eo4sd-urban project: cpl denpasar city report...this project has received funding from the european...

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.: 1.1

Date: 13.10.2017

EO4SD-Urban Project: CPL Denpasar City Report

Lead: Partners: Financed by:

Earth Observation for Sustainable Doc. No.: City-Operations Report

Development – Urban Project Issue/Rev-No.: 1.1

EO4SD-Urban Denpasar 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.



EO4SD-Urban Denpasar City Operations Report Page II

Summary

This document contains information related to the European Space Agency (ESA) supported project

“Earth Observation for Sustainable Development” Urban Applications (EO4SD-Urban) provision of

geo-spatial products to the World Bank City Planning Labs (CPL) programme for Denpasar, Indonesia.

Affiliation/Function Name Date

Prepared GAF AG S. Gomez, A. Broszeit 13/10/2017

Contributions Joanneum Research M. Hirschmugl, H. Proske 11/10/2017

Reviewed Quality Assurance Responsible K. Druessler 13/10/2017

Approved GAF AG, Project Coordinator T. Haeusler 13/10/2017

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

WB TTL G. Singh electronic copy

Document Status Sheet

Issue Date Details

1.0 30/08/2017 First Draft Document Issue

1.1 13/10/2017 Final Document Issue

Document Change Record

# Date Request Location Details

1 13/10/2017 Update of Draft

Version

New sections for

Flood Risk

product

description.

Report updated with Flood Risk product

description.



EO4SD-Urban Denpasar City Operations Report Page III

This page is intentionally left blank!



EO4SD-Urban Denpasar 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 City Planning Labs (CPL) programme and the counterpart City Authorities in

Denpasar (Indonesia). 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 current and historic

Urban Extent and Change

Transport Infrastructure - Road Network

Flood History Assessment

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. The

Derived products of Transport, Infrastructure and the Flood History product will be included in an

updated next Version of this Report.

This City Operations Report for Denpasar 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.



EO4SD-Urban Denpasar City Operations Report Page V




EO4SD-Urban Denpasar City Operations Report Page VI

Table of Contents

1 INTRODUCTION 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

2.3.1 Land Use/Land Cover Nomenclature ................................................................................................ 3

2.3.2 Transport Infrastructure – Road Networks ........................................................................................ 7

2.4 TERMS OF ACCESS ................................................................................................................... 7

3 SERVICE OPERATIONS ............................................................................................................ 8

3.1 SOURCE DATA ......................................................................................................................... 8

3.2 PROCESSING METHODS ........................................................................................................... 9

3.3 ACCURACY ASSESSMENT OF MAP PRODUCTS ........................................................................ 9

3.3.1 The Applied Sampling Design .......................................................................................................... 9

3.3.2 The Applied Response Design ........................................................................................................ 10

3.3.3 The Applied Analysis Design .......................................................................................................... 11

3.4 ACCURACY ASSESSMENT OF TRANSPORT NETWORK ........................................................... 12

3.5 QUALITY CONTROL/ASSURANCE .......................................................................................... 14

3.6 METADATA ............................................................................................................................ 15

4 ANALYSIS OF MAPPING RESULTS ..................................................................................... 17

4.1 URBAN EXTENT – DEVELOPMENTS FROM 2000, 2005, 2010 AND 2015 ................................ 17

4.2 LAND COVER/ LAND USE 2006 AND 2016 ............................................................................. 21

4.2.1 Spatial Distribution of Main LU/LC Change Categories ................................................................ 23

4.2.2 Changes of Agricultural Areas ........................................................................................................ 25

4.3 TRANSPORT NETWORK .......................................................................................................... 27

4.4 FLOOD RISK ASSESSMENT ..................................................................................................... 28

4.4.1 EO Data Used .................................................................................................................................. 28

4.4.2 Short Description of Methodological Approach ............................................................................. 28

4.4.3 Product Description and Accuracy Assessment .............................................................................. 30

4.4.4 Results ............................................................................................................................................. 32

5 REFERENCES ............................................................................................................................ 37

Annexes

Annex 1: Land Use and Land Cover Nomenclature and Definitions

Annex 2: Processing Methods for EO4SD-Urban Products

Annex 3: Filled Quality Control Sheets

Urban Land Use / Land Cover

Flood Risk Assessment



EO4SD-Urban Denpasar City Operations Report Page VII

List of Figures

Figure 1: Illustration of Core and Peri-Urban Areas for Mapping.........................................................2 Figure 2: Land Use Map of Denpasar, 2012 (Reference: World Bank, CPL Programme, 2017)..........4 Figure 3: Mapping result of the city of Denpasar of the year 2016 overlaid with randomly

distributed sample points for accuracy assessment ..............................................................11 Figure 4: Example of the applied sampling design to generate randomly distributed point for the

accuracy assessment of the road network .............................................................................13 Figure 5: 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 ................................................................................................................13 Figure 6: Quality Control process for EO4SD-Urban product generation; at each intermediate

processing step output properties are compared against pre-defined requirements .............15 Figure 7: Urban Extent developments in the epochs 2000 to 2005, 2005 to 2010 and 2010 to 2015

in Denpasar and surrounding region.....................................................................................18 Figure 8: Increase of Urban Extent between 2000 and 2015, in percent per administrative units.......18 Figure 9: Increase of Urban Extent between 2000 and 2005, in percent per administrative units.......19 Figure 10: Increase of Urban Extent between 2005 and 2010, in percent per administrative units.......19 Figure 11: Increase of Urban Extent between 2010 and 2015, in percent per administrative units.......20 Figure 12: Detailed Land Cover/ Land Use 2016 - Spatial distribution ................................................21 Figure 13: Detailed Land Cover/ Land Use 2006 structure: presented as Overall, Core City and in

Larger Urban Zone in % (left) and km2 (right) ....................................................................22 Figure 14: Detailed Land Cover/ Land Use 2016 structure: presented as Overall, Core City and in

Larger Urban Zone in % (left) and km2 (right) ....................................................................22 Figure 15: Spatial Distribution of Land Cover and Land Use Change Types from 2006 to 2016 ........24 Figure 16: Land Cover Land Use Change Types 2006-2016: presented as Overall, Core City and

in Larger Urban Zone in % (left) and km2 (right) ................................................................25 Figure 17: Spatial distribution of changes from Agricultural Areas to other classes between 2006

and 2016 ...............................................................................................................................26 Figure 18: Changes of Agricultural Areas into other LU classes between 2006 and 2016; presented

as Overall, Core City and in Larger Urban Zone in % (left) and km2 (right) .......................27 Figure 19: Transport Network of Denpasar in 2006 and 2016 ..............................................................27 Figure 20: Subset of Flood Hazard Map of Denpasar (Downtown area) (Image: Quickbird

2017/01/03) ..........................................................................................................................31 Figure 21: Subset of Flood Risk Map of Denpasar (Downtown area) (Image: Quickbird

2017/01/03) ..........................................................................................................................32 Figure 22: Proportion of Flood Hazard Zones in Denpasar urban (left) and peri-urban region (right)

..............................................................................................................................................32 Figure 23: Proportion of Flood Risk Zones in Denpasar urban (left) and peri-urban region (right) .....33 Figure 24: Proportion of Residential and Public Urban Fabric in flood prone areas

(high/medium/low/no hazard) – in Denpasar urban (left) and peri-urban region (right) .....33 Figure 25: Proportion of Residential and Public Urban Fabric in Flood Risk areas (very

high/high/medium/no risk) – in Denpasar urban (left) and peri-urban region (right) ..........34 Figure 26: Map of Residential and Public Urban Fabric combined with Flood Risk Zoning in the

centre of Denpasar (Image: Quickbird 20170103) ...............................................................34 Figure 27: Proportion of high-ranking infrastructure in flood prone areas (high/medium/low/no

hazard) – in Denpasar urban (left) and peri-urban region (right) .........................................35 Figure 28: Proportion of high-ranking infrastructure in Flood Risk areas (high/medium/no risk) –

in Denpasar urban (left) and peri-urban region (right) .........................................................35 Figure 29: Proportion of construction sites in flood prone areas (low/no hazard) – in Denpasar Core

City .......................................................................................................................................35 Figure 30: Proportion of commercial and industrial units in flood prone areas

(high/medium/low/no hazard) – in Denpasar Core City ......................................................36 Figure 31: Proportion of agricultural land in flood prone areas (high/medium/low/no hazard) – in

Denpasar urban (left) and peri-urban region (right) .............................................................36



EO4SD-Urban Denpasar City Operations Report Page VIII

Figure 32: Proportion of natural and semi-natural areas in flood prone areas (high/medium/low/no

hazard) – in Denpasar urban (left) and peri-urban region (right) .........................................36

List of Tables

Table 1: LU/LC Nomenclature used for Denpasar for the years 2006 and 2016 ................................ 5 Table 2: Number of sampling points for the EO4SD-Urban mapping classes after applied

sampling design ................................................................................................................... 10 Table 3: Validation result of the complemented transport network in Denpasar, which is based

on OSM data ........................................................................................................................ 14 Table 4: Overall LU/LC Statistics 2006 and 2016 ............................................................................ 23 Table 5: Statistics of change categories ............................................................................................. 25 Table 6: Statistics of changes of Agricultural Areas ......................................................................... 26 Table 7: Land use classes and reclassification to pre-defined damage levels ................................... 29 Table 8: Flood hazard and risk classification (Flood Risk Matrix) ................................................... 30

List of Abbreviations

AoI Area of Interest

AWEInsh Automated Water Extraction Index

CS Client States

DEM Digital Elevation Model

DLR German Space Agency

DOS Dark Object Subtraction

EEA European Environmental Agency

EGIS Consulting Company for Environmental Impact Assessment and Urban Planning, France

ENVISAT Environmental Satellite, ESA's successor to ERS

EO Earth Observation

EO4SD Earth Observation for Sustainable Development

ERS European Remote Sensing satellites by ESA

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

GPS Global Positioning System

GPS Global Positioning Systems

GUF Global Urban Footprint

HHWL Highest High Water Level

HR High Resolution

HRL High Resolution Layer

IFI International Financing Institute

InSAR Interferometric SAR

INSPIRE Infrastructure for Spatial Information in the European Community

ISGNSS International Symposium on Global Navigation Satellite System (GNSS)

ISO/TC 211 Standardization of Digital Geographic Information

JR JOANNEUM Research, Austria

LC / LU Land Cover/ Land Use



EO4SD-Urban Denpasar City Operations Report Page IX

LULUC Land Use and Land Use Change

MMU Minimum Mapping Unit

MSL Mean Sea Level

NDVI Normalized Difference Vegetation Index

NEO Geospatial Service Provider, The Netherlands

NNE North-Northeast

OSM OpenStreetMap

QA Quality Assurance

QC Quality Control

QM Quality Management

RTRW Rencana Tata Ruang Wilyah

SAR Synthetic Aperture Radar

SBAS Small Baseline Subset Method

SP Service Provider

SRTM Shuttle Radar Topography Mission

SSW South-Southwest

TEP Thematic Exploitation Platform

ToC Table of Contents

VHR Very High Resolution

WB World Bank

WBG World Bank Group



EO4SD-Urban Denpasar City Operations Report Page 1

1 Introduction 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 demonstrate the benefits of Earth Observation (EO)

in their operations. 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 and Multi-Lateral Development Banks (MDBs) 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 project with a focus on S. Asia, SE Asia, Latin America 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 MDBs and counterpart City

Authorities. The report provides a service description in Chapter 2, and then in Chapter 3 systematically

reviews the main production steps involved and importantly highlights whenever there are Quality

Control (QC) mechanisms involved; the related QC forms are provided in the Annexe of this report. The

description of the processes are kept intentionally at a top level and avoiding technical details as the

report is considered mainly for non-technical MDB staff/experts and City Authorities. Finally Chapter

4 presents the standard analytical work undertaken with the products which can provide further inputs

into urban development assessments, modelling and reports.

2 Service Description

The following Section summarises the service as it has been realised for the city of Denpasar in

Indonesia within the EO4SD-Urban Project and as it had been delivered to the User in the month of

August 2017.

2.1 Stakeholders and Requirements

The EO4SD-Urban products will be used as part of the World Bank City Planning Labs (CPL)

programme. CPL is a technical assistance program that is developing the capacity for evidence-based

planning in Indonesian cities. This will be achieved by developing municipal spatial data infrastructures

(MSDI) and an integrated ICT platform, or "geoportal", for data sharing. In addition, CPL is committed

to developing sustainable practices, so the programme will have a capacity building component. The

program will conduct training for municipal government personnel to use geospatial planning tools, and

further develop and maintain the MSDI and geoportal.

Currently, the city of Denpasar has partnered with the CPL technical assistance initiative to develop

Denpasar’s ability to perform geospatial planning and data sharing. Denpasar is interested to use remote

sensing to develop data about urban growth, Land Use/Land Cover changes, transportation

infrastructure and accessibility, population density/distribution, and flooding issues in the city. The




EO4SD-Urban project datasets and analyses will help the city develop planning policies, and inform its

medium-term and long-term plans towards the goal of making Denpasar a more resilient and sustainable

city.

2.2 Service Area Specification

The approach taken to demarcate the service area for map production for Denpasar was to present the

border (digitised) of the Core built up area and a wider extent for the Peri-Urban Area using Google

Earth as a background. These drafted boundaries were then sent to the World Bank team who shared it

with the City Authorities to fine tune the boundaries. The resulting Areas of Interest (AoIs) for both

Core and Peri-Urban agreed on for the service area for production is depicted in Figure 1 below.

Figure 1: Illustration of Core and Peri-Urban Areas for Mapping

The Core region has an area of 132km2 and the Peri-Urban an area of 409 km2 for a total service area of

541 km2.




2.3 Product List and Product Specifications

During the discussions related to the AoIs the planned geo-spatial products were also presented and

reviewed with the WB Team and users. It was noted that the Baseline Land Use 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 Denpasar the full list of products for both the Core and Peri-Urban

Areas are as follows:

Urban Land Use/Land Cover and the change between two points in time

Transport Infrastructure – Road Network

Flood History Assessment

This report will focus on the Baseline products of LU/LC as well as the change between two points in

time; for each EO4SD-Urban city two time slots of 2006 and 2016 were offered to the Users for the

assessment of changes in a 10 year epoch (for the different geo-spatial products). Due to EO data

availability the time epochs varied with plus or minus a year or two around these 2 time slots.

2.3.1 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 to 4 (with Level 4 being the most detailed classes). 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 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-SL > 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-2. However,

for the Core Urban Areas using the Very High Resolution (VHR) data with higher spatial resolution it

was possible to go down to Level 3 and 4. 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 production started.

In order to link the Urban Atlas classes described in the previous Section with the two Indonesian cities

of Semarang and Denpasar, the Consortium used the documentation provided by the cities for their

Master Plans. The urban planning in Indonesia requires cities to develop regional spatial planning or

Rencana Tata Ruang Wilayah (RTRW) for each city. The information provided for Denpasar from the

City Authorities included an English translated version of the RTRW for 2011, and geo-spatial data

provided included the Land Use Map for Denpasar from 2012 as presented in Figure 2.

Figure 2: Land Use Map of Denpasar, 2012 (Reference: World Bank, CPL Programme, 2017)

The main LU classes depicted in the RTRW, were reviewed to assess which classes could be linked to

the LU/LC Nomenclature from the Urban Atlas and therefore observed using the VHR data. See LU/LC

Nomenclature for Denpasar in Table 1.




Table 1: LU/LC Nomenclature used for Denpasar for the years 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 1100 Very

High Density

1000

Artificial Surfaces

1100 Residential 1100 Very

High Density

1120 High

Density

1120 High

Density

1130 Medium

Density

1130 Medium

Density

1140 Low

Density

1140 Low

Density

1150 Super

Low Density

1150 Super

Low 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


Public,

Military and Private Units

1211

Commercial Area

1212

Industrial Area

1213

Business

District

1214

Education

Facilities

1215

Military

1216 Health

Facilities

1217 Public

Buildings

1220 Roads 1221

Arterial Road

1220 Roads 1221

Arterial Road

1222

Toll Line

1222

Toll Line

1223 Collector

Road

1223 Collector

Road

12.3 Railway 123 Railway

124 Port 124 Port

125 Airport 125 Airport

1300 Mine,

Dump and

1310

Mineral

Extraction

1311

Mining

Area

1300 Mine,

Dump and

1310

Mineral

Extraction

1311

Mining

Area




2006 2016

Level I Level II Level III Level IV Level I Level II Level III Level IV

Construction

Sites

and Dump

Sites 1312

Landfill

Construction

Sites

and Dump

Sites 1312

Landfill

1320 Construction

Sites

1320 Construction

Sites

1330 Open

Space

1330 Open

Space

1400 Artificial

Non-Agricultural

Vegetated Areas

1410 Urban

Green

Spaces

1411 Urban

Parks

(Parks, Gardens)

1400 Artificial

Non-

Agricultural Vegetated Areas

1410 Urban

Green

Spaces

1411 Urban

Parks

(Parks, Gardens)

1412

Cemeteries

1412

Cemeteries

1420 Sport Facilities

(Sport

Fields, Sport Stadiums,

Golf

Courses, …)

1420 Sport Facilities

(Sport

Fields, Sport Stadiums,

Golf

Courses, …)

2000

Agricultural Area

2000

Agricultural Area

3000 Natural

and Semi-Natural areas

3100 Forest 3000 Natural

and Semi-Natural Areas

3100 Forest

3200 Fisheries 3200 Fisheries

3300 Grassland

and Shrubs/Natural

Area

3300 Grassland

and Shrubs/Natural

Area

3400 Bare Soil 3400 Bare Soil

3500 Mangrove 3500 Mangrove

3600 Beach 3600 Beach

4000 Peat/

Wetland/

Swamp

4000 Peat/

Wetland/

Swamp

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 and Open Street Map (OSM)

data. This aspect is further discussed in Chapter 3.

