Wednesday, December 23, 2015

Merry Christmas - and a new kind of movie

Firstly I wish a Merry Christmas to all readers. Here is a picture of a local solstice scene:




What AGW can do for you! As mentioned, I've been dabbling with NOMADS systems, and downloading data from GFS. I showed here a conventional animation of relative humidity, as it is convected around the world.

Regular readers will know that I have a strong preference for using spherical projections, in conjunction with active viewing (Javascript). The most powerful way of doing this with movies is with WebGL. I rigged up a system here with high resolution SST, but it's rather slow, and not in good repair. The problem is that too much data needs to be downloaded with each frame. I don't think it's practical to preload a whole movie, though I haven't really tried.

So I thought to go back to a pre-WebGL (and HTML 5) system that I had with displaying ordinary graphics files, as here. You can look at the world from a few viewpoints, typically faces of a Platonic solid. So in this movie, it's the faces of a cube - ie six views, 2 polar, 4 equator. Use the radio buttons to switch. It's again of surface relative humidity - this time of 33 days, starting November 11. It's GFS data from NOAA NOMADS, on a 0.5° grid. Below the fold.

You can vary the speed, in frames/sec, bottom left. Top left, the radio buttons are laid out in cube style, with North Pole view at the top. When you switch views, the movie will continue from current time. It runs in a loop. The date second row has day of month, then hour (6 hr intervals). The key at the right is in fractional RH units. Because there are four frames a day, you'll see a strong diurnal cycle over land, with RH going down as the air warms.

Although I'd like the download of new views to be seamless, it may take a second or two. However, once downloaded, it should reload quickly from cache when switching views.




























































14 comments:

  1. Replies
    1. Thanks, Eli. And my thanks to all who have contributed here over the year.

      Best wishes for the New Year
      Nick

      Delete
  2. What I wanted for Christmas 2014 was an ENSO-neutral warmest year. I got one. What I wanted for Christmas in 2015 was a PDO-El Nino warmest year. I'm getting one. What I want for Christmas in 2016 is a fadin' and laggin' El Nino warmest year. I can hope. Thanks Nick for all this stuff you have assembled.

    ReplyDelete
  3. Nick - Happy holidays and thanks for all the time and effort you spend trying to illuminate the darkness.

    ReplyDelete
    Replies
    1. kevin,
      Thanks for your contributions here and elsewhere - nd all the best for the New Year

      Delete
  4. Nick - excellent animations and visualizations. I agree that spherical projections are ideal, though more difficult to implement, especially for animations. The NP animation nicely shows the intrusion of water vapor from the Atlantic into the Arctic that dominated December (as it did with temperature also).

    The UM-CCI CFSR data for 2015 are now complete, in preliminary form. I calculated a 2015 global temperature anomaly estimate of 0.278C referenced to 1981-2010 based on the UM-CCI CFSR estimates, as compared to the preliminary WxBell CFSR estimate of 0.272C which does not yet include December 31. The NCEI estimate for Jan-Nov 2015 was 0.44C and I expect that including December will bump it up slightly (referenced to 1981-2010). The NCEI 2015 estimate will likely be the highest "ever" (based on their continuously adjusted estimates since 1880), whereas the CFSR 2015 estimate ranks 5th highest since 1979. In my view, all of these annual global temperature anomaly estimates have an overall uncertainty of at least 0.3C to 0.5C and thus I have low confidence in most resulting statistical comparisons and trends.

    ReplyDelete
    Replies
    1. Thanks, Bryan, and for all that information. I agree with you that the reanalysis data isn't so good for trend, or comparison with past years. I think it is useful for current detail, and of course daily information within the month. That said, the NCEP-NCAR analysis hasn't been updated for the last two days. I hope they come good in time to wrap up the year.

