How not to compare thunderstorm days


This is part of my story – you might see the category: My Story first if you haven’t already (begin at the bottom of the page and work yourself up).

Since my beliefs in Global Warming had crumbled I got the desire to be able to check out some things myself and not to rely on the opinion of others. But what could I do as an interested layman to achieve this goal? The first thing I intended to do was to take a closer look at the messages that appeared in the media, but not to take them on face value as I did before. My impression until then was that the mainstream media tells a very one sided story climate-wise and it could be interesting to discover the neglected data…

This is about the first story I checked. It made an impression back then. We write August 23, 2011. A big thunderstorm crossed Belgium. I experienced it when I was at work. It was surreal: at about 10 AM it gradually got darker and darker, until it was pitch black. It seemed like it was night time. Then came the wind, rain and thunder. Afterwards we learned that it was so dark because the thundercloud was about 16 kilometer high and not much sunlight could come through. A couple days before (August 18) we also had a heavy local thunderstorm which killed 5 music lovers at Pukkelpop festival when a tent collapsed and 140 others needed medical attention.

As expected, the next day I found articles in newspapers that made the connection of these storms with Global Warming. One such story was in the newspaper Het Laatste Nieuws of August 24, 2011. The title over two full pages was: “87,000 lightings and more rain than previous storm” and subtitle (over almost one page) “Warming of the earth increases storms” (translated from Dutch):

Coincidence? Yes and no
Two very severe thunderstorms in less than a week time: could this still be a coincidence? “Yes and no” according to Luc Debontridder. It is a coincidence that these thunderstorms are so severe. But it is certain that there are more thunderstorms than in the past. We reexamined the numbers and noticed that the number of detected thunderstorms in the previous decades did increase: from on average 88 days per year in the 1980s to 94 days now. Be careful: literary every thunder-activity on Belgian area is in this number. A thunderstorm that makes havoc in for example Vlaams-Brabant, but didn’t effect Limburg, will be counted as 1 thunderstorm day. But the increase of 88 to 94 thunderstorm days is irrefutable and according to Debontridder blamed on global warming.
The prominent Belgian climate expert Jean-Pascal van Ypersele confirms this. “When the climate warms by greenhouse gases and the temperature rises, there will be more evaporation from the oceans and seas, that is the way it is. Therefor clouds will be filled with more water vapor and those clouds will turn quicker into thurderstorm clouds. I see those weather extremes increase in the next 30 to 40 years anyway, says van Ypersele.

At first glance this seemed rather balanced: according to both of them, the severity of both thunderstorms was a coincidence, but not the frequency. There doesn’t seem to be strange looking statements. But yet some things caught my attention.

[…] We reexamined the numbers […]
What do they mean by this exactly? Did they simply looked at the numbers again when the last two thunderstorms came over? Or did they reexamined/adjusted/normalized/reconstructed the historical/current data? Whatever, after this “reexamining” they noticed an increase of 6 thunderstorm days per year in comparison with the 1980s.
[…] literary every thunder-activity on Belgian area is in this number […]
It seems the author left a back-door open. It tells us that these numbers alone will not say much. A local storm will have exactly the same weight as a heavy storm over the entire country. There are 6 storm days more per year on average. This could well be small local storms. With such a way of counting this doesn’t necessarily mean a significant change.
[…] from on average 88 days per year in the 1980s to 94 days now […]
The way this is stated reminded me of a statement I read some time ago. Can’t find the reference a the moment, but here is the gist: the mean temperature of the previous decade was compared with the mean temperature of the current decade and the conclusion was that the current average was higher. But when one looked at the current trend it went down. The fact the mean was higher was because the beginning of the period was much warmer and this pulled the mean upwards. So with this in mind this statement caught my attention. It also made me wonder if there was a specific reason why they skipped the 1990s altogether and took the 1980s to compare with?

I went searching what was known about thunderstorms in Belgium. I didn’t find much (what didn’t come as a surprise at all), but found that the climate experts of the KNMI(Royal Meteorologic Institute of the Netherlands) were not so sure about this connection (translated from Dutch):

[…] To what extend climate change effects thunderstorms and lightning is difficult to determine. The data series with observations of the number of thunderstorm days are of variable quality and don’t say anything about the frequency of lightning. […]

That is an interesting statement. “Variable quality”. Probably it means no consistent measurements over time? This made me wonder if this was the same for our country? After some searching, it seems KMI (Royal Metereologic Institute of Belgium) used manual systems (which only partly detected lighting activity) and those were replaced in August 1992 by SAFIR (Système d’Alerte Foudre par Interférométrie Radioélectrique) which can detect lightnings with a high resolution. SAFIR was later extended to BELLS (BElgian Lightning Location System) with an even higher resolution.

For those who don’t yet get it: 1992 is just after their base period of the 1980s. Let me rephrase that: KMI replaced in 1992 a low resolution system with a high resolution system. So in the 1980s they used a system that didn’t detect all thunderstorms. In the 2000s they used a automatic system that detects much more thunderstorms, if not all. How did they manage to compare those two correctly? Maybe they had to adjust the previous measurement data to be able to compare those two? Is that what they meant by “reexamining” the numbers?

It is true that if you count the observed number of thunderstorm days you will find more in this decade compared to the 1980s, there is no doubt about that. But they “forgot” to mention that there are better detection methods now than back then in the 1980s.


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