Predicting Extreme Heat

WHO: Brigitte Mueller and Sonia I. Seneviratne (Institute for Atmospheric and Climate Science, Eidgenössiche Technische Hochschule (ETH) Zurich, Switzerland)

WHAT:  Seeing if you can predict extreme heat from a lack of moisture in the ground.

WHEN: July 16 2012

WHERE: Proceedings of the National Academy of Sciences of the United States of America

TITLE: Hot days induced by precipitation deficits at the global scale (subs required)

Using data correlations to try and predict extreme heat – useful for drought and disaster planning

I’m sure many of you rush out each week to get the newest copy of the Proceedings of the National Academy of Sciences when it hits the stands (yeah, I don’t either) but this paper came to my attention as something a bit interesting.

The researchers were hoping that there might be a relationship between how dry the soil is in the lead up to heatwaves and the extreme heat that follows. If extreme heat can be better predicted, then it’s easier for local or state governments and emergency services to prepare for it.

To measure lack of rain, they used the Standardised Precipitation Index (SPI) which is a measure of drought that only looks at rainfall (so doesn’t count water supply levels, water demand or runoff losses). The index works on a number system where -2 is exceptionally dry, zero is normal amounts of rain, and +2 is exceptionally damp. This paper looked at soil moisture across the globe, but as an example, here’s the SPI map for the USA this May/June when they started their severe heatwave in several parts of the country.

The paper found that there is a relationship, where lack of surface moisture is generally followed by extreme heat a few weeks later. Keep in mind though, the favourite saying of statistics people around the world: correlation does not mean causation. So while there may be a connection between dry soil and coming heatwaves, that doesn’t mean that dry soil causes heat waves of course.

Interestingly, the researchers found that the relationship between lack of surface moisture and extreme heat was asymmetrical – the correlation was stronger for really extreme heat and weaker for average heat. Which is good because it means this method can help predict the really horrendously hot days better than the averagely uncomfortable hot days.

Predicting the days when it’s so hot even the Koalas will ask for a drink (Velovotee, Flickr)

The data used was SPI data for 3, 6 and 9-months between 1979-2010, and a ‘hot day’ was defined as a temperature that was warmer than 90% of all the other days. The strongest correlation between lack of soil moisture and extreme heat was found in North and South America, Europe, Australia and China. This means that in those areas, localised research could be done to look at how to best predict the potential for extreme heat before it occurs, which will allow for emergency services to prepare for things like more hospitalisations from heat stroke or bushfire conditions, and give local governments time to make public service announcements to try and stay indoors and hydrated, etc.

What’s this got to do with climate change? Well, currently for the North American summer of 2012, the heat records are beating the cold temperature records by 10 -1. This is what climate scientists are talking about when they say carbon pollution is loading the dice for more extremes – normally there should be one record high for every record low. So if extreme heat is going to become more common in some places as we experience greater levels of climate change from continuing carbon pollution being poured into the atmosphere, being able to plan for it is a really good idea.

1 thought on “Predicting Extreme Heat

  1. Pingback: Predicting Extreme Heat | Amy Huva

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 )

Connecting to %s