Robust quantification of carbon sequestration requires three steps: defining a baseline, measuring the increase due to given action, and accounting for all uncertainties involved.

Defining the baseline is simply a computation of how carbon stock develops through time, assuming prevailing practices (and legal requirements).

Quantifying the increase in carbon sequestration measures how the stocks increase above baseline, if a given carbon offset action is done.

Finally to ensure required level of conservatism, one must address all the embedded uncertainties: What they are, their size, cross effects etc.

This is particularly important, as the overriding principle is to choose a conservative value and methodology.

Though simple to understand, this is always a very detailed, physical computation process and in the context of boreal forests requires:

1. practical knowledge and understanding of prevailing forestry norms and management practices,

development of adequate

2. theoretical growth models,
3. applied and empirical forest-management models,

and to achieve the above, it would need to acquire.

4. large enough and varied enough physical research sites where these practices and models can be tested, verified, and their underlying uncertainties addressed.

Understanding the prevailing forestry practices and legal requirements is naturally the key in defining accurate baselines – by definition. They are also important in measuring increases as they define the set of forest-management techniques available, how they vary, ways to apply them, their cost, scale, impact etc. Finally, thorough understanding of these practices is crucial in understanding their intrinsic uncertainties.

Theoretical growth models are the first basis for computing the level of carbon stocks. It is fortunate that in forests tree growth and carbon stocks are highly correlated – also including the below ground biomass which is proportional to the above ground mass by allometry. Therefore the existing forest growth models can provide a starting point for estimating both the baseline and the increased growth levels.

Unfortunately, almost all the current models lack very long-term accuracy – and permanency is key for high-integrity carbon offsets. In many models this inaccuracy rises from design: they were designed to model single rotations that would then be repeated. If the growth periods are extended much above, the accuracy drops rapidly. Therefore, for proper and trustworthy carbon accounting, Aurora Forealis must contribute to the development/refinement of a new set of models (our concrete program work on this is described in the next section under Luke.) This model refinement is also needed for measuring model uncertainty. On one hand, the currently used simple, linear statistical models are easy to understand and prod, but they cannot count for the long-term effects. On the other hand, models that are non-linear, complex, and with a lot more parameters, can potentially be more accurate, but also a lot more uncertain. And from Bayesian perspective, this uncertainty increases very rapidly.¹Thus controlling both accuracy and predictability will always be a trade-off, and often a judgmental/empirical one in that.

1. Generally, the more non-linear the model is and the more parameters it has, the more accurate it can be made, but also a lot more uncertain. ↩︎

Applied and empirical forest-management models focus on the periodically applied forest-management techniques. In addition to modelling the prevailing practices and thereby baseline, they are fundamental in computing the increased carbon sequestration due to the isolated carbon offset action.

These applied models suffer from the same general problems as the growth models (lack of long-term accuracy, trade-off between predictability and uncertainty etc.), but also from very specific ones. They introduce periodicity to the problem – every so often a forest management technique is applied (fertilizing/weeding/thinning…) and these are techniques are optimized again in the subsequent period. Therefore, their underling modelling error is compounded by optimization, again in particularly over very long horizons. Also, and crucially from carbon perspective, they generally do not account for technique’s indirect impact on carbon stocks or unintended releases. Therefore, Aurora Forealis must also consider how these applied modelling approaches need to be refined (our actual approach described in the section below under Tapio).

Physical research and test sites since, as in any empirical research mixing practice and theory, it is important to experiment. During the research, it become clear to Aurora Forealis that in the case of boreal forest related carbon offsets, this is vital. In particular, any development of new models (theoretical and applied) will be impossible without the research sites. This is the only way the parameter values and correlations can be estimated, the models can be validated, and resulting hypotheses tested. Indeed, this is how all the existing t forest models have been developed.

Research and test sites are also key in creating separate test and control groups to measure the impact of different carbon offset treatments – i.e. very similar areas that differ only by a single controlled action. There is actually an emerging method in carbon offset validation based solely on this idea: The use of dynamic baselines (physical control forests) instead of projected baselines (model based).

These test sites are also crucial in understanding the embedded uncertainties – measurement methods, their variability, biases, and accuracy, variability in treatments, variability in effects, etc. They will also test many of the default assumptions – cost, scale and variability of the various techniques.

In terms of the exact characteristics of these research sites, for modelling the forestry-management techniques these areas must be large enough and close enough to form reasonable test and control groups. This allows comparison in close proximity (ground preparation, planting, fertilization, thinning etc.) For the theoretical growth models the size is less relevant (test plot sizes are only 30mx30m), but they must be varied enough. In other words, the underlying forest plots themselves must be varied enough – particularly in age range – so that full scale of the models can be fitted. Finally, for understanding and testing current forestry practices, the research and test sites must be both large enough and varied enough. (Auora Forealis 2023-2024 program in the section below under Research Sites Tervola).