Inconsistencies with drought models that don’t account for sea surface temperature changes mean that drought in a climate changed world could be worse than predicted.
WHO: Aiguo Dai, National Center for Atmospheric Research, Boulder, Colorado, USA
WHAT: Looking at the impact of sea surface temperature variability on drought
WHEN: January 2013
WHERE: Nature Climate Change, Vol. 3, No. 1, January 2013
TITLE: Increasing drought under global warming in observations and models (open access)
Climate deniers love it when the models are slightly wrong for predicting future climate changes, and believe me, I’d love it if climate change weren’t so verifiably real and we could all retire and live la dolce vita.
However, that’s not reality, and in the case of this paper, where the model doesn’t quite line up with the observed changes that’s because it’s worse than we previously though. Oh dear.
Global warming since the 1980s has contributed to an 8% increase in drought-ridden areas in the 2000s. It’s led to things like diminished corn crops and the steady draining of underground water aquifers in the USA, much of which is currently experiencing persistent drought. The letter L on the map below stands for long term drought.
What’s that got to do with climate models? Well, while the droughts in Southern Europe or my homeland of Australia are due to lack of rain drying things out, drought can also be from increased evaporation from warmer air temperatures, which the models don’t fully take into account.
These droughts are harder to measure because they’re related to sea surface temperature changes that take decades and can be hard to identify as a human forced signal rather than just natural variations. So this researcher compared sea surface temperatures with drought predictions and observed warming to try and work out what is going on.

Predicted changes in soil moisture globally for 1980–2080 (black dots are where 9 out of 11 models agree on data) (from paper)
There were two areas where the models differed from the observed changes – the Sahel area in Africa and the USA.
In the Sahel, the models predicted there would be warming in the North Atlantic Ocean which would lead to increased rain. What actually happened was that there was large warming in the South Atlantic Ocean compared to the North Atlantic and steady warming over the Indian Ocean which meant less rain, not more. Similarly, for the predicted patterns in the USA, the influence of the Pacific Multidecadal Oscillation was not known to be influenced by human climate forcing. However, it switched to a warm phase from above-normal sea surface temperature.
These sea surface variations that were missed in some of the previous models have some obvious consequences for planning for the slow pressure cooker of stress that drought is on anyone living through it, let alone trying to make a living from agriculture.
The researcher noted that there were also some differences from the models when looking at sulphate aerosols, however for the 21st Century the signal from greenhouse gases will be much stronger than those from aerosols, so shouldn’t mess with the data too much.
So what does this all mean? Well, it means that there are both regional and broader trends for drought in a changed climate. The broad patterns are fairly stable ‘because of the large forced trend compared with natural variations’, which is scientist for humans are making a large enough mess out of this to see the evidence clearly.
The paper ends quite bluntly stating that having re-worked the simulations to take into account the new data for sea surface temperature and other variables, that it’s only more bad news.
It’s likely to be ‘severe drought conditions by the late half of this century over many densely populated areas such as Europe, the eastern USA, southeast Asia and Brazil. This dire prediction could have devastating impacts on a large number of the population if the model’s regional predictions turn out to be true.’
Yes, a researcher actually used the word ‘dire’ in a scientific paper. Oh, and this was with an intermediate emissions scenario, not the business as usual path we’re currently all on. How about we all agree to stop burning carbon now?