Temperature change in Australia
The rates of temperature change in Australia at rural and non-rural weather stations are compared. Over those periods in which there were increasing temperatures, the rural stations appear to have warmed at about 60-70% of the warming rate of the non-rural stations.
Data
Data from the Australian Bureau of Meteorology (BOM) is used, and is assumed to be sufficiently accurate for the purpose. It is not known whether any original measurements have been adjusted in the BOM records.
The data values downloaded from BOM are monthly mean max and monthly mean min. These are used in preference to monthly max and monthly min, as being more representative of each month’s temperature.
Weather Stations
47 weather stations were active from 1940 to 2000, ie. from 1940 or earlier to 2000 or later. Of these, 10 were rural.
| Station Number | Station Name | Latitude | Longitude | Start Year | Last Year | Is Rural? |
| 3003 | BROOME AIRPORT | -17.9475 | 122.2353 | 1939 | 2009 | No |
| 4020 | MARBLE BAR COMPARISON | -21.1756 | 119.7497 | 1901 | 2006 | Yes |
| 8025 | CARNAMAH | -29.6889 | 115.8869 | 1940 | 2009 | No |
| 8093 | MORAWA | -29.2103 | 116.0089 | 1925 | 2005 | No |
| 9510 | BRIDGETOWN COMPARISON | -33.9575 | 116.1375 | 1907 | 2009 | No |
| 9518 | CAPE LEEUWIN | -34.3728 | 115.1358 | 1907 | 2009 | Yes |
| 9534 | DONNYBROOK | -33.5719 | 115.8247 | 1907 | 2009 | No |
| 9581 | MOUNT BARKER | -34.625 | 117.6361 | 1907 | 2009 | No |
| 10073 | KELLERBERRIN | -31.6183 | 117.7217 | 1910 | 2009 | Yes |
| 10111 | NORTHAM | -31.6508 | 116.6586 | 1907 | 2009 | No |
| 10579 | KATANNING COMPARISON | -33.6886 | 117.5553 | 1907 | 2009 | No |
| 10614 | NARROGIN | -32.9342 | 117.1797 | 1913 | 2009 | No |
| 10648 | WANDERING COMPARISON | -32.6814 | 116.6756 | 1901 | 2003 | No |
| 12038 | KALGOORLIE-BOULDER AIRPORT | -30.7847 | 121.4533 | 1939 | 2009 | No |
| 12071 | SALMON GUMS RES.STN. | -32.9869 | 121.6239 | 1932 | 2009 | Yes |
| 12074 | SOUTHERN CROSS | -31.2319 | 119.3281 | 1907 | 2007 | No |
| 17031 | MARREE COMPARISON | -29.6481 | 138.0637 | 1939 | 2009 | No |
| 17043 | OODNADATTA AIRPORT | -27.5553 | 135.4456 | 1940 | 2009 | No |
| 18012 | CEDUNA AMO | -32.1297 | 133.6976 | 1939 | 2009 | No |
| 18044 | KYANCUTTA | -33.1332 | 135.5552 | 1930 | 2009 | Yes |
| 18070 | PORT LINCOLN | -34.7225 | 135.8558 | 1892 | 2002 | No |
| 21046 | SNOWTOWN | -33.7844 | 138.2133 | 1908 | 2001 | No |
| 26026 | ROBE COMPARISON | -37.1628 | 139.756 | 1884 | 2009 | No |
| 29004 | BURKETOWN POST OFFICE | -17.7425 | 139.5475 | 1890 | 2009 | No |
| 30018 | GEORGETOWN POST OFFICE | -18.2922 | 143.5483 | 1894 | 2007 | No |
| 30045 | RICHMOND POST OFFICE | -20.7289 | 143.1425 | 1893 | 2009 | No |
| 32040 | TOWNSVILLE AERO | -19.2483 | 146.7661 | 1940 | 2009 | No |
| 33047 | TE KOWAI EXP STN | -21.1642 | 149.1192 | 1908 | 2009 | Yes |
| 37010 | CAMOOWEAL TOWNSHIP | -19.9225 | 138.1214 | 1939 | 2009 | No |
| 38003 | BOULIA AIRPORT | -22.9117 | 139.9039 | 1888 | 2009 | No |
| 39039 | GAYNDAH POST OFFICE | -25.6258 | 151.6094 | 1893 | 2009 | No |
| 39083 | ROCKHAMPTON AERO | -23.3753 | 150.4775 | 1939 | 2009 | No |
| 46037 | TIBOOBURRA POST OFFICE | -29.4345 | 142.0098 | 1910 | 2009 | No |
| 55023 | GUNNEDAH POOL | -30.9841 | 150.254 | 1876 | 2009 | No |
