CLIMATE MODELS are based on physics and run on supercomputers. They are indispensable for understanding climate, they have been tested, and they work.

• Model INPUTS
  

• Model TESTS from temperature data
  

• Model FORECASTING and scenarios
  

• Where can I GET MORE INFORMATION about models?
  

• How about a VIDEO on modeling?

• What about responses to FAQ on models?

MODEL INPUTS INCLUDE ALL EFFECTS

These models use natural (e.g. solar) and anthropogenic inputs. Here is a graphic for the inputs to the climate models. Note that both natural (including solar) and anthropogenic effects are included. 

Below is a graph of the time dependence of the various forcings relative to 1880, used as input to  NASA/GISS models:

Go to top.

--------------------------

MODEL TESTS FROM TEMPERATURE DATA

The temperature is an output of the models, and is not the result of a "fit" to parameters. As shown in the graphics below, the climate models have been successfully backtested against the last 100 years of temperature data (black lines) on a global level and at a continental level. The output of the models including all effects (anthropogenic/human and natural) are shown in red, which follow the data including recent global warming in the last 30 years. Natural effects only, in blue, are NOT sufficient to produce agreement with the data, especially in the last 30 years. Model uncertainties are exhibited by the bands. For all the supporting details, see the report.

 Go to top.

--------------------------

MODEL FORECASTING

Below is the graphic for the model forecasts for the average global temperature over the next 100 years. Forecasts require scenarios. A scenario represents a set of assumptions of human behavior in the future. All the forecasts predict increasing global temperature trends, depending on the scenario: rapidly increasing temperatures if "business as usual" (A2 red), less rapidly increasing temperatures if action is taken against global warming (B1 blue, A1B green). For each scenario, there are uncertainties in model behaviors, shown in the vertical "error bars" (gray) on the right. The baseline (orange) is irrelevant since it assumes no increase in greenhouse gases.

Note that the graph is in Centigrade (one degree C = 1.8 degrees F).

 Go to top.

--------------------------

WHERE CAN I GET MORE INFORMATION ON MODELS?

A thorough description of climate models, with references, can be found in the 2007 IPCC Science Report. For specific global models see, e.g. the Princeton GFDL/NOAA site and the GISS/NASA site (with the computer code here). HERE is an article showing the steady improvement of climate models in comparison with climate data. Here is an interesting VIDEO of recent development of modeling using desk-sized supercomputers.

IPCC Science Report Download Here: Climate Change 2007 - The Physical Science Basis: Working Group I Contribution to the Fourth Assessment Report of the IPCC.

 Go to top.

--------------------------

HERE is a great VIDEO describing the history of climate modeling:

 Go to top.

--------------------------

RESPONSES TO FAQ from RealClimate

RealClimate, the approachable scientific climate website run by professional climatologists, has responses to common Frequently Asked Questions FAQ about climate models.

FAQ Part I

What is the difference between a physics-based model and a statistical model?

Are climate models just a fit to the trend in the global temperature data? (Answer = NO)

Why are there ‘wiggles’ in the output?

What is robust in a climate projection and how can I tell?

How have models changed over the years?

What is tuning?

How are models evaluated?

Are the models complete? That is, do they contain all the processes we know about?

Do models have global warming built in?  (Answer = NO)

How do I write a paper that proves that models are wrong?

Can GCMs predict the temperature and precipitation for my home?

Can I use a climate model myself? (Answer = YES)

FAQ Part II

What are parameterisations?

How are the parameterisations evaluated?

Are clouds included in models? How are they parameterised?

What is being done to address the considerable uncertainty associated with cloud and aerosol forcings?

Do models assume a constant relative humidity? (Answer = NO)

What are boundary conditions?

Does the climate change if the boundary conditions are stable?

Does the climate change if boundary conditions change?

What is a forcing then?

What are the differences between climate models and weather models?

How are solar variations represented in the models?

What do you mean when you say a model has “skill”?

How much can we learn from paleoclimate?

Go to top.

--------------------------