Source: Flinders University
New Zealand and Flinders University experts have deployed artificial intelligence and 3D laser scanning to accurately map planted pine (radiata) forests for most of NZ’s North Island.
The results, which distinguish planted large estates, small woodlots and newly established stands as young as three years old, showcase a new way of using remote sensing with other technology to reveal forest growth and update growth information.
This approach is just as relevant for Australia, where radiata pine is also widely grown, says Dr Grant Pearse, Senior Lecturer in Remote Sensing and Geographic Information Systems (GIS) at Flinders University.
“In New Zealand, where radiata pine plantations dominate the forestry sector, the current national forest description lacks spatially explicit information and struggles to capture data on small-scale forests,” says Dr Pearse, from the College of Science and Engineering at Flinders University in Adelaide, South Australia.
“We combined deep learning-based forest mapping using high-resolution aerial imagery with regional airborne laser scanning data to map all planted forest and estimate key attributes.”
The spatially explicit forest description provides wall-to-wall information on forest extent, age, and volume for all sizes of forest. This facilitates stratification by key variables for wood supply forecasting, harvest planning, and infrastructure investment decisions – applications equally valuable for other forestry industries.
The research, with New Zealand timber industry researchers from Rotorua, Christchurch and Auckland, was carried out on planted forests in the Gisborne region, which has publicly available aerial imagery and airborne laser scanning data.
This region is particularly significant as it was severely impacted by Cyclone Gabrielle in early 2023, which caused widespread landslides and forest debris flows.
For such vulnerable terrain, knowing exactly where forests are located in the landscape, their age and condition is key to managing the risks of harvesting operations on the region’s steep slopes.
“We propose satellite-based harvest detection and digital photogrammetry to continuously update the initial forest description. This methodology enables near real-time monitoring of planted forests at all scales and is adaptable to other regions with similar data availability,” researchers say in a new article.
Along with the economic importance of NZ’s 1.8 million hectares of radiata pine forestry for export timber and fibre, these planted forests are a key part of the country’s emission trading scheme and are expected to play a significant role in achieving the government’s target of net-zero emissions by 2050.
The forest map derived from artificial intelligence can be viewed at: www.forestinsights.nz
In South Australia, plantation estates covering about 40,000 hectares support a $3 billion industry and employ 18,000 people as well as construction, manufacturing, tourism and regional communities.
The article. ‘Developing a forest description from remote sensing: Insights from New Zealand’ (2024) byGrant D Pearse (Flinders University), Sadeepa Jayathunga, Nicolò Camarretta, Melanie E Palmer, Benjamin SC Steer, Michael S Watt (all Scion), Pete Watt and Andrew Holdaway (both Indufor Asia Pacific) has been published in the journal Science of Remote Sensing. DOI: 10.1016/j.srs.2024.100183. (ref. https://www.forestinsights.nz/ )
Acknowledgements: This project was funded through the Ministry of Business, Innovation and Employment (MBIE) Strategic Science Investment Fund (administered by Scion, the New Zealand Forest Research Institute Ltd) and through the MBIE Programme (grant number C04X2101).