Support the Arctic Sea Ice Forum and Blog

Author Topic: Creating an Arctic Temperature Inversion plot  (Read 3381 times)

Niall Dollard

  • Grease ice
  • Posts: 929
    • View Profile
  • Liked: 372
  • Likes Given: 94
Creating an Arctic Temperature Inversion plot
« on: November 27, 2017, 11:35:58 PM »
Recent discussion about the measure of the Arctic temperature inversions in the freezing thread got me thinking about creating a plot of the difference between 850 hPa temperatures and the 2m temperatures.

Are such plots already available anywhere ? ESRL gives separate plots of both 850hPa temps and 2m temps for both RASM-ESRL and GFS data.

Perhaps A-Team or anyone else could have a look at this?. I have only got as far as downloading panoply, but am eager to learn! 

If 2m temps means 2m ASL (and not 2m above surface of the ground) a simple subtraction of 850 temps minus 2m temps would show the measure of the inversion. A preferred colour palate would be blues for negative results/no inversion (ie 850 temps < 2m temps) changing to green at zero and yellows/reds for positive/high positive results. 


  • Young ice
  • Posts: 2830
    • View Profile
  • Liked: 863
  • Likes Given: 35
Re: Creating an Arctic Temperature Inversion plot
« Reply #1 on: December 05, 2017, 09:31:47 PM »
a simple subtraction of 850 temps minus 2m temps would show the measure of the inversion. colour palate blue green yellow red
ESRL has quite a few items we have not made use of so far, notably the 'meteograms'  and 'xsections' in the archive of 'Reb Plots'. These are mostly for selected weather stations around the western Arctic, with the exception of one I was just looking at today, which seems to be an average the whole Arctic Ocean but only over current sea ice. (These are ten day predictions but include the initial state.)

It's somewhat a nuisance to animate these because of how they're archived, ie the whole bundle of files for each day of the time span has to be downloaded and the specific file fished out, below ArcticOcean_meteogram. Scraping off white space, these are 550x991 pixels.

Below, I'm testing whether the forum will display this because reducing size might adversely affect their thin colored lines. If it works, a faint image of ice extent for that day is easy to underlay.[edit: 700 pxl height works ok, went with that.]

These meteograms provide W/m2 of radiative transfer, ice volume, ice area, air pressure, temperature and precip/6hrs. The display is weirdly fascinating despite some layout erratics at their end.

These do not come with their underlying data files and so could not be reprocessed in Panoply. However if you could do the subtraction graphically in Gimp, one of the Panoply palettes could be brought in. It is probably best to go with ColorBrewer2; those can be butted up or discretized.


« Last Edit: December 05, 2017, 10:52:13 PM by A-Team »