Predictions are the essence of science. For a theory or a model to remain credible it has to predict something and that prediction must be compared to reality. Every scientist knows the remark that ‘many a beautiful theory was killed by an ugly fact.’ A particular theory can make a prediction about something that has already happened, or it can “hindcast” to see if it agrees with a preexisting dataset. To my mind though the most powerful and convincing proof of a theory or model is what it says about the future.
That is why it is good to have predictions about how the world’s various climate parameters might behave in the next few years. The UK Met Office has just released a 5-year forecast that is particularly interesting. The Met Office describes these 5-years as “near-term” climate predictions. Technically climate is a 30-year average, but if 5 years is useful period then there can be no more criticism that those who look at the period of the so-called warming pause – the last 15 years or so – are using too short a period to discern anything meaningful. Previously forecasts have been decadal, but the reduction to five years was, it seems, due to computational expense.
The Met Office predictions rest on the extraordinary year 2015 which was the warmest on record by a statistically significant margin, as opposed to changes seen in previous years which were not. As an aside I find it interesting that the year which is said to have the greatest anthropogenic contribution to warming is the same year that witnessed a record-breaking El Nino and an unprecedented “Pacific Blob.” It would be fair to say that some in the scientific community have gotten themselves in a bit of a mess about the contribution of El Nino and the “Pacific Blob” to the global annual temperature of 2015.
The Met Office says that 2016 will be warmer than 2015 because of the El Nino. 2017 will be cooler because of the La Nina and 2018-20 will see three years of warming. But there are the usual caveats, some of which dent considerably the value of such a prediction. For example Met Office scientist Doug Smith said, “We cannot say exactly how warm it will get but there is no doubt the overall upward trend of temperatures will continue.” This sounds contradictory to me. He adds: “Whether one of these years – 2018,2019 2020 – overtakes 2016 in terms of temperature is very hard to predict at this stage, we are looking quite far into the future, after all.”
Looking back at the 1998 El Nino, the subsequent La Nina resulted in 1999 and 2000 being significantly cooler so cooling for 2017 and 2018 cannot be ruled out. 2019 and 2020 may be warmer but two annual datapoints will not in themselves establish a warming trend. If the 2015 El Nino turns out to be like the 1998 one then the Met Office predictions will fail.
The Met Office’s forecast says that each year should fall (95% confidence) between 0.28°C – 0.77°C (above 1981-2010 average.) That is a very wide margin – 0.49°C – a little larger than any increase seen during any five year period in the instrumental record. It’s a rather large net.
The forecast is shown on Fig 1 (click on image to enlarge.) There are several points to note. The red are predictions and it can be seen that they have limited skill. In the 1980s the predictions were too cold and in the 2000s they were too hot. The blue section is the predicted course of the temperature trajectory, which is essentially the same shape as the period 2010 – 2015. I must point out the absence of errors for the temperature – a serious omission.
The Met Office says that “ten year global average warming rates are likely to return to late 20th century levels within the next two years.” I seriously doubt that. It would be unwise to use an El Nino-enhanced year as an end-point in temperature trend analysis. The Met Office also said there was a “pause”, but that it’s not a big deal in the long-term. The problem is that the “long-term” isn’t here yet.
As I have said these are valuable predictions, but nature has a way of humbling even the best “state-of-the-art” predictions as the Met Office’s Vicky Pope discovered after her 2007 forecast for the next seven years. It was wrong by a country mile.