See Annex 1 for finalised nomenclature and class definitions for Denpasar.




2.3.2 Transport Infrastructure – Road Networks

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.

2.4 Terms of Access

The dissemination of the digital data and the report was undertaken via FTP.




3 Service Operations

This Chapter presents all the main steps of the service operations including the necessary input data, the

processing methods, the accuracy assessment and Quality Control procedures. Methods are presented

in a top-level and standardised manner for all the EO4SD-Urban City Reports.

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 3.

A summary of the main EO data used overall in the project is described herewith.

High Resolution Optical EO Data

The major data sources for the Peri-Urban current and historic mapping of urban LULC, urban extent

and imperviousness were Sentinel-2 and Landsat data which were accessible and downloadable free of

charge.

Sentinel-2: the most recent data coverage comprises of 2 Sentinel-2 from 07 June 2016. Level-

1A data were downloaded.

Landsat 7: As a source of historical data two scenes of Landsat TM 7 from 09 December 2000

and 21 May 2002 has been acquired to get a cloud free image of the whole area of interest.

All images have been used as baseline data in the production process of the Peri-Urban LU/LC. The

ortho-rectification was achieved by an image to image registration using the Global Landcover-Ortho

image product. The NASA accuracy specification for the Global Landcover-Ortho product was 50m

RMSE, everywhere in the world.

Very High Resolution Optical EO Data

For Denpasar the VHR data used were WorldView-2 and Quickbird-2. 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 Space and European Space Imaging.

It has to be noted that under the current collaboration project the VHR EO data had to be purchased

under single-user license agreements between GAF AG and the EO Data Providers. If EO data would

have to be distributed to other stakeholders then further licence up-grades for multiple users would have

to be requested from the EO Data Providers which may result in additional costs.

The following VHR sensor data have been acquired:

WorldView-2: the most recent data coverage for the Core Urban Area comprises of 2 scenes,

both acquired on 08 September 2015.

Quickbird-2: for the historic data 2 scenes from 28 May 2008 and one scene 10 June 2006

were used.

Detailed lists of the used EO data as well as their quality is documented in the attached Quality

Control (QC) sheets in Annex 3.

Ancillary Data

A variety of reference data was provided by the City Authorities and the WB team at the inception of

production. This constituted existing VHR data as well as *.shp files for Land Use features and other

datasets such as Land Use Map 2012, spatial data for hospitals/clinics, schools/universities, parks, and

cemeteries. For the datasets that could be used as reference information to support the current

classification a QC was done and the results are provided in the QC sheets in Annex 3. It should be

noted that these datasets had limited metadata and no QC sheets provided and therefore there were

constraints to use them in an integrated manner during the production process in the current project.




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 2.

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 Olofson et al (2013) 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

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∗𝑞

(𝐸

𝑧)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 = 0.89, 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 2) that did not received samples in the first run, additionally 215

samples were randomly drawn across all these classes.

Table 2: Number of sampling points for the EO4SD-Urban mapping classes after applied sampling design

Class Name Class

ID

No. of

Sampling

Points

Km²

Coverage of

the Class

Residential 1100 167 202.55


Public, Education

1210 44 18.59

Roads 1220 1 5.83

Port 1240 1 0.39

Airport 1250 29 2.34

Mining Area, Landfill 1310 31 0.27

Construction Site 1320 26 0.27

Open Space 1330 5 0.88

Urban Parks, Cemeteries 1410 2 0.81

Sport Facilities 1420 3 0.66

Agriculture 2000 227 170.67

Forest 3100 135 56.56

Fisheries 3200 7 0.60

Grassland 3300 60 5.41

Bare Soil 3400 5 0.31

Mangroves 3500 9 10.98

Beach 3600 4 1.39

Wetlands 4000 63 0.76

Water 5000 201 62.23

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

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 3: Mapping result of the city of Denpasar of the year 2016 overlaid with randomly distributed sample

points 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

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:

�̂�𝑖𝑗 = (1

𝑀) ∑

1

𝜋𝑢ℎ∗

𝑥∈(𝑖,𝑗)

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 = √𝑝ℎ(1−𝑝ℎ)

𝑛ℎ 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 = √∑ 𝑤ℎ2. 𝜎ℎ

2

In which 𝑤ℎ is the proportion of the total area covered by each stratum. The 95% Confidence Interval

(CI) is +/- 1.96*𝜎.

The confusion matrices are provided Quality Control documentation the Annex 3 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 Denpasar has an overall mapping accuracy of 89.11% with

a Confidence Interval ranging from 87.14% to 91.07% at a 95% CI. The specific class accuracies

are given in Annex 3.

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 Lines 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 4 with the road network in grey, the grid

cells in black and the randomly selcted cells in green.




Figure 4: 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 5). 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 5: 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 280 sampling points were created and their differences recorded. The result is presented in Table

3 as histogram of deviations. For the entire sampling population a Mean Difference of 0.67 m and a

Standard Deviation of 1.14 m was calculated.

Table 3: Validation result of the complemented transport network in Denpasar, which is based on OSM

data

Distance in m Frequency

0.0 183

0-1 13

1-2 50

2-3 17

3-4 11

4-5 5

5-6 0

6-7 1

Above 7.5 meters 0

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.

3.5 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 below).




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 3.

3.6 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:

Figure 6: Quality Control process for EO4SD-Urban product generation; at each intermediate processing

step output properties are compared against pre-defined requirements

http://inspire-geoportal.ec.europa.eu/editor/

http://inspire-geoportal.ec.europa.eu/editor/




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.




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, the

Transport Infrastructure product and the Flood Risk 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 2015

Land Cover Land Use - Status and Trends between 2006 and 2016

Transport Infrastructure – Development 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. Furthermore, it has to be

acknowledged that this Chapter presents only a summary and overview of what is possible with the geo-

spatial datasets provided for Denpasar in the current EO4SD Urban project. Further geospatial analyses

with the presented products are beyond the scope of this project and could be part of future

collaborations.

4.1 Urban Extent – Developments from 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 Denpasar was first used to assess historical

developments from 2000-2015 and then overlaid with Ward boundaries 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 7 which shows the urban development in the Core and Peri-Urban Areas as well as surrounding

regions of Denpasar. 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 urban extent developments between 2000 and 2015 depicted in Figure 7, one can note that

urban extent development occurred in large patches specifically in the epochs between 2000 and 2005

(red patches); in the Core Urban Area it occurred notably in the west and northern borders and in the

Peri-Urban in the west. The urban extent between 2010 and 2015 then seemed to cluster around the

growth spots of 2005. And the growth in 2015 (blue patches) was notable in the southern Peninsula-

Bukit. A more quantitative depiction of the urban extent in percent is presented in Figure 8 which




illustrates the extent of built up area per administrative units between 2000 and 2015 and shows that the

main areas of growth occurred in the very north and south regions of Denpasar.

Figure 7: Urban Extent developments in the epochs 2000 to 2005, 2005 to 2010 and 2010 to 2015 in Denpasar

and surrounding region

Figure 8: Increase of Urban Extent between 2000 and 2015, in percent per administrative units




For the different 5 year epochs the analysis undertaken used the urban extent data overlaid with the

administrative units for the Core Urban Area of Denpasar and calculated the percent (%) change in

different classes with the low change categories ranging from 0-2% to 10% and the high change classes

being 40-69% or over 60%. Thus the Figure 9 illustrates that the urban extent started intensifying in the

northern and western parts in the epoch from 2000 to 2005 and a bit in the south. The urban extent

continued to be high in the northern part of the Core area in the next epoch from 2005 and 2010 (Figure

10).






Finally between 2010 and 2015 the urban extent showed the highest change again in the northern areas

but also increased in the south-west and south (see Figure 11).





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 12 and a cartographic version with map layout

design is provided as geo-pdf files with the digital data.

Figure 12: Detailed Land Cover/ Land Use 2016 - Spatial distribution

For the epoch 2006 the main/dominant LU/LC classes occurring in the overall Core and Peri-Urban

Regions (combined) were Agricultural Area (38.97% of total area), Residential (32.03% of total area)

and forest (8.03% of total area). By 2016 these classes remained the main/dominant LU/LC classes with

slight increases in overall area of the Residential class (36.68%) and the Forest are (10.44% of total

area) coupled with a decrease of the Agricultural Area (31.52% of total area). This information on the

classes are dis-aggregated and presented as percentages and area coverages in Figure 13 and Figure 14

for the epochs 2006 and 2016 respectively. The Table 3 also presents the detailed information on area

and percentage of total area for each class for 2006 and 2016 as well as the changes.




Figure 13: Detailed Land Cover/ Land Use 2006 structure: presented as Overall, Core City and in Larger

Urban Zone in % (left) and km2 (right)

Figure 14: Detailed Land Cover/ Land Use 2016 structure: presented as Overall, Core City and in Larger


In the Core Urban Area in both 2006 and 2016 as depicted in Figure 13 and Figure 14 the specific

Residential class of high (50-80%) and very high (80-100%) density and the Industrial, Commercial,

Public class also have significant percentage cover. The next Section will highlight the LU/LC change

information between the two epochs in more detail.

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 4 to get an overview.




Table 4: Overall LU/LC Statistics 2006 and 2016

LU/LC Classes

2015 2005 Change Change per Year

sqkm % of total

sqkm % of total

sqkm % sqkm %

Residential 0-10 3.83 0.71% 2.52 0.46% 1.31 0.24% 0.13 0.02%

Residential 10-30 11.99 2.21% 15.68 2.90% -3.69 -0.68% -0.37 -0.07%

Residential 30-50 57.10 10.54% 61.11 11.28% -4.00 -0.74% -0.40 -0.07%

Residential 50-80 99.99 18.46% 73.53 13.58% 26.46 4.89% 2.65 0.49%

Residential 80-100 29.64 5.47% 23.11 4.27% 6.53 1.21% 0.65 0.12%

Industrial, Commercial, Public, Military 18.59 3.43% 14.39 2.66% 4.20 0.78% 0.42 0.08%

Arterial Line 2.43 0.45% 2.08 0.38% 0.35 0.06% 0.03 0.01%

Collector Line 3.41 0.63% 3.40 0.63% 0.01 0.00% 0.00 0.00%

Port 0.39 0.07% 0.35 0.07% 0.04 0.01% 0.00 0.00%

Airport 2.34 0.43% 2.34 0.43% 0.00 0.00% 0.00 0.00%

Mining Area 0.27 0.05% 0.27 0.05% 0.00 0.00% 0.00 0.00%

Construction Site 0.27 0.05% 0.18 0.03% 0.09 0.02% 0.01 0.00%

Open Space 0.88 0.16% 0.48 0.09% 0.40 0.07% 0.04 0.01%

Urban Park 0.71 0.13% 0.66 0.12% 0.05 0.01% 0.00 0.00%

Cemeteries 0.10 0.02% 0.10 0.02% 0.00 0.00% 0.00 0.00%

Sport Facilities 0.66 0.12% 0.52 0.10% 0.15 0.03% 0.01 0.00%

Agricultural Area 170.67 31.52% 211.02 38.97% -40.36 -7.45% -4.04 -0.75%

Forest 56.56 10.44% 43.46 8.03% 13.10 2.42% 1.31 0.24%

Fisheries 0.60 0.11% 0.79 0.15% -0.18 -0.03% -0.02 0.00%

Grassland and Shrubs 5.41 1.00% 9.07 1.67% -3.65 -0.67% -0.37 -0.07%

Bare Soil 0.31 0.06% 0.87 0.16% -0.56 -0.10% -0.06 -0.01%

Mangroves 10.98 2.03% 10.92 2.02% 0.06 0.01% 0.01 0.00%

Beach 1.39 0.26% 1.37 0.25% 0.01 0.00% 0.00 0.00%

Peat, Wetlands, Swamps 0.76 0.14% 0.62 0.11% 0.14 0.03% 0.01 0.00%

Water 62.23 11.49% 62.67 11.57% -0.44 -0.08% -0.04 -0.01%

Total 541.52 100.00% 541.52 100.00%

The area statistics of the LU/LC classes show a lot of change dynamics within the residential density

classes. For example it can be noted that there are losses of area in the lower density classes (10-30%,

30-50%) and increases of area in the very low (0-10%) and higher (50-80%, 80-100%) density

residential classes as well in the class Industrial, Commercial, Public, and Military. Furthermore there

is an overall decrease in the agricultural areas and an increase in forested area which can be attributed

to development of plantations.

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 higher Residential

Density class

Urban Residential Extension: all changes from non-urban Residential classes to Residential




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

The overlay analysis of these aggregated categories of the epochs 2006 and 2016 is depicted in Figure

15.

Figure 15: Spatial Distribution of Land Cover and Land Use Change Types from 2006 to 2016

The spatial distribution of the change types as depicted in Figure 15 shows that the densification of the

Residential areas in the 10 year period, mainly happened in the peripheral areas of the urban Core area

(western and southern regions) and in the surrounding Peri-Urban Area in the west and south with a bit

in the north. There is notable Urban Residential Extension in the Core Urban Area in all regions except

the very central part; the in the Peri-Urban Area the Urban Residential Extension is notable in the south

Peninsula and in the west. The statistics of the Change categories are presented in Figure 16 and Table

5. The statistics in Figure 16 provide a quantitative aspects to the LU change classes; for example it’s

interesting to note that Urban Densification was dominant (49%) in the Peri-Urban zone whereas in the

Core City the class of Residential Extension was the dominant LU change class (47%).




Figure 16: Land Cover Land Use Change Types 2006-2016: presented as Overall, Core City and in Larger


The class of Urban Densification, i.e. changes from lower to higher Residential Density has the highest

amount of overall change in the Overall with 48.22 %. The amount of densification was slightly higher

in the Peri-Urban zone with 48.86% compared to the densification in the Core City with 46.89%. The

second largest block of changes can be observed with the classes Residential Extension as well as urban

Land Use into non-urban LU/LC classes. These changes occurred mainly in the Core Urban Area. The

quantitative data is given in Table 5.

Table 5: Statistics of change categories

Change Classes Change Overall Change Core City Change Larger Urban Zone

km² % km² % km² %

Urban Densification 45.30 48.22% 14.41 46.89% 30.89 48.86%

Urban Residential Extension 27.96 29.76% 9.58 31.17% 18.38 29.07%

Other Urban Land Use Extension 4.54 4.83% 4.17 13.57% 0.37 0.59%

Change within Natural and Semi-Natural Areas 16.15 17.19% 2.57 8.37% 13.58 21.48%

Total 93.96 100.00% 30.73 100.00% 63.23 100.00%

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, a 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 are displayed in Figure 17.




Figure 17: Spatial distribution of changes from Agricultural Areas to other classes between 2006 and 2016

The main overall changes is in the class of Agricultural areas converted to Residential (60%), and

Agricultural areas converted to Forest or Plantation developments. This pattern of land sue change is

reflects in both the Core and Peri-Urban Areas for Denpasar, with the conversion of Agriculture to

residential being the dominant LU change in the Core area (74%). In the Peri-Urban Area the conversion

of Agricultural areas to Forest (Plantation) is quite dominant as well (44%). The converted Agriculture

to Residential occurs all over the Core Urban Area with a concentration in the south-east area. In the

Peri-Urban zone the Agriculture to Residential conversion occurs mainly in the west, and the southern

Peninsula (Bikut) and to a lesser extent in the north. The main areas where Agriculture was converted

to Forest in the Peri-Urban zone is in south Peninsula.

The statistics in Figure 18 and Table 6 show that the majority of Agricultural area has been converted

into Residential area and Plantations in the ten year period between 2006 and 2016.

Table 6: Statistics of changes of Agricultural Areas

Change Classes Change Overall Change Core City Change Larger Urban Zone

km² % km² % km² %

Agricultural Area to Residential 23.18 60.04% 6.65 74.18% 16.53 55.76%

Agricultural Area to Industrial, Commercial, Public

or Military Area 1.62 4.19% 1.61 18.00% 0.00 0.01%

Agricultural Area to Forest 13.81 35.77% 0.70 7.83% 13.11 44.23%

Total 38.61 100.00% 8.97 100.00% 29.65 100.00%




Figure 18: Changes of Agricultural Areas into other LU classes between 2006 and 2016; presented as

Overall, Core City and in Larger Urban Zone in % (left) and km2 (right)

The next Section summarises the noted changes in the transport network for Denpasar.

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 for the Arterial roads and a buffer of 8m for the collector roads. Local roads

are only part of the vector data set, which are provided to the user.

Figure 19 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 main changes of the Transport

Network occurred in terms of densification of Local roads in 2016 in the peripheral regions of the Core

Urban Area such as in the north where urban densification has occurred.

Figure 19: Transport Network of Denpasar in 2006 and 2016




4.4 Flood Risk Assessment

Flood Hazard and Risk Mapping is a vital component of appropriate land use planning in flood-prone

areas. Flood Hazard and Risk Maps are designed to increase awareness of the likelihood of flooding

among the public, local authorities and other organisations, as well as support national disaster

management programmes.

Specific flood regimes and underlying causes for the flooding events in the area of interest have to be

analysed carefully, as these can be very different in different cities. For the Denpasar area generally

short-term local floods and river floods after heavy rainstorms are typical. In Bali, intense precipitation

events happen often between November and March. Areas that are prone to flooding are in most cases

along the rivers: Mati River, Badung River, Sidakarya, and smaller waterways (Gunawan et al. 2015).

According to local sources, flooding also occurs due to the poor quality of drainage and because of the

waste produced by housing and commercial activities not systematically disposed in waste dumps.

