ENDGame: the latest Met Office climate model update

Posted: July 30, 2014 by Andrew in atmosphere, climate, Forecasting, Psychobabble

imageThe Met Office has unveiled their latest update:

Even Newer Dynamics for General atmospheric modelling of the environment (ENDGame)

ENDGame is an evolution of the current dynamical core, the New Dynamics, and is based on a semi-implicit semi-lagrangain discretisation of the governing equations.

An interesting name choice for a climate model ?

The Met Office website describes its update:

In common with the New Dynamics, ENDGame is a finite-difference model discretised on a latitude-longitude grid and is based on the fully compressible, nonhydrostatic Euler equations. However, distinct from the New Dynamics, ENDGame has been designed to allow the code to be switchable between various options: a nonhydrostatic and a hydrostatic formulation; deep-atmosphere and shallow-atmosphere formulations, and use of spheroidal, spherical or Cartesian co-ordinates (as appropriate).

Exact mass conservation can be critical for long-term climate simulations. In the New Dynamics this is achieved by the application of a flux form Eulerian scheme for dry density. The use of an Eulerian scheme within an otherwise semi-Lagrangian model is undesirable. Therefore, a novel mass-conserving, semi-Lagrangian transport scheme (SLICE) has been developed that will be available within ENDGame for transport of dry density and other scalar variables that need exact conservation.

An important aspect of ENDGame is that it is designed around an iterative approach to solving the semi-implicit aspects of the scheme. This permits more accurate coupling of the scheme to the physics parametrizations. It also produces a simpler form of the Helmholtz problem that arises from the semi-implicit discretization. Building on recent research into the dispersion properties of Rossby waves on C-grids ENDGame has improved handling of these meteorologically significant waves whilst retaining the same optimal properties. This has required a modification to the staggering of variables with respect to the poles. These two aspects, the different Helmholtz problem and the changes to the polar discretization, are expected to improve the scaling of the model on massively parallel computer architectures.

Key aims
To improve the robustness, accuracy and efficiency of the dynamical core of the UM.
To maintain at least the functionality of the current dynamical core.
To develop and maintain a suite of test environments: a shallow-water version; a Cartesian vertical slice version; and the full 3D dynamical core version on a sphere.

Read full details of the ENDGame update here

  1. Stephen Richards says:

    Come on Betts, give us a laugh. Your forecast for 2014/5 winter, spring and summer. Should be easier for the “endgame”. Pathetic, truly pathetic.

  2. Schrodinger's Cat says:

    Is the MO buzz-word generator overheating?

  3. johnbuk says:

    I put this through Google translate and it came up with, “We have added mascara and stiletto heels to complement the lipstick”.
    Cynic? Moi?

  4. oldbrew says:

    Is there any logic to any of this?

    NB MetO paper says: ‘Research has already started on the next-generation dynamical
    core (named GungHo) which we expect to replace ENDGame in about 10 years.’

    Name seems apt 😉

    MetO says: ‘GungHo is an anglicised pronunciation of ”gong he”, where the Chinese characters ”gong” and ”he” are translatable individually as ”work” and ”together”, so this name reflects both the ambitious and the collaborative nature of the project.’

  5. Jaime says:

    Initial thoughts: very curiously named. ENDgame for who? I suspect they imagine the hoardes of climate change deniers unconvinced by the current accuracy of climate modeling and ensemble predictions of greenhouse gas warming who will be blown away by the improved resolution and iteratively achieved accuracy of the Even Newer dynamical core.
    The Met Office will finally be able to identify the anthropogenic fingerprint of extreme weather patterns no doubt.
    I note that in the list of improvements to the model’s physics, we have ” Improved absorption of longwave radiation by CO2 and O3 which improves heating and cooling in the stratosphere.” We also have a “new cirrus term” listed as an improvement to the representation of ice cloud.
    Finally, ENDgame is not the end for the development of computational dynamical cores. The next generation dynamical core is already in the pipeline, even more curious named Gung Ho:

    “Unthinkingly enthusiastic and eager, especially about taking part in fighting or warfare”

    “to be gungho. meaning to be so psyched out for war. a mentality,thinking you are bullet-proof. your so hyped up and pumped so much. you can’t wait to rush in and attack with your fellow soldiers.”
    [Asian origin, 1940’s battlefield slang]

    Very apt.

  6. Joe Public says:

    To whom do I apply for a grant to develop my brainchild?

    “Modelled Atmospheric Dynamics Necessitating Evermore System Subsidies”

  7. Joe Public says:

    “Before the endgame the gods have placed the middle game.”