| 66037 | SYDNEY AIRPORT AMO | -33.9411 | 151.1725 | 1939 | 2009 | No |
| 66062 | SYDNEY (OBSERVATORY HILL) | -33.8607 | 151.205 | 1859 | 2009 | No |
| 70014 | CANBERRA AIRPORT | -35.3049 | 149.2014 | 1939 | 2009 | No |
| 74128 | DENILIQUIN (WILKINSON ST) | -35.5269 | 144.952 | 1858 | 2003 | No |
| 78031 | NHILL | -36.3347 | 141.6367 | 1897 | 2008 | No |
| 82039 | RUTHERGLEN RESEARCH | -36.1047 | 146.5094 | 1912 | 2009 | Yes |
| 86071 | MELBOURNE REGIONAL OFFICE | -37.8075 | 144.97 | 1855 | 2009 | No |
| 90015 | CAPE OTWAY LIGHTHOUSE | -38.8556 | 143.5128 | 1864 | 2009 | Yes |
| 91057 | LOW HEAD (COMPARISON) | -41.0567 | 146.7883 | 1895 | 2001 | Yes |
| 91104 | LAUNCESTON AIRPORT COMPARISON | -41.5397 | 147.2033 | 1939 | 2009 | No |
| 94029 | HOBART (ELLERSLIE ROAD) | -42.8897 | 147.3278 | 1882 | 2009 | No |
| 200283 | WILLIS ISLAND | -16.2878 | 149.9652 | 1921 | 2009 | Yes |
| 200288 | NORFOLK ISLAND AERO | -29.0389 | 167.9408 | 1939 | 2009 | No |
Whether each station was rural or non-rural was determined using Google Earth. The decisions were necessarily subjective. The Google Earth views of the stations are shown in RSelectedStationsGoogleMaps (10.7mb PDF)
Note : Station 9581 Mount Barker is obscured by cloud in the Google Earth view. The population of Mount Barker is over 4,000, and the weather station appears to be in or very near the built-up area, so it is classified as Non-Rural.
The monthly mean max and monthly mean min temperatures for all these stations were downloaded from the BOM website http://www.bom.gov.au/. The temperatures were downloaded in January and February 2011, but unfortunately the December 2010 temperatures for some of the stations had not then been posted. For those stations, data was in fact complete to November 2010. Note that “Last Year” for stations that are still open is given as 2009, and that 10 of the stations closed between 2000 and 2009. A further 3 stations (29004, 39039, 91104) appear to have closed in 2009.
The data as downloaded is in http://members.westnet.com.au/jonas1/QMeanTemps.pdf
(2.8mb)
Please see the Acknowledgements and Licence information
Product code “IDCJAC0002″ is Monthly Mean Maximum Temperature, and “IDCJAC0004″ is Monthly Mean Minimum Temperature.
Calculations
Temperature trends (least-squares linear fit) over various periods were calculated separately for each calendar month, for each station, using only the monthly temperatures that were given. There was no attempt to fill in any missing data.
For each period, only the stations active for the whole of the period were used (there could be some missing data, but the test was that there was data for the first year of the period or for an earlier year and that there was data for the last year of the period or for a later year).
The overall trend for each period was then taken as the average of all trends for that period – ie, the average across all calendar months and all active stations.
The reasons for using this method are:
- If an attempt was made to fill in missing data, then there would be a risk that the filling-in method used could introduce errors or biases.
- Each calculated trend is for one calendar month only (eg. all Januaries), so has no seasonal bias.
