Talk:List of cities in Australia by population

Townsville population

Shouldn't the Townsville Statistical Division include Thuringowa? They are seperate Local Government Areas however they are no more two seperate cities than Sydney is ten cities. See dicsussion in Townsville, Queensland. WikiTownsvillian

Thanks for that edit Bongomanrae, however to be fair now that Townsville has been restored to 13th, Thuringowa should be removed from the list as its population is alredy counted, I have done this today, any objections please post here. WikiTownsvillian (8 Oct 06)

Just to note, Townsville and Thuringowa LGAs were merged in 2008, so this issue will not be ongoing for future ABS data releases ROxBo (talk) 11:26, 31 August 2009 (UTC)

Archive

This page is a bit long, and quite old mostly. It needs to be archived. Can someone do it, as my attempts at doing this myself have not been great! ROxBo (talk) 11:29, 31 August 2009 (UTC)

Done. --AussieLegend (talk) 11:45, 31 August 2009 (UTC)

Critical issue

The key issue with "Population data in Australia" is that the various classifications of the ABS are not immediately intuitive and overlap considerably. This is the driver behind the repetitive edit/redits that each town or city seems to go through, e.g. (new) editor finds LGA Townsville population is bigger than that listed in Townsville wiki page, so makes well-meaning change (again, and again and again). I think it would be great to have an image (or animated gif?) PLUS a distinct and separate subheading that briefly explains statistical district/division/lga/urban area. The text is sort of already there, but fragmented. Cheers ROxBo (talk) 11:36, 31 August 2009 (UTC)

Urban Centres

Does anyone know where I can find a list of the physical size (square kilometers) of Australia's "Urban Centres"? Are there any maps out there showing exactly what the Census defines as Urban Centres boundaries? --Criticalthinker (talk) 07:30, 19 September 2009 (UTC)

Yep. Just go to the Census search page, key in the UCL you want and highlight it, and a map springs up to the left. If you follow it through and download the "Basic Community Profile" (rather than going to Quickstats) which is a 200k compressed Excel spreadsheet and click on any page number link on the contents page of it, the header section of the resulting page will contain the area. It is laborious and I wish I knew a better way, but that is how I get the areas for suburb articles. We should lobby ABS to produce a nice clean list on their site :) Orderinchaos 08:10, 19 September 2009 (UTC)

Thanks. I did it for Melbourne and found an area of 2,152.8 square kilometers. There has to be somewhere on each major 'city' page where it'd help to put this fact, or perhaps to add it to the chart on this page to give an idea not just of population but density. BTW, I only put 'city' in quotes because what they call cities in Australia, we call urban areas or metro areas, it seems. --Criticalthinker (talk) 03:36, 21 September 2009 (UTC)

Missing cities

I noticed Ipswich and Gosford are missing and they are both over 150,000. matthewhicks93 —Preceding undated comment added 18:46, 22 November 2009 (UTC).

This is because neither are classified as Statistical Districts/Divisions or Urban Centres/Localities by the Australian Bureau of Statistics. Gosford (which is wholly included in Sydney's SD for some reason) is in 23rd for Local Government Areas, while Ipswich (partly included in Brisbane's SD) is smaller than the 25th so not included.

WP:OR and WP:RS prevent us from straying from this. It's also important not to confuse LGA boundaries with cities - the City of Gosford includes the urban areas of Woy Woy/Umina and Erina/Terrigal/Avoca, which are significantly disconnected from the Gosford metropolitan block, while Ipswich includes towns which aren't even part of an urban area such as Rosewood, and includes suburbs detached from Ipswich proper which are more outgrowths from Brisbane. Orderinchaos 16:49, 23 November 2009 (UTC)

Is there a list of cities that isn't based on statistical districts because there are so many left out? matthewhicks93

Unfortunately, no - if you want to see them changed for the ABS census in 2011 (and there is actually a fair few of us who do), I would suggest lobbying the ABS. I asked them regarding the 2006 census a while ago and they have advised they can't make changes to the structure of that data now, but obviously changes can be made ahead of time for the next one - they made literally thousands of improvements between 2001 and 2006. The contact details can be found here (select "Other"). No other organisation in Australia produces a comparable list which would meet Wikipedia's reliable sources guideline at the present stage. Orderinchaos 11:22, 25 November 2009 (UTC)

Thank you. Matthewhicks93 (talk) 05:10, 27 November 2009 (UTC)

Alice Springs not included anywhere, population at least 30,000. —Preceding unsigned comment added by 210.8.208.131 (talk) 04:23, 7 April 2010 (UTC)

According to the ABS the population of the Alice Springs LGA was only 23,893 in 2006 and now is only 27,877. The UC/L population was only 21,622 in 2006. There's no place for Alice Springs in this article. --AussieLegend (talk) 05:55, 7 April 2010 (UTC)

Sunshine Coast population

Since amalgamation the poopulation of the Sunshine Coast regional council is wrong, it is much closer to 300k which would put it in the top 10 —Preceding unsigned comment added by 123.211.151.57 (talk) 11:06, 4 January 2010 (UTC)

However, neither of them are based on council boundaries, they are defined by the ABS and actually span parts of all three former LGAs. The statistical district used for the first is here, whilst the UC/L population is defined here. Orderinchaos 15:12, 4 January 2010 (UTC)

I think you might find this link of interest [1] --Orestes1984 (talk) 06:21, 5 January 2010 (UTC)

That is local government areas. It even explicitly says so. By that standard, incidentally, Sunshine Coast isn't even a city, as it's a "region". Orderinchaos 06:24, 5 January 2010 (UTC)

Woolongong is a region, Newcastle is a region, the Gold Coast is a region, none of those places are cities in their own right either all of them cover a large regional area, for that matter Queenbeyan is part of New South Wales and thus can't be part of the ACT/Canberra at the same time, the same issue comes up for Tweed Heads. The citation in this article doesn't even state which CBD is considered the Sunshine Coast. Is it Noosa heads, Maroochydore, Caloundra, Gympie, Nambour?? more relevant citations are needed. --Orestes1984 (talk) 01:19, 7 January 2010 (UTC)

I'm a bit confused as to what you're referring to. The Bathurst Region is as far as I'm aware the only region in NSW, while QLD adopted the convention for a heap of local governments on 15 March 2008. Wollongong's LGA is the City of Wollongong, but the city that the ABS measures includes much of Shellharbour and a bit of Kiama, and not *all* of Wollongong (it excludes more rural bits). Newcastle combines parts of (not all of) the City of Newcastle and the City of Lake Macquarie. Gold Coast is entirely within the Gold Coast City LGA, but excludes large sections of it (rural sections).

I suggest using the "Browse" tool on the census stats for 2006 at the ABS - you'll see what is being referred to under these terms. The ABS measure LGAs separately from other quantities (Statistical Districts and Urban Centres/Localities). It's quite precise, and to use anything else violates WP:OR, a core policy at Wikipedia. Orderinchaos 01:51, 7 January 2010 (UTC)

This is very old information, Bendigo is listed as the forth largest city in Victoria when it is in fact the third largest city in the state, it has a population of over 103 thousand, yet no one will change this! Ballarat is not the third largest city in Victoria, it is the third, it only has about 89 thousand people, which is a big difference, it needs to be changed! —Preceding unsigned comment added by 203.171.70.115 (talk) 07:17, 19 May 2010 (UTC)

This article uses information published by the Australian Bureau of Statistics, which is the source for all population data in Australia. The information is current as of June 2009, because that's the most accurate estimate available for all cities. --AussieLegend (talk) 08:40, 19 May 2010 (UTC)

Wholesale edit for all australian cities!!

In this page there are 2 populations one urban and one metropolitan- shouldnt both populations be listed in the info bar in each cities article ie

Sydney Metro 4.5m Urban 3.5m

they do it for most cities accross the world in wiki but never any aussie ones! Why? —Preceding unsigned comment added by Luke193 (talk • contribs) 02:59, 6 December 2010 (UTC)

Very simple reason - how would you define "urban"? Australia (along with some other countries, mostly Commonwealth ones) does things a bit differently to most in two ways:

1. The City of Sydney or the City of Perth are tiny administrative areas whose borders essentially mean nothing in any real-world sense beyond defining which councillors represent which areas.
2. The Australian notion of "suburbs" is wiidly different to that used in Canada or the US to give one example - there is quite a big block of land simply called "Detroit" or "Los Angeles" or "Vancouver", with areas outside that forming a metropolitan region - whereas in Australia that would be subdivided into maybe 100 suburbs each with equal status to each other and managed by a raft of local authorities. One would have to manufacture an entity which does not exist in the real world just for Wikipedia's purposes to define Sydney, Perth or most other Australian cities more broadly. Orderinchaos 22:03, 28 July 2011 (UTC)

City vs Urban area

In the city list, there are "cities" like Gold Coast-Tweed, Canberra-Queanbeyan, Greater Hobart, Albury-Wodonga, Burnie-Devonport, Nowra-Bomaderry and Kalgoorlie-Boulder. Aren't these just the "urban areas"? Six of those seems to include two cities, and to my knowledge, Greater usually means the urban area. The city should just be Hobart.
Reading the text, I understand the listed ones are some kind of Statistical areas. Not sure though why these should be considered cities. 85.217.43.98 (talk) 18:26, 27 July 2011 (UTC)

