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
Session
09-03: Innovation Fair: Analyzing Land Data & Data Processing
Time:
Thursday, 23/Mar/2017:
8:30am - 10:00am

Session Chair: Christoph Aubrecht, European Space Agency (ESA) & The World Bank, United States of America
Location: MC 2-800

Streaming.


Show help for 'Increase or decrease the abstract text size'
Presentations

Better demographic data for improved planning and problem solving in South African cities

Sophie McManus1, Peter Magni2, Geci Karuri Sebina3, Matthew Adendorff4, Richard Gevers5

1Open Data Durban, South Africa; 2South African Cities Network; 3South African Cities Network; 4Open Data Durban, South Africa; 5Open Data Durban, South Africa

With unprecedented population growth predictions in cities, urban functions as we know them will be challenged and pushed beyond their current threshold (UN SDGs, 2015). This reality is driving the global urban narrative with concepts such as “sustainability”,“responsiveness” “resilience” and “inclusiveness”. Now more than ever, the use of evidence-based planning by local government and other urban role-planners is at a critical point in been able to shape future cities.

The proposed Lightning Talk presents an innovative tool developed by South African Cities Network by Open Data Durban to provide an evidence based approach to predict demographic trends at the ward level (the smallest South African administrative unit) of South African cities, and overcome limitations in current demographic data. In conjunction with demographic projections an open data planning tool and portal (SCODA) will be used to store the projections, and allow for government and a range of community stakeholders to store and combine other data sets for intensive analysis and modeling. SCODA is a user-designed tool, co-created with and for government and civil society, underpinned by the overarching theme of impacting on spatial transformation.



OpenLandContracts.org: A Tool for Facilitating Land Contract Disclosure

Jesse Coleman, Kaitlin Cordes, Sam Szoke-Burke

Columbia Center on Sustainable Investment, United States of America

A critical lack of transparency permeates the contracting processes that govern large-scale land investments. Despite widespread consensus that greater transparency of contractual terms constitutes best practice, few governments or investors have been willing or able to proactively disclose their land contracts. In the rare contexts in which contracts are disclosed, they are often difficult to find or comprehend. As the world’s first global repository of land contracts, OpenLandContracts.org helps address these challenges, offering a solution that combines technology with substantive technical expertise to make investor-state land contracts easier to disclose, access, and understand. Crucially, the platform also provides governments with the opportunity to develop country-specific repositories for contract disclosure, thereby providing an innovative and cost-effective solution for host governments and policy-makers seeking to improve their transparency and good governance efforts.

This Lightning Talk will include an interactive demonstration of the key features of OpenLandContracts.org, including its tools to help users search, compare, and understand documents, as well as its links to relevant data standards and complementary initiatives. The Talk will provide further information regarding the opportunity to develop country-specific repositories in partnership with the OpenLandContracts.org team, and will highlight examples of how information disclosed through OpenLandContracts.org has been used by a range of stakeholders.



Big Data Approaches to Market Analysis: SE Asian Palm Oil

Gabriel Thoumi. CFA. FRM

Climate Advisers, United States of America

Palm oil is an inexpensive and highly versatile oil derived from the fruit of the oil palm tree, a native of West Africa's tropical forests. It is found in half of all consumer goods on the shelves today in Western grocery stores. Palm kernel oil is also used as a bioeful to power vehicles, heat homes, and manufacture plastics. Due to its high yields and many uses, palm oil is the most actively traded edible oil in the world, with annual sales of $50 billion. For decades, however, the palm oil business has been criticized for its links to corruption, extinction, social injustice, and deforestation.

This presentation will show how big data tools can be applied to investigate the SE Asian palm oil industry. Based on a previously funded 4-month pilot project in 2016, analysis will show how Bloomberg Terminal data and independent research can be combined to show governance trends using two separate big data tools – Gephi (macro level) and LittleSis (micro-level).

The pilot will demonstrate what are the key investors and executive and board-level interlocks demonstrating industry-wide conflicts of interest – and at times even potentially illegal behavior.



Mapping Economic Indicators Using Commercial Satellite Imagery: A Case Study in Poverty Analysis of Sri Lanka to Explore Best Practices for Feature Extraction and Classification

Nick Hubing, Chris Lowe

LAND INFO Worldwide Mapping, LLC, United States of America

Typical sources of economic indicators such as household surveys and census data are time-consuming and expensive, some areas also have security concerns or access constraints. Lack of current information complicates the efforts of policy makers and aid organizations. This talk will use an extensive Sri Lanka poverty mapping project completed for the World Bank to examine the use of high-resolution satellite imagery, which has a relatively low cost and the potential for frequent updates, as a source to generate economic indicators. We will identify indicators such as building footprints, roof material type, transportation vectors and cultivation that can be extracted. We will highlight how to perform feature extraction and classification with a comparison of manual methods (hand digitizing) and automated classification techniques, including OBIA (Object Based Image Analysis) computer vision and cloud-processing. Output of custom mapping, including statistics, will be compared to open-source mapping (OpenStreetMap), and applications of map data layers extracted from imagery will be discussed.



Software Comes and Goes, Data Stays Forever!

Arnulf Christl

Metaspatial, Germany

Software developers complain about wrong data. It just does not comply to the specs. It is a complete mess. The owners of the data do not understand the agitation because they have worked with some odd, old software for ages and it sort of always worked out. Why change now? The metadata sits in the mind of so many people but has never been organized and fixed (to fit into your software). The data manager on the other hand has the same problems but the other way round. The data does not download, the encoding is wrong, the coordinate system is screwed and at the end all the decimals are cut off. Who is right? It is always the data owner! Because software comes and goes, the data stays. This talk is a plea to mind the data. Remember this: Never use software that does not work on open formats. Really open. Like in Open Source but even better.



Combining GIS And Remote Sensing Data Using Open Source Software To Assess Natural Factors That Influence Poverty

Markus Neteler, Carmen Tawalika, Till Adams, Hinrich Paulsen

mundialis GmbH & Co. KG, Germany

Worldwide livelihoods are at risk from a multitude of factors like drought, agricultural pests, pandemics, fluctuating commodity prices, or political instability. Monitoring vast areas is only possible with new Earth observation (EO) data including the European Copernicus Sentinel as well as the US Landsat-8 satellite data being available in almost real-time under an Open Data license. Yet there is a vast discrepancy between raw data availability and the uptake by policy makers due to lack of knowledge in processing satellite data, of computational power and fundamentally not knowing what type of information can be extracted to address societal challenges.

The company mundialis is a remote sensing and big geodata analyst startup from 2015, providing EO products through standardized web mapping services. Mundialis leverages the power of geospatial open source software brought, providing insights into time dependent events or gradual changes, hidden correlations and short term risk to support data driven decisions. The presentation will focus on temperature, a main driver for most ecological processes as well as human welfare. It is a key indicator for a multitude of applications from growing crops, assessing the potential spread of emerging disease vectors or agro-pests, to the detailed identification of urban heat islands.



Walking Spatio-Temporal Datacubes - Seamlessly from Laptop to Cloud to Federations

Peter Baumann

Jacobs University, Germany

rasdaman (“raster data manager”) enables agile analytics on massive spatio-temporal datacubes, including sensor, imagery, image timeseries, simulation, and statistics data. For land & poverty tasks rasdaman provides simple, integrated access and mix&match of satellite, weather, and further data dynamically. Working equally well in networked and standalone (e.g., rural) settings rasdaman supports intelligent farming, land use and environmental monitoring, disaster management, etc.

Flexibility, performance, scalability, and open standards set multi-award winning rasdaman apart, together with its adaptive mass data ingest. A plethora of open clients attaches itself to rasdaman, including OpenLayers, QGIS, NASA WorldWind, python, etc. Innovative enablers such as adaptive distributed storage, cloud parallelization/distribution, and use of heterogeneous hardware make rasdaman excel over, e.g., Spark in independent benchmarks. Laptops, clouds, and datacenters federate easily through their rasdaman installations offering a common single information space to users.

Open-source rasdaman is Reference Implementation for OGC & INSPIRE WCS and the blueprint for ISO Array SQL and OGC WCPS. Installed at leading data centers, such as NASA/US, ECMWF/Europe, and NCI/Australia, datacubes are exceeding 250 TB, growing towards PB. Hitechs utilize rasdaman for value-adding geo services.

Demos will include realtime federation, TB datacube queries, and on-the-fly addition of new datasets.



Innovative application of machine learning and AI using earth observation to provide dynamic data and actionable intelligence.

Conor Gerard Smyth, Alexis Smith

IMGeospatial.com/Intelligent Modelling Ltd, United Kingdom

As are an innovative company with a strong societal change ethos. We create AUTOMATICALLY SMARTER™ products based on machine learning and artificial intelligence. Our patented processing algorithms allow for automatic feature identification and extraction from earth observation and multi-scopic data without manual intervention and with the principal advantage of being able to provide near real-time topographic feature change detection, auto-map updates and feature classification and related information which are crucial in a range applications such as real-time monitoring, land-use change monitoring, disaster risk reduction and dynamic mapping of urban and other developments among many other end applications.

Our capability is more than just overcoming data gaps, we aim to provide actionable intelligence to assist our users, clients and partner organisations resolve many of their significant challenges whether immediate or strategic in nature. Our solutions to the many global and societal challenges are facilitated through our ability to:

• Create new insights and business intelligence

• Create sustainable profitability

• Resolve problems in near real-time

• Enhance planning and risk mitigation

• Integrate into existing geospatial platforms

• Facilitate positive societal change



High Accuracy, Low Cost Cadastral Mapping

Brent Jones

Esri, United States of America

Technology continues to evolve in all areas – ease of use, access to data, lower cost, and high accuracy. This presentation will explain how new GIS technology leverages survey accurate GPS using Android phones/tablets, satellite imagery and ArcGIS Online, all in a simple, accessible, low cost platform. This technology is secure, sustainable and enables organizations to get started and scale for the future.



Mapbox for Land

Mikel Maron

Mapbox, United States of America

Learn where Mapbox is pushing the edges of the possible with geospatial visualization, and where we can go together for the land sector