CENSUS CARTOGRAPHY: THE KENYAN EXPERIENCE

1
CENSUS CARTOGRAPHY THE KENYAN EXPERIENCE

• UN EXPERT GROUP MEETING ON CONTEMPORARY PRACTICES
  IN CENSUS MAPPING AND USE OF GEOGRAPHIC
  INFORMATION SYSTEMS
• UN HEADQUARTERS NEW YORK
• 29TH MAY TO 1ST JUNE 2007

2
PRESENTED

• by
• EMMA A.ODHIAMBO
• KENYA NATIONAL BUREAU OF STATISTICS
• Nairobi
• emma.odhiambo_at_gmail.com

3
INTRODUCTION

• The first census was conducted in 1948, followed
  by others in 1962, 1979, 1989 and the most recent
  in 1999.
• Kenya has been able to conduct censuses after
  every ten years.
• In the last three censuses Kenya has been able to
  carry out cartographic census mapping of the
  whole country.

4
INTRODUCTION CONTD

• The information collected includes
• Demographic
• Education
• Labour Force
• Housing and Sanitation

5
OBJECTIVES OF MAPS IN A CENSUS

• EA identification
• Assist in the delineation of the country into
  units known as Enumeration Area (EA)
• Assist enumerator to identify and plan how to
  logically move within the EA
• Omission and Duplication
• Maps ensure that errors of omission and
  duplication are avoided

6
OBJECTIVES CONTD

• Budgeting
• Budgeting for other census related requirements
  e.g personnel and materials.
• Data Dissemination
• Maps make it easier to analyze, present and
  disseminate census results

7
1969, 1979 and 1989 Censuses

• 1969 CENSUS
• Mapping activities done by Survey Department,
  Ministry of Lands and Settlement and Geography
  department, University of Nairobi.
• Topographic maps cuttings used (Scale 150,000)
• Very little field work done.

8
1979 CENSUS

• Efforts made to improve on the 1969 census
• Creation of cartography unit in 1976
• Enumeration Areas were delineated in the office.
• Attempt made to use satellite imageries ( not
  successful)

9
1989 Census

• Mapping drew heavily from 1979 experience and was
  considered a great success
• The officers were trained locally and abroad
• New concepts were introduced-
• Villages
• Household listing

10
1999 CENSUS MAPPING

• Geographic frame
• Administrative structure
• Intercensal changes
• In Administrative units and number of EAs

11
 
  Table 1 Changes in the number of
Administrative units between the last four
Censuses
12
FIELDWORK

• Training workshops (cartography,DSO Provincial
  Administration)
• Mapping teams
• Mapping

13
EA Delineation and EA size

• Dense rural settlements(Agricultural areas)
• Arid/ Semi Arid areas (Sparse population)
• Urban ( High population density)
• EA size- recommended is100 household (500
  persons)
• Composite EAs

14
Scales

• Rural Areas
• Dense rural settlments110,000, 120,000
• Sparsely populated areas 150,000, 1100,000,
  1250,000
• Urban 11000,12,500 and 15,000
• District Maps
• 1100,000, 1200,000

15
(No Transcript)
16
GEOCODING

• Coding-
• Admin Unit No Digits Sample code
• EA Type
• EAs - 4 3010101010011
• Sub location 2 301010101
• Locations 2 3010101
• Divisions 2 30101
• Districts 2 301
• Provinces 1 3
• EA TYPE CODE
• Rural 1 Urban 2 Peri Urban 3

17
Numbering of EAS
18
Ground Verification

• Checking of maps on the ground before the census
• In the history of census taking in Kenya
• this was achieved in 1989

19
Constraints

• Digital Mapping
• Late and erratic disbursement of funds which
  results to shortage of time for mapping
• Problem of urban slum mapping
• Use of sketch maps especially in urban areas
• Shortage of skilled manpower
• Lack of up to date topographic maps

20
RECOMMENDATIONS

• Early training of personnel in GIS/CAC
• Timely disbursement of funds
• Acquisition and use of satellite imageries for
  urban areas (Quick Bird)
• Use of Geographic Positioning System (GPS)

21
GEOGRAPHIC INFORMATION SYSTEM (GIS)

• The 2009 census will be the second to be
  undertaken using computer assisted cartography
  (CAC.)
• The UNFPA provided support for the establishment
  of a GIS facility which was commissioned in 1998
  and envisaged to culminate into a fully fledged
  Kenya Census Geographic Information System

22
GIS Contd

• However the implementation of the project
  experienced a number of setbacks that led to a
  slow pace in the process of preparing the
  enumeration area maps.
• As a result of the slackened pace, maps for 26
  districts were prepared using the GIS whereas the
  rest of the districts (43) were prepared using
  the conventional cartography.

