Conference Agenda

The conference agenda provides an overview and details of sessions. In order to view sessions on a specific day or for a certain room, please select an appropriate date or room link. You may also select a session to explore available abstracts and download papers and presentations.

Session Overview
11-08: Using Satellite Imagery for Urban Change Detection
Thursday, 22/Mar/2018:
2:00pm - 3:30pm

Session Chair: Ran Goldblatt, New Light Technologies, United States of America
Location: MC 8-100


Peering into Megacities from Space

Jon Kher Kaw1, Tomas Soukup2, Jan Kolomaznik2, Annie Bidgood1, Hyunji Lee1

1The World Bank Group, United States of America; 2GISAT

Drawing on detailed geo-spatial analysis of land use maps derived from very high-resolution (VHR) satellite imagery, this paper develops a methodology for mapping the characteristics of physical urban spaces and spatial growth of two megacities in two points in time - Karachi and Dhaka. The findings and analysis are subsequently applied towards actual World Bank operations on the ground by prioritizing and identifying infrastructure and public space investments.


Identifying Urban Areas Combing Data from the Ground and Outer Space: An Application to India

Yue Li1, Virgilio Galdo2, Martin Rama3

1World Bank, United States of America; 2World Bank, United States of America; 3World Bank, United States of America

We develop a tractable method to identify urban areas in India which differs from the previous literature. Instead of cells, we use officially defined cities, towns and villages as our unit of analysis. We rely on structured subjective assessments to make judgement for a large stratified sample administrative units. We propose a method that combines multiple sources of information to identify urban areas. We geo-reference population census data, which comes from the ground. We also incorporate data from open source satellite-imagery, which come from outer space, on both built-up areas and nighttime lights. Data are combined through a regression analysis conducted on the sample, that we then use to make prediction for out-sample units. This exercise yields a more accurate picture of urbanization in India than was available before. The analysis confirms the value added from a credible assessment based on what one “sees” and from combining multiple indicators.


Earth Observati On For Urban Sustainable Development: Advancements For Supporting Land Use Planning In Urban Areas

Thomas Haeusler, Sharon Gomez, Fabian Enssle

GAF AG, Germany

Many developing countries are lagging behind in the operational utility of Earth Observation (EO) for extraction of spatial data for urban planning. The EO for Sustainable Development Urban project started in 2016 and in collaboration with Multi-Lateral Development Banks, support urban development programmes with a suite of geo-spatial data. Sixteen global Cities were prioritised for the first year of the project and each City obtained Land Use/Land Cover (LU/LC) data as well as products such as Green Areas, Transport Networks, Informal Settlements, and Population Density. A total of 204 products were produced and provided to the Users. The overall accuracies achieved for the various products ranged from 85-95%. Examples of analytical work included assessment of urban growth over time, LU/LC change over time, and assessment of flood prone/flood risk areas. A capacity building component is included to address some of the important technology transfer to the City Authorities.


Less Greenery For The Poor? _ Social Inequity And Green Space Distribution In Tropical Asian Megacities

Yun Hye Hwang, Ivan Kurniawan Nasution, Deepika Amonkar

National University of Singapore, Singapore

Many studies on disparities in the distribution of green spaces have mainly focused on access to public open spaces that are usually for recreational purpose. However, when other types of urban nature beyond designated parks are accounted for, claims of green space distributive injustice may different in fast growing Asia cities context.

The research employs spatial regression to examine green space distribution in association with property value of districts in two Asia megacities, Mumbai and Jakarta. Green space provision is measured by four aspects: area, planting density, proximity, and type. Results show that positive association between poverty and green space coverage. These spaces provide important ecological and biophysical functions and may harbour significant biodiversity. The paradoxical situation of ‘more greens but less jobs’/ ‘poorer, but richer biodiversity’ requires further discussions on the role of planners and designers towards socio-ecologically sustainable cities.