    Siegbert Tarrasch

  8. steverichards1984 says:

    Developing more accurate and faster simulators is truly good.

    Its just a shame that fundamental physical processes which they model are not yet understood.

  9. craigm350 says:

    Reblogged this on the WeatherAction Blog and commented:
    Build a nice shiny skyscraper on shifting sands and tell everyone to worship how amazing you are. They really are in love with their own creations, Slingo recently saying they are a scientific wonder of our age. A ‘wonder’ yes that they cannot model the important things like clouds and oceans and yet are beheld over observations?! Another white elephant and they want a herd of the bloody things. They should change the name to PURSEsnatcher and just be as blatant as Operation Iraqi Liberation (OIL) was.

    Stop wasting public money on shiny knobs and do some basics like improving weather stations, sacking overpaid PR buffoons like Slingo and maybe improving our historic records as Lamb suggested long ago (which we could have had for a fraction of the cost of these modelling follies).

    I really wish someone would put their foot though the monitor, forcing these narcissists out into the sunlight.

  10. Roger Clague says:

    THe MO say

    However, distinct from the New Dynamics, ENDGame has been designed to allow the code to be switchable between various options: a nonhydrostatic and a hydrostatic formulation

    The troposphere, where we get weather, is hydrostatic, so why allow a non-hydrostatic option?

    There is no reason except so as to have another fudge factor. This is the usual computer- based climate science approach of Combining models with different physics.

  11. John, UK says:

    I look forward to seeing the website text as quoted being translated into English.

  12. M Simon says:

    johnbuk says:
    July 30, 2014 at 9:17 am

    I put this through Google translate and it came up with, “We have added mascara and stiletto heels to complement the lipstick”.
    Cynic? Moi?

    Combine that with:

    oldbrew says:
    July 30, 2014 at 9:22 am

    And you can see the name of an upcoming (heh) model: Big Ho.

  13. A C Osborn says:

    Could it be that the new Program will find some new data that persuades them it is not going to get quite so hot as the old one predicted?
    Thus alowing them “wriggle room” when the pause continues or it gets cooler.

  14. AlecM says:

    Betcha it still claims the surface emits to the atmosphere energy at the black body rate, when it’s really 1/6 that rate on average.

  15. They talk about Rossby wave modulation, if they only knew that this is the effects of the lunar declinational tides in the atmosphere, they could use past real data to forecast the future movement of the wave patterns and hence the frontal boundaries they create and control. This alone would make the models have better resolution in both the short term, (days and weeks) but also consistency into weeks and months ahead.
    As long as they neglect the “internal natural variation” as they call it, which is really the composite of solar/lunar tidal effects and solar wind influenced outer planetary interactions with the magnetosphere of the earth, then the short term outbreaks of severe weather they control will not show up in their models, nor will the longer term effects ENSO, NAO, PDO, that are also derived from the modulation of the global circulation by planetary effects, be understood or show up in their outputs.
    They will have the same problems the current crop of GCM’s have in not showing the big swings in the global modes and patterns of circulation.

  16. By the way I have daily forecasts for North America, and Australia out to November the 4th 2015 posted to my site, if you are interested in looking at the basic trends of the NH winter ahead.

  17. Curious George says:

    I can hardly believe that models still use a latitude-longitude grid. This improvement seems long overdue. That said, I am all for better models. An actually usable model would be a major step forward.

  18. DD More says:

    New Dynamics has served us well over more than a decade: not only have we continued to improve the skill of our large scale forecasts at the rate of 1 day lead time per decade (so for example today’s 3 day forecast is as accurate as the 2 day forecast was 10 years ago) but we have seen the introduction of a very high resolution (1-1/2 km) model over the UK which provides unprecedented levels of detail to our forecasters.

    So at this rate they will be able to get a 7 day forecast just a accurate at the 2 day forecast in only 40 more years. Now if they could just get an accurate 2 day forecast they might have something to sell.

    Hope it does not finish up like ‘Ender’s Game’ and an entire fleet is crashed into the planet to finish the war.

  19. David Schofield says:

    How lucky are we! Just as a problem that can only be identified by supercomputers comes along – we have super computers!

  20. oldbrew says:

    DD More says: ‘So at this rate they will be able to get a 7 day forecast just as accurate as the 2 day forecast in only 40 more years. ‘

    So for a 100-year climate projection to have any likelihood of being accurate we will have to wait a lot longer than anyone’s actual life span 😉