23
(No Transcript)
24
CONSTRAINTS

• Late acquisition and installation of the GIS
  facility.
• Lack of adequate skilled staff in Computer
  assisted cartography
• Inadequate software user licences
• Poor maintenance of the GIS facility

25
CONSTRAINTS Contd

• Servicing of equipment was not adequate
• Upgrading of software was tied to maintenance
  service contract
• The cost of maintenance service contract was and
  still is very high. If not on this contract you
  can not get any upgrade on the new software
  release

26
USE OF SATELLITE IMAGES IN MAPPING FOR THE 2009
KENYA POPULATION AND HOUSING CENSUS.

• One of the biggest challenges facing census
  mapping in the heavily settled and un planned
  areas in Kenya is lack of suitable base maps in
  terms of scale and currency.
• The heavily settled areas which are mainly in
  urban areas are associated with slums where the
  settlements patterns are very dynamic with rapid
  changes taking place

27
USE OF SATELLITE IMAGES/AERIAL PHOTOGRAPHS IN
CENSUS MAPPING Contd

• Current and very large scale maps are therefore
  compulsory to enable proper mapping and the
  creation of suitable Enumeration Areas.
• Unfortunately, those responsible are not able to
  provide such maps when required.

28
USE OF SATELLITE IMAGES IN CENSUS MAPPING Contd

• Such challenges are quite serious in countries
  such as Kenya where about 60 of the urban
  populations live in these heavily settled areas
  (Slums).
• There is hardily any town in Kenya which does not
  have slum settlements and this underlines the
  seriousness of the problem.

29
USE OF SATELLITE IMAGES IN CENSUS MAPPING Contd

• The above mentioned challenges are quite serious
  in Kenya where a high percentage of the
  population lives in the slum areas.

30
USE OF SATELLITE IMAGES/AERIAL PHOTOGRAPHS IN
CENSUS MAPPING Contd

• Such challenges are quite serious in countries
  such as Kenya where about 60 of the urban
  populations live in these heavily settled areas
  (Slums).
• There is hardily any town in Kenya which does not
  have slum settlements and this underlines the
  seriousness of the problem.

31
USE OF SATELLITE IMAGES/AERIAL PHOTOGRAPHS IN
CENSUS MAPPING Contd

• In the censuses of 1989 and 1999, an attempt was
  made to solve this problem by relying on the
  preparation of sketches.
• But after the censuses, the sketches could not be
  relied on during the design of NASSEP programs
  and in the conversion of the EA maps into digital
  formats especially after the 1999 census.

32
USE OF SATELLITE IMAGES/AERIAL PHOTOGRAPHS IN
CENSUS MAPPING Contd

• During the cartographic mapping for the 2009
  Census, satellite images and aerial photographs
  are being used so as to address this problem.
• These products were used in Kisumu during the
  pilot mapping and they have also been used during
  the phase one mapping covering a lager part of
  Nyanza province and which was concluded recently

33
USE OF SATELLITE IMAGES/AERIAL PHOTOGRAPHS IN
CENSUS MAPPING Contd

• The products have proved to be very useful. They
  have enabled the preparation of excellent maps of
  the covered towns
• Indeed they offer a complete solution to the
  problem sited above because the products are very
  current and therefore enabling the mapping of
  every part of any urban area and at any desired
  map scale.

34
USE OF SATELLITE IMAGES IN CENSUS MAPPING Contd

• The maps produced are of very high quality and
  accurate and therefore they will ensure a
  complete and accurate coverage of the census.
• Further, because of their high quality and
  accuracy, they will enable a better sampling
  frame to be put in place since the urban
  population can be demarcated precisely from the
  rural population. This was not possible in the
  last census as it later led into a gross under
  quotation of the urban population

35
(No Transcript)
36
(No Transcript)
37
CONSTRAINTS

• The cost of the images and shape files is too
  high
• The satellite images are usually not current
  hence not giving the true picture on the ground
• The time taken to access the images and to create
  shape file from them is also too long