Www.sei.se/relu Characterizing Rural England using GIS Steve Cinderby, Meg Huby, Anne Owen.

www.sei.se/relu

Characterizing Rural England using GIS

Steve Cinderby, Meg Huby, Anne Owen

www.sei.se/reluwww.sei.se/relu

Scoping study in the Rural Economy & Land Use Programme

Aim:

To integrate natural and social science data into a spatial

dataset that can be used for analysis to inform rural policy-

making and provide a knowledge base for furthering policy

integration

Characterizing Rural England Using GIS

www.sei.se/reluwww.sei.se/relu

www.sei.se/relu

The Super Output Area• This study uses the new Census Super Output Areas

(SOAs) as the base unit for aggregation.

• SOAs are a new geography designed to improve the reporting of small area census statistics. It is intended that they will eventually become the standard across UK National Statistics.

• Lower level SOAs have a minimum population of 1000 people with a mean of 1500 people. For rural SOAs, areas range from 0.16km2 to 684km2 with a mean size of 18.2km2.

www.sei.se/relu

The Rural Definition

• Classification based on underlying hectare square grid

• Each square classified into one of 9 “morphological” categories – e.g. small town, village, hamlet

• Each square assigned a score based on the sparsity of the surrounding area

www.sei.se/relu

The Rural Definition for OAs

www.sei.se/relu

The Rural Definition for SOAs

• Super Output Areas are either rural or urban

• SOA is either sparse or less sparse

• Rural SOAs are either town or village/hamlet

• 2 urban and 4 rural types

www.sei.se/relu

www.sei.se/relu

Spatial Integration• The 2001 Census and 2004 Indices of Deprivation use

the Super Output Area as their areal unit.

• Other variables, particularly environmental datasets, use a different geography, which need to be integrated at SOA level.

• The problems of geographic integration to a common base unit are well known.

• This project aims to characterise, minimise and represent errors and uncertainty when data is portrayed at SOA level.

www.sei.se/relu

Distribution of data

• Uniform

• Patchy

• Continuously varying

www.sei.se/relu

Geography of data

• Point

• Line

• Area

• Surface

www.sei.se/relu

Resolution of data

Low High

www.sei.se/relu

Distribution of non SOA level data

• Data that have not been collected at SOA level must be assigned to SOAs

• The nature of the assignation is determined according to the underlying distribution of the data

• Additional data are required to determine the geography of the distribution

www.sei.se/relu

Case Study I: Bird species richness

• Captured at 10km grid square level

• Resolution is low

• Assume uniform distribution throughout grid square

• Apply area weighted averaging technique to construct data at SOA level

www.sei.se/relu

23 29

35 41

60 m2 80 m2

30 m220 m2

(23 x 60)/190 = 7.26

(29 x 80)/190 = 12.21

(35 x 20)/190 = 3.68

(41 x 30)/190 = 6.47

190 m2

(7.26 + 3.68 + 12.21 + 6.47) = 30 (2 s.f.)

Area Weighted Technique

www.sei.se/reluwww.sei.se/relu

www.sei.se/relu

Case Study II: Voter participation

• Captured at parliamentary ward level

• Resolution is low

• Assume patchy distribution of population settlements

• Apply population weighted averaging technique to construct data at SOA level

www.sei.se/relu

0.72 x 900 = 648

0.53 x 600 =

3181500 people

(648 + 318) / 1500 = 64.4%

Population Weighted Technique

72% 53%

900

600

www.sei.se/relu

www.sei.se/relu

Case Study III: Air Pollution

• 1km grid square level

• Resolution is high

• Distribution is continuously varying

www.sei.se/relu

When ‘average’ is not appropriate

• A weighted average technique masks variation in the data and information on very high, or very low values is lost

• When considering pollution data, it may be more appropriate to consider maximum pollution found in an SOA rather than the mean

www.sei.se/relu

Pollution: averaging problem

www.sei.se/relu

Case Study IV: Impact of Tourism

• Calculate an indicator showing the effect of tourism on Rural SOAs

• Use point data of visitor numbers to tourist sites with line data of road network

• Aim to show tourist ‘intensity’ along area adjacent to roads

www.sei.se/relu

Tourist Influence

www.sei.se/relu

Legend

Tourist influence

High

Low

www.sei.se/relu

Tourist Influence along roads

www.sei.se/relu

Conclusion• Problems of combining data together spatially do not

arise because the data is either environmental or socio-economic

• They depend on the nature of the data

• Each type must therefore be considered on a case by case basis, using supplementary data on the underlying distribution for mapping to SOA level