Earth Observation for Sustainable Development
Urban Development Project
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 685761.
ESA Ref: AO/1-8346/15/I-NB
Doc. No.: City Operations Report
Issue/Rev.: 3.0
Date: 24.04.2018
EO4SD-Urban Project: Kigoma City Report
Lead: Partners: Financed by:
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page I
Consortium Partners
No. Name Short Name Country
1 GAF AG GAF Germany
2 Système d'Information à Référence Spatiale SAS SIRS France
3 GISAT S.R.O. GISAT Czech Republic
4 Egis SA EGIS France
5 Deutsche Luft- und Raumfahrt e. V DLR Germany
6 Netherlands Geomatics & Earth Observation B.V. NEO The Netherlands
7 JOANNEUM Research Forschungsgesellschaft mbH JR Austria
8 GISBOX SRL GISBOX Romania
Disclaimer:
The contents of this document are the copyright of GAF AG and Partners. It is released by GAF AG on
the condition that it will not be copied in whole, in section or otherwise reproduced (whether by
photographic, reprographic or any other method) and that the contents thereof shall not be divulged to
any other person other than of the addressed (save to the other authorised officers of their organisation
having a need to know such contents, for the purpose of which disclosure is made by GAF AG) without
prior consent of GAF AG.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page II
Summary
This document contains information related to the provision of geo-spatial products from the European
Space Agency (ESA) supported project “Earth Observation for Sustainable Development” Urban Applications (EO4SD-Urban) to the World Bank Urban Planning Study for Tanzania Programme for
the City of Kigoma.
Affiliation/Function Name Date
Prepared GAF AG J. Freitas-Santos, A.
Broszeit
19/04/2018
Reviewed GAF AG D. Angelova 20/04/2018
Approved GAF AG, Project Coordinator T. Haeusler 24/04/2018
The document is accepted under the assumption that all verification activities were carried out correctly
and any discrepancies are documented properly.
Distribution
Affiliation Name Copies
ESA Z. Bartalis electronic copy
World Bank Chyi-Yun Huang electronic copy
Document Status Sheet
Issue Date Details
1.0 14/09/2017 First Document Issue
2.0 15/11/2017 Second Document Issue
3.0 24/04/2018 Third Document Issue
Document Change Record
# Date Request Location Details
1 15/11/2017 Ch. 2.6, 2.7,
2.8, 3.5, 3.6,
4.4, 4.5, 4.6
and Annexe 2
Three products were added to the Service
Operations Report: 1. Urban Green Areas 2.
Planned and Unplanned Settlement Areas 3.
Population Distribution and Density
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page III
2 15/11/2017 Annexe 1 Description of processing methods included
for: Transport Network, Urban Green Areas,
Planned and Unplanned Settlement Areas,
Population Distribution and Density.
3 24/04/2018 Ch. 2.1, 3.3.1,
4.2, 4.3 and
4.4
After a User’s request, the spatial datasets for
the LULC, Transport network and Green
Areas products were adjusted. These updates
implied changes in the following Sections of
the City Operations Report.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page IV
Executive Summary
The European Space Agency (ESA) has been working closely together with the International Finance
Institutes (IFIs) and their client countries to demonstrate the benefits of Earth Observation (EO) in the
IFI development programmes. Earth Observation for Sustainable Development (EO4SD) is a new ESA
initiative, which aims to achieve an increase in the uptake of satellite based information in the regional
and global IFI programmes. The overall aim of the EO4SD Urban project is to integrate the application
of satellite data for urban development programmes being implemented by the IFIs or Multi-Lateral
Development Banks (MDBs) with the developing countries. The overall goal will be achieved via
implementation of the following main objectives:
To provide a service portfolio of Baseline and Derived urban-related geo-spatial products
To provide the geo-spatial products and services on a geographical regional basis
To ensure that the products and services are user-driven
This Report describes the generation and the provision of EO-based information products to the World
Bank (WB) supported Technical Assistance project “Urban Planning Study-Impact & Effectiveness
of Urban Planning on City Spatial Development” for Tanzania and the counterpart City authorities
in Kigoma. The Report provides a Service Description by referring to the user driven service
requirements and the associated product list with the detailed product specifications. The following
products were requested:
Urban Land Use/ Land Cover
Urban Extent
Urban Green Areas
Extent and Type of Informal Settlements
Population Distribution and Density
Transport Infrastructure - Road Network
The current Version of this Report contains the description of the generation and delivery of the Land
Use/Land Cover (LU/LC) and the LU/LC Change between two time periods of 2006 and 2016 as well
as the Transport Infrastructure product. The Derived products of Green Areas, Informal Settlements and
Population Density product are also included in this Report. Furthermore, it focuses on the
improvements of the LULC, Transport infrastructure and Green Areas products which have been done
based on the User Utility Review.
This City Operations Report for Kigoma systematically reviews the main production steps involved and
importantly highlights the Quality Control (QC) mechanisms involved; the steps of QC and the
assessment of quality is provided in related QC forms in the Annexe of this Report. There is also the
provision of standard analytical work undertaken with the products which can be further included as
inputs into further urban development assessments, modelling and reports.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page V
This page is intentionally left blank!
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page VI
Table of Contents
1 GENERAL BACKGROUND OF EO4SD-URBAN ................................................................... 1
2 SERVICE DESCRIPTION........................................................................................................... 1
2.1 STAKEHOLDERS AND REQUIREMENTS ..................................................................................... 1
2.2 SERVICE AREA SPECIFICATION ............................................................................................... 2
2.3 PRODUCT LIST AND PRODUCT SPECIFICATIONS ...................................................................... 3
2.4 LAND USE/LAND COVER NOMENCLATURE ............................................................................. 3
2.5 TRANSPORT INFRASTRUCTURE - ROAD NETWORK ................................................................. 6
2.6 URBAN GREEN AREAS ............................................................................................................. 7
2.7 PLANNED AND UNPLANNED SETTLEMENT AREAS .................................................................. 7
2.8 POPULATION DISTRIBUTION AND DENSITY ............................................................................. 8
2.9 TERMS OF ACCESS ................................................................................................................... 8
3 SERVICE OPERATIONS ............................................................................................................ 9
3.1 SOURCE DATA ......................................................................................................................... 9
3.2 PROCESSING METHODS ......................................................................................................... 10
3.3 ACCURACY ASSESSMENT OF MAP PRODUCTS ...................................................................... 10
3.3.1 The Applied Sampling Design ........................................................................................................ 10
3.3.2 The Applied Response Design ........................................................................................................ 11
3.3.3 The Applied Analysis Design .......................................................................................................... 12
3.4 ACCURACY ASSESSMENT OF TRANSPORT NETWORK ........................................................... 13
3.5 ACCURACY ASSESSMENT OF URBAN GREEN AREAS ............................................................ 15
3.6 ACCURACY ASSESSMENT OF PLANNED AND UNPLANNED SETTLEMENTS ........................... 17
3.7 QUALITY CONTROL/ASSURANCE .......................................................................................... 18
3.8 METADATA ............................................................................................................................ 19
4 ANALYSIS OF MAPPING RESULTS ..................................................................................... 21
4.1 URBAN EXTENT – DEVELOPMENTS 2000, 2005, 2010 AND 2015 .......................................... 21
4.2 LAND COVER/ LAND USE 2006 AND 2016 ............................................................................. 23
4.2.1 Spatial Distribution of Main LU/LC Change Categories ................................................................ 25
4.2.2 Changes of Agricultural Areas ........................................................................................................ 27
4.3 TRANSPORT NETWORK .......................................................................................................... 29
4.4 URBAN GREEN AREAS ........................................................................................................... 30
4.5 PLANNED AND UNPLANNED SETTLEMENT AREAS ................................................................ 31
4.6 POPULATION DISTRIBUTION AND DENSITY ........................................................................... 35
4.7 CONCLUDING POINTS ............................................................................................................ 37
5 REFERENCES ............................................................................................................................ 38
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page VII
Annexes
Annex 1: Processing Methods for EO4SD-Urban Products
Annex 2: Filled Quality Control Sheets
List of Figures
Figure 1: Illustration of Core and Peri-Urban Areas of Mapping for Kigoma. .................................. 2
Figure 2: Mapping result of the city of Kigoma of the year 2016 overlaid with randomly distributed
sample points used for accuracy assessment..................................................................... 12
Figure 3: Example of the applied sampling design to generate randomly distributed point for the
Accuracy Assessment of the road network. ...................................................................... 14
Figure 4: Secondary sampling grid to generate the sampling points at spatial intersection of roads
and grid cells. Roads are represented as white lines, grid as black lines and final sampling
point as black dots. ............................................................................................................ 14
Figure 5: Result of the Urban Green Area mapping in Kigoma (change product) with sampling points
used for product validation. .............................................................................................. 16
Figure 6: Result of the Informal Settlement Area mapping in Kigoma (change product) with sampling
points used for product validation. .................................................................................... 17
Figure 7: Quality Control process for EO4SD-Urban product generation. At each intermediate
processing step output properties are compared against pre-defined requirements. ......... 19
Figure 8: Urban Extent developments in the epochs 2000 to 2005, 2005 to 2010 and 2010 to 2015 in
Kigoma and surrounding region. ...................................................................................... 22
Figure 9: Urban Extent developments in the epochs 2000 to 2005, 2005 to 2010 and 2010 to 2015 in
Kigoma within the Core Urban Area. ............................................................................... 22
Figure 10: Detailed Land Cover Land Use 2016 in Kigoma. ............................................................. 23
Figure 11: Detailed Land Cover Land Use 2006 structure: Presented as Overall, Core Urban and in
Peri-Urban Zone in percentages (left) and square kilometres (right) ............................... 24
Figure 12: Detailed Land Cover Land Use 2016 structure: Presented as Overall, Core Urban and in
Peri-Urban Zone in percentages (left) and square kilometres (right). .............................. 24
Figure 13: Land Cover Land Use Change Types - Spatial Distribution ............................................ 26
Figure 14: Land Cover Land Use Change Types 2006-2016 - Overall, in Core Urban and in Peri-Urban
in % (left) and km² (right) in Kigoma. .............................................................................. 27
Figure 15: Spatial distribution of changes from Agricultural Areas into other Classes between 2006
and 2016. ........................................................................................................................... 28
Figure 16: Changes of Agricultural Areas into other LU classes between 2006 and 2016; Presented as
Overall, Core Urban and in Peri-Urban in % (left) and km2 (right).................................. 28
Figure 17: Transport Network of Kigoma in 2006 (left) and 2016 (right). ........................................ 29
Figure 18: Map overview of Urban Green Areas in Kigoma. Green area loss, gain and stable green
areas can be identified. The map is delivered as separate product of high resolution for
printing at paper size DIN A0, which is 84.1 cm x 118.9 cm. .......................................... 30
Figure 19: Percentage of urban green areas within the core area of Kigoma. The pie chart illustrates
the status and change of urban green areas in-between 2006 and 2016. ........................... 31
Figure 20: Bar charts for both points in time presenting the total area of urban greenery versus non-
green areas. ....................................................................................................................... 31
Figure 21: Planned settlement areas in Kigoma in 2016. ................................................................... 32
Figure 22: Unplanned settlement areas in Kigoma in 2016. .............................................................. 32
Figure 23: Planned and unplanned settlement areas received via email from MaryGrace Weber. .... 33
Figure 24: Map overview of changes in planned and unplanned settlement areas in Kigoma during the
years 2006 and 2016 ......................................................................................................... 33
Figure 25: Percentage of planned and unplanned areas within the core area of Kigoma. The pie chart
illustrates the status and change of planned and unplanned areas in-between 2006 and
2016………………………………………………………………………………………34
Figure 26: Bar charts for both points in time presenting the total area of planned and unplanned
settlement areas. ................................................................................................................ 34
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page VIII
Figure 27: Overview Map of Population Distribution Change in Kigoma (2005 – 2015). ................ 35
Figure 28: Population Distribution Change within the Core Urban Districts of Kigoma between 2005
and 2015. ........................................................................................................................... 36
Figure 29: Changes in Population Distribution, in relation to build up areas or soil sealing degree in
Kigoma between 2005 and 2015. ...................................................................................... 37
List of Tables
Table 1: LU/LC Nomenclature for 2006 and 2016. .......................................................................... 4
Table 2: Nomenclature used for the mapping and identification of Urban Green Areas. ................. 7
Table 3: Nomenclature used for the mapping and identification of Planned and Unplanned
Settlement Areas. ................................................................................................................ 7
Table 4: Nomenclature used for the mapping and identification of Population Distribution and
Density. ............................................................................................................................... 8
Table 5: Number of sampling points for the EO4SD-Urban mapping classes after applied sampling
design with information on overall land cover by class. ................................................... 11
Table 6: Validation result of the complemented Transport Network in Kigoma, which is based on
OSM data. ......................................................................................................................... 15
Table 7: Calculation of the minimum number of samples according Goodchild et al. 1994. ......... 15
Table 8: Results of the Accuracy Assessment of Urban Green Areas in Kigoma, 2006 ................ 16
Table 9: Results of the Accuracy Assessment of Urban Green Areas in Kigoma, 2016 ................ 16
Table 10: Calculation of the minimum number of samples according Goodchild et al. (1994). ...... 17
Table 11: Results of the Accuracy Assessment of Informal Settlement Areas in Kigoma, 2016. .... 18
Table 12: Detailed information on area and percentage of total area for each class for 2006 and 2016
as well as the changes. ...................................................................................................... 25
Table 13: Overall LU/LC Statistics. ................................................................................................. 27
Table 14: Statistics of changes of agricultural areas between 2006 and 2016. ................................. 29
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page IX
List of Abbreviations
CDS City Development Strategy
CS Client States
DEM Digital Elevation Model
DLR German Space Agency
EEA European Environmental Agency
EGIS Consulting Company for Environmental Impact Assessment and Urban Planning, France
EO Earth Observation
ESA European Space Agency
EU European Union
GAF GAF AG, Geospatial Service Provider, Germany
GIS Geographic Information System
GISAT Geospatial Service Provider, Czech Republic
GISBOX Romanian company with activities of Photogrammetry and GIS
GUF Global Urban Footprint
HR High Resolution
HRL High Resolution Layer
IFI International Financing Institute
INSPIRE Infrastructure for Spatial Information in the European Community
ISO/TC 211 Standardization of Digital Geographic Information
JR JOANNEUM Research, Austria
LC / LU Land Cover / Land Use
LULCC Land Use and Land Cover Change
MMU Minimum Mapping Unit
NDVI Normalized Difference Vegetation Index
NEO Geospatial Service Provider, The Netherlands
QA Quality Assurance
QC Quality Control
QM Quality Management
SP Service Provider
VHR Very High Resolution
WB World Bank
WBG World Bank Group
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 1
1 General Background of EO4SD-Urban
Since 2008 the European Space Agency (ESA) has worked closely together with the International
Finance Institutes (IFIs) and their client countries to harness the benefits of Earth Observation (EO) in
their operations and resources management. Earth Observation for Sustainable Development (EO4SD)
is a new ESA initiative, which aims to achieve an increase in the uptake of satellite based information
in the regional and global IFI programmes. The EO4SD-Urban project initiated in May 2016 (with a
duration of 3 years) has the overall aim to integrate the application of satellite data for urban
development programmes being implemented by the IFIs with the developing countries. The overall
goal will be achieved via implementation of the following main objectives:
To provide the services on a regional basis (i.e. large geographical areas); in the context of the
current proposal with a focus on S. Asia, SE Asia and Africa, for at least 35-40 cities.
To ensure that the products and services are user-driven; i.e. priority products and services to
be agreed on with the MDBs in relation to their regional programs and furthermore to implement
the project with a strong stakeholder engagement especially in context with the validation of the
products/services on their utility.
To provide a service portfolio of Baseline and Derived urban-related geo-spatial products that
have clear technical specifications, and are produced on an operational manner that are
stringently quality controlled and validated by the user community.
To provide a technology transfer component in the project via capacity building exercises in the
different regions in close co-operation with the MDB programmes.
This Report supports the fulfilment of the third objective which requires the provision of geo-spatial
Baseline and Derived geo-spatial products to various stakeholders in the IFIs and counterpart City
authorities. The Report provides a Service Description, and then in Chapter 3 systematically reviews the
main production steps involved and importantly highlights whenever there are Quality control (QC)
mechanisms involved with the related QC forms in the Annexe of this Report. The description of the
processes is kept intentionally at a top leave and avoiding technical details as the Report is considered
mainly for non-technical IFI staff and experts and City authorities. Finally Chapter 4 presents the
standard analytical work undertaken with the products which can be inputs into further urban
development assessments, modelling and reports.
2 Service Description
The following Section summarises the service as it has been realised for the city of Kigoma, Tanzania
within the EO4SD-Urban Project and as it had been delivered to the Task Team Leader (TTL) of the
World Bank programme Urban Planning Study for Tanzania in August 2017.
2.1 Stakeholders and Requirements
The EO4SD-Urban products described in this Report were provided to the World Bank (WB) Technical
Assistance project “Urban Planning Study-Impact & Effectiveness of Urban Planning on City
Spatial Development” for Tanzania. The Urban Planning Study is linked to the larger WB project
‘Tanzania Strategic Cities Project’ (TSCP). The TSCP supports the seven secondary cities: Tanga,
Arusha, Mwanza, Kigoma, Dodoma, Mbeya and Mtwara, in addition to the Capital Development
Authority (CDA) in Dodoma, the national capital. These cities are of strategic importance as they are
within the top ten most populous cities in Tanzania with high population growth rates. In the context of
the current EO4SD Urban project, three of the overall seven cities - Arusha, Kigoma, Dodoma - could
be provided with the geo-spatial datasets. The higher objective of the Urban Planning Study, as stated
in its Draft Concept Note (World Bank, 2016a) is, “to enhance the urban development and inform policies and development strategies of cities in Tanzania through gaining further insights on urban
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 2
planning system and development processes and the effectiveness of master and detailed urban
plans.” It aims to increase the appreciation of practitioners and local authorities on the importance of
urban plans for guiding urban growth and mitigating the potential problems and higher costs of
retrofitting unplanned development. In this context, the Bank is collaborating with the Consortium to
investigate the spatial development characteristics of Tanzanian cities with and without urban plans, and
attempt to assess the impact and effectiveness of such urban plans. This specific Version of the Report
focusses on providing the new results based on the improvements made of the LULC, Transport
infrastructure and Green areas products.
2.2 Service Area Specification
The Areas of Interest (AoI) for mapping the Urban and the Peri-Urban Areas for Kigoma was depicted
in a power point slide, and sent to the Users for verification. The boundaries depicted were based on the
municipality and administration boundaries of the cities. These boundaries were obtained from the
GADM database of Global Administrative Areas (http://www.gadm.org/).
As the city administrative boundary data availability is different for each country/city, the AoIs for the
Urban and Peri-Urban Area were in some cases adjusted to areas which could provide the best examples
of the geo-spatial products that Users may require. The adjustments are based on population distribution
data from LandScan (http://web.ornl.gov/sci/landscan/) and on visual interpretation of built-up areas as
evidenced on Google Earth. LandScan is the finest resolution global population distribution data (~1km
spatial resolution) available and represents an ambient population.
Figure 1: Illustration of Core and Peri-Urban Areas of Mapping for Kigoma.
The Core region has an area of 84 km2 and the Peri-Urban has an area of 209 km2, for a total service or
mapping area of 293 km2.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 3
2.3 Product List and Product Specifications
During the discussions related to the AoIs the potential geo-spatial products that could be provided for
the Cities were also reviewed with the WB Team and users. It was noted that the Baseline Land
Use/Land Cover (LU/LC) products (for the Core and Peri-Urban areas) were a standard product that
would be provided for all Cities as it is required for the derived products. In the case of Kigoma, the full
list of products for both the Core and Peri-Urban areas are as follows:
Urban Land Use/ Land Cover
Urban Extent
Urban Green Areas
Extent and Type of Informal Settlements
Population Distribution and Density
Transport Infrastructure - Road Network
For each of these products two time slots were used to provide historic and recent information; thus for
Kigoma it was 2006 and 2016. Due to EO data availability the time epochs varied with a plus/minus
year or two around these 2 time slots. The current Report will focus on the provision of the Baseline
LU/LC products and the Transport Infrastructure for Kigoma.
2.4 Land Use/Land Cover Nomenclature
A pre-cursor to starting production was the establishment with the stakeholders on the relevant Land
Use/Land Cover (LU/LC) nomenclature as well as class definitions. The approach taken was to use a
standard remote sensing based LU/LC nomenclature and then adapt it to the user’s LU requirements. Thus the remote-sensing based LU/LC classes in the urban context can be grouped into 5 Level 1 classes,
which are Artificial Areas, Natural/ Semi Natural, Agricultural, Wetland, Water Bodies. These classes
can then be sub-divided into several different more detailed classes such that the dis-aggregation can be
down to Level 2-4. This hierarchical classification system is often used in operational Urban mapping
programmes and is the basis for example of the European Commission’s Urban Atlas programme which provides pan-European comparable LU/LC data for with regular updates. A depiction of the way the
levels and classes are structured is presented as follows:
Level I Artificial surfaces
- Level II Urban Fabric
Level III
Continuous Urban Fabric (Sealing Layer-S.L. > 80%)
Discontinuous Urban Fabric (S.L. 10% - 80%)
Discontinuous Dense Urban Fabric (S.L. 50% - 80%)
Discontinuous Medium Density Urban Fabric (S.L. 30% - 50%)
Discontinuous Low Density Urban Fabric (S.L. 10% - 30%)
Discontinuous Very Low Density Urban Fabric (S.L. < 10%)
- Level II
Industrial, commercial, public, military, private and transport units
Level III
Industrial, commercial, public, military and private units zoning data
Road and rail network and associated land (Open Street Map or In-country data needed)
- Level IV Fast transit roads and associated land
(Reference: European Union, 2011)
It should be noted that in the current project, the Level 1 classes were used as the basis for classification
of the Peri-Urban areas using the High Resolution (HR) data such as Landsat or Sentinel. However, for
the Core Urban areas using the Very High Resolution (VHR) data it was possible to go down to Level
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 4
III and IV. The different Levels, classes and sub-classes from the remote sensing based urban
classification, were adapted to the User requirements based on existing Master Plans for cities and/or
direct advice from the User on critical classes required. The final LU/LC nomenclature had to be
endorsed by the user before productions started.
In order to link the Urban Atlas classes described in the previous Section with the Tanzanian cities, the
Consortium used the documentation provided by the Cities for their Master Plans, and assessed which
LU/LC classes could be mapped with remote sensing and linked to the Urban Atlas nomenclature. The
merging of LU/LC classes was provided to the WB Team and users for review and endorsement. For
Kigoma, the Consortium referred to the Kigoma-Ujiji (Draft) Master Plan 2017-2037, Ministry of
Lands, Housing and Human Development, Tanzania in order to assess which classes could be mapped
with remote sensing. See Table 1 for final LU nomenclature used for the different epochs.
Table 1: LU/LC Nomenclature for 2006 and 2016.
2006 2016
Level I Level II Level III Level
IV
Level I Level II Level III Level IV
1000
Artificial
Surfaces
1100
Residential
1110 Very
Low Density
1000
Artificial
Surfaces
1100
Residential
1110 Very
Low Density
1120 Low
Density
1120 Low
Density
1130 Medium
Density
1130 Medium
Density
1140 High
Density
1140 High
Density
1150 Very
High Density
1150 Very
High Density
1200
Industrial,
Commercial,
Public,
Military,
Private and
Transport
Units
1210
Industrial,
Commercial,
Public,
Military and
Private Units
- 1200
Industrial,
Commercial,
Public,
Military,
Private and
Transport
Units
1210
Industrial,
Commercial,
Public,
Military and
Private Units
1211 Commercial
- 1212 Industry
- 1213 University
- 1214 Schools
- 1215 Government
- 1216 Military
- 1217 Hospitals
- 1218 Public
Buildings
1220
Roads
1221
Arterial
1220
Roads
1221 Arterial
1222
Collector
1222 Collector
1230
Railway
1230
Railway
1240
Airport
1240
Airport
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 5
1250
Port
1250
Port
1300 Mine,
Dump and
Construction
Sites
1310
Mineral
Extraction
and Dump
Sites
1300
Mine, Dump
and
Construction
Sites
1310
Mineral
Extraction
and Dump
Sites
1320
Construction
Sites
1320
Construction
Sites
1330
Land without
Current Use
1330
Land without
Current Use
1400 Urban
Open Spaces
1410
Urban Parks
1400
Urban Open
Spaces
1410
Urban Parks
1420
Recreation
Facilities
(Sport
Facilities,
Stadiums,
Golf Courses,
etc.)
1420
Recreation
Facilities
(Sport
Facilities,
Stadiums,
Golf Courses,
etc.)
1430
Cemeteries
1430
Cemeteries
2000
Agricultura
l Area
2000
Agricultura
l Area
3000
Natural and
Semi-
natural
Areas
3100
Forest and
Shrub Lands
3000
Natural and
Semi-
natural
Areas
3100
Forest and
Shrub Lands
3200
Natural Areas
(Savannah,
Grassland)
3200
Natural Areas
(Savannah,
Grassland)
3300 Bare
Soil
3300 Bare
Soil
4000
Wetlands
4000
Wetlands
5000
Water
5000
Water
It is important to note that the possibility to classify at Level IV was highly dependent on the availability
of reliable Reference datasets from the City or sources such as Google Earth. This aspect is further
discussed in Chapter 3.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 6
2.5 Transport Infrastructure - Road Network
The road hierarchy used in the classification is based on the international road classification standards;
this is for example defined by the European Commission (https://ec.europa.eu/transport/road_safety/-
specialist/knowledge/road/designing_for_road_function/road_classification_en).
Roads are divided into three groups-Arterial or through traffic flow routes (in our case Arterial Roads),
distributor road (in our case Collector Roads), and access roads (in our case Local Roads). The three
road types are defined as follows:
Arterial Roads:
Roads with a flow function allow efficient throughput of (long distance) motorized traffic. All
motorways and express roads as well as some urban ring roads have a flow function. The number of
access and exit points is limited. (https://ec.europa.eu/transport/road_safety/specialist/knowledge/-
road/designing_for_road_function/road_classification_en)
Collector Roads:
Roads with an area distributor function allow entering and leaving residential areas, recreational areas,
industrial zones, and rural settlements with scattered destinations. Junctions are for traffic exchange
(allowing changes in direction etc.); road sections between junctions should facilitate traffic in flowing.
(https://ec.europa.eu/transport/road_safety/specialist/knowledge/road/designing_for_road_function/roa
d_classification_en)
Local Roads:
Roads with an access function allow actual access to properties alongside a road or street. Both junctions
and the road sections between them are for traffic exchange. (https://ec.europa.eu/transport/-
road_safety/specialist/knowledge/road/designing_for_road_function/road_classification_en).
Arterial roads and collector roads were the main focus of the classification. These types of roads were
identified for the entire AoI. Within the geospatial dataset the road features can be identified within the
attribute table. A value of 1 is assigned to the arterial roads and a value of 2 to the collector lines.
Spatial Accuracy:
The Collector roads and Arterial roads are integrated within the LULC mapping by applying a buffer
around the road centre lines of 12.0 m for the arterial roads and 7.5m for collector roads. The 7.5m are
set as the maximum allowable difference of the mapped centre line in comparison to the location in the
VHR imagery.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 7
2.6 Urban Green Areas
Urban Green Areas were separated within the Land Use/Land Cover product only in the land use classes
1411 Urban Parks (Parks, Gardens) and 1412 Cemeteries. However, all other urbanised land use classes,
like 1100 Residential, 1200 Industrial, Commercial, Public, Military, Private and Transport Units, and
1300 Mine, Dump and Construction Sites can also be partially vegetated. In order to get an overall
synopsis of all Green Areas within the urbanised classed, i.e. all sub-classes under 1000 Artificial
Surfaces, a further land cover classification of the VHR data was performed by applying a semi-
automated classification algorithm. Forested areas, Grassland and Agricultural areas which are part of
the LULC product were not further considered in this Urban Green Areas land cover. The applied
nomenclature and class coding is given in Table 2.
Table 2: Nomenclature used for the mapping and identification of Urban Green Areas.
Single date
Code 0 Non-green area
Code 1 Urban green area
Code 255 Non-urban areas. All Areas that do not fall in the Level I class of the
Land Use Land Cover product.
Change product
Code 0 Non-green area. No vegetated surfaces occurring on “Artificial Surfaces”, Level I, at both points in time
Code 1 Permanent urban green area. Vegetated surfaces in 2006 and 2016.
Code 2 Loss of urban green area. Vegetated areas in 2006, which changed to non-
vegetated areas in 2016.
Code 3 New urban green area. Non-vegetated surfaces in 2006 with vegetation
cover in 2016.
Code 255 Non-Urban Areas. All Areas that do not fall in the Level I class of the
Land Use Land Cover product.
2.7 Planned and Unplanned Settlement Areas
For the classification of Planned and Unplanned Settlements only the Core Urban area was assessed.
Furthermore the LU classes that were targeted were only the residential areas, including the LULC
classes “1100 Residential” and “1211 Commercial Area.” It has to be noted that within the Commercial
Area class only those polygons with a residential component, like in the CBD were classified as planned
or unplanned settlements. The applied nomenclature is given in Table 3.
Table 3: Nomenclature used for the mapping and identification of Planned and Unplanned Settlement
Areas.
Single date
Planned settlements
Unplanned settlements
Change product
No change in planned settlement area
No change in unplanned settlement area
Expansion of planned settlement area
Expansion of unplanned settlement area
Decrease of planned settlement area
Decrease of unplanned settlement area
Unplanned to planned settlement area
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 8
2.8 Population Distribution and Density
The Population Distribution and Density product is calculated only for the residential areas (LULC class
11) within the Core Urban Districts of Kigoma. The current product is derived from globally available
WorldPop data from 2015 with spatial resolution of 100m. Additionally, the population residuals are
adjusted, using interpolation of population counts from the official population census data. The historic
product is derived entirely based on interpolation of the official population census data. The applied
nomenclature is given in Table 4.
