Google Earth Engine revolutionized the way we monitor deforestation

The scale of nature can be very difficult to grasp. The largest rainforest on the planet, for example, the Amazon spans some 2.6 million square miles, crossing the borders of nine countries. It is more than twice the size of India and equal to 68 percent of the total area of ​​the United States.

Trying to protect something so large is a challenge of equal size and complexity, but it was made easier by Google’s Earth Engine, a geospatial processing service directed to researchers and other public policy experts. Launched in 2010, it completely changed the way forest loss is tracked around the world by making the process much faster and comprehensive – and it all started with one person’s curiosity.

In 2005, Google engineer Rebecca Moore found out about a new logging project near her home in the Santa Cruz Mountains of northern California. Although the plan for the project was shared by the developers with their community, it was hardly informative or transparent.

“It was a 400-page document with a sketchy black-and-white map that no one could understand. One could not even tell where it was going to happen, much less the details of what was being proposed,” she says.

A self-described map geek, he decided to use Google Earth, a global visualization tool released a few years earlier, to visualize the targeted 1,000-acre area. The high-resolution satellite images showed what the 400 pages did not: the project would not only affect the daily lives of the community’s 2,000 people but also residents of the wider Silicon Valley community, who get their drinking water from that area.

The area was made up of coastal redwood, an endangered sequoia species that contains some of the tallest living trees and the longest-lived on Earth.

“It turned out that helicopters were going to be landing and taking off and hauling logs over the nursery school and daycare center,” she recalls. “It was going to go right within 100 meters of three schools.”

With a flyover video of less than two minutes, she was able to mobilize the community and ultimately stop the logging project. Moore, who is now director of Google Earth, Earth Engine and Outreach, says, “Looking at the scale of the destruction, seeing where the deforestation is happening, gives you a sense of the reasons, which areas are more at risk.” are in need of more protection.”

Although that was a relatively small area and only dealt with a specific moment in time, using Google Earth to help stop a logging project got Moore the idea that something similar could be monitored for the world’s forests. . But for this to work, researchers would need access to a massive database and tools to process all this data—which was the seed of Google Earth Engine.

Until then, maps and satellite images will be stored in various institutions, such as NASA and other government agencies, and scientists will have to collect this information and download it to their computers before they can begin analysis. It was timely, taking weeks to assemble, and expensive due to the hardware required.

“The Google Earth engine changes this paradigm,” explains Gilberto Camara, computer scientist and former director of Brazil’s National Space Research Institute (INPE). “Instead of the data going where the software is, which is your computer, the software goes where the data is, which is Google.”

The tech giant collects free-access satellite imagery from research institutions around the world, from the US to Europe and Brazil, uploads it to its cloud service, and makes it available to academics with programming functions to process the data . “It clearly benefits greatly from Google’s ability to become a big data warehouse,” Camara says.

“We’re using Google data centers to bring in thousands and thousands of images every day from hundreds of satellites and other data sources and store them in a form that’s ready for analysis,” says Moore.

The first target of all that machine power was the world’s largest tropical forest. At the Amazon Institute of People and the Environment (Imazon), a non-profit organization, the first operational forest monitoring system was built on Earth Engine. It was integrated into its deforestation warning system to report the pace of forest degradation and deforestation in the Brazilian Amazon on a monthly basis.

After that work a successful analysis was done. published in Science in 2013It was the first high-resolution map of the world’s forests and how they had changed over the past 12 years.

“It had never been done before, and the reason was this: It required over a million hours of computation. But because we ran it on 10,000 computers in parallel, we got the result in a matter of days. A single computer But it would have taken 15 years,” explains Moore.

Energy is a big deal when talking about climate change because the world is still heavily dependent on fossil fuels. However, deforestation is also a major driver of global warming – not only because it reduces the amount of carbon storage available but also because of the carbon emitted by forest destruction. This is especially important in the case of Brazil.

According to latest official dataThe Brazilian Amazon lost more than 4,466 square miles of forest coverage from August 2021 to July 2022 (the number of deforestation is measured taking into account the region’s rainy season, which ends in July). To get a better sense of the scale of the destruction, which is roughly equivalent to 15 times the area of ​​New York City.