Terrain subsidence is to be expected in the future due to excessive groundwater extraction, and load

from buildings and structures but is not confirmed up to now (Abidin et al. 2012)

4.4.1 EO Data Used

In general, there are two types of data available for this purpose: optical and radar data. Available HR

optical data from the sensors Landsat 5, Landsat 7, Landsat 8, ASTER, and Sentinel-2 covering the

period from 1997 to 2017 was downloaded and analysed with regard to regional and/or local flooding.

VHR Imagery was provided by the Project Coordinators, covering the urban area:

Mosaic of Quickbird 20050911 and Quickbird 20060516

Quickbird Image 20170103

These datasets give a good impression of the rapid development and expansion of Denpasar. However,

there was no visual evidence for urban flooding. Additionally, numerous recent VHR images are

available in Google Earth. These images were also visually checked for indications of flooding, but no

flood events could be identified as well.

Radar data from the current European Space Agency (ESA) Sentinel-1 data is acquired constantly and

it is free of charge. The spatial resolution of this data is approximately 10 m. In order to go further back

in time (2002 – 2012), ERS/ENVISAT data could be used. Currently, these datasets are commercially

available, but for the future, there are concrete actions of ESA to make this data available also free of

charge during 2018. Sentinel-1 radar data covering the period from 2015-03-24 to 2017-04-06 (39

datasets) was downloaded and processed with regard to water extent. Flooded areas in the coastal zone

can be extracted but the data tend to give unreliable results in urban environments due to the high number

of low backscattering objects. In Peri-urban regions no relevant flood events were registered by this

data.

The flood risk product is a combination of flood hazard based on Land Use information derived from

the Very High Resolution (VHR) satellite data and thus these datasets are also indirectly used for this

product.

4.4.2 Short Description of Methodological Approach

Historic flood extent mapping

Flood extent mapping based on EO data heavily depends on available datasets as well as on the types of

floods. Whereas there is a good chance to identify large-scale river and coastal floods, normally no

information regarding short-term local floods (flash-floods) can be obtained from EO data due to the

temporal acquisition windows of the satellite systems and/or cloud cover. In some cases short-term local

floods can be recorded and localized based on reports (e.g. in social media) and press releases but such

inventory never will meet the claim to be complete.




The relevant optical datasets were corrected atmospherically applying the Dark Object Subtraction

(DOS) approach. For defining the water extent in coastal areas the water cover was classified by

applying the Automated Water Extraction Index AWEInsh (Feyisa et al. 2014) which makes use of the

reflectance values of Green, Near Infrared and Shortwave Infrared spectral bands of the Landsat 5,

Landsat 7, Landsat 8, ASTER and Sentinel-2 sensors. The AWEInsh is an index formulated to effectively

eliminate non-water pixels, including dark built surfaces in areas with urban background.

Regarding the occurrence of short-term local floods (flash-floods), point data of reported and localized

urban floods (2009 – 2017) were buffered with 200 m to roughly estimate potential flooding hotspots.

Additionally, the critical sections of the river and waterway network in the lower parts of the catchment

areas (sections where the general slope reduces to less than 3 degrees) were identified. Waterways were

taken from OSM Layer and complemented by visual interpretation of VHR data. After classification in

two classes based on Stream Order, the lines were buffered with 50 m (Stream Order 1 and 2, as defined

by Strahler 1952) and 100 m respectively (Stream Order 3 and 4) to roughly estimate potential flooding

zones.

Furthermore, areas classified as “Flood Zones” were provided by the City of Denpasar. Although there

is no documentation available and no verification possible, these areas were taken into consideration in

the analysis as well.

Flood hazard mapping

The flood hazard map is taking into account all available data sources (EO based water extents, reports

and press releases, results of modelling of potential flooding along rivers and waterways) aiming at

covering all types of floods and thus all areas which are potentially endangered by flooding. The

classification in three qualitative hazard levels is expert-based under consideration of observed and

reported frequencies of floods.

Flood risk mapping

Risk is defined as a combination of probability and consequences. A detailed and uniform land-use map

is an important prerequisite to perform flood risk calculations, since it determines what is damaged in

case of flooding. The original land-use map as provided by EO4SD Urban project was recoded to pre-

defined categories (as given in Table 7) to ensure consistent results.

The exposition is classified following an approach developed by NEO (based on: Dasgupta et al. 2015)

integrating economic costs, social damage, physical damage and flood duration. Four land use damage

levels (A, B, C, D) are defined based on this estimation.

Table 7: Land use classes and reclassification to pre-defined damage levels

Classes Damage Total Level

Economic

Costs

0-2

Social

Damage

0-2

Physical

Damage

0-2

Flood

Duration

0-2

Agricultural Land 1.5 0.5 0 1 3 B

Commercial and Industrial Units 2 0.5 1 0.5 4 B

Dump site 0 1.5 0.5 0 2 A

Construction Sites 1 0.5 0 0 1.5 A

Forests 0.5 0 0 0 0.5 A

Formal high density residential -

Continuous urban fabric

(Sealing level: 50%-100%)

1.5 1.5 2 1.5 6.5 D

Formal low density residential -

Discontinuous urban fabric

(Sealing level: 10%-50%)

1.5 1 2 1 5.5 C

Land Without Current Use, Bare Soil 0 0 0 0 0 A

Mineral Extraction site 1 0 0.5 1 2.5 B




Classes Damage Total Level

Economic

Costs

0-2

Social

Damage

0-2

Physical

Damage

0-2

Flood

Duration

0-2

Non-Residential Urban Fabric 1 1 0.5 1 3.5 B

Other Natural and Semi-Natural Areas

including Wetlands 0 0 0 0 0 A

Transportation units 1.5 1 2 1.5 6 C

Sports and leisure facilities, Beach 0.5 0.5 0 0.5 1.5 A

Other Urban / Artificial Area 1.5 1.5 1 0.5 4.5 C

Urban Greenery, Cemeteries 0.5 0.5 0.5 1 2.5 B

Village Settlements

(Sealing level 1-10%) 1.5 1.5 0.5 1.5 5 C

Water Bodies, Fisheries 0 0 0 0 0 A

The Flood Risk matrix (see Table 8) is generated based on the above code and Flood Hazard classified

into three hazard levels. The Flood Risk level is classified in four qualitative classes based on the

combination of flood hazard and land use damage.

Table 8: Flood hazard and risk classification (Flood Risk Matrix)

Damage cost on land use

A B C D

Flood Hazard

1 (low) 1A 1B 1C 1D

2 (medium) 2A 2B 2C 2D

3 (high) 3A 3B 3C 3D

Flood Risk classification

Low Risk 1A 1B 2A

Medium Risk 1C 1D 2B 2C 2D 3A 3B

High Risk 2D 3C

Very high Risk 3D

4.4.3 Product Description and Accuracy Assessment

There are three final layers in this product: (1) the raw data on past flood extents as derived from EO

data and ancillary data (flood history), (2) the Flood Hazard map which summarizes past flood events

and thus gives information about the likelihood of future events, and (3) the Flood Risk map combining

this data with information on urban and Peri-urban land use and its damage potential in case of flooding.

The delineation of coastal water extents is based on data from Landsat 8 (acquired on 15/04/2015,

12/01/2016 provided by the US Geological Survey), and Sentinel-2 (acquired on 21/10/2015,

30/12/2015, 06/08/2016, 25/10/2016 provided by the European Space Agency). The evaluation of the

data confirms the results of the Global Surface Water Explorer Product (Joint Research Centre) based

on Landsat Imagery. According to the current analysis and the latter product, no relevant increase of

permanent and seasonal water cover can be observed for the period from 1984 to 2015.

The accuracy assessment of the water extent classification was done based on stratified random

sampling. The following equation is applied to determine the number of points required for an accepted

standard error of the error of commission per class:

n_c=(p_c (1-p_c))/(σ_c^2 ), c = 1,…,L

n_c number of SSUs for category c

p_c estimated error rate for category c

σ_c accepted standard error of the error of commission for category c

L number of categories




The following values were applied: p_c=0.1, σ_c=0.05, L=2 (Water or No Water)

- => n_c = 36

In Denpasar, the validation extent was limited to the coastal area. The validation was based on the visual

comparison of 72 points (randomly distributed and situated in the coastal lowlands only) on each satellite

image and the respective classification result since no independent reference data were available.

Overall accuracies for the classification range from 90.3% to 97.2%.

The Flood Hazard map (subset see Figure 20) displays flood hazard information (delivered in vector

format). This data is based on the occurrence of floods of the past 10 years. It takes into account both

the hazard from tidal floods in coastal areas as well as from short-term local floods after heavy

rainstorms in urban areas. The classification in three qualitative hazard levels is expert-based under

consideration of reported frequencies of floods.

Since no independent reference data are available no accuracy assessment is possible. The plausibility

of the results nevertheless was evaluated on basis of local reports and press releases.

The Flood Risk Map (subset see Figure 21) displays flood risk information (delivered in vector format).

This data is based on the occurrence of floods of the past 10 years combined with information of the

land use map provided by EO4SD Urban project. It takes into account both the hazard level and potential

damages, based on different land uses. The damages are assessed on 4 aspects: economic, social,

physical and flood duration.

Since this product is a direct derivation and combination of the Hazard Classification and the Land Use

Classification, its plausibility can be rated high when the mentioned input datasets are rated as being

plausible and reliable.

Figure 20: Subset of Flood Hazard Map of Denpasar (Downtown area) (Image: Quickbird 2017/01/03)




Figure 21: Subset of Flood Risk Map of Denpasar (Downtown area) (Image: Quickbird 2017/01/03)

4.4.4 Results

With regard to the total calculated Flood Hazard zones it can be observed that the proportion of the total

affected area is significantly higher in Denpasar Core City than in the Peri-urban region: more than 16%

vs. approx. 9%. The proportion of medium and high hazard zones is significantly higher as well (cf.

Figure 22).

Figure 22: Proportion of Flood Hazard Zones in Denpasar urban (left) and peri-urban region (right)




As the extension of Flood Risk zones is equal to that of Hazard Zones this results in a similar picture.

The medium, high and very high risk categories occur disproportionately high in the Core City

(approximately 11% vs. 3% in the Peri-urban region, cf. Figure 23).

Figure 23: Proportion of Flood Risk Zones in Denpasar urban (left) and Peri-urban region (right)

Selected land use categories and groups of land use categories were analysed more detailed both in the

urban and in the Peri-urban area regarding their position in hazardous zones:

Residential and public urban fabric: public buildings, business district, education facilities,

health facilities, residential area

Transportation units: airport, arterial lines, collector lines

Construction sites

Commercial and industrial units

Agricultural land

Semi-natural and natural area: forests, grasslands and shrubs

The analysis of Residential and Public Urban Fabric shows that more than 6% of such Land Use is

situated in medium and high hazard zones in the urban region (Figure 24). Including the low hazard

areas in the analysis, a total of 16% of Residential and Public Urban Fabric is located in the calculated

hazard zones.

In the peri-urban region, the proportion of Residential and Public Urban Fabric located in the calculated

hazard zones is significantly lower with a total of 7.7%, most of it situated in low hazard zones.

Figure 24: Proportion of Residential and Public Urban Fabric in flood prone areas (high/medium/low/no

hazard) – in Denpasar urban (left) and Peri-urban region (right)

Focusing on the risk aspect, the applied methodology implies that no low Flood Risk zones can exist in

the Residential and Public Urban Fabric class (cf. Section 4.2.2 and Table 7 and 8). Figure 25 displays

the proportions of Residential and Public Urban Fabric in the medium, high, and very high Flood Risk




zones. As expected, the proportions of the high and very high Flood Risk zones are significantly higher

in the Core City.

Figure 25: Proportion of Residential and Public Urban Fabric in Flood Risk areas (very

high/high/medium/no risk) – in Denpasar urban (left) and Peri-urban region (right)

Figure 26: Map of Residential and Public Urban Fabric combined with Flood Risk Zoning in the centre of

Denpasar (Image: Quickbird 20170103)

The analysis of high-ranking infrastructure (airport, arterial lines, collector lines) reveals that the western

part of the airport is situated in a Peri-urban area classified as flood prone (low hazard). Moreover, many

main transport lines in the Peri-urban area are situated in coastal zones classified as flood prone. This

results in significantly higher percentages of high-ranking infrastructure in flood prone areas (most of it

classified as low hazard zones) for the Peri-urban region (approximately 32% versus 12% for the Core

City, cf. Figure 27).

Regarding the Flood Risk classification, most of the involved transportation units can be found in the

medium risk class (cf. Figure 27).




Figure 27: Proportion of high-ranking infrastructure in flood prone areas (high/medium/low/no hazard) –

in Denpasar urban (left) and Peri-urban region (right)

Figure 28: Proportion of high-ranking infrastructure in Flood Risk areas (high/medium/no risk) – in

Denpasar urban (left) and Peri-urban region (right)

The results of the analysis of the land use categories Construction Sites, Commercial and Industrial

Units, Agricultural Land, and Semi-Natural and Natural Area are displayed in Figure 29 to Figure 32.

Figure 29: Proportion of Construction Sites in flood prone areas (low/no hazard) – in Denpasar Core City




Figure 30: Proportion of Commercial and Industrial units in flood prone areas (high/medium/low/no

hazard) – in Denpasar Core City

Figure 31: Proportion of Agricultural Land in flood prone areas (high/medium/low/no hazard) – in

Denpasar urban (left) and Peri-urban region (right)

Figure 32: Proportion of Natural and Semi-Natural Areas in flood prone areas (high/medium/low/no

hazard) – in Denpasar urban (left) and Peri-urban region (right)




5 References

Abidin, H.Z., Andreas, H., Gumilar, I., Sidiq, T.P. & Fukuda, Y. (2012): On the Roles of Geospatial

Information for Risk Assessment of Land Subsidence in Urban Areas of Indonesia.-The 8th

International Conference on Geo‐information for Disaster Management, 13‐14 December, 2012,

Enschede, The Netherlands.

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.).

Dasgupta, S., Asif, Z., Subhendu, R., Mainul, H., Sarwar, J. & Ainun, N. (2015): Urban Flooding of

Greater Dhaka in a Changing Climate: Building Local Resilience to Disaster Risk. Directions in

Development. Washington, DC: World Bank. doi:10.1596/978-1-4648-0710-7. License: Creative

Commons Attribution CC BY 3.0 IGO

European Union (2011). Mapping Guide for a European Urban Atlas, Version 11.0

European Union (2013). INSPIRE Metadata Implementing Rules: Technical Guidelines

Feyisa, L.G., Meilby, H., Fensholt, R. & Proud, S.R. (2014): Automated Water Extraction Index: A new

technique for surface water mapping using Landsat imagery.- Remote Sens. Environ, 140 (2014), pp.

23-35.

Gunawan, I., Sagala, S., Amin, S., Zawani, H. & Mangunsong, R. (2015): City Risk Diagnostic for

urban Resilience.-World Bank.

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

RTRW (2011). RTRW Denpasar City Year 2011-2031, Denpasar, Indonesia (English translated

version)

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.

Strahler, A. N. (1952): "Dynamic basis of geomorphology". Geological Society of America Bulletin 63:

pp. 923–938.



Annex 1 to EO4SD-Urban Denpasar City Operations Report Page 1

Annex 1 – Land Use and Land Cover Nomenclature and Definitions




Nomenclature according to User Requirements, including definitions from a Remote Sensing perspective.

Level and

Land Use Class Sub-Classes Definition

1.1 Residential

- Very High Density

- High Density

- Medium Density

- Low Density

- Super Low

Density

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.

Super 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.

1.2.1. Industrial,

Commercial,

Public, Military

and Private Units

- Commercial Area

- Industrial Area

- Education

Facilities

- Business District

(Government,

Offices)

- Military

- Public Building

Factories, warehouses, CBD, shopping malls, markets, other commercial

facilities, Schools and University and associated land, Government, Offices.

military and associated land, “Big” public buildings like churches, bibliotheca,

…

(Distinction depends strongly on availability of ancillary data)

1.2.2 Roads - Arterial Road

- Toll Line

- Collector Road

Transportation units such as roads with a minimum width of 4 m (Collector

Road) and 6 m (Arterial Road).

1.2.3 Railway Railway facilities including stations, cargo stations and associated land.

1.2.4 Port Infrastructure of port areas, including quays, dockyards, transport and storage

areas and associated area.

1.2.5 Airport Administrative area of airports, mostly fenced.

Included are all airport installations: runways, buildings and associated land.

1.3.1 Mineral

extraction and

dump sites

- Mining Area

- Landfill

Open pit extraction sites; public, industrial or mine dump sites.

1.3.2 Construction

Sites

Spaces under construction or development. Clear evidence of actual

construction needs to be identifiable in the data.

1.3.3 Open Space Areas in the vicinity of artificial surfaces still waiting to be used or re-used. No

actual agriculture or recreational use.

1.4.1 Urban Green

Space

- Urban Parks

- Parks

- Public Gardens

- Cemeteries

Urban Parks, Parks, Public Gardens:

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.

1.4.2 Sport

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.

2. 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.

3.1 Forest High woody vegetation in natural forests; transitional woodland; low

vegetation cover with bushes and shrubs.

3.2 Fisheries Fishery Areas

3.3 Grassland and

Shrubs

Natural area where there is low to medium high vegetation




3.4 Bare Soil Bares Soil Areas (not within the urban area)

3.5 Mangroves Areas where the Mangrove Ecosystem occurs

3.6 Beach Beaches including small commercial facilities like restaurants.

4. 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.

5. Water Visible water areas like lakes, rivers, ponds (natural, artificial).




Annex 2 – Processing Methods for EO4SD-Urban Products




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.





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.






In order to generate the Flood Risk product, several processing steps are needed, which are shown in the

graph below. These processing steps generally lead to three sub-products: (1) Historic flood extent

products; (2) Flood hazard product and (3) Flood risk product.