Table 4: Nomenclature used for the mapping and identification of Population Distribution and Density.
Single date
Population Density between 0 – 150 inhabitants/ km2
Population Density between 151 – 300 inhabitants/ km2
Population Density between 301 – 600 inhabitants/ km2
Population Density between 601 – 1500 inhabitants/ km2
Population Density between 1501 – 3000 inhabitants/ km2
Population Density between 3001 – 5000 inhabitants/ km2
Population Density between 5001 – 9000 inhabitants/ km2
Population Density between 9001 – 18000 inhabitants/ km2
Population Density between 18001 - 41000 inhabitants/ km2
Change product
Unchanged Population Distribution
Up to -100% decrease
Up to 200% increase
201% - 400% increase
401% - 600% increase
601% - 800% increase
801% - 1000% increase
More than 1000% increase
2.9 Terms of Access
The Dissemination of the digital data and the Report was undertaken via FTP.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 9
3 Service Operations
The following Sections present all steps of the service operations including the necessary input data, the
processing methods, the accuracy assessment and Quality Control procedures.
3.1 Source Data
This Section presents the remote sensing and ancillary datasets that were used. Different types of data
from several data providers have been acquired. A complete list of source data as well as a quality
assessment is provided in Annex 2.
A summary of the main data used is provided in the following Sections.
High Resolution Optical EO Data
The major data sources for the Peri-Urban current and historic mapping of urban LULUC, urban extent
and imperviousness were Landsat and Sentinel-2 data which were accessible and downloadable free of
charge.
Landsat-5: As a source of historical data one scene of Landsat TM 5 from 02 September 2005
has been acquired which covers the whole area of interest.
Sentinel-2: the most recent data coverage comprises one Sentinel-2 data set. The data was
downloaded and processed at Level 1C.
Very High Resolution Optical EO Data
The VHR data for the core urban area mapping had to be acquired and purchased through commercial
EO Data Providers such as Airbus Defence and European Space Imaging.
It has to be noted that under the current collaboration project the VHR EO data had to be purchased
under mono-license agreements between GAF AG and the EO Data Providers. If EO data would have
to be distributed to other stakeholders then further licences for multiple users would have to be
purchased.
The following VHR sensor data have been acquired to cover the entire AoI:
Pleiades-1B o 2 scenes for 2016
Quickbird-2 o 4 scenes for 2005
Detailed lists of the used EO data as well as their quality is documented in the attached Quality
Control Sheets in Annex 2.
Ancillary Data
Data Provided by User:
o Kig_Rivers_v2.shp: Spatial location of rivers. A slight geometric shift in the data was
identified. Data was integrated in the LULC mapping by using the spatial location of
that data set. The polygons for the water class are based on the VHR imagery.
o KIG_roads.shp: Similar to OSM data. Data was used to verify the Open Street Map
(OSM) data, which was used as basis for the road classes.
o tanzania_health_facility_registry6322wgs84.shp: 6323 Points of Interest (POIs) on
Health Centres, Hospitals and Dispensaries. The point based information was overlaid
on the LULC mapping, but no information was available on spatial precision and
completeness.
o updated_schools_november_2014_Dodoma.shp: Location of 12723 schools in
Tanzania. Data was used for map overlay as provided by User. No information was
available on spatial precision and completeness.
Open Street Map (OSM) data: OSM data is freely available and generated by volunteers across
the globe. The so called crowd sourced data is not always complete, but has for the most parts
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 10
of the world valuable spatial information. Data was downloaded to complement the Transport
Network layer and further enhanced. The spatial location of the OSM based streets was used a
geospatial reference.
Detailed lists of the used ancillary data as well as their quality is documented in the attached Quality
Control Sheets in Annex 2.
3.2 Processing Methods
Data processing starts at an initial stage with quality checks and verification of all incoming data. This
assessment is performed in order to guarantee the correctness of data before geometric or radiometric
pre-processing is continued. These checks follow defined procedures in order to detect anomalies,
artefacts and inconsistencies. Furthermore all image and statistical data were visualised and interpreted
by operators.
The main techniques and standards used for data analysis, processing and modelling for each product
are described in Annex 1.
3.3 Accuracy Assessment of Map Products
Data and maps derived from remote sensing contain - like any other map - uncertainties which can be
caused by many factors. The components, which might have an influence on the quality of the maps
derived from EO include quality and suitability of satellite data, interoperability of different sensors,
radiometric and geometric processing, cartographic and thematic standards, and image interpretation
procedures, post-processing of the map products and finally the availability and quality of reference
data. However, the accuracy of map products has a major impact on secondary products and its utility
and therefore an accuracy assessment was considered as a critical component of the entire production
and products delivery process. The main goal of the thematic accuracy assessment was to guarantee the
quality of the mapping products with reference to the accuracy thresholds set by the user requirements.
The applied accuracy assessments were based on the use of reference data, and applying statistical
sampling to deduce estimates of error in the classifications. In order to provide an efficient, reliable and
robust method to implement an accuracy assessment, there are three major components that had to be
defined: the sampling design, which determines the spatial location of the reference data, the response
design that describes how the reference data is obtained and an analyses design that defines the accuracy
estimates. These steps were undertaken in a harmonised manner for the validation of all the geo-spatial
products.
3.3.1 The Applied Sampling Design
The sampling design specifies the sample size, sample allocation and the reference assessment units
(i.e. pixels or image blocks). Generally, different sampling schemes can be used in collecting accuracy
assessment data including: simple random sampling, systematic sampling, stratified random sampling,
cluster sampling, and stratified systematic unaligned sampling. In the current project a single stage
stratified random sampling based on the method described by Olofsson et al. (2015) was applied which
used the map product as the basis for stratification. This ensured that all classes even very minor ones
were included in the sample.
However, in complex LU/LC products with many classes, this usually results in a large number of
strata (one stratum per LU/LC classes), of which some classes cover only very small areas (e.g. sport
fields, cemeteries) and not being adequately represented in the sampling. In order to achieve a
representative sampling for the statistical analyses of the mapping accuracy it was decided to extend
the single stage stratified random sampling. At the first stage the number of required samples was
allocated within each of the Level I strata (see Table 2). In the second stage all Level III classes that
were not covered by the first sampling, were grouped into one new stratum. Within that stratum the
same number of samples was randomly allocated as the Level I strata received. To avoid a clustering
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 11
of point samples within classes and to minimise the effect of spatial autocorrelation a minimum
distance in between the sample points was set to be 150 m. The final sample size for each class can be
considered to be as close as possible to the proportion of the area covered by each stratum considering
that the target was to determine the overall accuracy of the entire map.
The total sample size per stratum was determined by the expected standard error and the estimated
error rate based on the following formula which assumes a simple random sampling (i.e. the
stratification is not considered):
n = 𝑃∗𝐸𝑧 ² n = number of samples per strata / map class
p = expected accuracy
q = 1 – p
E = Level of acceptable (allowable) sample error
Z = z-value (the given level of significance)
Hence, with an expected accuracy of p = 0.85, a 95% confidence level and an acceptable sampling
error of 5%, the minimum sample size is 196. A 10% oversampling was applied to compensate for
stratification inefficiencies and potentially inadequate samples (e.g. in case of cloudy or shady
reference data). For each Level I strata 215 samples have been randomly allocated. Afterwards, within
all classes of Level III (see Table 5) that did not received samples in the first run, additionally 215
samples were randomly drawn across all these classes.
Table 5: Number of sampling points for the EO4SD-Urban mapping classes after applied sampling
design with information on overall land cover by class.
Class Name Class
ID
No. of
Sampling
Points
Km²
coverage
Residential 1100 174 48.2
Commercial 1210 14 3.0
Roads 1220 1 2.3
Railway 1230 50 0.3
Airport 1240 3 1.7
Port 1250 1 0.1
Mining 1310 4 0.3
Construction Site 1320 2 0.3
Land without current use 1330 1 0.9
Urban Parks 1410 15 0.4
Recreational Facilities 1420 3 0.3
Cemetery 1430 1 0.2
Agriculture 2000 220 1045
Forest 3100 137 56
Natural Areas 3200 72 20.1
Bare Soil 3300 5 0.1
Wetlands 4000 210 7.6
Water 5000 216 35.8
Total -- 1129 292.9
3.3.2 The Applied Response Design
The response design determines the reference information for comparing the map labels to the reference
labels. Collecting reference data on the ground by means of intensive fieldwork is both costly and time
consuming and in most projects not feasible. The most cost effective reference data sources are VHR
satellite data with 0.5 m to 1 m spatial resolution. Czaplewski (2003) indicated that visual interpretation
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 12
of EO data is acceptable if the spatial resolution of EO data is sufficiently better compared to the
thematic classification system. However, if there are no EO data with better spatial resolution available,
the assessment results need to be checked against the imagery used in the production process.
The calculated number of necessary sampling points for each mapping category was randomly
distributed among the strata and overlaid to the VHR data of each epoch. The following Figure is
showing the mapping result with the overlaid sample points.
Figure 2: Mapping result of the city of Kigoma of the year 2016 overlaid with randomly distributed
sample points used for accuracy assessment.
In this way a reference information could be extracted for each sample point by visual interpretation of
the VHR data for all mapped classes. The size of the area to be observed had to be related to the
Minimum Mapping Unit (MMU) of the map product to be assessed. The reference information of each
sampling point was compared with the mapping results and the numbers of correctly and not-correctly
classified observations were recorded for each class. From this information the specific error matrices
and statistics were computed (see next Section).
3.3.3 The Applied Analysis Design
Each class usually has errors of both omission and commission, and in most situations, these errors for
a class are not equal. In order to calculate these errors as well as the uncertainties (confidence intervals)
for the area of each class a statistically sound accuracy assessment was implemented.
The confusion matrix is a common and effective way to represent quantitative errors in a categorical
map, especially for maps derived from remote sensing data. The matrices for each assessment epoch
were generated by comparing the “reference” information of the samples with their corresponding
classes on the map. The Reference represented the “truth”, while the Map provided the data obtained
from the map result. Thematic accuracy for each class and overall accuracy is then presented in error
matrices (see Tables below). Unequal sampling intensity resulting from the random sampling approach
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 13
was accounted for by applying a weight factor (p) to each sample unit based on the ratio between the
number of samples and the size of the stratum considered (Selkowitz & Stehman, 2011): �̂� = ( 1𝑀) ∑ 1𝜋 ℎ∗𝑥∈ ,
Where i and j are the columns and rows in the matrix, M is the total number of possible units (population)
and π is the sampling intensity for a given sample unit u in stratum h.
Overall accuracy and User and producer accuracy were computed for all thematic classes and 95%
confidence intervals were calculated for each accuracy metric.
The standard error of the error rate was calculated as follows: 𝜎ℎ = √ ℎ − ℎ𝑛ℎ where nh is the sample
size for stratum h and ph is the expected error rate. The standard error was calculated for each stratum
and an overall standard error was calculated based on the following formula: 𝜎 = √∑ 𝑤ℎ . 𝜎ℎ
In which 𝑤ℎ is the proportion of the total area covered by each stratum. The 95% Confidence Interval
(CI) is +/- 1.96.
The confusion matrices are provided within the Annex 2 and showing the mapping error for each
relevant class. For each class the number of samples which are correctly and not correctly classified are
listed, which allows the calculation of the user and producer accuracies for each class as well as the
confidence interval at 95% confidence levels based on the formulae above.
The Land Use/Land Cover product for Kigoma has an overall mapping accuracy of 97.25% with
a CI ranging from 96.26% to 98.25% at a 95% Confidence Level. The specific class accuracies
are given in Annex 2.
3.4 Accuracy Assessment of Transport Network
The road network was partially integrated in the LULC map by selecting first and second level roads.
These are the Arterial Roads and the Collector Lines. For the accuracy assessment of the Road network
it should be noted that the sampling design, response design and analyses design are different from the
one used for validating the LULC maps. The Accuracy Assessment of the Transport Network is related
to the geospatial precision of the collected and digitised centerlines of the roads.
Sampling Methodology
A systematic random sampling was applied to define the primary and secondy sampling units. Over the
entire AoI a regular grid of 450m by 450m was created. Based on these grid cells a random selection of
2% samples cells were selected. An example is given in Figure 3 with the road network in grey, the grid
cells in black and the randomly selcted cells in green.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 14
Figure 3: Example of the applied sampling design to generate randomly distributed point for the
Accuracy Assessment of the road network.
Within the randomly selected cells another grid of 150m distance was created (see Figure 4). All
intersection between the created road layer and the 150m grid were extracted as points. At all points the
road locations were visually checked and if any, the differences between spatial location on VHR
imagery and spatial location of the digitised lines recorded.
Figure 4: Secondary sampling grid to generate the sampling points at spatial intersection of roads and
grid cells. Roads are represented as white lines, grid as black lines and final sampling point as
black dots.
Overall 336 sampling points were created and their differences recorded. The result is presented in Table
6 as histogram of deviations. For the entire sampling population a Mean Difference of 1.21 m and a
Standard Deviation of 1.79 m was calculated.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 15
Table 6: Validation result of the complemented Transport Network in Kigoma, which is based on OSM
data.
Distance in m Frequency
0 174
0 - 2 75
2 - 4 60
4 - 6 17
6 – 7.5 4
Above 7.5 meters 5
By setting a maximum allowable difference of 7.5 m the distances are separated into two classes. Correct
street locations and in-correct street locations. The statistical analysis for the two classes reveal an
overall accuracy of 98.5% was achieved.
3.5 Accuracy Assessment of Urban Green Areas
The validation of the Green Area mapping results is done in a similar way as the validation for the Land
Use Land Cover product. The necessary amount of sampling points are calculated according to the
formula of Goodchild et al. (1994), which is given in Table 7.
Table 7: Calculation of the minimum number of samples according Goodchild et al. 1994.
Variables Values
p 0,85
q 0,15
E 0,05
z 1,96
196
n with 10% oversampling 215
with:
p=required accuracy of the data
q=1-p
E=Level of acceptable (allowable) sample error
z=value from table (for the given level of significance)
The calculated number of 215 sample points was randomly distributed among the entire map and
overlaid to the VHR data of each epoch. The following Figure is showing the mapping result with the
overlaid sample points.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 16
Figure 5: Result of the Urban Green Area mapping in Kigoma (change product) with sampling points
used for product validation.
At each sample point location the reference data was collected by visual interpretation of the VHR data.
The size of the area to be observed had to be related to the Minimum Mapping Unit (MMU) of the map
product to be assessed. Finally, visual interpreted land cover type was compared with the mapping
results and the numbers of correctly and not-correctly classified observations were recorded. From this
information the specific error matrices and statistics were computed.
The confusion matrices show the mapping error for each relevant class. For each class the number of
samples which are correctly and not correctly classified are listed in the Tables below. They allow the
calculation of the user and producer accuracies for each class as well as the confidence interval at 95%
confidence levels based on the formulae above. The results of the Accuracy Assessment are listed in
Table 8 and Table 9 below, for 2006 and 2016 respectively.
Table 8: Results of the Accuracy Assessment of Urban Green Areas in Kigoma, 2006
Overall Accuracy 89.8%.
2006 Reference Data
Totals 0 - Non-Urban Green Area 1 - Urban Green Area
0 - Non-Urban Green Area 122 15 137
1 - Urban Green Area 7 71 78
Totals 129 86 215
Table 9: Results of the Accuracy Assessment of Urban Green Areas in Kigoma, 2016
Overall Accuracy 90.7%
2016 Reference Data
Totals 0 - Non-Urban Green Area 1 - Urban Green Area
0 - Non-Urban Green Area 116 9 125
1 - Urban Green Area 11 79 90
Totals 127 88 215
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 17
The confusion matrices are additionally provided within the Quality Control documentation in Annex 2
and showing the mapping error for each relevant class. For each class the number of samples which are
correctly and not correctly classified are listed, which allows the calculation of the user and producer
accuracies for each class as well as the confidence interval at 95% confidence levels.
3.6 Accuracy Assessment of Planned and Unplanned Settlements
The validation of the Planned and Unplanned Settlement mapping results is done in a similar way as the
validation for the Land Use Land Cover and the Urban Green Area product. The necessary amount of
sampling points are calculated according to the formula of Goodchild et al. (1994), which is given in
Table 10.
Table 10: Calculation of the minimum number of samples according Goodchild et al. (1994).
Variables Values
p 0.85
q 0.15
E 0.05
z 1.96
196
n with 10% oversampling 215
with:
p = required accuracy of the data
q = 1-p
E = Level of acceptable (allowable) sample error
Z = value from table (for the given level of significance)
The calculated number of 215 sample points was randomly distributed among the entire Core Urban
area and overlaid on the VHR data of each epoch. The following Figure shows the mapping result with
the overlaid sample points.
Figure 6: Result of the Informal Settlement Area mapping in Kigoma (change product) with sampling
points used for product validation.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 18
At each sample point location the reference data was collected by visual interpretation of the VHR data.
The size of the area to be observed had to be related to the Minimum Mapping Unit (MMU) of the map
product to be assessed. Finally, visual interpreted land cover type was compared with the mapping
results and the numbers of correctly and not-correctly classified observations were recorded. From this
information the specific error matrices and statistics were computed.
The confusion matrices show the mapping error for each relevant class. For each class the number of
samples which are correctly and not correctly classified are listed in the Tables below. They allow the
calculation of the user and producer accuracies for each class as well as the confidence interval at 95%
confidence levels based on the formulae above. The results of the Accuracy Assessment is listed in
Table 11 for 2016.
Table 11: Results of the Accuracy Assessment of Informal Settlement Areas in Kigoma, 2016.
Overall Accuracy 96.73 %.
Settlements 2016 Reference Data
Totals Planned Settlement Area Unplanned Settlement Area
Planned Settlement Area 101 6 107
Unplanned Settlement Area 1 106 107
Totals 102 112 214
The confusion matrices are additionally provided within the Quality Control documentation in Annex 2
and showing the mapping error for each relevant class. For each class the number of samples which are
correctly and not correctly classified are listed, which allows the calculation of the user and producer
accuracies for each class as well as the confidence interval at 95% confidence levels.
3.7 Quality Control/Assurance
A detailed Quality Control and Quality Assurance (QC/QA) system has been developed which records
and documents all quality relevant processes ranging from the agreed product requirements, the different
types of input data and their quality as well as the subsequent processing and accuracy assessment steps.
The main goal of the QC/QA procedures was the verification of the completeness, logical consistency,
geometric and thematic accuracy and that metadata are following ISO standards on geographic data
quality and INSPIRE data specifications. These assessments were recorded in Data Quality Sheets
which are provided in Annex 2. The QC/QA procedures were based on an assessment of a series of
relevant data elements and processing steps which are part of the categories listed below:
Product requirements;
Specifications of input data: EO data, in-situ data, ancillary data;
Data quality checks: EO data quality, in-situ data quality, ancillary data quality;
Geometric correction, geometric accuracy, data fusion (if applicable), data processing;
Thematic processing: classification, plausibility checks;
Accuracy: thematic accuracy, error matrices
Delivery checks: completeness, compliancy with requirements
After each intermediate processing step a QC/QA was performed to evaluate products appropriateness
for the subsequent processing (see Figure 7).
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 19
Figure 7: Quality Control process for EO4SD-Urban product generation. At each intermediate processing
step output properties are compared against pre-defined requirements.
After the initial definition of the product specifications (output) necessary input data were defined and
acquired. Input data include all satellite data and reference data e.g. in-situ data, reference maps,
topographic data, relevant studies, existing standards and specifications, statistics. These input data were
the baseline for the subsequent processing and therefore all input data had to be checked for
completeness, accuracy and consistency. The evaluation of the quality of input data provides
confidence of their suitability for further use (e.g. comparison with actual data) in the subsequent
processing line. Data processing towards the end-product required multiple intermediate processing
steps. To guarantee a traceable and quality assured map production the QC/QA assessment was
performed and documented by personnel responsible for the Quality Control/Assurance. The results of
all relevant steps provided information of the acceptance status of a dataset/product.
The documentation is furthermore important to provide a comprehensive and transparent summary of
each production step and the changes made to the input data. With this information the user will be able
to evaluate the provided services and products. Especially the accuracy assessment of map products and
the related error matrices are highly important to rate the quality and compare map products from
different service providers.
The finalised QC/QA forms are attached in Annex 2.
3.8 Metadata
Metadata provides additional information about the delivered products to enable it to be better
understood. In the current project a harmonised approach to provide metadata in a standardised format
applicable to all products and end-users was adopted. Metadata are provided as XML files, compliant
to the ISO standard 19115 "Metadata" and ISO 19139 "XML Scheme Implementation". The metadata
files have been created and validated by the GIS/IP-operator for each map product with the Infrastructure
for Spatial Information in Europe (INSPIRE) Metadata Editor available at: http://inspire-
geoportal.ec.europa.eu/editor/.
The European Community enacted a Directive in 2007 for the creation of a common geo-data
infrastructure to provide a consistent metadata scheme for geospatial services and products that could
be used not only in Europe but globally. The geospatial infrastructure called INSPIRE was built in a
close relation to existing International Organization for Standardization (ISO) standards. These are ISO
191115, ISO 19119 and ISO 15836. The primary incentive of INSPIRE is to facilitate the use and
sharing of spatial information by providing key elements and guidelines for the creation of metadata for
geospatial products and services.
The INSPIRE Metadata provides a core set of metadata elements which are part of all the delivered geo-
spatial products to the users. Furthermore, the metadata elements provide elements that are necessary to
perform queries, store and relocate data in an efficient manner. The minimum required information is
specified in the Commission Regulation (EC) No 1205/2008 of 3 December 2008 and contains 10
elements:
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 20
Information on overall Product in terms of: Point of contact for product generation, date of
creation
Identification of Product: Resource title, Abstract (a short description of product) and Locator
Classification of Spatial Data
Keywords (that define the product)
Geographic information: Area Coverage of the Product
Temporal Reference: Temporal extent; date of publication; date of last revision; date of
creation
Quality and Validity: Lineage, spatial resolution
Conformity: degree of conformance to specifications
Data access constraints or Limitations
Responsible party: contact details and role of contact group/person
These elements (not exhaustive) constitute the core information that has to be provided to meet the
minimum requirements for Metadata compliancy. Each element and its sub-categories or elements have
specific definitions; for example in the element “Quality” there is a component called “Lineage” which has a specific definition as follows: “a statement on process history and/or overall quality of the spatial data set. Where appropriate it may include a statement whether the data set has been validated or quality
assured, whether it is the official version (if multiple versions exist), and whether it has legal validity.
The value domain of this element is free text,” (INSPIRE Metadata Technical Guidelines, 2013). The
detailed information on the Metadata elements and their definitions can be found in the “INSPIRE Metadata Implementing Rules: Technical Guidelines,” (2013). Each of the EO4SD-Urban products will
be accompanied by such a descriptive metadata file. It should be noted that the internal use of metadata
in these institutions might not be established at an operational level, but the file format (*.xml) and the
web accessibility of data viewers enable for the full utility of the metadata.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 21
4 Analysis of Mapping Results
This Chapter will present and assess all results which have been produced within the framework of the
current project, in the context of presentation of the Urban Extent product, the LU/LC products and the
Transport Infrastructure product. Furthermore the Sections that follow will provide the results of some
standard analytics undertaken with these products including the following:
Urban Extent – Developments from 2000, 2005, 2010 to 2016
Land Cover Land Use - Status and Trends between 2006 and 2016
Transport Infrastructure - Status and Change between 2006 and 2016
Urban Green Areas – Status and Change between 2006 and 2016
Planned and Unplanned Settlement Areas – Status and Change between 2006 and 2016
Population Distribution and Density – Status and Change between 2006 and 2016
It is envisaged that these analytics provide information on general trends and developments in the Core
and Peri-Urban areas which can then be further interpreted and used by urban planners and the City
Authorities for city planning.
It should be noted that all digital data sets for these products are provided in concurrence with this City
Report with all the related metadata and Quality Control documentation
4.1 Urban Extent – Developments 2000, 2005, 2010 and 2015
The Urban Extent product in the EO4SD-Urban project is provided by the German Aerospace Centre
(DLR) and is provided for 4 points in time; the 2015 Global Urban Footprint (GUF) Plus product has
been produced jointly exploiting multi-temporal 30m Landsat-8 and ESA Sentinel-1 data with 10m
resolution acquired in 2014-2015. And for the years 2000, 2005 and 2010, the Urban Extent products
generated – given the unavailability of freely and easily accessible multi-temporal radar data at high
resolution – were based only on multi-temporal 30m Landsat-5 and Landsat-7 imagery, and scaled up
to 10m resolution.
As the Urban Extent 2015 product was based on the ESA Sentinel-1 dataset which is a Synthetic
Aperture Radar (SAR) in C band it should be noted that some structures which are flat in nature such as
airport runways were not classified; this is due to the fact that radar relies on backscatter which is more
prominent from vertical features. The GUF+ 2015 products will be validated and available as public
domain data from October 2017 onwards on the Urban Thematic Exploitation Platform (TEP) supported
by the DLR.
In the current project the Urban Extent product for Kigoma was first used to assess historical
developments from 2000-2015. Further analysis by overlaying administrative boundaries can be
performed to assess urbanisation extent patterns based on administrative units.
Results:
The first result provided using the different Urban Extent products from 2000 to 2015 is illustrated in
Figure 6 which shows the urban development in the Core and Peri-Urban areas as well as surrounding
regions of Kigoma. The Urban Extent developments after 2000 can be examined by Urban Planners to
identify different patterns of growth such as “Edge Growth” or “Leapfrog Growth” depending on the
location of the developments.
From the depiction of urban extent developments between 2000 and 2015 in Figure 8 and Figure 9, one
can note that urban extent development occurred in Kigoma in small blocks around existing residential
areas. It stretches towards north, west and south-west in the Core Urban and mainly towards north-east
in the Peri-Urban.
The urban development till 2005 stretches towards east in very dense patches in most of the areas. For
the next years, 2005 until 2010 the residential areas stretched in small fragmented patches scattered
through the entire Core Urban, especially around the existing residential areas. Additionally, in the
eastern, north-eastern and in the very south parts of Peri-Urban the extension of residential areas between
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 22
the years 2005 and 2015 were predominant. The last time interval, 2010 until 2015, presents a very
dense extension concentrated in the northern part of the Core Urban.
Figure 8: Urban Extent developments in the epochs 2000 to 2005, 2005 to 2010 and 2010 to 2015 in Kigoma
and surrounding region.
Figure 9: Urban Extent developments in the epochs 2000 to 2005, 2005 to 2010 and 2010 to 2015 in Kigoma
within the Core Urban Area.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 23
4.2 Land Cover/ Land Use 2006 and 2016
This Section will present the results of the LU/LC mapping for 2006 and 2016 as well the statistical
information on the changes between these two epochs. The LU/LC overview map for 2016 is depicted
in Figure 10 and a cartographic version with map layout design is provided as geo-pdf files in addition
to the geo-spatial product.
Figure 10: Detailed Land Cover Land Use 2016 in Kigoma.
For the epoch 2006 the most dominant LU/LC classes occurring in the Overall area were Agriculture
(41.68% of total area), Forests (15.62%) and Residential (14.06% of total area). By 2016 these classes
remained as the most dominant in the Overall area with a slight increase of 2.37% in the Residential
class (accounting for 16.44 % of the total area by 2016), and minor increase in Forest area with 3.54%.
The Agricultural class decreased by 5.85%. Further information on the class disaggregation and area
coverage is presented in Figure 11 and Figure 12 for the epochs 2006 and 2016 respectively. Detailed
information on the area, percentage distribution and changes can be further observed in Table 12.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 24
Figure 11: Detailed Land Cover Land Use 2006 structure: Presented as Overall, Core Urban and in Peri-
Urban Zone in percentages (left) and square kilometres (right)
Similarly to the Overall area, Agriculture was the most dominant class in 2006 in both the Core and the
Peri-Urban areas with 34.33% and 44.65%, respectively. The main difference in these two areas is the
coverage and density of the Residential class. For instance, 31.56% (26.60 km2) from the Core Urban
area in 2006 was represented by the Residential class, whereas the class accounted only for 7% (14.61
km2) from the Peri-Urban area during the same year. Additionally, residential areas with high (50 – 80%
of soil sealing) and medium densities (30 – 50% of soil sealing) were predominant in the Core Urban.
Considering the Peri-Urban, residential areas are comprised of the low (10 – 30% of soil sealing) and
very low (less than 10% of soil sealing) density classes and these are scattered over larger areas.
By 2016 the land use distribution patterns within the Core Urban, Overall and Peri-Urban areas are
similar to those in 2006, except that the agricultural class decreased in all three areas. Furthermore, the
Residential class represented the largest area coverage in the Core Urban area by 2016 with 39.82%
(33.56 km2) of the area.
Figure 12: Detailed Land Cover Land Use 2016 structure: Presented as Overall, Core Urban and in Peri-
Urban Zone in percentages (left) and square kilometres (right).
The next Section will highlight the LU/LC change information between the two epochs in more detail.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 25
Description of LULC Changes:
In addition to the overall LU/LC classification for the two epochs it is interesting to assess the different
trends between classes over the 10 year time period. The quantitative figures for each class (combined
Core and Peri-Urban) are first provided in Table 12 to get an overview.
Table 12: Detailed information on area and percentage of total area for each class for 2006 and 2016 as
well as the changes.
