Copyright © 2016 MIT Joint Program on the Science and Policy of Global Change.
Welcome to the Web Climate Lab!
This application is built around a simple climate model,
which demonstrates the process used by more sophisticated
models to help us understand climate change. Click “Start” for
an introduction to the model, or click “Skip”
to start using it immediately.
Choose a policy scenario
First, choose which climate policy you think the world will
Choose your model parameters
Move these two sliders to change how your model simulates the
Run your model
Now, click “Run Model” to see your results!
Your simple model projects that by 2100, the Earth's average
surface temperature will increase
Now, let's compare your result with results from state of the
Comparison with other models
Your simple model's projection for average temperatures between 2081
and 2100 is
We should also
see how well your model fits past observations.
Comparison with past observations
These graphs show four different measurements of past
climate. If the red line is almost always in the
blue range on each graph, your simple model matches past observations
well, so it's also likely to project future temperatures well.
If it doesn't, your choice of parameters might have led to
an inaccurate model!
You now have access to all the features of this application.
Try varying your policy selections to understand
the implications of each decision!
The options listed here correspond to the Representative
Concentration Pathways, or RCPs, created by the Intergovernmental
Panel on Climate Change. Rather than representing specific policies,
these pathways describe possibilities for atmospheric greenhouse
gas concentrations, which are affected by both policy and natural
variation. We have summarized them in terms of policy for convenience.
The RCPs are a worldwide standard for concentration scenarios,
and therefore enable comparisons between many different models.
For more information, see
To simulate the behavior of the climate, models use equations
derived from observations and experiments. Some of these equations
include numbers, called “parameters,” which scientists can't
measure exactly. Instead, they estimate the likelihood of each
possible value, producing the green curve shown on the graphs
If you change the sliders below, your model's results will change.
However, not all of the results you can produce are reasonable,
as you can see by comparing your model's results to past
observations and the results of other models. In particular,
using two “very unlikely” values can produce wildly improbable
The green bar shows the 5th to 95th percentile range of Coupled
Model Intercomparison Project phase 5 (CMIP5) runs for the
selected scenario. CMIP5 includes 42 models from research groups
around the world, all of which are atmosphere-ocean general
circulation models, the most sophisticated type available. Its
results were used for the Intergovernmental Panel on Climate
Change's most recent report.