- If trends were calculated on annual averages, instead of by calendar month, then any year with a missing month’s data could not be used, because that would introduce a selection bias.
- If trends were calculated on all months, instead of by calendar month, then the least-squares algorithm would bias the result towards the hottest and coldest calendar months. The previous issue – a missing month’s data affecting a whole year – also applies.
- Each calculated trend is for one station only. If station data was averaged before calculating the trend, then it would not be possible to use any month for which any station’s data was missing, because it would introduce a selection bias.
- The station / calendar month trends for each period were only calculated for stations active in that calendar month over the whole of the period (individual months’ data may still be missing). Thus when trends are averaged for any given period, there is no station selection bias. Note that the number of selected stations can be different for different periods.
It is possible that there may be a geographical bias in the data, ie. the stations used may have different geographical distributions for different periods, or that the geographical distribution of stations is not representative of the region. It is also possible that missing data could affect results, especially if missing values are clustered (eg. same calendar month missing over consecutive years, for a given station).
The full results, where there were at least 5 stations used in each trend average, are given in this table:
Also in: http://members.westnet.com.au/jonas1/RAverageMeanTempTrendsSummary5.pdf (42kb)
and are as follows:
Trends are given in deg C per year. So, for example, the first number, 0.0036, for the period 1910-2000, means that the average trend of Monthly Mean Max temperatures from 1910 to 2000 for Non-Rural stations was 0.0036 deg C per year. (All results are given to 4 decimal places, so there may be rounding errors).
For stations whose Dec 2010 data was missing, the average trend shown to 2010 is the average of the Jan-Nov trends. Stations are only included for periods over which they have both “Max” and “Min” temperature data.
These figures indicate that over those periods in which there were increasing temperatures, the rural stations only warmed at about 60-70% of the warming rate of the non-rural stations.
There are some curious numbers in the results, which could deserve further analysis – for example, the rural trends for 1980-2000 and 1990-2000 are much lower than the non-rural, yet for 2000-2010 they are about the same as the non-rural. Also, there is no warming trend from 1910-1940, a period of quite rapid global warming (see, eg.,
http://www.cru.uea.ac.uk/cru/info/warming/gtc.gif).
Where there were at least 2 stations used in each trend average, the results are given in
http://members.westnet.com.au/jonas1/RAverageMeanTempTrendsSummary2.pdf (43kb)
and all trend averages are given in
http://members.westnet.com.au/jonas1/RAverageMeanTempTrendsSummary.pdf (45kb)
All of the individual station / calendar-month trends (rounded to 4 dec places) are given in this document:
http://members.westnet.com.au/jonas1/QMeanTempTrends.pdf (734kb)
This too could deserve further analysis, eg. there may be significant summer / winter trend differences.
=============================================
For further information and discussion, see
http://members.westnet.com.au/jonas1/AustraliaTemperatureChangeNotes.pdf
This document includes:
- List of the 47 weather stations.
- Google Earth maps used for Rural/Non-Rural identification. [link]
- BOM website (data source). [link]
- Source data. [link]
- Individual station / calendar-month trends. [link]
Ian: The “why” is to produce a positive slope, and apparency of warming.
Scott: here’s a correction: “
ShamefullyShamelessly they call it ‘raw data’ ”There, all fixed! 😉
Mindbuilder,
would you like to show us how many stations were researched to come up with the homogenization adjustment that is so popular in the official temp series data?
This is pretty poor methodology. Most of these stations ought to be classified rural, exceptions being Sydney Canberra Melbourne Rockhampton Launceston Hobart and possibly a few others depending on specific site characteristics. Unless the result holds up with a more sensible station classification, it doesnt mean anything
“Looks like the Aussies had the reverse of the North American temperature trend 1910-1940. Sure would prefer a temp vs. time graph”
That is net surprising if you analyse the CRU data.