I have to agree with you but there's a long standing consensus that we use the ABS statistical areas to define city boundaries, despite what the individual state definitions may be. --AussieLegend (talk) 09:04, 28 July 2011 (UTC)

Two of the six, Nowra-Bomaderry and Kalgoorlie-Boulder, are not "two cities" each - Bomaderry is a suburb of Nowra, Boulder is a suburb of Kalgoorlie, but there's historical reasons why the names are used together. I agree with you on Hobart. We're effectively confined here by Wikipedia's original research policy, which means we have to use what is provided to us even if it's not perfect. FTR, I put in a submission to ABS about some of these issues ahead of the 2011 Census, so hopefully they'll be nice and address some of them for us :) Orderinchaos 21:58, 28 July 2011 (UTC)

Didn't even think about it back then, but these so-called cities do not even have articles as "whole" cities. Only one of those, Kalgoorlie-Boulder, links to Kalgoorlie and ignores Boulder. All others have separate articles for both parts, also the new "double city": Newcastle-Maitland, to which Maitland has been added since then. 82.141.67.208 (talk) 00:49, 18 December 2013 (UTC)

The Newcastle-Maitland figure doesn't actually represent anything. The old figure used in the article was effectively the whole lower Hunter Region and included areas that nobody in their right mind would associate with Newcastle, so at Newcastle, New South Wales we used the UCL population, which was 308,308 at the 2011 census but even it covers a lot of Lake Macquarie and one Port Stephens suburb while missing out on several Newcastle suburbs. Newcastle-Maitland SUA is 398,770 and covers all of Newcastle, less than a lot of Lake Macquarie, most of Maitland and some of Port Stephens (but not the suburb included in the Newcastle UCL). The actual combined population of Newcastle (148,535) and Maitland (67,478) is only 216,013 with most of the difference being residents of lake Macquarie which, as always, is denied its identity as a city even though it is the largest city in the region with an area more than twice that of Newcastle and a population much higher than Newcastle. --AussieLegend (✉) 11:21, 18 December 2013 (UTC)

Where Is Alice Springs?

According to the census it has a population of around 30,000 but is not on the list. — Preceding unsigned comment added by 120.17.60.47 (talk) 10:04, 21 October 2011 (UTC)

Alice Springs is gazetted (NT Place Names Register) as a town, no matter what its population is. Darwin and Palmerston are the only gazetted cities in the NT. Bidgee (talk) 10:10, 21 October 2011 (UTC)

Perth population

I would appreciate it if someone with more statistical nouse and skill than I can explain how the relevant citation points to the figure 1,738,807 for Perth. I have only been able to find the 'rounded figure'of 1.74 million here. Also, if we are not using a rounded or estimated figure, perhaps we should be including the actual date at which the population was exactly 1,738,807, eh? Cheers, Bjenks (talk) 05:29, 1 April 2012 (UTC)

It's in the Downloads tab at that link, but the actual XLS can't be directly linked because the ABS don't allow for it. They should - I've written to them about this some time ago. Orderinchaos 06:02, 1 April 2012 (UTC)

The figures for all of the capital cities may be found at the cited source in the table titled "ESTIMATED RESIDENT POPULATION, States and Territories - Capital City and Balance of State/Territory". At the very bottom of the page are three tables. It's the first of these three tables. The populations are always as of 30 June in the year prior to release of the document so, for the 1,738,807 figure it's 30 June 2011, although we round down to just June 2011 in the table, since the actual day is not essential. It's only an estimate anyway. --AussieLegend (talk) 08:44, 1 April 2012 (UTC)

Aha, so it is, duh! The figure didn't show up in my searches simply because it uses spaces, not commas. So shouldn't your excellent guideline, above, be pasted into footnote #4 for the benefit of inexpert readers like myself? Cheers, Bjenks (talk) 09:56, 1 April 2012 (UTC)

Central Coast

This region of Gosford Wyong Shires has been left out and adds up to around 300,000 people Could this oversite be rectified please. — Preceding unsigned comment added by 220.233.26.53 (talk) 09:00, 7 May 2012 (UTC)

2011 Census

The 2011 kind of makes these figures inaccurate as the ABS has found its interim estimates were vastly out. This list needs to be updated with all three columns removed and replaced by the most recent two official census figures - 2006 and 2011. --Biatch (talk) 10:18, 24 June 2012 (UTC)

No, census figures, as admitted by the ABS, are always very out. Why is this? Because there is always a significant undercount, given people who are away/in the car/not answering on the night. This is why the ABS' official estimates will be higher than census figures. Every 5 years based on census data, they will alter their predictions slightly based on the findings of the census. You are right that when they come out with their next set of official estimates, they will be altered down slightly as it is true that they found fewer people than expected. This does not change the fact, however, that the official population estimates are the most accurate numbers and the ones which should be used. Saruman-the-white (talk) 02:03, 28 June 2012 (UTC)

Correct - the estimates are what governments use on a day-to-day basis for planning, while the census gives them more detailed information on a five-yearly basis. Orderinchaos 05:14, 29 June 2012 (UTC)

So what of the "50 largest Urban Centres by population" section then. It uses 2006 Census figures. Shouldn't they be updated with 2011 Census figures ? --Biatch (talk) 04:13, 29 June 2012 (UTC)

Still have to sort out what to do with that - they've done away with UC/Ls and come up with new groupings like "Statistical Area Level 3" and "Level 4". Newcastle is a Level 4, Townsville is a Level 3, the "Greater" areas for metro areas include all sorts of unexpected areas (Melbourne includes Bacchus Marsh and Macedon for example, Sydney contains the entire Blue Mountains and Central Coast, Perth includes Mandurah) and no longer accord with state definitions of metropolitan areas, and there's now no meaningful statistics for Mandurah, WA's second city (you can get them by a two-step process, but that's OR from a Wikipedia point of view). In short, the 2006 areas and definitions simply don't exist any more and for a fair few places there isn't even an equivalent. If anyone has any ideas which do not contradict OR, feel free to suggest them. Orderinchaos 05:14, 29 June 2012 (UTC)

Actually, the old Sydney SD is not much different to the new Greater Sydney. It's a completely unrealistic representation of the real Sydney.[2] --AussieLegend (talk) 17:59, 30 June 2012 (UTC)

They have been done away with so there is no use keeping them. As every year passes they will become more and more out of date and basically constitute completely useless, unusable figures. I say we have one for the Greater Capital City Regions which we already have for 2011, and one for "cities proper" (otherwise known as LGAs in Australia). This would be the most logical approach, but the 'urban regions' really need to be removed as they are already thoroughly out of date and will continue to become more out of date as time goes on - useless. Saruman-the-white (talk) 11:15, 29 June 2012 (UTC)

Part of the problem is that with the new designations, there's no way to actually compare metropolitan regions any more - we must be one of the only civilised countries in the world where this is the case! (I'm half beginning to wonder if lobbying ABS is the answer, this *really* is a mess.) The "Greater" regions are useless as a measure as they include all sorts of things which have nothing to do with the metropolis according to anyone else apart from the ABS. Orderinchaos 14:54, 30 June 2012 (UTC)

I raised the no 2011 UC/L data on the Australian noticeboard talk page and User:Mattinbgn stated that the 2006 UC/L data was released in a second batch later in the year, I guess we have to hope that is the case. Bidgee (talk) 02:42, 1 July 2012 (UTC)

On the contrary, the metropolitan regions are based on commutership and shared labour markets. Brisbane's is in fact not extensive enough, as it does not include the Gold Coast (mere 30-40 km away from Brisbane CBD) whereas the Sydney region has always included the Central Coast. These are the official figures so they are the ones that must be used. If we want to go any smaller, we could have a second table for cities proper (LGAs) which is what other countries tend to do. — Preceding unsigned comment added by Saruman-the-white (talk • contribs) 02:35, 1 July 2012 (UTC)

I find it hard to believe that the metropolitan regions are really based on commutership and shared labour markets. The Sydney figures include places like Glen Alice, which is 128km as the crow flies and 223km by road from the Sydney CBD. Sydney also extends north to Catherine Hill Bay, which is 121km from the Sydney CBD, but only 34km from Newcastle. I'm convinced the regions are determined by some gimp that they keep locked in an airless dungeon below the ABS offices in Botswana. --AussieLegend (talk) 05:15, 1 July 2012 (UTC)

Agreed with AussieLegend in totality. LGAs are meaningless as a basis for comparison, except to compare LGAs. Perth for instance has just 11k! Bunbury (C) has half of Bunbury's population and actually looks to be smaller than Albany, Geraldton or Bunbury despite being over twice the size. And there was a huge war on Wiki over Townsville and Thuringowa until those two merged in 2008, basically because they're one metro area. Orderinchaos 13:03, 1 July 2012 (UTC)