      Delete
  5. Nick, I haven't followed your blog long enough to know if you have seen this web site, so I'm passing it along just in case you haven't. They produce a variety of near-real-time displays of GFS output on user selectable map projections. They also appear to have archived old initial conditions, although I have not yet explored to see how far back. The user interface is very simple and fairly intuitive and a lot of different model information can be accessed quickly. It is easy to edit the URL to change the date/time. The only problem I see with their streamline animations is that it implies air motion following the streamlines, which is not correct except for steady-state weather patterns. But it still is an interesting visualization. This view is looking toward Australia on a spherical projection and you can zoom in using the roller on your mouse.
    http://earth.nullschool.net/#2016/01/01/1200Z/wind/surface/level/overlay=temp/orthographic=-226.89,-11.63,414

    ReplyDelete
    Replies
    1. Bryan,
      Thanks for that. I was aware of the nullschools site, though following your link I read more of the documentation, which has lots of good documentation. They seem to make extensive use of WebGL, which has been an enthusiasm of mine, and they have linked it to the huge amount of reanalysis data, which I am currently trying to tap into. Also Javascriot code, which handles stuff in a more advanced way than I do. I'm interested in doing more along these lines but daunted by the data handling involved - maybe those ideas will help.

      The wind tracking I find to be an annoying gimmick - as you say, it isn't physical, but a pseudo visualisation of instant wind values artificially continued. I used to do that nearly thirty years ago, when I could only do steady state CFD. I think it is a distraction from the extensive underlying information.

      Delete
    2. It is a daunting amount of data to deal with, too daunting for me. But overall they do a nice job of it and I get the impression the code is open source. I have not seen a display system that is so flexible anywhere else. For instance, on the spherical view you can flip the poles and put the south pole at the top of the screen like this: http://earth.nullschool.net/#current/wind/surface/level/overlay=temp/orthographic=-44.64,209.71,414

      Although when you do that, pulling to the right with the cursor moves the globe to the left, which takes some getting used to.

      I like their color scheme for air temperatures, which I can't say for a lot of other weather animation web sites. It really helps to visualize where it's hot and cold.

      I suppose another route to flexible spherical views would be to map onto Google Earth. I have seen mapping onto Google Maps, like on WunderMap by the Weather Underground, but I don't like their childish data displays. They have managed to tap into a very large amount of hobby weather data that can be useful at times to see details that otherwise would be missing. But the hobby data comes with much lower QA/QC and must be used cautiously for that reason.

      Delete
    3. Bryan,
      "I get the impression the code is open source"
      Yes, I've downloaded the code. It is open, but uses rather general structures from libraries, which makes it harder to follow.

      The flexibility of the display is really just WebGL at work. You can drag a globe to any position. It can get messy, and I have found it necessary to provide an orient button which lets you get the central longitude back to N-S.

      "to map onto Google Earth"
      CCI-Maine does this using a G-E facility that lets you overlay a lat-lon flat graphics, even for whole earth. I've used GE with kmz files, mainly for showing stations (eg here). I now use Google Maps (eg here) because I can use Javascript controls and selection.

      Delete
  6. Obviously Nick is from Australia. Who here has been to Australia? My Dad was a veterinarian, and he was invited to give a talk to an Australian cattlemen's association. He went back about 6 times, and my family accompanied him on one of those trips. We landed in Sydney around 1 AM their time... after about 19 sleepless hours on an airplane. By the time we got into our rental car it was 2 AM. Like Chevy Chase, I headed out of the airport and went around the first round about the wrong way. We headed right out to the middle of the outback to do some work at a huge cattle station. The whole trip was an adventure. My son and I went scuba diving at the Great Barrier Reef. A cowboy went under a steer and they had an air ambulance fly him to Townesville to fix all of his broken bones. My favorite drive was the road into Gloncurry: http://www.jazclass.aust.com/blog/photos21/blog208.jpg

    ReplyDelete
    Replies
    1. JCH,
      Well, you certainly did some travelling. I've never been to Cloncurry (or NW Qld), though the kind of landscape in your photo is familiar.

      We have family from NYC staying with us at the moment, so we're very conscious of the long plane trip. It's about 14-15 hrs now from LA, and they won't let planes land at Sydney at 1am. It's always amazing to me that in that time you can cross the whole Pacific.

      Delete
    2. Based on Mel Gibson movies, I expected that every Australian spent a lot of time racing across the outback. The reality seemed to be quite the opposite. It did not appear that any Australians drove fast... they seemed to assiduously obey the speed limits, which were tortoise-like by Texas standards. I was speeding like Mel in hi squad car, and the occasional cop just went on by. I was never pulled over (long story). It seemed like most of the Australians I met had done very little traveling around greater Australia. I would ask at the gas stations about what to see in the next town, and they would always say they had never been to the next town.

      Delete