Very little data from 1900-1940 was actually used and 4years of data was picked from another station in the same grid.
e.g. Kalgoorlie was filled with 1890’s figures 200km away
http://members.westnet.com.au/rippersc/kaljones1990.jpg
http://members.westnet.com.au/rippersc/kaljones1999sg.jpg
Geraldton was filled in the 1890’s from Hamlem pool (200km away)
http://members.westnet.com.au/rippersc/gerojones1999.jpg
Hall’s Creek used only 1899 from the old town 12km away
http://members.westnet.com.au/rippersc/hcjones1999.jpg
Then after climategate the Met office released the CRU2010 details that for Australia purport to be “Based on the original temperature observations sourced from records held by the Australian Bureau of Meteorology”.
That was real eyeopener as that showed less warming/cooling on the sites I have checked and lower numbers for 2007,2008,2009.
http://members.westnet.com.au/rippersc/hccru2010.jpg
http://members.westnet.com.au/rippersc/meekacru2010.jpg
So which is the original data?
The old Bom site or the CRU2010v data?
The only way to find out is to go back thought the original paper forms IMO.
E.M. Smith has explained why this is a superficial view of what’s done to bias the record.
I don’t see a way to enter anything in tips and notes, so here goes.
The Wolfram Alpha weather data is summarized in the link. They list:
Among current principal sources for WeatherData are:
Citizen Weather Observer Program. “Citizen Weather Observer Program (CWOP).” 2008. »
National Oceanic and Atmospheric Administration. “National Weather Service.” 2008. »
United States National Climatic Data Center. “Global Surface Summary of Day.” 2008. »
United States National Climatic Data Center. “Integrated Surface Database.” 2008. »
Does anyone believe they realize the data might not be so good? I noted to them that they need to plot error bars and the source(s) of the data. They pointed out the Sources link at the bottom of the temp chart. (For example, in WolframAlpha, you can type in “temperature Houston” and get a temp record for Houston. http://www.wolframalpha.com/input/?i=Houston+temperature
The data source list is here:
http://reference.wolfram.com/mathematica/note/WeatherDataSourceInformation.html
This is pretty poor methodology. Most of these stations ought to be classified rural
No, it is good reasoning, although the terminology is poor.
The issue is not whether a site is “rural” or “urban”. The issue is whether the land use in the nearby area has changed over the last 100 years.
A site out of town, but by the international airport, is not “rural” for climate purposes. No matter how few people live nearby. The UHI effects are what matter, not the population.
Do try to keep up with the actual issues. The pretence that because a site not inside a town means it is “rural” is a key feature in the inflated land temperature values we get from GISS etc.
Did I miss the conclusions? A summary of some sort?
Mt Barker is non-rural, but it’s also non-urban. It’s only around 30kms from Adelaide but nestled in the heart of the Adelaide Hills. It is elevated well above adelaide though and gets a lot more rain and a constant temperature of 2 degrees less than adelaide. (Yes I Live near by)
Heavily OT, but I am anxious about the folks in New Zealand whether some have suffered from the earthquake of this morning.
Are you OK?
The coordinates of Mt Barker in the first table make it the West Australian town, not the SA town.
MOUNT BARKER ,01/1886, -34.6250, 117.6361 BoM #009581
The 2006 census population was 1622 people. The given coordinates place the weather station about 1,600 km west of the edge of town. There have been up to 3 station moves since recording began in 1886.
Darren Parker says:
February 21, 2011 at 9:48 pm
Mt Barker is non-rural, but it’s also non-urban. It’s only around 30kms from Adelaide but nestled in the heart of the Adelaide Hills. It is elevated well above adelaide though and gets a lot more rain and a constant temperature of 2 degrees less than adelaide. (Yes I Live near by)
——————————————–
I thought it was established that it was Mt Barker in WA?
I commend the researchers, but suggest that more work is needed to bring this up to publishable standard. Going back to taws (there’s one for the international English brigade) we need to know whether, and if so how and why, the BOM data has been adjusted.
It is good to see that people are out there working on this stuff.
If you still have the data in suitable form, it might be interesting to sort those stations elevated more than 150 m asl and compare trends of the low ones with the high ones. On a smaller set of data, I found this to be the most significant extraneous factor in the last 40 years.
