Getting Started

Using climate information can be a daunting prospect when you’re first getting started – even for users with a climate background, navigating the complexities of datasets generated by climate models can be difficult. Here is a brief explainer that may help you determine some priorities for your climate modeling journey!

What are climate models?

Climate models are highly sophisticated numerical simulations of the planet, including processes happening in the atmosphere, ocean, land, and even sea ice. They are sometimes referred to as “general circulation models” or, if they include representation of biological processes, as “Earth system models”. They were created to allow scientists to understand how the climate system works – and in particular, how it will respond to ongoing changes brought about by human activities. Climate models are one of the main tools used to generate future projections of the impacts of climate change, which inform international assessments like the Intergovernmental Panel on Climate Change (IPCC).

Climate models have been around for a few decades now, and their capabilities have increased dramatically as computers have gotten faster and cheaper. They are now able to simulate hundreds of variables like temperature, pressure, and winds, at every location on Earth, over time periods of hundreds of years!

If you would like to know more about what climate models are and how they work, check out “What are Climate Models?” on our climate model overview page!

What is climate model output used for?

Climate model output was originally designed for academic research purposes, and is still widely used by researchers. Since the advent of the IPCC Assessment Reports and the adoption of the Coupled Model Intercomparison Project (CMIP) protocols by most climate modeling centers, the output of climate model simulations have become much more widely used by communities other than ‘traditional’ climate researchers. Some of the more common types of applications include:

• Adaptation planning for climate change impacts
• Risk assessment
• Natural resource planning (e.g. water resources)
• Ecological impacts (habitat suitability, etc.)

If you would like to know more about some common applications where climate model output has been useful across sectors of society, check out “When are Climate Models Useful?” on our climate model overview page!

How much should I trust projections of the future from climate models?

Climate models are reliable tools for estimating future impacts of climate change, which are capable of capturing most of the large-scale aspects of the real world’s climate. However, no tool is perfect, and if you are planning on using climate models to assess future impacts to a particular place, it’s important to understand the caveats and biases associated with these data. 

Some of the more common things to be aware of when working with climate models are:

• Climate models are not weather forecasting models! Since they are initialized from a different starting point than observations, and there is chaos in the climate system, the timing of climate variability will be different in a climate model than the real world.
• There are biases in the spatial patterns of climate as simulated in models, which lead to slight mismatches between models and observations that can start to matter if you are working with a very small spatial area.
  

If you would like to know more about the reliability of climate models and general “pro tips” for using them appropriately, check out our “How Good are Climate Models?” on our climate model overview page!

What types of climate model data are out there, and how can I get it?

Information from climate models is generally organized into data files corresponding to the different variables saved out by the model: think temperature, winds, soil moisture, and many other measurable quantities. Data is available for many different model simulations run in different ways: for applications purposes, usually the most relevant simulations are the ones which try to replicate the historical climate and to project it into the future. (Of course, there are other ways to set up the model that are relevant for other purposes!)

There are lots of Web resources where you can access model data in various ways: you can download climate information to your local computer for analysis; you can use post-processed output to automatically generate selected graphics; and most recently, you can access model output remotely using cloud computing to do your own analyses.

If you would like to know more about places to obtain and/or work with climate model output, check out our resources page!

What technical details should I be aware of when starting to use climate model output?

If you’re interested in doing your own analyses with climate model output, you’ll need to know something about the data files and how they’re organized. The standard format for climate model data is the “Network Common Data Format”, or netCDF, file, which is well suited for large datasets with lots of metadata like climate model output! There are many netCDF files which get produced by all climate model simulations: these are usually organized by creating one file per variable, and sometimes even breaking up variables into multiple files which contain different chunks of time to make the files smaller and easier to manipulate. 

If you would like to know more about the ins and outs of how climate model output is organized, check out our Model Naming Conventions and Climate Model Data Organization pages.

Where can I find examples of how to do my own analyses with climate model output?

There are LOTS of ways to analyze climate model output! The details of what you might want to do are of course going to depend on your application, but some common ways that researchers and environmental managers create analyses using climate models are:

• Making maps of changes in physical variables over time (e.g. amount of warming in a given region)
• Creating regional averages of different variables to allow their temporal behavior to be plotted more easily
• Relating multiple climate variables to one another to learn about the causes for changes in particular places (e.g. linear regression, or correlation)

If you would like to know more about how to take climate model output and use it to create analyses useful for your own purposes, check out our tutorials page!