The nature of this project is carbon, we communicate in tonnes CO2 and therefore how we measure tonnes of CO2 is very important. One of the most important things is measuring the actual trees standing in our project. There are loads of different ways to measure carbon dioxide. Recently, a lot of companies are pushing for the most modern techniques backed by AI, some companies suggest you don’t even have to go to your forest. But interaction with the land and the community is important for Forestbase. Our piloting forest inventory consisted of 0.1 ha plots where all species with a breast height greater than 10 cm were identified and measured.
Boots on the ground
A forest inventory is not desk work, it’s physically hard, human work carried out in real conditions. Reaching sampling locations can require long travel days, changing weather windows, and careful navigation through terrain where equipment, time, and energy are always limited. In this context, consistency matters: the value of the dataset depends on collecting the same measurements the same way, even under pressure. That is why we pair fieldwork with training and quality control. Team members are aligned on protocols before entering the forest, measurements are checked in the field to reduce error, and records are reviewed to ensure data is complete and comparable across sites. Beyond the numbers, this approach also builds local capacity, strengthens practical skills, shared routines, and long-term ownership of monitoring, so the system can be maintained and improved over time.
Design of the inventory
We used a random sampling design to reduce selection bias and make the dataset more representative of the wider forest. At this stage of the project, we also designed the inventory to be logistically realistic: access, time in the field, and operational constraints shape what can be measured consistently. The trade-off is clear: more complete coverage is more desirable, but a repeatable protocol that can be executed well now (and expanded over time) is more valuable than a one-off effort that cannot be maintained. Finally, we focused measurements on trees above 10 cm diameter, a widely used threshold that captures the main components of forest structure while keeping data collection efficient and robust. Alongside structure, we also recorded species information where possible, because species composition provides critical context on forest condition and resilience that structure alone can’t capture, and it strengthens how we track biodiversity outcomes over time.
Sneak peek of the results
Early results from our forest inventory already provide a clear snapshot of forest condition and recovery dynamics. Across randomly distributed 0.1 ha plots, our team recorded 595 trees (≥10 cm DBH) representing 79 species, 63 genera, and 32 botanical families, giving us a solid first view of species composition and structure.
Most recorded stems (around 88%) fall within the 10–37 cm diameter range, suggesting relatively uniform stand conditions and a forest that is actively regenerating after past disturbance. Height measurements tell a similar story: the average tree height recorded so far is 14 m (with values ranging from 5 to 35 m), which is consistent with predominantly secondary forest rather than mature primary forest. Even at this stage, the dataset highlights both diversity and complexity, capturing large individuals, documenting key biomass contributors, and identifying potential seed-producing trees and seedbeds that can support long-term natural regeneration. As we expand sampling into additional plots, these insights will become more representative of the wider forest and strengthen our ability to track change over time.
Eventually, this dataset will be combined with the national forest inventory and Verra-aligned risk mapping to estimate tonnes per hectare for our forest. By grounding these estimates in field measurements and recognized reference datasets, we strengthen both the quality and credibility of our results and credits.
