An easier way to browse, discover, and use Earth Engine datasets
By Chris Herwig, Geo Data Engineer, Google Earth Engine
Today, the Earth Engine public data catalog contains nearly 600 datasets from more than 50 different data providers from across the world, including governments, intergovernmental organizations, nonprofits, and businesses. These datasets are based on observations from over 30 satellites/instruments (including the Landsat program, PROBA-V, Cryosat-2, TOMS, OMI, GRACE, Sentinel constellation, Hyperion, ALOS PALSAR, AVHRR, TRMM/GPM/GSMAP).
In total, the data catalog size now exceeds 20 petabytes (that’s 20 quadrillion bytes, or almost double the size of all the photos uploaded to Google Photos in its first year). That’s a lot of information, but luckily, the Earth Engine Data Catalog is now a little easier to navigate.
We know how important it is for scientists, journalists, and researchers to have access to the right datasets for their analyses in Earth Engine (in fact, user demand is the main way we prioritize ingestion of new datasets). As we have increased the volume and variety of datasets in our data catalog, we have also focused on improving the user experience of interacting with the catalog along two primary dimensions — finding new datasets and actually using the new datasets. The result of these efforts is the new Earth Engine Data Catalog, built right into our existing developer documentation.
Usability
Getting started with new datasets can be tough, so we focused on reducing the cost of learning to use a new dataset. For each dataset, we generated ready-to-use code samples to get you up and running quickly — with literally no coding required. Just click the “Open in Code Editor” button below the code snippets at the bottom of each dataset page, and the script will open in the Earth Engine Code Editor. Click “Run” to dive into the dataset.
Dataset discoverability
The new data catalog makes it easy to find what datasets are available in Earth Engine (and to discover new ones you may not have known about before). Previously, this was somewhat challenging, as the data catalog information was spread out across several places. Now, Earth Engine datasets have a central, publicly accessible location. You can also browse by dataset tags, like climate, Landsat-derived, weather, or JAXA.
For some tags (for example, those representing Landsat raw and derived products), we have a lot of datasets. For Landsat, for example, there are over 40 datasets in Earth Engine, so we decided to take a more curated approach, which you can see in the dedicated Landsat, Sentinel, and MODIS pages.
Google Dataset Search
You can also discover new datasets in Earth Engine using Google Dataset Search, a new tool launched today that we’re all pretty excited about. Similar to how Google Scholar works, Dataset Search lets you find datasets wherever they’re hosted, whether it’s on a publisher’s site, in a digital library, or on the author’s personal web page. Dataset Search relies on an open standard for describing datasets (schema.org) and anybody who publishes data on the web can describe their dataset in this way. You can read more about Google Dataset Search here.
We hope today’s changes to our data catalog save you a few clicks and keystrokes and make your Earth Engine workflow more pleasant and productive! Check out the new data catalog here.
