Being a resident of the Northern Hemisphere, all I seem to hear about climate change Down Under is that (1) Tony Abbott is the Great Satan for his very reasonable skepticism over climate hysteria, and (2) much of Australia is turning into a dustbowl and a tinderbox due to a lack of precipitation brought on by anthropogenic climate change.

A recent study featured at The Guardian discusses how the "Roaring Forties' [Southern Ocean winds] shift south means more droughts for southern Australia."  The gist of the article is as follows:

Droughts across southern Australia are to continue increasing as the Roaring Forties get stronger and closer to Antarctica, a study has found. It also explains why Antarctica is bucking the global warming trend.

The researchers "found increasing greenhouse gases were strengthening the Southern Ocean's Roaring Forties – known as the Southern Annual Mode (SAM) – that delivered rains to southern Australia."  According to the work, "this resulted in the decreasing rainfall across southern Australia, particularly in Western Australia. 'This isn't good news for farmers in southern Australia who are reliant on winter rains,' said the report, published on Monday in Nature Climate Change."

Thankfully, Australia posts its official climate data online for the public and other interested scientists to investigate.

There is "decreasing rainfall across southern Australia" due to "increasing greenhouse gases"?  Here is the annual rainfall anomaly for southern Australia since the dataset begins in 1900.

Do you see an anthropogenic signature for declining rainfall in southern Australia?  I sure don't.  Actually, there is a modestly significant (p=0.068) trend towards increasing – repeat, increasing – rainfall in southern Australia since 1900.

There aren't any trends whatsoever (p=1.00 – i.e., perfect non-correlation) since 1990, nor are there any significant trends since 1980, 1970, 1960, 1950, 1940, 1930, 1920, or 1910.

In the Murray-Darling river basin of southeastern Australia – which drains one-seventh of the continent's land mass, contains much of the nation's important agricultural area, and supplies three million people with drinking water – there has been an explosion of alarm among the media and scientific communities about supposedly declining precipitation in the large watershed.

Strangely enough, there is – once again – a near statistically significant increase in annual precipitation for the Murray-Darling basin since 1900, and absolutely no evidence of any declining trends if we move back at 10-year intervals to the start of the dataset.  Summer rainfall in the basin has a highly statistically significant increasing trend over the past 114 years, and there are no trends in autumn, winter, or spring.  But yet we read that "now it is feared that the [Murray-Darling] rivers will run dry from a long-term decline in rainfall associated with climate change."  The historical datasets simply do not agree with such alarmist claims in any form.

So why all the hysteria over the Murray-Darling system?  There are periodic water shortages in this basin, but that doesn't appear to be due to any declining precipitation trends from climate change – as there aren't any such declining trends.  Rather, there are probably too many industries and individuals trying to take water from a river in a natural dry region of the planet.  If you stick enough straws into any glass of water, it will eventually go dry.

And about those winter rains in southern Australia that were apparently declining due to climate change?  I don't see any significant trend in winter rainfall for southern Australia since 1900.  Nor is there one since 1990, 1980, 1970...all the way back to the beginning of the 20^th century.

Australia as a whole is getting much, much wetter.  Annual precipitation is increasing at extremely high statistical confidence, as is precipitation during the summer.  No clear trends are evident during the remaining seasons.

Perhaps Australia's precipitation is just getting more extreme?  Nope.  The Australian government's Bureau of Meteorology also maintains a "climate extremes" database.  No significant trends in any of the following precipitation-based extreme climate variables since records begin in 1900: average number of wet days, average number of heavy rain days, average number of very heavy rain days, average wettest day total, average wettest 5-day total, average total rain >95^th percentile (i.e., very wet days), average total rain >99^th percentile (i.e., extremely wet days), average total rain on wet days, average daily intensity, average consecutive dry days, or average consecutive wet days.

There is a very small region of southwestern Australia that has been drying since 1900, as the figure below shows.  But it is a very small area.  Effectively, all of the nation is seeing a 114-year-long trend toward more precipitation, or at the very least, no trend at all.

One caveat with these types of trend maps commonly shown in the media is that they illustrate both statistically significant and non-significant trends, and do not discriminate between them.  In practice, only statistically significant trends should be shown on these types of maps.

Note to NOAA: I'm talking to you in particular, with your classically alarmist and statistically substandard mapping, such as this apparently doubly-subtractively normalized (how is that even possible?) non-statistically significant misleading masterpiece here.  In contrast, there has unequivocally been no significant change in annual temperatures for the United States since 1991.  Claims or suggestions to the contrary are simply false.

Thus, if a trend is non-significant, you can't say it is any different from zero with any reasonable level of confidence.  What is being shown on such maps is the slope of the linear regression, but if the regression is non-significant, then the slope is of little interest.

Also in The Guardian's article, we read this:

The southern region of Western Australia had a 20% decline in rainfall since the 1960s, said Abram.

Sure, there is a significant declining trend in southwestern Australia's annual rainfall since the 1960s, but the trend isn't significant after 1970, nor after 1980, 1990, or 2000.  It is not clear how the lack of a trend during the past four-and-a-half decades is unequivocal evidence of anthropogenic climate change reducing rainfall in this tiny region of the country – and we clearly should not be extrapolating from what appears to be a very small isolated area to all of southern Australia.

As well, there is no significant trend in annual precipitation, or during any of the individual seasons, since 1900 in southeastern Australia.

So is Tony Abbott correct in adopting a very cautious policy line when it comes to the claims of climate alarmism?  Absolutely.  We need to take more time to get the science right and determine what is really going on, and why.