Monday, April 16, 2007

Statistical breakthrough in city measuring

Statistical breakthrough in city measuring

As a true big city person and leading scientist, I have been frustrated with the current prevalent method of determining a cityÂ's size. Lame cities like Istanbul, Sao Paulo and Mexico City unjustly score high in these lists giving the reader the impression that these cities can measure themselves with real big and buff cities like New Yawk or Tokyo. This popular and misleading way of listing the worldÂ's Â'biggestÂ' cities is done simply by stating the population of the city or entire metropolis.

The buffness (i.e. Â'realÂ' size) of a city should be measured in a way that is more in line with that what makes a real city a real city. What makes a real city a real city, and not just a sprawl of houses, is the simultaneous presence of entertainment, shopping and offices. Using this as a basic concept, I have come up with a new, more reliable method of measuring city size.

Methodology

With the aid of the fabulous programme Google earth, I have measured the surface areas of Arnhem (in the Netherlands, my original home town) and Bangkok (the city in which I currently reside) that have simultaneous presence of entertainment, shopping and offices.

To calculate the surface of these areas of Arnhem and Bangkok, the Google Earth measure tool was used to draw a rectangular shape containing the central areas as accurately as possible (see images). Drawing polygons would be even more accurate, but it takes ages to calculate their surfaces.

Because the population density effectively influences a cityÂ's buffness (more people in one area means more buidlings in one area, which means more city for the same amount of space), I have also set a benchmark level of population density to compare the actual densities with and used the resulting ratioÂ's as multipliers for the citiesÂ' surfaces, resulting in the true size of a city.

In addition, I made a comparison between the comparison between the two cities using my method and the comparison between the two cities using the method of determining a cityÂ's size using its official city population. The findings are most striking.

Data

Arnhem

• Perimeter of the area: 3.98 km
• Length of the area: 1.17 km
• Width of the area: 0.85 km
• Surface (S): 0.99 km2

Bangkok

• Perimeter of the area: 34.33 km
• Length of the area: 9.24 km
• Width of the area: 7.87 km
• Surface (S): 72.72 km2

Divide the surface of BangkokÂ's centre with the surface of ArnhemÂ's centre, and you will find that Bangkok is exactly 73.5 times bigger and buffer than Arnhem.

Now letÂ's add the density variable (density correction, or DC) to the equation:

Density benchmark (DB): 2,000 inhabitants per km2
Population density (D) Arnhem: 1,428 inhabitants per km2 ^1
Population density (D) Bangkok: 4,051 inhabitants per km2 ^2

So:

City size (DC) = (L*W) or S * (D/DB)
City size (DC) Arnhem = 0.99 * (1,428/2,000) = 0.71 km2
City size (DC) Bangkok = 72.72 * (4,051/2,000) = 147.3 km2

With density correction, Bangkok turns out to be effectively 207.5 times bigger and buffer than Arnhem.

Let us compare this result with the result we would have if we would compare the citiesÂ' sizes using the traditional method that uses the number of inhabitants:

• Arnhem # of inhabitants: 142,201 ^3
• Bangkok # of inhabitants: 6,320,000 ^4

According to these figures, Bangkok is a mere 44.4 times bigger and buffer than Arnhem.

Conclusion

Current statistics concerning the size of cities are misleading. True big and buff cities often lose to large, boring conglomerates of suburbs and shantytowns, which is utterly atrocious. In this experiment I proved that Bangkok is effectively 207.5 times bigger and buffer than Arnhem. As of today, this method shall be known as the 'Sir P von method of determining city sizeÂ' and it shall replace the method based on population figures that has been used by liars for over 400 years.

Footnotes:

1. http://www.cbs.nl/nl-NL/menu/themas/...002-arnhem.htm
2. http://www.answers.com/topic/list-of...lation-density
3. http://www.sdu.nl/staatscourant/gemeentes/gem30g.htm
4. http://www.answers.com/Bangkok


>

No comments: