Climate change predictions

Climate change predictions abound, they are everywhere. And we seem to use every slightly unusual weather event to justify them.

It will get hotter, wetter, drier, more stormy and sea levels will rise.

Ice caps will melt, ice sheets will break apart and polar bears will go extinct.

Elsewhere huge cyclones will strike at the coasts.

These are examples of predictions for global warming and they are educated guesses as to the shape of the future. For this is what a climate change prediction is, an estimate of what will happen to climate in the future.

First some thoughts about predictions

Toss a coin and you can predict with certainty that the coin will fall as either heads or tails. 

A miracle could happen and the coin would land and stick on its edge but most times it falls one side or the other.

As a normal coin has no bias we can safely predict that it will land 'heads' side up 50% of the time.

Toss the coin enough times and very close to half the total number of tosses will be 'heads'.

The next song to play on your iPod when it is set to random play will be...?

Well, near impossible to guess if you have a few hundred tracks stored, even if you know how many tracks and you have the list in front of you.

There is a precise probability for any given track that is based on this total number of tracks and the assumption that the iTunes randomness algorithm is truly random (although I am sure you can smell the bias for Supertramp songs).

Even if we calculate this prediction, all probability can do is tell us is the chance of a given track being played, not which track will come next.

Some predictions are a little easier.

The next party to be elected to government in a democracy is not quite as uncertain. In most cases it will be a two party race. So, with the help of some political understanding, information from opinion polls and some attention to current affairs, the prediction should be more precise.

It still has some uncertainty, but is far less than the 50% of the coin toss.

Whatever the situation, predictions are are calculation of chance (measured as a probability) and are only as reliable as the understanding we have of

1. the number of potential outcomes
2. the processes that influence those outcomes
3. the reliability of the probabilities we compute from knowledge of outcomes and pocesses

Predictions in climate change

When it comes to climate change predictions

1. the number of potential outcomes is very large
2. even though there is arguably one primary driver of climate change (the net energy retained within the atmosphere) the number of processes that are triggered by this energy balance are many, consequently
3. computation of probabilities is hugely complex

This means that we are dealing with a great deal of uncertainty. Exactly how much depends on who you talk to and their understanding of climate science.

There are two main approaches taken to climate change predictions

1. Analyse trends in measurements of climate [e.g. sea surface temperatures] and project the trend into the future
2. Input observations into models of how the atmosphere and oceans work, run the models for different scenarios of energy balance, and then use the model outputs as predictions.

The first is rather crude when it comes to climate because we don't know if the key drivers of climate change such as the concentration of greenhouse gases in the atmosphere will deliver proportional effects.

If there is a threshold concentration that cause the rules to shift or new processes are triggered then the trend prediction will be out, possibly by a long way.

The second approach of process modelling is the preferred option and the one adopted by the Intergovernmental Panel on Climate Change [IPCC

Why do we need predictions?

We know that the world is changing.

Even without climate issues, 7 billion humans have and will continue to change things.

And we also know that a response to climate change is prudent. There will have to be action taken because climate change effects will be significant.

Only what action makes the most sense?

We can choose to

• cope with the change
• adapt to it
• try to prevent it

So far we have not embraced the magnitude of this choice.

Cope means we let the change happen and press on. After the flood we clean up and return to normal until the next flood.

Adapt means that we plan to build on higher ground or put new buildings on stilts to protect the upper floors. We still experience floods but we are better prepared for their effects.

Prevention means we either try to stop it happening or deflect any effects. We build a flood barrier around every town.

What climate change predictions should do is help guide the choice of response because — if they are good enough — they can tell us where, when and how severe the effects of climate change will be.

Most importantly they will help understand how effective our interventions will be. It makes no sense to invest heavily in actions that are not needed or are unlikely to be effective.

Will the climate change predictions come true?

The likelihood of real climate change effects is strong enough for us to pay attention.

Predictions tell us that it will become harder to grow food, maintain water supplies and avoid the vagaries of severe weather.

We need to pay more attention to the choice of response.

Each option has costs, risk and benefit. It is the classic balancing act. Do nothing and the risk and cost rise with benefit only if we get lucky and nothing happens.

Do too much and the costs outweigh the benefits even if the worst of the predictions comes to pass.

Balance is everything and can only be found when we put our minds and or feelings into it.

Ostriches we cannot be.

Back to the top of Climate change predictions|Back to What is climate change? | Back to climate-change-wisdom home page