NOAA and Google Cloud: A data match made in the cloud
Director, NOAA Big Data Program
Business Director, NOAA Big Data Program
With Valentine’s Day upon us, there is nothing the U.S. National Oceanic and Atmospheric Administration (NOAA) loves more than having our environmental data open and accessible to all—and the cloud is the perfect match for NOAA's goal to disseminate its environmental data more broadly than ever before.
In 2019, as part of the Google Cloud Public Datasets Program and NOAA's Big Data Program, NOAA and Google signed a contract with the potential to span 10 years, so we could continue our partnership and expand our efforts to provide timely, open, equitable, and useful public access to NOAA’s unique, high-quality environmental information.
Democratizing data analysis and access for everyone
NOAA sits on a treasure trove of environmental information, gathering and distributing scientific data about everything from the ocean to the sun. Our mission includes understanding and predicting changes in climate, weather, oceans, and coasts to help conserve and manage ecosystems and natural resources. But like many federal agencies, we struggle with data discoverability and adopting emerging technologies. The reality is that on our own, it would be difficult to share our massive volumes of data at the rate people want it.
Partnering up with cloud service providers such as Google and migrating to cloud platforms like Google Cloud lets people access our datasets without driving up costs or increasing the risks that come with using federal data access services. It also unlocks other powerful processing technologies like BigQuery and Google Cloud Storage that enhance data analysis and improve accessibility.
Google Cloud and other cloud-based platforms help us achieve our vision of making our data free and open and also aligns well with the overall agenda of the U.S. Government. The Foundations for Evidence-Based Policy Making Act, signed in January 2019, generally requires U.S. Government data to be open and available to the public. Working with cloud service providers such as Google Cloud helps NOAA democratize access to NOAA data—it’s truly a level playing field. Everyone has the same access in the cloud, and it puts the power of data in the hands of many, rather than a select few.
Another critical benefit of data dissemination public-private partnerships, like our relationship with Google Cloud, is their ability to jumpstart the economy and promote innovation. In the past, the bar for an entrepreneur to enter a market like the private weather industry was extremely high. You needed to be able to build and maintain your own systems and infrastructure, which limited entry to larger organizations with the right resources and connections available to them.
Today, to access our data on Google Cloud, all you need is a laptop and a Google account to get started. You can spin up your own HPC cluster on Google Cloud, run your model, and put it out into the marketplace without being burdened with the long-term maintenance. As a result, we see small businesses being able to leverage our data and operate in areas where previously they simply didn’t exist.
Public-private data partnerships at the heart of innovation
NOAA's datasets have contributed to a number of innovative use cases that highlight the benefits of public-private data partnerships. Here are some projects to date:
Acoustic detection of humpback whales
Using over 15 years of underwater audio recordings from the Pacific Islands Fisheries Science Center of NOAA, Google helped develop algorithms to identify humpback whale calls. Historically, passive acoustic monitoring to identify whales was done manually by somebody sitting with a pair of headphones on all day, but using audio event analysis helped automate these tasks—and moved conservation goals forward by decades. Researchers now have new techniques at their disposal that help them automatically identify the presence of humpback whales so they can mitigate anthropogenic impacts on whales, such as ship traffic and other offshore activities. Our National Centers for Environmental Information established an archive of the full collection of multi-year acoustic data, which is now hosted on Google Cloud as a public dataset.
Megaptera novaeangliae, the humpback whale, and a spectrogram of its call, one of the audio events found in the dataset, with time on the x-axis and frequency on the y-axis.
Weather forecasting for fire detection
One of the most important aspects of our mission is the protection of life—and the cloud and other advanced technologies are driving the discovery of new potential life-saving capabilities that keep people informed and safe. NOAA’s GOES-16 satellite and GOES-17 satellite provide critical datasets that help detect fires, identify their locations, and track their movements in near real-time. Combining our data and Google Earth Engine’s data analysis capabilities, Google recently introduced a new wildfire boundary map to provide deeper insights for areas impacted by ongoing wildfires.
Using data from NOAA's GOES satellites and Google Earth Engine, Google creates a digital polygon to represent the approximate wildfire impact area on Search and Google Maps.
Start exploring and experimenting with NOAA’s datasets, including those found on Google Cloud Public Datasets. If you’re already using our public datasets, we’d love to hear from you. What data are you using and how? What are you looking forward to using the most?