Coastal below 150 m average trend deg C/yr Tmax 0.0084 Tmin 0.003
Inland above 150 m average trend deg C/yr Tmax0.0219 Tmin 0.0206
The other extraneous factor that needs investigation is the change from Hg to electronic temperature sensors. This started in the late 1980s. It might help explain your convergence of rural with non-rural in the decades after this.
Ed Waage : I think a few of the stations’ classifications are marginal, and there are a number of other interesting features worth pursuing, but right now I’m not putting any more time into it other than responding to some comments. The data is readily available, so anyone can work on it, but be prepared for it to be very time-consuming!
Ian George : I have no idea why the temperatures have been adjusted. If I do get going on the next phase of this exercise, it would be nice to have access to the raw data. Do you have a link? When looking at temperature trends in Oz, I think you should look at minimums [minima?] as well as maximums. Eyeballing the trends as posted, it looks like minimum temperatures went up much more than maximums soon after 1970, but that the trend reversed around 2000. More work needed …..
Mindbuilder : Yes the sample size is small, and probably too small to reach firm conclusions (which is why I said “… appear to …”). I certainly don’t want to try to read into it any more than is justifiable. I don’t think that my study proves anything, but combined with other tests for UH (such as have been reported on WUWT in the past and by Graham here) it does suggest that a proper evaluation of UH is needed.
Graham : It seems that however one looks at the data, there is an indication of UH. But as I said above, I’m not sure that the sample size that was available to me is large enough to come to a definitive conclusion.
Bill : Anthony has shown that it not the absolute size of a settlement that matters. Its growth or encroachment on the weather station is a more important factor. (Mooloo has put it well a few comments down from yours).
Darren Parker : It seems there are two Mt Barkers. The relevant one here is at (-34.625, 117.6361) just N of Albany, WA.
Oodnadatta is clasified as non-rural, it has a scattered population of 280, it is in the middle of the Simpson desert, has Finke, an Aboriginal township 130 miles (8 hours) to the north, Coober Pedy, a mining town 130 miles (5 hours) to the southwest, Birdsville 450 miles (2 days if you are lucky) to the east and Kalbarri 1100 miles (5 days) to the west. If that isn’t rural I don’t know what is.
The classic study of Barrow, AK, showed that a mere couple thousand population can cause a significant UHI effect. It depends on the details of the site and the changes that have occurred.
With that in mind, without a detailed site analysis I suspect many stations classified as rural, really aren’t. Truly rural stations are probably quite rare. A proper analysis would eliminate ALL non-rural stations, even if it eliminated almost all of them.
None of this UHI research matters if the satellites show the same warming as the thermometers. Do they? Are they a little different than the thermometers? Are they a lot different? Are the satellites calibrated against the ground stations from time to time to account for the drift or change in the satellite instruments? Calibrating the satellites against the ground stations would be a stupid thing to do if you wanted to use the satellites as evidence to corroborate the ground stations. Is there anybody doing measurements in the satellite community who we can trust not to redo the calculations as many times as it takes to get the “right” answer?
old44 : It might seem odd that a little remote place like Oodnadatta is classified non-rural, but Station 17043 is Oodnadatta AIRPORT. The weather station there appears to be right alongside where the planes taxi in and out (see the Google Maps link). The test isn’t whether the place is a major urban centre, but whether the temperature record there is likely to be contaminated by development.
Mike Jonas says:
February 21, 2011 at 11:17 pm
You are correct about minimums – they seem to be increasing (maybe due to cloud cover or increasing CO2? – just speculating.). Of course, more cloud cover can cool daytime max temps.
Link to raw data is:
http://www.bom.gov.au/climate/data/
Link to high-quality data sites (which I presume they use for their official data sets is:
http://reg.bom.gov.au/cgi-bin/climate/hqsites/site_networks.cgi?variable=maxT&period=annual&state=aus
Ian George : Thanks for the links. The “raw data” is what I used, but I don’t think it is stated anywhere that it is raw – I assumed that it could have been adjusted (but that it was the best available).
I see their high quality climate sites are classified urban or rural, with only 9 of the 100-odd sites being classified urban. Port Macquarie (Bellevue Gardens) and Bathurst Agricultural Station are among the urban sites, yet Brisbane Aero and Canberra Airport are classified rural.
