Weather, climate and land-use

This is a story about land-use and weather, or perhaps climate. We don’t know which, because the rainfall data only start with European occupation. In 1850 Thomas Elder sent a small ship from Adelaide up to the end of Spencer Gulf, its trip the beginning of the town of Port Augusta. The reports were excellent — lots of open grassy land, abundant fresh water in lakes and streams. Elder himself sent men and cattle north, and farmers arrived to till the soil. The winter rainfall continued predictably for a decade and more. Wheat was sent to Adelaide, and before long it was being sent to England.

Those with money took up large leases. One Hugh Proby, a young men whose father was a lord, took up what would later be known as Kanyaka Station, south of today’s township of Hawker. Not long afterwards, he was drowned in Willochra Creek attempting to get his cattle to higher ground during a flash flood. The lease was taken up by others, notably John Phillips, who decided that sheep were a better bet than cattle, and built a substantial homestead area at Kanyaka.

At its peak Kanyaka was virtually a township itself. Phillips had more than 70 families working there. There was abundant stone, and little usable native wood, so everything was made out of stone. The woolshed remains show the skill of the stonemasons he employed. The homestead had sixteen rooms, and there were dozens of other structures, the remans of which you can see today. His workers built around 40 kilometres of dry-stone walls. It was a big place.

Kanyaka was set up to be virtually independent with its own vegetable gardens, dairy, and piggery, while it had the last post office in the north of South Australia, with tens of thousands of postal articles in its best year, 1864. Because of its healthy level of supplies Kanyaka was also the departure point for a number of exploratory expeditions to the north.

In 1864 shearers shore 41,000 sheep. Three years later, after continual and unprecedented drought, only half of the sheep were left. Eventually, everyone left, and all you can see today are the ruins, in stone. It is one of the saddest places I have ever visited in Australia, and I have been there five times.

The drought affected a large part of occupied South Australia, and George Goyder,  the Surveyor-General was sent to determine where it was sensible to grow wheat, and where you simply shouldn’t. His Goyder line, from the 1860s, has proved to be a pretty good marker, even today. But the development of the railway line to Alice Springs (it got as far as Oodnadatta) encouraged more people to believe that once the railway was there the good seasons must return, and wheat farming was tried again and again above Goyder’s line. It never lasted. There was another serious drought in the 1880s.

By 1882 the railway had reached an outpost called Government Gums, later renamed Farina (Latin for ‘flour’), more than 100 km to the north of Kanyaka. Before long there were 100 people working there, and the occasional wet winter produced confidence that the area would become a vast agricultural region. No way. Goyder was right. All you can see in Farina today are more stone ruins. That part of South Australia is littered with them.

Until the 1864-67 drought there was a delusion that ‘rain follows the plough’, an early example of the hubris that sometimes besets people who think that humanity has the capacity and the power to fundamentally affect climate. Mind you, when you see the effect that winter rain has on wheat-growing in South Australia and Victoria, which I have witnessed in the past couple of weeks, you can almost excuse the farmers of 150 years ago. ‘It happened last year at exactly this time,’ they would say to one another, ‘and look, it’s happened again.’

Three years of drought ruined Kanyaka, though Phillips did come back one more time before drought drove him out forever. The little township surveyed five miles north of Kanyaka homestead, and the two-storey pub — they’ve all gone. Humans need abundant water if their life is to be endurable and productive, especially if they’re running sheep or cattle, or growing grain.

I spent ten or so years of my life studying farmers and their politics, and that took me around some of inland Australia. Again and again I would come across examples of over-reliance on rain. Some people do learn. A grazier near Canberra used his ‘winnings’ from the wool boom in the early 1950s to build the biggest dam he could scoop out. He’s never been troubled by drought since.

Weather can seem to come in cycles. Major droughts have occurred in southeastern Australia at the time of Federation, during and after the Second World War, and more recently around the Millennium. Rainfall can come in waves too. South Australia has no major streams that flow into the sea, other than the Murray, so floods there refer to the Murray. Further upstream, so to speak, there have been 21 major floods at Gundagai in the past 150 years. This marker shows their relative heights.


We have annual averages of rainfall, but it’s a rare place that has average rainfall every year, let alone the month’s average very month. Temperature? There’s a good deal of evidence that some of the hottest years experienced in Australia occurred in the 1890s, and the notion of an Australian ‘average’ is about as useful as that of a ‘global average’. We haven’t been here long enough to have good data that tell us much about long-term patterns in weather.

So what caused  the good weather and then the drought in central South Australia in the mid 18th century? We don’t know. We do know that some of the water that people found had come down Cooper Creek from Queensland, because Goyder found Lake Blanche full of fresh water, and that big lake is fed by the Cooper. But there must have been some good local rain as well, and for about fifteen years. To the best of my knowledge there has never been another such high-rainfall period in that area since.

What we also know is that droughts, fires and floods are part of our past, present and future. We need to prepare for them. We should not be building homes (or anything much) on flood-plains, or in the midst of eucalyptus forests. We should be conserving water wherever we can, and we should be thinking hard about how we will supply water to our every-growing cities. I know, I’ve said all this before.

None of it is easy. People don’t like being told what to do. There will be huge protests if any government proposes to build another water-supply dam. Many of those who lost their homes in forest fires want to build there again. I think they’re mad, but people are individuals who have their own preferences, and their own sense of risk.

But visit Kanyaka if you want to experience the sadness of the hard work and defeat that has accompanied so much of the pioneering of inland Australia, much of it because of a lack of knowledge of rainfall. We know much more now, and do not have to repeat the mistakes of the past.