Here's your problem! Located in the Canberra suburb of Belconnen (across the road from Westfield Belconnen), no doubt they just looked at a map and thought that has to be right! Bidgee (talk) 14:23, 1 July 2012 (UTC)

The name of the article is list of largest cities in Australia. As such, strictly, it should be about the largest LGAs, as these are the official cities. City of Sydney, City of Melbourne, City of Brisbane, etc. This is what most other countries do, and yes, the boundaries of these cities are totally arbitrary - not just in Australia. They usually just encompass an inner core of a metro area and exclude a very large proportion. If you want to include a second figure aside from the greater capital city regions (metro areas) - which are quite well done in my opinion with the exception of Brisbane's which should include the Gold Coast - then a secondary table with the cities (ie "cities proper" or "LGAs") would be the logical way to go. Also this is consistent with the practice in the vast majority of other countries, which use a figure for a metro area which extends a long way out and includes vast suburban areas across a large sprawl, as well as a figure for the official city proper extending to city limits - however small and arbitrary, given the extent to which the metro areas have grown out of them, may seem. Saruman-the-white (talk) 15:59, 1 July 2012 (UTC)

You are confusing a few things. The use of the term "City" in administration is not the same as in common language or human geography or urban planning, where the term "city" is independent of the former, and it is the metro area of Sydney or Melbourne which is meant. Also local administrative divisions are quite different in Australia from many other countries, and an LGA can be any type of area, rural or mixed. Unfortunately the ABS doesn't seem to care much about comparability and every year changes many aspects of its methodology. We can either simply reflect the ABS data and note the differences (change of the reference area, etc) or present comparable figures relative to the same metro area (for instance as used in urban planning, simple sum of metropolitan LGAs). Note that simple calculations are not OR per WP:CALC. --ELEKHH^T 16:36, 1 July 2012 (UTC)

In NSW, legislation establishes that city boundaries are officially defined by the LGA boundaries. That's why the city of Cessnock, which is really just what's inside the red line in this map, covers all of this area. If my town were to become large enough to become a city, the city would be called "City of Port Stephens", not "City of Raymond Terrace" and would cover all of the Port Stephens LGA. Sydney is unique in the state, as it is the only city that is comprised of multiple LGAs. Saruman is correct in saying that we should compare LGAs, at least as far as NSW goes. Any other comparison requires a WP:OR definition of a city, as the legislated city boundaries are the official boundaries under the law. However, the NSW city definition doesn't extend to other states. WA, for example, uses a quite different method of defining city boundaries so comparing LGAs is not appropriate. Of course, even in NSW, comparison of LGAs is meaningless because of the variations between them and the "inconsistencies" such as Cessnock. As to what we should do, at Medowie, New South Wales I've added a "Historical population data" data showing the populations at the 2001, 2006 and 2011 census, with citations for each using the map=yes option so that readers can compare the maps used at each census. Perhaps we should do something like that for each city. --AussieLegend (talk) 02:19, 2 July 2012 (UTC)

Yes, but that is a restriction of the meaning of "official" to local government administration, and as you note that is a very narrow and inconsistent use of the term, most readers wouldn't be interested in. And is not that the urban area definition for cities wouldn't be used by officials, for instance in regional and urban planning, if one really needs to follow the "official" line. Furthermore I thought for being neutral, actually Wikipedia should rely more on science than politics. --ELEKHH^T 07:26, 2 July 2012 (UTC)

Part of an email I received today from the ABS: "The geographic boundaries that will give you overall census information for the cities you mention (the Urban Centre and Locality boundaries) have yet to be completed, and are not due for release until October this year as part of the second release of Census data. This is in line with what has happened in the past. The Urban Centre and Locality boundaries (UCLs) use the Census data to define a town or city on the ground, and represent the ABS consideration of what comprises these centres. They were never going to be output with the first release data."

"[...] The SA4s are specifically designed for the release of Labour Force statistics, and define separate labour markets. These have been defined from the analysis of Journey To Work data from the 2006 Census. In the larger cities, such as Perth, the SA4s represent localised sub-labour regions, of which the SA4 of Mandurah is one."

So it seems we need to wait for the urban centre stuff. Orderinchaos 08:52, 2 July 2012 (UTC)

Replacing defunct geographical classifications

This page has been tardy in updating from the defunct and no longer released old geographical classification to the new one used by the ABS since 2011. I have so far updated the capital cities' SAs to the new GCCSAs, in order that the newest ABS figures can be used. I have not as of yet replaced the regional cities' SAs with the new SA4s. This change will be necessary though as the old SAs are no longer used by the ABS, so we will not be able to gain future statistics relating to these areas from the ABS. I have also removed the "urban centres" table, which has not been used since 2006, and as such provided useless statistics that will not be update-able in future. Updating to the ABS' new geography is of course unavoidable in order to cite the official stats and even merely to collect statistics from 2011 as this is how the areas have been redefined. Anyone willing to help with the (more time consuming) task of replacing the old regional city SA's with the new SA4's would be greatly appreciated. Saruman-the-white (talk) 03:15, 2 August 2012 (UTC)

If you bothered to look the Urban Centre/Locality data will be updated later in the year when it is released as part of the second release of the Census data. I've restored it and don't remove it again since it is updateable. Bidgee (talk) 07:50, 2 August 2012 (UTC)

My mistake, thanks. Saruman-the-white (talk) 12:30, 2 August 2012 (UTC)

Incidentally, you appear to have gotten confused between Statistical Districts and Statistical Divisions. The latter are SA4s. In some cases these are a very significant proportion of a state, not a metropolitan area as the SDs were. In fact, *none* of the SA4s represent a metropolitan area as far as I can see (apart from Gold Coast and Sunshine Coast). Where are you getting this information from? Orderinchaos 05:58, 3 August 2012 (UTC)

In speaking with the ABS on the phone, I'm not sure that they are actually going to produce a comparable product any more to Statistical Districts for non-capital cities. Looking at the WA cities, Bunbury includes three unrelated large towns significant distances away; Mandurah includes one (this creates a disparity which makes Mandurah look 10% smaller than Bunbury when it's in fact 30% larger); while Geraldton and Albany can't even be united under a single district (one has to add up four in one case and five in another). Looking around Australia, most pose a similar problem or require OR to interpret (Mackay and Townsville are SA3, Mount Iis SA2, Gold Coast and Sunshine Coast are SA4, Cairns is split between two SA3s and Toowoomba does not have a discrete unit - the SA3 *and* SA4 contain completely separate areas), and as the ABS indicated in their email that I reproduced above, the UC/L is the only product they have which tries to define metro areas. On looking at this, my opinion has become, sadly, to disband the first table for all except capital city areas as we otherwise end up comparing apples and oranges in almost any pair of cases imaginable, and comparison is at the root of why this table exists. Orderinchaos 06:09, 3 August 2012 (UTC)

The capital cities suffer from the same problems that you've discovered with Bunbury. See my comments above regarding Glen Alice and Catherine Hill Bay.[3] Sydney can't be compared to other places when it includes places that are clearly not in Sydney, like Glen Alice, Catherine Hill Bay and Gosford. About all you can do is compare cities within states based on that state's definition of a city. The only common denominator between states seems to be the LGAs and even then the comparison needs to include population densities. --AussieLegend (talk) 07:46, 3 August 2012 (UTC)

Given that in the new scheme some regional cities are listed under SA4, some under SA3, etc. it will clearly be a pain trying to reassemble the list based on the post-2011 scheme. The Greater Capital City Statistical Areas (roughly corresponding to a US CSA or European Metro Area) are however clearcut and self-explanatory though and should be used. Whether or not we personally think that an area should or should not be included in these divisions is immaterial (I happen to think that GC should be included in Brisbane's GCSSA, for example), and we can't do original research, but must simply use the official statistical areas published by the ABS based on their extensive research of labour markets. Aside from the GCCSAs which are easy to deal with and can have their own table, I suppose we could have seperate tables for SA4s, SA3s, SA2s, not to mention a final table for LGAs. This would probably be the most clear-cut way of doing it. Also, "cities" usually refers to "cities proper", ie. LGAs. What we have here is a list of LGAs, and a list of metro areas (GCCSAs, formerly statistical divisions) which are much more commonly used to gauge the population of cities (otherwise Sydney would be listed as having barely 100 thousand and being a tenth the size of Brisbane!) As such it would probably be more accurate to rename this article "list of populated areas in australia" or similar, as it includes LGAs (cities proper), GCSSAs (metro areas), etc. Saruman-the-white (talk) 12:56, 3 August 2012 (UTC)

Perth's contains Mandurah as well, which is actually *defined* to be outside Perth by the state government. The problem is that Townsville's SA3, Mount Isa's SA2 and Gold Coast's SA4 are directly comparable. If you tried to compare the same level for just those three you'd already be comparing very different entities indeed - in many cases SA4 is a massive and arbitrary regional area with only meaning to the ABS, and SA3 sometimes is, sometimes isn't. So the entire point of the comparison is lost. I really wish they'd kept the old statistical units, even if just as a grouping category for new ones. Orderinchaos 13:19, 3 August 2012 (UTC)

"We can't do original research, but must simply use the official statistical areas published by the ABS based on their extensive research of labour markets" - Really, that's contradictory because the ABS doesn't define city boundaries. That's left up to the states. The ABS just provides statistical data that doesn't necessarily match anything that's legislated. As for labour markets, please again refer to my earlier comments regarding Glen Alice and Catherine Hill Bay. This isn't a case of "personally think[ing] that an area should or should not be included in these divisions". Sydney is defined as this area and Glen Alice, Catherine Hill Bay, Gosford and a lot of other places are not in that area. It's OR to say that that this is Sydney. In fact it's just a projection by the gimp in the airless basement of the the ABS offices of what he thinks Sydney will one day be, regardless of any practical considerations. The ABS sets all boundaries on what is convenient to its data collection practices. For example, the red line in this diagram shows the gazetted borders of my town but the ABS only includes the shaded area in its definition of the town.[4] That doesn't mean that the ABS is correct.