Geoff Sherrington : Apoologies, I missed your comment earlier. The individual station trends by period are all in
http://members.westnet.com.au/jonas1/QMeanTempTrends.pdf (734kb)
if you want to sort them by altitude.
The trends are rounded to 4 dec places but hopefully that will be sufficient.
Let me know if you need a different format.
Interesting about the change from Hg to electronic temperature sensors from the late 1980s. All temperature analyses are in trouble if that made a significant difference.
Mike, I’m in the middle of a similar project, using several hundred stations. Like you, I’m using Google earth to define Pristine, Rural with sensor away from town, rural with sensor within town (and population from several past censuses). I have the advantage that I’ve visited very many of these places over the last 55 years and I have retained some local knowledge. Indeed, I was part of setting up some stations.
It is likely that the USA categories of rural and non-rural based on population are too insensitive. I’m listing populations from 10 people upwards , hoping to end up with composite graphs of Tmax and Tmin trends, dissected into whatever other factors appear influential. Warwick Huhes did this a few years ago. One of the factors I’ve so far found for non-urban sites is that airports seldom affect the trend. Too few flights, aircraft too small, often dirt strips with no asphalt.
Because I’m so tied up with this, I suggested the look at trends versus altitude. It’s not that I’m lazy, just thought you might be able to do it with a flick of a familar wrist on the data. You might be surprised to find how large (and how inexplicable) it is.
BTW, I too made an error with Mt Barker WA. The recording station is 1,600 m, not km, west of the edge of town.
Finally, re the changeover from Hg to electronic sensors, this was about the time of a WMO get together on data gathering, 1990 or so. If you look at various spaghetti maps like Darwin below, you often see that the various data adjusters seem to come to closest concordance about 1990, then diverge again.
http://i260.photobucket.com/albums/ii14/sherro_2008/Spaghetti_darwin_necking.jpg
In the Google maps, the position of each weather station is given. If it’s among the buildings of a settlement, even if the settlement is small, I have tended to classify it as non-rural. Anthony has made the point before that it isn’t the size of the settlement that matters, but its growth. If the station is well away from the buildings, runways, etc, then I could classify it as rural. But it still in the end can’t be anything but a subjective process.
I have a couple of hopefully constructive thoughts about this. It sounds like what you are trying to say is that the difference between the rural/non-rural classifications is in the distance between the weather station and other man-made features. If that is the case, I think you should say that, because the use of the titles “rural” and “non-rural” is misleading. More importantly, shouldn’t you show us exactly how you made the rural/non-rural classifications? How far does a weather station need to be from other mad-made features so it becomes a “rural” station as opposed to a “non-rural” station? “Well away” could mean 10m to one person and 1 km to another. And how was that distance determined? Is there research that informed that decision?
Using that set of criteria, could a properly sited weather station in the middle of park in a large urban area be classified as “rural” if there are no neighboring structures? If I put a weather station in the middle of the Great Lawn in Central Park (in New York), would that be a rural station?
Peter – the process was imprecise. In nearly all cases, the classification was pretty obvious, but some of them did come down to a subjective decision. The basic criterion was whether it looked like the station might have been affected by nearby development after it first went into operation. An airport therefore immediately put the station into “non-rural”. But a position next to a lighthouse, for example, would still qualify as “rural”. I don’t have to know whether airports do or don’t affect temperatures, the argument is that they might therefore a non-rural classification is the safe one. I accept earlier comments that “rural” and”non-rural” are not fully descriptive either, but they or “rural” and “urban” are the terms that seem to be generally used, and “possibly affected by nearby development or activity” and “probably not affected by nearby development or activity” seemed a bit unwieldy.
There were some other factors that I couldn’t allow for, such as stations moving or having changed equipment, as I didn’t have that information. Another issue that I didn’t address explicitly is this: I don’t know for sure whether the Australian temperatures used in the “official” historical global temperature calculations (GISS, Hadcrut etc) were indeed the ones I accessed, or whether the “official” history uses some other set – for example the BOM “high quality” stations. In that case, comparison of the rural trends to the “high quality” stations’ trends might have been more useful.