LU/LC Classes 2016 2006 Change Change per Year
sqkm % of
total sqkm
% of
total sqkm % sqkm %
Residential 0-10% 18.56 6.34% 21.11 7.21% -2.55 -0.87% -0.26 -0.09%
Residential 10-30% 14.04 4.79% 8.98 3.07% 5.06 1.73% 0.51 0.17%
Residential 30-50% 10.51 3.59% 7.74 2.64% 2.76 0.94% 0.28 0.09%
Residential 50-80% 4.83 1.65% 3.29 1.12% 1.54 0.53% 0.15 0.05%
Residential 80-100% 0.21 0.07% 0.07 0.03% 0.14 0.05% 0.01 0.00%
Industrial, Commercial, Public, Military 2.99 1.02% 2.58 0.88% 0.41 0.14% 0.04 0.01%
Arterial Line 0.79 0.27% 0.79 0.27% 0.00 0.00% 0.00 0.00%
Toll Line 1.47 0.50% 1.43 0.49% 0.03 0.01% 0.00 0.00%
Railway 0.28 0.10% 0.28 0.10% 0.00 0.00% 0.00 0.00%
Airport 1.74 0.59% 1.43 0.49% 0.31 0.11% 0.03 0.01%
Port 0.09 0.03% 0.09 0.03% 0.00 0.00% 0.00 0.00%
Mining Area 0.28 0.10% 0.11 0.04% 0.17 0.06% 0.02 0.01%
Construction Site 0.32 0.11% 0.30 0.10% 0.02 0.01% 0.00 0.00%
Land Without Current Use 0.89 0.30% 0.84 0.29% 0.05 0.02% 0.00 0.00%
Urban Parks 0.38 0.13% 0.16 0.06% 0.21 0.07% 0.02 0.01%
Recreation Facilities 0.34 0.11% 0.30 0.10% 0.04 0.01% 0.00 0.00%
Cemeteries 0.21 0.07% 0.20 0.07% 0.01 0.00% 0.00 0.00%
Agricultural Area 104.98 35.84% 122.11 41.68% -17.13 -5.85% -1.71 -0.58%
Forest 56.11 19.15% 45.76 15.62% 10.36 3.54% 1.04 0.35%
Natural areas (non-forested) 30.39 10.37% 36.23 12.37% -5.84 -1.99% -0.58 -0.20%
Bare Soil 0.09 0.03% 0.05 0.02% 0.03 0.01% 0.00 0.00%
Wetlands 7.62 2.60% 3.28 1.12% 4.34 1.48% 0.43 0.15%
Water 35.82 12.23% 35.79 12.22% 0.03 0.01% 0.00 0.00%
Total 292.94 100.00% 292.94 100.00%
The area statistics of the LU/LC classes show little change dynamics within residential density classes.
The changes mostly happened in the medium and low density classes with an increase in the 10 - 30%
and 30 - 50% density classes and decrease in area of the 0 - 10% class. Main area loss in the 10 year
period happened in agricultural (loss of 17.13% of the total area) and natural areas/non-forested (loss of
5.84% of total area). On the other hand, main area gain in the 10 year period happened in forested areas.
4.2.1 Spatial Distribution of Main LU/LC Change Categories
In order to better analyse the growth trend and the spatial distribution of changes meaningful
aggregations of the LU/LC classes in both epochs were used. The following categories were developed:
Urban Densification: Changes from lower Residential Density Class into a higher Residential
Density class;
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 26
Urban Residential Extension: all changes from Non-Urban Residential classes to a Residential
class;
Other Urban Land Use Extension: all changes from Non-Residential Urban classes to Other
Urban and Non-Urban classes.
Changes within Natural and Semi-Natural Areas: all changes in between the natural and semi-
natural classes (e.g. Forest into agriculture).
The overlay analysis of these aggregated categories of the epochs 2006 and 2016 is depicted in Figure
13.
Figure 13: Land Cover Land Use Change Types - Spatial Distribution
The spatial distribution of the change types as depicted in Figure 13 shows that the urban densification
in Kigoma mainly happened along the main roads and was predominant in the north-eastern part of the
Peri-Urban area of the city. The extension of the Residential areas (red colour in Figure 13) occurred
mainly in the north, east and south of the Core Urban area following the pattern of urban densification.
The changes between the natural and semi-natural classes occurred expressively in the north and south
of the Peri-Urban area and some parts on the north and east of the Core area.
The statistics of the change categories are presented in Figure 14 and Table 13. Figure 14 provides
quantitative aspects to the LU change classes; for example the largest changes occurred between the
natural and semi-natural areas with 53.66% of the Overall area and 82.89% in the Peri-Urban. For the
Core Urban, the densification of urban areas is the dominant change type with 38.30% as presented in
Table 13. Furthermore, extension of residential areas was the second largest change 35.35% in the Core
Urban.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 27
Figure 14: Land Cover Land Use Change Types 2006-2016 - Overall, in Core Urban and in Peri-Urban in
% (left) and km² (right) in Kigoma.
As the Figure 14 and Table 13 show, urban related changes were significantly presented in the entire
area. Few areas in the Peri-Urban Zone were used for residential extension and densification therefore
land conversion into urban related classes are almost exclusively processes related to the Core Urban
areas.
Table 13: Overall LU/LC Statistics.
Change Classes Change Overall Change Core
Urban Change Peri-Urban
sqkm % sqkm % sqkm %
Urban Densification 11.01 18.75% 7.67 38.30% 3.34 13.80%
Urban Residential Extension 7.90 17.86% 7.07 35.35% 0.82 3.40%
Other Urban Land Use
Extension 1.59 3.60% 1.59 7.95% 0.00 0.01%
Change within Natural and
Semi-Natural Areas 23.73 53.66% 3.68 18.41% 20.05 82.80%
Total 44.23 93.87% 20.01 100.00% 24.21 100.00%
4.2.2 Changes of Agricultural Areas
In order to analyse the relatively large loss of Agricultural areas as noted in the earlier part of Section
4.2, further change analysis was performed. The following change categories were developed for this
analysis:
Agriculture to Residential area;
Agriculture to Industry, Commercial; Public or Military area;
Agriculture to Plantations.
The spatial distribution of these changes is displayed in Figure 15.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 28
Figure 15: Spatial distribution of changes from Agricultural Areas into other Classes between 2006 and
2016.
The conversion of agricultural areas into other LULC classes was mainly caused by natural or artificial
reforestation in the Peri-Urban area. This conversion type is shown in Figure 15 in green colour and as
can be seen exclusively in the northern part, large agricultural areas were replaced by forest. Within the
Core Urban, as noted in blue colour in Figure 15, the conversion of agricultural areas into residential
had especially occurred in those areas along main roads which are scattered through-out the entire city.
The third change class in Figure 15 depicts changes from agricultural to industrial or commercial land
use which represents few changes but especially in the northern part of the Core Urban of Kigoma (see
Figure 15, red colour) these type of changes occurred.
Figure 16: Changes of Agricultural Areas into other LU classes between 2006 and 2016; Presented as
Overall, Core Urban and in Peri-Urban in % (left) and km2 (right).
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 29
The statistics in Figure 16 and Table 14 show that the majority of agricultural areas in the Core Urban
area has been converted into residential area in the ten year period between 2006 and 2016. In the Peri-
Urban, the agricultural areas mainly were converted into forested land. As the absolute number of the
agricultural areas converted into residential areas within the Overall Area and the Core Urban (5.84 km²
and 5.57 km², respectively) are similar, it can be noted that the overall conversion into residential area
occurs mainly in the Core Urban area of Kigoma.
Table 14: Statistics of changes of agricultural areas between 2006 and 2016.
Change Classes Change Overall Change Core Urban Change Peri-Urban
km² % km² % km² %
Agricultural Area to
Residential 5.84 35.01% 5.57 79.09% 0.27 2.79%
Agricultural Area to
Industrial,
Commercial, Public
or Military Area
0.36 2.15% 0.36 5.09% 0.00 0.00%
Agricultural Area to
Forest 10.47 62.84% 1.11 15.81% 9.36 97.21%
Total 16.67 100.00% 7.04 100.00% 9.63 100.00%
4.3 Transport Network
The Transport Network was created for both points in time (2006 and 2016) using three levels for the
road type classification. The Arterial roads and Collector roads were integrated in the LULC map by
applying a buffer of 12 m and 8 m for the Arterial and Collector roads, respectively. Local roads are
only part of the vector data set, which are provided to the user.
Figure 17 depicts the Transport Network for both points in time. The left figure presents the Transport
Network in 2006 and the right Figure the Transport Network in 2016. The railway is shown as well. The
main changes of the Transport Network occurred in terms of densification of Local roads in 2016 in the
southern part and northern parts of the Core Urban area. Primary roads and Secondary roads remained
the same through the ten years period.
Figure 17: Transport Network of Kigoma in 2006 (left) and 2016 (right).
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 30
4.4 Urban Green Areas
The location and extent of green areas are determined within the product of urban land use/ land cover
at Level I. Urban green areas refer to land within and on the edges of a city that is partly or completely
covered with grass, trees, shrubs, or other vegetation. The product delivered provides accurate
information (1 m resolution) on the spatial location and extent of green areas located within the Urban
Extent (Level I class: 1000) derived from the baseline LULC information product. Detecting and
monitoring urban green coverage needs very high resolution optical satellite images, which explains the
product generation over the Core Urban Area of Kigoma only.
The overview map in Figure 18 shows loss, gain and stable urban green areas in Kigoma. The map is
delivered as separate product of high resolution for printing at paper size DIN A0, which is 84.1 cm x
118.9 cm. Figure 18 gives an overview of the mapping result and the structure of the map, which is
designed for a print out at DIN A0 paper size (84.1 x 118.9 cm).
Figure 18: Map overview of Urban Green Areas in Kigoma. Green area loss, gain and stable green areas
can be identified. The map is delivered as separate product of high resolution for printing at
paper size DIN A0, which is 84.1 cm x 118.9 cm.
The spatial distribution of the change types as depicted in Figure 19 show that changes are distributed
across the entire Core Urban area of Kigoma, whereas most of the non-green areas can be found in the
denser built up parts of the city. Overall 18.51% of the entire area was covered by vegetation in 2006
and 2016. The percentage of gain of green area (17.14%) is a little bit higher than the percentage of loss
of green area (13.62%) resulting in a general increase of urban green areas in Kigoma. About 50.72%
of the artificial areas (Level I class) were not covered by vegetation in 2006 and 2016.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 31
Figure 19: Percentage of urban green areas within the core area of Kigoma. The pie chart illustrates the
status and change of urban green areas in-between 2006 and 2016.
The changes within the urban green areas can be illustrated as overall area coverage. Figure 20
represents the green area coverage in square kilometres in 2006 and 2016, compared to the amount of
the artificial areas class. Both classes demonstrated a relatively equal growth within the 10 year period.
Urban areas showed an increase of 4.3 km² (from 23.02 km² to 27.32 km²) and green areas an increase
of 3.91 km² (from 10.90 km² to 14.81 km²).
Figure 20: Bar charts for both points in time presenting the total area of urban greenery versus non-green
areas.
4.5 Planned and Unplanned Settlement Areas
The location and extent of planned and unplanned settlement areas are determined within the classes
“1100 Residential” and “1211 Commercial Areas” of the urban land use/ land cover product. Polygons
with the LU class “Commercial Areas” were only included for classification in this product, when they
also contained residential buildings.
In an effort to use spatial patterns to depict the class of “planned settlements” (see Figure 21) there was
a focus on residential areas where the houses are oriented in the same direction, similar in size and shape
and have a well-defined road network, often in a grid pattern which provides almost direct access to
every house. For “unplanned settlements” (see Figure 22) houses are not oriented in the same direction
50.72%
18.51 %
13.62 %
17.14 %
non green stable green green loss green gain
0
5
10
15
20
25
30
2006 2016
area
in k
m²
urban (km2) green (km2)
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 32
and streets, if existing are curved, small and appear to have been constructed in an ad hoc manner (no
distinct patterns). Also not every building can be reached by a road.
Figure 21: Planned settlement areas in Kigoma in 2016.
Figure 22: Unplanned settlement areas in Kigoma in 2016.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 33
To further support the classification effort, the Kigoma-Ujiji Master Plan 2017-2037, Draft Report
(2016) page 49 was used to orient in locating of planned and unplanned settlement areas. There was also
an information exchange with the World Bank (WB) team on potential locations of planned/unplanned
settlements in Kigoma; feedback on locations of these 2 settlements was obtained from MaryGrace
Weber of the WB Team after her trip to Kigoma and in consultation with the local Authorities (received
on 18th May 2017); this information was also included. The input is illustrated in Figure 23 and
highlights areas as planned and unplanned to support the interpretation process.
Figure 23: Planned and unplanned settlement areas received via email from MaryGrace Weber.
Due to the complexity of identifying these classes, the detection and monitoring planned and unplanned
settlements needs VHR optical satellite images; thus the product was generated only over the Core
Urban Area of Kigoma.
The overview map in Figure 24 shows the spatial distribution of planned and unplanned areas between
the two years 2006 and 2016 in Kigoma. The map is delivered as separate product of high resolution for
printing at paper size DIN A0, which is 84.1 cm x 118.9 cm. gives an overview of the mapping result
and the structure of the map, which is designed for a print out at DIN A0 paper size (84.1 x 118.9 cm).
Figure 24: Map overview of changes in planned and unplanned settlement areas in Kigoma during the
years 2006 and 2016
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 34
From the Figure it can be seen that planned settlements are mainly located in the inner city, to the north,
south-west and south-east. Only the surrounding residential areas have unplanned settlements.
An analysis of the changes is depicted in Figure 25. Overall about 40% of the residential area is and
remained planned settlement (40.46%) in 2006 and 2016. Only 27.78% of the whole residential area is
unplanned settlement. However, during the 10 year period, more unplanned settlements (18.79%)
developed than planned settlements (10.05%).
Figure 25: Percentage of planned and unplanned areas within the core area of Kigoma. The pie chart
illustrates the status and change of planned and unplanned areas in-between 2006 and 2016.
The changes in the overall area coverage for both the planned and unplanned settlement areas are
illustrated in Figure 26; the bar chart represents the planned and unplanned settlement areas in square
kilometres in 2006 and 2016. Both classes increased overall in area within the 10 years: in 2006 planned
settlements covered an area of 13.11km2, and this area extended in 2016 to 15.77km2; unplanned
settlements covered 13.70km2 in 2006, and this settlement type increased in area in the following 10
years more than the planned settlement type to 18.09 km2.
Figure 26: Bar charts for both points in time presenting the total area of planned and unplanned settlement
areas.
27.78%
40.46%
18.79%
2.91%0.89%
10.05%
No Change in Unplanned Settlement Area No Change in Planned Settlement Area
Expansion of Unplanned Settlement Area Expansion of Planned Settlement Area
Decrease of Unplanned Settlement Area Unplanned to Planned Settlement Area
0
2
4
6
8
10
12
14
16
18
20
2006 2016
Are
a [
km
2]
Unplanned Settlement Area Planned Settlement Area
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 35
4.6 Population Distribution and Density
The Population Distribution and Density Change product contains spatial explicit information about
population distribution within the Core Urban Districts of Kigoma. The product is derived for all units
which belong to the residential class (LULC class 11), using five different data sources: 1) Land
Use/Land Cover Baseline Product; 2) WorldPop data (2015) with a spatial resolution of 100m; 3)
Official Population census data from 2002 and 2012; 4) Imperviousness/Soil Sealing layer (see Annex
1 for a full description of this product) and 5) Administrative boundaries. An important note is that the
product has been always developed for the period 2005 – 2015, due to data availability (the global
WorldPop dataset was only available for the year 2015), and to meet the user requirements, regarding
time interval of 10 years between the historic and current state of the product.
The overview map in Figure 27 shows eight different change classes which were identified, based on a
frequency distribution analysis of the changes that occur: Unchanged Population Distribution; Up to –100% decrease; Up to 200% increase; 201% - 400% increase; 401% - 600% increase; 601% - 800%
increase; 801% - 1000% increase; More than 1000% increase. The Unchanged Population Distribution
class is defined as the residential units which experienced stable population distribution between 2005
and 2015, with values within the overall annual population growth rate (-4.09% to 4.09%). The map is
delivered as separate product of high resolution for printing at paper size DIN A0, which is 84.1 cm x
118.9 cm. It has to be noted that the GIS data set allows map printouts at any other scale which however
would require different map sheet layouts and cartographic designs.
Figure 27: Overview Map of Population Distribution Change in Kigoma (2005 – 2015).
The spatial distribution of the change types as depicted in Figure 27 can be grouped in 3 main categories:
1) unchanged areas; 2) areas which experienced population decrease and 3) areas which increased in
terms of population distribution at a different degree. The overview map shows that most of the
residential units in Kigoma experienced increase of up to 200% (depicted in light yellow colour). The
decrease in population (depicted in green colour on the map) is mainly observed on the south of the city
in the Bangwe district, which is closely linked by the Decrease of Unplanned Settlement Areas and the
conversion of Unplanned to Planned Settlement Areas, as observed in Figure 24.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 36
On the other hand, the highest degree of increase (More than 1000%, shown in dark brown colour) is
observed mainly on the north east (in Buhanda and Businde districts), and also in the very north part of
Bangwe district. Based on the analysis of the mapping results, several conclusions can be drawn which
justify the trends of extreme population growth in the above mentioned 3 districts, such as: 1) residential
expansion in Buhanda and Businde districts (see Figure 13); 2) urban densification in Bangwa (see
Figure 13); 3) conversion of lots of agricultural plots into residential areas, especially in Buhanda and
Businde districts (see Figure 15); 4) existence of unplanned settlement in Buhanda and Businde districts;
5) expansion of Unplanned Settlements, especially in Buhanda district (see Figure 24); 6) attractiveness
of the areas in the very north part of Bangwa district, which are located closer to the Central Business
District of Kigoma and the port.
An analysis of the change types depicted in Figure 28, shows that 55.97% of the area was subject to
population increase of up to 200%, followed by areas which experienced decrease in population
distribution (12.07% from the total area). The third most common class was represented by areas which
experienced population growth between 201% and 400% (11.31% from the total area). Only 2.28%
remained unchanged (within the range of the annual growth rate of -4.09% to 4.09%)
Figure 28: Population Distribution Change within the Core Urban Districts of Kigoma between 2005 and
2015.
Figure 29 provides further detail on the population distribution changes between 2005 and 2015, in
relation to build up areas which are based on the amount of soil sealing degree in each Core Urban
District.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 37
Figure 29: Changes in Population Distribution, in relation to build up areas or soil sealing degree in
Kigoma between 2005 and 2015.
For instance, it is noted that in Kibirizi (the second biggest district in terms of population) more than
9000 people moved to build up areas with a sealing degree of 30 – 50%. Similarly, a high proportion of
the residents in Gungu, Mwanga Kaskazini and Mwanga Kusini moved to areas with sealing degree of
30 – 50%. The population growth in most of the Core Urban Districts was mainly distributed among the
Very Low to Medium Density classes. A common trend is observed in the Very High Density class (80
– 100%) as well, which decreased in most of the districts. This phenomena can be driven by different
factors, such as: rent price in central districts, lack of land availability, and conversion of residential
units to commercial, industrial, central business districts in the city centre, etc. However, the information
presented in Figure 29 cannot explicitly define those driving forces. Therefore, further research is needed
to identify and justify the drivers behind such phenomena, thus helping urban planners and local officials
in the decision – making process.
4.7 Concluding Points
This Chapter 4 presented only a summary and overview of what is possible in term of analytics with the
geo-spatial datasets provided for Kigoma in the current project. This Report is a living document and
will be complemented with further analysis during the project. Important would be to further analyse
the EO4SD Urban datasets with the City Master Plan for Kigoma in order to enhance the latter for
planning purposes.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
EO4SD-Urban Kigoma City Operations Report Page 38
5 References
Kigoma-Ujiji Draft Master Plan 2017-2037 (2016). Ministry of Lands, Housing and Human
Development, Tanzania.
Czaplewski, R. L. (2003). Chapter 5: Accuracy assessment of maps of forest condition: statistical design
and methodological considerations, pp. 115–140. In Michael A. Wulder, & Steven E. Franklin (Eds.),
Remote sensing of forest environments: concepts and case studies. Boston: Kluwer Academic Publishers
(515 pp.).
European Union (2011). Mapping Guide for a European Urban Atlas, Version 11.0
Goodchild, M., Chih-Chang, L. and Leung, Y. (1994): Visualizing fuzzy maps, pp. 158-67. In
Heamshaw, H.H. and Unwin, D.J. (Eds.), Visualization in geographical information systems.
Chichester: Wiley.
Olofsson, P., Foody, G. M., Stehman, S. V., & Woodcock, C. E. (2013). Making better use of accuracy
data in land change studies: Estimating accuracy and area and quantifying uncertainty using stratified
estimation. Remote Sensing of Environment, 129, 122–131. doi:10.1016/j.rse.2012.10.031
Selkowitz, D. J., & Stehman, S. V. (2011). Thematic accuracy of the National Land Cover Database
(NLCD) 2001 land cover for Alaska. Remote Sensing of Environment, 115(6), 1401–1407.
doi:10.1016/j.rse.2011.01.020.
Internet
Road classification, European Commission, 2017, https://ec.europa.eu/transport/road_safety/specialist-
/knowledge/road/designing_for_road_function/road_classification_en, last accessed 2017.08.17
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
Annex 1 to EO4SD-Urban Kigoma City Operations Report Page 1
Annex 1 – Processing Methods for EO4SD-Urban Products
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
Annex 1 to EO4SD-Urban Kigoma City Operations Report Page 2
This page is intentionally left blank!
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
Annex 1 to EO4SD-Urban Kigoma City Operations Report Page 3
Summary of Processing Methods
Urban and Peri-Urban Land Use/Land Cover and Change
The input includes Very High Spatial Resolution (VHR) imagery from different sensors acquired at
different time. The data is pre-processed to ensure a high level of geometric and radiometric quality
(ortho-rectification, radiometric calibration, pan-sharpening).
The complexity when dealing with VHR images comes from the internal variability of the information
for a single land-use. For instance, an urban area is represented by a high number of heterogeneous pixel
values hampering the use of automated pixel-based classification techniques.
For these VHR images, it is possible to identify textures (or pattern) inside an entity such as an
agricultural parcel or an urban lot. In other words, whereas pixel-based techniques focus on the local
information of each single pixel (including intensity / DN value), texture analysis provides global
information in a group of neighbouring pixels (including distribution of a group intensity / DN values
but also spatial arrangement of these values). Texture and spectral information are combined with a
segmentation algorithm in an Object Based Image Analysis (OBIA) approach to reach a high degree of
automation for most of the Peri-Urban rural classes. However, within urban land, land use information
is often difficult to obtain from the imagery alone and ancillary/in situ data needs to be used. The
heterogeneity and format of these data mean that another information extraction method based on
Computer Aided Photo-Interpretation techniques (CAPI) need to be used to fully characterise the LULC
classes in urban areas. Therefore, a mix of automated (OBIA) and CAPI are used to optimise the
cost/quality ratio for the production of the LULC/LUCC product. The output format is typically in vector
form which makes it easier for integration in a GIS and for subsequent analysis.
Level 4 of the nomenclature can be obtained based on additional information. These can be generated
by more detailed CAPI (e.g. identification of waste sites) or by an automated approach based on
derived/additional products. An example is illustration by categorising the density of the urban fabric
which is related to population density and can then subsequently used for disaggregating population
data.
Information on urban fabric density can be obtained through several manners with increasing level of
complexity. The Imperviousness Degree (IMD) or Soil Sealing (SL) layer (see separate product) can be
produced relatively easily based on the urban extent derived from the LULC product and a linear model
between imperviousness areas and vegetation vigour that can be obtained from Sentinel 2 or equivalent
NDVI time series. This additional layer can be used to identify continuous and discontinuous urban
fabric classes. Five urban fabric classes can be extracted based on a fully automated procedure:
Continuous urban fabric (IMD > 80%)
Discontinuous dense urban fabric: (IMD 50-80 %)
Discontinuous medium density urban fabric (IMD: 30-50 %)
Discontinuous low density urban fabric (IMD 10-30 %)
Discontinuous very low density urban fabric (IMD < 10 %)
Manual enhancement is the final post-processing step of the production framework. It will aim to
validate the detected classes and adjust classes’ polygon geometry if necessary to ensure that the correct MMU is applied. Finally, a thorough completeness and logical consistency check is applied to ensure
the topological integrity and coherence of the product.
Change detection: Four important aspects have to be considered to monitor land use/land cover change
effectively with remote sensing images: (1) detecting that changes have occurred, (2) identifying the
nature of the change, (3) characterising the areal extent of the change and (4) assessing the spatial pattern
of the change.
The change detection layer can be derived based on an image-to-image approach provided the same
sensor is used. An original and efficient image processing chain is promoted to compare two dates’ images and provide multi-labelled changes. The approach mainly relies on texture analysis, which has
the benefits to deal easily with heterogeneous data and VHR images. The applied change mapping
approach is based on spectral information of both dates’ images and more accurate than a map-to-map
comparison.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
Annex 1 to EO4SD-Urban Kigoma City Operations Report Page 4
Summary of Processing Methods
Urban Extent and Change
Reliably outlining urban areas is of high importance since an accurate characterization of the urban
extent is fundamental for accurately estimating, among others, the population distribution, the use of
resources (e.g., soil, energy, water, materials), infrastructure and transport needs, socioeconomic
development, human health and food security. Moreover, monitoring the change in the extent of urban
areas over time is of great support for properly modelling the spatial-temporal patterns of urbanisation
evolution and, thus, better estimating future trends and implementing suitable planning strategies.
The product is a binary mask outlining in the area of interest the urban areas (intended as built-up
structures) with respect to all other land-cover classes merged together into a single information class.
The urban class and the non-urban class are associated with value “255” and “0”, respectively. Regrouping of relevant LULC thematic classes can be used to depict urban extent precisely. Instead, if
a detailed LULC product is not available for the selected study region, then the information will be
derived by the following approach.
The product is generated at 30 m spatial resolution by properly exploiting Landsat-4/5/7/8 multi-
temporal imagery acquired over the peri-urban and urban area within a given time interval of interest in
which no relevant changes are expected to occur (typically a time period of 1-2 years allows to obtain
very accurate results). For all the considered scenes, cloud masking and, optionally, atmospheric
correction are performed. Next, a series of features specifically suitable for delineating urban areas are
derived for each image. These include both spectral indexes (e.g., the normalized different vegetation
index (NDVI), the atmospherically resistant vegetation index (ARVI), the normalized difference water
index (NDWI), etc.) and texture features (e.g., occurrence textures, co-occurrence texture, local
coefficient of variation, etc.). The core idea is then to compute per each pixel key temporal statistics for
all the extracted features, like temporal maximum, minimum, mean, variance, median, etc. This allows
compressing all the information contained in the different multi-temporal acquisitions, but at the same
time to easily and effectively characterize the underlying dynamics. It is worth noting that for different
pixels in the study area, different number of scenes might be available. However, in the hypothesis of a
sufficient minimum number of acquisitions for computing consistent statistics, this does not represent
an issue. Moreover, in this framework it is also possible to obtain spatially consistent datasets to be
employed for the desired analyses even when investigating large areas. Training data for the urban and
non-urban class are then extracted by employing a strategy based on the analysis of the DLR Global
Urban Footprint (GUF) layer (which varies depending whether the target period of interest refers to a
time interval before or after that which the GUF refers to). Afterwards, a Support Vector Machines
(SVM) classifier is employed where a Radial Basis Function (RBF) kernel is used.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
Annex 1 to EO4SD-Urban Kigoma City Operations Report Page 5
Summary of Processing Methods
Transport Network
The transportation network is mainly manually digitised and mapped on the basis of very high resolution
(VHR) optical satellite imagery. Requested features are obtained by integration of auxiliary data such
as OpenStreetMap (OSM) or local datasets as a starting point for the product generation. The revision
and update of the auxiliary data is realised by using up-to-date VHR data. The workflow can be specified
as follows:
Quality check of available ancillary data such as local data sets.
Processing of optical satellite data – dependent on satellite data product level (geometric,
atmospheric and radiometric corrections, enhancements – colour optimization, mosaicking,
tiling).
Identification, collection and integration of available ancillary data (e.g. Open Street Map)
Identification and adjustment of spatial inconsistencies. The OSM data is used as spatial
reference. Upon User request other data sets can be used.
Update of the network by visual photo-interpretation according road hierarchy (see description
below).
Update of attributes by photointerpretation.
Generalization, application of MMU (minimum allowable dangling length)
Quality control and accuracy assessment
o Statistical sampling of check points
o Independent evaluation of products (second interpreter, third party assessment)
Change detection
The road hierarchy used in the classification is based on the international road classification
standards. One definition is specified by the European Commission. Roads are divided into three
groups - arterial or through traffic flow routes (in our case Arterial Roads), distributor road (in our
case Collector Roads), and access roads (in our case Local Roads). The three road types are defined
as follows: Arterial Roads - roads with a flow function allow efficient throughput of (long distance)
motorized traffic. All motorways and express roads as well as some urban ring roads have a flow
function. The number of access and exit points is limited. Collector Roads - roads with an area
distributor function allow entering and leaving residential areas, recreational areas, industrial zones,
and rural settlements with scattered destinations. Local Roads - roads with an access function allow
actual access to properties alongside a road or street. Arterial roads and collector roads were the
main focus of the classification. These types of roads were identified for the entire Area of Interest.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
Annex 1 to EO4SD-Urban Kigoma City Operations Report Page 6
Summary of Processing Methods
Urban Green Areas
The location and extent of green areas are determined within the product of urban land use/ land cover
at Level I. Urban green areas refer to land within and on the edges of a city that is partly or completely
covered with grass, trees, shrubs, or other vegetation. This includes public parks, private gardens,
cemeteries, forested areas as well as trees, river alignments, hedges etc. The product delivered within
EO4SD-Urban project thus provides accurate information (1 m resolution) on the spatial location and
extent of the green areas located within the Urban Extent (Level I class: 1000) derived from the baseline
LULC information product.
Detecting and monitoring urban green coverage needs very high resolution optical satellite images,
which explains the product generation over the Core Urban Area of AOI only. The same images have
been logically used for generating the LULC information product. Consequently, the usual preliminary
quality check and pre-processing tasks were already implemented.
Urban Green Areas have been detected using automated non-supervised classification method. More
precisely, each single multispectral VHR scene has been classified by specifying the most appropriate
algorithm and class number. Then, pixel units from the classes considered as representing green areas
have been combined into 1 single class. From this operation results the required binary raster product.