"Also, "cities" usually refers to "cities proper", ie. LGAs." - That's correct in NSW, but not in WA and probably not correct in other states. Even in NSW the definition is deceptive because of legislation. For example, the originally declared city of Cessnock is actually Cessnock, New South Wales, but changes to the Local Government Act resulted in small towns and hamlets in the LGA that were never part of the city, and which are up to 70km from Cessnock itself, being included in the new city boundaries literally with the stroke of a pen. --AussieLegend (talk) 14:06, 3 August 2012 (UTC)

Agreed with AL. As long as we are very rigid about the use of state definitions, and have rigid state definitions to go from (I know Perth has the MRS and Melbourne has M2030, and I've seen one for Brisbane in planning documents, haven't bothered to research anywhere else) WP:CALC tells us that one way to compare is simply to use ABS to tell us how many people are in those defined areas. But it's unfortunate that we should have to. The approach thus detailed, btw, is exactly what two WA government departments are doing for Perth and the five regional cities.

Of interest also is this quote from an ABS email: "The SA4 boundaries that you mention are not designed to represent "metropolitan areas" - in fact, the ABS does not define metropolitan." Orderinchaos 14:42, 3 August 2012 (UTC)

So you want to use separate state definitions for each city, which may exist for some states and not others? This seems pretty bizarre for at least the capital cities, given that the figures that are always cited and generally accepted are the official ABS figures and areas. I agree that it is more complicated when it comes to regional cities but using boundary definitions from some planning documents of various ages originating from state governments (not to mention the fact that you will not accurately be able to work out the current and up-to-date populations, nor census populations, of these areas since this is not how the ABS defines them) seems very inconsistent. The ABS defines the capital cities in a particular way, this is how the census data and official projections are released, and these are the figures which are commonly used and accepted. For at least the capital cities, the ABS definitions should be used. For the regional cities, I'm sure the SA3's/SA4's/SA2's which correspond can be ranked in a single list. We don't want to end up erratically using separate definitions for each state from random planning documents, relating to areas for which statistics are not released by the ABS and which are not commonly accepted. Saruman-the-white (talk) 16:51, 3 August 2012 (UTC)

While the figures may be commonly accepted, they're commonly accepted because people don't know any better and generally don't care. They don't realise that the ABS figures don't actually represent the area that they're being attributed to. A quick tally of the populations of all of the LGAs that are actually part of Sydney results in a figure that is about 600,000 less than the ABS figure for Sydney. The most recent figures for the Hunter Region, ie Hunter Valley exc Newcastle + Newcastle show the entire region has a population of of 398,983, (the 11 LGAs have a combined population of 620,000) but the population of Newcastle in this article is 540,002. We should be giving readers the facts without perpetuating misconceptions. --AussieLegend (talk) 17:44, 3 August 2012 (UTC)

We've been placed in an odd situation by the new census that the ABS no longer defines cities, but labour markets (!). This was not the case in 1976-2006. A labour market is not a city, and can be influenced by all sorts of things that have nothing to do with metropolitan areas. "Various ages" is not true at all, these things are fixed by governments - for instance Perth's MRS, despite the massive growth of the city, has survived unchanged from 1955 onward. Orderinchaos 20:49, 3 August 2012 (UTC)

I've just tried now to get populations for the cities listed in the table. Only 10 seem to correspond with a SA2 (1), SA3 (7) or SA4 (2); four can't be measured at all (Coffs Harbour, Gladstone, Rockhampton and Devonport-Burnie). Most are measurable by either gluing together SA2s, or taking an SA3 and taking a couple of SA2s away. While it is comparing apples with apples, finally, it is very hard work and rather unlikely to scrape in under WP:OR. The regions just can't be used by themselves. I do intend on writing (again) to the ABS about this. Orderinchaos 21:45, 3 August 2012 (UTC)

AussieLegend who are you to say which LGAs are "actually part of Sydney"? Doing this or using definitions which may or may not exist in a particular state and then trying to tally up the populations for these areas yourself by adding areas together (it will often not be as clear cut as merely tallying LGAs but tallying some pieces or LGAs or single suburbs) constitutes OR. Whatsmore, no in-depth stats will be available for these areas making it impossible to write about the demographics of our cities on their respective articles. I recall seeing "definitions" of Brisbane for the purpose of a planning document in one of these large state govt future urban development planning documents. It included ONLY the LGA of Brisbane City, leaving out huge swathes of the metro area such as Logan and the South-Western suburbs which have been considered part of Brisbane for a very, very long time. This was done because it is easier to co-ordinate localised planning and development by LGA rather than forcing them to cooperate for the issues which the state gov has no control over I suppose. Anyhow, a useless definition no less subjective (in fact, far more subjective and not used by anyone) is come up with. When it comes to Syd, Melb and Perth, etc, piecing together small areas to try and calculate total numbers based on state planning docs will be much more complicated, not to mention to the fact that stats are not released for these areas and these figures are never quoted. The point of wikipedia is not to try and change things to what we think they should be but to present the information that is commonly used and accepted. I guess we will have to wait for the ABS to reply to Orderinchaos, however until then, the capital cities at least, are clearly defined and the definitions are only very minor variations on the definitions which have been used since 1976 and always used here (thus minimal change in population when moving over to the new scheme). The issue at hand is not the capital cities but rather the fact that some of the regional cities do not correspond to an SA4, 3 or 2. An alternative would be to merely have one table for the most populated GCSSAs, SAs, one for the most populated urban centres (being released in October and solving the regional cities problem) and one for LGAs. Saruman-the-white (talk) 01:56, 4 August 2012 (UTC)

"who are you to say which LGAs are "actually part of Sydney"?" - I guess you missed the numerous references to File:Sydney councils.png which shows all the Sydney LGAs, and which reflects Local government areas of New South Wales#Sydney, which lists the Sydney LGAs. I've previously confirmed these against various official documents to confirm the accuracy so no, it's not me, it's the reliable sources that Wikipedia reflects that determine the Sydney boundaries. You won't find any reliable sources that confirm Glen Alice, Gosford or Catherine Hill Bay are part of Sydney. Only the ABS seems to believe that.

"trying to tally up the populations for these areas yourself by adding areas together (it will often not be as clear cut as merely tallying LGAs but tallying some pieces or LGAs or single suburbs) constitutes OR." - Firstly, WP:CALC allows us to do that because in NSW the borders of cities are legally defined by the LGA boundaries. I've addressed this previously.[5] Because the city boundaries follow the LGA boundaries, it's a simple matter of combining the LGA populations. "Tallying some pieces or LGAs or single suburbs"[sic] doesn't even come into it, as all of the LGAs in File:Sydney councils.png and Local government areas of New South Wales#Sydney are fully inside the area that the ABS defines as being Sydney.

"it is easier to co-ordinate localised planning and development by LGA rather than forcing them to cooperate for the issues which the state gov has no control over I suppose" - Not applicable to NSW. In NSW LGAs and the local councils that manage them exist at the state government's pleasure; they can be resized or disbanded as the state government desires. This works well for Sydney, where multiple LGAs have to be administered and where major projects such as motorways cross multiple LGAs.

