Title: The New Definition of Rural Areas of England and Wales and its Application to the 2001 Census
1The New Definition of Rural Areas of England and
Wales and its Application to the 2001 Census
- John Shepherd
- Defra Rural Evidence Research Centre, Birkbeck
College - University of London
2 The ODPM Review of Urban and Rural Definitions
in Use for Policy Purposes 2001
- Aims
- to identify policies that required definitions or
classifications of urban and rural geographies, - to assess how these definitions were used and how
well they performed relative to the policy needs, - to identify a set of core definitions that could
meet a wide range of user needs. - to suggest whether any new approaches to
definitions were required and what techniques
might be used to develop new definitions.
3Rural was defined in two quite different ways
- as not urban
- via the Urban Areas (land use) definition,
- gave 92 of the area and 10 of the population of
England as rural, - produces an un-differentiated rurality,
- rural towns defined as those below c. 10,000
pop. (itself arbitrary), but no populations for
places below c. 1500 population, - difficult and confusing to apply to
administrative areas, and in any case these can
divide some towns.
4 and as a socio-economic classification of
administrative areas
The Tarling definition
- produced in 1993 for RDC
- no conceptual basis, poorly defined and
un-documented, - accessible and remote rural not quantified,
and - anomalies due to lack of inclusion of internal
(e.g LA districts) settlement structure.
5The Review recommended that for rural policy a
new definition should
- move away from social/economic (e.g. agricultural
employment) criteria supposedly identifying
rurality, - extend the land use definition to below the c.
1500 population cut-off to include villages and
hamlets, - identify the actual settlement pattern of rural
areas, along with rural population and identified
distinctions between different types of rural
area, and - incorporate directly the geographical context of
rural policy expressed in such terms as as
accessible, urban fringe, village envelope,
etc.
6Extending the Logic of the Land Parcel (Urban
Areas) Approach meant
- Selecting a basic building block from
- Census Output Areas,
- Land parcels (TOIDS), or
- Grid cells at 100m resolution.
- Advantages of the latter
- a neutral boundary system
- a flexible geographical aggregation system
- enabled geographical tagging of a wide range of
policy data.
7Creating a new rural/urban definition Work began
in September 2002, was completed in March 2004
and is currently undergoing validation at
www.rural-urban.org.uk
The Clients DEFRA ODPM The Countryside
Agency The Welsh Assembly Government Office for
National Statistics
The Consultants SERRL (now RERC), Birkbeck
College, London Town and Regional Planning,
University of Sheffield Geowise, Edinburgh
8Three dimensions of the duality urban / rural
form a conceptual basis for the new definition
a distinction between land that is built over
and land that is not (usually associated with a
population size), the wider context of
physically defined settlements e.g. a
free-standing city, a market town, a sub-regional
mix of villages and hamlets, a sparsely
populated area, and economic separation from
the land as a direct source of income/wealth i.e.
the functional significance of settlements.
The new definition emphasises the MORPHOLOGY and
CONTEXT of settlements. This approach ensures
that the focus remains on the enduring aspects of
settlement. However, we also explored the
FUNCTIONAL attributes of rural settlements.
9Summary what the new definition does
- focuses on the enduring elements of rurality the
settlement pattern - extends the logic of the ODPM Urban Areas (land
use) definition to all settlements from towns,
through villages and hamlets to isolated
farms/dwellings, - classifies each postal address on the basis of
the 1ha grid cell in which it falls (there are 35
million for England and Wales) and build these
into small towns, villages, hamlets etc. - creates a computable map of settlement
structure (i.e. form, pattern and hierarchy),
across England and Wales - incorporates geographic context and settlement
form in the definition for all settlements, - classifies and aggregates COAs and wards on the
basis of context and settlement form.
10However
- the new definition does not include places with
10,000 population or more in 2001, which are
defined on the basis of Urban Areas - this was in the terms of reference for the
project and probably based on previous usage of
this cut-off, - a small point perhaps but it is something users
will have to be aware of (an example later).
11Form Building Settlements from Postal Addresses
- The Idea
- dwellings (addresses) are the basic units to be
classified - builds settlements from residential addresses.
- a ready made postcode look up,
- a 6482 x 5329 grid (34.5million cells).