The destruction of the Amazon rainforest affects not only the region and the communities living there, but the entire world. It helps regulate the global climate by storing carbon and releasing billions of tons of water into the atmosphere. Protecting its biodiversity also helps prevent new disease outbreaks and potential sources of drugs.

To try to roll back the high rate of deforestation, The Brazilian government recently released a new plan This ranges from fighting environmental crime to promoting bio-economic activities that generate income for local communities while keeping the forest standing.

Latin America’s largest country, Brazil, is the seventh-largest annual emitter of greenhouse gases, accounting for 3 percent of the world total, according to 2019 data compiled by the World Resources Institute. The top emitters are China (25.2 percent), the United States (12 percent) and India (7 percent).

while around 71 percent of the world’s electricity comes from non-renewable sourcesthat number falls to 17 percent in Brazil, Most of the country’s carbon emissions come from deforestation and land-use change, accounting for about 49 percent of their emissions in 2021.

This figure is generated every year by System of Greenhouse Gas Emissions and Removal Estimations, a non-governmental initiative. The system has a sister platform called mapbiomass (“MapBiomes,” in a literal translation), which tracks deforestation, fire, mining, surface water, infrastructure, and pasture quality across the country.

Created by a network of scientists and research institutions, MapBiomas was first launched in 2015 at the United Nations Climate Conference in Paris.

“We need to make a map every year to see whether public policies and recommendations have affected emissions and removals of greenhouse gases,” says Marcos Rosa, MapBiomass’ technical coordinator.

It was difficult because it required a huge dataset (daily, high-resolution photos of the entire country), lots of processing power and machine learning (to recognize and classify changes on satellite images of the Amazon and five other biomes in Brazil). The demand was made. ,

This is where Google Earth Engine came into play. “It was essential. The Google Earth engine made it possible[for researchers]to work remotely in each biome without having to buy a huge computational infrastructure,” says Rosa.

Such monitoring was taken to a worldwide scale by the World Resources Institute’s Global Forest Watch (GFW), launched in 2014. Mikaela Weiss, the platform’s director, points out that the institute has been involved in providing transparent forest information since the 1990s, but, until then, did not have the technology to globalize it.

She says, “Many advances have made this possible – including free access to satellite imagery and better internet access around the world – but a significant part of it has come from cloud computing and the power to analyze satellite images using the Google Earth Engine.” Is.”

Some of his colleagues are researchers at the University of Maryland and the University of Wageningen, who use Google’s system to create deforestation alerts in near-real time in order to detect forest changes as soon as possible to prompt a response.

Still, Weisse sees some gaps in the available data. “Tropical regions are particularly difficult to monitor using satellite imagery because they are often cloudy. There is also less historical and on-the-ground information available in those locations,” she says, calling on public and public to address the issue. There are efforts of both private organizations.

Although the scientists we talked to ledge While not expressing particular concern about it, Google is a huge company that, naturally, has economic interests of its own – which won’t always align with environmental issues. In the case of Earth Engine, issues may arise regarding the reliability of the data being hypothetically promoted or the company’s interest in keeping the platform running, for example.

Google’s Rebecca Moore emphasizes that all images she provides are in the public domain. “This is data coming from government satellites, for example, in the US, Brazil, Europe, Japan,” she says. “And all the data is scientifically peer-reviewed.”

“I don’t see a conflict of interest, which would mean any impact on data generated by independent researchers with the engine,” says Weiss. “The biggest risk is that if Google decides it is not in their economic interest to continue offering this service – it would be a great loss to the research community.”

Even though Google’s service was the first of its kind, today, there are alternatives like Microsoft Planetary Computing and Amazon’s Earth on AWS. “Many countries have also started efforts to create their own Google-like platforms,” ​​says INPE’s Gilberto Camara.

He claims that Google has democratized access to information, but believes that Earth Engine has not evolved much since 2010.

Mapbiomass’s Marcos Rosa says that, despite having a partnership with Google for digital storage space, the company has never tried to meddle with the content they produce.

“We’ve always had a choice about the data we generate with Google Earth Engine. So, if there is a change in Google’s policy – which has never happened – we can replicate[the model]on another platform like Amazon.