At this stage, it only remains necessary to apply some post-processing steps:
Morphological filter is applied to fill small gaps within the green areas (caused by shadow)
Resampling of the data to the provided spatial resolution of 1m
Removing small pixel groups under the minimum mapping unit.
Integrating the information provided by the LULC product (e.g. class Urban Parks, Cemeteries).
Validation of Mapping results
Furthermore, using archive very high resolution images, current and historic extent of urban green areas
are compared to identify their temporal evolution – extent growth or reduction. Quality control and
accuracy assessment tasks are performed by means of visual interpretation considering also the LULC
dataset.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
Annex 1 to EO4SD-Urban Kigoma City Operations Report Page 7
Summary of Processing Methods
Planned and Unplanned Settlements
The location and extent of planned and unplanned settlement areas are determined within the classes
“1100 Residential” and “1211 Commercial Areas” of the urban land use/ land cover product. Polygons
with the LU class “Commercial Areas” were only included for classification in this product, when they also contained residential buildings.
Two distinct between the two types following rules were applied:
In planned settlement areas (see Figure 1 left side) houses are oriented in the same direction, are
similar in size and shape and show a good road network reaching every house.
In unplanned settlements (see Figure 1 right side) houses are not oriented in the same direction
and streets, if existing are curved and small. Not every building can be reached by a road.
Figure 1: Planned (left) and unplanned (right) settlement areas in Kigoma in 2016.
To further improve the results, the City Urban Master plans were used to help in the distinction in
planned and unplanned settlement areas. Detecting and monitoring planned and unplanned settlements
need very high resolution (VHR) optical satellite images; therefore this product was only generated over
the Core Urban Area of Kigoma only. As the same images VHR images were used for generating the
LULC information product, the usual preliminary quality check and pre-processing tasks were already
implemented. Planned and unplanned areas have been detected by visual interpretation of the actual and
historic VHR image by following the rules described above and with the help of the City Master Plans
of the three cities.
Quality control and accuracy assessment tasks are also performed by means of visual interpretation
considering also the LULC dataset.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
Annex 1 to EO4SD-Urban Kigoma City Operations Report Page 8
Summary of Processing Methods
Population Distribution and Density
This product aims at providing information on the spatial distribution of residents in a specified Area of
Interest (AOI). Additionally, it depicts: a) location of people; b) estimated population in a specific spatial
unit; and c) changes in population distribution throughout the years.
Data sources and types
To derive the Population Distribution and Density product, five main ancillary data types are used: 1)
Land Use/Land Cover Baseline Product; 2) WorldPop data; 3) Population census data; 4)
Imperviousness/Soil Sealing layer produced by DLR and 5) Administrative boundaries.
1) LULC product is used to extract the residential units within the Core Urban Districts of the city.
2) WorldPop data is a globally available dataset with spatial resolution of 100m, which provides
disaggregated population counts at a specific spatial unit (on a pixel level) for the year 2015.
Therefore, this dataset is used to interpolate the population counts for each residential unit in
the current state of the product, whereas the historic state is modelled, based on soil sealing
information and official population census data.
3) Population census data is further applied to improve the quality of the current state of the product
by adjusting population residuals, and to extrapolate the population size in the historic state of
the product.
4) Imperviousness/Soil Sealing layer has been produced by the German Space Agency (DLR). The
product provides for each pixel identified as urban the corresponding estimated Percentage
Impervious Surface (PIS). When used with the population data this PIS layer supports the
disaggregation of the population counts from district level to pixel level, using the mean sealing
degree for each urban fabric class.
5) Administrative boundaries (district level) are downloaded from the Global Administrative
Areas Website. Furthermore, they are clipped with the specified AOI and used to inform the
total number of inhabitants in each district, and to estimate the overall population density on a
district level.
Methodology
This section describes the general structure of the methodology which is used to compute Population
Distribution and Density for the specific study period. It covers the following two parts: 1) Interpolation
and projection of the population data; and 2) Spatial disaggregation and adjustment of the population
residuals.
1. Interpolation and projection of the population data
The first step from this methodological approach is to interpolate and to project the total number of
inhabitants per district for the specific reference year, based on official population census data. To
support this analysis, population census datasets from 2002 and 2012 are acquired from the Integrated
Public Use Microdata Series (IPUMS) and the United Republic of Tanzania, Ministry of East African
Cooperation, in order to estimate the population size in the Core Urban Districts of Kigoma for 2005
and 2015, respectively(IPUMS; MEAC, 2013) . First, the overall annual population growth rate is
calculated based on the total population size, using the following formula as in Kindu et al. (2015): 𝑃 = 𝑃 𝑒 (1)
where: 𝑃0 is the total population in 2002, 𝑃 is the total population in 2012,
represents the number of years between the two periods,
is the average annual growth rate.
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
Annex 1 to EO4SD-Urban Kigoma City Operations Report Page 9
Next, the average annual population growth rate per district is derived, based on the available census
data (2002 – 2012). Thus, the population size for each district is projected and estimated for 2015, using
equation (1). Finally, the population size in each district in 2005 is interpolated, based on the average
annual population growth rate per district or the overall annual growth rate (when census data for the
initial reference year (2002) was not available).
The population census datasets are used as a supportive parameter in the computation of the current state
of the product, in order to re-distribute the population residuals, created from the interpolation of the
WoldPop data. In contrast, it is the main source for population disaggregation in the historic state, since
no historic WorldPop data is available.
2. Spatial disaggregation and adjustment of the population residuals
In order to improve the quality of the current state of the product (based on WorldPop data), and to
interpolate the historic, further disaggregation analysis is conducted, following the general approach of
Batista e Silva et al. (2013): 𝐾 = 𝑃𝑆∑ 𝑈𝑐∗𝑆𝑐 𝑐 (2)
where: 𝐾 represents the number of inhabitants per unit (%) of sealed surface in each district,
Ps accounts for the total population in the specific district, based on the official census data,
Uc is the number of urban pixels (in raster data), or the sum of class area within the district (in
a vector data),
Sc is the mean soil sealing degree of each urban fabric class c.
Next, the estimated number of residents for each urban fabric class c (𝑃𝑐 is derived, as in Batista e
Silva et al. (2013): 𝑃𝑐 = 𝑘 ∗ 𝑆𝑐 (3)
Finally, a dasymetric population technique is applied to disaggregate the number of residents per spatial
unit and to adjust the population residuals. This approach takes into consideration the residential
polygons from the LULC product, denoted as ‘target’ zone, and the number of inhabitants per district, based on the official census data as ‘source’ zone, using the following formula (Batista e Silva, et al. 2013, p.17): 𝑃 ’ = 𝑃 ∗ 𝐴𝑖∗𝑊𝑖∑𝑖 𝐴𝑖∗𝑊𝑖 (4)
where: 𝑃𝑖′ refers to the estimated population in the target zone i, 𝑃 is the known population in the ‘source’ zone s
Ai is the area of the ‘target’ zone polygons
Wi is the weighting parameter related to the population density of the ‘target’ zone. n corresponds to the number of transitional polygons within each source polygon.
Output product
The output product is delivered in a vector format, presenting number of inhabitants and population
density (inhabitants/sqkm) for each residential spatial unit from the LULC product.
Validation
The final product, which is developed by a disaggregated procedure such as the one herein described is
never less accurate than the original source data (JRC Technical Report, 2013). Further detail on the
quality and accuracy of the product is directly linked to the JRC Technical Report (2013), as follows:
‘By disaggregating numerical data from one coarse geometry to a finer geometry, we always gain detail
and approximate ground truth without the risk of deteriorating the source information. The degree to
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
Annex 1 to EO4SD-Urban Kigoma City Operations Report Page 10
which the disaggregation approximates reality, however, varies greatly, and it depends chiefly on: 1) the
quality of the ancillary data and 2) the appropriateness of the disaggregation algorithm and its
parameters.’
Reference
Batista e Silva, F., Gallego, J. & Lavalle, C. (2013). A high-resolution grid map for Europe. Journal of
Maps, 9(1), 16-28.
Batista e Silva, F. (DG JRC), Poelman, H. (DG Regional Policy), Martens, V. (DG Regional Policy),
Lavalle, C. (DG JRC). (2013). " Population Estimation for the Urban Atlas Polygons" Joint
Research Center (JRC) Technical Reports.
Kindu, M., Schneider, T., Teketay, D., & Knoke, T. (2015). Drivers of land use/land cover changes in
Munessa-Shashemene landscape of the south-central highlands of Ethiopia. Environmental
monitoring and assessment, 187(7), 452.
Minnesota Population Center. Integrated Public Use Microdata Series, International (IPUMS –
International): Version 6.5. Tanzania – Population and Housing Census 2002. Minneapolis:
University of Minnesota, 2017. http://doi.org/10.18128/D020.V6.5.
United Republic of Tanzania, Ministry of East African Cooperation (MEAC). (2013). 2012
POPULATION AND HOUSING CENSUS TANZANIA. Population Distribution by
Administrative Areas. Downloaded from: http://meac.go.tz/
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
Annex 2 to EO4SD-Urban Kigoma City Operations Report Page 1
Annex 2 – Filled Quality Control Sheets
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
Annex 2 to EO4SD-Urban Kigoma City Operations Report Page 2
This page is intentionally left blank!
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 1
Earth Observation for Sustainable Development - Urban
Quality Assurance and Quality Control Sheets These QA/QC Templates were prepared by GAF AG compliant with ISO 9001:2008 Quality Management System
standards and can only be used by Partners in the current EO4SD-Urban Project.
Project Title: EO4SD-Urban
Project Leader: GAF AG
Service Provider: GAF AG Editor: AB, JF
Client: WB Date: 24.04.2018
Product: Urban Land Use/ Land Cover
Overview of QC-Sheets and Processing Steps Sheet used
Sheet filled in
Requirements
0.1 Requirements Yes Yes
Specifications of Input Data
1.1 List of EO Data Yes Yes
1.2 List of In-situ Data Yes Yes
1.3 List of Ancillary Data Yes Yes
Data Quality Checks
2.1 EO Data Quality Yes Yes
2.2 In-situ Data Quality Yes Yes
2.3 Ancillary Data Quality Yes Yes
Pre-Processing of EO Data
3.1 Geometric Correction Yes Yes
3.1.1 Data Fusion Yes Yes
3.2 Data Processing Yes Yes
Thematic Processing
4.1 Classification Yes Yes
4.2 Intermediate Quality Control of Land Use Data Yes Yes
Accuracy Assessment
5.1 Thematic Accuracy Yes Yes
5.2 Error Matrices Yes Yes
Delivery Checks / Delivery
6.1 Completeness Yes Yes
6.2 Compliancy Yes Yes
Glossary (index numbers in the QA/QC tables refer to the glossary at the end of this document) Further QC-relevant Documents:
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 2
Comments / Characteristics:
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 3
0.1 Requirements
Product 1 (28) Urban Land Use/ Land Cover and Change
Abstract
Land Use/Land Cover (LU/LC) information product contains spatial explicit information on different land use and land cover occurring in both the Core and Peri-Urban areas of the City of Kigoma. The Core area has detailed LU/LC nomenclature that is either at Level 3 or 4 whereas the Peri-Urban area LU/LC nomenclature is at an aggregated Level 1 or 2. The input data for the Core area was the Very High Resolution (VHR) data of Pleiades (2016), Quickbird (2005) and the input data for the Peri-Urban area was Sentinel-2 (2016), Landsat 5 (2005). The LU/LC product is the Baseline Product from which various derived products (such as Green Areas and Informal Settlements) are produced.
Service / Product Specifications
Area Coverage
Country: Tanzania A) Wall-to-wall:
City: Kigoma Selected Sites: 1
Area km² Core Urban: 84 B) Sampling based: n/a
Peri-Urban: 209
Time Period - Update Frequency
A) Baseline Year(s): B) Update Frequency
2006 and 2016 2 points in time
Comments:
Geographic Reference System
WGS 84 UTM zone 35S
Mapping Classes and Definitions
Residential (Built-Up Areas)
Built-up areas and their associated land, such as gardens, parks, planted areas and non-surfaced public
areas and the infrastructure, if these areas are not suitable to be mapped separately with regard to the
minimum mapping unit size.
Very High: Average degree of soil sealing: 80 -100% Residential buildings, roads and other artificially
surfaced areas.
High: Average degree of soil sealing: > 50 - 80% Residential buildings, roads and other artificially surfaced
areas.
Medium: Average degree of soil sealing: > 30 - 50% Residential buildings, roads and other artificially
surfaced areas. The vegetated areas are predominant, but the land is not dedicated to forestry or
agriculture.
Low: Average degree of soil sealing: 10 - 30% Residential buildings, roads and other artificially surfaced
areas. The vegetated areas are predominant, but the land is not dedicated to forestry or agriculture.
Very Low: Average degree of soil sealing: <10 % Residential buildings, roads and other artificially surfaced
areas. The vegetated areas are predominant, but the land is not dedicated to forestry or agriculture.
Example: exclusive residential areas with large gardens.
Commercial* Warehouses, CBD, shopping malls, markets, other commercial facilities
Industrial* Factories, … and associated land
University* University and associated land
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 4
Schools* Schools and associated land including sport fields.
Government* Governmental Buildings.
Military* Military and associated land.
Hospitals* Hospitals and associated land.
Public Buildings*
“Big” public buildings like churches, bibliotheca, etc.
Arterial Line Highways, connecting the city with other cities.
Collector Line Bigger Connecting roads within the city.
Railway Railway facilities including stations, cargo stations and associated land.
Airport Administrative area of airports, mostly fenced. Included are all airport installations: runways, buildings and associated land.
Port Port and associated area.
Mining Area and Dump Sites
Open pit extraction sites (sand, quarries) including water surface, if < Minimum Mapping Unit (MMU), open-cast mines, inland salinas, oil and gas fields.
Construction sites
Spaces under construction or development, soil or bedrock excavations for construction purposes or other earthworks visible in the image.
Land without current use
Areas in the vicinity of artificial surfaces still waiting to be used or re-used. The area is obviously in a transitional position, “waiting to be used”.Waste land, removed former industry areas, (“brown fields”) gaps in between new construction areas or leftover land in the urban context (“green fields”). No actual agricultural or recreational use. No construction is visible, without maintenance, but no undisturbed fully natural or semi-natural vegetation (secondary ruderal vegetation).
Urban Parks
Public green areas for predominantly recreational use such as gardens, zoos, parks, castle parks. Suburban natural areas that have become and are managed as urban parks. Forests or green areas extending from the surroundings into urban areas are mapped as green urban areas when at least two sides are bordered by urban areas and structures, and traces of recreational use are visible.
Recreation facilities
All sports and leisure facilities including associated land, whether public or commercially managed: Golf courses, Sports fields (also outside the settlement area), Camp grounds, Riding grounds, Racecourses, Amusement parks, Swimming resorts etc., Glider or sports airports.
Cemeteries Cemeteries and associated area.
Agricultural Area
Cultivated areas non-irrigated or permanently irrigated including rice fields: arable land (annual crops), permanent crops, complex or mixed cultivation, orchards; pasture and meadow under agricultural use, grazed or mechanically harvested.
Forest and Shrub lands
High woody vegetation in natural forests; transitional woodland; low vegetation cover with bushes and shrubs.
Natural Areas Natural area where there is little vegetation and does not serve as construction site.
Bare Soil Natural Area with no vegetation.
Wetlands Areas flooded or liable to flooding during a large part of the year by fresh, brackish or standing water with specific vegetation coverage made of low shrub, semi-ligneous or herbaceous species; shallow water areas covered with reed.
Water Visible water areas like lakes, rivers, ponds (natural, artificial).
Comment: * These classes have to be map only in 2015. For the mapping of the historic land cover these classes were merged to the class Commercial, Public, Military and Private Units.
Cloud and Cloud Shadow Detection and Removal
Information for the entire AOI is required and therefore clouded areas have to be replace by other Earth Observation (EO) data.
Spatial Resolution
n.a. (Product provided as Shapefile)
Minimum Mapping Unit (MMU)
Minimum Mapping Unit is 0.25 ha for the urban area, 0.5 ha for the peri-urban area.
Data Type & Format
Shapefile *.shp and GeoPDF
Bit Depth
n.a.
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 5
Class Coding
Class Code Class Name RGB Code
1100 Residential (Built-Up Areas) 255; 0; 0 (main class only)
1211 Commercial Area 197; 0; 255
1212 Industrial Area 132; 0; 168
1213 University 194; 158; 215
1214 Schools 232; 190; 255
1215 Government 0; 255; 197
1216 Military 115; 115; 0
1217 Hospitals 115; 223; 255
1218 Public Buildings 192; 252; 234
1221 Arterial Line 78; 78; 78
1222 Collector Line 120; 120; 120
1230 Railway 52; 52; 52
1240 Airport 168; 0; 132
1250 Port 0; 168; 132
1310 Mining Area and Dump Sites 115; 76; 0
1320 Construction sites 255; 115; 223
1330 Land without current use 242; 242; 242
1410 Urban Parks 85; 255; 0
1420 Recreation facilities 255; 170; 0
1430 Cemeteries 230; 230; 0
2000 Agricultural Area 255; 235; 175
3100 Forest and Shrub lands 38; 115; 0
3200 Natural Areas 180; 215; 158
3300 Bare Soil 204; 204; 204
4000 Wetlands 76; 0; 115
5000 Water 0; 112; 255
Metadata
Provided as INSPIRE conformant *.xml data set, covering at least the mandatory elements.
Service / Product Quality
Thematic Accuracy
Overall Accuracy: > 80%
Positional Accuracy
RMSE < 15 m
Delivery Procedure
Service Provision
Online via FTP.
Delivery Date
End of April 2018
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 6
1.1 List of EO Data
Sensoren (8) Sentinel-2, Landsat 5, PLX, QB-2
Incoming Date
Acquisition Date
Proc. Level
Path / Row
AOI - City / Region / Country
Spatial Res. No. of Bands
Cloud Cover (Data Provider)
Projection / Spheroid (16)
Data Format (3)
Bit Depth (5)
Header / Metadata (2)
File Name [e.g yymmdd; tbd...]
Sentinel-2
1. S2A_OPER_MSI_L1C_TL_MPS__20160722T115748_A005652_T35MQQ
23.03.2017
22.07.2016
1B 35/MQQ
Kigoma, TZA
10m 13 3.9%
WGS 84 / UTM zone 35S
.tiff 16 Bit
.xml
2. S2A_OPER_MSI_L1C_TL_MPS__20160722T115748_A005652_T35MRQ
23.03.2017
22.07.2016
1B 35/MRQ
Kigoma, TZA
10m 13 0%
WGS 84 / UTM zone 35S
.tiff 16 Bit
.xml
Landsat-5
3. LT51720632005245JSA00 14.03.2017
02.09.2005
L1 168/ 064
Kigoma, TZA
30m 5 0% WGS 84 / UTM zone 36
.tiff 8 Bit
.txt
Pleiades 1B
4. IMG_PHR1B_MS_201606110819501_SEN_2226863101-002_R1C1
17.03.2017
11.06.2016
primary R1C1 Kigoma, TZA
MUL: 2m; PAN: 0.5m
4 0% WGS84 .tiff 16 Bit
.xml
5. IMG_PHR1B_P_201606110819501_SEN_2226863101-001_R1C1
17.03.2017
11.06.2016
primary R1C1 Kigoma, TZA
MUL: 2m; PAN: 0.5m
4 0% WGS84 .tiff 16 Bit
.xml
Quickbird-2
6. 05APR26083225-M2AS_R1C1-056358854030_01_P001
31.03.2017
26.04.2005
LV2A R1C1 Kigoma, TZA
MUL: 2m; PAN: 0.6m
4 0% UTM 36S GeoTIFF
16 Bit
.img,
.rpb,
.til, xml
7. 05APR26083225-M2AS_R1C2-056358854030_01_P001
31.03.2017
26.04.2005
LV2A R1C2 Kigoma, TZA
MUL: 2m; PAN: 0.6m
4 0% UTM 36S GeoTIFF
16 Bit
.img,
.rpb,
.til, xml
8. 05APR26083225-M2AS_R2C1-056358854030_01_P001
31.03.2017
26.04.2005
LV2A R2C1 Kigoma, TZA
MUL: 2m; PAN: 0.6m
4 0% UTM 36S GeoTIFF
16 Bit
.img,
.rpb,
.til, xml
9. 05APR26083225-M2AS_R2C2-056358854030_01_P001
31.03.2017
26.04.2005
LV2A R2C2 Kigoma, TZA
MUL: 2m; PAN: 0.6m
4 0% UTM 36S GeoTIFF
16 Bit
.img,
.rpb,
.til, xml
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 7
1.2 List of In-situ Data
Dataset 1 (14) Incoming Date Acquisition Date AOI - City / Region / Country
Data Type (4)
Projection / Spheroid (16)
No. of Sample Plots
Sampling Design (22)
Positional Accuracy (11)
Purpose (9)
No In-situ Data Used
Lineage:
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 8
1.3 List of Ancillary Data
Dataset 1 (14)
Schools Date of Receipt from Client
Metadata (2)
Reference Mapping Date / Date of Creation
Geographic Area/- City / Region / Country
Area Coverage
Data Type (4)
Data Format (3)
Projection / Spheroid (16)
Positional Accuracy (11)
No. of Classes
Thematic Accuracy
Availability of Class Definitions
updated_schools_november_2014.shp
12.04.2016 not
available 2014/n.a. Tanzania (country level) 100% Vector *.shp GCS_WGS_1984 unknown n.a. unknown No
Lineage(29): List of schools in Tanzania.
Source (30):: Received from MaryGrace Weber.
Dataset 2 (14)
Health Facilities
Date of Receipt from Client
Metadata (2)
Reference Mapping Date / Date of Creation
Geographic Area / City / Region / Country
Area Coverage
Data Type (4)
Data Format (3)
Projection / Spheroid (16)
Positional Accuracy (11)
No. of Classes
Thematic Accuracy
Class Definitions
tanzania_health_facility_registry6322wgs84.shp
12.04.2016 not
available n.a. / n.a. Tanzania (country level) 100% Vector *.shp GCS_WGS_1984 unknown n.a. unknown No
Lineage(29): List of Health Facilities.
Source (30):: Received from MaryGrace Weber.
Dataset 4 (14)
Roads Date of Receipt from Client
Metadata (2)
Reference Mapping Date / Date of Creation
Geographic Area/- City / Region / Country
Area Coverage
Data Type (4)
Data Format (3)
Projection / Spheroid (16)
Positional Accuracy (11)
No. of Classes
Thematic Accuracy
Availability of Class Definitions
KIG_Roads 21.12.2016 not
available n.a./n.a.
Kigoma (municipal level)
80% Vector *.shp Arc_1960_UTM_Zone_35S
unknown n.a. unknown No
Lineage(29): Road network of Kigoma
Source (30):: Received from MaryGrace Weber.
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 9
Dataset 3 (14)
OSM Data Date of Receipt from Client
Metadata (2)
Reference Mapping Date / Date of Creation
Geographic Area / City / Region / Country
Area Coverage
Data Type (4)
Data Format (3)
Projection / Spheroid (16)
Positional Accuracy (11)
No. of Classes
Thematic Accuracy
Class Definitions
OSM_Data_Kigoma_Tanz
ania n.a.
not available
n.a. /n.a. Kigoma (municipality boundary)
100% Vector *.shp WGS 84 unknown n.a.
unknown n.a.
Lineage(29): OSM data were used as basis for the road network as well as for the classification. Railway and Rivers were also extracted and enhanced.
Source (30): https://www.openstreetmap.org/search?query=kigoma#map=13/-4.8841/29.6451
Dataset 4 (14)
River Network
Date of Receipt from Client
Metadata (2)
Reference Mapping Date / Date of Creation
Geographic Area / City / Region / Country
Area Coverage
Data Type (4)
Data Format (3)
Projection / Spheroid (16)
Positional Accuracy (11)
No. of Classes
Thematic Accuracy
Class Definitions
KIG_Rivers_v2.shp
n.a. not
available n.a. /06.07.2017
Kigoma 100% Vector *.shp Arc_1960_UTM_Zone_35S
unknown n.a.
unknown n.a.
Lineage(29): River lines in and around Kigoma city. Due to missing Metadata the lineage of the data set and its accuracy/completeness are unknown.
Source (30): Received from MaryGrace Weber.
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 10
2.1 EO Data Quality
Sensoren(8) Sentinel-2, Landsat 5, QB-2, PLX
Ba
cku
p
Re
ad
ab
ility
(1
)
Ch
eck
He
ad
er
/
Me
tad
ata
(2)
Band Specifications
Pro
ject
ion
/ S
ph
ero
id (1
6)
Sce
ne
Lo
cati
on
Co
mp
lete
ne
ss o
f
Ad
dit
ion
al D
ata
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Clo
ud
Co
ver
(in
tern
al
che
ck)
Ra
dio
me
try
To
po
gra
ph
y
Dro
pp
ed
Lin
es
/ A
rte
fact
s (7
)
Acc
ep
tan
ce S
tatu
s
Comment
s File Name [e.g yymmdd; tbd...]
No
. o
f B
an
ds
Ba
nd
Re
gis
tra
tio
n
Sp
ect
ral/
Sp
ati
al
Re
solu
tio
n
Sentinel-2
1. S2A_OPER_MSI_L1C_TL_MPS__20160722T115748_A005652_T35MQQ
Y - Q Y - V Q Y - V Q Y - V Q Y - V Q Y - Q Y - V Y - V Y - V Q Y - Q Y - Q
Y - V Q
Y - V Q
Y - V
N - V Q
Yes none
2. S2A_OPER_MSI_L1C_TL_MPS__20160722T115748_A005652_T35MRQ
Y - Q Y - V Q Y - V Q Y - V Q Y - V Q Y - Q Y - V Y - V Y - V Q Y - Q Y - Q
Y - V Q
Y - V Q
Y - V
N - V Q
Yes none
Landsat-5
3. LT51720632005245JSA00 Y - Q Y - V Q Y - V Q Y - V Q Y - V Q Y - Q Y - V Y - V Y - V Q Y - Q Y - Q
Y - V Q
Y - V Q
Y - V
N - V Q
Yes none
Pleiades 1B
4. IMG_PHR1B_MS_201508280834575_SEN_2226864101-002
Y - Q Y - V Q Y - V Q Y - V Q Y - V Q Y - Q Y - V Y - V Y - V Q Y - Q Y - Q
Y - V Q
Y - V Q
Y - V
N - V Q
Yes none
5. IMG_PHR1B_MS_201606110819501_SEN_2226863101-002
Y - Q Y - V Q Y - V Q Y - V Q Y - V Q Y - Q Y - V Y - V Y - V Q Y - Q Y - Q
Y - V Q
Y - V Q
Y - V
N - V Q
Yes none
Quickbird-2
6. 05APR26083225-M2AS-056358854030_01_P001
Y - Q Y - V Q Y - V Q Y - V Q Y - V Q Y - Q Y - V Y - V Y - V Q Y - Q Y - Q
Y - V Q
Y - V Q
Y - V
N - V Q
Yes none
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 11
2.2 In-situ Data Quality
Dataset 1 (14)
Ba
cku
p
Re
ad
ab
ility
(1)
Ch
eck
He
ad
er
/
Me
tad
ata
(2)
Ext
en
t
Pro
ject
ion
/ S
ph
ero
id
(16
)
Sp
ati
al R
eso
luti
on
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Lo
cati
on
Co
mp
lete
ne
ss
Ge
om
. M
isa
lign
me
nt
Pla
usi
bili
ty
Dro
pp
ed
Lin
es
/ A
rte
fact
s (7
)
Acc
ep
tan
ce S
tatu
s
Comments
File Name [e.g yymmdd; tbd...]
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 12
2.3 Ancillary Data Quality
Dataset 1 (14) LU Map
Re
ad
ab
ility
(1)
Ch
eck
He
ad
er
/Me
tad
ata
(2)
Da
ta C
ove
rage
Ma
tch
es
Se
rvic
e
Are
a (
%)
Pro
ject
ion
/ S
ph
ero
id
, E
PS
G(1
6)
Sp
ati
al R
eso
luti
on
an
d u
nit
(e
.g.
m, k
m)
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Co
mp
lete
ne
ss
(Ve
cto
r: A
ttri
bu
te
Ta
ble
; R
ast
er:
Th
em
ati
c V
alu
es)
Ge
om
. M
isa
lign
me
nt
Dro
pp
ed
Lin
es
/ A
rte
fact
s (
EO
da
ta
on
ly)(
7)
Uti
lity
for
Cu
rre
nt
Pro
ject
File Name [e.g yymmdd; tbd...]
updated_schools_november_2014.shp
☒ Yes
☐ No
☐ Complete (INSPIRE/ISO19119)
☐ Incomplete
☒ not available
☒ Yes
☐ No
☐ Unknown If no: xxx%
☒ Correct
☐ Incorrect
☐ Unknown EPSG: 32749
n.a. *.sh
p n.a.
☒ Complete
☐Incomplete
☐ No
☐ Yes
☒ Unknown If yes: XX m
☐ No
☐ Yes
☒ n.a. If yes: xxx %
☐ None
☒ Partial
☐ Full
Comments: Due to missing Metadata the Accuracy and Completeness of the data set is unknown.
Dataset 2,3,4 (14)
Re
ad
ab
ility
(1)
Ch
eck
He
ad
er
/Me
tad
ata
(2)
Da
ta C
ove
rage
Ma
tch
es
Se
rvic
e A
rea
(%
)
Pro
ject
ion
/ S
ph
ero
id
EP
SG
(16
)
Sp
ati
al R
eso
luti
on
an
d u
nit
(e
.g.
m, k
m)
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Co
mp
lete
ne
ss
(Ve
cto
r: A
ttri
bu
te
Ta
ble
; R
ast
er:
Th
em
ati
c V
alu
es)
Ge
om
. M
isa
lign
me
nt
Dro
pp
ed
Lin
es
/ A
rte
fact
s (
EO
da
ta
on
ly)(
7)
Uti
lity
for
Cu
rre
nt
Pro
ject
File Name [e.g yymmdd; tbd...]
tanzania_health_facility_registry6322wgs84.shp
☒ Yes
☐ No
☐ Complete (INSPIRE/ISO19119)
☐ Incomplete
☒ not available
☒ Yes
☐ No
☐ Unknown If no: %
☒ Correct
☐ Incorrect
☐ Unknown EPSG: 32749
n.a. *.sh
p n.a.
