This innovative method differentiates between large-scale plantations, small woodlots, and newly established stands as young as three years old. The technique demonstrates how remote sensing technology, combined with AI, can track forest development and update essential growth data.

Dr. Grant Pearse, a Senior Lecturer in Remote Sensing and Geographic Information Systems (GIS) at Flinders University, emphasizes that this approach holds great potential for Australia, where radiata pine is also widely cultivated.

"In New Zealand, where radiata pine plantations dominate the forestry industry, current national forest inventories lack detailed spatial information and often fail to capture data on smaller forests," explains Dr. Pearse, from Flinders University's College of Science and Engineering.

"Our study integrates deep learning techniques for forest mapping with high-resolution aerial imagery and regional airborne laser scanning data to map all planted forests and estimate critical attributes."

The new spatially explicit forest mapping system offers comprehensive data on forest extent, age, and volume. This wealth of information supports more precise forecasting of wood supply, harvest planning, and infrastructure development-applications that are also highly relevant to forestry industries worldwide.

Collaborating with New Zealand forestry researchers from Rotorua, Christchurch, and Auckland, the study focused on planted forests in the Gisborne region. This area, which benefits from publicly accessible aerial and airborne laser scanning data, suffered significant damage from Cyclone Gabrielle in early 2023, resulting in landslides and extensive forest debris.

For such at-risk landscapes, understanding the precise location, age, and condition of forests is crucial for managing the risks associated with harvesting on steep terrain.

"We advocate for the integration of satellite-based harvest detection and digital photogrammetry to maintain continuous updates of forest inventories. This approach allows near real-time monitoring of planted forests at all scales and can be adapted to regions with similar data availability," the researchers highlight in their latest publication.

Beyond its economic significance-NZ's 1.8 million hectares of radiata pine plantations play a crucial role in export timber and fibre industries-these forests are also integral to the country's emissions trading scheme. They are expected to be key contributors to achieving New Zealand's net-zero emissions target by 2050.

The AI-generated forest mapping data can be accessed here

Research Report:Developing a forest description from remote sensing: Insights from New Zealand

Related Links
Flinders University
Forestry News - Global and Local News, Science and Application