Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Climate Models

  1. Jul 14, 2007 #1

    wolram

    User Avatar
    Gold Member

    One of the factors that can produce a winning F1 car is the understanding of turbulent flow, the people that design these cars spend millions on wind tunnels, simply because turbulent flow has not yielded to mathematics, my way of thinking is ,that it is essential to understand turbulent flow before any one can predict the out come of any new input to any thing as dynamic as the Earths weather.
     
  2. jcsd
  3. Jul 14, 2007 #2

    Ivan Seeking

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    On what do you base your opinion?
     
  4. Jul 14, 2007 #3

    Astronuc

    User Avatar

    Staff: Mentor

    It certainly has. The field of computational fluid dynamics (CFD) has exploded in the last couple of decades, and simulation of turbulent flow of liquids and gases is certainly a major application. Every automobile designer uses CFD in their design process. It's also used in engine design and performance analysis.

    We have much better computers and software these days, and improvements in understanding of fluid dynamics and energy transport are continuous.

    Certainly the earth's climate is extremely complex and better models are needed to capture local effects.


    With respect to turbulence, one should keep in mind that the Reynolds number of atmospheric flows is quite low compared to turbulence produced by high speed systems like aircraft, turbomachinery or forced convection systems. Also, the dimensional scales are much, much greater in the atmosphere than man-made systems.

    But Wooly is correct in asserting that we need much better models, which only comes from better understanding of the physics of the atmosphere - both short term and long.
     
    Last edited: Jul 15, 2007
  5. Jul 15, 2007 #4

    Ivan Seeking

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    To what extent can we calculate the margin of error for such models?

    To what limit must we understand local effects in order to make global projections? I don't know if the analogy works, but one that comes to mind is that I don't need to predict the precise path of a Pachinko ball to know that it will end up at the bottom of the machine.
     
  6. Jul 15, 2007 #5

    Astronuc

    User Avatar

    Staff: Mentor

    I don't know. Error bars that I've seen on graphs seem much larger than the magnitude of change in temperature, i.e. is could be 1°C +/- 5°C, or the uncertainty could be greater.

    Last year, climate experts were predicting a very active hurricane season in the Atlantic. Well, we had a fairly quiet season, and many hurricanes turned north to northeast and did not reach the US. Clearly something was missing in the models.

    I think the models need to include sources and sinks of thermal energy and atmospheric composition, and that is certainly easier said than done.

    During the month of June, we had relatively cool evenings and nights, but the days were apparently warmer than normal, July seems to be hotter.

    We've had periods of drought punctuated by heavy rains (thunderstorms), and we've had periods of record precipitiation and flooding.

    Evo has mentioned the cold weather lingering in the central part of the country, and now parts of the US midwest, SW and SE are having drought conditions.

    http://drought.unl.edu/dm/archive/20070710/pics/conus_dm_070710.jpg
    from http://drought.unl.edu/dm/archive.html (images change weekly)

    Dandelions, Poison Ivy Grow With Global Warming
    http://www.npr.org/templates/story/story.php?storyId=11903786
    by John Nielsen
    We have seen an increase in poison ivy in our region, and that also seems to be coupled with increased stress on various deciduous and coniferous trees.


    NPR's Global Warming Page - http://www.npr.org/templates/story/story.php?storyId=5394326

    Florida Joins Climate Change Campaign
    http://www.npr.org/templates/story/story.php?storyId=11945348
    by Greg Allen
     
  7. Jul 15, 2007 #6

    Ivan Seeking

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I don't think any climate scientist claims that any GW model predicts yearly variations. Again and again what I hear is that any particular year or event might be indicative of the effects of GW, but there is no way to know. Only the long term trends can be evaluated with any confidence. Of course, I can't help but wonder if given a bit less energy and moisture content in Katrina, would New Orleans have escaped the wrath of the storm and still be a thriving city? Might we have already lost a city to GW? Could be, but no way to know.

    Yes, it is very complex and certainly in the exclusive domain of the experts.
     
    Last edited: Jul 15, 2007
  8. Jul 21, 2007 #7
    I have also heard this. A 'climate model' is a solution to a bunch of equations, one of which being the Navier-Stokes. The term u.gradu in this equation is not fully understood, this is the term that encompasses turbulence. If we understood turbulence we would better understand this term, hence we would better understand one of the equations which comprise a climate model, and hence we would have better climate models.
     
  9. Jul 22, 2007 #8
    General Circulation Models have improved significantly, but they are far from being able to predict yearly climate changes. But like Ivan's Pachinko ball, we know that as the earth continues to absorb more energy than it emits, which is and has been the case since the 70's, that the average temperature will increase over time.

    GCMs are attempting to predict how climate will react to the added energy, but as is evident from the "predictions" we hear about in the news, there is still far to much uncertainty. Even though it was primitive by today's standards, Hansen's climate model was surprisingly accurate in predicting the cooling from Pinatubo, as well as how much temperatures would rise in 20 years.

    I always find it astonishing, that the processes scientists are most confident of, ie CO2 and other GHGs, are the issues that the skeptics attack as "unfounded." If I were going to dispute the IPCC 4AR, I would point out that the margin for error with aerosols could put the radiative budget in the negative.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?