"The point of wikipedia is not to try and change things to what we think they should be but to present the information that is commonly used and accepted." - Ummm, no. Wikipedia is an encyclopaedia and presents facts neutrally regardless of what may or may not be commonly accepted. There are enough reliable sources to confirm that the cities of Lithgow, Gosford and Lake Macquarie are not part of Sydney so including their populations, or even parts of their populations, in the Sydney population is quite ridiculous. --AussieLegend (talk) 04:26, 4 August 2012 (UTC)

Are you trying to imply that the ABS is not a reliable source? We are dealing with statistics here. Population statistics. The ABS happens to be the nation's official gatherer and publisher of statistics. It is the premier, and only widely used source for doing so in Australia. I just cannot understand why you would try tacking together LGAs to comprise Greater Sydney when it has already been defined for us by the national bureau of statistics in a form which is almost exactly the same (from memory, with the change of boundaries in 2011, the population of Sydney increased by 79 people) as the one which has been used in statistical reporting since 1976. Saruman-the-white (talk) 05:26, 4 August 2012 (UTC)

Very simple - our readers are just that, not geeks. We like to not mislead them. The ABS *themselves* have said that their definition has nothing to do with metropolitan areas, and if you look up in this very thread, you can see the quote from them about that. It's not like we're calling them out on something they're claiming to do - they have specifically said they don't do it. We would very much *like* them to do it, which is why I am writing to the ABS. Orderinchaos 06:12, 4 August 2012 (UTC)

Agree with OIC. The ABS doesn't have the responsibility or the authority to define the boundaries of Sydney; that is the responsibility of the NSW state government, which has decided that Sydney comprises the 38 LGAs shown in File:Sydney councils.png and Local government areas of New South Wales#Sydney. Labour markets, which the ABS claims to use, do not define city boundaries, so using them to support population figures makes no sense. --AussieLegend (talk) 06:54, 4 August 2012 (UTC)

Why did you leave it as it was for the last decade, in that case, as the previous capital city SDs are also based on labour markets and are hardly changed at all in terms of population. Given that these are the figures that everyone in Australia is used to and quotes (for the last 40 years) if you ask them the population of x city, it will seem quite odd for Wikipedia to be giving contrary information to that which is received everywhere else, with Sydney suddenly having as few people as Melbourne for example! Most people will simply dismiss it. Saruman-the-white (talk) 07:16, 4 August 2012 (UTC)

Very simple. The statistical districts reflected metropolitan boundaries. That category has now been completely abolished, for reasons that aren't completely clear, and replaced by new categories that require OR and a herculean effort by editors in some cases to stitch together. The specific issue with Sydney I'm unsure about as to previous use - however, the previous definitions of Melbourne, Brisbane, Adelaide and Perth matched exactly with state definitions, and now don't for reasons which aren't completely clear. In short, we have been chucked into a situation not of our own making, and finding our way out isn't as simple as pretending black is white or that entire rural sections of a state are in fact part of a distant "city". Unless the ABS come to the table on SDs, it is quite likely that the first table in this article will have to be removed and the October 2012 release of data on urban centres (which is entirely inadequate for basic comparison purposes - large parts of the metro area of Perth are not within the UC/L for example) will be our only meaningful base for comparison. As I've already said a few times, I really hope we can convince the ABS to add a layer, even if it's just the 2006 SD layer or an adaptation thereof. Orderinchaos 07:39, 4 August 2012 (UTC)

Sydney has always been a victim of "We use SDs for capital cities", despite the SD not representing the actual city. --AussieLegend (talk) 07:47, 4 August 2012 (UTC)

And on adding up the 2011 census populations, yes, Melb and Sydney are about equal, at around 3.9 million apiece. This statement in the Regional Population Growth summary for 2010-11 probably indicates why: "Melbourne had the largest growth of all Australian capital city SDs in 2010-11, as it has had for the last ten years." Faster growth + relatively small difference in population = inevitable catchup. Orderinchaos 07:59, 4 August 2012 (UTC)

Errors in Urban Centre Listing

Some of the urban centre data is incorrect for 2011. For example, see Darwin, where the LGA is used, as well as Palmerstaon. Palmerston is in the Darwin urban centre. List needs a good edit. — Preceding unsigned comment added by 75.132.161.1 (talk • contribs) 23:28, 2 January 2013‎ (UTC)

Thanks for identifying the problem with Darwin–Palmerston. This has now been corrected. If you have any other specific errors, please comment directly here; instead of a generalised comment that "Some of the urban centre data is incorrect...". Thanks Rangasyd (talk) 05:03, 4 January 2013 (UTC)

A waste of wiki space

Waste of wiki space folks - its just serves to offend people who live in these areas to see the errors herein and confusion it creates - just delete the page contents and instead put links to the original source, the ABS website, where the data will be current and the statistical codes explained. — Preceding unsigned comment added by 106.68.171.131 (talk) 11:38, 26 November 2013 (UTC)

Queanbeyan "city"

Queanbeyan is addressed as a "city" in its own article, and given a population rank of #37. It links to this article, strangely, though, because the lists only has 'Canberra-Queanbeyan'. Is there (or have there been) a list where a separate entry of Queanbeyan is? 82.141.67.208 (talk) 23:15, 17 December 2013 (UTC)

There were errors in the Queanbeyan article. The population had been updated incorrectly in October 2013, which I've now fixed. "37th" was a carry over from the 2006 census and should have been removed, as it is not listed in this article other than as part of Canberra-Queanbeyan. --AussieLegend (✉) 08:57, 18 December 2013 (UTC)

And I just noticed that Dubbo article claims to use the same 2011 data, and also it has the rank #37... It has over 5,000 less of population, though. 82.141.67.208 (talk) 22:53, 18 December 2013 (UTC)

I checked this article's version history, and in first version, there is section "Capital city Statistical Divisions and Statistical Districts by population" which has Canberra as separate entry and Canberra-Queanbeyan unranked and marked with *. To answer to myself: apparently there never were a list Queanbeyan was on its own. 82.141.67.208 (talk) 01:44, 19 December 2013 (UTC)

Ipswich, Queensland

Someone wrote to OTRS alerting us to the fact that the above is missing from this list. I figured it was a straightforward add to the table(s), but looking at the inclusion criteria here I'm not so sure. So instead I'm just passing along the request to be actioned if merited and appropriate. Thanks! §FreeRangeFrog^croak 22:38, 6 September 2014 (UTC)

Hi, Ipswich is certainly a Local Government Area (a council area) in its own right, but it is included in the Brisbane GCCSA (metro area) and the Brisbane urban centre by the Australian Bureau of Statistics. I realise the area around Ipswich, like many other suburban areas in Brisbane and Australia's other major cities has something of its own identity but it is officially very much a part of Brisbane's metro and urban areas, and has been for decades. Australia's major cities are however made up of many, many 'Local Government Areas' (councils) which range from one suburb to many. There is an LGA of Ipswich which is included in the LGA list.--Saruman-the-white (talk) 00:52, 8 September 2014 (UTC)

By refering to Wikipedia's own page on the LGA/City of Ipswich which states it has a population of 177,323 based on 2011 Census, it should be listed in the section titled '30 largest local government areas by population', at Number 23 between Penrith & Townsville. As this is the LGA population section and not the ABS defined metropolitan regions (thus Logan & Moreton Bay showing up in the LGA list), Ipswich is a glaring omission. User:ChenBot (talk) — Preceding unsigned comment added by 203.35.135.165 (talk) 04:57, 21 July 2016 (UTC)

GCCSA/SUA

Could it be more explicitly spelled out on the page what the component building blocks are for GCCSA/SUA? Are each of these a collection of local government areas into which the contiguous urban area extends? This is a very basic, but important piece of information for those reading and wondering what constitutes a "city" in the Australian sense of the word. The "urban centres" and "local government area" definitions are easy enough to understand because they are explained fairly well. It's the GCCSA/SUA which gives no component building block definition from what I can tell. --Criticalthinker (talk) 08:57, 31 January 2015 (UTC)

Wait, what little I can find is that "cities" aren't a collection of local government areas but of what would amount in America to something like a Census tract? Someone help me with this, and then find out a way to make it more clear in the definition on the page. --Criticalthinker (talk) 06:48, 3 February 2015 (UTC)

The definition of city varies from state to state. In New South Wales, a city's boundaries are determined by the local government area (LGA) boundaries. If a population centre within the LGA becomes large enough to reach city status, the whole LGA becomes the city. For example, if Raymond Terrace reached city status, the city boundaries would not be those of Raymond Terrace, they'd follow the boundaries of the Port Stephens Council LGA and places as far afield as Fingal Bay (45km to the east) and Duns Creek (a rural suburb about 30km to the north-west) would become part of the city of Port Stephens. This has already happened elsewhere. Cessnock was declared a city in the early part of last century, but the city includes small hamlets consisting of a few houses on the side of a country road within its boundaries. Sydney is an exception to the rule. Its boundaries include multiple LGAs, as seen in File:Sydney councils.png. Other states have different rules. and then we have the Australian Bureau of Statistics (ABS), whose definitions are consistent across the country but often mean very little. For example, Newcastle is a city of 148,535 people, but one of the ABS definitions of "Newcastle" includes most of the City of Newcastle LGA, much of the City of Lake Macquarie and a bit of the Port Stephens Council LGA. Another ABS definition extends this to cover most of the Lower Hunter Region, or what is commonly called Greater Newcastle. The Sydney ABS definition has some ridiculous boundaries, as I mentioned in this discussion. When it comes to the ABS definitions, they can be difficult to evaluate. For example, the definition of GCCSA is "The GCCSAs represent the socioeconomic extent of each of the eight state and territory capital cities. This provides a stable and relevant geographic definition for the release of socioeconomic survey data collected only within capital cities as well as other survey data requiring large population output regions. Within each state and territory, the area not defined as being part of the greater capital city is represented by a rest of state region." I'm not sure how to write that better. --AussieLegend (✉) 11:15, 3 February 2015 (UTC)