Join the discussion 22 Comments

  • PeterE says:

    Very interesting, particularly as I was born at Port Augusta and lived at various times at Iron Knob and Whyalla. The Goyder line is well known in SA and I once did some work on the early railways od SA. Still, just this week I heard an authority discussing the ‘new developments’ in the ‘unprecedented’ climate changes we have recently seen: for example, he pointed to droughts followed suddenly by flooding rains, to wildfires never seen before and so on. The ignorance of our landscape history is stunning. Many people seem to have a reference range of only about twenty years.

    • Don Aitkin says:

      My reference for temperatures is, which I have recommended before. It uses all the data sets, and offers both mapping and graphing displays. For july, and using GISS data, the report goes as follows:

      ‘July 2015 global surface air temperatures

      General: The average global air temperature was
      close to the average for the last ten years.

      The Northern Hemisphere was characterised by
      regional air temperature contrasts, especially in the
      Northern Hemisphere, as usual. Much of North
      America, the North Atlantic and Europe with
      western Russia had below average temperatures.
      Siberia and Alaska had had temperatures near or
      above average. Most of the Arctic had below
      average temperatures.

      Near the Equator temperatures were above average
      in most of the Pacific and the Indian Ocean. The
      western Pacific and the Atlantic sector had below
      average temperature. Otherwise, this region
      generally had temperatures near the 1998-2006

      The Southern Hemisphere temperatures were
      mainly near average 2005-2014 conditions. Most of
      South America and parts of Africa had above
      average temperatures, while most of Australia had
      below average temperature. Most of the Antarctic
      continent had below average temperatures.’

      I commented on the change in the NOAA data changes in a recent post.

      • David says:

        Selecting sub-sets of climate data, to prosecute your argument is called cherry picking.

        • Don Aitkin says:

          Did you go to the site? Humlum explains clearly what he does and why he does it. If you have a disagreement with him about his methodology, you might set it out for all of us.

          It’s not clear, in any case, what you mean by ‘sub-sets’. What do you have in mind? What, in climate data, is not a sub-set?

          • David says:

            Sub-sets such as particular regions e.g. “South America and parts of Africa” instead of analyzing the mean

          • dlb says:

            He did mention the mean:
            “General: The average global air temperature was
            close to the average for the last ten years.”
            He then gave some analysis on the mean by stating how the temperature change was geographically distributed.
            David, are you bringing up valid issues or just arguing for the sake of it?

          • David says:

            Valid issues.

            “Mentioning” the mean is no substitute for analyzing the mean.

            The US National Oceanic and Atmospheric Administration has reported that the global mean temperature for July 2015 was the highest for any month since record keeping began in 1880.”

            Telling me that the 2015 temperature was a little higher in some regions and a little little lower in others is hardly surprising. Nor am I surprised that the 2015 record was “close” to 10-year mean for 2005 to 2014.

          • Don Aitkin says:

            Why do you hang on to this little factoid?

            The NOAA statement clearly lays the cause to the strong el Nino condition, which has nothing to do with carbon dioxide. If you study climate4you you’ll see that there is nothing particularly interesting in the July estimates from all the datasets. We are likely to have a high global average temperature for 2015 if the el Nino continues, but so what? Such conditions are usually followed swiftly by la Nina conditions, which bring the temperature back.

            The orthodox argument is that carbon dioxide emissions are heating the atmosphere. And indeed they might be. But as we have seen for a long time now, the effect cannot always be powerful in itself. There are other factors that can offset the warming, or enhance it. We don’t know what they are, hence the black-box term ‘natural variability’.

          • David says:

            More cherry picking. The have been el Nino’s in the past. The bottom line is that this July, July 2015, had the highest global mean temperature on record.

            And I just saw this

            “There are other factors that can offset the warming, or enhance it. We don’t know what they are, hence the black-box term ‘natural variability’.”

            Exactly! Common on Don, you have done some statistics. What is the expected value of the error term (aka “natural variability) ? Its zero!

            Which is to say natural variability will have no effect on the trend.

          • Don Aitkin says:

            Oh dear. You do want to hang to it, don’t you! Does it worry you that none of the five standard datasets reports the same?

          • David says:

            If by “it” you mean one of the tenets of the Gauss-Markov, which has been the foundation of statistical analysis since 1806, yes I do. 🙂

          • David says:

            … Gauss-Markov [theorem] …

          • dlb says:

            Tell me Mr stats expert, in the data set or sets that show that this July is the warmest on record, is it significantly (statistically) different from the second warmest July?

          • David says:

            Common sense will tell you that the highest month in 100 years is likely to be only, “a little bit” higher than the second highest month on record.

            Statistical significance is not magical. Whether that difference is “statistically significant” will depend on how many data points you include in the data set. So if we included a sufficiently large number of July temperatures from places all over the world and a equally large number of temperatures for the second highest month, the difference will be statistically significant.

          • dlb says:

            But if the variance in temperature anomalies measured all over the world is larger than the increase of the mean temperature from second to warmest July, then statistically it is insignificant.
            Common sense as well as statistics will tell you this “warmest July” is nothing but rhetoric.

          • David says:

            To conduct a test for statistical significance for difference in means you need three pieces of information about.

            (i) the means
            (ii) the variance
            (iii) the sample size


          • dlb says:

            Yes, agree.

            If you look at The Hadley data their error bars for median global temperature anomalies are typically around + – 0.15 C. Their latest latest monthly anomaly (June) was 0.728 C. You would have to go back to 2008 to find a June where the error bars don’t overlap.
            I would imagine the NOAA data is much the same, though I have not been able to find their error estimates.

    • dlb says:

      When I read who employed Quiggin I thought of you David…. Cherry Picking.

Leave a Reply