• News

Forest AI – measuring levels of carbon capture around the world

*Photo credit: FAO/Marlondag

A scientist from Indiana’s Purdue University is working with a large team of global collaborators to develop an artificial intelligence (AI) model that will combine information collected about billions of trees measured on-site in the world’s forests using satellite and other geospatial data.

This information will then be used to map carbon accumulation rates across the globe.

Associate Professor Jingjing Liang has been enabled to conduct this research thanks to a two-year, US$870,000 grant from the World Resources Institute – a non-profit research organisation based in Washington D.C. He is Associate Professor of quantitative forest ecology and co-director of the Forest Advanced Computing and Artificial Intelligence Lab at Purdue.

“To accurately capture the carbon accumulation rates of forested ecosystems across the world has always been a challenging task,” Associate Professor Liang said.

These challenges are the result of the difficulties around capturing the sheer volume of ground-sourced data required to complete this type of project, and the limited amount of such data currently available to the scientific community.

According to Nancy Harris, Research Director of the Land and Carbon Lab at the World Resources Institute, this is a much more difficult undertaking than the mapping of carbon emissions resulting from forest losses.

“With emissions, there’s a clear signal in satellite imagery when trees are cut, leading to a big drop in forest carbon stocks and a relatively abrupt pulse of emissions to the atmosphere,” Dr Harris said.

Conversely, when it comes to carbon sequestration, forests and trees accumulate it gradually and in a non-linear way, meaning even the most advanced ground-based sensors are unable to reliably capture the data required to make estimations.

Carbon accumulation rates in forests are therefore estimated using three key measures:

Ingrowth – the number of seedlings that have reached the threshold size required to be considered as trees

Upgrowth – the increase in tree diameters which is aided through photosynthesis

Mortality – the number of trees that reach the end of their lives.

Ground-sourced data measured in an ongoing capacity at various intervals over time is the only truly reliable means the industry has at its disposal to collect accurate information relating to these three factors.

As a result, rates of ingrowth, upgrowth and mortality in individual forest stands have never been able to be estimated on a global scale, meaning significant uncertainties around the amounts and trends of carbon sequestration at different locations around the world.

To overcome this knowledge gap, Associate Professor Liang is developing an AI model to combine satellite and other geospatial data relating to billions of trees from forest growers around the world, meaning the required local forest information needed to estimate carbon sink can be captured on a global scale.

While being the first AI-based forest growth model to be deployed at a global scale, and thus able to accurately quantify levels of carbon capture, the model will also be able to gather information on factors such as biodiversity and timber quality.

“The spatially granular data this new project will provide will help us better understand the role our planet’s forests play in local, nature-based solutions to mitigate global climate change,” Dr Harris said.

While Purdue’s state-of-the-art, high-performance computing clusters will play a crucial role in enabling this work, achieving comprehensive global coverage of ground-sourced plot data requires collaboration from other collectors of data based around the world.

“Through the newly established network of Science-I and its sister consortium, the Global Forest Biodiversity Initiative, we already have working relationships with a large number of scientists across the world who are collecting and sharing those data,” Associate Professor Laing said.

Science-I, which was founded by Associate Professor Liang, is a web-based collaboration platform involving more than 300 scientists across the world.

The Global Forest Biodiversity Initiative, also founded by Associate Professor Laing, has a database of 1.3 million sample plots and 55 million trees and, as such, will provide the basis for this new project.

“We are going to collect much more data, especially from the global south, to fill those data gaps. We will get more people involved, especially those from underrepresented groups,” he said.

Project collaborators already involved include representatives from Indigenous groups across North America, Amazonia and Africa, to name just a few. Rural communities, members of the world’s forestry industries and an army of citizen scientists will also play a significant role in providing data.

“In Science-i, everyone collaborates with each other as equal partners on all projects,” he said.

“We openly share our findings with transparent real-time discussions across the whole team. Then we cross-evaluate and consolidate our research findings at the end. This is a brand-new way to do international collaborative forest research.”

Source: Purdue University

Posted Date: July 17, 2023

Related Resources