As climate-related disasters become more frequent, the need for actionable climate intelligence has never been greater. Earth observation technologies offer critical insights into our rapidly changing environment and the interconnected dynamics of Earth’s systems.
By 2032, satellite Earth observation is expected to generate over 2 exabytes (2 billion gigabytes) of data cumulatively. The volume and complexity of this data has historically prevented it from being translated into actionable climate solutions.
Large volumes of data require sophisticated processing and analysis to create insights that can easily be integrated into operational decision-making processes. To turn Earth observation data into even more useful climate and weather information, efficient and effective data processing and analysis is essential.
Advances in satellites, artificial intelligence (AI) and other synergistic technologies are helping Earth observation data to become more accessible and impactful than ever before. A new report published by the World Economic Forum and MIT Media Lab, Charting the Future of Earth Observation: Technology Innovation for Climate Intelligence, examines 10 key Earth observation technology trends unlocking unprecedented climate insights.
1. Advanced sensor technology on satellites
Recent advancements in satellite Earth observation sensors are providing improved global coverage, resolution and accuracy, and a wider range of observable measurements.
Satellites equipped to take “superspectral” imagery can collect more detailed and refined data, producing two to three times the temporal, spatial and spectral resolution of current multispectral imagery. These enhanced spectral resolutions can detect differences in plant health and show wildfire burn severity, allowing for more precise water management and accurate post-disaster recovery efforts.
2. AI, machine learning and deep learning
Sophisticated AI and machine learning (ML) algorithms are accelerating the processing and analysis of Earth observation data. For example, ML-based models trained on existing data can generate estimates up to 1,000 times faster than traditional climate models. This reduces the time it takes to generate weather forecast models such as flood maps by up to 80%.
This also makes it possible to conduct detailed post-disaster assessments in hours or minutes following climate-related events such as hurricanes or floods – traditional models or on-site inspections can take weeks. The unprecedented speed and accuracy offered by the computational efficiency of AI models used with Earth observation data is crucial for timely decision-making.
3. Satellite edge computing
Satellite edge computing processes Earth observation data directly in-orbit on the satellite. This cuts the time it takes to get from data collection to actionable insights. The technology reduces latency and the need for data download, allowing for faster transmission of critical information to emergency responders in disaster scenarios.
4. Miniaturization of Earth observation sensors
Miniaturized sensors, as well as reduced manufacturing and launch costs, have enabled more nations to manufacture and launch their own Earth observation satellites. This increases publicly available Earth observation data.
Advancements in microelectronics and semiconductor technologies have integrated greater processing power into smaller chips. This facilitates data analysis by the sensor hardware itself, rather than relying on heavy, energy-intensive equipment.
5. Larger satellites with advanced capabilities
Alongside miniaturization, there is currently a parallel trend for larger satellites equipped with advanced sensors and enhanced data transmission capabilities. These larger platforms are more reliable, can provide more features and can house larger and more complex instruments.
6. Climate ML-based models
Traditional Earth system models use complex numerical simulations to help us understand climate dynamics and predict future climate scenarios. They are often computationally intensive, consuming up to 10 megawatt hours of energy to simulate a century of climate activity – roughly equivalent to powering a home for a year.
In contrast, climate models that integrate physics-informed ML have the ability to process enormous, petabyte-scale datasets to deliver accurate and fast weather and climate predictions. These ML-based models are particularly effective for localized studies, offering high-resolution forecasts at significantly lower computational costs, with studies showing up to 100 times more energy-efficiency.
7. Geospatial AI foundational models
This technology is designed to detect high-level patterns from large amounts of satellite Earth observation data. Trained on many different datasets in a self-supervised way, geospatial AI models can be used for a wide range of applications and are highly effective at creating accurate models of global patterns.
8. Digital twins
Digital twins are dynamic, digital replicas of Earth systems such as climate, oceans and ecosystems. They enable users to better understand, predict and investigate complex Earth system phenomena.
Digital twin technology allows users to analyse various “what if” climate scenarios. They can use it to visualize and test the potential impacts of different climate-related strategies.
