Calculating Carbon:
Harder Than It Looks
Decoding the Complexity of Carbon Removal
Getting the Numbers Right
Accurately quantifying this carbon sequestration is challenging due to several factors:
1. Diverse Ecosystems
Natural ecosystems’ carbon sequestration capabilities vary due to tree species, age, soil, climate, and management practices. Tropical, temperate, and boreal forests sequester carbon differently, making a one-size-fits-all approach difficult.
2. Variety of Allometric Equations
Allometric equations estimate tree biomass and carbon from diameter at breast height (DBH), tree height, and wood density. Tailored to regions, forest types, and species, these equations reflect diverse growth patterns of trees. Even in a tropical climate, different equations apply for forests, mangroves and peatlands. As an example, Chave’s equation is a widely-used, pan-tropical allometric equation commonly employed to estimate AGB in tropical forests.
Chave’s standardised equation is typically used for tropical forests:
Biomass = 0.0673 × (Wood Density × DBH² × Height)0.976
The choice of an allometric equation depends on the specific objectives, data availability, and local conditions of the study area. Arkadiah validates the AGB range by using both region-specific and pan-tropical equations to understand the range of AGB estimates and the reasons for variations, thereby increasing confidence in the estimates.
3. Measurement Techniques
Several measurement methods are used, each with its strengths and limitations. Direct methods, such as biomass sampling and soil carbon analysis, provide more precise data but are labour-intensive. Remote sensing technologies, like satellite imagery, offer broader coverage but may lack the granularity needed.
Case Study: Measuring Brunei Peatland Forest Carbon
Brunei’s peat swamps are unique ecosystems with exceptionally high biomass and carbon stock. Remote sensing methods have been insufficient, relying on sparse data that often deviates significantly from actual on-ground measurements.
This discrepancy poses a challenge for accurate carbon estimation and forest management in these critical areas.
Arkadiah’s Approach
To address the issue of carbon estimation accuracy, our science and engineering team combined ground truth with AI-driven analysis. The analysis and findings were previously presented at the Brunei Darussalam Conference on Biodiversity in Jun 2024 by Gerry Ong and Dr Deepthi Chimalakonda.
Step 1:
We collected sample plot data from a ½-hectare area in Badas.
We employed both manual measurements and terrestrial laser scanning (TLS) LiDAR for comprehensive data collection and comparison to estimate aboveground biomass (AGB).
Step 2:
We uploaded the data measurements into NatureOS and scaled up carbon estimation to an extensive project plot encompassing 77,000 hectares of peat swamp forests.
Step 3:
Using AI and machine learning, we incorporated Arkadiah’s proprietary models to analyse the scaled land area.
Our Findings
#1: Consistent AGB Values from Ground Truth Measurements
Applying various allometric equations yielded consistent AGB values ranging from 470 to 560 tonnes per hectare from both Manual and TLS measured data. We have observed in manual measurement, equipment handling may have attributed to the potential overestimation of tree height, underscoring the advantage of LiDAR technology to provide precise, consistent measurements free from human error and physical limitations.
#2: Higher AGB Values vs Other Asian Tropical Forests
The AGB values from Brunei’s peatland forests significantly surpassing the average AGB of 280-350 tonnes per hectare observed in other tropical rainforests in Asia. This highlights that Brunei’s peat swamp forests are old growth forests and hold high biomass stocks.
#3: Ground Truth Closes Carbon Underestimation Gap
Arkadiah’s proprietary AI model, trained with ground truth data from the ½-hectare sample plot, revealed an over 70% underestimation by pre-trained remote sensing data alone. Our model’s values are aligned consistently with AGB measured by TLS.
Dr Borhan, Lunima Sdn Bhd, Brunei Darussalam, “As forestry consultants to Brunei Forest Department (BFD) conducting the National Forest Resources Inventory (NFRI 23/25), the collaboration between BFD and Arkadiah has taught us on a better and faster way of addressing the challenges of carbon measurement in Brunei’s complex, undisturbed and unique peat swamp forests. It is not surprising when BFD holds Arkadiah in high regard in this respect. The precision of their AI-driven models and LiDAR technology sets a new benchmark for large-scale carbon estimation and dramatically improves our ability to quantify, manage and protect Brunei’s unique peatlands.”