Of course, one would use ABS definitions since they are consistent across the country. I'm still not getting what the base components are the for the ABS are called for building "cities" in Australia, that's all I'm asking and asking to be made clear. If these are statistical measurements and not local government areas (apart from when a settlement becomes large enough where the entire local government area is included) what are these base statistical divisions called that make up the larger statsitical divisions? --Criticalthinker (talk) 19:29, 3 February 2015 (UTC)

Unfortunately, the ABS doesn't seem to make it clear what they actually use to determine the "socioeconomic extent of each" blah blah. Prior to the 2011 census everything was reasonably clear but the 2011 census made everything a bit hazy. We know that a GCCSA represents a capital city, even if it's a little meaningless, but when referring to anything smaller, it's necessary to find the acronym that best matches the map of the location that you're searching for. --AussieLegend (✉) 19:45, 3 February 2015 (UTC)

Doing a bit more research, I was able to find a few links explaining components of "cities". In the case of GCCSAs, they are made up of whole Statistical Areas Level 4 (SA4s).

http://www.abs.gov.au/websitedbs/D3310114.nsf/4a256353001af3ed4b2562bb00121564/6b6e07234c98365aca25792d0010d730/$FILE/Greater%20Capital%20City%20Statistical%20Area%20-%20Fact%20Sheet.pdf

The SA4s are designed to reflect labor markets, which used travel to work data from 2006. They say that the introduction of the GCCSAs is more consistent than former definitions of metropolitan cities, though, they definitely define each of these cities more broadly than the former definitions. To my American ear, it sounds like the difference between MSAs and CSAs.

http://www.abs.gov.au/websitedbs/D3310114.nsf/home/Australian+Statistical+Geography+Standard+%28ASGS%29

This second link is something I'll have to delve more into as it describes the other metropolitan cities. Maybe others can read through them. At the end of the day, if anyone can find a better way to word definitions for GCCSAs and SUAs on the mainpage, that would be great. I do think the ABS really messed this up, though. It appears that GCCSAs and SUAs are not directly comparable, which maybe means that they should be on different lists to avoid confusion. I guess what I'm saying is that simply bolding the capital cities to show them as different might not be as clear a diffentiation as is needed. At first blush, it appears that using different components that the ABS exaggerates the size of the large capital cities relative to the other cities of the country. --Criticalthinker (talk) 19:55, 5 February 2015 (UTC)

"At first blush, it appears that using different components that the ABS exaggerates the size of the large capital cities relative to the other cities of the country." Indeed it does, as well as to each other. For example, Melbourne is actually larger than Sydney, but Sydney's size is exaggerated by including places like Glen Alice, which is 128 km (80 mi) as the crow flies and 223 km (139 mi) by road from the Sydney CBD and could not possibly be considered to be part of Sydney. --AussieLegend (✉) 12:50, 6 February 2015 (UTC)

See, this is where I kind of have to take issue with some of the whining on this page, particularly of the heated Sydney vs. Melbourne rivalry that I can only imagine must be contentious in Australia to save the least. I guess it could be debate whether the ABS changed the way they measured the capital cities to favor one city over the other, but I don't think it can be argued that measurements of the GCCSA category of cities aren't consistent acrosss/within that definition; it's still apples to apples. It's not as if they made Sydney a GCCSA and made SUA. Does that make sense? --Criticalthinker (talk) 16:17, 6 February 2015 (UTC)

For the record, I hate Melbourne, but that has nothing to do with what I've presented. When you look at the actual populations of the cities, the ABS populations just don't make sense. The same is applicable to Newcastle, where I was born. The ABS figures put Newcastle as the seventh largest city in Australia, but it's actually smaller, both in population and area than adjacent Lake Macquarie. The ABS has said it doesn't report populations of actual places, and that's borne out in the population data. It covers statistical areas, but I sometimes wonder who decides on these areas. Unfortunately, we seem to be stuck with them. --AussieLegend (✉) 16:29, 6 February 2015 (UTC)

But, see, that's defining "city" as a local government area. As a settlement, "Lake Macquarie" doesn't seem to exist as physically seperate from metropolitan Newcastle, so it can't be a bigger "city" because it's not a city beyound just happening to have a local government area. In fact, some of the areas within this "City" are actually part of the statistical/metropolitan Newcastle city. I admit that you guys have some confusing definitions, but I do find them generally consistent. I have to say that "city" is a fairly simple concept here in America. "City" in America is a local government area with boundaries which generally corresponds to a single and historical city proper settlement in most cases, whether it be a rural settlement or urban one. In Australia, it doesn't seem that there is much rhyme or reason, often times, to the boundaries of LGAs. Sometimes, they include a single settlement and huge swaths of rural or even undeveloped land. Sometimes, they even seem to split a historical settlement in half and pare if off between two different local governments. Just generally looking over it, I wouldn't say that the metropolitan or statistical city of Newcastle seems to be unusually overdone relative to any other. --Criticalthinker (talk) 18:29, 6 February 2015 (UTC)

More research seems to yield an answer to my original question (http://www.abs.gov.au/ausstats/abs@.nsf/Products/3218.0~2011~Main+Features~Population+estimates+and+Australia%27s+new+statistical+geography?OpenDocument):

Statistical Areas Level 2 (SA2s) - are medium-sized general purpose areas which aim to represent communities that interact together socially and economically. SA2s are based on officially gazetted suburbs and localities. In urban areas SA2s largely conform to one or more whole suburbs, while in rural areas they generally define the functional zone of a regional centre. Statistical Areas Level 3 (SA3s) - are aggregations of whole SA2s and reflect a combination of widely recognised informal regions as well as administrative regions such as state government regions in rural areas and LGAs in urban areas. Statistical Areas Level 4 (SA4s) - are made up of whole SA3s and are designed to reflect labour markets. In rural areas, SA4s generally represent aggregations of small labour markets with socioeconomic connections or similar industry characteristics. Large regional city labour markets are generally defined by a single SA4. Within major metropolitan labour markets SA4s represent sub-labour markets. Greater Capital City Statistical Areas (GCCSAs) - are built from whole SA4s and represent a broad socioeconomic definition of each of the eight state and territory capital cities. They contain not only the urban area of the city, but also the surrounding and non-urban areas where much of the population has strong links to the capital city, through for example, commuting to work.

So, the basic building blocks for these cities are actually gazetted suburbs and localities which then go on to make up larger and larger statistical areas. Also, while I'm not sure if these are readily available or published, it looks like these statistical areas aren't actually new measurements - save for the GCCSAs - and have something to be directly compared to:

To assist users in the transition to the new geography, a set of historical population estimates based on the ASGS is included in this issue. Estimates have been provided at the SA2 level annually back to 1991, the SA3 and SA4 levels back to 1981, and the GCCSA level back to 1971. Feedback on these historical estimates can be emailed to regional.population@abs.gov.au.'

Interesting. I hope this helps others who have similar questions. --Criticalthinker (talk) 18:14, 6 February 2015 (UTC)

50 largest urban centres by population

According to the table, the four largest urban centres in NSW are, in order, Sydney, Gold Coast-Tweed Heads, Canberra-Queanbeyan, and Newcastle. But the only part of Gold Coast-Tweed Heads that is in NSW is Tweed Heads, which has a population of only 7,525. (And that assumes that Tweed Heads is part of the Gold Coast, which it isn't.) And the only part of Canberra-Queanbeyan that is in NSW is Queanbeyan, which has a population of 37,991. So the ordering in the table is ludicrous. Each should either be listed only in the state where most of the population lies, or split and listed separately. As presented here the information is misleading. — Preceding unsigned comment added by 125.255.16.233 (talk) 22:29, 28 February 2016 (UTC)

We can't and shouldn't split them as we are taking them as they are from the source. Not up to us to go splitting them. They are listed as Qld-NSW and ACT-NSW are they not? This is accurate. The state with the majority of population can go first. Albury-Wodonga is another one where the majority is in NSW and minority in Vic.--Saruman-the-white (talk) 23:31, 28 February 2016 (UTC)

Saruman-the-white is correct. Even if we were to try to split them, we couldn't, because we don't know the individual populations. The wikilinks to articles are just guides, we can't use the data from those articles, as they actually cover a different area to this one. For example this shows the map for the 2014 Gold Coast-Tweed Heads SUA. The Gold Coast article uses an outdated figure from 2010. None of the other areas available from the 2011 ABS QuickStats covers the same area as the SUA, even the SA4, which is most similar.[6] That means we'd be getting different population types from different areas. While I have very little faith in the data in this article, at least it's consistent, more or less. --AussieLegend (✉) 04:37, 29 February 2016 (UTC)

I am concerned that this data is mis-used to claim that "Tweed Heads is the sixth largest city in New South Wales". That article is about Tweed Heads only, not about the larger ABS statistical area, so it is not appropriate to call it "the sixth largest city" just because it is a small part of a region which has a larger population. I tried to corerct that in the article, but it was reverted by Doctorhawkes because of the entry here for "Gold Coast-Tweed Heads". Is there a guideline for how to handle these cases in articles about the individual towns that constitute these larger areas?--Gronk Oz (talk) 22:49, 13 November 2016 (UTC)

