A Data Revolution. From Above.
Last week I participated in the inaugural Satsummit, convened by our friends at Development Seed and Mapbox. The gathering brought together a mix of professionals from both the commercial satellite industry and the humanitarian and development sectors to discuss opportunities for further collaboration in the use of earth observation to enhance global development.
While satellite imagery has been utilized in the defense industry for decades, with the advent of “CubeSat” hardware technology, the subsequent growth of the commercial space industry, and the decreasing cost of data analytics there is an emerging opportunity to explore how an increasing flow of high value imagery can benefit humanity. The conversation highlighted, by the diverse set of actors present, how we must overcome several ongoing challenge in order to realize the full potential of this exciting new technology.
Understanding of the Technology
There are many different types of imagery available that all serve distinct functions. In order to properly engage with industry, development practitioners should have a baseline understanding of these various capabilities.
- Temporal Resolution—refers to the precision of a measurement with respect to time. The partnership between The Rockefeller Foundation and Planet Labs provided high frequency imagery to teams participating in the Global Resilience Challenge so that they would be able to monitor landscape change in a defined geographic area.
- Spatial Resolution—refers to the number of pixels utilized in construction of a digital image. Images having higherspatial resolution are composed with a greater number of pixels than those of lower spatial resolution, thus a crisper image at a more refined level. Digital Globe‘s provision of high resolution imagery to first responders after Typhoon Haiyan demonstrated how a very sharp resolution was necessary in order to assess storm damage.
- Spectral Resolution—refers to the ability to resolvespectral features and bands into their separate components, including infrared and ultraviolet light not visible to the human eye. Although technical in nature, a simple way to understand this concept is by breaking an image down into its invisible light bands certain characteristics can be highlighted. Examples include distinguishing muddy water from muddy land or pinpointing forest fires under cloud cover.
For the new user, the World Bank Group recently released this comprehensive landscape of satellite capabilities to increase understanding among potential new users.
In order to access imagery, users must obtain proper licensing from the provider. License types range for different types of users and usage bases. Given that each satellite company has its own form of agreement, it can become difficult for development practitioners to navigate the full landscape in order to assess the best partner for a particular project. There is an acknowledgement that standardization may be difficult due to each imagery provider maintaining an effective business model, however, initial efforts are currently underway to streamline the process for obtaining and publishing imagery.
There is a need for easier to access Application Programming Interfaces (APIs) made available so that those organizations working on global development challenges with the appropriate technical analytical capabilities are able to access and process the data that is currently available for their use. Landsat, collected by the USGS, is the largest free archive of earth observation imagery available for the public. While in the past this imagery has been incredibly complex to access, due to a collaborative effort by several organizations, all current and future Landsat 8 imagery will now be available for any public user through a new Amazon Web Services portal. As the multitude of stakeholders continue to sort out the complex licensing issues, the demand for user friendly platforms to both access and process this imagery will increase.
Highlighting Strong Use Cases
At The Rockefeller Foundation, earth observation imagery is increasingly becoming an important component of our grantee’s work, especially within our resilience portfolio. Although analytical challenges remain outstanding, the promise of these applications make them worth pursuing.
- Grameen Foundation – Working in the Philippines to geospatially register smallholder farmers as they enter into an insurance market using a combination of mobile services and community members as agents who personalize registrations. Planet Labs is partnering with them to explore how detecting health and storm damage to crops on a timely basis using Planet imagery, combined with robust ground data collection, could be used to more accurately model risk assessments for insurance providers.
- 100 Resilient Cities – As rapid changes in regional and local climate affect Boulder, Colorado, the city seeks to understand the nature of its urban forest canopy. Collaborating with platform partners Digital Globe and Trimble, the city is using high resolution satellite imagery to feed as data into Trimble’s e-Cognition software. This first ever forest canopy inventory will then provide the necessary baseline for the development of an Urban Forest master plan. Additional benefits may accrue from multi-spectral imagery that will assess individual tree health, providing a pathway towards active management strategies.
2015 has been a milestone year in terms of mobilizing the data revolution for sustainable development with us here at The Rockefeller Foundation active participants in this series of global policy dialogues. As the humanitarian and development community gain more clarity on how these new resources can add value to their work, we look forward to partnering with both industry and practitioners to break down siloes so that our eyes from above can improve the well being of humanity here on the ground.