Climate Change Predictions and Mathematical Decay

Happy Sunday! This may be a short post–not sure yet.

Forecasters, pundits and scientists have been happily engaged in predicting the state of our climate in 2100. Century timescales seem appropriate for discussing something that can’t even be accurately measured on less than 30-year timescales, so I think that’s appropriate.

The IPCC thinks that sea level rise will fall somewhere between 26 and 98 centimeters by2100. That’s a very wide range, which should protect their sense of self worth somewhat. It’s one thing to predict the Seattle Seahawks will win the next Super Bowl. It’s a bit different to predict that the next Super Bowl champion will come from the Western Division of the NFC.

Similar predictions abound for temperature rise, atmospheric sensitivity, frequency and intensity of storms, drought and flood, the number of climate refugees, etc.

But it’s no longer a century timescale. We are in 2015. No matter how precise the calculations and how liberal the range used, the fact that we are in 2015 should be affecting those predictions now. Honest scientists should be publishing new versions of the same forecasts based on the actual behavior of the system being studied. Even Las Vegas changes the odds on a football game right up to game time.

If a forecast was published first in 2000 saying that a meter of sea level rise was probably by 2100, it is completely legitimate to note that 15% of the time frame involved has passed with about 1% of the expected sea level occurring to date. As sea level is not expected to change in a linear fashion it would not even harm the forecaster’s reputation, assuming s/he phrased it properly.

That is called ‘mathematical decay.’ If you say something has a 85% chance of happening in the next five years and it doesn’t happen in the first year, if you’re smart you’ll recalculate the odds accordingly.

The same should be true for predictions about temperature, storms, etc. One would assume that the best scientists would incorporate the information gathered in the past 15 years to adjust the elements driving their prediction. The worst would just adjust the timeline, saying disaster would strike in 2115, rather than 2100.

But there should be changes. I’m writing this short essay because I haven’t seen studies that say ‘We predicted this in 2000. Based on what has happened since, we have adjusted our model in such and such a way and our new prediction in 2015 is this.’

The IPCC AR5 has attempted to use current scientific work in preparing their Assessment Report. But as it comes from many different models and voluminous work from many scientists, it doesn’t really capture changes effectively.

Perhaps I suffer from tunnel vision and haven’t seen glaring examples of this kind of adjustment. In which case I hope readers will show me where I’m wrong.

What’s important is to understand that if these changes are not occurring, is it because of some inclination to not give skeptics ammunition, internal pressure to be consistent, group pressure to conform or something else? I can’t think of real world legitimate reasons not to recalculate, but again that might just be tunnel vision on my part.

But this is the type of exercise that could be undertaken painlessly, without much in the way of embarrassment, and it could serve to give policy makers better information and the public more in the way of understanding.

We all know predicting stuff is hard. Smart predictors take advantage of new information to change their story. As for not-so-smart predictors…


2 responses to “Climate Change Predictions and Mathematical Decay

  1. So, in short, one might be inclined to say …

    Their predilection for prediction
    Yields nothing more than science fiction!

  2. Tom,
    The true believers don’t readjust to facts or new information. They just shout louder and manipulate their pet politicians and tame journalists to move more quickly and write more hysterically.
    By the way, I love the concept of “mathematical decay”, but don’t find a definition exactly like yours. Similar by implication, but not precise. Where did you get this idea? I would like to work with it more.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s