It's really just a case of using the correct figures for the particular area. Newcastle has the same issue. People keep referring to parts of Lake Macquarie as suburbs of Newcastle when they are not, the two cities just share a common border, so the populations are often lumped together to make ridiculous claims. Look through the figures available on Quickstats and pick the one that best represents Tweed Heads which is not a city despite the claim in the article. --AussieLegend (✉) 07:56, 14 November 2016 (UTC)

Thanks for that, AussieLegend - that has just raised a new question: is there a guideline for what is classed as a "city"?--Gronk Oz (talk) 11:17, 14 November 2016 (UTC)

Sadly, no, as it changes from state to state. In NSW a city is legally the area of the LGA, but this is different elsewhere. --AussieLegend (✉) 14:54, 14 November 2016 (UTC)

Ellenbrook

Why is Ellenbrook on the list of Greater Capital City Statistical Areas/Significant Urban Areas by population? Isn't it a suburb of Perth? If Ellenbrook is entitled to be listed separately, wouldn't Mandurah be a more appropriate entry? 2001:8003:8656:7600:AD72:5E43:C8E9:19F2 (talk) 10:04, 1 October 2017 (UTC)

Areas and densities

Not entirely sure about the inclusion of areas and densities in a population article. These can be misleading as (for example) the Brisbane CGGSA has an area of 15000 square km which is almost 3 times that of Perth or 5 times that of Adelaide. In reality of course, about 95% of that population is concentrated in about 2500 square km of that 15000 square km and urban Brisbane is vastly denser than urban Perth or Adelaide. Similarly, Sydney has a very large GCCSA area but Sydney is vastly denser than Melbourne (let alone Perth or Adelaide) — Preceding unsigned comment added by Saruman-the-white (talk • contribs) 06:44, 2 October 2017 (UTC)

I let it go because it didn't seem to do any harm but you do have valid points. Sydney's GCCSA extends to a ridiculous point beyond what any rational person would consider to be part of Sydney so the population density is not truly representative by any means. This is only supposed to be a list of populations and adding this extra information is beyond the scope of the article. I'm going to be bold and remove it. If the editor wishes to restore, he can argue the merits here. --AussieLegend (✉) 07:17, 2 October 2017 (UTC)

I would love to see areas and density come back. If using the GCCSA (sp?) does not make sense, perhaps there is another more reasonable data source? Here in Canada we have 2 articles that list population and land area together by municipality and population centre, and I think it would be useful to be able to compare cities on this metric. 19:23, 17 January 2023 (UTC) — Preceding unsigned comment added by 184.71.108.66 (talk)

Mildura

38,000 to 60,000 and 9% growth? One or more of these numbers is wrong. 14.0.230.63 (talk) 06:54, 18 September 2018 (UTC)

New ABS Statistics Released

The new Australian Bureau of Statistics dataset for population through to June 2019 has been released: https://www.abs.gov.au/AUSSTATS/abs@.nsf/Lookup/3218.0Main+Features12018-19?OpenDocument — Preceding unsigned comment added by Happypup398 (talk • contribs) 04:05, 28 March 2020 (UTC)

See my map, see my map, see it's hopef'ly not too crap

First off, great job to all on keeping this list up-to-date with the ABS's ever-changing and too-complicated different definitions of statistical areas. I've noticed in the years since I created this article nearly 15 years ago that there's never-ending debate and confusion about the different statistical concepts under the post-2011 ASGS, so I tried to illustrate them all on one map, using the greater Sydney area as an example. It's my hope this map could go in the article along with a clearer definition of GCCSAs, SUAs, and urban centres/localities.

As can be seen from the map, the GCCSA for Sydney (black line) is HUGE. It contains the Significant Urban Areas of both Sydney and the Central Coast, and these in turn (along with the other SUAs all marked orange) contain collections/agglomerations of the individual urban centres (in pink), which in Sydney's UC case is still large, including Richmond-Windsor and the lower Blue Mountains, but nowhere near the size of the GCCSA, which as has been pointed out in this talk section, includes vast rural regions.

One thing I've thought about this article recently was that perhaps we should be using the SUA figures in the list for the capital cities as well, as they represent much tighter-defined areas, in Sydney's case not including the Central Coast, which is always contentious. This would require revising the figures of only a few cities on the list. For completeness' sake you could have a 4th list, with the populations of only the GCCSAs. As it stands we've got an awkward inclusion of the GCCSA for each capital along with the SUA for every other city in the land, which I understand was a compromise when the ASGS came out, but has never really made sense to me, as they're defined very differently (labour market areas, vs agglomerations of urban centres less than 5 km apart sharing links).

What does everyone think? I haven't been on this site in 15 years so buggered if I can remember how to sign my name. ҉ Randwicked ҉ 09:35, 27 April 2020 (UTC) (no, I figured it out)

Maybe a list at the top for GCCSAs only.then one for SUAs then one for LGAs. I agree including Central Coast in Sydney is controversial but at the same time GCCSAs are far and away the premier and primary area that the ABS uses for the capital cities with all the capital city population stats, annual and quarterly regional population statistics publications, projections etc released for GCCSAs. So for the sake of getting updated and the best data by definition the capital cities have to have a table for GCCSAs primarily, as they are the ABS definitive measure for capital city statistics. After that can be a table of all SUAs incl capital and non capital.StormcrowMithrandir 08:40, 27 April 2020 (UTC)

Regarding the black lines, why are there so many disjointed lines in peculiar places that aren't Sydney? I'm also concerned about apparently randomly selected areas like Brandy Hill, a small suburb in Port Stephens Council. --AussieLegend (✉) 09:43, 27 April 2020 (UTC)

Thanks for the comments both!Stormcrow, I agree this would be the best solution, retaining a list for the GCCSA, cos as you point out they are regarded as primary (even by the wikipedia articles for each capital city). AussieLegend, the black lines are showing through in places like the coast because they also form the coastline. They just show there because there aren't overlaid boundaries. It might look better if I changed the colour of the black lines to grey. Brandy Hill is labeled because it's a separate urban centre/locality. (The labeling placement is random, done by the GIS software.) - ҉ Randwicked ҉ 00:00, 28 April 2020 (UTC)

I thought the examples might labelled for that reason but it looks weird, Brandy Hill especially because it's just a rural locality - no shops, no urban etc, just farms and large rural residential blocks. --AussieLegend (✉) 15:16, 28 April 2020 (UTC)

Thanks for the further feedback. What might look better would be if I just removed the labels for the individual urban centres. It would reduce the general clutter of the map, and it's obvious that the biggest urban centres are the relevant cities. I'll try revising it tonight. - ҉ Randwicked ҉ 04:51, 29 April 2020 (UTC)

And edited! I got rid of the tiny urban centre/locality labels, enlarged the other unit labels slightly, made the GCCSA/coastal outline grey to be less prominent, and turned the solid shaded LGAs into outlines for better legibility. - ҉ Randwicked ҉ 10:47, 30 April 2020 (UTC)

Definitely an improvement, although I did like the colouring in the previous version. --AussieLegend (✉) 13:43, 30 April 2020 (UTC)

Thanks! I've tweaked it slightly once more. Made the labels bigger and higher contrast so they'd be more legible in a thumbnail, and other minor changes. - ҉ Randwicked ҉ 13:33, 3 May 2020 (UTC)

seems there's been no more feedback the past few weeks, so I might look at expanding on this in the article this weekend, godwilling I have time. ҉ Randwicked ҉ 11:26, 29 May 2020 (UTC)

Going to revise this article as discussed a few months ago, and noticed some errors

E.g. we've still got Mandurah listed as a separate Urban Centre, whereas all the ABS sources I've checked seem to say it merged into Perth's UC from the 2016 census. The next one down would be Alice Springs, if I'm reading the data cubes correctly, but it'll require review of the other entries and units to make sure they actually still exist. Will have a look at it all over the next few days. ҉ Randwicked ҉ 06:45, 31 October 2020 (UTC)

I've updated the first section, splitting the GCCSAs into a new table as I originally proposed, given their difference in definition and construction from SUAs. All the units have their formal titles. A couple of units (e.g. the Central Coast, Melton) have appeared from under the shadow of their capital cities. Yanchep appears to be a brand new SUA. There are 101 SUAs now, with populations updated to June 2019. ҉ Randwicked ҉ 09:00, 31 October 2020 (UTC)

I'm going to try to tackle the messy Urban Centres/Localities section next. We really should try to get it in order well before the 2021 census happens and new UCs are generated. ҉ Randwicked ҉ 08:28, 14 February 2021 (UTC)

GCCSA vs. SUA

Was this page recently changed? I didn't realize that GCCSA and SUA used different measurements, rather I thought the GCCSA were just the formal names for the capital SUAs. It'd be nice if there can be some explanation in the paragraphs for both as to what the base units are for measurements. The only thing I can compare it to is that GCCSA are kind of equivalent to the United State's CSA and the the SUAs are closer to United States MSAs? As in that both American measurements use counties as the base, but different levels and calculations of commuters to determine these areas. Do GCCSAs and SUAs use the name building blocks but maybe different populations thresholds for said building blocks, or are the two vastly different measurements?

