How dense?

James Murdoch's office

Not actually a post about my intellect or lack thereof, but on the density of residents in the area (yes I've chosen to write a blog on more demographic guff).

The map below plots the population density of Wapping and surrounding areas, and takes in all parts of Wapping and a few neighbouring LSOAs (Lower Layer Super Output Area - how the Office for National Statistics breaks the country down).

Population density is measured by taking population and dividing by the area in question (in this case hectares, or 10,000 sq metres, equivalent to a square 100 metres by 100 metres).

My colour code goes from very light green (zero people per hectare) to dark red (320 people per hectare). For reference Greater London (32 boroughs) is around 50 people per hectare, Kensington and Chelsea is 140 and Westminster 118. However, depending on the mix between residential, commercial, industrial and open space will interact to affect population density, so you can have low population density but very little greenspace, which makes simple comparison difficult.

In my map you can see there's a big variation in density, with the area around Shadwell station packing in 304 people per hectare and the Glamis / Cable Street area achieving 288, compared to parts of Wapping with only 38.7 (albeit with a marina in the middle).

Now, a word of caution; the LSOAs are broadly drawn to capture 1,000 people (see below for details), so density of different LSOAs may vary drastically, whereas, if larger blocks of a standard area were used, density wouldn't fluctuate as much, so we need to avoid reaching tautological conclusions. However, the data does let us see were residential populations are built up. One flaw with my data is that it excludes any increase in population arising from 21 Wapping Lane. However, I think that in someways that it's appropriate not to include such a significant development, as if a series of such developments were built, as here we're looking at relative population density in a mature settled community. To include such a large development in the analysis might lead to perverse conclusions when we don't know yet what the overall drain on the community's amenities will be.

Density of parts of LBTH

Apologies for the tables - at some point I will be motivated to write them in HTML rather than copying and pasting from my workings in Excel.

What we can tell from my dataset (see below) is that density in LBTH as a whole is quite a bit higher than for Wapping. However, looking at the 4 LSOAs which overlap or are directly touching the News International site (I label them 'NI Cluster' and are LSOAs TH 026A-D) shows that the existing population density is not much lower than LBTH as a whole, but relatively higher than Wapping, because of the presence of the St George's estate. I've also calculated densities excluding the area of the NI site to see if that affects the population density for the rest of the area - the answer is not massively.

Now, a big caveat here - Wapping has a lot of water - whether or not it should be excluded is a diffcult question. Clearly, you can't build on it, so renders comparing urban population density difficult, but does lead to a feeling of openess.

	LBTH	NI Cluster	LBTH less Wapping & NI cluster	Wapping& NI	Wapping
Area	2,157	58	1,980	176	168
Population	234,765	5,981	221,607	13,158	11,716
Density	109	103	112	75	70
Excluding NI site
Area	2,151	52	1,974	170	162
Density	109	115	112	77	72

 

Populating the NI site

So, realistically, how many residents should be housed on the NI site. Whats in Wapping stated it was 15.16 acres when St George plc acquired the site - which is 6.135 hectares. Assuming that the developers wish to have a mixed use development, rather than make it purely a housing estate with no additional facillities, we get the following figures:

Number of residents based on maintaining existing densities to area of site
	LBTH	NI Cluster	LBTH less Wapping & NI cluster	Wapping+NI	Wapping
Including NI site in existing density calc	668	633	686	458	428
Excluding NI site in existing density calc	670	708	689	475	444

Now if this gives a baseline increase in population, the question arises of how many units of housing. Clearly the answer depends on how many people live in each one. My preference would be for a larger number of 3 bed properties, of which I have a feeling from my initial research that there are a lack of in the area, catering for famillies rather than transient young professionals. I'll need to have a look into if there's any info on profitability of different sized housing units to see what motivates developers.

I've also calculated the extra number of residents that the site would need to accommodate to bring the overall density of Wapping upto the rest of LBTH, but I need to format these tables first.

About output areas

MSOAs are made up of LSOAs and LSOAs are made up of OAs - you at the back, keep up and pay attention.

MSOAs, LSOAs and OAs are defined by the Office of National Statistics and broadly come together to match Boroughs and Wards, but not necessarily(!). As a handy hint, OAs contain a minimum of 100 people, LSOAs 1000 and MSOAs 5,000 with the average size of each about 50% higher than the minimum. See the ONS website for more details.

Data source

http://data.london.gov.uk/datafiles/demographics/lsoa-data-2012.xls

 

Posted with Blogsy