Climate change in the Faustmann forest economics model: a qualitative review Open Access

Climate change and forest economic management are tightly linked because forests are major carbon sinks and providers of other ecosystem services, many of which are often not priced in markets. A prominent suggestion to reduce the consequences of global warming is to manage forests to offset carbon emissions (Englin and Callaway, 1993). Management choices – rotation length, species selection and silvicultural treatments – directly affect net CO₂ fluxes and, therefore, climate outcomes. At the same time, climate change alters the biophysical conditions under which those choices are made: it is associated with more severe forest disturbances (e.g. fire, insects, windstorms) and reduced net carbon uptake, while also shifting the ecological zones and growing conditions of species leading to sizeable economic consequences for forest owners. Rapid forest loss amplifies this by weakening the capacity of forests for carbon sequestration and increasing near-term emissions.

A central forest economic benchmark for addressing these questions is the Faustmann model, which was proposed in its classical form by Faustmann in 1849 (Sedjo, 2017), and whose optimal-rotation condition was later derived by Bertil Ohlin in 1921 (Ohlin, 1921), as highlighted in Löfgren’s historical note (Löfgren, 1983). The optimal rotation age, as defined by the model, is the harvest time that maximizes land expectation value (LEV), i.e. the present value of an infinite sequence of rotations. In its standard form, this can be written as follows (Amacher et al., 2009):

where:

1. $p$ is stumpage price,

2. $f(T)$ is timber volume at rotation age $T$

3. $r$ is the discount rate, and

4. $c$ is regeneration cost

The corresponding first-order condition can be written as (Amacher et al., 2009):

where:

1. $pf’(T)$ is the marginal value of timber growth

2. $rpf(T)$ is the opportunity cost of delaying timber revenue

3. $rV$ is the opportunity cost of delaying the next rotation

The economic intuition behind the classic Faustmann rule is that forest owners seek to increase the value of forest land by comparing the marginal benefit of allowing the stand to grow for one more period with the opportunity cost of delaying harvest. The opportunity cost consists of the return that the forest owner could have earned by investing the timber revenue elsewhere and the land rent forgone because the site remains occupied by the current stand instead of being regenerated and used for the next rotation (Amacher et al., 2009). Thus, the Faustmann rotation is not determined only by biological growth but can be understood as the point at which additional stand growth no longer compensates for the delayed receipt of timber revenue and the delayed start of the next rotation. It then follows that stumpage price, regeneration cost and interest rate are fundamental parameters that determine optimal rotation because they affect the harvest revenue and the opportunity cost of waiting, and thus the value of forest land (Amacher et al., 2009). Since it links forest valuation to optimal rotation and related management choices, the model can also inform broader questions of forest management, from the ecological to the climatic, and those regarding policymaking. Its value, in the context of climate change, lies in its ability to be extended to represent the multitude of economic (timber price, costs, carbon payments, risk and uncertainty) and natural forces (growth rates, destructive events, biodiversity, amenities) that are directly affected by climate change.

One of the earliest major extensions of this framework was Hartman's (1976) model, which generalized the Faustmann–Ohlin rule by showing that the optimal harvest decision changes when the standing forest itself provides valuable non-timber services in addition to timber revenues. Later on, a pivotal step in this broader development was Chang's extension of the Faustmann framework to uneven-aged stands, which showed that the same Faustmann formula can serve as the theoretical basis for both even- and uneven-aged management, at least when it comes to timber production (Chang, 1981), followed by his plantation-management model (Chang, 1983). Subsequently, Chang's landmark paper introduced the generalized Faustmann formula, which allowed for the harvest age, timber yield, regeneration cost and stumpage price to vary in each rotation (Chang, 1998). These seminal contributions are important because uneven aged forestry, and by extension continuous cover forestry, is one of the more recent trends in the Faustmann literature and it is linked to the valuation of ecosystem services (Mpekiri and Papaspyropoulos, 2026a), which are highly relevant to climate-change mitigation and adaptation. Modern Faustmann-based work commonly expands the objective function and constraints to include carbon sequestration with multiple carbon pools and post-harvest carbon dynamics (Hoel et al., 2014), joint production regimes (e.g. timber plus carbon or biofuel) (Bjørnstad and Skonhoft, 2002; Indrajaya, 2020; Ning and Sun, 2019), age-dependent disturbance risk (Loisel, 2014), policy instruments (carbon taxes, subsidies, penalties, insurance) and uncertainty in prices (Yu et al., 2026) and hazards (Amacher et al., 2009). This makes the Faustmann framework an invaluable economic reference point for comparing how different climate modeling techniques and policy designs change both optimal rotation and forestland value.

