AI and LiDAR Join Forces in Space to Map Forests for Climate Action

In the battle against climate change, an unexpected partnership has formed between cutting-edge space technologies and artificial intelligence (AI), creating new pathways in climate science. By harnessing AI with satellite LiDAR imagery from trailblazers like NASA and the European Space Agency (ESA), researchers can now quickly and accurately map forest carbon, providing vital insights into global carbon cycles.

Traditionally, estimating forest biomass has been a cumbersome process—deeply reliant on ground-based methods that are both time-consuming and geographically limited. However, inspired by technologies that have helped archaeologists uncover hidden ruins, a new, innovative method now applies satellite tools to the challenge of mapping forest carbon, achieving unparalleled speed and precision. Leading these advancements is Hamdi Zurqani from the University of Arkansas, who emphasizes the significant role forests play in climate regulation, noting their storage of approximately 80% of Earth’s terrestrial carbon.

At the core of this method is NASA’s GEDI LiDAR system, which operates from the International Space Station using three lasers to capture 3D data of forest canopies, such as canopy height and surface elevation. By merging this data with imagery from ESA’s Sentinel satellites, Zurqani has notably enhanced the precision of biomass estimations. Among various machine learning algorithms tested, gradient tree boosting emerged as the most accurate tool, though other models like random forest, Classification and Regression Trees (CART), and support vector machines also offered valuable support.

Zurqani underscores the vast importance of accurate forest biomass mapping for both carbon accounting and forest management. Precise carbon storage and emission measurements are crucial for governments and conservation groups dedicated to climate policy and action plans.

Yet, challenges remain. Variables such as weather conditions can impact satellite data collection, and many regions still lack high-resolution LiDAR coverage. Looking forward, future advancements aim to address these challenges by employing even more advanced AI models, potentially incorporating sophisticated neural networks.

Key Takeaways:

• AI, paired with satellite LiDAR technology, has dramatically improved the speed and accuracy of forest carbon mapping.
• This technological advancement is vital for carbon tracking and forest management, supporting global efforts to combat climate change.
• Gradient tree boosting was identified as the most effective model, benefiting from the strategic integration of NASA and ESA satellite data.
• Although obstacles persist, ongoing enhancements in AI technology promise sustained improvements, ensuring its crucial role in global environmental conservation efforts.