Historic flood extent mapping

Flood extent mapping based on EO data heavily depends on available data sets. This depends on cloud-

free weather conditions as well as on types of floods in focus. Whereas there is a good chance to identify

large-scale river and coastal floods, normally no information regarding short-term local floods (flash-

floods) can be obtained from EO data due to short duration. In some cases short-term local floods can

be recorded and localized based on reports (e.g. in social media) and press releases but such inventory

never will meet the claim to be complete.

The relevant EO optical datasets were atmospherically corrected. For defining the extent of tidal floods

in coastal areas the water cover was classified by applying a water index such as for example the

Normalized Difference Water Index (NDWI) or the Automated Water Extraction Index AWEInsh

(Feyisa et al. 2014) which makes use of the reflectance values of Green, Near Infrared and Shortwave

Infrared spectral bands of the Landsat 5, Landsat 7, Landsat 8, ASTER and Sentinel-2 sensors. The

index is used to effectively eliminate non-water pixels, including dark built surfaces in areas with urban

background.

Additionally, to overcome the drawback of missing significant EO data in the urban area, the usability

of the Hazard Component of the “Tidal Floods Risk Online Map” (Nugraha et al. 2015,

http://geodesi.undip.ac.id/gis/) was verified. The plausibility verification was based on available reports

and press releases and resulted in the decision to include this data in the hazard assessment. Regarding

the occurrence of short-term local floods (flash-floods) point data of reported and localized urban floods

(2009 – 2017) were buffered with 200 m to roughly estimate potential flooding hotspots. Additionally

the critical sections of the river and waterway network in the lower parts of the catchment areas (sections

where the general slope reduces to less than 3 degrees) were identified. Waterways were taken from

OSM Layer and complemented by hydrologic modelling of potential catchment areas and flow routes

based on the combined ALOS-SRTM DEM and visual interpretation of VHR data. After classification

in two classes based on Stream Order the lines were buffered with 50 m (Stream Order 1 and 2, as

defined by Strahler 1952) and 100 m respectively (Stream Order 3 and 4) to roughly estimate potential

flooding zones. If available, local maps and data from the cities were also included.

Flood hazard mapping

The flood hazard map is taking into account all available data sources (EO based flood extents, reports

and press releases, results of modelling of potential flooding along rivers and waterways, ancillary data

– online tidal flooding map) aiming at covering all types of floods and thus all areas which are potentially

Review of occurring flood regimes

Review existing

flood data (EO and

non EO)

Localization of non-

EO data

Processing of EO data

Pre-processing

Flood classification

Flood

history

products

Flood

hazard

Flood

Risk

Land Use

Map

Subsidenc

e map (if

existing)

Existing

maps




endangered by flooding. Where available, a ten years period of EO data was used. The classification in

three qualitative hazard levels is expert-based under consideration of observed and reported frequencies

of floods. In principle, higher frequency means higher hazard. In case of occurrence and availability,

results of recent InSAR subsidence calculation were also taken into account, as for example done in the

city of Semarang.

Flood risk mapping

Risk is defined as a combination of probability and consequences. Probability follows from the flood

hazard product. Consequences were determined by damage assessment based on the land-use map. A

detailed and uniform land-use map is therefore an important prerequisite to perform flood risk

calculations, since it determines what is damaged in case of flooding. The original land-use map was

recoded to pre-defined categories to ensure consistent results.

The exposition is classified following an approach developed by NEO (based on: Dasgupta et al. 2015)

integrating economic costs, social damage, physical damage and flood duration. Four damage levels (A,

B, C, D) are defined based on this estimation. Flood risk follows from the combination of flood hazard

level and damage level.




Annex 3 – Filled Quality Control Sheets

Urban Land Use / Land Cover


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

Client: WB Editor: AB; FE

Product: Urban Land Use/ Land Cover Date: 30.08.2017

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

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:







Page 2

Comments / Characteristics:







Page 3

0.1 Requirements

Product 1 (28) Urban Land Use/ Land Cover Status 2006 and 2016

Abstract

The Land Use/Land Cover (LU/LC) product contains spatial explicit information on the different

occurring land use and land cover in both the Core and Peri-Urban areas of the City of Denpasar for the

years 2006 and 2016. The Core area has detailed LU/LC nomenclature whereas the Peri-Urban area

LU/LC nomenclature is at an aggregated Level. The input data for the Core area was the Very High

Resolution data (WorldView-2 for 2017, Quickbird-2 for 2005, 2006 ) and the input data for the Peri-

Urban area was Sentinel (2017) and Landsat-7 (2000, 2002) data. 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: Indonesia A) Wall-to-wall: yes

City: Denpasar Selected Sites: Core-Urban and Peri-Urban

Area km² Core Urban: 132 B) Sampling based: n.a.

Peri-Urban: 409

Time Period - Update Frequency

A) Baseline Year(s): B) Update Frequency

2006 and 2016 2 points in time, no future update

Comments: None

Geographic Reference System

EPSG: 32749. WGS 84 / UTM zone 50S

Mapping Classes and Definitions

Residential

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


dedicated to forestry or agriculture.

Super Low: Average degree of soil sealing: <10 % Residential buildings, roads and other


dedicated to forestry or agriculture. Example: exclusive residential areas with large gardens.

Commercial Area* Warehouses, Central Business Districts (CBD), shopping malls, markets, other commercial

facilities.

Industrial Area* Factories and associated land

Business District* Government, Offices.







Page 4

Education Facilities* Schools and Universities and associated land (including sport fields)

Military* Military and associated land

Health Facilities* Hospitals and associated land

Public Buildings* “Big” public buildings like churches, bibliotheca, …

Arterial Line Highways, connecting the city with other cities

Toll Road Public or private roadway for which a fee is assessed for passage.

Collector Line Bigger Connecting roads within the city

Railway Railway facilities including stations, cargo stations and associated land.

Port Port and associated area.

Airport

Administrative area of airports, mostly fenced. Included are all airport installations: runways,

buildings and associated land.

Mining Area

Open pit extraction sites (sand, quarries) including water surface, if <Minimum Mapping Unit

(MMU), open-cast mines, inland salinas, oil and gas fields;

Landfill Dump sites and associated land

Construction Site

Spaces under construction or development, soil or bedrock excavations for construction

purposes or other earthworks visible in the image.

Open Space

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.

Cemeteries Cemeteries and associated area

Sport 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.

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

High woody vegetation in natural forests; transitional woodland; low thickets. Includes

Plantations

Fisheries Fishery Areas

Grassland and Shrubs Natural area where there is little vegetation

Bare Soil Bare Soil Areas (not within the urban area)

Mangroves Areas where the Mangrove Ecosystem occurs

Beach Beaches including small commercial facilities like restaurants.







Page 5

Peat_Wetlands_Swamps

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 only to be mapped 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 replace by other Earth Observation

(EO) data.

Spatial Resolution

n.a. (Product provided as Shapefile)

Minimum Mapping Unit (MMU)

Minimum Mapping Unit within urban core area is 0.25 ha. Peri-urban area is mapped with a MMU of 0.5 ha

Data Type & Format

Shapefile *.shp and GeoPDF

Bit Depth

n.a.

Class Coding

Class Code Class Name RGB Code [R; G; B]

1100 Residential 255; 0; 0 (main class only)

1211 Commercial Area 197; 0; 255

1212 Industrial Area 132; 0; 168

1213 Business District 0; 255; 197

1214 Education Facilities 232; 190; 255

1215 Military 115; 115; 0

1216 Health Facilities 115; 223; 255

1217 Public Buildings 192; 252; 234

1221 Arterial Line 78; 78; 78

1222 Toll Road 100; 100; 100

1223 Collector Line 120; 120; 120

1230 Railway 52; 52; 52

1240 Port 0; 168; 132

1250 Airport 168; 0; 132

1311 Mining Area 115; 76; 0

1312 Landfill 190; 232; 255

1320 Construction site 255; 115; 223

1330 Open space 242; 242; 242

1411 Urban Parks (Urban Green Spaces) 85; 255; 0

1412 Cemeteries 230; 230; 0

1420 Sport Facilities 255; 170; 0

2000 Agricultural Area 255; 235; 175

3100 Forest 38; 115; 0

3200 Fisheries 158; 170; 215

3300 Grassland and Shrubs 180; 215; 158

3400 Bare Soil 204; 204; 204

3500 Mangroves 185; 255; 150

3600 Beach 255; 255; 0

4000 Peat_Wetlands_Swamps 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%







Page 6

Positional Accuracy

RMSE < 30 m.

Delivery Procedure

Service Provision

Online via FTP

Delivery Date

End of August 2017



© 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

1.1 List of EO Data

Sensoren (8) Sentinel-2, Landsat 7, WV-2, QB-

2 Incoming

Date

Acquisit

ion

Date

Proc.

Level

Path /

Row

AOI - City

/ Region

/ Country

Spatial Res. No. of

Bands

Cloud

Cover

(Data

Provid

er)

Projectio

n /

Spheroid (16)

Data

Forma

t (3)

Bit

Depth (5)

Head

er /

Meta

data (2)

File Name [e.g yymmdd; tbd...]

Sentinel-2

1. S2A_OPER_MSI_L1C_TL_SGS__20160607T

091621_A005005_T50LKR

23.03.20

17

07.06.2

016 1C

50/L

KR

Denpasar

, IND 10m 4 1.1%

WGS 84

/ UTM

zone

50S

.tiff 16

Bit .xml

2. S2A_OPER_MSI_L1C_TL_SGS__20160607T

091621_A005005_T50LLR_N02.02

23.03.20

17

07.06.2

016 1C

50/LL

R

Denpasar

, IND 10m 4 1.2%

WGS 84

/ UTM

zone

50S

.tiff 16

Bit .xml

Landsat-7

3. LE07_L1TP_116066_20001209_2017020

8_01_T1

08.02.20

17

09.12.2

000 L1TP

116/

066

Denpasar

, IND

MUL: 30m;

PAN: 15m 5 28%

WGS 84

/ UTM

zone 50

.tiff 8 Bit .txt

4. LE07_L1TP_116066_20020521_2017013

0_01_T1

08.02.20

17

21.05.2

002 L1TP

116/

066

Denpasar

, IND

MUL: 30m;

PAN: 15m 5 4%

WGS 84

/ UTM

zone 50

.tiff 8 Bit .txt

WorldView-2

6.

17FEB21023512-M2AS_R1C1-

056358854060_01_P001

31.03.20

17

21.02.2

017 LV2A R1C1

Denpasar

, IND

MUL: 2m;

PAN: 0.5m 4

7.0e-

03% UTM 50S .tiff

16

Bit .xml

17FEB21023512-M2AS_R1C2-

056358854060_01_P001

31.03.20

17

21.02.2

017 LV2A R1C2

Denpasar

, IND

MUL: 2m;

PAN: 0.5m 4

7.0e-

03% UTM 50S .tiff

16

Bit .xml

17FEB21023512-M2AS_R2C1-

056358854060_01_P001

31.03.20

17

21.02.2

017 LV2A R2C1

Denpasar

, IND

MUL: 2m;

PAN: 0.5m 4

7.0e-

03% UTM 50S .tiff

16

Bit .xml

17FEB21023512-M2AS_R2C2-

056358854060_01_P001

31.03.20

17

21.02.2

017 LV2A R2C2

Denpasar

, IND

MUL: 2m;

PAN: 0.5m 4

7.0e-

03% UTM 50S .tiff

16

Bit .xml

17FEB21023512-M2AS_R3C1-

056358854060_01_P001

31.03.20

17

21.02.2

017 LV2A R3C1

Denpasar

, IND

MUL: 2m;

PAN: 0.5m 4

7.0e-

03% UTM 50S .tiff

16

Bit .xml

Quickbird-2

7. 06MAY16030316-M2AS_R1C1-

056358854070_01_P001

31.03.20

17

16.05.2

006 LV2A R1C1

Denpasar

, IND

MUL: 2m;

PAN: 0.6m 4

2.0e-

03% UTM 50S

GeoTI

FF

16

Bit

.imd,

rpb,

.til,

xml






Page 8

06MAY16030316-M2AS_R2C1-

056358854070_01_P001

31.03.20

17

16.05.2

006 LV2A R2C1

Denpasar

, IND

MUL: 2m;

PAN: 0.6m 4

2.0e-

03% UTM 50S

GeoTI

FF

16

Bit

.imd,

rpb,

.til,

xml

06MAY16030316-M2AS_R2C2-

056358854070_01_P001

31.03.20

17

16.05.2

006 LV2A R2C2

Denpasar

, IND

MUL: 2m;

PAN: 0.6m 4

2.0e-

03% UTM 50S

GeoTI

FF

16

Bit

.imd,

rpb,

.til,

xml

8. 05SEP11025755-M2AS-

056358854070_01_P002

31.03.20

17

11.09.2

005 LV2A

Denpasar

, IND

MUL: 2m;

PAN: 0.6m 4 0% UTM 50S

GeoTI

FF

16

Bit

.imd,

rpb,

.til,

xml






Page 9

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:






Page 10

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

elementary_schools.sh

p 23.03.2017 incomplete

2016 /

26.09.2016

Denpasar

(municipality

boundary)

100% Vector *.shp WGS_1984_UTM

_Zone_50S unknown 1 unknown No

junior_high_schools.sh


2016 /

26.09.2016.

Denpasar

(municipality

boundary)



senior_high_schools.sh


2016 /

26.09.2016

Denpasar

(municipality

boundary)



vocational_schools.shp 23.03.2017 incomplete 2016 /

26.09.2016

Denpasar

(municipality

boundary)



Lineage(29): This file contains the location and attribute information of the public elementary schools, junior high schools, senior high schools and vocational schools in Denpasar.

Source (30): Received from World Bank City Planning Lab programme; provided by CPL Team-Seetha Raghupathy on 25.11.2016

basic_education.shp 23.03.2017 incomplete n.a. /

22.07.2015

Denpasar

(municipality

boundary)



education.shp 23.03.2017 incomplete 2017 /

01.07.2015

Denpasar

(municipality

boundary)



Lineage(29): This shapefile includes basic education schools on the territory of Denpasar, which are divided into 2 classes and the other shapefile includes educational facilities on the territory of

Denpasar, which are divided into 5 classes.

Source (30): Received from World Bank City Planning Lab programme: provided by CPL team-Ms. Aurora Dias Lokita on 07.03.2017






Page 11

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

hospitals.shp 23.03.2017 incomplete 2016 /

26.09.2016

Denpasar

(municipality

boundary)



clinics.shp 23.03.2017 Incomplete 2016 /

26.09.2016

Denpasar

(municipality

boundary)



Lineage(29): This files contains the locations and attributes of the community health centers and hospitals in Denpasar.

Source (30): Received from Ms. Aurora Dias Lokita and Ms. Seetha Raghupathy.

poi_sarana_kesehatan.

shp 23.03.2017 not available

2017 /

22.03.2017

Denpasar

(municipality

boundary)

100% Vector *.shp GCS_WGS_1984 unknown 3 unknown No

Lineage(29): This shapefile includes health facilities within Denpasar, which are divided into the following 3 classes: hospitals, clinics and pharmacies.


Dataset 3 (14)

Street

Layers

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

jalan_wgs84UTM50s.sh

p 23.03.2017 Incomplete

2010 /

12.08.2016

Denpasar

(municipality

boundary)

80% of

core_are

a

Vector *.shp WGS_1984_UTM

_Zone_50S unknown n.a. unknown No

Lineage(29): This file contains the road network of Denpasar from the year 2010. There are topological errors in this file where lines from streets may not connect as they do in reality.

Source (30): Data received from Ms. Seetha Raghupathy.

Status &Fungsi Jln Dps

Plan_polyline.shp 23.03.2017 not available

2009 /

19.11.2015

Denpasar

(municipality

boundary)


_Zone_50S unknown n.a. unknown No

Lineage(29): This shapefile includes secondary streets which were identified by the user.







Page 12

Dataset 4 (14)

Commer

cial

areas

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

poi_sarana_inudustri-

perdagangan.shp

23.03.2017 not available 2012 /

22.03.2017

Denpasar

(municipality

boundary)

100% Vector *.shp GCS_WGS_1984 unknown 2

unknown No

Lineage(29): This shapefile includes commercial areas on the territory of Denpasar, which are divided into 2 classes.

Source (30): Received from Ms. Aurora Dias Lokita

poi_sarana_perdagang

an_jasa_Markets.shp 23.03.2017 not available

2012 /

22.03.2017

Denpasar

(municipality

boundary)


unknown No

Lineage(29): This shapefile includes commercial areas on the territory of Denpasar, which are divided into 7 classes.







Page 13

Dataset 5 (14)

Public

Building

s

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

poi_sarana_olahraga.sh

p 23.03.2017 not available

2012 /

19.11.2015

Denpasar

(municipality

boundary)


unknown No

Lineage(29): This shapefile includes sport facilities on the territory of Denpasar.


poi_sarana_perkantora

n_dan_pelayanan_mas

yarakat.shp

23.03.2017 Incomplete 2012 /

22.03.2017

Denpasar

(municipality

boundary)


unknown No

Lineage(29): This shapefile includes public facilities on the territory of Denpasar, which are divided into 7 classes.


poi_sarana_ibadah.shp 23.03.2017 not available 2012 /

22.03.2017

Denpasar

(municipality

boundary)


unknown No

Lineage(29): This shapefile includes religious buildings on the territory of Denpasar, which are divided into 4 classes.


poi_kuburan.shp 23.03.2017 not available n.a. /

22.03.2017

Denpasar

(municipality

boundary)


unknown No

Lineage(29): This shapefile includes cemeteries within Denpasar.

