Google Earth Engine, an amazing tool for scientists and researchers, released at COP 16

Yesterday at the International Climate Change Conference, Google released a mountain of data for scientists around the world to analyze and use. Known as “Google Earth Engine“, it gives researchers access to this huge pile of data, along with computing power to deal with all of it.

The data includes Landsat satellite data and “trillions of scientific measurements” covering the past 25 years.

So what can be done with all of this data? Thanks to some launch partners, there are already a handful of projects underway. For example, below is a map showing the loss of forest cover in the Congo for the past 10 years, taken from the Earth Engine Map Gallery.

congo.jpg

To go along with this, Google is also donating 10 million CPU-hours for each of the next two years, to help developing nations track the state of their forests. This is in anticipation for REDD (Reducing Emissions from Deforestation and Forest Degradation in Developing Countries), which offers financial incentives for tropical nations that protect their forests.

Do you plan to use Google Earth Engine? If not, what would you use it for if you had the chance?

Introducing Google Earth Engine

[Cross-posted from the Google.org blog]

Today, we launched a new Google Labs product called Google Earth Engine at the International Climate Change Conference in sunny Cancun, Mexico. Google Earth Engine is a new technology platform that puts an unprecedented amount of satellite imagery and data—current and historical—online for the first time. It enables global-scale monitoring and measurement of changes in the earth’s environment. The platform will enable scientists to use our extensive computing infrastructure—the Google “cloud”—to analyze this imagery. Last year, we demonstrated an early prototype. Since then, we have developed the platform, and are excited now to offer scientists around the world access to Earth Engine to implement their applications.

Why is this important? The images of our planet from space contain a wealth of information, ready to be extracted and applied to many societal challenges. Scientific analysis can transform these images from a mere set of pixels into useful information—such as the locations and extent of global forests, detecting how our forests are changing over time, directing resources for disaster response or water resource mapping.

The challenge has been to cope with the massive scale of satellite imagery archives, and the computational resources required for their analysis. As a result, many of these images have never been seen, much less analyzed. Now, scientists will be able to build applications to mine this treasure trove of data on Google Earth Engine, providing several advantages:

  • Landsat satellite data archives over the last 25 years for most of the developing world available online, ready to be used together with other datasets including MODIS. And we will soon offer a complete global archive of Landsat.
  • Reduced time to do analyses, using Google’s computing infrastructure. By running analyses across thousands of computers, for example, unthinkable tasks are now possible for the first time.
  • New features that will make analysis easier, such as tools that pre-process the images to remove clouds and haze.
  • Collaboration and standardization by creating a common platform for global data analysis.

Google Earth Engine can be used for a wide range of applications—from mapping water resources to ecosystem services to deforestation. It’s part of our broader effort at Google to build a more sustainable future. We’re particularly excited about an initial use of Google Earth Engine to support development of systems to monitor, report and verify (MRV) efforts to stop global deforestation.

Deforestation releases a significant amount of carbon into the atmosphere, accounting for 12-18% of annual greenhouse gas emissions. The world loses 32 million acres of tropical forests every year, an area the size of Greece. The United Nations has proposed a framework known as REDD (Reducing Emissions from Deforestation and Forest Degradation in Developing Countries) that would provide financial incentives to tropical nations to protect their forests. Reaching an agreement on early development of REDD is a key agenda item here in Cancun.

Today, we announced that we are donating 10 million CPU-hours a year over the next two years on the Google Earth Engine platform, to strengthen the capacity of developing world nations to track the state of their forests, in preparation for REDD. For the least developed nations, Google Earth Engine will provide critical access to terabytes of data, a growing set of analytical tools and our high-performance processing capabilities. We believe Google Earth Engine will bring transparency and more certainty to global efforts to stop deforestation.

Over the past two years, we’ve been working with several top scientists to fully develop this platform and integrate their desktop software to work online with the data available in Google Earth Engine. Those scientists—Greg Asner of the Carnegie Institution for Science, Carlos Souza of Imazon and Matt Hansen of the Geographic Information Science Center at South Dakota State University—are at the cutting edge of forest monitoring in support of climate science.

In collaboration with Matt Hansen and CONAFOR, Mexico’s National Forestry Commission, we’ve produced a forest cover and water map of Mexico. This is the finest-scale forest map produced of Mexico to date. The map required 15,000 hours of computation, but was completed in less than a day on Google Earth Engine, using 1,000 computers over more than 53,000 Landsat scenes (1984-2010). CONAFOR provided National Forest Inventory ground-sampled data to calibrate and validate the algorithm.