☒ Complete
☐Incomplete
☐ No
☐ Yes
☒ Unknown If yes: XX m
☐ No
☐ Yes
☒ n.a. If yes: xxx %
☐ None
☒ Partial
☐ Full
KIG_Roads ☒ Yes
☐ No
☐ Complete (INSPIRE/ISO19119)
☐ Incomplete
☒ not available
☒ Yes
☐ No
☐ Unknown If no: %
☒ Correct
☐ Incorrect
☐ Unknown EPSG: 32749
n.a. *.sh
p n.a.
☐ Complete
☒Incomplete
☐ No
☐ Yes
☒ Unknown If yes: XX m
☐ No
☐ Yes
☒ n.a. If yes: xxx %
☐ None
☒ Partial
☐ Full
OSM_Data_Kigoma_Tanzania ☐ Yes
☐ No
☐ Complete (INSPIRE/ISO19119)
☐ Incomplete
☒ not available
☒ Yes
☐ No
☐ Unknown If no: %
☒ Correct
☐ Incorrect
☐ Unknown EPSG: 4326
n.a. *.sh
p n.a.
☐ Complete
☒Incomplete
☐ No
☐ Yes
☒ Unknown If yes: XX m
☐ No
☐ Yes
☒ n.a. If yes: xxx %
☐ None
☒ Partial
☐ Full
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 13
Kig_Rivers_v2.shp ☒ Yes
☐ No
☐ Complete (INSPIRE/ISO19119)
☒ Incomplete
☒ not available
☒ Yes
☐ No
☐ Unknown If no: %
☒ Correct
☐ Incorrect
☐ Unknown EPSG: 32749
n.a. *.sh
p n.a.
☐ Complete
☒Incomplete
☐ No
☐ Yes
☒ Unknown If yes: XX m
☐ No
☐ Yes
☒ n.a. If yes: xxx %
☐ None
☒ Partial
☐ Full
Comments:
tanzania_health_facility_registry6322wgs84.shp: Due to missing Metadata the Accuracy and Completeness of the data set is unknown. KIG_Roads.shp: Due to missing Metadata the Accuracy and Completeness of the data set is unknown.
Kig_Rivers_v2.shp: Due to missing Metadata the Accuracy and Completeness of the data set is unknown. OSM_Data_Kigoma_Tanzania: Were used in addition to the Road layers received from the city of Kigoma. They were visually checked and corrected
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 14
3.1 Geometric Correction
Sensoren (8) Sentinel-2, Landsat 5, QB-2, PLX
Pro
cess
ing D
ate
AO
I
Cit
y /
Re
gio
n /
Co
un
try
Pro
ject
ion
/ S
ph
ero
id (1
6)
No
. &
RM
S (
m)
of
GC
Ps
(17
)
No
. &
RM
S (
m)
of
TP
s (1
8)
No
. &
RM
S (
m)
of
CP
s (1
9)
Dig
ita
l E
leva
tio
n M
od
el
(DE
M)
Mo
de
l /
Alg
ori
thm
(13
)
Re
sam
plin
g
Me
tho
d (2
0)
Va
lida
tio
n R
ep
ort
s
Acc
ep
tan
ce S
tatu
s
Output File Name [e.g yymmdd; tbd...] N
o. File Name [e.g yymmdd; tbd...]
Sentinel-2
1. S2A_OPER_MSI_L1C_TL_MPS__20160722T115748_A005652_T35MQQ
12.04.2017
Kigoma TZA
UTM35S / WGS84
N/A N/A N/A N/A N/A N/A N/A Yes S2A_OPER_MSI_L1C_TL_MPS__20160722T115748_A0
05652_T35MQQ
2. S2A_OPER_MSI_L1C_TL_MPS__20160722T115748_A005652_T35MRQ
12.04.2017
Kigoma TZA
UTM35S / WGS84
N/A N/A N/A N/A N/A N/A N/A Yes S2A_OPER_MSI_L1C_TL_MPS__20160722T115748_A005652_T35MRQ
Landsat-5
3. LT51720632005245JSA00_MTL_toa 12.04.2017
Kigoma TZA
UTM36N/ WGS84
N/A N/A N/A N/A N/A N/A N/A Yes LT51720632005245JSA00_MTL_toa
Pleiades 1B
4. DIM_PHR1B_MS_201508280834575_SEN_2226864101-002_toa_PSH
12.04.2017
Kigoma TZA
UTM35S / WGS84
MGCPs 12; RMSE x0,79; y0,69
4; RMSE x0,73; y 0,88
N/A SRTM
30
0 order poynom
CC Yes Yes oDIM_PHR1B_MS_201508280834575_SEN_2226864101-002_toa_PSH
5. DIM_PHR1B_MS_201606110819501_SEN_2226863101-002_toa_PSH
12.04.2017
Kigoma TZA
UTM35S / WGS84
MGCPs 12; RMSE x0,79; y0,69
4; RMSE x0,73; y 0,88
N/A SRTM 30
0 order poynom
CC Yes Yes oDIM_PHR1B_MS_201606110819501_SEN_2226863101-002_toa_PSH
Quickbird-2
6. 05APR26083225-M2AS-056358854030_01_P001_toa_PSH
12.04.2017
Kigoma TZA
UTM35S / WGS84
MGCPs 7; RMSE x1,07; y0,89
N/A N/A SRTM 30
Affin, 1st order poynom
CC Yes Yes o05APR26083225-M2AS-056358854030_01_P001_toa_PSH
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 15
3.1.1 Data Fusion
Dataset 4 (14) Input A Input B Output
Acc
ep
tan
ce S
tatu
s
Comments
Filename DIM_PHR1B_MS_201606110819501_SEN
_2226863101-002 DIM_PHR1B_P_201606110819501_SEN
_2226863101-001 DIM_PHR1B_MS_201606110819501_SEN_222
6863101-002_toa_PSH
none
Sensor Pleiades 1B Pleiades 1B Pleiades 1B
Method PCI / Pansharp2 PCI / Pansharp2 PCI / Pansharp2
Spatial Resolution / (MS/Pan)
2m MS 0,5m Pan 0,5m MS
Band Combination RGB NIR PAN RGB NIR
Data Format (3) & Bit Depth (5)
*.tif 16 Bit u *.tif 16 Bit u *.pix 16 Bit u Yes
Dataset 5 (14) Input A Input B Output
Acc
ep
tan
ce S
tatu
s
Comments
Filename DIM_PHR1B_MS_201508280834575_SEN
_2226864101-002 DIM_PHR1B_P_201508280834575_SEN
_2226864101-001 DIM_PHR1B_MS_201508280834575_SEN_222
6864101-002_toa_PSH
none
Sensor Pleiades 1B Pleiades 1B Pleiades 1B
Method PCI / Pansharp2 PCI / Pansharp2 PCI / Pansharp2
Spatial Resolution / (MS/Pan)
2m MS 0,5m Pan 0,5m MS
Band Combination RGB NIR PAN RGB NIR
Data Format (3) & Bit Depth (5)
*.tif 16 Bit u *.tif 16 Bit u *.pix 16 Bit u Yes
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 16
Dataset 7 (14) Input A Input B Output
Acc
ep
tan
ce S
tatu
s
Comments
Filename 05APR26083225-M2AS-
056358854030_01_P001 05APR26083225-P2AS-
056358854030_01_P001 05APR26083225-M2AS-
056358854030_01_P001_toa_PSH
none
Sensor Quickbird-2 Quickbird-2 Quickbird-2
Method PCI / Pansharp2 PCI / Pansharp2 PCI / Pansharp2
Spatial Resolution / (MS/Pan)
2m MS 0,6m Pan 0,6m MS
Band Combination RGB NIR PAN BGR NIR
Data Format (3) & Bit Depth (5)
*.tif 16 Bit u *.tif 16 Bit u *.pix 16 Bit u Yes
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 17
3.2 Data Processing
Sensoren (8)
Landsat 5, QB-2, PLX Atmospheric Correction
Radiometric Processing (15)
Topographic Normalisation
OTHER Calculation:
Acc
ep
tan
ce S
tatu
s
Ba
cku
p
Comment
No.
File Name [e.g yymmdd; tbd...] Processing
Date
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
Sentinel-2
S2A_OPER_MSI_L1C_TL_MPS__20160722T115748_A005652_T35MQQ
10.04.2017 No N/A No N/A No N/A No N/A Yes Yes none
S2A_OPER_MSI_L1C_TL_MPS__20160722T115748_A005652_T35MRQ
10.04.2017 No N/A No N/A No N/A No N/A Yes Yes none
Landsat-5
3. LT51720632005245JSA00 11.04.2017 No N/A No N/A No N/A Yes GAFmap /
TOA Yes No none
Pleiades 1B
4. DIM_PHR1B_MS_201508280834575_SEN_2226864101-002
11.04.2017 No N/A No N/A No N/A Yes GAFmap / TOA
Yes No none
5. DIM_PHR1B_MS_201606110819501_SEN_2226863101-001
11.04.2017 No N/A No N/A No N/A Yes GAFmap / TOA
Yes No none
Quickbird-2
6. 05APR26083225-M2AS-056358854030_01_P001
11.04.2017 No N/A No N/A No N/A Yes GAFmap / TOA
Yes No none
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 18
4.1 Classification
Sensors (8) Sentinel-2, Landsat, Pleiades, QB
Processing Date
Cloud Masking Thematic
Classification Manual
Enhancement Mosaicking
Acc
ep
tan
ce S
tatu
s
Ba
cku
p
Comment No
. File Name [e.g yymmdd; tbd...]
pro
cess
ed
Software /
Method
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
pro
cess
ed
Software /
Method
1.
S2A_OPER_MSI_L1C_TL_MPS__20160722T115748_A005652_T35MQQ S2A_OPER_MSI_L1C_TL_MPS__20160722T115748_A005652_T35MRQ
09.06.2017
no No clouds
Yes
eCognition/semi-automatic approach
Yes
eCognition and ArcGIS 10 / Visual Interpretation
Yes
ArcGIS 10 / Mosaic Dataset with manual neatline selection
Yes
Yes
2. LT51720632005245JSA00 21.08.2
17 no
No clouds
No NA No NA No NA
Only used to assist visual
interpretation
3.
DIM_PHR1B_MS_201508280834575_SEN_2226864101-002 DIM_PHR1B_MS_201606110819501_SEN_2226863101-001
30.05.2017
no No clouds
Yes GAFMap and ArcMap/visual mapping
Yes
GAFMap and ArcMap/Visual interpretation
Yes
ArcGIS 10 / Mosaic Dataset with manual neatline selection
Yes
Yes
4. 05APR26083225-M2AS-056358854030_01_P001 21.08.2
17 no
No clouds
Yes GAFMap and ArcMap/visual mapping
Yes
GAFMap and ArcMap/Visual interpretation
Yes
ArcGIS 10 / Mosaic Dataset with manual neatline selection
Yes
Yes
Earth Observation for Sustainable
Development – Urban
Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 19
4.2 Intermediate Quality Control of LCLU Data
No.
x = ok (checked/performed) o = not performed
n/a = not applicable
Co
mp
lian
t
Each item was checked during production and if necessary the source of error was identified and
corrected.
Description of QC item Comment
1 Data in predefined coordinate system?
(e.g. UTM WGS84 system) x
2 Do all features have attributes? x
3
All feature attributes are valid (attribute
value range, flags, correct comments)?
1) Do all features have a valid class
name?
x
4 All feature attributes present and filled
in? x
5 Data coherence at the border of the
urban area checked? x
6 Positional Accuracy of the features
according to User Specifications? x
7 All feature geometries valid? x
8
Vertices of feature geometries adjusted?
(e.g. missing, duplicated/ very close
vertices)
x
9 Multipart features resolved (Ring loops)? x
10 No data gaps? (Clean Gaps only if there
are over ~200 gaps) x
11 No data overlaps? (Clean Overlaps only if
there are over ~200 gaps)
12
Feature geometries optimized (e.g.
spikes, cutbacks minimized) (Find acute
angles, remove angles < 15°)
x
13 Minimum Mapping Area checked? (Urban
0.25ha, peri-urban 0.5ha) x
14 Minimum Mapping Width checked? x
15 Data checked for unnecessary polygon
boundaries? (after final dissolve) x
16 Service Area completely covered by
requested data (no data gaps)? x
17 No data outside delivery AoI? x
18
Add two new columns with the code and
the name of the class.(Name them:
code_year and name_year of the actual
classification”
19
Feature attribution and feature relations
plausible? (Plausibility Check ) --> Change
product
x
20 Calculation of area [AREA_km2] adjusted
to final data? x
21 Re-calculation of [ID] (FID + 1)? x
22
LCLU product file naming and versioning
according to specifications?
(City_program_product_referenceDate_C
oordinateSystem)
x
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 20
5.1 Thematic Accuracy
No. Product Name
Processing Date
Area Coverage
No
. o
f C
lass
es
Re
fere
nce
Da
ta
Sa
mp
le S
ize
Sa
mp
ling U
nit
(21
)
Sa
mp
ling
De
sign
(22
)
Sa
mp
le E
xclu
sio
n
Cri
teri
a (2
3)
Th
em
ati
c A
ccu
racy
(2
4)
Acc
ep
tan
ce S
tatu
s
Comment
1.
Land Cover Map 2016
01.09.2017 Full AOI (urban and peri-urban)
Mapped. 25
classes
VHR data sources for urban area. Bing Maps by Microsoft and Google Imagery for Peri-urban area.
215 per strata (Level 1 classes)
Point Stratified random sampling None, all used 97.25% Y
One class did not occure (‘1216 Military’)
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 21
5.2 Error Matrices
Class Name (columns = Ground Truth; Rows =
Mapped Class)
Re
sid
en
tia
l
Co
mm
erc
ial
Ro
ad
s
Ra
ilwa
y
Air
po
rt
Po
rt
Min
ing
Co
nst
ruct
ion
La
nd
wit
ho
ut
curr
en
t u
se
Urb
an
Pa
rks
Re
cre
ati
on
al
Fa
cilit
ies
Ce
me
tery
Agri
cult
ure
Fo
rest
Na
tura
l A
rea
s
Ba
re S
oil
We
tla
nd
s
Wa
ter
User Accuracy and Confidence Interval at 95%
Confidence Level
Class ID 11 121 122 123 124 125 131 132 133 141 142 143 2 31 32 33 4 5
Totals
Residential 11 173 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 174 99.43% ±1.4
Commercial 121 1 13 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 14 92.86 % ±17.6
Roads 122 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 100.00% ±50.0
Railway 123 0 0 0 50 0 0 0 0 0 0 0 0 0 0 0 0 0 0 50 100.00% ±1.0
Airport 124 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 3 100.00% ±16.7
Port 125 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 100.00% ±50.0
Mining 131 0 0 0 0 0 0 3 1 0 0 0 0 0 0 0 0 0 0 4 75.00% ± 54.94
Construction 132 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 2 50.00% ±94.3
Land without current use 133 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 100.0 % ± 50.0
Urban Parks 141 0 0 0 0 0 0 0 0 0 15 0 0 0 0 0 0 0 0 15 100.00% ± 3.3
Recreational Facilities 142 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 3 100.00% ±16.67
Cemetery 143 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 100.00% ±50.0
Agriculture 2 3 0 0 0 0 0 0 0 1 2 0 0 202 3 4 1 2 2 220 100.00% ± n.a.
Forest 31 0 0 0 0 0 0 0 0 0 0 0 0 0 135 2 0 0 0 137 91.82% ±3.85
Natural Areas 32 0 1 0 0 0 0 0 0 2 0 0 0 0 2 66 0 1 0 72 98.54% ±2.37
Bare Soil 33 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 0 0 5 91.67% ±7.1
Wetlands 4 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 209 0 210
100.00% ±10.0 99.52% ±1.17
Water 5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 216 216 100.0% ±0.23
Totals 178 14 1 50 3 1 3 2 4 18 3 1 202 141 72 6 212 218 1129
Producer Accuracy and Confidence Interval at 95%
Confidence Level 97
.19
%
±2
.7
92
.86
%
±1
7.6
10
0.0
0%
±
50
.0
10
0.0
0%
±
50
.0
10
0.0
0%
±
1.0
10
0.0
0%
±
16
.7
10
0.0
0%
±
50
.0
50
.00
%
±9
4.3
25
.00
%
±5
4.9
83
.33
%
±2
0.0
10
0.0
0%
±
16
.7
10
0.0
0%
±
50
.0
10
0.0
0%
±
0.3
95
.74
%
±3
.7
91
.67
%
±7
.1
83
.33
%
±3
8.2
98
.58
%
±1
.8
99
.08
%
±1
.5 Overall Accuracy: 97.25%
Confidence Interval 96.26% - 98.25%
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 22
6.1 Completeness
INPUT DATA
No.
Item AOI Coverage [km²]
Area Coverage [km²]
Completeness of Coverage (25)
No. of Scenes Scenes used in Production
Completeness of Verification (27)
Metadata Comments
1. HR EO Data
Kigoma
209 km² 100% 3 1 100% Yes none
2. VHR EO Data 84 km² 100% 6 6 100% Yes none
3. In-situ Data N/A N/A N/A N/A N/A N/A none
4. Ancillary Data N/A N/A N/A N/A N/A N/A none
PRODUCTS
No.
Product (28) AOI Coverage [km²]
Product Coverage [km²]
Completeness of Coverage (25)
Completeness of Classification (26)
Unclassifiable Area [%]
Completeness of Verification (27)
Metadata Comments
1. Urban and Peri Urban Areas 2006 and 2016
293 km² 293 km² 100% 100% 0% 100% Yes none
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 23
6.2 Compliancy
Product 1 (28) EO4SD_Kigoma _WB_LULC_2006_2016
Abstract
Land Use/Land Cover (LU/LC) information product contains spatial explicit information on different land use and land cover occurring in both the Core and Peri-Urban areas of the City of Kigoma. The Core area has detailed LU/LC nomenclature that is either at Level 3 or 4 whereas the Peri-Urban area LU/LC nomenclature is at an aggregated Level 1 or 2. The input data for the Core area was the Very High Resolution (VHR) data of Pleiades (2016), Quickbird (2005) and the input data for the Peri-Urban area was Sentinel-2 (2016), Landsat 5 (2005). The LU/LC product is the Baseline Product from which various derived products (such as Green Areas and Informal Settlements) are produced.
Service / Product Specifications
Area Coverage
Requirements Achieved Specifications Compliancy Comments
Country 1: Tanzania Country 1: Tanzania Yes ---
Country 2: N/A Country 2: N/A --- ---
A) Wall-to-wall: A) Wall-to-wall:
Yes --- Kigoma city Core (84 km²) and Peri-Urban area (209 km²)
Full coverage of AOIs, Core and Peri-Urban area
Area of Interest. 293 km², defined by the national user.
B) Sampling based: B) Sampling based: --- ---
N/A N/A
Time Period - Update Frequency
Requirements Achieved Specifications Compliancy Comments
Baseline Year(s): Baseline Year(s): Yes
See EO data acquisition dates. 2005 and 2015 (+/- 1 years) 2006 and 2016
B) Update Frequency B) Update Frequency Yes ---
No update No update
Geographic Reference System
Requirements Achieved Specifications Compliancy Comments
EPSG:32735; WGS84 / UTM 35S EPSG:32735; WGS84 / UTM 35S Yes ---
Mapping Classes and Definitions (Definitions see REDD+ MRV Design Document)
Requirements Achieved Specifications Compliancy
26 Main classes (see 0.1 Requirements sheet) All classes were mapped according to their definition. yes
Cloud and Shadow Detection and Removal
Requirements Achieved Specifications Compliancy Comments
Cloud and shadow covered areas shall be removed.
Full AoI coverage with spatial explicit LULC
cover information Yes ---
Spatial Resolution
Requirements Achieved Specifications Compliancy Comments
Na Na Na
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 24
Only applicable for raster data
Minimum Mapping Unit (MMU)
Requirements Achieved Specifications Compliancy Comments
0.25 ha and 0.5 ha MMU of 0.25 ha and 0.5 ha Yes ---
Data Type
Requirements Achieved Specifications Compliancy Comments
Vector data yes Yes ---
GeoPDF None were set Na ---
Bit Depth
Requirements Achieved Specifications Compliancy Comments
None Na Na ---
Data Format
Requirements Achieved Specifications Compliancy Comments
*.shp Shapefile Shapefile Yes ---
(open cross-platform format)
Class Coding (Raster Data Only)
Requirements Achieved Specifications Compliancy
Na Na ---
Metadata
Requirements Achieved Specifications Compliancy Comments
Metadata ISO compliant INSPIRE compliant and attached to
product Yes ---
Service / Product Quality
Thematic Accuracy
Requirements Achieved Specifications Compliancy Comments
Overall Accuracy: >80% 97% Yes ---
Positional Accuracy
Requirements Achieved Specifications Compliancy Comments
RMSE < 15 m 15 m Y Positional Accuracy of
VHR data is < 5m.
Delivery Procedure
Service Provision
Requirements Achieved Specifications Compliancy Comments
online via FTP Uploaded to FTP Yes ---
Delivery Date
Requirements Achieved Specifications Compliancy Comments
Middle of September Third Delivery April 2018 Yes ---
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 25
Glossary
Quality Checks
Prefix Suffix Explanation
Y (Yes) - V Visually checked
N (No) - Q Quantitatively/qualitatively checked
N/A - V Q Visually and quantitatively/qualitatively checked
- N/A Not applicable
Readability (1) Check readability of all required input data. Can the data be stored again?
Header / Metadata (2) Check Image Header Information and/or Metadata for completeness / distinctive features.
Data Format (3) For digital data, please give file format (e.g. *.tiff, *.shp).
Data Type (4) Please specify the type of the data (e.g. raster, vector or analogue).
Bit Depth (5) Please give pixel depth and sign of raster data (e.g. 8 bit unsigned integer).
Dynamic Range (6) Check dynamic range of all image bands. Visual check of dynamic range should be accompanied by histograms and statistics.
Dropped Lines & Artefacts (7) Check Image for dropped lines and other artefacts. If such occurs, please give description of extent and influence in the Comments section.
Sensor 1, 2, … (8) Please delete / add additional sensor sections as necessary.
Purpose (9)
The purpose and use of the given auxiliary or reference data should be stated:
ORHTOrectification, GEOmetric correction, REFerence, VERification source, POSitional ACCuracy assessment, THEmatic ACCuracy assessment.
Sampling Methodology (10)
Methodology of in situ or reference/auxiliary data sampling scheme should be outlined. In case of sample plots, also state how the plot positions have been determined (e.g. from GPS measurements, topographic maps, terrestrial triangulation, EO data, etc.) and how the sampling grid was established.
Positional Accuracy (11)
Positional accuracy of collected in situ data should be given. For reference/auxiliary data it MUST be given. If unknown, the data’s use must be explained. For DEM or other data with 3D information please specify both vertical and horizontal Positional Accuracy. For analogue data (e.g. maps) try to give approximate accuracy related to mapping scale.
Completeness (12) Data should be checked for spatial/temporal/content gaps.
Model / Algorithm (13)
Give the name of the software and its version. Specify the software module/algorithm used for: a) geometric correction, e.g., Polynomial and its degree, Rational Functions, Thin Plate Spline, etc. b) classification, e.g., ISODATA, Maximum Likelihood, Neural Networks, etc.
Dataset 1, 2, … (14) Please delete / add additional dataset sections as necessary.
Radiometric Processing (15) State whether (and which) radiometric processing was applied (e.g., contrast enhancement, histogram matching, bundle block adjustment, radiometric normalisation, filtering, etc.).
Projection; Spheroid / Ellipsoid (16)
Always specify completely, i.e. at minimum the Projection (+Zone, if applicable), Spheroid / Ellipsoid, Map Datum. Give additional information if necessary to unambiguously define the reference system.
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 26
Ground Control Points (GCPs) (17)
Give the number, distribution and RMS of used Ground Control Points (as average per scene) as obtained from the Geometric Correction, in meters [m]. Optionally, also RMSx and RMSy may be given. Distribution of GCPs should be attached as a snapshot, or described in the Comments section.
Tie Points (TPs) (18)
In case of mosaicking, give the number of used Tie Points and their total RMS (as average per scene) as obtained from the Geometric Correction, in meters [m]. Optionally, also RMSx and RMSy may be given. Distribution of TPs should be attached as a snapshot, or described in the Comments section.
Check Points (CPs) (19)
The real measure of positional accuracy and the only measure which should be examined as to its Acceptable Range. Give the independent Check Points’ total RMS (as average per scene) as obtained from the Geometric Correction, in meters [m]. Optionally, also RMSx and RMSy may be given. Distribution of CP's should be attached as a snapshot, or described in the Comments section.
Resampling Method (20) Specify, if / which resampling algorithm has been used (e.g. NN=Nearest Neighbour, BIL=Bilinear Interpolation, CC=Cubic Convolution).
Sampling unit (21) Specify sampling unit of Accuracy Assessment as POINT, FRAME, POLYGON and how it is treated (e.g. pixel center, polygon centre, etc.).
Sampling Design (22) SYST=Systematic, RAND=Random, STRAT=Stratified, SBCLASS=Stratified by class, SBAREA=Stratified by area
Sample exclusion criteria (23)
Describe which criteria you apply for sampling point selection resp. exclusion of certain points. For example, if the point is too close to a class boundary (less than 1 pixel), it is excluded. If the selected sample point is not representative of the class , it is excluded. For these reasons, it is recommended to oversample by 10% to compensate for sample point exclusion.
Thematic Accuracy (24)
Provide a detailed description of the Accuracy Assessment results in the form of error matrices showing commission and omission errors, user’s , producer’s and overall accuracies and other measures of Thematic Accuracy, as the confidence level (usually fixed at 95%) and the respective confidence interval, at least for the overall accuracy. If classification is done in a phased approach, e.g. if Forest Area and subsequently Forest Type are mapped, independent reports have to be produced.
Completeness of Coverage (25) State whether the coverage is limited to a subset, or portion of the final product.
Completeness of Classification (26)
State whether classification was constrained to a subset, or portion of the final product.
Completeness of Verification (27)
State whether verification applied to lineage, positional, or thematic accuracy is constrained to a subset, or portion of the final product.
Product (28) Please delete / add additional product sections as necessary.
Lineage (29)
Definition (INSPIRE 2015): Lineage is “a statement on process history and/or overall quality of the spatial data set. Where appropriate it may include a statement whether the data set has been validated or quality assured, whether it is the official version (if multiple versions exist), and whether it has legal validity. The value domain of this element is free text.”
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 27
Source(30) Definition (INSPIRE, 2015): “This is the description of the organisation responsible for the establishment, management, maintenance or distribution of the resource. This description shall include: name of the organisation and contact email address.”
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
Annex 2 to EO4SD-Urban Kigoma City Operations Report Page 3
This page is intentionally left blank!
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 1
Earth Observation for Sustainable Development - Urban
Quality Assurance and Quality Control Sheets These QA/QC Templates were prepared by GAF AG compliant with ISO 9001:2008 Quality Management System
standards and can only be used by Partners in the current EO4SD-Urban Project.
Project Title: EO4SD-Urban
Project Leader: GAF AG
Service Provider: GAF Editor: A. Broszeit
Client: WB Date: 24.04.2018
Product: Urban Green Areas
Overview of QC-Sheets and Processing Steps Sheet used Sheet filled
in
Requirements
0.1 Requirements Yes Yes
Specifications of Input Data
1.1 List of EO Data Yes Yes
1.2 List of In-situ Data Yes Yes
1.3 List of Ancillary Data Yes Yes
Data Quality Checks
2.1 EO Data Quality Yes Yes
2.2 In-situ Data Quality Yes Yes
2.3 Ancillary Data Quality Yes Yes
Pre-Processing of EO Data
3.1 Geometric Correction Yes Yes
3.1.1 Data Fusion Yes Yes
3.2 Data Processing Yes Yes
Thematic Processing
4.1 Classification Yes Yes
Accuracy Assessment
5.1 Thematic Accuracy Yes Yes
5.2 Error Matrices Yes Yes
Delivery Checks / Delivery
6.1 Completeness Yes Yes
6.2 Compliancy Yes Yes
Glossary (index numbers in the QA/QC tables refer to the glossary at the end of this document) Further QC-relevant Documents:
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 2
Comments / Characteristics:
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 3
0.1 Requirements
Product 1 (28) Urban Green Areas 2006 and 2016
Abstract
Urban Green Areas information product contains spatial explicit information on green areas within the urban extent. Green areas include linear green features such as river alignments, hedges and trees, as well as public parks, private gardens, forested areas, etc. Urban Green Areas & Change dataset is based on Very High Resolution (VHR) satellite imagery by means of automated classification processing techniques.
Service / Product Specifications
Area Coverage
Country: Tanzania A) Wall-to-wall: n/a
City: Kigoma Selected Sites: Area of Interest defined with end-users
Area km² Core Urban: 84 B) Sampling based: n/a
Time Period - Update Frequency
A) Baseline Year(s): B) Update Frequency
2006 and 2016 2 points in time
Comments: None
Geographic Reference System
WGS84 UTM Zone 35S
Mapping Classes and Definitions
Single date
0 Non-urban green area
1 Urban green area
Change product
0 Non-urban green area
1 Permanent urban green area
2 Loss of urban green area
3 New urban green area
Cloud and Cloud Shadow Detection and Removal
n/a
Spatial Resolution
1 m
Minimum Mapping Unit (MMU)
1 m²
Data Type & Format
Raster data in GEOTIFF
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 4
Bit Depth
8bit for *.tif raster files
Class Coding
Class Code Class Name RGB Code
Single date
0 Non-urban green area 240/240/240
1 Urban green area 56/168/0
Change product
0 Non-urban green area 255/245/235
1 Permanent urban green area 56/168/0
2 Loss of urban green area 255/100/100
3 New urban green area 152/230/0
Metadata
ISO standard 19115 "Metadata" and ISO 19139 "XML Scheme Implementation"
Service / Product Quality
Thematic Accuracy
> 80%
Positional Accuracy
3 m (CE90)
Delivery Procedure
Service Provision
Online via FTP
Delivery Date
End of April 2018
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 5
1.1 List of EO Data
Sensoren (8) PLX, QB-2
Incoming Date
Acquisition Date
Proc. Level
Path / Row
AOI - City / Region / Country
Spatial Res. No. of Bands
Cloud Cover (Data Provider)
Projection / Spheroid (16)
Data Format (3)
Bit Depth (5)
Header / Metadata (2)
File Name [e.g yymmdd; tbd...]
