|Rank||Country||Population (Millions)||Area km2||Density pop/km2|
Source: Google "Public Data" — "World development ..." — Environment — Population Density — Choose country.
This chart shows the density for the top and bottom Counties, Authorities and Towns in the UK.
|Top 10||All Land||Bottom 10||All Land|
|Southampton||4733||Isles of Scilly||134|
Population density is the total number of people residing in a country on a particular date divided by the land area.
The land area is the total land mass of a country less any water over 1sq km. This is an EU directive. It actually makes little sense as it includes mountains, marshes, roads etc. which are not usable for housing. A good example is Netherlands as it is smaller than Switzerland but has more than twice the population density. This is because the Netherlands is flat and Switzerland is one big mountain.
Land Usage Surveys break down land usage into categories so it should be possible to quantify how much land is used to house people together with their gardens. The resulting land area would not include lands which would be impractical to live on such as mountains, rivers, defense, roads, offices, factories, farmland, woodlands, heaths, marshes etc.
No sources for the UK and Regions give the Urban Land Usage directly; the % data below comes from the Countryside Survey 2007, it is for “Built-up and Gardens” and “Unsurveyed urban land”
|UK||63,181,800||242,509||12.1 4 29,101||2,172|
|England||53,012,500||130,279||16.5 5 21,235||2,496|
|Scotland||5,295,000||77,932||6.3 3 4,910||1,074|
|Wales||3,063,500||20,735||6.9 6 1,431||2,141|
|N Ireland||1,810,900||13,562||10.5 7 1,424||1272|
1. Population Census 2011 from ONS
2. UK land coverage 2007 from ONS
3. Data for Scotland, 80% live on 5% of land from SNH (6.3 = 5/0.8) Not available.
4. UK Urban usage is calculated from the 4 regions.
5. England. Built-up Urban & Gardens + Unsurveyed Urban land Countryside Survey 2007 England
6. Wales. Built-up Urban & Gardens + Unsurveyed Urban land. Countryside Survey 2007 Wales
7. Northern Ireland. Built-up Urban & Gardens + Building & Curtilage. Countryside Survey 2007 N Ireland
8. UK Summary Chapter 9
|North West||8,590||City of London||44,210||Barnet||10,870|
|Yorkshire & Humberside||7,900||Kensington||37,330||Harrow||10,290|
|East of England||5,930||Hackney||35,580||Hillingdon||9,640|
Population from 2011 census.
Standard Land Area Measurements.
The land area data is for Buildings & Gardens only 2007.
2005 Land % Useage from Census ward level GLUD 2005 Excel table
Full data from UK & London Population Density PDF or Excel
Revised with census data
244,146.6 UK To high water mark in km2
14,129.7 N Ireland
New mid 2010 population stats and revised UK land areas Jan 2011
I thought that this would be the easiest section to start with. How wrong could I be? The starting point was to find the area of the UK and Regions. As an engineer use to using Standard facts, I was amazed to find very few answers and then there was no consistency at all.
From there it got worse. Every Population Density figures I found were fundamentally incorrect as the
area used in the calculations included all land, such as mountains, roads, rivers that would be impractical to
I eventually found some data for London. Tower Hamlets 60,000/km sq. A lot different to the usual quote figure of 254.
The Countryside Survey looks beautiful but the data categories were inconsistent, no overall land sizes were referred to and Northern Ireland was found on its own website with no links to it.
The Next stage is to find correct data and put it in the public domain.
In researching data on UK population density, using your data, I have calculated that the population density of Tower
Hamlets is 60,300/km sq.
Am I correct?
My argument is simple, it concerns the way in which you calculate the land area.
Your figures include areas which would be impossible to live on such as mountains, rivers, roads, railways etc and areas which would be inappropriate to use to live on such as woodlands, farmlands, green belt etc.
This leaves land used for housing and their gardens. My figures for B & G come from Land Use GLUD 2005.
I will try and address your concerns about the population density figures. I have used the information you have included in your e-mail and the table you attached.
Your first e-mail concerned your calculation of the population density of
Tower Hamlets. You calculated that the population density of Tower Hamlets
is 60,300/km sq, and asked us if you are correct.
From your e-mail and attached file, I deduced that you used the "Area of Domestic Buildings" + "Area of domestic gardens". From the Land Use GLUD 2005 file you are using this is equal to 3,657 metres squared, or about 3.657 km sq. The population of Tower Hamlets is estimated at 234,800 for 2009. This means the population density for Tower Hamlets using your chosen definition of land area is 64,203.7 /km sq (or 64,194/km sq using the unrounded 2009 population estimate). This is close to your figure of 64,150 (Please can you let me know how you arrived at your figure? I think the small difference is due to rounding?)
So in this respect the figures are roughly similar. However, the difference between definitions of land area is what is causing the difference. ONS Population density figures are calculated using land areas derived from Standard Area Measurement figures created by ONS. Areas are calculated to the Mean High Water mark, and exclude inland water bodies greater than 1 sq km. These land area definitions are the standard used for comparability across the UN and the EU and consistent with standard methodology for calculating population density. This standard is most likely chosen because it better reflects the occupation of space per person and is therefore more useful.
If we only use "Area of Domestic Buildings" + "Area of domestic gardens"
and ignore green spaces, parks, woods, forests, farmed land, national
parks, mountains etc, we can wind up with a situation where the population
density of, say, the Isle of Skye in the Inner Hebrides (Highland Council
area) is considered to have a similar population density as Hampshire
(because the size of domestic buildings and gardens might be roughly the
same between the two areas). This is just a theoretical example. Including
areas such as green spaces, parks, cultivated areas, national parks, woods
and forests in the population density calculations provides a more
meaningful indicator of population density for many reasons.
The use of only "Area of Domestic Buildings" + "Area of domestic gardens" for population density calculations is also subject to a large margin of error because of differences in housing. The area occupied by a high-rise flat building in London may be the same as the area occupied by a detached house with a garden in Hampshire, but there is a big differences in the number of people who live in the two areas and individuals living the flats may have the same amount of living space each as the family living in the house.
The population density of Tower Hamlets is 11,876 people per sq. km.
On the Land use GLUD spreadsheet I calculated the % of land used for
Buildings + Gardens in relation to the total land area.
City of London 8%
Tower Hamlets 15%
These figures reinforce my argument that the land area should be the land that is actually used for Dwellings not the total area. The City 7% is empty at night because most of the land is taken up with offices. Residents are confined to a small area. Harrow 44% is "Metroland" leafy suburbs covered with Semi's and Terraced houses with gardens over a large area.
People can see this, it's not about statistics.
Having been to Switzerland and the Netherlands it is obvious that there will be a population difference as Switzerland is just one big mountain; but in reality, I would guess that Switzerland's density will be much higher than the Netherlands. The logic being that they are all crowded into a small space.
I understand your view point as you need to use statistical data that is reliable, consistent and is common to the EU and others for comparison. My background is in engineering, design, sales and marketing so my work is consumer orientated.
I understand the reasons for your point of view, that the land area should be the land that is actually used for Dwellings not the total area. Perhaps it would be most useful to clarify the differences by stating that the population densities you have calculated are based on 'dwelling area' for these reasons, and that ONS population densities are based on total land area (excluding water bodies larger than 1km2)