For teaching purposes I am looking for an atmospheric model that can be easily installed and the input data can be easily read and prepared. Preferably I would like to run a global low resolution simulation of ~5 years. Naturally, I know this model will not have high forecasting accuracy. Is there a model that suits this criteria?

Thanks for the help!

  • $\begingroup$ Do you have any constraints more concrete than "easily" ? A simulation that's stable over years does not exist. 10 days is the max for a 50% accuracy for most meteorological models, even the latest high res ones. Or are you looking for climate models ? en.wikipedia.org/wiki/Climate_model Atmosphere alone is not enough. There's a limit to simplification before it becomes unrealistic. $\endgroup$ – user20217 Apr 5 '20 at 0:31
  • $\begingroup$ Thanks for your help! what I intend to do is manipulate the input, so I will naturally run those several years in (probably) 5 day cycles. For the duration of several years. Thats why I also wrote "input data can be easily read and prepared." if someone already did interpolations to model grid it will be the perfect thing. Thanks again! Yair $\endgroup$ – yair suari Apr 5 '20 at 16:28
  • $\begingroup$ Should it be a global or regional model? It should a rather a climate model? Does the model mentioned in this answer help you? $\endgroup$ – daniel.heydebreck Apr 11 '20 at 13:25
  • $\begingroup$ Thanks for that. The planet simulator is also a climate model but I am looking for more of a short term forecast. I think I will start with WRF Portal $\endgroup$ – yair suari Apr 12 '20 at 12:29

I think you are asking a question with a variety of different constraints. I'll tackle a couple of them.

What is the simplest atmospheric model to operate?

That would be the Zero-dimensional energy balance model. It has almost zero resolution and no temporal capacity.

What is an atmospheric model that can be easily installed and run?

The Weather Research and Forecasting (WRF) model is a very flexible, limited area model. This model has successfully been run on what amounts to a child's toy.

What is a global atmospheric model that can be run for a long period of time, but can easily be run?

While I have never run it myself, it sounds like you may be looking for SPEEDY. I have not heard it is the most numerically stable model, and I cannot attest to how easy it is to run. From the documentation, it seems like it can be run one year at a time, but I was once told that it is typically stable in the range of months or even days.

Atmospheric models are not like a game you can just install and run like MS Word or a game. With almost any atmospheric model, a knowledge of programming in a Linux/Unix framework is paramount.

  • 1
    $\begingroup$ This was very helpfull! I am currenty testing with WRF, apparently when using WRFPortal this is managable. I will also try SPEEDY $\endgroup$ – yair suari Apr 12 '20 at 12:26

Weather forecasting (meteorology) is not climate modelling. As such, a meteorological model to forecast temperatures over years is not the right tool for the job.

But there is indeed a very, very simple climate "model" that only accounts for CO2 and temperature.

You find it here:


with a little guide:


and more info here:


The question was "simple", and this is simple. It only shows a single dependency but may be useful to demonstrate a relationship. Of course "real" quantitative modelling depends on ridiculously much more and is constantly improved and updated.

  • $\begingroup$ @Fred, how did you do the subscript 2 ? $\endgroup$ – user20217 Apr 11 '20 at 22:35
  • $\begingroup$ For subscript 2 in CO2 do this, CO<sub>2</sub>, for superscript 2 for m2 (meters squared) do this, m<sup>2</sup>. $\endgroup$ – Fred Apr 12 '20 at 5:27
  • $\begingroup$ Thanks for your help. Actually, I dont intend to run a climate model, but rather a weather forecasting model. Yair $\endgroup$ – yair suari Apr 12 '20 at 12:27
  • 1
    $\begingroup$ @yairsuari Do as you please, but you cannot run a meteorological model over years. Well, you can, but it is not made for your case and it will not be stable for more than a week or two. I suggest that, in order to avoid spreading incorrect information, you do not rely on the information presented here and instead contact your countrie's meteorologal service or other officials and ask for teaching material. $\endgroup$ – user20217 Apr 12 '20 at 13:14
  • $\begingroup$ Sorry I was not clear. The intention is to rerun it. Meaning to creat new initialization and boundary conditions once every w4h. Like the way a forecast model would run and just change the surface input. I think this should work. $\endgroup$ – yair suari Apr 13 '20 at 19:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.