How the science of cities can help European policy makers ... _urbanisation... · How the science of cities can help European policy makers: new analysis ... intensity, distance •

Regional & Urban Policy

How the science of cities can help European policy

makers: new analysis and perspectives

By Lewis Dijkstra, PhD

Deputy Head of the Economic Analysis Unit,

DG Regional and Urban Policy

European Commission


Overview

• Data revolution

• Defining cities

• New analysis and links to the urban sustainable development goal indicators

• Density

• Land use changes

• Green space

• Public transport


A data revolution

• More micro data (population register, business register, patents)

• More geo-coded data (building register, census…)

• More remote sensing data (water, green, imperviousness, buildings…)

• More big data (smart phones, geotagged pictures, messages…)


Copernicus Urban Atlas

• Thematic classes based on CORINE Land Cover nomenclature

• But more specific for built-up areas, and less specific outside urban areas

• Geometric resolution of 1:10,000

• Minimum mapping unit of 0.25 ha in urban areas, 1 ha in other areas

• Imagery reference year: 2006 and 2012


CORINE Land Cover


Urban Atlas


27 May, 2015

Sealed Surface Layer


27 May, 2015

GHSL


What makes a city?

• Buildings, mass, proximity

• People, density, size

• Exchange, intensity, distance

• Functions, specialised, variety

• Political

• Economically linked

• Labour market, commuting zone


Population distribution within a city

1. To find out IF a municipality contains a city

2. To define an 'urban centre'

3. To measure access to transport, green space…

4. To measure weighted density instead of average density

5. To measure exposure to air quality


Urban centre versus administrative city


EU-OECD city and commuting zone definition in three steps

1. Define an urban centre of 50 000 or more

2. Define a city based on this urban centre (consisting of one or more municipalities)

3. Define a commuting zone based on this city (including check for polycentric cities)

IMPORTANT! Cities are selected based on the population of their centre, not total population


One, two, three


Three grid concepts

1. Urban centres = contiguous (excluding diagonals) cells with a density of at least 1500 inhab/km2 and a minimum of 50 000 inhabitants (after gaps filled with majority rule)

2. Urban clusters = contiguous (including diagonals) cells with a density of at least 300 inhab/km2 and a minimum of 5 000 inhabitants (no gap filling)

3. Rural grid cells = cells outside urban clusters


Three degrees of urbanisation Three grid concepts (Cork, IE)

Three types of municipalities


Three types of municipalities

Cities > 50% pop. in urban centres

Towns and suburbs

> 50% pop. in urban clusters

< 50% pop. in urban centres

Rural area > 50% pop. in rural grid cells

Urban areas = Cities + Towns and Suburbs


Density drops away from the centre

0

100

200

300

400

0

100

200

300

400

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Popu

lati

on p

er h

ecta

re

Population density profile of selected mid-sized European capital cities, 2006

Distance from city centre (Km) Source: Batista e Silva, F. et al. (2012)

Wien

Lisbon

Budapest

Brussels

Stockholm

Dublin Amsterdam


Share of built-up area drops away from the centre

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Distance to the city centre

10

9

8

7

6

5

4

3

2

1

0


Share of built-up area drops away from the centre (cumulative)

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Maximum distance to the centre

10

9

8

7

6

5

4

3

2

1

0


Density or land use indicators

• Target 11.b holistic disaster risk management

• Population density measured over continuous urban footprint

• Target 11.3 sustainable urbanization (& 11.a)

• Ratio of land consumption rate to population growth rate at comparable scale

• Problems

• Urban footprint or building footprint

• What is land consumption?

• What scale?


Proposal land use efficiency indicator

• Measure built-up area (building footprint) per inhabitant based on GHSL for

• Cities following the EU-OECD definition and

• Commuting zone (if commuting is available) or

• Suburbs following degree of urbanisation or

• A buffer based on population size of a city

• Monitor the changes in built-up area per capita over time (land use efficiency)

• Cities with a high efficiency can reduce it, cities with low efficiency should increase it.



Measuring access to public transport: input data

• Location of all public transport stops

• Timetables of services: 2 groups:

• bus and tram

• train and metro

• Population per building block based on:

• detailed population grids

• census tracts

• neighbourhood statistics

• plus disaggregation using land use data and/or imperviousness if needed


Spatial distribution of population matters

No location awareness: assuming uniform population density throughout the city

High-resolution spatial distribution of population: Opportunities for new indicators


Frequency of departures

• Average stops an hour from 6:00 to 20:00 on a normal week day

Very high

More than ten departures an hour for both medium- and high-speed modes

High More than ten departures an hour for one mode, but not both

Medium

Between four and ten departures an hour on one or both modes, but no access to more than ten departures and hour

Low less than four departures an hour for one or both modes, but no access to more than four departures an hour

Null No access within walking distance


Typology of frequency classes

Very high Access to more than ten departures an hour for both medium- and high-speed modes

High Access to more than ten departures an hour for one mode, but not both

Medium Access to between four and ten departures an hour on one or both modes, but no access to more than ten departures and hour

Low less than four departures an hour for one or both modes, but no access to more than four departures an hour

Null No access within walking distance


Stockholm: areas and population by access to public transport and its frequency

844,000 1,135,000 1,542,000 2,042,000 inh. inh. inh. inh.


Access to public transport in Brussels

0

10

20

30

40

50

60

70

80

90

100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Sh

are o

f p

op

op

ula

tio

n,

in %

Distance to the city centre in km

Null

Low

Medium

High

Very high

Access to a

public

transport by

frequency of


Target 11.2 Public transport

• Share of people living within 0.5 km of public transit [running at least every 20 minutes] in cities with more than 500,000 inhabitants

• Specify the city definition to be used

• Km of high capacity (BRT, light rail, metro) public transport per person for cities with more than 500,000 inhabitants

• Why not measure access to high capacity public transport?



Green spaces in Brussels, 2012

0%

10%

20%

30%

40%

50%

60%

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Distance to city centre in km

forests

agriculture and natural areas

sports and leisure

green urban areas


Access to green spaces by size

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1 2 3 4 5 6 7 8 9

Shar

e o

f p

op

ula

tio

n

Maximum distance to the city centre in km

extra large

large

medium

small

very small

(almost) none

Access according to the size of green space


Target 11.7 Green and public space

• Area of public space as a % of total city space

• Share of residents within 0.5 km of accessible green and public space

• Accessible is extremely difficult to determine

• Public space: roads, sidewalks, squares?

• Proposal: share of residents with

• Almost no open space in a buffer of 0.5 km

• No green space of at least x m2 within 0.5 km

• This avoids the problem of measuring access, but it will be a subset of the population with no access


Conclusion

• Data revolution is in full swing, but we need

• a universe of cities using a single methodology

• Understand population distribution within cities

• Be aware of the modifiable area unit problem:

• Use uniform building blocks (like grid cells)

• Use population with access rather than area share

• When using area shares, use a grid definition, not an administrative one

• Take full advantage of new continuous, high resolution data sets (vs coarse and binary data)


More information

• EU-OECD City definition

• http://ec.europa.eu/regional_policy/sources/docgener/focus/2012_01_city.pdf

• New degree of urbanisation

• http://ec.europa.eu/regional_policy/sources/docgener/work/2014_01_new_urban.pdf

http://ec.europa.eu/regional_policy/sources/docgener/focus/2012_01_city.pdf



http://ec.europa.eu/regional_policy/sources/docgener/work/2014_01_new_urban.pdf



How the science of cities can help European policy makers ... _urbanisation... · How the science of cities can help European policy makers: new analysis ... intensity, distance •

Documents