What’s the (unnerving) similarity between 21st century forecasts for sea level rise and forecasts for world population growth?

They increase with time. And that suggests something is wrong with the science – something that needs to be uncovered.

In hindsight, forecasting 21st century population growth is really easy: just predict business as usual – continued growth in a straight line. Yet somehow, the demographic models structurally think otherwise – and therefore keep revising upwards every couple of years. Smells like sloppy science. Graph from The World Population Prospects – 2015 Revision.

In 2007 the IPCC won a Nobel Peace Prize for a report that truly managed to put climate change on the global political agenda. Depending on emissions scenario this report predicted a sea level rise of between 18 to 59 centimetres by the end of the century – based on models that assume largely linear melting of ice sheets, therefore linear sea level rise.

This year ‘the Hansen study’ passed peer-review. This publication predicts a sea level rise of between 2 and 5 metres over this same century, because it tried to incorporate complicated feedbacks. Indeed – two extremes. Now you might have a tendency to think ‘truth must be in the middle’ (which would be disastrous enough) – but then wait until you’ve watched the below chronology of model-based forecasts for another sustainability problem: the amount of people.

From decimetres to metres, and – oops – another billion people. We as science communicators should ask the right questions. How sloppy are these models?

While we as science communicators replace decimetres with metres (writing about sea level rise forecasts for 2100) – we apparently also have to add another billion people to the year 2050. Because somehow, back in 2002 and all the years after, no one had anticipated the world population would stick at business as usual growth – while it did – and still we are asked to believe that sometime in the near future it will deviate, slowing the growth – while demographic policy is totally non-existent.

We really don’t want to be cynical, but let’s just be honest and say we are starting to lose trust. Trust in the models – and trust in overly optimistic assumptions that are apparently made, structurally, by those that publish world population forecasts.

A short summary of leading world population forecasts for the year 2050…

In 2002 the UN Population Division made a world population growth forecast for the year 2050 – stating (medium variant) we’d be at 8.9 billion people then. Two years later (2004) the UN said – oops – make that 9.1 billion. Then in 2006 they said hang on, 9.2 – only to revise that number upwards to 9.3 in 2010.

And then just two years later (2012) that same UN Population Division made another forecast. The target year stayed the same – yes, 2050 – but now we were to believe world population would by that time have risen to 9.6 billion people.

And just when you won’t believe it anymore last year, in 2015, they again upped their forecast by another 100 million people – to 9.7 billion (medium variant) for again 2050.

The news of the day? It’s no longer 2015, but 2016 – so time for a new forecast. This time the projection does not come from the UN Population Division, but from the ‘Population Reference Bureau’ – an agency that thinks the UN has been overly optimistic still about the rate of the global population growth – and that the world population forecast for 2050 should be increased with another 200 million – to 9.9 billion in 2050 (see report – PDF).

Now… we are almost tempted to make another chatter plot and do a regression – to predict what forecasts will be in 2 years time, 5 years time.

All we know now is that the rate of increase of population growth forecasts happens at exactly the same speed of the actual global population growth (one billion extra in 14 years time) – and that is an underperformance that ‘demographic science’ should really be held accountable for…

Related articles: