In some disasters, economies appear to rebound better than expected. Occasionally, however, disasters will cause damage to critical assets that will impede recovery to such a large degree that the fiscal impact will overshadow the primary event. Examples include:
- The levee breach in New Orleans following Hurricane Katrina
- The nuclear power plant meltdown in Fukushima following the Tohoku tsunami
- Disruptions to the supply of vital computer and automotive equipment following the floods in Thailand in 2011.
Geospatial data gives the location of critical infrastructure and can be used to identify and mitigate damage. However, in many cases the location of key places are unmapped or unshared, particularly in developing countries.
With funding from the NASA Research Opportunities in Space and Earth Science (ROSES) Program, ImageCat has been working with research partners, CIESIN and HOTOSM to find where regional risks from critical infrastructure disruptions could impact the economy and reverse the progress of developing countries. Their project Disaster Forecasting, Mitigation and Response is one of the recently announced Group on Earth Observations (GEO) - Google Earth Engine (GEE) Program winners granted funding to tackle environmental and social challenges using open Earth data.
We spoke with the Principal Investigator on the project, Charles Huyck from ImageCat, to learn more about how the team will be using Google Earth Engine (GEE) to expand their ability to model the catastrophic impacts of infrastructure disruptions in their project.
Charles, can you tell us about your project?
Our team at ImageCat along with our partners at CIESIN, have established methods of developing building exposure databases for modeling damage from natural disasters. This is achieved by harnessing sampling technologies, moderate resolution satellite imagery, cloud computing, and Artificial Intelligence (AI).
To date, we have developed exposure databases for risk studies in almost 100 countries and regions. We are expanding these methods to critical infrastructure, and early results indicate that the data will be suitable for the types of risk studies that are prioritised by the Sendai Framework for Disaster Risk Reduction and implemented by Non-Governmental Organisations (NGOs) in developing countries.
What challenge is your project addressing?
The project aims to address the inconsistent global databases available that characterise critical infrastructure and key economic assets such as industrial facilities. As with global population datasets, a global critical infrastructure database would not be able to provide facility-level detail, but would provide a much-needed consistent estimate of the geographic distribution of assets. Achieving this basic goal is going to require rethinking the way we model risks to critical infrastructure. It simply is not possible to develop detailed asset-specific data required for supply chain modeling at a national level, for example. To rethink the process of modeling cascading impacts, we have turned to macroeconomic modeling methods and are using global scale hazard data and EO data.
How has this challenge impacted communities?
Cities are complex systems with interconnected “lifeline networks” that are enabled by critical infrastructure. When these are severely damaged or destroyed in a disaster, the resulting economic stagnation is felt far beyond the limited direct damage to the facilities themselves.
The potential for cascading effects is recognised but has largely been unmodeled due to lack of data, and thus, the risk is left unmitigated and uninsured. An example of this is following the 2011 floods in Thailand, the damage to critical systems impeded the recovery process and the country's GDP dropped by over 10% in the following year.
How does Google Earth Engine help you achieve your project-related goals?
Google Earth Engine is allowing the team to make rapid technical progress through its built-in catalogue of geospatial data and powerful parallel-processing capabilities, as well as the ability to ingest our own large datasets. We will also be able to:
- Rapidly prototype and refine our code by quickly adjusting the input datasets and classification parameters.
- Gain immediate visual feedback that is available during each change iteration.
- Integrate with AI processing capabilities to allow for useful experimentation.
- Host our data and techniques on a ready-made platform so that they will be usable and modifiable by the worldwide risk community.
How will the GEO-GEE funding help your project?
The training and support provided by the GEO-GEE program will help our developers get up to speed with the GEE platform and ensure that we get the most out of the platform to meet the needs of our project.
The funding will enable us to experiment with more advanced AI image classification techniques to determine if better results can be achieved through different image classification processes.
What does success look like to you?
At the end of the project, we hope to have a global database indicating regions associated with critical infrastructure and industrial production, paired with an index of the risk of cascading economic impact that communities worldwide can use to address Disaster Risk Reduction and Climate Change Adaptation challenges.
We would like to see this used by those interested in global risk identification and management, including NGOs in developing countries, CAT modellers, and other agencies that are involved in addressing the economic impact of disasters.
We would also like to make sure that the source materials used to develop the dataset are openly available for experimentation and modification by those that would like to use and repurpose the data for a variety of purposes, much like a global population dataset.
EO Data Science’s role in the GEO - GEE Program
EO Data Science partnered with Google Earth Engine and the Group on Earth Observations to launch the GEO-GEE Program, which supports GEO member countries to operationalise their science as they strive to tackle the world’s biggest sustainable development challenges.
In July 2020, 32 projects across 22 countries were selected into the program which offers 3 million USD towards product licenses and 1 million USD in technical support from EO Data Science. This funding and support will help these projects tackle global challenges using open Earth data. Read the announcement and list of winners here.
Follow us on Twitter, LinkedIn or Facebook to hear about more GEO-GEE winning projects!
Back To News Stories