Pleiades 1B
1. IMG_PHR1B_MS_201606110819501_SEN_2226863101-002_R1C1
17.03.2017
11.06.2016
primary R1C1 Kigoma, TZA
MUL: 2m; PAN: 0.5m
4 0% WGS84 .tiff 16 Bit
.xml
2. IMG_PHR1B_P_201606110819501_SEN_2226863101-001_R1C1
17.03.2017
11.06.2016
primary R1C1 Kigoma, TZA
MUL: 2m; PAN: 0.5m
4 0% WGS84 .tiff 16 Bit
.xml
Quickbird-2
3. 05APR26083225-M2AS_R1C1-056358854030_01_P001
31.03.2017
26.04.2005
LV2A R1C1 Kigoma, TZA
MUL: 2m; PAN: 0.6m
4 0% UTM 36S GeoTIFF
16 Bit
.img,
.rpb,
.til, xml
4. 05APR26083225-M2AS_R1C2-056358854030_01_P001
31.03.2017
26.04.2005
LV2A R1C2 Kigoma, TZA
MUL: 2m; PAN: 0.6m
4 0% UTM 36S GeoTIFF
16 Bit
.img,
.rpb,
.til, xml
5. 05APR26083225-M2AS_R2C1-056358854030_01_P001
31.03.2017
26.04.2005
LV2A R2C1 Kigoma, TZA
MUL: 2m; PAN: 0.6m
4 0% UTM 36S GeoTIFF
16 Bit
.img,
.rpb,
.til, xml
6. 05APR26083225-M2AS_R2C2-056358854030_01_P001
31.03.2017
26.04.2005
LV2A R2C2 Kigoma, TZA
MUL: 2m; PAN: 0.6m
4 0% UTM 36S GeoTIFF
16 Bit
.img,
.rpb,
.til, xml
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 6
1.2 List of In-situ Data
Dataset 1 (14) Incoming Date Acquisition Date AOI - City / Region / Country
Data Type (4)
Projection / Spheroid (16)
No. of Sample Plots
Sampling Design (22)
Positional Accuracy (11)
Purpose (9)
No In-situ Data Used
Lineage:
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 7
1.3 List of Ancillary Data
Dataset 1 (14)
LULC Date of Receipt from Client
Metadata (2)
Reference Mapping Date / Date of Creation
Geographic Area/- City / Region / Country
Area Coverag
e
Data Type (4)
Data Format (3)
Projection / Spheroid (16)
Positional Accuracy (11)
No. of Classes
Thematic
Accuracy
Availability of Class Definitions
EO4SD-Urban Land Use/Land Cover Baseline Product
2017-08-30 Yes
(INSPIRE) 2006 and 2016
Kigoma (urban area) 100% Vector *.shp EPSG:32736, WGS 84 / UTM zone 36S
< 3 m 26 97% Yes
Lineage(29): /
Source (30):: GAF AG
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 8
2.1 EO Data Quality
Sensoren(8) QB-2, PLX
Ba
cku
p
Re
ad
ab
ility
(1
)
Ch
eck
He
ad
er
/
Me
tad
ata
(2)
Band Specifications
Pro
ject
ion
/ S
ph
ero
id (1
6)
Sce
ne
Lo
cati
on
Co
mp
lete
ne
ss o
f
Ad
dit
ion
al D
ata
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Clo
ud
Co
ver
(in
tern
al
che
ck)
Ra
dio
me
try
To
po
gra
ph
y
Dro
pp
ed
Lin
es
/ A
rte
fact
s (7
)
Acc
ep
tan
ce S
tatu
s
Comment
s File Name [e.g yymmdd; tbd...]
No
. o
f B
an
ds
Ba
nd
Re
gis
tra
tio
n
Sp
ect
ral/
Sp
ati
al
Re
solu
tio
n
Pleiades 1B
1. IMG_PHR1B_MS_201508280834575_SEN_2226864101-002
Y - Q Y - V Q Y - V Q Y - V Q Y - V Q Y - Q Y - V Y - V Y - V Q Y - Q Y - Q
Y - V Q
Y - V Q
Y - V
N - V Q
Yes none
2. IMG_PHR1B_MS_201606110819501_SEN_2226863101-002
Y - Q Y - V Q Y - V Q Y - V Q Y - V Q Y - Q Y - V Y - V Y - V Q Y - Q Y - Q
Y - V Q
Y - V Q
Y - V
N - V Q
Yes none
Quickbird-2
3. 05APR26083225-M2AS-056358854030_01_P001
Y - Q Y - V Q Y - V Q Y - V Q Y - V Q Y - Q Y - V Y - V Y - V Q Y - Q Y - Q
Y - V Q
Y - V Q
Y - V
N - V Q
Yes none
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 9
2.2 In-situ Data Quality
Dataset 1 (14) No Data used
Ba
cku
p
Re
ad
ab
ility
(1)
Ch
eck
He
ad
er
/ M
eta
da
ta (2
)
Ext
en
t
Pro
ject
ion
/ S
ph
ero
id
(16
)
Sp
ati
al R
eso
luti
on
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Lo
cati
on
Co
mp
lete
ne
ss
Ge
om
. M
isa
lign
me
nt
Pla
usi
bili
ty
Dro
pp
ed
Lin
es
/
Art
efa
cts
(7)
Acc
ep
tan
ce S
tatu
s
Comments File Name [e.g yymmdd; tbd...]
No In-situ Data Used
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 10
2.3 Ancillary Data Quality
Dataset 1 (14)
LULC
Re
ad
ab
ility
(1
)
Ch
eck
He
ad
er
/Me
tad
ata
(2)
Da
ta C
ove
rage
M
atc
he
s S
erv
ice
Are
a
(%)
Pro
ject
ion
/ S
ph
ero
id
, E
PS
G(1
6)
Sp
ati
al R
eso
luti
on
an
d u
nit
(e
.g.
m, k
m)
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Co
mp
lete
ne
ss
(Ve
cto
r: A
ttri
bu
te
Ta
ble
; R
ast
er:
Th
em
ati
c V
alu
es)
Ge
om
. M
isa
lign
me
nt
Dro
pp
ed
Lin
es
/ A
rte
fact
s (
EO
da
ta
on
ly)(
7)
Uti
lity
for
Cu
rre
nt
Pro
ject
File Name [e.g yymmdd; tbd...]
EO4SD-Urban LULC Product ☒ Yes
☐ No
☒ Complete (INSPIRE/ISO19119)
☐ Incomplete
☐ not available
☒ Yes
☐ No
☐ Unknown If no: xxx%
☒ Correct
☐ Incorrect
☐ Unknown EPSG: 32749
n.a. *.shp n.a. ☒ Complete
☐Incomplete
☒ No
☐ Yes
☐ Unknown If yes: XX m
☒ No
☐ Yes
☐ n.a. If yes: xxx %
☐ None
☐ Partial
☒ Full
Comments: /
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 11
3.1 Geometric Correction
Sensoren (8) QB-2, PLX
Pro
cess
ing D
ate
AO
I
Cit
y /
Re
gio
n /
Co
un
try
Pro
ject
ion
/ S
ph
ero
id (1
6)
No
. &
RM
S (
m)
of
GC
Ps
(17
)
No
. &
RM
S (
m)
of
TP
s (1
8)
No
. &
RM
S (
m)
of
CP
s (1
9)
Dig
ita
l E
leva
tio
n M
od
el
(DE
M)
Mo
de
l /
Alg
ori
thm
(13
)
Re
sam
plin
g
Me
tho
d (2
0)
Va
lida
tio
n R
ep
ort
s
Acc
ep
tan
ce S
tatu
s
Output File Name [e.g yymmdd; tbd...] N
o. File Name [e.g yymmdd; tbd...]
Pleiades 1B
1. DIM_PHR1B_MS_201508280834575_SEN_2226864101-002_toa_PSH
12.04.2017
Kigoma TZA
UTM35S / WGS84
MGCPs 12; RMSE x0,79; y0,69
4; RMSE x0,73; y 0,88
N/A SRTM
30
0 order poynom
CC Yes Yes oDIM_PHR1B_MS_201508280834575_SEN_2226864101-002_toa_PSH
2. DIM_PHR1B_MS_201606110819501_SEN_2226863101-002_toa_PSH
12.04.2017
Kigoma TZA
UTM35S / WGS84
MGCPs 12; RMSE x0,79; y0,69
4; RMSE x0,73; y 0,88
N/A SRTM 30
0 order poynom
CC Yes Yes oDIM_PHR1B_MS_201606110819501_SEN_2226863101-002_toa_PSH
Quickbird-2
3. 05APR26083225-M2AS-056358854030_01_P001_toa_PSH
12.04.2017
Kigoma TZA
UTM35S / WGS84
MGCPs 7; RMSE x1,07; y0,89
N/A N/A SRTM 30
Affin, 1st order poynom
CC Yes Yes o05APR26083225-M2AS-056358854030_01_P001_toa_PSH
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 12
3.1.1 Data Fusion
Dataset 1 (14) Input A Input B Output
Acc
ep
tan
ce S
tatu
s
Comments
Filename DIM_PHR1B_MS_201606110819501_SE
N_2226863101-002 DIM_PHR1B_P_201606110819501_SEN
_2226863101-001 DIM_PHR1B_MS_201606110819501_SEN
_2226863101-002_toa_PSH
none
Sensor Pleiades 1B Pleiades 1B Pleiades 1B
Method PCI / Pansharp2 PCI / Pansharp2 PCI / Pansharp2
Spatial Resolution / (MS/Pan)
2m MS 0,5m Pan 0,5m MS
Band Combination RGB NIR PAN RGB NIR
Data Format (3) & Bit Depth (5)
*.tif 16 Bit u *.tif 16 Bit u *.pix 16 Bit u Yes
Dataset 2 (14) Input A Input B Output
Acc
ep
tan
ce S
tatu
s
Comments
Filename DIM_PHR1B_MS_201508280834575_SE
N_2226864101-002 DIM_PHR1B_P_201508280834575_SEN
_2226864101-001 DIM_PHR1B_MS_201508280834575_SEN
_2226864101-002_toa_PSH
none
Sensor Pleiades 1B Pleiades 1B Pleiades 1B
Method PCI / Pansharp2 PCI / Pansharp2 PCI / Pansharp2
Spatial Resolution / (MS/Pan)
2m MS 0,5m Pan 0,5m MS
Band Combination RGB NIR PAN RGB NIR
Data Format (3) & Bit Depth (5)
*.tif 16 Bit u *.tif 16 Bit u *.pix 16 Bit u Yes
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 13
Dataset 3 (14) Input A Input B Output
Acc
ep
tan
ce S
tatu
s
Comments
Filename 05APR26083225-M2AS-
056358854030_01_P001 05APR26083225-P2AS-
056358854030_01_P001 05APR26083225-M2AS-
056358854030_01_P001_toa_PSH
none
Sensor Quickbird-2 Quickbird-2 Quickbird-2
Method PCI / Pansharp2 PCI / Pansharp2 PCI / Pansharp2
Spatial Resolution / (MS/Pan)
2m MS 0,6m Pan 0,6m MS
Band Combination RGB NIR PAN BGR NIR
Data Format (3) & Bit Depth (5)
*.tif 16 Bit u *.tif 16 Bit u *.pix 16 Bit u Yes
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 14
3.2 Data Processing
Sensoren (8)
QB-2, PLX Atmospheric Correction
Radiometric Processing (15)
Topographic Normalisation
OTHER Calculation:
Acc
ep
tan
ce S
tatu
s
Ba
cku
p
Comment
No.
File Name [e.g yymmdd; tbd...] Processing
Date
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
Pleiades 1B
1. DIM_PHR1B_MS_201508280834575_SEN_2226864101-002
11.04.2017 No N/A No N/A No N/A Yes GAFmap / TOA
Yes No none
2. DIM_PHR1B_MS_201606110819501_SEN_2226863101-001
11.04.2017 No N/A No N/A No N/A Yes GAFmap / TOA
Yes No none
Quickbird-2
3. 05APR26083225-M2AS-056358854030_01_P001
11.04.2017 No N/A No N/A No N/A Yes GAFmap / TOA
Yes No none
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 15
4.1 Classification
Sensors (8) Pleiades, QB
Processing Date
Cloud Masking Thematic
Classification Manual
Enhancement Mosaicking
Acc
ep
tan
ce S
tatu
s
Ba
cku
p
Comment No
. File Name [e.g yymmdd; tbd...]
pro
cess
ed
Software /
Method
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
1.
DIM_PHR1B_MS_201508280834575_SEN_2226864101-002 DIM_PHR1B_MS_201606110819501_SEN_2226863101-001
15.09.2017
no No clouds
Yes ERDAS IMAGINE
Yes
GAFMap and ArcMap/Visual interpretation
Yes
ArcGIS 10 / Mosaic Dataset with manual neatline selection
Yes
Yes
2. 05APR26083225-M2AS-056358854030_01_P001 15.09.2
17 no
No clouds
Yes ERDAS IMAGINE
Yes
GAFMap and ArcMap/Visual interpretation
Yes
ArcGIS 10 / Mosaic Dataset with manual neatline selection
Yes
Yes
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 16
5.1 Thematic Accuracy
No. Product Name
Processing Date
Area Coverage
No
. o
f C
lass
es
Re
fere
nce
Da
ta
Sa
mp
le S
ize
Sa
mp
ling U
nit
(21
)
Sa
mp
ling
De
sign
(22
)
Sa
mp
le E
xclu
sio
n
Cri
teri
a (2
3)
Th
em
ati
c A
ccu
racy
(2
4)
Acc
ep
tan
ce S
tatu
s
Comment
1.
Urban Green Areas 2006
26.09.2017 Full AOI (urban core area)
4 VHR EO Data
215 per strata
POINT Stratified random sampling None, all used 89.77% Yes None
2.
Urban Green Areas 2016
26.09.2017 Full AOI (urban core area)
4 VHR EO Data
215 per strata
POINT Stratified random sampling None, all used 90.70% Yes none
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 17
5.2 Error Matrices
Urban Green Area 2016 Reference Data
Totals 0 - Non-Urban Green Area 1 - Urban Green Area
0 - Non-Urban Green Area 116 9 125
1 - Urban Green Area 11 79 90
Totals 127 88 215
Accuracy Statistics z= 1.96 Overall Accuracy: 90.70%
95% Confidence Interval: 86.58% 94.81%
Class Name Producer’s Accuracy User’s Accuracy
0 - Non-Urban Green Area 91.34% 92.80%
1 - Urban Green Area 89.77% 87.78%
Urban Green Area 2006 Reference Data
Totals 0 - Non-Urban Green Area 1 - Urban Green Area
0 - Non-Urban Green Area 122 15 137
1 - Urban Green Area 7 71 78
Totals 129 86 215
Accuracy Statistics z= 1.96 Overall Accuracy: 89.77%
95% Confidence Interval: 85.48% 94.05%
Class Name Producer’s Accuracy User’s Accuracy
0 - Non-Urban Green Area 94.57% 89.05%
1 - Urban Green Area 82.56% 91.03%
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 18
6.1 Completeness
INPUT DATA
No.
Item AOI Coverage [km²]
Area Coverage [km²]
Completeness of Coverage (25)
No. of Scenes Scenes used in Production
Completeness of Verification (27)
Metadata Comments
1. VHR EO Data Kigoma
85 km² 100% 6 6 100% Yes none
2. Ancillary Data N/A N/A N/A N/A N/A Yes none
PRODUCTS
No.
Product (28) AOI Coverage [km²]
Product Coverage [km²]
Completeness of Coverage (25)
Completeness of Classification (26)
Unclassifiable Area [%]
Completeness of Verification (27)
Metadata Comments
1. Urban Green Areas 2006 and 2016 and Change
Kigoma 85 km² 100% 100% 0% 100% Yes none
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 19
6.2 Compliancy
Product 1 (28) Urban Green Areas
Abstract
Urban Green Areas information product contains spatial explicit information on green areas within the urban extent. Green areas include linear green features such as river alignments, hedges and trees, as well as public parks, private gardens, forested areas, etc. Urban Green Areas & Change dataset is based on Very High Resolution (VHR) satellite imagery by means of automated classification processing techniques.
Service / Product Specifications
Area Coverage
Requirements Achieved Specifications Compliancy Comments
Country 1: Tanzania Country 1: Tanzania YES None
A) Wall-to-wall: Kigoma city A) Wall-to-wall: Kigoma city
Core Urban Area (85 km²) Core Urban Area (85 km²)
Area of Interest: Kigoma 85 km², defined by the national user.
B) Sampling based: B) Sampling based:
N/A N/A
Time Period - Update Frequency
Requirements Achieved Specifications Compliancy Comments
Baseline Year(s): Baseline Year(s): Yes
See EO data acquisition dates. 2005 and 2015 (+/- 1 years) 2006 and 2016
B) Update Frequency B) Update Frequency Yes --
No update No update
Geographic Reference System
Requirements Achieved Specifications Compliancy Comments
WGS84 / UTM 35S WGS84 / UTM 35S YES None
Mapping Classes and Definitions
Requirements Achieved Specifications Compliancy
Single date
0 Non-urban green area 0 Non-urban green area YES
1 Urban green area 1 Urban green area YES
Change product
0 Non-urban green area 0 Non-urban green area YES
1 Permanent urban green area 1 Permanent urban green area YES
2 Loss of urban green area 2 Loss of urban green area YES
3 New urban green area 3 New urban green area YES
255 No data available due to absence of historic VHR imagery
255 No data available due to absence of historic VHR imagery
YES
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 20
Cloud and Shadow Detection and Removal
Requirements Achieved Specifications Compliancy Comments
N/A N/A
Spatial Resolution
Requirements Achieved Specifications Compliancy Comments
1 m 1 m YES None
Minimum Mapping Unit (MMU)
Requirements Achieved Specifications Compliancy Comments
1 m² 1 m² YES None
Data Type
Requirements Achieved Specifications Compliancy Comments
Raster data Raster data YES None
Bit Depth
Requirements Achieved Specifications Compliancy Comments
N/A N/A
Data Format
Requirements Achieved Specifications Compliancy Comments
GEOTIFF GEOTIFF YES None
Class Coding
Requirements Achieved Specifications Compliancy
N/A N/A
Metadata
Requirements Achieved Specifications Compliancy Comments
ISO standard 19115 "Metadata" and ISO 19139 "XML Scheme Implementation"
ISO standard 19115 "Metadata" and ISO 19139 "XML Scheme Implementation"
YES None
Service / Product Quality
Thematic Accuracy
Requirements Achieved Specifications Compliancy Comments
> 80% > 80% YES None
Positional Accuracy
Requirements Achieved Specifications Compliancy Comments
3 m (CE90) 3 m (CE90) YES None
Delivery Procedure
Service Provision
Requirements Achieved Specifications Compliancy Comments
Online via FTP Online via FTP YES None
Delivery Date
Requirements Achieved Specifications Compliancy Comments
End of October 2017 Third Delivery April 2018 YES None
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 21
Glossary
Quality Checks
Prefix Suffix Explanation
Y (Yes) - V Visually checked
N (No) - Q Quantitatively/qualitatively checked
N/A - V Q Visually and quantitatively/qualitatively checked
- N/A Not applicable
Readability (1) Check readability of all required input data. Can the data be stored again?
Header / Metadata (2) Check Image Header Information and/or Metadata for completeness / distinctive features.
Data Format (3) For digital data, please give file format (e.g. *.tiff, *.shp).
Data Type (4) Please specify the type of the data (e.g. raster, vector or analogue).
Bit Depth (5) Please give pixel depth and sign of raster data (e.g. 8 bit unsigned integer).
Dynamic Range (6) Check dynamic range of all image bands. Visual check of dynamic range should be accompanied by histograms and statistics.
Dropped Lines & Artefacts (7) Check Image for dropped lines and other artefacts. If such occurs, please give description of extent and influence in the Comments section.
Sensor 1, 2, … (8) Please delete / add additional sensor sections as necessary.
Purpose (9)
The purpose and use of the given auxiliary or reference data should be stated:
ORHTOrectification, GEOmetric correction, REFerence, VERification source, POSitional ACCuracy assessment, THEmatic ACCuracy assessment.
Sampling Methodology (10)
Methodology of in situ or reference/auxiliary data sampling scheme should be outlined. In case of sample plots, also state how the plot positions have been determined (e.g. from GPS measurements, topographic maps, terrestrial triangulation, EO data, etc.) and how the sampling grid was established.
Positional Accuracy (11)
Positional accuracy of collected in situ data should be given. For reference/auxiliary data it MUST be given. If unknown, the data’s use must be explained. For DEM or other data with 3D information please specify both vertical and horizontal Positional Accuracy. For analogue data (e.g. maps) try to give approximate accuracy related to mapping scale.
Completeness (12) Data should be checked for spatial/temporal/content gaps.
Model / Algorithm (13)
Give the name of the software and its version. Specify the software module/algorithm used for: a) geometric correction, e.g., Polynomial and its degree, Rational Functions, Thin Plate Spline, etc. b) classification, e.g., ISODATA, Maximum Likelihood, Neural Networks, etc.
Dataset 1, 2, … (14) Please delete / add additional dataset sections as necessary.
Radiometric Processing (15) State whether (and which) radiometric processing was applied (e.g., contrast enhancement, histogram matching, bundle block adjustment, radiometric normalisation, filtering, etc.).
Projection; Spheroid / Ellipsoid (16)
Always specify completely, i.e. at minimum the Projection (+Zone, if applicable), Spheroid / Ellipsoid, Map Datum. Give additional information if necessary to unambiguously define the reference system.
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2018 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 22
Ground Control Points (GCPs) (17)
Give the number, distribution and RMS of used Ground Control Points (as average per scene) as obtained from the Geometric Correction, in meters [m]. Optionally, also RMSx and RMSy may be given. Distribution of GCPs should be attached as a snapshot, or described in the Comments section.
Tie Points (TPs) (18)
In case of mosaicking, give the number of used Tie Points and their total RMS (as average per scene) as obtained from the Geometric Correction, in meters [m]. Optionally, also RMSx and RMSy may be given. Distribution of TPs should be attached as a snapshot, or described in the Comments section.
Check Points (CPs) (19)
The real measure of positional accuracy and the only measure which should be examined as to its Acceptable Range. Give the independent Check Points’ total RMS (as average per scene) as obtained from the Geometric Correction, in meters [m]. Optionally, also RMSx and RMSy may be given. Distribution of CP's should be attached as a snapshot, or described in the Comments section.
Resampling Method (20) Specify, if / which resampling algorithm has been used (e.g. NN=Nearest Neighbour, BIL=Bilinear Interpolation, CC=Cubic Convolution).
Sampling unit (21) Specify sampling unit of Accuracy Assessment as POINT, FRAME, POLYGON and how it is treated (e.g. pixel center, polygon centre, etc.).
Sampling Design (22) SYST=Systematic, RAND=Random, STRAT=Stratified, SBCLASS=Stratified by class, SBAREA=Stratified by area
Sample exclusion criteria (23)
Describe which criteria you apply for sampling point selection resp. exclusion of certain points. For example, if the point is too close to a class boundary (less than 1 pixel), it is excluded. If the selected sample point is not representative of the class , it is excluded. For these reasons, it is recommended to oversample by 10% to compensate for sample point exclusion.
Thematic Accuracy (24)
Provide a detailed description of the Accuracy Assessment results in the form of error matrices showing commission and omission errors, user’s , producer’s and overall accuracies and other measures of Thematic Accuracy, as the confidence level (usually fixed at 95%) and the respective confidence interval, at least for the overall accuracy. If classification is done in a phased approach, e.g. if Forest Area and subsequently Forest Type are mapped, independent reports have to be produced.
Completeness of Coverage (25) State whether the coverage is limited to a subset, or portion of the final product.
Completeness of Classification (26)
State whether classification was constrained to a subset, or portion of the final product.
Completeness of Verification (27)
State whether verification applied to lineage, positional, or thematic accuracy is constrained to a subset, or portion of the final product.
Product (28) Please delete / add additional product sections as necessary.
Lineage (29) Definition (INSPIRE 2015): Lineage is “a statement on process history and/or overall quality of the spatial data set. Where appropriate it may include a statement whether the data set has been validated or quality assured, whether it is the official version (if multiple versions exist), and whether it has legal validity. The value domain of this element is free text.”
Source(30) Definition (INSPIRE, 2015): “This is the description of the organisation responsible for the establishment, management, maintenance or distribution of the resource. This description shall include: name of the organisation and contact email address.”
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 1
Earth Observation for Sustainable Development - Urban
Quality Assurance and Quality Control Sheets These QA/QC Templates were prepared by GAF AG compliant with ISO 9001:2008 Quality Management System
standards and can only be used by Partners in the current EO4SD-Urban Project.
Project Title: EO4SD-Urban
Project Leader: GAF AG
Service Provider: GAF Editor: D. Angelova
Client: WB Date: 02.11.2017
Product: Planned/Unplanned Settlement Areas
Overview of QC-Sheets and Processing Steps Sheet used
Sheet filled in
Requirements
0.1 Requirements Yes Yes
Specifications of Input Data
1.1 List of EO Data Yes Yes
1.2 List of In-situ Data Yes Yes
1.3 List of Ancillary Data Yes Yes
Data Quality Checks
2.1 EO Data Quality Yes Yes
2.2 In-situ Data Quality Yes Yes
2.3 Ancillary Data Quality Yes Yes
Pre-Processing of EO Data
3.1 Geometric Correction Yes Yes
3.1.1 Data Fusion Yes Yes
3.2 Data Processing Yes Yes
Thematic Processing
4.1 Classification Yes Yes
Accuracy Assessment
5.1 Thematic Accuracy Yes Yes
5.2 Error Matrices Yes Yes
Delivery Checks / Delivery
6.1 Completeness Yes Yes
6.2 Compliancy Yes Yes
Glossary (index numbers in the QA/QC tables refer to the glossary at the end of this document) Further QC-relevant Documents:
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 2
Comments / Characteristics:
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 3
0.1 Requirements
Product 1 (28) Planned and Unplanned Settlement Areas 2006 and 2016
Abstract
The product Planned and Unplanned Settlements contains spatial explicit information on the two settlement types within the core urban area. The distinction into the two settlement types is restricted to the residential area (LULC class 11), and to commercial areas (LULC class 1211) which also have a residential component. The Planned and Unplanned Settlements, and Change dataset is developed, based on Very High Resolution (VHR) satellite imagery by means of visual interpretation.
Service / Product Specifications
Area Coverage
Country: Tanzania A) Wall-to-wall: n/a
City: Kigoma Selected Sites: Area of Interest defined with end-users
Area km² Core Urban: 85 B) Sampling based: n/a
Time Period - Update Frequency
A) Baseline Year(s): B) Update Frequency
2006 and 2016 2 points in time
Comments: None
Geographic Reference System
WGS84 UTM Zone 35S
Mapping Classes and Definitions
Single date
Planned settlements
Unplanned settlements
Change product
No change in planned settlement area
No change in unplanned settlement area
Expansion of planned settlement area
Expansion of unplanned settlement area
Decrease of planned settlement area
Decrease of unplanned settlement area
Unplanned to planned settlement area
Cloud and Cloud Shadow Detection and Removal
n/a
Spatial Resolution
n.a. (Product provided as Shapefile)
Minimum Mapping Unit (MMU)
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 4
0.25 ha
Data Type & Format
Shapefile *.shp
Bit Depth
n.a.
Class Coding
Class Name RGB Code
Single date
Planned settlements 132/0/168
Unplanned settlements 168/0/0
Change product
No change in planned settlement area 132/0/168
No change in unplanned settlement area 168/0/0
Expansion of planned settlement area 197/0/255
Expansion of unplanned settlement area 255/0/0
Decrease of planned settlement area 168/132/167
Decrease of unplanned settlement area 255/100/100
Unplanned to planned settlement area 255/0/197
Metadata
ISO standard 19115 "Metadata" and ISO 19139 "XML Scheme Implementation"
Service / Product Quality
Thematic Accuracy
> 80%
Positional Accuracy
RMSE < 15 m
Delivery Procedure
Service Provision
Online via FTP
Delivery Date
End of October 2017
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 5
1.1 List of EO Data
Sensoren (8) PLX, QB-2
Incoming Date
Acquisition Date
Proc. Level
Path / Row
AOI - City / Region / Country
Spatial Res. No. of Bands
Cloud Cover (Data Provider)
Projection / Spheroid (16)
Data Format (3)
Bit Depth (5)
Header / Metadata (2)
File Name [e.g yymmdd; tbd...]
