The second bi-annual UN World Data Forum took place in Dubai a couple of weeks ago. It was a three-day gathering of over 2,000 people, including statisticians, nonprofits, private companies, foundations, UN and governmental representatives. The focus of the conference was to share ideas about the use of data in support of Sustainable Development Goals (SDGs).
The SDGs are a collection of 17 goals set by the United Nations General Assembly in 2015 and are a “blueprint to achieve a better and more sustainable future for all.” They include universally agreeable, if ambitious in number and scope, goals that include eliminating poverty, achieving good health and well-being, and protecting life on land. But how will we know when we have achieved the “protection of life on land?” Each SDG has several targets. And each target has between 1 – 3 indicators. Find plenty of detail here.
Data will be used to assess the indicators to measure success towards the SDGs. Collaboration between countries for monitoring SDGs is clearly beneficial because every country has the same set of SDGs to measure. Some indicators are more challenging to measure than others. Each requires different tools and expertise. Some countries have the resources and systems in place that makes additional data collection possible, some do not.
One of the exciting parts for me is understanding how geospatial technology can be applied to measure indicators. Azavea was invited by the UK Office of National Statistics to attend the Forum because of our contribution to the UN Data Platform, a project of the UN Global Working Group on Big Data. It brings together tools through public-private partnerships that will enable data analysis and the production of official statistics in support of SDG monitoring. Azavea was invited to contribute to the platform based on our expertise in cloud-based imagery processing through GeoTrellis and Raster Foundry.
I had never given much thought to the generation of official statistics prior to a month ago. But an improved ability to measure progress on SDG indicators means we can improve our understanding of what is working where and better allocate resources to where they are needed. There is a lot of potential for our work at Azavea to help make meaningful contributions in this space.
So what is the point of the forum? To encourage dialogue and connections between people from countries around the world that are collecting and analyzing similar datasets for similar purposes.
The program was packed with presentations and panels of expert practitioners that shared their lessons learned for one specific dimension of data and SDGs. There were presentations (here and here) that taught me about the world of National Statistics Offices. I realized it’s a much larger community than I could have anticipated after speaking with a Statistician from Albania’s Institute of Statistics and he said they were a small office of only 250 people. There were also invaluable presentations on data finance and questions around trust in data.
The presentations I was most interested in were those that discussed how geospatial data, and particularly earth imagery, are being used to measure SDG success. These two (here and here) in particular stood out.
There is an abundance of Earth Imagery data. Parsing the signal in the noise is a constant challenge that will only continue to grow as the rate of data generation grows. The ability to use the data requires tools that can bring together different datasets and provide the analytical tools to query the data in meaningful ways.
Learn how to access the best satellite imagery for your project in our latest white paper.
From the talks, I found the most direct link between an indicator and earth imagery data from Brian Sullivan, Senior Program Manager for Google’s “Geo For Good” program. He showed a beta release of a tool built on Google Earth Engine that will help respond directly to SDG 6.1.1. They have analyzed 30 years of Landsat scenes from the Google Earth Data repository and created a time series water classification for every pixel on the planet.
Resource Watch is another good example of a platform for data synthesis. It is the result of collaboration between World Resources Institute (WRI) and over 30 partners to collate global datasets to help visualize and provide access to these datasets in a way that will facilitate their utilization.
There was some good discussion about using Earth Imagery for reporting on indicators, but I heard less resounding success stories than I might have hoped. I don’t think this is a failing of the community, but rather, I think it remains a significant technical and logistical challenge of global proportions. The first challenge is simply processing data at global scales. Then there are challenges of ground truth quality control – tackling geographical heterogeneity and generating realistic accuracy measures.
People in this space are already talking about the application of Earth Imagery for SDGs. This CEOS report claims that Satellites have the potential to support national reporting against a quarter of the associated SDG targets, which is fairly substantial. The GeoGroup produced this report, but even better, generated this diagram to show where Earth Observation Data can support SDGs:
It’s a little overwhelming, which in its own sense is encouraging – there are numerous ways that Earth Imagery can support SDGs. But for brevity’s sake, I’ll just focus on two: tracking forests (SDG 13) and tracking urbanization (SDG 11). Each comes with a set of challenges. But the development of a methodology that would support national reporting for every country on earth would be a significant accomplishment.
The first target for the “Life on land” SDG aims for total conservation, restoration, and sustainable use of ecosystems. By 2020. That’s certainly ambitious, and measuring the land cover percentages of various ecosystem types is not easy, but there has been significant work done in this area. The first indicator for this goal is measuring the forest area as a proportion of total land area. This does seem straightforward by comparison and is ideally suited to satellite imagery analysis. Initially launched in 1997, WRI’s Global Forest Watch is the premier example of what is possible with decades of focus on this problem space and how technology can change the scale of what is possible.
Target 3 for the “Sustainable cities and communities” SDG is to enhance sustainable urbanization in all countries by 2030. The main indicator for this target is the ratio of land consumption rate to the population growth rate. The goal here is to monitor the impact of sprawling population growth. There are many challenges associated with measuring and defining urbanization. First of all, experts cannot agree on a definition of urban extent that applies universally, nor can they identify a method that should be used for coupling urban extent maps with population data.
There are also technical challenges to measuring the impact of population growth. A global and robust method for providing accurate mapping results in all conditions and for all types of human settlements does not yet exist. Even more challenging is the need to process the vast amount of satellite data and derive temporal statistics required by the urban extent classifiers. Such high computing and storage needs can only be achieved by running the algorithms on high-performance computing platforms.
Subjective questions like what is urban extent and how do we couple this data are more challenging to answer. However, once those are in place, the technical challenges of processing vast amounts of data are challenges that have been solved. It’s exciting to me that Azavea is well positioned to provide the expertise required to overcome these technical challenges.
It was incredibly invigorating to attend the UN Forum. I learned about a whole new development space to which Earth Imagery can be applied. I was surrounded by experts in the room who are tackling the world’s biggest needs with cutting edge technology. And I met people who are thinking about problems at a global scale with technical challenges that we at Azavea are uniquely poised to help address. I am grateful to the UK Office of National Statistics for their drive behind the creation of the UN Global Data Platform. I think it has the potential to support National Statistics offices across the world to utilize Big Data to monitor SDGs. I am excited to continue thinking about Earth Imagery’s application to SDGs and I hope Azavea can join in on the good work being done.