9. AR/VR data immersive platforms
Augmented reality (AR) and virtual reality (VR) platforms offer users an immersive experience. These intuitive platforms are transforming how Earth observation data is accessed and understood by diverse stakeholders and are encouraging data literacy through interactive learning.
10. Data cubes
By organising Earth observation data along various dimensions – spatial, temporal and variable grids – data cubes allow users to extract useful insights and conduct complex analyses of Earth's environment and its changes at different scales and levels of detail.
Data from various Earth observation sources are standardized to a uniform resolution and shared characteristics, which simplifies their use for computations without further transformation. This approach is especially useful when quick access to analysis-ready data is needed.
A path toward proactive climate action
Advancements in technologies used in conjunction with satellite Earth observation data will help us to be more proactive in preventing the effects of climate change. Leading Earth observation data providers, users and experts are working with the World Economic Forum to researching ways to harness Earth observation’s transformative potential for addressing climate and environmental challenges.
These technologies aren’t just about better data, they give access to critical climate insights that empower communities, businesses and policymakers to build resilience against climate change. As these technologies continue to evolve, they will allow us to profoundly shape how we monitor and respond to climate change.
Weforum
Why EU tariffs are unlikely to dent Chinese EV makers’ European expansion
Chinese electric vehicles will remain competitive in Europe despite the EU’s additional tariffs on autos made in the country, particularly after they were revised lower last month.
In the latest tariff revisions at end August, BYD, China’s behemoth automaker, saw tariffs cut to 17% from 17.4%, Geely to 19.3% from 19.9%, and SAIC
saw a reduction to 36.3% from 37.6%.
To make the European market unattractive for Chinese EV exporters, tariffs have to be as high as 50%, according to research group Rhodium. It said that number might need to be even higher for vertically integrated manufacturers such as BYD.
The current tariffs will not be a significant deterrent to China’s EV-makers, said Joseph McCabe, president and CEO of global auto research company AutoForecast Solutions. “Tariffs on Chinese-made EVs will create a hurdle, but not a barrier to entry,” he added.
He pointed out that the EU’s tariffs were not as severe as those announced by North America because European and Chinese original equipment manufacturers are heavily interconnected. The U.S. announced a 100% tariff on Chinese EVs in May this year. Canada followed suit last month.
“It is a delicate balance to promote domestic European production without severely impacting their Chinese operations,” McCabe said.
Chinese EV makers are coming up with newer, cheaper offerings even as the EU strives to curtail imports via tariffs.
At a conference in May this year, Chinese behemoth BYD
announced its Dolphin model to the European market at less than $21,550. The model is a rebrand of the Chinese Seagull model.
In comparison, Western EV-maker Tesla’s
Model 3, the brand’s cheapest offering, is being sold for $44,480 in the United Kingdom. Electric vehicles made by Tesla in China also face a 9% tariff on imports to the EU.
Even with the 17% levy, BYD’s Dolphin model will still be about $23,270 cheaper than the China-imported Tesla Model 3.
To better compete with fierce Chinese rivals, German brand Volkswagen has announced plans to develop a low-cost electric vehicle for the European market at a comparable price of around $21,476 by 2027.
“Now, profitability takes a back seat to market share. The investment community rewards new, innovative EV players on the promise what they could be rather than short-term financial performance that legacy manufacturers are measured,” said McCabe.
“If they really have to kill the EV industry in China, they have to put in 300% of tariffs ... which, you know, doesn’t make sense from my perspective,” William Ma, CIO of GROW Investment Group told CNBC’s “Street Signs Asia” on Tuesday.
If the Chinese original equipment manufacturing sector is affected, the risk of retaliatory tariff measures from China against Europe is high, McCabe warned.
EU tariff talks started in June as a response to “unfair subsidies” to Chinese EV makers, which pose “a threat of economic injury” to European EV counterparts.
“This geopolitical or sanction will not go away easily for the next year or two,” Ma said.
CNBC