Thursday, August 11, 2011

Global surface temp for July - no change.


10 comments:

  1. Quite interesting to see that UAH is higher, I've created a few global temp maps going from Jan-2009 through Jul-2011 and combined them into a megaplot
    http://img847.imageshack.us/img847/8510/uahtempmaps20092011.png

    The tropics are above the base period, I also noticed that the Antarctic region was particular warm for July, this could also contribute to the divergence between RSS and UAH, since RSS has less coverage at the south pole.

    PS I also have a Quicktime movie of the sequence, not sure how to share this on a blog in a an easy way.

    MP

    ReplyDelete
  2. Maybe QBO also affects the anomalies, this oscillation mainly affects the poles at higher altitudes and has an approximate period of two years. Lean used it for a regression analysis of the lower stratosphere temperature. It peaks around July.

    http://en.wikipedia.org/wiki/Quasi-biennial_oscillation

    also see http://www.cpc.ncep.noaa.gov/data/indices/
    http://www.cpc.ncep.noaa.gov/products/CDB/Tropics/figt3.gif

    MP

    ReplyDelete
  3. MP,
    That's a great plot of the LT monthly data. It does bring out the oscillations. These were a big feature of the surface temp plot for July, especially in the NH. I was a bit worried that the marked NH periodicity was an artefact of the spherical harmonic fitting, but I see that your plot has similar features.

    I'd be happy to help make the quicktime available - I could certainly host it here, but I have no experience with actual embedding. I'll check.

    ReplyDelete
  4. Dear Nick,
    Thanks, I've sent you the file with wetransfer.com; I used your westnet e-mail address.

    Cheers,
    MP

    ReplyDelete
  5. MP,
    I had to embed it through the Blogger system - if this poses any restrictions, the download address should work. It did work for me. It runs rather quickly.

    ReplyDelete
  6. Nick,

    Yes it runs rather quickly, I've tried to fix that before but could not get it to work for a lossless quicktime movie (I use the ffmpeg encoder to convert images to a movie). I mainly use these type of movies as an image browser, with the slider in quicktime you can go back and forth, which gives a good impression.

    MP

    ReplyDelete
  7. Nick,

    I wrote a matlab script that allows me to generate maps of single months, seasons, years, moving averages, trends (linear slope) and moving trends. I visualize these maps with GMT (generic mapping tool) using cygwin. I can choose different baselines, and also set the colormap (exponential or linear) It is a messy script, not so portable and a bit slow but it seems to work pretty ok.

    I've implemented all major temperature indices, Hadley/CRU (land, ocean or land+ocean), GISS (250, 1200 land+ocean), NCDC (land+ocean), JMA (land+ocean), UAH (t2lt, t2, t4 and t24-Zou,Fu), RSS (TLT, TMT, TTS, TLS, t23-Zou,Fu and t24-Zou,Fu), STAR (t2, t3, t4, t23-Zou,Fu and t24-Zou,Fu) and a few non temperature indices, like TOMS total column ozone, KNMI MSR ozone and ocean heat content. I was also looking at aerosols, but this is a bit of a jungle.

    If you are interested in a particular set of maps I can try to make it.

    MP

    ReplyDelete
  8. Nick,
    I had a closer look at the spectrum of QBO (data : http://www.geo.fu-berlin.de/en/met/ag/strat/produkte/qbo/) and SH temp, QBO shows a peak around 2.3 years, a hint of this is present in SH temp, however not very convincing. Unfortunately I am not an expert in spectrum analysis. Lean also did regression analysis of temperature against forcings/enso etc on maps, I have been playing with the idea to do this on zonal means, need to find time to do it...

    I've created two more graphs of zonal temperature anomalies for UAH and RSS

    http://img684.imageshack.us/img684/2053/uahzonal.png

    http://img695.imageshack.us/img695/9757/rsszonal.png

    I've chosen zones which have approximately equal areas. My impression is that the difference in coverage does affect the global mean of RSS.

    MP

    ReplyDelete
  9. MP,
    Thanks for all that. I'll make this the subject of a new post, using your images (if that's OK) - esp GISS.

    ReplyDelete