Ridge Filter Surface Temps

Paul note 2013 Himalay has stronger renditiong:

Ridge Filter 2013

And as you can see the regions are actually ridges of rings or braids or knots!

2010 lesser intensity for Himalaya ring:

Ridge Filter 2010

Also in both years you could clearly see the Mahler patterns, I strongly believe that these are standing-waved-echos off the South American coasts. Since the Ridge filter is actually deals with curvature of data at each pixel, and that repeated pattern looks like the waving echos off the cost of Chile?

Programmer's Manual:

Ridge Filter


Principal curvature


  • Options

    Dara, that is a pretty convincing view of a Sudden Stratospheric Warming (SSW) event around the Himalayas

    RG gives an interesting take on what is going on at Neven's blog http://neven1.typepad.com/blog/2013/04/sudden-stratospheric-warmings-causes-effects.html


    Also, talking about standing wave patterns in the southern east Pacific, there is a very interesting theory making the rounds concerning Rapa Nui (Easter Island). What several research articles are finding is that Rapa Nui may actually be located at a null location or node in the ENSO standing-wave pattern. The lack of strong weather disturbances in the climatology history of Rapa Nui is supporting this viewpoint. Those waves coming off of Chile may overlap Rapa Nui if extended.

    [1] Genz, Joseph, and Terry L. Hunt. "El Nino/southern oscillation and Rapa Nui prehistory." Rapa Nui Journal 17.1 (2003): 7-14. [2] Mann, Daniel, et al. "Drought, vegetation change, and human history on Rapa Nui (Isla de Pascua, Easter Island)." Quaternary Research 69.1 (2008): 16-28.

    Both of these are available as PDFs on Google Scholar.

  • Options

    Thanx Paul I read the refs.

    So the application of filters and differential operators to volumetric data is quite useful, mind you I did not take a single moving average of any kind, what you see is what is in the data (transformed or operated on per pixel).

    I was thinking we could come up with equations that govern these operators or filters on volumetric data as opposed to 1D or 2D regions.


  • Options

    Dara, Your work looks very promising indeed for uncovering the underlying dynamics.

    Reminds me of the E-M fields that occur in a waveguide http://en.wikipedia.org/wiki/Waveguide


  • Options

    Actually the original volume rendering of all I post here looks like the E-M fields gif you posted in #4, but requires a Slicer to view and people did not seem to like to operate a slicer.

    So I rewrote the code to dump a flipbook into Youtube for ease of access.

    I am almost convinced that something like Maxwell Eq could be formulated to get some good models for some of the data patterns we see. In particular the braided solutions to differential equations.

    If you like that E-M field rendering I could make a large number of Slicers you could view to get ideas how these volumetric data are put together.

    We are working on automating this process, so you could login into an account and most of the code is there you just need to change a few thigns and fire off a large number of Slicers.

  • Options

    Dara, I am getting most of your Mathematica CDF plugins to operate, so kudos to you for trying out that functionality.

  • Options

    ONE. Paul it is great that you use the CDFs, I will make a lot more, and hopefully much of it will be computed and delivered by our servers

    TWO. I see interconnected loops in the Ridge filter! I was expecting all kinds of curves but not so much looped and interconnected.

  • Options

    Any standing wave patterns are dynamite for understanding and simplifying the math. When the standing wave pattern shows clearly fixed nodes in the spatial dimension, I start to feel more comfortable that the temporal dimension may in fact be separable from the spatial dimension.
    In particular I was counting on this separability when I started to look at the sloshing dynamics -- if Tahiti and Darwin are fairly strong fixed-position anti-node dipoles then it may indeed be possible to try a temporal nonlinear wave equation as a first cut.

    As a rule, aren't physicists always seeking symmetry and patterns?

    So keep on doing what you are doing and I will continue to try to keep up.

  • Options


    Here is a paper describing biennial variations centered on the Himalayas

    Bertolani, Laura, Massimo Bollasina, and Gianni Tartari. "Recent biennial variability of meteorological features in the Eastern Highland Himalayas." Geophysical research letters 27.15 (2000): 2185-2188. PDF

    Compare this to the odd/even year pairing here, which is much more apparent:


  • Options

    WOW! Paul, downloded the paper I go and download more years and run the filter

Sign In or Register to comment.