Pleiades 1B
1. IMG_PHR1B_MS_201606110819501_SEN_2226863101-002_R1C1
17.03.2017
11.06.2016
primary R1C1 Kigoma, TZA
MUL: 2m; PAN: 0.5m
4 0% WGS84 .tiff 16 Bit
.xml
2. IMG_PHR1B_P_201606110819501_SEN_2226863101-001_R1C1
17.03.2017
11.06.2016
primary R1C1 Kigoma, TZA
MUL: 2m; PAN: 0.5m
4 0% WGS84 .tiff 16 Bit
.xml
Quickbird-2
3. 05APR26083225-M2AS_R1C1-056358854030_01_P001
31.03.2017
26.04.2005
LV2A R1C1 Kigoma, TZA
MUL: 2m; PAN: 0.6m
4 0% UTM 36S GeoTIFF
16 Bit
.img,
.rpb,
.til, xml
4. 05APR26083225-M2AS_R1C2-056358854030_01_P001
31.03.2017
26.04.2005
LV2A R1C2 Kigoma, TZA
MUL: 2m; PAN: 0.6m
4 0% UTM 36S GeoTIFF
16 Bit
.img,
.rpb,
.til, xml
5. 05APR26083225-M2AS_R2C1-056358854030_01_P001
31.03.2017
26.04.2005
LV2A R2C1 Kigoma, TZA
MUL: 2m; PAN: 0.6m
4 0% UTM 36S GeoTIFF
16 Bit
.img,
.rpb,
.til, xml
6. 05APR26083225-M2AS_R2C2-056358854030_01_P001
31.03.2017
26.04.2005
LV2A R2C2 Kigoma, TZA
MUL: 2m; PAN: 0.6m
4 0% UTM 36S GeoTIFF
16 Bit
.img,
.rpb,
.til, xml
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 6
1.2 List of In-situ Data
Dataset 1 (14) Incoming Date Acquisition Date AOI - City / Region / Country
Data Type (4)
Projection / Spheroid (16)
No. of Sample Plots
Sampling Design (22)
Positional Accuracy (11)
Purpose (9)
No In-situ Data Used
Lineage:
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 7
1.3 List of Ancillary Data
Dataset 1 (14)
LULC Date of Receipt from Client
Metadata (2)
Reference Mapping Date / Date of Creation
Geographic Area/- City / Region / Country
Area Coverag
e
Data Type (4)
Data Format (3)
Projection / Spheroid (16)
Positional Accuracy (11)
No. of Classes
Thematic
Accuracy
Availability of Class Definitions
EO4SD-Urban Land Use/Land Cover Baseline Product
2017-08-30 Yes
(INSPIRE) 2006 and 2016
Kigoma (urban area) 100% Vector *.shp EPSG:32736, WGS 84 / UTM zone 35S
< 3 m 26 97% Yes
Lineage(29): /
Source (30):: GAF AG
Dataset 2 (14)
Report Date of Receipt from Client
Metadata (2)
Reference Mapping Date / Date of Creation
Geographic Area/- City / Region / Country
Area Coverage
Data Type (4)
Data Format (3)
Projection / Spheroid (16)
Positional Accuracy (11)
No. of Classes
Thematic Accuracy
Availability of Class Definitions
KIGOMA-UJIJI MASTER PLAN 2017-2037
2017-05-18 n.a. December, 2016
Kigoma/ Tanzania Kigoma Document
pdf n.a. n.a. n.a. n.a. n.a.
Lineage(29): The Master Plan of Kigoma was used as a reference while mapping planned and unplanned settlements in 2016.
Source (30): Received from MaryGrace Weber.
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 8
2.1 EO Data Quality
Sensoren(8) QB-2, PLX
Ba
cku
p
Re
ad
ab
ility
(1
)
Ch
eck
He
ad
er
/
Me
tad
ata
(2)
Band Specifications
Pro
ject
ion
/ S
ph
ero
id (1
6)
Sce
ne
Lo
cati
on
Co
mp
lete
ne
ss o
f
Ad
dit
ion
al D
ata
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Clo
ud
Co
ver
(in
tern
al
che
ck)
Ra
dio
me
try
To
po
gra
ph
y
Dro
pp
ed
Lin
es
/ A
rte
fact
s (7
)
Acc
ep
tan
ce S
tatu
s
Comment
s File Name [e.g yymmdd; tbd...]
No
. o
f B
an
ds
Ba
nd
Re
gis
tra
tio
n
Sp
ect
ral/
Sp
ati
al
Re
solu
tio
n
Pleiades 1B
1. IMG_PHR1B_MS_201508280834575_SEN_2226864101-002
Y - Q Y - V Q Y - V Q Y - V Q Y - V Q Y - Q Y - V Y - V Y - V Q Y - Q Y - Q
Y - V Q
Y - V Q
Y - V
N - V Q
Yes none
2. IMG_PHR1B_MS_201606110819501_SEN_2226863101-002
Y - Q Y - V Q Y - V Q Y - V Q Y - V Q Y - Q Y - V Y - V Y - V Q Y - Q Y - Q
Y - V Q
Y - V Q
Y - V
N - V Q
Yes none
Quickbird-2
3. 05APR26083225-M2AS-056358854030_01_P001
Y - Q Y - V Q Y - V Q Y - V Q Y - V Q Y - Q Y - V Y - V Y - V Q Y - Q Y - Q
Y - V Q
Y - V Q
Y - V
N - V Q
Yes none
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 9
2.2 In-situ Data Quality
Dataset 1 (14) No Data used
Ba
cku
p
Re
ad
ab
ility
(1)
Ch
eck
He
ad
er
/ M
eta
da
ta (2
)
Ext
en
t
Pro
ject
ion
/ S
ph
ero
id
(16
)
Sp
ati
al R
eso
luti
on
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Lo
cati
on
Co
mp
lete
ne
ss
Ge
om
. M
isa
lign
me
nt
Pla
usi
bili
ty
Dro
pp
ed
Lin
es
/
Art
efa
cts
(7)
Acc
ep
tan
ce S
tatu
s
Comments File Name [e.g yymmdd; tbd...]
No In-situ Data Used
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 10
2.3 Ancillary Data Quality
Dataset 1 (14)
LULC
Re
ad
ab
ility
(1
)
Ch
eck
He
ad
er
/Me
tad
ata
(2)
Da
ta C
ove
rage
M
atc
he
s S
erv
ice
Are
a
(%)
Pro
ject
ion
/ S
ph
ero
id
, E
PS
G(1
6)
Sp
ati
al R
eso
luti
on
an
d u
nit
(e
.g.
m, k
m)
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Co
mp
lete
ne
ss
(Ve
cto
r: A
ttri
bu
te
Ta
ble
; R
ast
er:
Th
em
ati
c V
alu
es)
Ge
om
. M
isa
lign
me
nt
Dro
pp
ed
Lin
es
/ A
rte
fact
s (
EO
da
ta
on
ly)(
7)
Uti
lity
for
Cu
rre
nt
Pro
ject
File Name [e.g yymmdd; tbd...]
EO4SD-Urban LULC Product ☒ Yes
☐ No
☒ Complete (INSPIRE/ISO19119)
☐ Incomplete
☐ not available
☒ Yes
☐ No
☐ Unknown If no: xxx%
☒ Correct
☐ Incorrect
☐ Unknown EPSG: 32749
n.a. *.shp n.a. ☒ Complete
☐Incomplete
☒ No
☐ Yes
☐ Unknown If yes: XX m
☒ No
☐ Yes
☐ n.a. If yes: xxx %
☐ None
☐ Partial
☒ Full
Comments: /
Dataset 2 (14) Report
Re
ad
ab
ility
(1)
Ch
eck
He
ad
er
/Me
tad
ata
(2)
Da
ta C
ove
rage
M
atc
he
s S
erv
ice
Are
a (
%)
Pro
ject
ion
/ S
ph
ero
id
, E
PS
G(1
6)
Sp
ati
al R
eso
luti
on
a
nd
un
it (
e.g
. m
, k
m)
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Co
mp
lete
ne
ss
(Ve
cto
r: A
ttri
bu
te
Ta
ble
; R
ast
er:
Th
em
ati
c V
alu
es)
Ge
om
. M
isa
lign
me
nt
Dro
pp
ed
Lin
es
/
Art
efa
cts
(E
O d
ata
o
nly
)(7
)
Uti
lity
for
Cu
rre
nt
Pro
ject
File Name [e.g yymmdd; tbd...]
KIGOMA-UJIJI MASTER PLAN 2017-2037
☒ Yes
☐ No
☐ Complete
☐ Incomplete
☒ not available
☒ Yes
☐ No
☐ Unknown If no: xxx%
n.a. n.a. *.pd
f n.a. n.a. n.a. n.a.
☐ None
☐ Partial
☒ Full
Comments: No Metadata is available for the Master Plan of Kigoma.
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 11
3.1 Geometric Correction
Sensoren (8) QB-2, PLX
Pro
cess
ing D
ate
AO
I
Cit
y /
Re
gio
n /
Co
un
try
Pro
ject
ion
/ S
ph
ero
id (1
6)
No
. &
RM
S (
m)
of
GC
Ps
(17
)
No
. &
RM
S (
m)
of
TP
s (1
8)
No
. &
RM
S (
m)
of
CP
s (1
9)
Dig
ita
l E
leva
tio
n M
od
el
(DE
M)
Mo
de
l /
Alg
ori
thm
(13
)
Re
sam
plin
g
Me
tho
d (2
0)
Va
lida
tio
n R
ep
ort
s
Acc
ep
tan
ce S
tatu
s
Output File Name [e.g yymmdd; tbd...] N
o. File Name [e.g yymmdd; tbd...]
Pleiades 1B
1. DIM_PHR1B_MS_201508280834575_SEN_2226864101-002_toa_PSH
12.04.2017
Kigoma TZA
UTM35S / WGS84
MGCPs 12; RMSE x0,79; y0,69
4; RMSE x0,73; y 0,88
N/A SRTM
30
0 order poynom
CC Yes Yes oDIM_PHR1B_MS_201508280834575_SEN_2226864101-002_toa_PSH
2. DIM_PHR1B_MS_201606110819501_SEN_2226863101-002_toa_PSH
12.04.2017
Kigoma TZA
UTM35S / WGS84
MGCPs 12; RMSE x0,79; y0,69
4; RMSE x0,73; y 0,88
N/A SRTM 30
0 order poynom
CC Yes Yes oDIM_PHR1B_MS_201606110819501_SEN_2226863101-002_toa_PSH
Quickbird-2
3. 05APR26083225-M2AS-056358854030_01_P001_toa_PSH
12.04.2017
Kigoma TZA
UTM35S / WGS84
MGCPs 7; RMSE x1,07; y0,89
N/A N/A SRTM 30
Affin, 1st order poynom
CC Yes Yes o05APR26083225-M2AS-056358854030_01_P001_toa_PSH
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 12
3.1.1 Data Fusion
Dataset 1 (14) Input A Input B Output
Acc
ep
tan
ce S
tatu
s
Comments
Filename DIM_PHR1B_MS_201606110819501_SE
N_2226863101-002 DIM_PHR1B_P_201606110819501_SEN
_2226863101-001 DIM_PHR1B_MS_201606110819501_SEN
_2226863101-002_toa_PSH
none
Sensor Pleiades 1B Pleiades 1B Pleiades 1B
Method PCI / Pansharp2 PCI / Pansharp2 PCI / Pansharp2
Spatial Resolution / (MS/Pan)
2m MS 0,5m Pan 0,5m MS
Band Combination RGB NIR PAN RGB NIR
Data Format (3) & Bit Depth (5)
*.tif 16 Bit u *.tif 16 Bit u *.pix 16 Bit u Yes
Dataset 2 (14) Input A Input B Output
Acc
ep
tan
ce S
tatu
s
Comments
Filename DIM_PHR1B_MS_201508280834575_SE
N_2226864101-002 DIM_PHR1B_P_201508280834575_SEN
_2226864101-001 DIM_PHR1B_MS_201508280834575_SEN
_2226864101-002_toa_PSH
none
Sensor Pleiades 1B Pleiades 1B Pleiades 1B
Method PCI / Pansharp2 PCI / Pansharp2 PCI / Pansharp2
Spatial Resolution / (MS/Pan)
2m MS 0,5m Pan 0,5m MS
Band Combination RGB NIR PAN RGB NIR
Data Format (3) & Bit Depth (5)
*.tif 16 Bit u *.tif 16 Bit u *.pix 16 Bit u Yes
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 13
Dataset 3 (14) Input A Input B Output
Acc
ep
tan
ce S
tatu
s
Comments
Filename 05APR26083225-M2AS-
056358854030_01_P001 05APR26083225-P2AS-
056358854030_01_P001 05APR26083225-M2AS-
056358854030_01_P001_toa_PSH
none
Sensor Quickbird-2 Quickbird-2 Quickbird-2
Method PCI / Pansharp2 PCI / Pansharp2 PCI / Pansharp2
Spatial Resolution / (MS/Pan)
2m MS 0,6m Pan 0,6m MS
Band Combination RGB NIR PAN BGR NIR
Data Format (3) & Bit Depth (5)
*.tif 16 Bit u *.tif 16 Bit u *.pix 16 Bit u Yes
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 14
3.2 Data Processing
Sensoren (8)
QB-2, PLX Atmospheric Correction
Radiometric Processing (15)
Topographic Normalisation
OTHER Calculation:
Acc
ep
tan
ce S
tatu
s
Ba
cku
p
Comment
No.
File Name [e.g yymmdd; tbd...] Processing
Date
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
Pleiades 1B
1. DIM_PHR1B_MS_201508280834575_SEN_2226864101-002
11.04.2017 No N/A No N/A No N/A Yes GAFmap / TOA
Yes No none
2. DIM_PHR1B_MS_201606110819501_SEN_2226863101-001
11.04.2017 No N/A No N/A No N/A Yes GAFmap / TOA
Yes No none
Quickbird-2
3. 05APR26083225-M2AS-056358854030_01_P001
11.04.2017 No N/A No N/A No N/A Yes GAFmap / TOA
Yes No none
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 15
4.1 Classification
Sensors (8) Pleiades, QB
Processing Date
Cloud Masking Thematic
Classification Manual
Enhancement Mosaicking
Acc
ep
tan
ce S
tatu
s
Ba
cku
p
Comment
No. File Name [e.g yymmdd; tbd...]
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
1.
DIM_PHR1B_MS_201508280834575_SEN_2226864101-002 DIM_PHR1B_MS_201606110819501_SEN_2226863101-001
30.05.2017
no No clouds Yes
GAFMap and ArcMap/visual mapping
Yes
GAFMap and ArcMap/Visual interpretation
Yes
ArcGIS 10 / Mosaic Dataset with manual neatline selection
Yes Yes
2. 05APR26083225-M2AS-056358854030_01_P001
21.08.2017
no No clouds Yes
GAFMap and ArcMap/visual mapping
Yes
GAFMap and ArcMap/Visual interpretation
Yes
ArcGIS 10 / Mosaic Dataset with manual neatline selection
Yes Yes
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 16
5.1 Thematic Accuracy
No. Product Name
Processing Date
Area Coverage
No
. o
f C
lass
es
Re
fere
nce
Da
ta
Sa
mp
le S
ize
Sa
mp
ling U
nit
(21
)
Sa
mp
ling
De
sign
(22
)
Sa
mp
le E
xclu
sio
n
Cri
teri
a (2
3)
Th
em
ati
c A
ccu
racy
(2
4)
Acc
ep
tan
ce S
tatu
s
Comment
1. Settlements 2016
25.10.2017 Full AOI (urban core area)
2 VHR EO Data
107 per strata
POINT Stratified random sampling None, all used 96.73% Yes
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 17
5.2 Error Matrices
Settlements 2016 Reference Data
Totals Planned Settlement Area Unplanned Settlement Area
Planned Settlement Area 101 6 107
Unplanned Settlement Area 1 106 107
Totals 102 112 214
Accuracy Statistics z= 1.96 Overall Accuracy: 96.73%
95% Confidence Interval: 94.11 99.35%
Class Name Producer’s Accuracy User’s Accuracy
Planned Settlement Area 99.02% 94.39%
Unplanned Settlement Area 94.64% 99.07%
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 18
6.1 Completeness
INPUT DATA
No. Item AOI Coverage [km²]
Area Coverage [km²]
Completeness of Coverage (25)
No. of Scenes Scenes used in Production
Completeness of Verification (27)
Metadata Comments
1. VHR EO Data Kigoma
85 km² 100% 6 6 100% Yes none
2. Ancillary Data N/A N/A N/A N/A N/A Yes none
PRODUCTS
No. Product (28) AOI Coverage [km²]
Product Coverage [km²]
Completeness of Coverage (25)
Completeness of Classification (26)
Unclassifiable Area [%]
Completeness of Verification (27)
Metadata Comments
1.
Planned and Unplanned Settlement Areas 2006 and 2016 and Change
Kigoma 85 km² 100% 100% 0% 100% Yes
Originally, 7 change classes were required to be mapped. However, only 6 classes were present in the change map of Kigoma.
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 19
6.2 Compliancy
Product 1 (28) Planned and Unplanned Settlement Areas 2006 and 2016
Abstract
The product Planned and Unplanned Settlements contains spatial explicit information on the two settlement types within the core urban area. The distinction into the two settlement types is restricted to the residential area (LULC class 11), and to commercial areas (LULC class 1211) which also have a residential component. The Planned and Unplanned Settlements, and Change dataset is developed, based on Very High Resolution (VHR) satellite imagery by means of visual interpretation.
Service / Product Specifications
Area Coverage
Requirements Achieved Specifications Compliancy Comments
Country 1: Tanzania Country 1: Tanzania YES None
A) Wall-to-wall: Kigoma city A) Wall-to-wall: Kigoma city
Core Urban Area (85 km²) Core Urban Area (85 km²)
Area of Interest: Kigoma 85 km², defined by the national user.
B) Sampling based: B) Sampling based:
N/A N/A
Time Period - Update Frequency
Requirements Achieved Specifications Compliancy Comments
Baseline Year(s): Baseline Year(s): Yes
See EO data acquisition dates. 2005 and 2015 (+/- 1 years) 2006 and 2016
B) Update Frequency B) Update Frequency Yes --
No update No update
Geographic Reference System
Requirements Achieved Specifications Compliancy Comments
WGS84 / UTM 35S WGS84 / UTM 35S YES None
Mapping Classes and Definitions
Requirements Achieved Specifications Compliancy
Single date
Planned settlements Planned settlements YES
Unplanned settlements Unplanned settlements YES
Change product
No change in planned settlement area No change in planned settlement area YES
No change in unplanned settlement area No change in unplanned settlement area YES
Expansion of planned settlement area Expansion of planned settlement area YES
Expansion of unplanned settlement area Expansion of unplanned settlement area YES
Decrease of planned settlement area Decrease of planned settlement area YES
Decrease of unplanned settlement area Decrease of unplanned settlement area YES
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 20
Unplanned to planned settlement area Unplanned to planned settlement area YES
Cloud and Shadow Detection and Removal
Requirements Achieved Specifications Compliancy Comments
N/A N/A
Spatial Resolution
Requirements Achieved Specifications Compliancy Comments
N/A N/A N/A None
Minimum Mapping Unit (MMU)
Requirements Achieved Specifications Compliancy Comments
0.25 ha 0.25 ha YES None
Data Type
Requirements Achieved Specifications Compliancy Comments
Vector data Vector data YES None
Bit Depth
Requirements Achieved Specifications Compliancy Comments
N/A N/A
Data Format
Requirements Achieved Specifications Compliancy Comments
ESRI SHP ESRI SHP YES None
Class Coding
Requirements Achieved Specifications Compliancy
N/A N/A
Metadata
Requirements Achieved Specifications Compliancy Comments
ISO standard 19115 "Metadata" and ISO 19139 "XML Scheme Implementation"
ISO standard 19115 "Metadata" and ISO 19139 "XML Scheme Implementation"
YES None
Service / Product Quality
Thematic Accuracy
Requirements Achieved Specifications Compliancy Comments
> 80% > 80% YES None
Positional Accuracy
Requirements Achieved Specifications Compliancy Comments
RMSE < 15 m RMSE < 15 m YES None
Delivery Procedure
Service Provision
Requirements Achieved Specifications Compliancy Comments
Online via FTP Online via FTP YES None
Delivery Date
Requirements Achieved Specifications Compliancy Comments
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 21
End of October 2017 End of October 2017 YES None
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 22
Glossary
Quality Checks
Prefix Suffix Explanation
Y (Yes) - V Visually checked
N (No) - Q Quantitatively/qualitatively checked
N/A - V Q Visually and quantitatively/qualitatively checked
- N/A Not applicable
Readability (1) Check readability of all required input data. Can the data be stored again?
Header / Metadata (2) Check Image Header Information and/or Metadata for completeness / distinctive features.
Data Format (3) For digital data, please give file format (e.g. *.tiff, *.shp).
Data Type (4) Please specify the type of the data (e.g. raster, vector or analogue).
Bit Depth (5) Please give pixel depth and sign of raster data (e.g. 8 bit unsigned integer).
Dynamic Range (6) Check dynamic range of all image bands. Visual check of dynamic range should be accompanied by histograms and statistics.
Dropped Lines & Artefacts (7) Check Image for dropped lines and other artefacts. If such occurs, please give description of extent and influence in the Comments section.
Sensor 1, 2, … (8) Please delete / add additional sensor sections as necessary.
Purpose (9)
The purpose and use of the given auxiliary or reference data should be stated:
ORHTOrectification, GEOmetric correction, REFerence, VERification source, POSitional ACCuracy assessment, THEmatic ACCuracy assessment.
Sampling Methodology (10)
Methodology of in situ or reference/auxiliary data sampling scheme should be outlined. In case of sample plots, also state how the plot positions have been determined (e.g. from GPS measurements, topographic maps, terrestrial triangulation, EO data, etc.) and how the sampling grid was established.
Positional Accuracy (11)
Positional accuracy of collected in situ data should be given. For reference/auxiliary data it MUST be given. If unknown, the data’s use must be explained. For DEM or other data with 3D information please specify both vertical and horizontal Positional Accuracy. For analogue data (e.g. maps) try to give approximate accuracy related to mapping scale.
Completeness (12) Data should be checked for spatial/temporal/content gaps.
Model / Algorithm (13)
Give the name of the software and its version. Specify the software module/algorithm used for: a) geometric correction, e.g., Polynomial and its degree, Rational Functions, Thin Plate Spline, etc. b) classification, e.g., ISODATA, Maximum Likelihood, Neural Networks, etc.
Dataset 1, 2, … (14) Please delete / add additional dataset sections as necessary.
Radiometric Processing (15) State whether (and which) radiometric processing was applied (e.g., contrast enhancement, histogram matching, bundle block adjustment, radiometric normalisation, filtering, etc.).
Projection; Spheroid / Ellipsoid (16)
Always specify completely, i.e. at minimum the Projection (+Zone, if applicable), Spheroid / Ellipsoid, Map Datum. Give additional information if necessary to unambiguously define the reference system.
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 23
Ground Control Points (GCPs) (17)
Give the number, distribution and RMS of used Ground Control Points (as average per scene) as obtained from the Geometric Correction, in meters [m]. Optionally, also RMSx and RMSy may be given. Distribution of GCPs should be attached as a snapshot, or described in the Comments section.
Tie Points (TPs) (18)
In case of mosaicking, give the number of used Tie Points and their total RMS (as average per scene) as obtained from the Geometric Correction, in meters [m]. Optionally, also RMSx and RMSy may be given. Distribution of TPs should be attached as a snapshot, or described in the Comments section.
Check Points (CPs) (19)
The real measure of positional accuracy and the only measure which should be examined as to its Acceptable Range. Give the independent Check Points’ total RMS (as average per scene) as obtained from the Geometric Correction, in meters [m]. Optionally, also RMSx and RMSy may be given. Distribution of CP's should be attached as a snapshot, or described in the Comments section.
Resampling Method (20) Specify, if / which resampling algorithm has been used (e.g. NN=Nearest Neighbour, BIL=Bilinear Interpolation, CC=Cubic Convolution).
Sampling unit (21) Specify sampling unit of Accuracy Assessment as POINT, FRAME, POLYGON and how it is treated (e.g. pixel center, polygon centre, etc.).
Sampling Design (22) SYST=Systematic, RAND=Random, STRAT=Stratified, SBCLASS=Stratified by class, SBAREA=Stratified by area
Sample exclusion criteria (23)
Describe which criteria you apply for sampling point selection resp. exclusion of certain points. For example, if the point is too close to a class boundary (less than 1 pixel), it is excluded. If the selected sample point is not representative of the class , it is excluded. For these reasons, it is recommended to oversample by 10% to compensate for sample point exclusion.
Thematic Accuracy (24)
Provide a detailed description of the Accuracy Assessment results in the form of error matrices showing commission and omission errors, user’s , producer’s and overall accuracies and other measures of Thematic Accuracy, as the confidence level (usually fixed at 95%) and the respective confidence interval, at least for the overall accuracy. If classification is done in a phased approach, e.g. if Forest Area and subsequently Forest Type are mapped, independent reports have to be produced.
Completeness of Coverage (25) State whether the coverage is limited to a subset, or portion of the final product.
Completeness of Classification (26)
State whether classification was constrained to a subset, or portion of the final product.
Completeness of Verification (27)
State whether verification applied to lineage, positional, or thematic accuracy is constrained to a subset, or portion of the final product.
Product (28) Please delete / add additional product sections as necessary.
Lineage (29) Definition (INSPIRE 2015): Lineage is “a statement on process history and/or overall quality of the spatial data set. Where appropriate it may include a statement whether the data set has been validated or quality assured, whether it is the official version (if multiple versions exist), and whether it has legal validity. The value domain of this element is free text.”
Source(30) Definition (INSPIRE, 2015): “This is the description of the organisation responsible for the establishment, management, maintenance or distribution of the resource. This description shall include: name of the organisation and contact email address.”
Earth Observation for Sustainable Doc. No.: City-Operations Report
Development – Urban Project Issue/Rev-No.: 3.0
Annex 2 to EO4SD-Urban Kigoma City Operations Report Page 4
This page is intentionally left blank!
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 1
Earth Observation for Sustainable Development - Urban
Quality Assurance and Quality Control Sheets These QA/QC Templates were prepared by GAF AG compliant with ISO 9001:2008 Quality Management System
standards and can only be used by Partners in the current EO4SD-Urban Project.
Project Title: EO4SD-Urban
Project Leader: GAF AG
Service Provider: GAF Editor: D. Angelova
Client: Product:
WB
Population Distribution and Density Date: 02.11.2017
Overview of QC-Sheets and Processing Steps Sheet used
Sheet filled in
Requirements
0.1 Requirements Yes Yes
Specifications of Input Data
1.1 List of EO Data Yes Yes
1.2 List of In-situ Data Yes Yes
1.3 List of Ancillary Data Yes Yes
Data Quality Checks
2.1 EO Data Quality Yes Yes
2.2 In-situ Data Quality Yes Yes
2.3 Ancillary Data Quality Yes Yes
Pre-Processing of EO Data
3.1 Geometric Correction Yes Yes
3.1.1 Data Fusion Yes Yes
3.2 Data Processing Yes Yes
Thematic Processing
4.1 Classification Yes Yes
Accuracy Assessment
5.1 Thematic Accuracy Yes Yes
5.2 Error Matrices Yes Yes
Delivery Checks / Delivery
6.1 Completeness Yes Yes
6.2 Compliancy Yes Yes
Glossary (index numbers in the QA/QC tables refer to the glossary at the end of this document) Further QC-relevant Documents:
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 2
Comments / Characteristics:
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 3
0.1 Requirements
Product 1 (28) Population Distribution and Density Change 2005 and 2015
Abstract
The Population Distribution and Density Change product contains spatial explicit information about population distribution changes at a certain degree, within the Core Urban Districts of Kigoma. The following eight change classes are identified, based on a frequency distribution analysis of the changes that occur: Unchanged Population Distribution; Up to –100% decrease; Up to 200% increase; 201% - 400% increase; 401% - 600% increase; 601% - 800% increase; 801% - 1000% increase; More than 1000% increase. The Population Distribution and Density Change product is estimated for all objects which belong to the residential class (LULC class 11), using 5 different data sources: LULC Product, WorldPop data (2015) with a spatial resolution of 100m, official population census data from 2002 and 2012, Soil Sealing layer and Administrative Boundaries.
Service / Product Specifications
Area Coverage
Country: Tanzania A) Wall-to-wall: n/a
City: Kigoma Selected Sites: Area of Interest defined with end-users
Area km² Core Urban: 85 km2 B) Sampling based: n/a
Time Period - Update Frequency
A) Baseline Year(s): B) Update Frequency
2005 and 2015 2 points in time
Comments: None
Geographic Reference System
WGS84 UTM Zone 35S
Mapping Classes and Definitions
Single date
1 Population Density between 0 – 150 inhabitants/ km2
2 Population Density between 151 – 300 inhabitants/ km2
3 Population Density between 301 – 600 inhabitants/ km2
4 Population Density between 601 – 1500 inhabitants/ km2
5 Population Density between 1501 – 3000 inhabitants/ km2
6 Population Density between 3001 – 5000 inhabitants/ km2
7 Population Density between 5001 – 9000 inhabitants/ km2
8 Population Density between 9001 – 18000 inhabitants/ km2
9 Population Density between 18001 - 41000 inhabitants/ km2
Change product
1 Unchanged Population Distribution
2 Up to -100% decrease
3 Up to 200% increase
4 201% - 400% increase
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 4
5 401% - 600% increase
6 601% - 800% increase
7 801% - 1000% increase
8 More than 1000% increase
Cloud and Cloud Shadow Detection and Removal
n/a
Spatial Resolution
n.a. (Product provided as Shapefile)
Minimum Mapping Unit (MMU)
n.a.
Data Type & Format
Shapefile *.shp
Bit Depth
n.a.
Class Coding
Class Code Class Name RGB Code
Single date
1 Population Density between 0 – 150 inhabitants/ km2 242/241/162
2 Population Density between 151 – 300 inhabitants/ km2 250/245/105
3 Population Density between 301 – 600 inhabitants/ km2 255/255/38
4 Population Density between 601 – 1500 inhabitants/ km2 255/183/0
5 Population Density between 1501 – 3000 inhabitants/ km2 255/85/0
6 Population Density between 3001 – 5000 inhabitants/ km2 247/5/122
7 Population Density between 5001 – 9000 inhabitants/ km2 240/7/228
8 Population Density between 9001 – 18000 inhabitants/ km2 161/14/230
9 Population Density between 18001 - 41000 inhabitants/ km2 93/26/201
Change product
1 Unchanged Population Distribution 156/156/156
2 Up to -100% decrease 112/153/89
3 Up to 200% increase 242/237/194
4 201% - 400% increase 237/215/114
5 401% - 600% increase 242/176/99
6 601% - 800% increase 237/175/175
7 801% - 1000% increase 196/113/96
8 More than 1000% increase 120/35/35
Metadata
ISO standard 19115 "Metadata" and ISO 19139 "XML Scheme Implementation"
Service / Product Quality
Thematic Accuracy
> 80%
Positional Accuracy
RMSE < 15 m
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 5
Delivery Procedure
Service Provision
Online via FTP
Delivery Date
End of October 2017
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 6
1.1 List of EO Data
Sensoren (8) No Data used
Incoming Date
Acquisition Date
Proc. Level
Path / Row
AOI - City / Region / Country
Spatial Res.
