances of temperatures being normal (white), above normal (red) and below normal (blue) for the June-July-August period at DCA. As you can see, this chart gives roughly equal chances of it being warm, average and cool. But wait... on CPC's own national map, DCA is shaded in orange, which would suggest greater chances of being hot! So which is the real CPC outlook? The local pie-chart or the national map? I would love to know.
Probability of it being near normal (white), above normal (red), and below normal (blue) at BWI in June-July-August according to CPC. Courtesy NOAA CPC.
Taking a look at the pie-chart for nearby BWI raises further questions. Somehow BWI has a 52% chance of being warmer than normal while DCA -- a mere 35 miles to the southeast -- only has a 33% chance. Why in the world are the probabilities so different? No science can credibly support a difference of that magnitude over that distance, so I suspect there's an error. Furthermore, the pie-chart (like the DCA pie-chart) is inconsistent with the national map. On the national map, there's no contour for representing a 50% or greater probability of warmer than normal temperatures in the entire mid-Atlantic and Northeast region. Will the real probabilities please stand up?
My last complaint about these pie charts is that they are difficult to find. I was unable to locate a quick link to them from CPC's main Web site and they are pretty buried within the local National Weather Service pages.
Bottom line: I applaud NOAA's CPC for developing these pie-charts. They clearly convey probabilities. But to have any real application, they need to be easy to find, internally consistent and consistent with related products. Right now, they seem to be none of the above...
The Capital Weather Gang's summer outlook comes out tomorrow. Stay tuned...
By Jason Samenow |
May 27, 2008; 10:30 AM ET
Local Climate
Previous: Forecast: Return to Rainy Reality |
Next: How Many 90 Degree Days This Summer?
Posted by: weatherdudeVA (Lake Ridge) | May 27, 2008 11:07 AM
According to Ed O'Lenic, Chief, Operations Branch
NOAA-NWS-Climate Prediction Center, this product does appear to be inconsistent and understands we (CPC) has to do something about it. The format of this product was recently taken over by Ed's group from the NOAA/NWS Climate Services Division, and recognizes its now his responsibility.
For additional information on the full suite of CPC's forecast products, see: http://www.emc.ncep.noaa.gov/research/NCEP-EMCModelReview2007/CPC-Olenic_Annual_Review_2007.pdf
Posted by: Steve Tracton, Capital Weather Gang | May 27, 2008 12:49 PM
The comments to this entry are closed.
I'm eagerly waiting the CWG Summer Outlook. Hopefully the words "blizzard" and "week-off-of-school" are in there.
What? Can't someone wish? Hehehe.
I remember last summer when it hit 103 here in Lake Ridge. I took like 20 pictures of the thermometer, LOL.