Simpler way to ask this I guess it: Why and how is the Sydney GCCSA different than the Sydney SUA, for instance? --Criticalthinker (talk) 23:25, 27 December 2020 (UTC)

Hi Criticalthinker, I added a 'definitions' section, along with a map, a few months ago, to try to explain the ABS's multitude of confusing units. Essentially, when they changed their statistical geography in 2011, everything became based on 'SA units', SA 1/2/3/4, rather than on the old local governmental boundaries. Why, heaven knows, but now everything is made of these base units. GCCSAs exist only for the capital cities, are built from SA4s (labour market areas), and more or less equal the old capital city statistical divisions pre-2011. At the same time they created a unit called 'SUA' out of the SA2 units (much smaller units representing one or two suburbs), to replace the old statistical areas for the towns and cities, but they created these ALSO for the capital cities, so now each capital city has THREE different population figures (GCCSA, SUA, urban centre), or four if you count its central LGA ('city proper' elsewhere) as well. Hopefully my map clears it up a bit. Sometimes it seems the ABS's primary goal is confusing people. ҉ Randwicked ҉ 02:42, 28 December 2020 (UTC)

Okay, I scrolled down to see it. I guess my suggestion is that the definitions come first in the article so you're not just tossing people into a bowl of different terms they don't know. But that's up to you all. --Criticalthinker (talk) 04:23, 28 December 2020 (UTC)

Melbourne's Significant Urban Area population surpasses Sydney's

and this goes completely uncommented on in the media, probably because the ABS's conflicting population units are an unintelligible morass and no-one really gets them. LMFAO. Anyway, I've updated the SUA list from the lastest Regional Population figures released today. Will continue the revisions as I get a chance. ҉ Randwicked ҉ 11:41, 30 March 2021 (UTC)

tl;dr: Sydney's literally only the bigger city because the ABS counts the Central Coast in its population now. ҉ Randwicked ҉ 11:43, 30 March 2021 (UTC)

Confused about how the general population views what their cities are

This is one of my favorite pages to visit. However, I have recently been getting more and more confused about how we define cities in Australia. Each official state government does not reference the Gold Coast, Wollongong or Victor Harbor as anything other than centres. I know that this page uses ABS definitions, however does the general population (who know nothing about what the ABS does and define) consider these urban areas or UCL as cities? When they do, are they taking into account the greater urban land space or just the original core town?

More importantly, how do the state governments use the ABS definitions ?

If the states or general population do not define the cities based on ABS figures, why does Wikipedia not use what the official authorities do? — Preceding unsigned comment added by Sdinesh2222 (talk • contribs) 23:36, 29 August 2021 (UTC)

@Sdinesh2222: I am not at all surprised that you are confused. Wikipedia decided to use the ABS figures because they are generally consistent across the country while the definition of cities varies from state to state and even within states there can be confusion. For example, in NSW (where I live), city borders consistently follow LGA boundaries which you would think would be best. However, that's not the case. Cessnock became a city when it was a well populated mining centre but the population has decreased now to about 15,000 and it's now just a large town. BUT, once somewhere has become a city it never loses that status. Instead, the City of Cessnock LGA boundaries now include various towns, suburbs and even small groups of rural houses outside of Cessnock proper bringing the population back up to more than 55,000. The ABS definition of Sydney includes one rural place 145km from the Sydney CBD while the generally accepted borders of Sydney stop far short of that. In the recent lock-downs, the NSW government included suburbs in the central coast (not part of Sydney!) as part of Greater Sydney. The ABS definition of Newcastle has included parts of adjacent Lake Macquarie and Port Stephens which is why Newcastle, New South Wales doesn't just talk about actual Newcastle. These are but a few examples but they show the inconsistencies that exist. Regarding Wollongong and the Gold Coast, these are both officially cities. I'm sorry I can't give you a better answer, but I don't think there is one. I'd also note that the ABS is a statistical body and does not necessarily follow state definitions of places. --AussieLegend (✉) 02:40, 29 August 2021 (UTC)

Cities and Metro areas

Recently had this debate on the Brisbane talk page, and settled on the fact that cities in Australia are ranked by the metropolitan populations and NOT city lgas. To put this bluntly the City of Melbourne is not 5 million people but about 150k, therefore this page should be retitled to reflect that as the term "City" in Australia if often used to define LGAs, whereas this article is not listing cities primarily by their City LGAs but rather metropolitan areas. The article should be retitled to reflect that

List of cities in Australia by metropolitan population--Caltraser5 (talk) 04:14, 16 January 2022 (UTC)

I've removed the template you added because it is used for section renames, not for renaming whole articles. Your proposed name is not appropriate because not all LGAs are cities, and they are listed in this article. To be fair, "List of cities" is not really correct either, because non-cities are listed. --AussieLegend (✉) 13:36, 17 January 2022 (UTC)

The trouble with this is that if you ask most Aussies to describe a city they'll describe the whole urban area, not the central LGA. It's just the way we define things. This article uses Australian English, and the title is correct in Australian English. IMO. Standard caveats re: a particular city's mileage may vary, etc. etc. ҉ Randwicked ҉ 07:52, 20 January 2022 (UTC)

What about different Urban areas within the single Greater Capital City area? For example are Melton or Central Coast considered cities of their own or are they ignored and absorbed under the Capital city  ? 14.201.220.44 (talk) 13:55, 13 May 2022 (UTC)

2021 estimates

the ABS has released the Regional Population Growth series for 2021. Interesting is Sydney pulling ahead of Melbourne again (in SUA population), mainly due to Melbourne losing over 60,000 people net in emigration during the pandemic. I will update things in a day or so. - ҉ Randwicked ҉ 11:59, 29 March 2022 (UTC)

Do UCL figures get taken into account for anything anymore or is it ignored if a Urban Centre is part of a SUA already  ? Sdinesh2222 (talk) 08:54, 30 April 2022 (UTC)

They are still released, but are calculated for the census only (i.e. every five years). The updated UCLs won't be published until later this year, see https://www.abs.gov.au/statistics/standards/australian-statistical-geography-standard-asgs-edition-3/latest-release#release-schedule. ҉ Randwicked ҉ 06:45, 1 May 2022 (UTC)

Sounds like another update is coming as the ABS has updated the SUA definitions. Melbourne SUA back in front of Sydney SUA - maybe the missing population moved to Melton? https://www.smh.com.au/national/melbourne-tops-sydney-as-australia-s-biggest-city-on-a-technicality-20230413-p5d04g.html Tml au (talk) 12:31, 17 April 2023 (UTC)

Sydney's Greater Cities Commission 6 region plan

Am curious if this will be reflected in future GCCSA counts in that Newcastle and the Gong will now start getting included?

https://greatercities.au/news/greater-cities-commission-to-deliver-six-cities-region — Preceding unsigned comment added by Sdinesh2222 (talk • contribs) 21:17, 14 May 2022 (UTC)

good question. I sure hope not, as that would just be excessive, but it'd be a matter for the ABS. I understand they developed the GCCSA units with the intention not to have to alter them unless really necessary. - ҉ Randwicked ҉ 06:37, 15 May 2022 (UTC)

Urban centre 2021 data

Is there any reason that the data on urban centres on this page only includes data up to 2016 and doesn't include 2021 cencus data? 59.100.112.44 (talk) 23:28, 25 April 2023 (UTC)

Because no one bothers to update the list. 2001:8003:900C:5301:18B6:F60A:BA46:2339 (talk) 13:34, 3 August 2023 (UTC)

Melbourne has overtaken Sydney's population!

Melbourne has overtaken Sydney's population! 131.172.226.227 (talk) 18:13, 1 May 2023 (UTC)

So? It would never catch Sydney in house price. 2001:8003:900C:5301:18B6:F60A:BA46:2339 (talk) 13:35, 3 August 2023 (UTC)

Semi-protected edit request on 23 August 2023

Wrongname123 (talk) 11:20, 23 August 2023 (UTC)Melbourne has overtaken Sydney as Australia's most populous city for the first time since the 19th Century gold rush, following a boundary change.

Sydney has proudly held the title for more than 100 years.

But with populations rapidly growing on Melbourne's fringe, the city limits have been expanded to include the area of Melton.

The latest government figures, from June 2021, put Melbourne's population at 4,875,400 - 18,700 more than Sydney.

The Australian Bureau of Statistics (ABS) defines a city's "significant urban area", by including all connecting suburbs with more than 10,000 people. Source BBC

 Not done: it's not clear what changes you want to be made. Please mention the specific changes in a "change X to Y" format and provide a reliable source if appropriate. — Paper9oll (🔔 • 📝) 14:33, 23 August 2023 (UTC)

This is way out of date

All of this data is nearly ten years old and needs an update 180.150.101.170 (talk) 00:59, 29 June 2024 (UTC)