No. of Bands
Cloud Cover (Data Provider)
Projection / Spheroid (16)
Data Format (3)
Bit Depth (5)
Header / Metadata (2)
File Name [e.g yymmdd; tbd...]
No EO Data Used
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 7
1.2 List of In-situ Data
Dataset 1 (14) Incoming Date Acquisition Date AOI - City / Region / Country
Data Type (4)
Projection / Spheroid (16)
No. of Sample Plots
Sampling Design (22)
Positional Accuracy (11)
Purpose (9)
No In-situ Data Used
Lineage:
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 8
1.3 List of Ancillary Data
Dataset 1 (14)
LULC Date of Receipt from Client
Metadata (2)
Reference Mapping Date / Date of Creation
Geographic Area/- City / Region / Country
Area Coverage
Data Type (4)
Data Format (3)
Projection / Spheroid (16)
Positional Accuracy (11)
No. of Classes
Thematic Accuracy
Availability of Class Definitions
EO4SD-Urban Land Use/Land Cover Baseline Product
2017-08-30 Yes
(INSPIRE) 2006 and 2016
Kigoma (urban area) 100% Vector *.shp EPSG:32736, WGS 84 / UTM zone 35S
< 3 m 26 97% Yes
Lineage(29): The Urban Land Use/Land Cover Baseline Product and the derived Soil Sealing layer are used to interpolate the population distribution and density for each residential unit, within the Core Urban Districts of Kigoma.
Source (30):: GAF AG
Dataset 2 (14)
Report Date of Receipt from Client
Metadata (2)
Reference Mapping Date / Date of Creation
Geographic Area/- City / Region / Country
Area Coverage
Data Type (4)
Data Format (3)
Projection / Spheroid (16)
Positional Accuracy (11)
No. of Classes
Thematic Accuracy
Availability of Class Definitions
KIGOMA-UJIJI MASTER PLAN 2017-2037
2017-05-18 n.a. December, 2016
Kigoma/ Tanzania Kigoma Document
pdf n.a. n.a. n.a. n.a. n.a.
Lineage(29): /
Source (30): Received from MaryGrace Weber.
Dataset 3 (14)
Population Raster
Dataset
Date of Receipt from Client
Metadata (2)
Reference Mapping Date / Date of Creation
Geographic Area/- City / Region / Country
Area Coverage
Data Type (4)
Data Format (3)
Projection / Spheroid (16)
Positional Accuracy (11)
No. of Classes
Thematic Accuracy
Availability of Class Definitions
WorldPop Dataset 2017-09-01 Yes 2015 Kigoma/Tanzania 100% Raster Geotiff Geographic, WGS84
n.a. n.a. n.a. n.a.
Lineage(29): The WorldPop Dataset is used to interpolate the current population data for each residential object in the Core Urban Districts of Kigoma.
Source (30): Downloaded from WorldPop: http://www.worldpop.org.uk/
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 9
Dataset 4 (14)
Report Date of Receipt from Client
Metadata (2)
Reference Mapping Date / Date of Creation
Geographic Area/- City / Region / Country
Area Coverage
Data Type (4)
Data Format (3)
Projection / Spheroid (16)
Positional Accuracy (11)
No. of Classes
Thematic Accuracy
Availability of Class Definitions
2012 POPULATION AND HOUSING CENSUS
2017-09-01 n.a. 2012-08-26 Tanzania National coverage
Document
pdf n.a. n.a. n.a. n.a. n.a.
Lineage(29): The 2012 Population and Housing Census report is used to project the number of inhabitants per district, and to adjust the population residuals in the current Population Distribution and Density product.
Source (30): Downloaded from United Republic of Tanzania, Ministry of East African Cooperation: http://meac.go.tz/
Dataset 5 (14)
Report Date of Receipt from Client
Metadata (2)
Reference Mapping Date / Date of Creation
Geographic Area/- City / Region / Country
Area Coverage
Data Type (4)
Data Format (3)
Projection / Spheroid (16)
Positional Accuracy (11)
No. of Classes
Thematic Accuracy
Availability of Class Definitions
2002 POPULATION AND HOUSING CENSUS
2017-09-01 Yes
(The World Bank)
2002-08-22 Tanzania National coverage
Document
pdf n.a. n.a. n.a. n.a. n.a.
Lineage(29): The 2002 Population and Housing Census report is used to interpolate the population data in the historic Population Distribution and Density product.
Source (30): Downloaded from the IPUMS – International Website: https://international.ipums.org/international/
Dataset 6 (14)
Administrative
Boundaries
Date of Receipt from Client
Metadata (2)
Reference Mapping Date / Date of Creation
Geographic Area/- City / Region / Country
Area Coverage
Data Type (4)
Data Format (3)
Projection / Spheroid (16)
Positional Accuracy (11)
No. of Classes
Thematic Accuracy
Availability of Class Definitions
TZA_adm3 2017-09-01 Yes
2015 Tanzania
National coverage
Vector *.shp WGS 84 unknown n.a. unknown Yes
Lineage(29): This shapefile provides with information, regarding the administrative boundaries at a district level in Tanzania.
Source (30): Downloaded from Global Administrative Areas: http://www.gadm.org/country
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 10
2.1 EO Data Quality
Sensoren(8) No Data used
Ba
cku
p
Re
ad
ab
ility
(1
)
Ch
eck
He
ad
er
/
Me
tad
ata
(2)
Band Specifications
Pro
ject
ion
/ S
ph
ero
id (1
6)
Sce
ne
Lo
cati
on
Co
mp
lete
ne
ss o
f A
dd
itio
na
l D
ata
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Clo
ud
Co
ver
(in
tern
al
che
ck)
Ra
dio
me
try
To
po
gra
ph
y
Dro
pp
ed
Lin
es
/ A
rte
fact
s (7
)
Acc
ep
tan
ce S
tatu
s
Comments File Name [e.g yymmdd; tbd...]
No
. o
f B
an
ds
Ba
nd
Re
gis
tra
tio
n
Sp
ect
ral/
Sp
ati
al
Re
solu
tio
n
No EO Data Used
n.a n.a n.a n.a n.a n.a n.a n.a n.a n.a n.a n.a n.a n.a n.a n.a n.a n.a
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 11
2.2 In-situ Data Quality
Dataset 1 (14) No Data used
Ba
cku
p
Re
ad
ab
ility
(1)
Ch
eck
He
ad
er
/ M
eta
da
ta (2
)
Ext
en
t
Pro
ject
ion
/ S
ph
ero
id
(16
)
Sp
ati
al R
eso
luti
on
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Lo
cati
on
Co
mp
lete
ne
ss
Ge
om
. M
isa
lign
me
nt
Pla
usi
bili
ty
Dro
pp
ed
Lin
es
/
Art
efa
cts
(7)
Acc
ep
tan
ce S
tatu
s
Comments File Name [e.g yymmdd; tbd...]
No In-situ Data Used n.a n.a n.a n.a n.a n.a n.a n.a n.a n.a n.a n.a n.a n.a /
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 12
2.3 Ancillary Data Quality
Dataset 1 (14)
LULC
Re
ad
ab
ility
(1
)
Ch
eck
He
ad
er
/Me
tad
ata
(2)
Da
ta C
ove
rage
M
atc
he
s S
erv
ice
Are
a
(%)
Pro
ject
ion
/ S
ph
ero
id
, E
PS
G(1
6)
Sp
ati
al R
eso
luti
on
an
d u
nit
(e
.g.
m, k
m)
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Co
mp
lete
ne
ss
(Ve
cto
r: A
ttri
bu
te
Ta
ble
; R
ast
er:
Th
em
ati
c V
alu
es)
Ge
om
. M
isa
lign
me
nt
Dro
pp
ed
Lin
es
/ A
rte
fact
s (
EO
da
ta
on
ly)(
7)
Uti
lity
for
Cu
rre
nt
Pro
ject
File Name [e.g yymmdd; tbd...]
EO4SD-Urban LULC Product ☒ Yes
☐ No
☒ Complete (INSPIRE/ISO19119)
☐ Incomplete
☐ not available
☒ Yes
☐ No
☐ Unknown If no: xxx%
☒ Correct
☐ Incorrect
☐ Unknown EPSG: 32749
n.a. *.shp n.a. ☒ Complete
☐Incomplete
☒ No
☐ Yes
☐ Unknown If yes: XX m
☒ No
☐ Yes
☐ n.a. If yes: xxx %
☐ None
☐ Partial
☒ Full
Comments: /
Dataset 2 (14) Report
Re
ad
ab
ility
(1)
Ch
eck
He
ad
er
/Me
tad
ata
(2)
Da
ta C
ove
rage
M
atc
he
s S
erv
ice
Are
a (
%)
Pro
ject
ion
/ S
ph
ero
id
, E
PS
G(1
6)
Sp
ati
al R
eso
luti
on
a
nd
un
it (
e.g
. m
, k
m)
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Co
mp
lete
ne
ss
(Ve
cto
r: A
ttri
bu
te
Ta
ble
; R
ast
er:
Th
em
ati
c V
alu
es)
Ge
om
. M
isa
lign
me
nt
Dro
pp
ed
Lin
es
/
Art
efa
cts
(E
O d
ata
o
nly
)(7
)
Uti
lity
for
Cu
rre
nt
Pro
ject
File Name [e.g yymmdd; tbd...]
KIGOMA-UJIJI MASTER PLAN 2017-2037
☒ Yes
☐ No
☐ Complete
☐ Incomplete
☒ not available
☒ Yes
☐ No
☐ Unknown If no: xxx%
n.a. n.a. *.pd
f n.a. n.a. n.a. n.a.
☐ None
☒ Partial
☐ Full
Comments: /
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 13
Dataset 3 (14) Population Raster Dataset
Re
ad
ab
ility
(1)
Ch
eck
He
ad
er
/Me
tad
ata
(2)
Da
ta C
ove
rage
M
atc
he
s S
erv
ice
Are
a
(%)
Pro
ject
ion
/ S
ph
ero
id
, E
PS
G(1
6)
Sp
ati
al R
eso
luti
on
an
d u
nit
(e
.g.
m, k
m)
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Co
mp
lete
ne
ss
(Ve
cto
r: A
ttri
bu
te
Ta
ble
; R
ast
er:
Th
em
ati
c V
alu
es)
Ge
om
. M
isa
lign
me
nt
Dro
pp
ed
Lin
es
/ A
rte
fact
s (
EO
da
ta
on
ly)(
7)
Uti
lity
for
Cu
rre
nt
Pro
ject
File Name [e.g yymmdd; tbd...]
WorldPop Dataset ☒ Yes
☐ No
☐ Complete
☒ Incomplete
☐ not available
☒ Yes
☐ No
☐ Unknown If no: xxx%
☒ Correct
☐ Incorrect
☐ Unknown EPSG: 32749
100m *.tif 32 Bit ☒ Complete
☐Incomplete
☒ No
☐ Yes
☐ Unknown If yes: XX m
☒ No
☐ Yes
☐ n.a. If yes: xxx %
☐ None
☒ Partial
☐ Full
Comments: The WorldPop Dataset is used only for the estimation of the current Population Distribution and Density product.
Dataset 4 (14) Report
Re
ad
ab
ility
(1
)
Ch
eck
He
ad
er
/Me
tad
ata
(2)
Da
ta C
ove
rage
M
atc
he
s S
erv
ice
Are
a (
%)
Pro
ject
ion
/ S
ph
ero
id
, E
PS
G(1
6)
Sp
ati
al R
eso
luti
on
an
d u
nit
(e
.g.
m, k
m)
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Co
mp
lete
ne
ss
(Ve
cto
r: A
ttri
bu
te
Ta
ble
; R
ast
er:
Th
em
ati
c V
alu
es)
Ge
om
. M
isa
lign
me
nt
Dro
pp
ed
Lin
es
/
Art
efa
cts
(E
O d
ata
o
nly
)(7
)
Uti
lity
for
Cu
rre
nt
Pro
ject
File Name [e.g yymmdd; tbd...]
2012 POPULATION AND HOUSING CENSUS
☒ Yes
☐ No
☐ Complete
☐ Incomplete
☒ not available
n.a. n.a. n.a. *.pdf n.a. n.a. n.a. n.a.
☐ None
☐ Partial
☒ Full
Comments: The 2012 Population Census report is used to derive the number of inhabitants per district in 2015 in Kigoma.
Dataset 5 (14) Report
Re
ad
ab
ility
(1)
Ch
eck
He
ad
er
/Me
tad
ata
(2)
Da
ta C
ove
rage
M
atc
he
s S
erv
ice
A
rea
(%
)
Pro
ject
ion
/ S
ph
ero
id
, E
PS
G(1
6)
Sp
ati
al R
eso
luti
on
an
d u
nit
(e
.g.
m, k
m)
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Co
mp
lete
ne
ss
(Ve
cto
r: A
ttri
bu
te
Ta
ble
; R
ast
er:
Th
em
ati
c V
alu
es)
Ge
om
. M
isa
lign
me
nt
Dro
pp
ed
Lin
es
/ A
rte
fact
s (
EO
da
ta
on
ly)(
7)
Uti
lity
for
Cu
rre
nt
Pro
ject
File Name [e.g yymmdd; tbd...]
2002 POPULATION AND HOUSING CENSUS
☒ Yes
☐ No
☒ Complete
☐ Incomplete
☐ not available
n.a. n.a. n.a. *.pdf n.a. n.a. n.a. n.a.
☐ None
☐ Partial
☒ Full
Comments: The 2002 Population Census report is used to derive the number of inhabitants per district in 2005 in Kigoma.
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 14
Dataset 6 (14)
Administrative Boundaries
Re
ad
ab
ility
(1
)
Ch
eck
He
ad
er
/Me
tad
ata
(2
)
Da
ta C
ove
rage
Ma
tch
es
Se
rvic
e A
rea
(%
)
Pro
ject
ion
/ S
ph
ero
id ,
E
PS
G(1
6)
Sp
ati
al R
eso
luti
on
an
d u
nit
(e.g
. m
, k
m)
Da
ta F
orm
at
(3)
Bit
De
pth
(5)
Co
mp
lete
ne
ss (
Ve
cto
r:
Att
rib
ute
Ta
ble
; R
ast
er:
Th
em
ati
c V
alu
es)
Ge
om
. M
isa
lign
me
nt
Dro
pp
ed
Lin
es
/ A
rte
fact
s
(EO
da
ta o
nly
)(7
)
Uti
lity
for
Cu
rre
nt
Pro
ject
File Name [e.g yymmdd; tbd...]
TZA_adm3 ☒ Yes
☐ No
☐ Complete
☒ Incomplete
☐ not available
☒ Yes
☐ No
☐ Unknown If no: xxx%
☒ Correct
☐ Incorrect
☐ Unknown EPSG: 32749
n.a. *.shp n.a. ☒ Complete
☐Incomplete
☒ No
☐ Yes
☐ Unknown If yes: XX m
☒ No
☐ Yes
☐ n.a. If yes: xxx %
☐ None
☐ Partial
☒ Full
Comments: /
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 15
3.1 Geometric Correction
Sensoren (8) No Data used
Pro
cess
ing D
ate
AO
I
Cit
y /
Re
gio
n /
Co
un
try
Pro
ject
ion
/ S
ph
ero
id (1
6)
No
. &
RM
S (
m)
of
GC
Ps
(17
)
No
. &
RM
S (
m)
of
TP
s (1
8)
No
. &
RM
S (
m)
of
CP
s (1
9)
Dig
ita
l E
leva
tio
n M
od
el
(DE
M)
Mo
de
l /
Alg
ori
thm
(13
)
Re
sam
plin
g
Me
tho
d (2
0)
Va
lida
tio
n R
ep
ort
s
Acc
ep
tan
ce S
tatu
s
Output File Name [e.g yymmdd; tbd...] N
o. File Name [e.g yymmdd; tbd...]
No EO Data Used
3.1.1 Data Fusion
Dataset 1 (14) Input A Input B Output
Acc
ep
tan
ce S
tatu
s
Comments
Filename n.a n.a n.a
none
Sensor n.a n.a n.a
Method n.a n.a n.a
Spatial Resolution / (MS/Pan) n.a n.a n.a n.a n.a n.a
Band Combination n.a n.a n.a
Data Format (3) & Bit Depth (5) n.a n.a n.a n.a n.a n.a n.a
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 16
3.2 Data Processing
Sensoren (8)
No Data used Atmospheric Correction
Radiometric Processing (15)
Topographic Normalisation
OTHER Calculation:
Acc
ep
tan
ce S
tatu
s
Ba
cku
p
Comment
No. File Name [e.g yymmdd; tbd...] Processing
Date
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
No EO Data Used
4.1 Classification
Sensors (8) No Data used
Processing Date
Cloud Masking Thematic
Classification Manual Enhancement Mosaicking
Acc
ep
tan
ce S
tatu
s
Ba
cku
p
Comment
No. File Name [e.g yymmdd; tbd...]
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
pro
cess
ed
Software / Method
No EO Data Used
5.1 Thematic Accuracy
No. Product Name Processing Date
Area Coverage
No
. o
f C
lass
es
Re
fere
nce
Da
ta
Sa
mp
le S
ize
Sa
mp
ling U
nit
(21
)
Sa
mp
ling
De
sign
(22
)
Sa
mp
le E
xclu
sio
n
Cri
teri
a (2
3)
Th
em
ati
c A
ccu
racy
(2
4)
Acc
ep
tan
ce S
tatu
s
Comment
1.
Population Distribution and Density Change 2005 and 2015
13.10.2017
Full AOI (Core Urban Districts)
8 n.a n.a n.a n.a n.a n.a Yes
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 17
5.2 Error Matrices
The estimated population per district corresponds to the 2012 population and housing census data.
In a disaggregation procedure, such as the one herein described, the outcome of the process is never less accurate than the original source data. By disaggregating
numerical data from one coarse geometry to a finer geometry, we always gain detail and approximate ground truth without the risk of deteriorating the source
information. The degree to which the disaggregation approximates reality, however, varies greatly, and it depends chiefly on: 1) the quality of the ancillary data
and 2) the appropriateness of the disaggregation algorithm and its parameters.
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 18
6.1 Completeness
INPUT DATA
No. Item AOI Coverage [km²]
Area Coverage [km²]
Completeness of Coverage (25)
No. of Scenes Scenes used in Production
Completeness of Verification (27)
Metadata Comments
1. Ancillary Data Kigoma 85 km² 100% 1 1 100% Yes none
PRODUCTS
No. Product (28) AOI Coverage [km²]
Product Coverage [km²]
Completeness of Coverage (25)
Completeness of Classification (26)
Unclassifiable Area [%]
Completeness of Verification (27)
Metadata Comments
1.
Population Distribution and Density Change 2005 and 2015
Kigoma 85 km² 100% 100% 0% 100% Yes none
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 19
6.2 Compliancy
Product 1 (28) Population Distribution and Density Change 2005 and 2015
Abstract
The Population Distribution and Density Change product contains spatial explicit information about population distribution changes at a certain degree, within the Core Urban Districts of Kigoma. The following eight change classes are identified, based on a frequency distribution analysis of the changes that occur: Unchanged Population Distribution; Up to –100% decrease; Up to 200% increase; 201% - 400% increase; 401% - 600% increase; 601% - 800% increase; 801% - 1000% increase; More than 1000% increase. The Population Distribution and Density Change product is estimated for all objects which belong to the residential class (LULC class 11), using 5 different data sources: LULC Product, WorldPop data (2015) with a spatial resolution of 100m, official population census data from 2002 and 2012, Soil Sealing layer and Administrative Boundaries.
Service / Product Specifications
Area Coverage
Requirements Achieved Specifications Compliancy Comments
Country 1: Tanzania Country 1: Tanzania YES None
A) Wall-to-wall: Kigoma city A) Wall-to-wall: Kigoma city
Core Urban Area (85 km²) Core Urban Area (85 km²)
Area of Interest: Kigoma 85 km², defined by the national user.
B) Sampling based: B) Sampling based:
N/A N/A
Time Period - Update Frequency
Requirements Achieved Specifications Compliancy Comments
Baseline Year(s): Baseline Year(s): Yes
See EO data acquisition dates. 2005 and 2015 (+/- 1 years) 2005 and 2015
B) Update Frequency B) Update Frequency Yes --
No update No update
Geographic Reference System
Requirements Achieved Specifications Compliancy Comments
WGS84 / UTM 35S WGS84 / UTM 35S YES None
Mapping Classes and Definitions
Requirements Achieved Specifications Compliancy
Single date
Population Density between 0 – 150 inhabitants/ km2 0 – 150 inhabitants/ km2 YES
Population Density between 151 – 300 inhabitants/ km2 151 – 300 inhabitants/ km2 YES
Population Density between 301 – 600 inhabitants/ km2 301 – 600 inhabitants/ km2 YES
Population Density between 601 – 1500 inhabitants/ km2 601 – 1500 inhabitants/ km2 YES
Population Density between 1501 – 3000 inhabitants/ km2 1501 – 3000 inhabitants/ km2 YES
Population Density between 3001 – 5000 inhabitants/ km2 3001 – 5000 inhabitants/ km2 YES
Population Density between 5001 – 9000 inhabitants/ km2 5001 – 9000 inhabitants/ km2 YES
Population Density between 9001 – 18000 inhabitants/ km2 9001 – 18000 inhabitants/ km2 YES
Population Density between 18001 - 41000 inhabitants/ km2 18001 - 41000 inhabitants/ km2 YES
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 20
Change product
Unchanged Population Distribution Unchanged Population Distribution YES
Up to -100% decrease Up to -100% decrease YES
Up to 200% increase Up to 200% increase YES
201% - 400% increase 201% - 400% increase YES
401% - 600% increase 401% - 600% increase YES
601% - 800% increase 601% - 800% increase YES
801% - 1000% increase 801% - 1000% increase YES
More than 1000% increase More than 1000% increase YES
Cloud and Shadow Detection and Removal
Requirements Achieved Specifications Compliancy Comments
N/A N/A
Spatial Resolution
Requirements Achieved Specifications Compliancy Comments
N/A N/A N/A None
Minimum Mapping Unit (MMU)
Requirements Achieved Specifications Compliancy Comments
N/A N/A N/A None
Data Type
Requirements Achieved Specifications Compliancy Comments
Vector data Vector data YES None
Bit Depth
Requirements Achieved Specifications Compliancy Comments
N/A N/A
Data Format
Requirements Achieved Specifications Compliancy Comments
ESRI SHP ESRI SHP YES None
Class Coding
Requirements Achieved Specifications Compliancy
N/A N/A
Metadata
Requirements Achieved Specifications Compliancy Comments
ISO standard 19115 "Metadata" and ISO 19139 "XML Scheme Implementation"
ISO standard 19115 "Metadata" and ISO 19139 "XML Scheme Implementation"
YES None
Service / Product Quality
Thematic Accuracy
Requirements Achieved Specifications Compliancy Comments
> 80% The estimated population per district corresponds to the 2012 population and housing census data.
YES None
Positional Accuracy
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 21
Requirements Achieved Specifications Compliancy Comments
RMSE < 15 m 15m YES None
Delivery Procedure
Service Provision
Requirements Achieved Specifications Compliancy Comments
Online via FTP Online via FTP YES None
Delivery Date
Requirements Achieved Specifications Compliancy Comments
End of October 2017 End of October 2017 YES None
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 22
Glossary
Quality Checks
Prefix Suffix Explanation
Y (Yes) - V Visually checked
N (No) - Q Quantitatively/qualitatively checked
N/A - V Q Visually and quantitatively/qualitatively checked
- N/A Not applicable
Readability (1) Check readability of all required input data. Can the data be stored again?
Header / Metadata (2) Check Image Header Information and/or Metadata for completeness / distinctive features.
Data Format (3) For digital data, please give file format (e.g. *.tiff, *.shp).
Data Type (4) Please specify the type of the data (e.g. raster, vector or analogue).
Bit Depth (5) Please give pixel depth and sign of raster data (e.g. 8 bit unsigned integer).
Dynamic Range (6) Check dynamic range of all image bands. Visual check of dynamic range should be accompanied by histograms and statistics.
Dropped Lines & Artefacts (7) Check Image for dropped lines and other artefacts. If such occurs, please give description of extent and influence in the Comments section.
Sensor 1, 2, … (8) Please delete / add additional sensor sections as necessary.
Purpose (9)
The purpose and use of the given auxiliary or reference data should be stated:
ORHTOrectification, GEOmetric correction, REFerence, VERification source, POSitional ACCuracy assessment, THEmatic ACCuracy assessment.
Sampling Methodology (10)
Methodology of in situ or reference/auxiliary data sampling scheme should be outlined. In case of sample plots, also state how the plot positions have been determined (e.g. from GPS measurements, topographic maps, terrestrial triangulation, EO data, etc.) and how the sampling grid was established.
Positional Accuracy (11)
Positional accuracy of collected in situ data should be given. For reference/auxiliary data it MUST be given. If unknown, the data’s use must be explained. For DEM or other data with 3D information please specify both vertical and horizontal Positional Accuracy. For analogue data (e.g. maps) try to give approximate accuracy related to mapping scale.
Completeness (12) Data should be checked for spatial/temporal/content gaps.
Model / Algorithm (13)
Give the name of the software and its version. Specify the software module/algorithm used for: a) geometric correction, e.g., Polynomial and its degree, Rational Functions, Thin Plate Spline, etc. b) classification, e.g., ISODATA, Maximum Likelihood, Neural Networks, etc.
Dataset 1, 2, … (14) Please delete / add additional dataset sections as necessary.
Radiometric Processing (15) State whether (and which) radiometric processing was applied (e.g., contrast enhancement, histogram matching, bundle block adjustment, radiometric normalisation, filtering, etc.).
Projection; Spheroid / Ellipsoid (16)
Always specify completely, i.e. at minimum the Projection (+Zone, if applicable), Spheroid / Ellipsoid, Map Datum. Give additional information if necessary to unambiguously define the reference system.
Earth Observation for Sustainable
Development – Urban Project QA/QC Sheets developed by GAF AG
© 2017 GAF AG All Rights Reserved. Unless otherwise indicated, the templates of these QA/QC pages are copyrighted by GAF AG. No part of these pages, either text or image may be used for any purpose other than use in the EO4SD-Urban Project. Therefore, reproduction, modification, storage in a retrieval system or retransmission, in any form or by any means, electronic, mechanical or otherwise, for reasons other than personal use, is strictly prohibited without prior written permission Page 23
Ground Control Points (GCPs) (17)
Give the number, distribution and RMS of used Ground Control Points (as average per scene) as obtained from the Geometric Correction, in meters [m]. Optionally, also RMSx and RMSy may be given. Distribution of GCPs should be attached as a snapshot, or described in the Comments section.
Tie Points (TPs) (18)
In case of mosaicking, give the number of used Tie Points and their total RMS (as average per scene) as obtained from the Geometric Correction, in meters [m]. Optionally, also RMSx and RMSy may be given. Distribution of TPs should be attached as a snapshot, or described in the Comments section.
Check Points (CPs) (19)
The real measure of positional accuracy and the only measure which should be examined as to its Acceptable Range. Give the independent Check Points’ total RMS (as average per scene) as obtained from the Geometric Correction, in meters [m]. Optionally, also RMSx and RMSy may be given. Distribution of CP's should be attached as a snapshot, or described in the Comments section.
Resampling Method (20) Specify, if / which resampling algorithm has been used (e.g. NN=Nearest Neighbour, BIL=Bilinear Interpolation, CC=Cubic Convolution).
Sampling unit (21) Specify sampling unit of Accuracy Assessment as POINT, FRAME, POLYGON and how it is treated (e.g. pixel center, polygon centre, etc.).
Sampling Design (22) SYST=Systematic, RAND=Random, STRAT=Stratified, SBCLASS=Stratified by class, SBAREA=Stratified by area
Sample exclusion criteria (23)
Describe which criteria you apply for sampling point selection resp. exclusion of certain points. For example, if the point is too close to a class boundary (less than 1 pixel), it is excluded. If the selected sample point is not representative of the class , it is excluded. For these reasons, it is recommended to oversample by 10% to compensate for sample point exclusion.
Thematic Accuracy (24)
Provide a detailed description of the Accuracy Assessment results in the form of error matrices showing commission and omission errors, user’s , producer’s and overall accuracies and other measures of Thematic Accuracy, as the confidence level (usually fixed at 95%) and the respective confidence interval, at least for the overall accuracy. If classification is done in a phased approach, e.g. if Forest Area and subsequently Forest Type are mapped, independent reports have to be produced.
Completeness of Coverage (25) State whether the coverage is limited to a subset, or portion of the final product.
Completeness of Classification (26)
State whether classification was constrained to a subset, or portion of the final product.
Completeness of Verification (27)
State whether verification applied to lineage, positional, or thematic accuracy is constrained to a subset, or portion of the final product.
Product (28) Please delete / add additional product sections as necessary.
Lineage (29) Definition (INSPIRE 2015): Lineage is “a statement on process history and/or overall quality of the spatial data set. Where appropriate it may include a statement whether the data set has been validated or quality assured, whether it is the official version (if multiple versions exist), and whether it has legal validity. The value domain of this element is free text.”
Source(30) Definition (INSPIRE, 2015): “This is the description of the organisation responsible for the establishment, management, maintenance or distribution of the resource. This description shall include: name of the organisation and contact email address.”