Lies, damned lies and BoM maps! How Australia’s denial movement can’t read a map

As we mentioned a few weeks back, the joint report put out by the CSIRO and Bureau of Meteorology (“The State of the Climate”) was greeted by most of the denial movement with shock, and as can be expected, with disbelief.

Some such as Andrew Bolt went straight into an apoplectic meltdown and attacked the credibility of the CSRIO itself.

Others where curiously silent, and I did wonder when they would begin their counter attack. Well their assault on reason has begun but it’s more a whimper than a bang. The denial movements efforts can be best described as comical. I hesitate to use the word, but in reality their response shows a  failure to understand basic statistics and an inability to read a map.

Kenskingdom, Anrew Bolt and Jo Nova: trend, what trend!?!

Bolt is Australia’s most prominent denier, Nova operates on the fringes of the debate and the blogger at Kenskingdom is a rough untrained amateur. And yet in their attempt to discredit the report, they they each offer up the same argument. I’d suggest you have a look at each of the authors separate, but identical claims:

Andrew Bolt’s article directly lifts the claim from Jo Nova’s blog, and best paraphrases the separate arguments:

“…JoNova points out yet another trick the CSIRO pulled last week in its deceitful pamphet to whip up warmist panic. It published this Bureau of Meteorology graphic to show how our more populated areas have got much drier since 1960, thanks to (it claims) man-made warming…

So why did the CSIRO choose to show only the rainfall changes from 1960, when the BoM’s records go back many decades earlier?

Answer: perhaps befcause if it showed the rainfall changes from, say, 1900, you’d see that most of Australia has got wetter over the century…”

Two maps are produced, both of which are directly taken from the BoM’s website:

Annual rainfall trends 1900-2010

The above is annual rainfall trends over the last 110 years as recorded by the BoM. It is a statistical average showing precipitation over this time. The deniers then contrast this map with the map used in the the CSIRO report and are “shocked” by the difference:

Annual rainfall trends 1960-2010

The conclusion of all three is that it must be a “trick” and that the CSIRO and BoM are trying to pull a fast one. All I can say is “Oh dear guys, you really can’t even read a map can you?”

Clearly Bolt thinks the map showing the distribution for rainfall over the 1900-2010 period demonstrates Australia is getting wetter everywhere: in otherwords an upward trend in rainfall distribution across the continent. What Andrew fails to grasp is that this map shows a statistical average over the 110 year period. In order to observe how the trend varies we need to compare it to other data sets. BoM does just that.

Failing to understand trends: about the BoM trend maps

The BoM maps need to be viewed as a time line. Each map provides a statistical average over a period of time. As stated the map for 1900-2010 is the average over 110 years.

But how do you demonstrate any changes in rainfall patterns over time? You shorten the time periods incrementally and then calculate your rainfall averages. The BoM does so by reducing the average time period in ten year increments  and then calculates the average rain fall patterns. This is how to look for long term trends.

This is not a “trick”, it’s basic statistics. As the BoM states on their website:

The trend maps are a useful way to compare how the rainfall has changed in different regions of Australia over time.

Trend maps: misrepresenting the BoM’s data

One of the basic premises of climate change is how we have observed statistically valid changes in climate over time. The deniers can’t seem to grasp this basic concept. Let’s look at how Kenskingdom reads the maps:

Now this is technically correct, but it is also less than honest.  Australia has rainfall records going back in some places to the 1870s.  BOM’s charts and graphs are reliable apparently from 1900.  Why not show these?  Because they do NOT support the implication that Global Warming is causing much of Australia to become drier.

Sorry Kenskingdom, as I’ve stated the map is a static representation of data averaged over 110 years. The point of trend maps is to compare different data sets over time. This why BoM gives us to different trend maps covering different time periods. This is why you need to look at all them, in isolation they tell us very little about the trends themselves.

Let’s run the same maps, reducing the period in which you calculate your average rainfall  in ten year increments. When we do that we see a very a clear trend in changing rainfall patterns:

Trends in total rainfall 1900-2009

The above is the map that the deniers thinks somehow compromises CSIRO/BoM statements on changing rainfall. OK, for the last time this is a statistical average over a 110 year period. Now, lets slightly shorten that period by ten years as BoM and see if there are any emergent trends:

Trends in total rainfall 1910-2009

And again, let’s shorten that period by another ten years…

1920-2010

…and another ten years…

1930-2010

…and yes, that’s right, another ten years…

1940-2010

…keep going…

1950-2010

…just a bit more…

1960-2010

…are we there yet? No a bit more…

1970-2010

And there you go!

That’s a clearly recognisable trend. In simple terms there has been change in rainfall patterns. This is exactly what the science of climate change states. See all that brown stuff? As time has progressed parts of Australia are getting drier and receiving less rain. Some parts are now recieving more rain. This accords with legitimate climate science.  Nothing to see at all guys – right?

Back to denying reality then…

Logical fallacies and cherry picking data: the deniers old standbys

Had Nova, Bolt or Kenskingdom actually spent time on the BoM website where these reports can be freely generated by anyone, they would have seen how the long term trends are clearly demonstrated.

Instead they chose to look at very isolated maps: this is simply cherry picking data. I’ve reproduced the entire sequence of maps in order to show how they’ve used only a small selection of the data available.

Indeed, I’d strongly suggest all three of them fell for the cognitive bias of hunting for anomalies: looking for facts or data that somehow support their predetermined opinion.

4 thoughts on “Lies, damned lies and BoM maps! How Australia’s denial movement can’t read a map

  1. […] Lies, damned lies and statistics: how Australia's denial movement … […]

  2. substanti8 says:

    Thanks for debunking more denier disinformation.

    You might want to fix the sentence below the second map. It should be as follows:

    “In order to an observe how the trend varies we need to compare it to other data sets.”

  3. […] of their conspiracy laden world view and how their basic scientific illiteracy: Andrew Bolt can’t read a map! […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com 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

%d bloggers like this: