Go Boldly and Smash all Preconceptions: Steve Goddard

WE all have heroes. British biologist Thomas Henry Huxley (1825-1895) is one of mine. The contemporary of Charles Darwin wrote: “Sit down before facts as a little child, be prepared to give up every conceived notion, follow humbly wherever and whatever abysses nature leads, or you will learn nothing.”

I have done this for many temperature series: especially maximum temperature series and for many different localities around Australia, even the Bathurst jail because it has an exceptionally long record beginning in 1857. I have also trawled through some archives boxes, and wished I had time to examine more.

I have slowly developed a picture in my mind of what an amalgamation of all the records might look like. I like to develop my ideas slowly and from the bottom-up. I like detail and am often convinced, until I double check again, that I might have got something wrong.

Then along comes well-known American cyclist, blogger, geologist and electrical engineer, oh and also sceptic, Steve Goddard aka Tony Heller. He announced on Twitter a couple of days ago: I am bored with US temperatures, and have turned my attention to climate fraud in Oz.

He starts with a few odd charts at his Twitter feed. I provided some feed back and then he posts ‘Australian Afternoons Used to Be Much Hotter’. The title and the chart excited me like only an entomologist can get excited when someone has suddenly completed the collection for them.

The chart is as I imaged the maximum temperature trends for, at least eastern Australia, would look like after all my station data was amalgamated. Steve’s methodology is very straightforward: There are 1,655 Australian stations in the GHCN (Global Historical Climatology Network) database that include temperature data. The database is located here: ftp://ftp.ncdc.noaa.gov/pub/data/ghcn/daily/all/. The station list is at bottom. The database includes maximum temperatures. He averages the daily data for each month at each station over the lifespan of the station to generate a monthly mean, and then calculates anomalies from the monthly mean. The numbers in his charts are the average anomaly across all stations for all records during a year.

Screen Shot 2014-09-04 at 9.40.01 PM

What Steve’s chart shows is that it was much hotter in the late 1800s, than at any time since and by a significant margin. He has suggested that 1878 is the hottest year on record. Coincidentally my scrutiny of data from the Bathurst jail weather station had already turned-up the hottest day in that record as being January 12, 1878.

There is a cooling trend to about 1960, and then temperatures start to warm again. But they never reach the highs of the late 1800s and early 20th Century.

Ten years ago, when I started this blog, my daughter wanted to be the first to leave a comment and she wrote: “Do not go where the path may lead, go instead where there is no path, and leave a trail.” Ralph Waldo Emerson.

Steve Goddard has just done this with the Australian temperature record. He has gone where no-one has dared go before. Thank you, Steve.

28 Responses to Go Boldly and Smash all Preconceptions: Steve Goddard

  1. Bryce Letcher September 4, 2014 at 11:10 pm #

    We need many more intelligent educated people with open minds (and time) to look at these data sets and analyse them independently. Currently there are far to many people who are to willing to believe what they are told in the Murdoch and Fairfax press rather than consider that the press may be biased. Having many more opinions from respected people will help redress this situation.

  2. Colorado Wellington September 5, 2014 at 4:21 am #

    “Then along comes well-known American cyclist, blogger, geologist and electrical engineer, oh and also sceptic …”

    You missed metric football aficionado and Whole Foods stammgast, Jennifer.

    The man is large. He contains multitudes.

  3. Emily Frazer September 5, 2014 at 5:29 am #

    Graham Lloyd continues to smash preconceptions in the OZ.


    This time he is quoting Dr David Stockwell who is calling for an audit.

  4. Mikky September 5, 2014 at 6:14 am #

    A similar graph (for South East Australia) is shown in a published paper, see Fig. 6 of Ashcroft et al, “Temperature variations of southeastern Australia, 1860–2011”, available here (click on VIEW for the pdf):


    Homogenisation drops all early temps by around 1C, not quite as blatant as the 2C applied by BoM.

  5. Svend Ferdinandsen September 5, 2014 at 6:42 am #

    Because these temperature measurements and averages is used so extensively in climate science they need to be gold standard. It should not be possible to find any questionable records.
    Anyway you find it when you dig into the data like you and now Steven Goddard.
    I wonder how many findings in climate science that might turn out to be wrong because they used these homogenised temperatures.
    The DMI (Meteorology in Denmark) keeps a faily good statistics of temperatures, and from that it is clear that large differences in daily, monthly and even yearly means can exist in a few hundred kilometers. Remember that Denmark is mostly surrounded by water. It means that large differences can exist, and large changes in mean values of various timescales does not mean that any station needs adjustment. It is just how weather is.
    I really hope BOM some time comes up with an explanation.

  6. Steven Mosher September 5, 2014 at 8:09 am #

    nope. he did it wrong again.

  7. jennifer September 5, 2014 at 8:55 am #

    As I explained in the above post, I worry about the detail. I was worrying about the 1878 hot days in the Bathurst Jail record. I worry about outliers and I worry about tossing out such outliers. In my work on rainfall forecasting, we keep all the outliers in the record and make the artificial neural network work hard to reproduce them. These outliers are often the big flood events… like Brisbane January 2011.

    Anyway, I know a lot of statistician just discard them when dealing with temp and other records.

    Now I see Steve Goddard, is giving me more reason to have more confidence in the original Bathurst jail record.

    Fantastic old blog post by him here… http://stevengoddard.wordpress.com/2014/04/04/1878-the-year-earth-went-crazy/

    Some quotes… including from the thread…

    “Jason Calley says:
    April 5, 2014 at 12:42 am
    From 1860 to 1877 scientists and meteorologists all around the Northern Hemisphere were stupid. They mismeasured temperatures, such that even when their figures were collated together, their errors refused to average out, leaving them with readings .2 degrees too hot. This went on year after year, but then, 1878 came along, and that year, the scientists were… stupid again. But THAT year, they were all so overwhelmingly bad that they consistently measured temperature too hot, and not just by .2 degrees like always. No, that year they were twice as far off and in the other direction. For that single year, the worlds scientists swung .6 degrees from the previous errors — but not to worry! The year after that, 1879, they returned to their old errors as if the wild year of 1878 never happened.”

    Anything is possible says:
    April 5, 2014 at 1:59 am
    Actually, the Earth did go kind of crazy in 1878:

    This is worth filing away for when our “friends” start babbling away about “unprecedented catastrophe” the next time there’s a major El-Nino :

    Jason Calley says:
    April 5, 2014 at 5:24 pm
    The question is not whether the Earth went crazy; it may well have done so. The question is whether the measurements made that year went crazy; were the readings in error by .4 degrees?

    1878 must have been very warm as the first commercial transit of the Northern Sea Route was made in that year according to Wikipedia

    “However, it was only in 1878 that Finnish-Swedish explorer Nordenskiöld made the first complete passage of the North East Passage from west to east, in the Vega expedition. The ship’s captain on this expedition was Lieutenant Louis Palander of the Swedish Royal Navy.”


    Studiodad says:
    April 5, 2014 at 3:25 pm
    The earliest ice out date for Lake Minnetonka in MN in 136 years is 1878. http://climate.umn.edu/doc/ice_out/ice_out_historical.htm

  8. egg September 5, 2014 at 9:36 am #

    ‘…akin to pouring a single bottle of water on a fire dance of raging global warming alarmists.’

    Well done, its the thin edge of the wedge and it works.

  9. Peter September 5, 2014 at 10:23 am #

    I was a Weather Observer for a couple of years in the late 60s. About a year ago I took a look at the actual temperature records for Sydney and another place in country NSW which had a long history (I forget which, it was over 100 years) and saw no discernible difference in max and min.

    I email the climate contact at the BOM and was told this was because of technical differences in the accuracy of thermometers and placement.

    I still don’t agree that there would be differences of the significance they claim.

    Don’t underestimate your contribution. You have fulfilled your daughters quote to you, got this topic on the front page of The Australian and got the ball rolling.

    Really, really well done to you.

  10. Siliggy September 5, 2014 at 11:23 am #

    This seems to be all about maximum temperatures. They were more extreme then with far hotter heat waves but what did the minimums do? Me thinks they were more extreme too! As Co2 was meant to increase extreme weather it seems clear the opposite has happened. Co2 was to have more effect on minimums but that does not translate to maximums going wildly down. So surely there is another cause for the changes and humans are likely to be uninvolved.

  11. egg September 5, 2014 at 4:11 pm #

    Going back further to the time of William Dawes, Australia’s first meteorologist (1788-1791), I calculated that his observations were at least a degree above the current average at Observatory Hill. He had no Stevenson Screen, but he knew what he was doing.

    The result surprised me, as we were about to sink into the Dalton Minimum.

  12. Neville September 5, 2014 at 8:00 pm #

    Here are two trends covering the last 114 years of the HAD 4 data set. The first is from 1900 to 1950 and the second is from 1950 to 2013. The post 1950 trend (64 years) is the same as the earlier 50 year trend even though the later trend should show an impact from AGW.


    The next is the PDO index graph over the last 114 years with all the trend lines shown from HAD 4.


    I’m sure that most of the warm and cool phases of the PDO correlate well with the temp trends over the last 114 years. Not co2 emissions.

  13. Johnathan Wilkes September 5, 2014 at 8:50 pm #

    What I find most annoying is that some, if not all ppl involved in the climate change business on the warmist’s side regard the record keepers/observers of the past as incompetent imbeciles who could not make sense of a thermometer reading or didn’t think it was important to keep accurate records.

    May I remind them that all our current technology and scientific achievements are based on their work?
    If anything their work ethic was a lot higher than what we have today.

    What we have now are instruments that are easier to read, being digital, but their accuracy still depends on constant recalibration of the sensors, and I’m sure that Gavin would back me up on this.
    Hope you are still with us Gav?

  14. DaveMyFace September 5, 2014 at 9:17 pm #

    A bit off topic, but you did bring up 1878 as a crazy year

    A year when in the Wimbledon Mens Tennis Final

    Hadow beat Gore (7-5- 6-1 9-7)

    Now Hadow is here


    And we all know about GORE

    Were these tennis players the genetic relations of the current CAGW boys?

    Maybe they changed the rules then?

    CAGW – a game only the CLIMATE boys club are allowed to play
    For ever

  15. Steve Goddard September 5, 2014 at 10:40 pm #

    Steven Mosher,

    Berkeley Earth has temperature data for Nobby Newcastle going back to the 1860’s, but GHCN reports no temperatures there before 1957. GHCN also shows much more warming than Berkeley Earth.

    What is wrong with the GHCN database? Almost all of the pre-1957 data is missing.

  16. Colorado Wellington September 5, 2014 at 11:33 pm #

    Steven Goddard posted today: Major Monkey Business In Australia

    “A few days ago I noted that the GHCN database for Australia has very little data prior to 1957, and wondered if the jump was due to Sputnik.

    Well it turns out that lack of data prior to 1957 is due to GHCN simply not including it. Berkeley Earth has Nobbys data going back to the 1860’s, but the GHCN data is missing prior to 1957 …”


    I’d like to know how much Steven Mosher knew about this when he dismissed Goddard in his comment above.

  17. Sophie September 5, 2014 at 11:46 pm #

    Steven Mosher September 5, 2014 at 8:09 am #
    nope. he did it wrong again.

    In light of your reply, I remember your famous comment on WUWT:
    ‘ You’ve got to fix the data.’

    I’m so glad that Steve/Tony uses the ‘real data’ and not the ‘fixed data.’ 🙂

  18. jennifer September 5, 2014 at 11:49 pm #


    If you want to get familiar with the ‘raw’ data held by the Australian Bureau it is here…

    But it would be more immediately useful to me if you could do an analysis of some data downloaded from an FTP file at the BOM some years ago… it is the data used to construct the first homogenised data set for Australia… I think.

    I shall give you access to a dropbox by email.

  19. Jennifer Marohay September 6, 2014 at 12:11 am #


    Once I give you access to the Dropbox you should have data for all the stations listed in Table 1 of this key paper… http://www.waclimate.net/torok_study.pdf

    This paper by Torok and Nicholls (1996) was a seminal paper and I suspect they did a fair job of the homogenising. Also for key locations of interest to me they have homogenised data that goes way back… see for example 63005 Bathurst (Agricultural Experimental Research Station)… which is actually an amalgamation of the jail and research station with a homogenised data set that starts in 1858.

    It is past midnight in Eastern Oz, so I need to go to bed. But I had wanted to look and see what a plot of this data looked like ASAP. If you could email me a chart/post at your site… so that is the long series 63005… and after that any of the other really long series.

    Also it would be good to see all the data in the file (so all the data for all the stations listed in table 1) amalgamated as per the methodology you used in the above plot that you have done of the GHCN station data. But the title would be along the lines… Maximum Temps as per Torok and Nicholls 1996 Homogenised Data Sets…. but we can work on this… the trends post 1910 should look like Fig 6 of the paper… but we would also have data that started back in the mid 1800s… which is what I am desperate to see.

    So these are my ideas… which I have been wanting to implement… but I’m a bit time poor. But happy to work with you on this over the weekend. Jen xo

  20. jennifer September 6, 2014 at 12:45 am #

    So, I’ve dropped the zipped folder… for Steve… but will I be able to sleep?

    John Sayers… I will phone you about this in the morning.

  21. Colorado Wellington September 6, 2014 at 2:40 am #

    I thing that Goddard’s inquiry above was not visible yet when I ask my question about Mosher, Berkeley Earth and GHCN:


    [That is correct. Steve’s comment was initially caught by the filter. I had to release it from pending. Jen]

  22. Colorado Wellington September 6, 2014 at 7:33 am #

    Thanks, Jen. I’ve wondered if I’d missed it.

  23. jennifer September 6, 2014 at 4:36 pm #

    So, I didn’t get much sleep but I did extract the Bathurst series from the file (downloaded from an FTP site at the BOM about 6 years ago. all credit to JS) with all of the data as detailed in Table 1 of Torok and Nicholls 1996.

    Regarding the homogenisation undertaken by Torok and Nicholls… well it was much milder than the two versions (HQ and ACORN-SAT) that have followed.

    You want numbers… Nah. I’m going to write this up for peer review. 😉

  24. Neville September 7, 2014 at 9:15 am #

    Interesting info from blogger Greg Goodman at WUWT about the increase in SST in the N pacific. IOW producing more ice in the Arctic releases a lot more heat.

    Greg Goodman

    September 6, 2014 at 8:38 am

    As Bob noted in his last post on this, almost all of the warming in SST is in N. Pacific.

    To be more precise it is centred on Berring Sea area and also Labrador Str between Canada and Greenland.

    This article mentions the effects of evaporation but it is the latent heat of fusion ( freezing/melting ) of water that may be more pertinent.

    A quick back of envelope calculation using phycical properties of pure water rather than sea water shows that freesing 1kg of water releases enough heat to warm 100kg of water by 0.8K. That is about the current anomaly in those regions.

    The SST shown has been climbing since 2012. That was the low point of arctic sea ice coverage. There was a recovery in ice volume of 50% in 2013. That implies an enormous transfer of energy, initally to surrounding water.

    It looks like that may be the origin of the recent rise.

    Maps shown in Bob’s post show most of the rest of the world is showing very neutral SST anomalies.

    Most of this warming could be due to the Arctic ice recovery.

  25. Neville September 7, 2014 at 11:50 am #

    More on the leadership from the Abbott govt that help other people overseas to question the pause in warming for the last 17 years.


  26. Maureen Coffey September 9, 2014 at 5:16 am #

    You see, this can’t be science. My logic: if people with millions in grant money cannot find this, this can’t be science. See? As an aside I would like to know what would happen if we could find a very good statistician who knows absolutely NOTHING about the current climate debate and gave him/her all the data that went into the NOAA series and models and let him/her think it’s about the incidence of some epidemic or shoe sizes or whatever. Then let that person do his or her best to mine these data and come up with trends and correlations. While I may be biased I have an inkling where this impartial audit might lead …

  27. Larry Fields September 9, 2014 at 3:19 pm #

    Jennifer wrote:
    “Then along comes well-known American cyclist, blogger, geologist and electrical engineer, oh and also sceptic, Steve Goddard aka Tony Heller. He announced on Twitter a couple of days ago: I am bored with US temperatures, and have turned my attention to climate fraud in Oz.”

    He’s also a footy aficionado.

  28. omanuel September 17, 2014 at 1:27 am #

    Thank you for your brave efforts with Steven Goddard to unravel the five-year (2014 – 2009 = 5 yr) old Climategate mystery.

    From information in the autobiographies of two great scientists – Fred Hoyle (astronomer, astrophysicist and cosmologist) and Paul Kazuo Kuroda (nuclear geochemist) – I am convinced that the mystery began sixty-four years (2009 – 1945 = 64 yr) earlier, in the chaotic closing days of the Second World War.

    The evidence is summarized here: “Solar energy,” Advances in Astronomy(submitted 1 Sept 2014) https://dl.dropboxusercontent.com/u/10640850/Solar_Energy.pdf

Website by 46digital