Source (30): Received from Ms. Aurora Dias Lokita.

slums_point.shp 23.03.2017 not available n.a. /

19.11.2015

Denpasar

(municipality

boundary)


unknown No

Lineage(29): This shapefile includes slum areas on the territory of Denpasar.







Page 14

Dataset

6(14)

Land

Use Map

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

PL_DENPASAR_GEO.sh


2012 /

26.09.2016

Denpasar

(municipality

boundary)

100% Vector *.shp WGS_1984_UT

M_Zone_50S unknown 28 unknown No

Lineage(29): This shapefile is a LU map of the territory of Denpasar.


Dataset

6(14)

Census

Data,

Ward

Boundar

ies

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

kelurahan_census.shp 23.03.2017 incomplete 2013 (?) /

n.a.

Denpasar

(municipality

boundary)

100% Vector *.shp WGS_1984_UT

M_Zone_50S unknown / unknown No

Lineage(29): This shapefile includes census data for the years 2004 to 2013 as well as the word boundaries of the territory of Denpasar.







Page 15

2.1 EO Data Quality

Sensoren(8) Sentinel-2, Landsat 7, QB-2, WV-2

Ba

ck

up

Re

ad

ab

ilit

y (1

)

Ch

eck

He

ad

er

/

Me

tad

ata

(2

)

Band Specifications

Pro

jecti

on

/ S

ph

ero

id (

16

)

Sce

ne

Lo

ca

tio

n

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

ve

r (i

nte

rna

l

ch

eck

)

Ra

dio

me

try

To

po

gra

ph

y

Dro

pp

ed

Lin

es /

Art

efa

cts

(7

)

Acce

pta

nce

Sta

tus

Com

ment

s File Name [e.g yymmdd; tbd...]

No

. o

f B

an

ds

Ba

nd

Re

gis

tra

tio

n

Sp

ectr

al/

Sp

ati

al

Re

so

luti

on

Sentinel-2

1. S2A_OPER_MSI_L1C_TL_SGS__20160607T0

91621_A005005_T50LKR 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_SGS__20160607T0

91621_A005005_T50LLR_N02.02 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-7

3. LE07_L1TP_116066_20001209_20170208

_01_T1 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

4. LE07_L1TP_116066_20020521_20170130

_01_T1 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

Worldview-2

6. 17FEB21023512-M2AS-

056358854060_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

Quickbird-2

7. 06MAY16030316-M2AS-


Y -

Q Y - V

Q

Y - V

Q

Y -

V N - V

Q Yes none

8. 05SEP11025755-M2AS-


Y -

Q Y - V

Q

Y - V

Q

Y -

V N - V

Q Yes none






Page 16

2.2 In-situ Data Quality

Dataset 1 (14)

Ba

ck

up

Re

ad

ab

ilit

y (1

)

Ch

eck

He

ad

er

/

Me

tad

ata

(2)

Ext

en

t

Pro

jecti

on

/ S

ph

ero

id

(16

)

Sp

ati

al R

eso

luti

on

Da

ta F

orm

at

(3)

Bit

De

pth

(5)

Lo

ca

tio

n

Co

mp

lete

ne

ss

Ge

om

. M

isa

lign

me

nt

Pla

usib

ilit

y

Dro

pp

ed

Lin

es /

Art

efa

cts

(7)

Acce

pta

nce

Sta

tus

Comments


No in-situ data used






Page 17

2.3 Ancillary Data Quality

Dataset 1 (14) Schools

Re

ad

ab

ilit

y (1

)

Ch

eck

He

ad

er

/M

eta

da

ta (2

)

Da

ta C

ove

rage

Ma

tch

es S

erv

ice

Are

a (

%)

Pro

jecti

on

/ 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

aste

r:

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

on

ly)(

7)

Uti

lity

fo

r C

urr

en

t

Pro

ject


elementary_schools.shp ☒ Yes

☐ No

☐ Complete (INSPIRE/ISO19119)

☒ Incomplete

☐ not available

☒ Yes

☐ No

☐ Unknown If no: 100%

☒ Correct

☐ Incorrect

☐ Unknown EPSG: 32750

n.a. *.shp n.a. ☒ Complete

☐Incomplete

☐ No

☐ Yes

☒ Unknown If yes: XX m

☐ No

☐ Yes

☒ n.a. If yes: xxx %

☐ None

☒ Partial

☐ Full

junior_high_schools.shp ☒ Yes

☐ No


☒ Incomplete

☐ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect



☐Incomplete

☐ No

☐ Yes


☐ No

☐ Yes


☐ None

☒ Partial

☐ Full

senior_high_schools.shp ☒ Yes

☐ No


☒ Incomplete

☐ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect



☐Incomplete

☐ No

☐ Yes


☐ No

☐ Yes


☐ None

☒ Partial

☐ Full

vocational_schools.shp ☒ Yes

☐ No


☒ Incomplete

☐ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect



☐Incomplete

☐ No

☐ Yes


☐ No

☐ Yes


☐ None

☒ Partial

☐ Full

basic_education.shp ☒ Yes

☐ No


☒ Incomplete

☐ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect



☐Incomplete

☐ No

☐ Yes


☐ No

☐ Yes


☐ None

☒ Partial

☐ Full

education.shp ☒ Yes

☐ No


☒ Incomplete

☐ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect



☐Incomplete

☐ No

☐ Yes


☐ No

☐ Yes


☐ None

☒ Partial

☐ Full

Comments: Accuracy of the dataset is unknown. Some points show schools, others not.






Page 18

Dataset 2 (14) Health Facility

Re

ad

ab

ilit

y (1

)

Ch

eck

He

ad

er

/M

eta

da

ta (2

)

Da

ta C

ove

rage

Ma

tch

es S

erv

ice

Are

a (

%)

Pro

jecti

on

/ 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

aste

r:

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

on

ly)(

7)

Uti

lity

fo

r C

urr

en

t

Pro

ject


hospitals.shp ☒ Yes

☐ No


☒ Incomplete

☐ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect



☐Incomplete

☐ No

☐ Yes


☐ No

☐ Yes


☐ None

☐ Partial

☒ Full

clinics.shp ☒ Yes

☐ No


☒ Incomplete

☐ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect



☐Incomplete

☐ No

☐ Yes


☐ No

☐ Yes


☐ None

☐ Partial

☒ Full

poi_sarana_kesehatan.shp ☒ Yes

☐ No


☐ Incomplete

☒ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect



☐Incomplete

☐ No

☐ Yes


☐ No

☐ Yes


☐ None

☒ Partial

☐ Full

Comments: Accuracy of the dataset is unknown. This files contains the locations and attributes of the hospitals, community health centers, clinics and pharmacies in Denpasar.

Dataset 3 (14) Street Layers

Re

ad

ab

ilit

y (1

)

Ch

eck

He

ad

er

/M

eta

da

ta (2

)

Da

ta C

ove

rage

Ma

tch

es S

erv

ice

Are

a (

%)

Pro

jecti

on

/ 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

aste

r:

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

on

ly)(

7)

Uti

lity

fo

r C

urr

en

t

Pro

ject


jalan_wgs84UTM50s.shp ☒ Yes

☐ No


☒ Incomplete

☐ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect



☐Incomplete

☐ No

☐ Yes


☐ No

☐ Yes


☐ None

☒ Partial

☐ Full

Status &Fungsi Jln Dps

Plan_polyline.shp ☒ Yes

☐ No


☐ Incomplete

☒ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect



☐Incomplete

☐ No

☐ Yes


☐ No

☐ Yes


☐ None

☒ Partial

☐ Full

Comments: Accuracy of the dataset is unknown. Streets were identified by the user.






Page 19

Dataset 4 (14) Commercial

Areas

Re

ad

ab

ilit

y (1

)

Ch

eck

He

ad

er

/M

eta

da

ta (

2)

Da

ta C

ove

rage

Ma

tch

es S

erv

ice

Are

a

(%)

Pro

jecti

on

/ 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

aste

r:

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

on

ly)(

7)

Uti

lity

fo

r C

urr

en

t

Pro

ject


poi_sarana_inudustri-

perdagangan.shp ☒ Yes

☐ No


☐ Incomplete

☒ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect



☐Incomplete

☐ No

☐ Yes


☐ No

☐ Yes


☐ None

☒ Partial

☐ Full

poi_sarana_perdagangan_jasa_Mar

kets.shp ☒ Yes

☐ No


☐ Incomplete

☒ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect



☐Incomplete

☐ No

☐ Yes


☐ No

☐ Yes


☐ None

☒ Partial

☐ Full

Comments: Accuracy of the dataset is unknown. This shapefiles includes commercial areas on the territory of Denpasar.






Page 20

Dataset 5 (14) Public

Buildings

Re

ad

ab

ilit

y (1

)

Ch

eck

He

ad

er

/M

eta

da

ta (2

)

Da

ta C

ove

rage

Ma

tch

es S

erv

ice

Are

a

(%)

Pro

jecti

on

/ 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

aste

r:

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

on

ly)(

7)

Uti

lity

fo

r C

urr

en

t

Pro

ject


poi_sarana_olahraga.shp ☒ Yes

☐ No


☐ Incomplete

☒ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect



☐Incomplete

☐ No

☐ Yes


☐ No

☐ Yes


☐ None

☒ Partial

☐ Full

poi_sarana_perkantoran_dan_pelay

anan_masyarakat.shp ☒ Yes

☐ No


☒ Incomplete

☐ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect



☐Incomplete

☐ No

☐ Yes


☐ No

☐ Yes


☐ None

☒ Partial

☐ Full

poi_sarana_ibadah.shp ☒ Yes

☐ No


☐ Incomplete

☒ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect



☐Incomplete

☐ No

☐ Yes


☐ No

☐ Yes


☐ None

☒ Partial

☐ Full

poi_kuburan.shp ☒ Yes

☐ No


☐ Incomplete

☒ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect



☐Incomplete

☐ No

☐ Yes


☐ No

☐ Yes


☐ None

☒ Partial

☐ Full

slums_point.shp ☒ Yes

☐ No


☐ Incomplete

☒ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect



☐Incomplete

☐ No

☐ Yes


☐ No

☐ Yes


☐ None

☒ Partial

☐ Full

Comments: Accuracy of the dataset is unknown. This shapefile includes sport facilities, public facilities, religious buildings, cemeteries and slum areas on the territory of Denpasar.

Dataset 6 (14) Land Use

Map 2012

Re

ad

ab

ilit

y (1

)

Ch

eck

He

ad

er

/M

eta

da

ta (2

)

Da

ta C

ove

rage

Ma

tch

es S

erv

ice

Are

a

(%)

Pro

jecti

on

/ 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

aste

r:

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

on

ly)(

7)

Uti

lity

fo

r C

urr

en

t

Pro

ject


PL_DENPASAR_GEO.shp ☒ Yes

☐ No


☐ Incomplete

☒ Yes

☐ No

☐ Unknown

☒ Correct

☐ Incorrect

☐ Unknown

n.a. *.shp n.a. ☐ Complete

☒Incomplete

☐ No

☐ Yes

☒ Unknown

☐ No

☐ Yes

☒ n.a.

☐ None

☒ Partial

☐ Full






Page 21

☒ not available If no: 100% EPSG: 4326 If yes: XX m If yes: xxx %

Comments: Accuracy of the dataset is unknown. This shapefile is a Land Use Map with 28 classes.

Dataset 7 (14)

Census Data,

Ward

Boundaries

Re

ad

ab

ilit

y (1

)

Ch

eck

He

ad

er

/M

eta

da

ta (

2)

Da

ta C

ove

rage

Ma

tch

es

Se

rvic

e A

rea

(%

)

Pro

jecti

on

/ S

ph

ero

id ,

EP

SG

(16

)

Sp

ati

al R

eso

luti

on

an

d

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:

Att

rib

ute

Ta

ble

; R

aste

r:

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

on

ly)(

7)

Uti

lity

fo

r C

urr

en

t P

roje

ct


kelurahan_census.shp ☒ Yes

☐ No


☐ Incomplete

☒ not available

☒ Yes

☐ No


☒ Correct

☐ Incorrect


n.a. *.shp n.a. ☐ Complete

☒Incomplete

☐ No

☒ Yes

☐ Unknown If yes: XX m

☐ No

☐ Yes


☐ None

☒ Partial

☐ Full

Comments: Accuracy of the dataset is unknown. This shapefile was used for statistical analyses. Overlaps and gaps had to be corrected first, before it could be used for the analyses.






Page 22

3.1 Geometric Correction

Sensoren (8) Sentinel-2, Landsat 7, QB-2,

WV-2

Pro

ce

ssin

g D

ate

AO

I

Cit

y / R

egio

n /

Co

un

try

Pro

jecti

on

/ S

ph

ero

id (

16

)

No

. &

RM

S (

m)

of

GC

Ps (

17

)

No

. &

RM

S (

m)

of

TP

s (

18

)

No

. &

RM

S (

m)

of

CP

s (

19

)

Dig

ita

l E

leva

tio

n M

od

el

(DE

M)

Mo

de

l / A

lgo

rith

m (

13

)

Re

sa

mp

lin

g

Me

tho

d (

20

)

Va

lid

ati

on

Re

po

rts

Acce

pta

nce

Sta

tus

Output File Name [e.g

yymmdd; tbd...] N

o. File Name [e.g yymmdd; tbd...]

Sentinel-2

1. S2A_OPER_MSI_L1C_TL_SGS__201606

07T091621_A005005_T50LKR

13.04.

2017

Denpas

ar IDN

UTM50S

/ WGS84 N/A N/A N/A

S2A_OPER_MSI_L1C_TL

_SGS__20160607T091

621_A005005_T50LLR


07T091621_A005005_T50LLR_N02.02

13.04.

2017

Denpas

ar IDN

UTM50S

/ WGS84 N/A N/A N/A

S2A_OPER_MSI_L1C_TL

_SGS__20160607T091

621_A005005_T50LKR

Landsat-7

3. le07_l1tp_116066_20001209_201702

08_01_t1_b1-8_stack_toa_PSH

13.04.

2017

Denpas

ar IDN

UTM50S

/ WGS84 N/A N/A N/A

le07_l1tp_116066_200

01209_20170208_01_

t1_b1-8_stack_toa_PSH

4. le07_l1tp_116066_20020521_201701

30_01_t1-b1-8_stack_toa_PSH

13.04.

2017

Denpas

ar IDN

UTM50S

/ WGS84 N/A N/A N/A

le07_l1tp_116066_200

20521_20170130_01_

t1-b1-8_stack_toa_PSH

Worldview-2

6. 17FEB21023512-M2AS-

056358854060_01_P001_toa_PSH

13.04.

2017

Denpas

ar IDN

UTM50S

/ WGS84 N/A N/A N/A

SRTM

30

0 order

poynom CC Yes Yes

o17FEB21023512-

M2AS-

056358854060_01_P0

01_toa_PSH

Quickbird-2

7. 05SEP11025755-M2AS-

056358854070_01_P002_toa_PSH

13.04.

2017

Denpas

ar IDN

UTM50S

/ WGS84

GCPs 6;

RMSE

x0,96;

y0,64

TPs 4;

RMSE

x0,47;

y1,17

N/A SRTM

30

Affin, 1st

order

poynom

CC Yes Yes

o05SEP11025755-

M2AS-

056358854070_01_P0

02_toa_PSH

8. 06MAY16030316-M2AS-

056358854070_01_P001_toa_PSH

13.04.

2017

Denpas

ar IDN

UTM50S

/ WGS84

GCPs 6;

RMSE

x0,96;

y0,64

TPs 4;

RMSE

x0,47;

y1,17

N/A SRTM

30

Affin, 1st

order

poynom

CC Yes Yes

o06MAY16030316-

M2AS-

056358854070_01_P0

01_toa_PSH






Page 23

3.1.1 Data Fusion

Dataset 1 (14) Input A Input B Output

Acce

pta

nce

Sta

tus

Comment

s

Filename 17FEB21023512-M2AS-

056358854060_01_P001

17FEB21023513-P2AS-

056358854060_01_P001

17FEB21023512-M2AS-

056358854060_01_P001_toa_PSH

none

Sensor Worldview-2 Worldview-2 Worldview-2

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

Ye

s


Acce

pta

nce

Sta

tus

Comment

s

Filename 06MAY16030316-M2AS-

056358854070_01_P001

06MAY16030316-P2AS-

056358854070_01_P001

06MAY16030316-M2AS-

056358854070_01_P001_toa_PSH

none

Sensor Quickbird2 Quickbird2 Quickbird2







Ye

s


Acce

pta

nce

Sta

tus

Comment

s

Filename 05SEP11025755-M2AS-

056358854070_01_P002

05SEP11025754-P2AS-

056358854070_01_P002

05SEP11025755-M2AS-

056358854070_01_P002_toa_PSH

none

Sensor Quickbird2 Quickbird2 Quickbird2










Page 24



Ye

s


Acce

pta

nce

Sta

tus

Comment

s

Filename LE07_L1TP_116066_20001209_201702

08_01_T1_B8

LE07_L1TP_116066_20001209_201702

08_01_T1_MTL

le07_l1tp_116066_20001209_20170208_01_t1_

b1-8_stack_toa_PSH

none

Sensor Landsat-7 Landsat-7 Landsat-7



(MS/Pan) 30m MS 15m Pan 0,5m MS




Y

e

s


Acce

pta

nce

Sta

tus

Comment

s

Filename LE07_L1TP_116066_20020521_201701

30_01_T1_MTL

LE07_L1TP_116066_20020521_201701

30_01_T1_B8

le07_l1tp_116066_20020521_20170130_01_t1-

b1-8_stack_toa_PSH

none

Sensor Landsat-7 Landsat-7 Landsat-7



(MS/Pan) 30m MS 15m Pan 0,5m MS




Y

e

s






Page 25

3.2 Data Processing

Sensoren (8)

Sentinel-2, Landsat 7, QB-2,

WV-2

Atmospheric

Correction

Radiometric

Processing (15)

Topographic

Normalisation OTHER Calculation:

Acce

pta

nce

Sta

tus

Ba

ck

up

Comment No

. File Name [e.g yymmdd; tbd...]

Processing

Date

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software

/ Method

Sentinel-2


607T091621_A005005_T50LKR 10.04.2017 N N/A N N/A N N/A N N/A Y Y

Data use as

delivered

2.

S2A_OPER_MSI_L1C_TL_SGS__20160

607T091621_A005005_T50LLR_N02.

02

10.04.2017 N N/A N N/A N N/A N N/A Y Y Data used as

delivered

Landsat-7

3. LE07_L1TP_116066_20001209_201

70208_01_T1_MTL 12.04.2017 N N/A N N/A N N/A Y

GAFmap /

TOA Y Y none

4. LE07_L1TP_116066_20020521_201

70130_01_T1_MTL 12.04.2017 N N/A N N/A N N/A Y

GAFmap /

TOA Y Y none

Worldview-2

6. 17FEB21023512-M2AS-

056358854060_01_P001 12.04.2017 N N/A N N/A N N/A Y

GAFmap /

TOA Y Y none

Quickbird-2

7. 06MAY16030316-M2AS-

056358854070_01_P001 12.04.2017 N N/A N N/A N N/A Y

GAFmap /

TOA Y Y none

8. 05SEP11025755-M2AS-

056358854070_01_P002 12.04.2017 N N/A N N/A N N/A Y

GAFmap /

TOA Y Y none






Page 26

4.1 Classification

Sensors (8) Sentinel-2, Landsat 7, WV-2,

QB-2

Processing

Date

Cloud

Masking Thematic Classification Manual Enhancement Mosaicking

Acce

pta

nce

Sta

tus

Ba

ck

up

Comme

nt N

o. File Name [e.g yymmdd; tbd...]

pro

ce

sse

d

Softwa

re /

Metho

d pro

ce

sse

d

Software / Method

pro

ce

sse

d

Software / Method

pro

ce

sse

d

Software /

Method

1.

S2A_OPER_MSI_L1C_TL_SGS__20160607T

091621_A005005_T50LKR

S2A_OPER_MSI_L1C_TL_SGS__20160607T

091621_A005005_T50LLR_N02.02

12.06.2017 N N/A Y eCognition/semi-

automatic approach Y

eCognition and

ArcGIS 10 / Visual

Interpretation

Y

ArcGIS 10 /

Mosaic Dataset

with manual

neatline selection

Y Y

Cloud

free

data.

2.

LE07_L1TP_116066_20001209_20170208

_01_T1_MTL

LE07_L1TP_116066_20020521_20170130

_01_T1_MTL

22.06.2017 N N/A Y

GAFMap and

ArcMap/visual

mapping

Y

GAFMap and

ArcMap/Visual

interpretation N N/A Y Y

Cloud

free

data.

4. 17FEB21023512-M2AS-

056358854060_01_P001

12.06.2017

N N/A Y

GAFMap and

ArcMap/visual

mapping

Y

GAFMap and

ArcMap/Visual

interpretation

Y

ArcGIS 10 /

Mosaic Dataset

with manual

neatline selection

Y Y

Cloud

free

data.

5.

06MAY16030316-M2AS-

056358854070_01_P001

05SEP11025755-M2AS-

056358854070_01_P002

22.06.201

7 N N/A Y

GAFMap and

ArcMap/visual

mapping

Y

GAFMap and

ArcMap/Visual

interpretation

Y

ArcGIS 10 /

Mosaic Dataset

with manual

neatline selection

Y Y

Cloud

free

data.


Development – Urban

Project QA/QC Sheets developed by GAF AG





Page 27

4.2 Intermediate Quality Control of LCLU Data

No.

x = ok (checked/performed)

o = not performed

n/a = not applicable

Co

mp

lia

nt

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)

11 No data overlaps? (Clean Overlaps only if

there are over ~200 gaps) x

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”

x

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






Page 28

5.1 Thematic Accuracy

No. Product

Name

Processing

Date Area Coverage

No

. o

f C

lasse

s

Re

fere

nce

Da

ta

Sa

mp

le S

ize

Sa

mp

lin

g U

nit

(2

1)

Sa

mp

lin

g

De

sig

n (2

2)

Sa

mp

le E

xclu

sio

n

Cri

teri

a (

23

)

Th

em

ati

c A

ccu

racy

(24

)

Acce

pta

nce

Sta

tus

Comment

1.

Land

Cover

Map

2015

18.05.2017

Entire AOI

(Urban Core and

Peri-Urban

areas)

17 VHR imagery

of 2017

215 per strata

(overall 1019

points were

sampled with

a minimum

distance of

150m).

points Stratified Random Sampling None 89,11 % Yes






Page 29

5.2 Error Matrices

Class Name

(columns = Ground

Truth; Rows =

Mapped Class) Re

sid

en

tia

l

Co

mm

erc

ial

Ro

ad

s

Po

rt

Air

po

rt

Min

ing

Co

nstr

ucti

on

Op

en

Sp

ace

Gre

en

Are

as

Sp

ort

Agri

cu

ltu

re

Fo

rest

Fis

he

rie

s

Gra

ssla

nd

/S

h

rub

s

Ba

re S

oil

Ma

ngro

ve

s

Be

ach

We

tla

nd

s

Wa

ter

User Accuracy

and Confidence

Interval at 95%

Confidence

Level

Clas

s ID 11

12

1 122 124 125 131 132 133 141 142 2 31 32 33 34 35 36 4 5 Totals

Residential 11 158 10 0 0 0 0 0 1 0 1 1 1 0 0 0 0 0 0 0 172 91.86% +4.4% Commercial 121 2 32 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 36 88.89% +11.7% Roads 122 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 100.00% n.a. Port 124 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 100.00% n.a. Airport 125 1 0 0 0 29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 30 96.67% +8.1% Mining 131 0 0 0 0 0 30 0 0 0 0 0 0 0 0 0 0 0 0 0 30 100.00% n.a. Constructio

n 132 0 1 0 0 0 1 26 2 0 0 0 0 0 0 0 0 0 0 0

30 86.67% +13.8%

Open Space 133 0 0 0 0 0 0 0 2 1 0 0 0 0 0 0 0 0 0 0 3 66.67% +70% Green Areas 141 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 100.00% n.a. Sport 142 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 100.00% n.a. Agriculture 2 4 0 0 0 0 0 0 0 0 0 197 12 0 2 0 0 0 0 0 215 91.63% +3.9% Forest 31 2 1 0 0 0 0 0 0 0 0 15 113 0 4 0 0 0 1 0 136 83.09% +6.7% Fisheries 32 0 0 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 3 100.00% n.a. Grassland 33 0 0 0 0 0 0 0 0 0 0 6 6 0 47 0 1 0 0 0 60 78.33% +11.3% Bare Soil 34 0 0 0 0 0 0 0 0 0 0 0 0 0 1 4 0 0 0 0 5 80.00% +45.1% Mangroves 35 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6 0 1 0 7 85.71% +33.1% Beach 36 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 0 0 4 100.00% n.a. Wetlands 4 0 0 0 0 0 0 0 0 0 0 8 1 2 6 0 0 0 52 0 69 75.36% +10.9% Water 5 0 0 0 0 0 0 0 0 0 0 0 1 2 0 0 2 0 9 201 215 93.49% +3.5%

Totals 167 44 1 1 29 31 26 5 2 3 227 135 7 60 4 9 4 63 201 1019

Producer Accuracy

and Confidence

Interval at 95%

Confidence Level 94

.61

%

+3.7

%

72

.73

%

+14

.3%

10

0.0

0%

n

.a.

10

0.0

0%

n.a

.

10

0.0

0%

n.a

.

96

.77

%

+7.8

%

10

0.0

0%

n.a

.

40

.00

%

+52

.9%

50

.00

%

+94

.3%

33

.33

%

+70

.0%

86

.78

%

+4.6

%

83

.70

%

+6.6

%

42

.86

%

+43

.8%

78

.33

%

+11

.3%

10

0.0

0%

n.a

.

66

.67

%

+36

.4%

10

0.0

0%

n.a

.

82

.54

%

+10

.2%

10

0.0

0%

n.a

. Overall Accuracy: 89.11%

Confidence Interval

87.14% - 91.07%






Page 30

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

541 km²

409 km² 100% 4 4 100% Yes none

2. VHR EO Data 132 km² 100% 3 3 100% Yes none

3. In-situ Data None None None None 100% Na Na

4. Ancillary Data Na Na Na Na 100% Incomplete

Ancillary data was

evaluated and

partially used,

Metadata often

incomplete.

PRODUCTS

No

. Product (28)

AOI Coverage

[km²]

Product

Coverage

[km²]

Completeness of

Coverage (25)

Completeness of

Classification (26)

Unclassifiable

Area [%]

Completeness of


1. Land Use 2006

541 km²

541 km² 100% 100% 0% 100% Yes none

2. Land Use 2016 541 km² 100% 100% 0% 100% Yes none







Page 31

6.2 Compliancy

Product 1 (28) EO4SD_Denpasar_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 Denpasar. 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 data of WorldView-2 for 2017, Quickbird-2 for 2005, 2006 and the input data for the Peri-Urban

area was Sentinel (2017) and Landsat-7 (2002, 2002). The LU/LC product is the Baseline Product from

which various derived products (such as Green Areas and Informal Settlements) are produced.


Area Coverage

Requirements Achieved Specifications Compliancy Comments

Country 1: Indonesia Country 1: Indonesia 100%

Country 2: N/A Country 2: N/A



A) Wall-to-wall: A) Wall-to-wall: 100%

Sub-national, Denpasar Sub-national, Denpasar

Test area of approx. 26,000 km², defined by the national user.

B) Sampling based: B) Sampling based:

N/A N/A



Baseline Year(s): Baseline Year(s):

Y

EO data used differs

from specified

Baseline Years,

within accepted

range

2005 (+/- 1 year)

2015 (+/- 1 year)

2005/2006

2017

B) Update Frequency B) Update Frequency

Na Na



EPSG: 32749. WGS 84 / UTM zone

50S EPSG: 32749. WGS 84 / UTM zone 50S Yes --

Mapping Classes and Definitions (Definitions see REDD+ MRV Design Document)

Requirements Achieved Specifications Compliancy

30 classes (see 0.1 Requirements sheet) All classes were mapped according to their

definition. Y

Cloud and Shadow Detection and Removal


Cloud free Yes Y --







Page 32

Spatial Resolution


Na -- -- --



0.25 ha and 0.5 ha MMU 0.25 ha and 0.5 ha Y

Data Type


Vector data yes Y --

GeoPDF None were set na --

Bit Depth


None Na Na --

Data Format


*.shp Shapefile Shapefile Y --

(open cross-platform format)

Class Coding


Na Na --

Metadata


INSPIRE compliant INSPIRE compliant and attached to

product Y


Thematic Accuracy


Overall Accuracy: >80% 89.11 % Y

Positional Accuracy


30m 15m Y

Delivery Procedure







Page 33

Service Provision


online via FTP Uploaded to FTP Yes online via FTP

Delivery Date


August 2017 August 2017 Yes







Page 34

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.







Page 35

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


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







Page 36

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.”







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: JR Editor: H. Proske

Client: ADB Date: 14.09.2017

Product: Flood Risk Assessment

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 No No

1.3 List of Ancillary Data Yes Yes

Data Quality Checks

2.1 EO Data Quality Yes Yes

2.2 In-situ Data Quality No No

2.3 Ancillary Data Quality Yes Yes

Pre-Processing of EO Data

3.1 Geometric Correction Yes Yes

3.1.1 Data Fusion Yes No

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:







Page 2

Comments / Characteristics:

In-situ data for thematic accuracy assessment is not available







Page 3

0.1 Requirements

Product 1 (28) Water Extent Denpasar (Water_Extent_RS_Denpasar)

Abstract

For the Denpasar area short-term flooding close to rivers and waterways in the rainy season (November

– March) is typical. Though Denpasar is prone to flooding according to available local sources, this

cannot be confirmed by the analysis of EO data.

No relevant increase of permanent and seasonal water cover can be observed for the period from 1984

to 2015 according to the Global Surface Water Explorer Product (Joint Research Centre), based on

Landsat Imagery.

For demonstration purposes, water extents of recent dates as derived from HR EO data are included in

the present Geodatabase.


Area Coverage

Country: Indonesia

City: Denpasar Wall-to-wall

Area km² Core Urban: 141,51 km²

Peri-Urban: 540,94 km²



2015 N/A

Comments: Update Frequency depends on flooding events


UTM 50S


Water Extent of recent dates


Data selection is aiming at low cloud coverage in flood exposed areas

Relevant clouds manually digitized and replaced using additional EO data

Spatial Resolution

Vector files

Input raster resolution: 10 m (S2), 30 m (LS8)


1.000 m²

Data Type & Format

Vector, shapefile format

Bit Depth

N/A

Class Coding

Class Code Class Name RGB Code

0 No Water Cover

1 Water Cover

Metadata

ISO compliant








Page 4

Thematic Accuracy

Overall Accuracy: not specified

Positional Accuracy

RMSE < 30 m

Delivery Procedure

Service Provision

Online via FTP

Delivery Date

End of May 2017

Product 2 (28) Flood History River Floods (Flood History Rivers_Den)

Abstract




To overcome this shortcoming, point data of reported and localized urban floods (2008 – 2017) are

included in this product.


Area Coverage

Country: Indonesia

City: Denpasar Core urban

Area km² Core Urban: 141,51 km²²



not specified depends on frequency of flooding events

Comments: data analysis of short-term river floods (hours to days) depends on availability of datasets to be

analysed


UTM 50S


Localization and date of flooding from river floods (point data)


N/A

Spatial Resolution

Vector files


N/A

Data Type & Format


Bit Depth

N/A

Class Coding







Page 5


1 Localization of flood reports

Metadata

ISO compliant


Thematic Accuracy


Positional Accuracy

Not specified – depends on available input data

Delivery Procedure

Service Provision

Online via FTP

Delivery Date

End of May 2017







Page 6

Product 3 (28) Flood Hazard Short-term River Floods

(Flood_Hazard_Denpasar.gdb)

Abstract


– March) is typical.

The present Geodatabase includes two shp-files for rough estimation of potential flooding zones:

a. Denpasar_Flood_Hazard_Rivers

Waterways are taken from OSM Layer and complemented by hydrologic modelling of potential

catchment areas and flow routes based on SRTM and visual interpretation of VHR data. After

classification in two classes based on Stream Order the lines were buffered with 50 m and 100 m

respectively to roughly estimate potential flooding zones.

b. Denpasar_Flood_Reports_buff200

Point data of reported and localized urban floods (2008 – 2017) are buffered with 200 m to roughly

estimate potential flooding hot-spots


Area Coverage

Country: Indonesia

City: Denpasar Wall-to-wall







analysed


UTM 50S


potential flooding zones close to rivers and waterways

potential flooding hot-spots


N/A

Spatial Resolution

Point Shape file


N/A

Data Type & Format


Bit Depth

N/A

Class Coding


1 potential flooding zones close to rivers and waterways

1 potential flooding hot-spots

Metadata

ISO compliant







Page 7


Thematic Accuracy


Positional Accuracy


Delivery Procedure

Service Provision

Online via FTP

Delivery Date

End of May 2017







Page 8

Product 4 (28) Flood Hazard Denpasar Combined

(Flood_Hazard_Denpasar_Combined.shp)

Abstract






Landsat Imagery.

As land subsidence is expected in Denpasar due to observed decrease in groundwater level coastal

lowlands are classified as flood prone at a distance of 200 m. This does not include Tsunami flooding.


and press releases, results of modelling of potential flooding along rivers and waterways, ancillary data)

aiming at covering all types of floods and thus all areas which are potentially endangered by flooding.

The classification in three qualitative hazard levels is expert-based under consideration of observed and



Area Coverage

Country: Indonesia

City: Denpasar







analysed


UTM 50S


1 low hazard

2 medium hazard

3 high hazard


N/A

Spatial Resolution

Vector file


N/A

Data Type & Format


Bit Depth

N/A

Class Coding

Class

Code Class Name RGB Code

1 low hazard 255-255-115







Page 9

2 medium hazard 255-211-127

3 high hazard 255-127-127

Metadata

ISO compliant


Thematic Accuracy


Positional Accuracy


Delivery Procedure

Service Provision

Online via FTP

Delivery Date

September 2017







Page 10

Product 5 (28) Flood Risk Classification Denpasar

(Flood_Risk_Classification_Denpasar.shp)

Abstract



case of flooding. The land-use map was provided by GAF AG and recoded to pre-defined categories to

ensure consistent results.

The exposition is classified integrating economic costs, social damage, physical damage and flood

duration. Four land use damage levels (A, B, C, D) are defined based on this estimation.

The Flood Risk matrix is generated based on these results code and on flood hazard classified into three

hazard levels. The flood risk level is classified in four qualitative classes based on the combination of

flood hazard and land use damage.


Area Coverage

Country: Indonesia

City: Denpasar

Area km² Core

Urban: 141,51 km²

Peri-

Urban: 540,94 km²



not specified depends on frequency of flooding events and on availability of land-use classification

(5 to 7 years)


UTM 50S


1 low risk

2 medium risk

3 high risk

4 very high risk


N/A

Spatial Resolution

Vector file


0.25 ha for the urban area, 0.5 ha for the peri-urban area.

Data Type & Format


Bit Depth

N/A

Class Coding

Class

Code Class Name RGB Code

1 low risk 255-255-190

2 medium risk 255-211-127







Page 11

3 high risk 255-170-0

4 very high risk 255-0-0

Metadata

Not specified


Thematic Accuracy


Positional Accuracy


Delivery Procedure

Service Provision

Online via FTP

Delivery Date

September 2017






Page 12

1.1 List of EO Data

Sensor 1 (8)

Landsat 5

Incoming

Date

Acquisition

Date

Proc.

Level

Path /

Row

AOI - City /

Region /

Country

Spatia

l Res.

No. of

Band

s

Cloud

Cover

(Data

Provider

)

Projectio

n /

Spheroid (16)

Data

Forma

t (3)

Bit

Depth (5)

Header /

Metadata (2) File Name

1. LT51160661997151DKI00 20170509 19970531 L1TP 116/66 Denpasar 30,

30 7 3

UTM50S/

WGS84 *.tif 8 Bit u *.mtl


30 7 42

UTM50S/



30 7 68

UTM50S/


Lineage: USGS Earth Explorer Collection 1 Level-1/L4-5 TM C1 Level 1

Sensor 2 (8)

Landsat 8

Incoming

Date

Acquisition

Date

Proc.

Level

Path /

Row

AOI - City /

Region /

Country

Spatia

l Res.

No. of

Band

s

Cloud

Cover

(Data

Provider

)

Projectio

n /

Spheroid (16)

Data

Forma

t (3)

Bit

Depth (5)

Header /


1. LC81160662013115LGN0

2 20170504 20130425 L1TP 116/66 Denpasar

30,

30 11 6.97

UTM50S/

WGS84 *.tif

12 Bit

u *.mtl

2. LC81160662013307LGN0

1 20170504 20131103 L1TP 116/66 Denpasar

30,

30 11 14.59

UTM50S/

WGS84 *.tif

12 Bit

u *.mtl

3. LC81160662014006LGN0

1 20170504 20140106 L1TP 116/66 Denpasar

30,

30 11 47.52

UTM50S/

WGS84 *.tif

12 Bit

u *.mtl

4 LC81160662015105LGN0

1 20170504 20150415 L1TP 116/66 Denpasar

30,

30 11 1.91

UTM50S/

WGS84 *.tif

12 Bit

u *.mtl

5. LC81160662016012LGN0

2 20170504 20160112 L1TP 116/66 Denpasar

30,

30 11 6.32

UTM50S/

WGS84 *.tif

12 Bit

u *.mtl

Lineage: USGS Earth Explorer Collection 1 Level-1/L8 OLI/TIRS C1 Level 1

Sensor 3 (8)

Sentinel 2

Incoming

Date

Acquisition

Date

Proc.

Level

Path /

Row

AOI - City /

Region /

Country

Spatia

l Res.

No. of

Band

s

Cloud

Cover

(Data

Provider

)

Projectio

n /

Spheroid (16)

Data

Forma

t (3)

Bit

Depth (5)

Header /


1.

S2A_OPER_MSI_L1C_TL_EP

A__20161213T095422_A0

01716_T50LKR_N02.04

20170512 20151021 L1C 3/n/a Denpasar 10,

10 13

13.281

7

UTM50S/

WGS84 *jp2

12 Bit

u *.safe






Page 13

2.

S2A_OPER_MSI_L1C_TL_S

GS__20151230T091952_A

002717_T50LKR_N02.01

20170512 20151230 L1C 3/n/a Denpasar 10,

10 13 5

UTM50S/

WGS84 *jp2

12 Bit

u *.safe

3.

S2A_OPER_MSI_L1C_TL_S

GS__20160806T092156_A

005863_T50LKR_N02.04

20170512 20160806 L1C 3/n/a Denpasar 10,

10 13 5.9869

UTM50S/

WGS84 *jp2

12 Bit

u *.safe

4.

S2A_OPER_PRD_MSIL1C_P

DMC_20161025T230615_

R003_V20161025T02175

2_20161025T021752

20170512 20161025 L1C 3/n/a Denpasar 10,

10 13

69.808

2

UTM50S/

WGS84 *jp2

12 Bit

u *.safe

Lineage: ESA Copernicus Open Access Hub






Page 14

1.2 List of In-situ Data N/A

1.3 List of Ancillary Data

Dataset 1 (14)

DEM Incoming

Date

Acquisition

Date

AOI - City / Region /

Country

Data

Type (4)

Data

Format (3)

Projection

/

Spheroid (16)

Positional

Accuracy (11)

No. of

Classes

Area

Coverage Header / Metadata (2)

SRTM 1arcsec 20170426 11.02.2000 Denpasar Raster *.tif UTM50S

/ WGS84

Horizontal:

CE90 <20m

Vertical:

CE90 <16m

N/A 100% *.mta

Lineage:

The Shuttle Radar Topography Mission (SRTM) was flown aboard the space shuttle Endeavour on February 11 - 22, 2000. The National Aeronautics and

Space Administration (NASA) and the National Geospatial-Intelligence Agency (NGA) participated in an international project to acquire radar data which

were used to create detailed topographic maps. SRTM data are intended for scientific use with a Geographic Information System (GIS) or other special

application software. Endeavour orbited Earth 16 times each day during the 11-day mission completing 176 orbits. SRTM successfully collected radar

data over 80% of the Earth's land surface between 60° north and 56° south latitude with data points posted every 1 arc-second (approximately 30

meters).

Source: USGS Earth Explorer https://earthexplorer.usgs.gov/

Dataset 1 (14)

Water Cover Incoming

Date

Acquisition

Date


Country

Data

Type (4)

Data

Format (3)

Projection

/

Spheroid (16)

Positional

Accuracy (11)

No. of

Classes

Area


Global Surface Water

Explorer Product 20170502

1984 -

2015 Denpasar / Indonesia Raster *.tif

UTM50S

/ WGS84

RMSE < 30

m.

depending

on

dataset

100% *.xml

Lineage:






Page 15

The European Commission’s Joint Research Centre developed this new water dataset in the framework of the Copernicus Programme. This maps the

location and temporal distribution of water surfaces at the global scale over the past 32 years (1984 – 2015) and provides statistics on the extent and

change of those water surfaces. The dataset, produced from Landsat imagery, will support applications including water resource management, climate

modelling, biodiversity conservation and food security. The following six datasets are downloadable: Occurrence, Occurrence change intensity,

Seasonality, Recurrence, Transitions, Maximum water extent. Each of the downloadable files contains a colormap which will display the files in desktop

GIS tools using the symbology that has been used in the Global Surface Water Explorer website.

Source: Joint Research Centre; https://global-surface-water.appspot.com

Dataset 1 (14)

OSM Incoming

Date

Acquisition

Date


Country

Data

Type (4)

Data

Format (3)

Projection

/

Spheroid (16)

Positional

Accuracy (11)

No. of

Classes

Area


OSM Street Layer 20170502 N/A Denpasar Vector *.shp UTM50S

/ WGS84 N/A 1 100% N/A

Dataset 1 (14)

OSM Incoming

Date

Acquisition

Date


Country

Data

Type (4)

Data

Format (3)

Projection

/

Spheroid (16)

Positional

Accuracy (11)

No. of

Classes

Area


OSM Waterways Layer 20170502 N/A Denpasar Vector *.shp UTM50S

/ WGS84 N/A 1 100% N/A

Lineage: Up-to-date data extracts from the OpenStreetMap project

https://global-surface-water.appspot.com/






Page 16

Dataset 4 (14)

LULC map Incoming

Date

Acquisition

Date

AOI - City /

Region / Country

Data

Type (4)

Data

Format (3)

Projection

/

Spheroid (16)

Positional

Accuracy (11)

No. of

Classes

Area


Land Use Map 28.08.2017 +/- 2016 Denpasar Shapefile *.shp UTM50S

/ WGS84 85% 27 100% *.mta

Lineage:

Land Use/Land Cover Information was derived from Very High Resolution (VHR) and High Resolution (HR) satellite imagery. The Land Use/Land Cover

(LU/LC) product contains spatial explicit information on the different occurring land use and land cover in both the Core and Peri-Urban areas of the City of

Denpasar for the years 2006 and 2016. The Core area has detailed LU/LC nomenclature whereas the Peri-Urban area LU/LC nomenclature is at an

aggregated Level. The input data for the Core area was Very High Resolution data (Worldview, Quickbird) and the input data for the Peri-Urban area was

HR Sentinel and Landsat data. The LU/LC product is the Baseline Product from which various derived products are produced.






Page 17

2.1 EO Data Quality

Sensor 1 (8) Landsat 5

Ba

ck

up

Re

ad

ab

ilit

y (1

)

Ch

eck

He

ad

er

/

Me

tad

ata

(2

)

Band Specifications

Pro

jecti

on

/ S

ph

ero

id (

16

)

Sce

ne

Lo

ca

tio

n

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

ve

r (i

nte

rna

l

ch

eck

)

Ra

dio

me

try

To

po

gra

ph

y

Dro

pp

ed

Lin

es /

Art

efa

cts

(7

)

Acce

pta

nce

Sta

tus

Comments File Name

No

. o

f B

an

ds

Ba

nd

Re

gis

tra

tio

n

Sp

ectr

al/

Sp

ati

al

Re

so

luti

on

1. LT51160661997151DKI00 Y - Q Y - V Y- V 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



Sensor 2 (8) Landsat 8 B

ack

up

Re

ad

ab

ilit

y (1

)

Ch

eck

He

ad

er

/

Me

tad

ata

(2)

Band Specifications

Pro

jecti

on

/ S

ph

ero

id (

16

)

Sce

ne

Lo

ca

tio

n

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

ve

r (i

nte

rna

l

ch

eck

)

Ra

dio

me

try

To

po

gra

ph

y

Dro

pp

ed

Lin

es /

Art

efa

cts

(7)

Acce

pta

nce

Sta

tus

Comments File Name

No

. o

f B

an

ds

Ba

nd

Re

gis

tra

tio

n

Sp

ectr

al/

Sp

ati

al

Re

so

luti

on

1. LC81160662013115LGN02 Y - Q Y - V Y- V 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










Page 18

Sensor 3 (8)

Sentinel 2

Ba

ck

up

Re

ad

ab

ilit

y (1

)

Ch

eck

He

ad

er

/ M

eta

da

ta

(2)

Band Specifications

Pro

jecti

on

/ S

ph

ero

id (

16

)

Sce

ne

Lo

ca

tio

n

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

ve

r (i

nte

rna

l

ch

eck

)

Ra

dio

me

try

To

po

gra

ph

y

Dro

pp

ed

Lin

es /

Art

efa

cts

(7

)

Acce

pta

nce

Sta

tus

Commen

ts File Name

No

. o

f B

an

ds

Ba

nd

Re

gis

tra

tio

n

Sp

ectr

al/

Sp

ati

al

Re

so

luti

on

1. S2A_OPER_MSI_L1C_TL_EPA__

20161213T095422_A001716_

T50LKR_N02.04

Y -

Q

Y -

V

Y-

V

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 Ye

s none

2. S2A_OPER_MSI_L1C_TL_SGS__

20151230T091952_A002717_

T50LKR_N02.01

Y -

Q

Y -

V

Y-

V

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 Ye

s none

3. S2A_OPER_MSI_L1C_TL_SGS__

20160806T092156_A005863_

T50LKR_N02.04

Y -

Q

Y -

V

Y-

V

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 Ye

s none

4.

S2A_OPER_PRD_MSIL1C_PDMC_

20161025T230615_R003_V20161025T02

1752_

20161025T021752

Y -

Q

Y -

V

Y-

V

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 Ye

s none






Page 19

2.3 Ancillary Data Quality

Dataset 1 (14) DEM

Ba

ck

up

Re

ad

ab

ilit

y (1

)

Ch

eck

He

ad

er

/

Me

tad

ata

(2)

Ext

en

t

Pro

jecti

on

/ S

ph

ero

id

(16

)

Sp

ati

al R

eso

luti

on

Da

ta F

orm

at

(3)

Bit

De

pth

(5)

Lo

ca

tio

n

Co

mp

lete

ne

ss

Ge

om

. M

isa

lign

me

nt

Pla

usib

ilit

y

Dro

pp

ed

Lin

es /

Art

efa

cts

(7)

Acce

pta

nce

Sta

tus

Comments File Name

SRTM1S09E115V3 Y - Q Y - Q Y - Q Y - V Y - Q Y - Q Y - Q Y - Q Y - V Q Y - V N - V Y - V N - V Yes none

Dataset 2 (14)

Global Surface

Water Explorer

Product

Ba

ck

up

Re

ad

ab

ilit

y (1

)

Ch

eck

He

ad

er

/

Me

tad

ata

(2

)

Ext

en

t

Pro

jecti

on

/ S

ph

ero

id (

16

)

Sp

ati

al R

eso

luti

on

Da

ta F

orm

at

(3)

Bit

De

pth

(5

)

Lo

ca

tio

n

Co

mp

lete

ne

ss

Ge

om

. M

isa

lign

me

nt

Pla

usib

ilit

y

Dro

pp

ed

Lin

es /

Art

efa

cts

(7

)

Acce

pta

nce

Sta

tus

Comments

File Name

change.tif Y - Q Y - V Y - V Q Y - V Y - V Y - Q Y - Q Y - Q Y - V Y - V Q N - V Q Y - V N - V Q Yes none

extent.tif Y - Q Y - V Y - V Q Y - V Y - V Y - Q Y - Q Y - Q Y - V Y - V Q N - V Q Y - V N - V Q Yes none

occurrence.tif Y - Q Y - V Y - V Q Y - V Y - V Y - Q Y - Q Y - Q Y - V Y - V Q N - V Q Y - V N - V Q Yes none

recurrence.tif Y - Q Y - V Y - V Q Y - V Y - V Y - Q Y - Q Y - Q Y - V Y - V Q N - V Q Y - V N - V Q Yes none

seasonality.tif Y - Q Y - V Y - V Q Y - V Y - V Y - Q Y - Q Y - Q Y - V Y - V Q N - V Q Y - V N - V Q Yes none

transitions.tif Y - Q Y - V Y - V Q Y - V Y - V Y - Q Y - Q Y - Q Y - V Y - V Q N - V Q Y - V N - V Q Yes none






Page 20

Dataset 3 (14) OSM

Ba

ck

up

Re

ad

ab

ilit

y (1

)

Ch

eck

He

ad

er

/

Me

tad

ata

(2)

Ext

en

t

Pro

jecti

on

/ S

ph

ero

id

(16

)

Sp

ati

al R

eso

luti

on

Da

ta F

orm

at

(3)

Bit

De

pth

(5)

Lo

ca

tio

n

Co

mp

lete

ne

ss

Ge

om

. M

isa

lign

me

nt

Pla

usib

ilit

y

Dro

pp

ed

Lin

es /

Art

efa

cts

(7)

Acce

pta

nce

Sta

tus

Comments File Name

gis.osm_roads_free_1.shp Y - Q Y - Q N/A Y – V Q Y - Q N/A Y - Q N/A Y - V Q Y - V N - V Y - V N - V Yes none

gis.osm_waterways_free_1 Y - Q Y - Q N/A Y – V Q Y - Q N/A Y - Q N/A Y - V Q Y - V N - V Y - V N - V Yes none

Dataset 4 (14) Land Use

Ba

ck

up

Re

ad

ab

ilit

y (1

)

Ch

eck

He

ad

er

/

Me

tad

ata

(2)

Ext

en

t

Pro

jecti

on

/ S

ph

ero

id

(16

)

Sp

ati

al R

eso

luti

on

Da

ta F

orm

at

(3)

Bit

De

pth

(5)

Lo

ca

tio

n

Co

mp

lete

ne

ss

Ge

om

. M

isa

lign

me

nt

Pla

usib

ilit

y

Dro

pp

ed

Lin

es /

Art

efa

cts

(7)

Acce

pta

nce

Sta

tus

Comments Land Use/Land Cover Information

Denpasar_WB_LULC_2006_2016_Total_2_2.shp Y - Q Y - Q Y - Q Y - V Y - Q NA Y - Q NA Y - V Q Y - V N - V Y - V N - V Yes






Page 21

3.1 Geometric Correction


Pro

ce

ssin

g D

ate

AO

I

Cit

y / R

egio

n /

Co

un

try

Pro

jecti

on

/ S

ph

ero

id (

16

)

No

. &

RM

S (

m)

of

GC

Ps (

17

)

No

. &

RM

S (

m)

of

TP

s (

18

)

No

. &

RM

S (

m)

of

CP

s (

19

)

Dig

ita

l E

leva

tio

n M

od

el

(DE

M)

Mo

de

l / A

lgo

rith

m (

13

)

Re

sa

mp

lin

g

Me

tho

d (

20

)

Va

lid

ati

on

Re

po

rts

Acce

pta

nce

Sta

tus

Output File Name No. File Name

1. LT51160661997151DKI00 Landsat 5 Imagery was downloaded as Level 1 TP Products

(radiometrically calibrated and orthorectified),

refer to https://landsat.usgs.gov/geometry

2. LT51160661999317DKI00

3. LT51160662000064DKI00


Pro

ce

ssin

g D

ate

AO

I

Cit

y / R

egio

n /

Co

un

try

Pro

jecti

on

/ S

ph

ero

id (

16

)

No

. &

RM

S (

m)

of

GC

Ps (

17

)

No

. &

RM

S (

m)

of

TP

s (

18

)

No

. &

RM

S (

m)

of

CP

s (

19

)

Dig

ita

l E

leva

tio

n M

od

el

(DE

M)

Mo

de

l / A

lgo

rith

m (

13

)

Re

sa

mp

lin

g

Me

tho

d (2

0)

Va

lid

ati

on

Re

po

rts

Acce

pta

nce

Sta

tus


1. LC81160662013115LGN02

Landsat 8 Imagery was downloaded as Level 1 TP Products

(radiometrically calibrated and orthorectified),

refer to https://landsat.usgs.gov/geometry

2. LC81160662013307LGN01

3. LC81160662014006LGN01

4. LC81160662015105LGN01

5. LC81160662016012LGN02

https://landsat.usgs.gov/geometry

https://landsat.usgs.gov/geometry






Page 22

Sensor 3 (8) Sentinel 2

Pro

ce

ssin

g D

ate

AO

I

Cit

y / R

egio

n /

Co

un

try

Pro

jecti

on

/ S

ph

ero

id (

16

)

No

. &

RM

S (

m)

of

GC

Ps (

17

)

No

. &

RM

S (

m)

of

TP

s (

18

)

No

. &

RM

S (

m)

of

CP

s (

19

)

Dig

ita

l E

leva

tio

n M

od

el

(DE

M)

Mo

de

l / A

lgo

rith

m (

13

)

Re

sa

mp

lin

g

Me

tho

d (2

0)

Va

lid

ati

on

Re

po

rts

Acce

pta

nce

Sta

tus


1.