12Creating Settlement Morphology
- from the density of residential addresses,
- using a moving window focused on each 1 ha
(100m) grid square - applied at 200m, 400m, 800m and 1600m to give an
average density for any particular grid square, - different types of settlement typically have
different density profiles or relationship
between different average densities.
Dark Green area for calculating the 200m
density Pale Green area for calculating the
400m density Pink area for calculating the 800m
density Beige area for calculating the 1600m
density
13Some Typical Household Densities and Settlement
Form in a Rural Area
Typical household densities (dwellings/ha)
Redditch
Henley in Arden
A Compact Village 360 dwellings 800m scale
0.73 400m scale 2.94 200m scale 11.08.
Alcester
A Rural Town c 1400 dwellings 1600m scale
2.1 800m scale 7.3 400m scale 18.3
Stratford
14But the Approach Promises Much More Rural
Settlement Structure in Detail
Ascott
Hook Norton
Moreton in Marsh
Long Compton
Urban Area urban shelf small urban urban
fringe nucleated village village
envelope scattered dwellings
Longborough
Evenlode
Chipping Norton
Stow on the Wold
Churchill
Hyde Hill
Dean
Spelsbury
Charlbury
15The Settlement Context Measure
- it was agreed not to measure accessibility to
services as in Scotland - context is represented as a combination of larger
scale density ratios, which measures average
accessibility to households within a given radius
of a 100m grid square - the radii give average density at 10,000m,
20,000m and 30,000m - a proxy for rural service delivery conditions a
work journey, local social services, emergency
services - what is the significance of a place relative to
aggregate accessibility? - the terms sparse and less sparse have been
agreed to describe context
16Settlement Context (Aggregate Accessibility)
Aggregate Household Density at 10,000m
Aggregate Household Density at 30,000m
17The Combined Densities Map Sparsity
Dark Blue sparse at all three scales Mid
Blue sparse at 30,000m and 20,000m Light
Blue sparse at 10,000m
18(No Transcript)
19Classification of 2001 Census Output
Areas Settlement Form and Context Areas
classified by a majority rule for the
proportion of households.
20Classification of 2001 Census Wards Rural Towns
in Context
21Classification of 2001 Census Wards Rural
Settlements rural towns and villages and
scattered settlements
22The Population of Rural England (New Definition,
OA Based)
The Rural Population Total 9.5m not 13m 19 not
23 Type 4.4m in rural towns 3.5m in villages.
23Migration 2001 in Rural England (New Definition,
Ward Based)
Rural towns gaining most of the migration growth
of rural wards (3 times that for villages and
scattered settlements Sparsity has a different
effect for rural towns as opposed to villages and
scattered settlements
24Outcome of the Validation Exercise
Just over 800 visitors to the website 71 online
and 26 written comments Three quarters
satisfied/very satisfied with the method,
slightly fewer with the outcome. Need for good
description/documentation and for advice on
policy application Need to understand impact of
uninhabited areas on density measures. Need for
clear policy on updating because of postcode
basis
25A Rural/Urban Classification of Local Authority
Districts and Unitary Authorities?
- Arguments Against
- the logic of the COA/ward classification doesnt
work at this level - raises the issue of market towns
- impending changes to local government structure
- few (if any) rural policies work at this level
- Arguments For
- there has been one since 1993
- much useful data only reported at this level
- Defra already has disadvantaged districts
- if Defra doesnt do it, someone will
- nice maps at this level
26What Might a District Classification Look Like?
Should be based on the new definition but depends
upon
- the choice of measure (binary or more?)
- the choice of cut off (50 too severe?)
- whether/how to include market towns over 10,000
population - whether/how to include rural territory
27To summarise, the new definition of rural areas
- extends the current Government land use based
definition of urban areas to include rural
settlements of different sizes and kinds -
- recognises the differences between rural areas
and moves away from crude urban / rural splits
given increasing difficulty in drawing a sharp
line between the two -
- bases the description of rural on factors that
people recognise as important rural
characteristics, for example the nature and
distribution of towns, villages and dispersed
areas -
- allows fine grain analysis of localities within
rural areas, for rural delivery and targeting
purposes and -
- provides a means of basing social and economic
data from, for example, the Census, on the more
enduring features of rurality - namely the
settlement pattern.