What is a computer model anyway?

Ever made an Excel spreadsheet, or know someone who has? Most people reading this would have. Lotus 123 (an earlier spreadsheet software program) became popular in the Eighties because it gave people with a little bit of computer expertise a simple way to do modelling.

• Work out the household budget and see what effects a change in rent/ pay/ school fees has.


• Work out the the effect on the household budget if interest rates are x% compared to y%.


• Work out the Sales Department's net profit if sales of widget A are 100 units compared to 500 compared to 1000, including the different marketing costs to achieve each level of sales.

These are all examples of very simple computer models. What ALL computer models have in common is that some factors stay the same and some factors can be changed to see "what if".

Presumably the global warming/ climate change computer models are much more complex. Regardless of the complexity, the value of some factors are built in and fixed, and some are changed to see "what if".

The computer model, simple or ultra complex, is only as good as the value of the factors that are used and the accuracy of the formula for predicting the effect of the changed value.

Therefore computer models for climate change effects are only as good as the numbers and the equations that are put into the model. They are not magic or superior to human judgement. They are only as good as the research and open mindedness of the person who creates the model. And computer models only show "what if". The result depends on the "if" factor.

There is a very technical computer term which applies here - GIGO. This stands for Garbage In, Garbage Out.