S2A_OPER_MSI_L1C_TL_EPA__

20161213T095422_A001716_

T50LKR_N02.04

Sentinel 2 Imagery was downloaded as Level 1C Products.

Level-1C processing includes radiometric and geometric corrections including ortho-rectification

and spatial registration on a global reference system with sub-pixel accuracy.

(refer to https://sentinel.esa.int/web/sentinel/user-guides/sentinel-2-msi/processing-levels/level-1)

2.

S2A_OPER_MSI_L1C_TL_SGS__

20151230T091952_A002717_

T50LKR_N02.01

3.


20160806T092156_A005863_

T50LKR_N02.04

4.


20161025T230615_R003_V20161025T021752_

20161025T021752

https://sentinel.esa.int/web/sentinel/user-guides/sentinel-2-msi/processing-levels/level-1






Page 23

3.1.1 Data Fusion

Natural colour and false colour composites were produced from all EO data listed in Section 1.1., merging visible bands (R, G, B) and NIR band

3.2 Data Processing

Sensor 1 (8) Landsat 5 Atmospheric

Correction

Radiometric

Processing (15)

Topographic

Normalisation

OTHER

Calculation:

Acce

pta

nce

Sta

tus

Ba

ck

up

Comment

No. File Name Processing

Date

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

1. LT51160661997151DKI00 20170518 Yes QGIS/DOS1 No downloaded

Level 1 TP

Products

radiometrically

calibrated

No topographic

effects in

flood-prone

areas

negligible

No N/A Yes Yes none

2. LT51160661999317DKI00 20170518 Yes QGIS/DOS1 No No No N/A Yes Yes none

3. LT51160662000064DKI00 20170518 Yes QGIS/DOS1 No No No N/A Yes Yes none

Sensor 2 (8) Landsat 8 Atmospheric

Correction

Radiometric

Processing (15)

Topographic

Normalisation

OTHER

Calculation:

Acce

pta

nce

Sta

tus

Ba

ck

up

Comment

No. File Name Processing

Date

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

1. LC81160662013115LGN02 20170518 Yes QGIS/DOS1 No downloaded

Level 1 TP

Products

radiometrically

calibrated

No topographic

effects in

flood-prone

areas

negligible

No N/A Yes Yes none

2. LC81160662013307LGN01 20170518 Yes QGIS/DOS1 No No No N/A Yes Yes none









Page 24

Sensor 3 (8)

Sentinel 2 Atmospheric

Correction

Radiometric

Processing (15)

Topographic

Normalisation

OTHER

Calculation:

Acce

pta

nce

Sta

tus

Ba

ck

up

Commen

t No. File Name

Processing

Date

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Softwar

e /

Method

1.


20161213T095422_A001716_

T50LKR_N02.04

2017051

9

Ye

s

QGIS/DOS

1

N

o

downloaded

Level 1C

Products

radiometricall

y calibrated

N

o

topographi

c effects in

flood-prone

areas

negligible

N

o N/A

Ye

s

Ye

s none

2.


20151230T091952_A002717_

T50LKR_N02.01

2017051

9

Ye

s

QGIS/DOS

1

N

o

N

o

N

o N/A

Ye

s

Ye

s none

3.


20160806T092156_A005863_

T50LKR_N02.04

2017051

9

Ye

s

QGIS/DOS

1

N

o

N

o

N

o N/A

Ye

s

Ye

s none

4.


20161025T230615_R003_V20161025T021752

_

20161025T021752

2017051

9

Ye

s

QGIS/DOS

1

N

o

N

o

N

o N/A

Ye

s

Ye

s none






Page 25

4.1 Classification


Processing

Date

Cloud Masking Thematic

Classification

Manual

Enhancement Mosaicking

Acce

pta

nce

Sta

tus

Ba

ck

up

Comment

No. File Name

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

1. LT51160661997151DKI00 20170520 No N/A Yes

ArcGIS 10 /

AWEI-

calculation

No N/A No N/A Yes Yes none

2. LT51160661999317DKI00 20170520 Yes ArcGIS 10 /

manual Yes

ArcGIS 10 /

AWEI-

calculation

No N/A Yes

ArcGIS 10 /

Mosaic

Dataset

Yes Yes none

3. LT51160662000064DKI00 20170520 Yes ArcGIS 10 /

manual Yes

ArcGIS 10 /

AWEI-

calculation

No N/A Yes

ArcGIS 10 /

Mosaic

Dataset

Yes Yes none


Processing

Date


Classification

Manual


Acce

pta

nce

Sta

tus

Ba

ck

up

Comment

No. File Name

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

1. LC81160662013115LGN02 20170520 No N/A Yes

ArcGIS 10 /

AWEI-

calculation


2. LC81160662013307LGN01 20170520 Yes ArcGIS 10 /

manual Yes

ArcGIS 10 /

AWEI-

calculation

No N/A Yes

ArcGIS 10 /

Mosaic

Dataset

Yes Yes none

3. LC81160662014006LGN01 20170520 Yes ArcGIS 10 /

manual Yes

ArcGIS 10 /

AWEI-

calculation

No N/A Yes

ArcGIS 10 /

Mosaic

Dataset

Yes Yes none

4. LC81160662015105LGN01 20170520 No N/A Yes

ArcGIS 10 /

AWEI-

calculation


5. LC81160662016012LGN02 20170520 No N/A Yes

ArcGIS 10 /

AWEI-

calculation







Page 26

Sensor 3 (8) Sentinel 2

Processing

Date


Classification

Manual


Acce

pta

nce

Sta

tus

Ba

ck

up

Comment

No. File Name

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

pro

ce

sse

d

Software /

Method

1.


20161213T095422_A001716_

T50LKR_N02.04

20170522 No N/A Yes

ArcGIS 10 /

AWEI-

calculation


2.


20151230T091952_A002717_

T50LKR_N02.01

20170522 No N/A Yes

ArcGIS 10 /

AWEI-

calculation


3.


20160806T092156_A005863_

T50LKR_N02.04

20170522 Yes ArcGIS 10 /

manual Yes

ArcGIS 10 /

AWEI-

calculation

No N/A Yes

ArcGIS 10 /

Mosaic

Dataset

Yes Yes none

4.


20161025T230615_R003_V20161025T021752_

20161025T021752

20170522 Yes ArcGIS 10 /

manual Yes

ArcGIS 10 /

AWEI-

calculation

No N/A Yes

ArcGIS 10 /

Mosaic

Dataset

Yes Yes none






Page 27

5.1 Thematic Accuracy

Qualitative (visual) Comparison of Water Extents based on Automated Water Extraction Index (AWEI) with Source Imagery







Page 28

5.2 Error Matrices

Class No Reference Data

Totals

1 2

Water Extent

20150415

No water

cover

water

cover

1 No

water

cover

35 1

36

2 water

cover 2 34 36

Totals 37 35 72

Accuracy Statistics z= 1,96

Overall

Accuracy: 95,83%

95%

Confidence

Interval:

N/A N/A

Kappa

(Cohen): 0,91667

Class Name Producer

Accuracy

95% Confidence

Interval

User

Accuracy 95% Confidence Interval

Class 1 94,59% N/A N/A 97,22% N/A N/A

Class 2 97,14% N/A N/A 94,44% N/A N/A


Totals

1 2

Water Extent

20151024

No water

cover

water

cover

1 No

water

cover

34 2

36

2 water

cover 5 31 36

Totals 39 33 72








Page 29

Overall

Accuracy: 90,28%

95%

Confidence

Interval:

N/A N/A

Kappa

(Cohen): 0,80556

Class Name Producer

Accuracy

95% Confidence

Interval

User


Class 1 87,18% N/A N/A 94,44% N/A N/A

Class 2 93,94% N/A N/A 86,11% N/A N/A


Totals

1 2

Water Extent

20161025

No water

cover

water

cover

1 No

water

cover

36 0

36

2 water

cover 2 34 36

Totals 38 34 72


Overall

Accuracy: 97,22%

95%

Confidence

Interval:

N/A N/A

Kappa

(Cohen): 0,94444

Class Name Producer

Accuracy

95% Confidence

Interval

User


Class 1 94,74% N/A N/A 100,00% N/A N/A

Class 2 100,00% N/A N/A 94,44% N/A N/A






Page 30

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


1. HR EO Data

Denpasar

peri-urban

(540,94 km²)

540,94 km² 100% 12 3 100% Yes none

2. VHR EO Data 141,51 km² 26,2 % 3 0 N/A Yes none

3. In-situ Data N/A N/A N/A N/A N/A N/A none

4. Ancillary Data 540,94 km² 100 % 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


1. Water Extent

Denpasar

Denpasar

peri-urban

(540,94 km²)

540,94 km² 100% 100% N/A 100% Yes none

2. Flood History

River Floods

Denpasar urban

(141,51 km²) 141,51 km² 100% 100% N/A N/A Yes none

3. Flood Hazard

River Floods

Denpasar

peri-urban

(540,94 km²)

540,94 km² 100% 100% N/A N/A Yes none

4.

Flood Hazard

Denpasar

Combined

Denpasar

peri-urban

(540,94 km²)

540,94 km 100% 100% N/A N/A Yes none

5.

Flood Risk

Classification

Denpasar

Denpasar

peri-urban

(540,94 km²)

540,94 km 100% 100% N/A N/A Yes none







Page 31

6.2 Compliancy

Product 1 (28) Water Extent Denpasar

Abstract






Landsat Imagery.

For demonstration purposes, water extents of recent dates as derived from HR EO data are included in

the present Geodatabase.


Area Coverage


Country 1: Indonesia/Denpasar Country 1: Indonesia/Denpasar yes

A) Wall-to-wall: A) Wall-to-wall: yes

Urban and peri-urban area Urban and peri-urban area


B) Sampling based: B) Sampling based: N/A

N/A N/A



Baseline Year(s): Baseline Year(s): yes

2000 2000


Update Frequency depending on frequency of flooding events



WGS84 / UTM 50S WGS84 / UTM 50S yes



Class 1: Extent of flooding Class 1: Extent of flooding yes



Cloud covered areas shall be

removed.

Data selection was aiming at low cloud

coverage in flood exposed areas.

Relevant clouds manually digitized and

replaced using additional EO data.

yes

Spatial Resolution


10m – 30 m 10 m – 30 m yes







Page 32



0.5 ha 0.1 ha yes

Data Type


Thematic Vector or Raster File Thematic Vector File yes

Bit Depth


N/A N/A N/A

Data Format


*.shp Shapefile Shapefile yes

Class Coding


0 = No Water Cover 0 = No Water Cover yes

1 = Water Cover 1 = Water Cover yes

Metadata


not specified not specified N/A


Thematic Accuracy


Overall Accuracy: > 80% Overall Accuracy: > 80% yes

Positional Accuracy


RMSE < 30 m RMSE < 30 m yes

Delivery Procedure

Service Provision


online via FTP online via FTP yes

Delivery Date


End of May 2017 31.05.2017 yes







Page 33

Product 2 (28) Flood History River Floods

Abstract




To overcome this shortcoming, point data of reported and localized urban floods (2008 – 2017) are

included in this product.


Area Coverage




Urban area Urban area



N/A N/A



Baseline Year(s): Baseline Year(s): N/A

not specified 2008 - 2017








Class 1: Localization and date of flooding Class 1: Localization and date of flooding yes



N/A N/A N/A

Spatial Resolution


N/A N/A N/A







Page 34



N/A N/A N/A

Data Type



Bit Depth


N/A N/A

Data Format



Class Coding


1 = Localization and date of flooding 1 = Localization and date of flooding yes

Metadata


N/A N/A N/A


Thematic Accuracy


N/A N/A N/A

Positional Accuracy


N/A N/A N/A

Delivery Procedure

Service Provision



Delivery Date


End of May 2017 31.05.2017 yes







Page 35

Product 3 (28) Flood Hazard River Floods

Abstract


– March) is typical.

The present Geodatabase includes two shp-files for rough estimation of potential flooding zones:

a. Denpasar_Flood_Hazard_Rivers

Waterways were taken from OSM Layer and complemented by hydrologic modelling of potential

catchment areas and flow routes based on SRTM and visual interpretation of VHR data. After

classification in two classes based on Stream Order the lines were buffered with 50 m and 100 m

respectively to roughly estimate potential flooding zones.

b. Denpasar_Flood_Reports_buff200

Point data of reported and localized urban floods (2008 – 2017) were buffered with 200 m to roughly

estimate potential flooding hot-spots


Area Coverage




Peri-urban area Peri-urban area



N/A N/A












1: Potential flooding zones close to

rivers and waterways 1:

Potential flooding zones close to rivers

and waterways yes

2 Potential flooding hot-spots 2 Potential flooding hot-spots yes



N/A N/A N/A

Spatial Resolution


N/A N/A N/A







Page 36



N/A N/A N/A

Data Type



Bit Depth


N/A N/A N/A

Data Format



Class Coding


Potential flooding zones close to rivers and waterways Potential flooding zones close to rivers and

waterways yes

Potential flooding hot-spots Potential flooding hot-spots yes

Metadata


not specified not specified N/A


Thematic Accuracy


N/A N/A N/A

Positional Accuracy


N/A N/A N/A

Delivery Procedure

Service Provision



Delivery Date


End of May 2017 31.05.2017 yes







Page 37

Product 4 (28) Flood Hazard Denpasar Combined

Abstract


– March) is typical. Even though Denpasar is prone to flooding according to available local sources, this




Landsat Imagery.

As land subsidence is expected in Denpasar due to observed decrease in groundwater level coastal

lowlands are classified as flood prone at a distance of 200 m. This does not include Tsunami flooding.


and press releases, results of modelling of potential flooding along rivers and waterways, ancillary data)

aiming at covering all types of floods and thus all areas which are potentially endangered by flooding.

The classification in three qualitative hazard levels is expert-based under consideration of observed and



Area Coverage







N/A N/A





B) Update Frequency B) Update Frequency N/A

N/A N/A







1 low hazard 1 low hazard yes

2 medium hazard 2 medium hazard yes

3 high hazard 3 high hazard yes



N/A N/A N/A

Spatial Resolution








Page 38

Vector file Vector file yes



N/A N/A N/A

Data Type



Bit Depth


N/A N/A N/A

Data Format



Class Coding


low hazard: RGB Code 255-255-115 low hazard: RGB Code 255-255-115 yes

medium hazard: RGB Code 255-211-127 medium hazard: RGB Code 255-211-127 yes

high hazard: RGB Code 255-127-127 high hazard: RGB Code 255-127-127 yes

Metadata


ISO compliant ISO compliant yes


Thematic Accuracy


N/A N/A N/A

Positional Accuracy


Not specified – depends on

available input data N/A N/A

Delivery Procedure

Service Provision









Page 39

Delivery Date


End of May 2017 15.09.2017 no

Product 5 (28) Flood Risk Classification Denpasar

Abstract



case of flooding. The land-use map was provided by GAF AG and recoded to pre-defined categories to

ensure consistent results.

The exposition is classified integrating economic costs, social damage, physical damage and flood

duration. Four land use damage levels (A, B, C, D) are defined based on this estimation.

The Flood Risk matrix is generated based on these results code and on flood hazard classified into three

hazard levels. The flood risk level is classified in four qualitative classes based on the combination of

flood hazard and land use damage.


Area Coverage







N/A N/A





B) Update Frequency B) Update Frequency N/A

not specified N/A

depends on frequency of flooding events and on availability of land-use classification (5 to 7 years)






1 low risk 1 low risk yes

2 medium risk 2 medium risk yes

3 high risk 3 high risk yes

4 very high risk 4 very high risk yes



N/A N/A N/A







Page 40

Spatial Resolution


Vector file Vector file yes



0.25 ha for the urban area, 0.5 ha

for the peri-urban area.

0.25 ha for the urban area, 0.5 ha for

the peri-urban area. yes

same as landuse / landcover classification

Data Type



Bit Depth


N/A N/A N/A

Data Format



Class Coding


low risk: RGB Code 255-255-190 low risk: RGB Code 255-255-190 yes

medium risk: RGB Code 255-211-127 medium risk: RGB Code 255-211-127 yes

high risk: RGB Code 255-170-0 high risk: RGB Code 255-170-0 yes

very high risk: RGB Code 255-0-0 very high risk: RGB Code 255-0-0 yes

Metadata


ISO compliant ISO compliant yes


Thematic Accuracy


not specified depends on available input data N/A

Positional Accuracy


depends on available input data N/A







Page 41

Not specified – depends on

available input data

Delivery Procedure

Service Provision



Delivery Date


End of May 2017 15.09.2017 no

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.







Page 42

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.

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


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


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.







Page 43

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.







Page 44