Thus, the purpose of this study is to conduct a qualitative review of the Faustmann literature that explicitly incorporates climate change, with the aim of examining what these extensions of the classical model imply for the core output of the model, optimal rotation age. Specifically, it traces (1) the main modeling techniques used to bring climate change into the Faustmann framework (carbon accounting, climate-driven disturbance risk and productivity and species shifts), (2) the recurring patterns regarding optimal rotation and LEV under carbon objectives, risk and stochastic prices, and (3) the resulting policy implications, emphasizing how conclusions depend on accounting rules, instrument design (taxes, subsidies, penalties, insurance) and uncertainty. The important implication is that it is not enough to design policies to address the ecological aspects of climate change (Hanewinkel et al., 2010). The results of the analysis call for a transition to more complete, ecosystem-based methods of carbon accounting and policy design that will assist in the implementation of mitigation strategies, such as species diversification, insurance against timber and carbon price volatility and adaptations to combat the increasing uncertainties we face due to climate change.

The literature search was conducted in Scopus on March 27th, 2026, using the keywords “climat*” and “faustmann” connected through the Boolean operator AND, in order to identify publications addressing climate-related issues, including climate change or climatic change, in relation to the Faustmann model. The broader keyword “climat*” was used instead of a more specific term, such as “climate change”, to capture a wider range of climate-related studies related to the Faustmann model. Further, other terms that are often associated with climate change (e.g. “carbon sequestration” or “ecosystem services”) were not used as keywords, because their selection would require a subjective predefinition of climate-related topics and could potentially introduce a topic-selection bias (Haunschild et al., 2016). No time restrictions were applied. The initial search returned 66 records, spanning publication years 1993–2025.

Study selection was based on literature-review methods commonly used to identify, screen and select relevant studies, including methods previously applied in a bibliometric and thematic review of the Faustmann literature (McIntosh and Zhang, 2024; Mpekiri and Papaspyropoulos, 2026a). For screening we followed the PRISMA methodology (Figure 1) (Liberati et al., 2009). From the initial results, one (n = 1) duplicate was removed as well as nine (n = 9) non-English records. Subsequently, the results were screened by document type, retaining only journal articles, conference papers and review papers, excluding four (n = 4) records featuring books and book chapters, and two papers using the word climatic but not in relation to climate change. This process resulted in a final dataset of 50 papers which were read carefully and analyzed qualitatively.

The selected papers were reviewed qualitatively, based on optimal-rotation theory rather than as a systematic literature review, with emphasis on how each study incorporates climate change into the Faustmann framework and how this is linked to outcomes for optimal rotation and policy. The dataset of papers was then carefully analyzed and information was extracted on (1) the modeling technique used to represent climate-related effects (e.g. carbon-related components, disturbance risk, productivity/species changes), (2) the way rotation is derived within the model setting and (3) any explicit policy-relevant implications tied to accounting rules, instruments or uncertainty. In the following sections, the results are presented in a thematic organization to highlight recurring patterns uncovered across the selected studies.

The results of the review show that climate change has been an important driver of expansion within the Faustmann research landscape. To address the complex demands of climate change in terms of economic forest management, the Faustmann model has been expanded from a “timber-only” valuation, into a framework that incorporates the explicit treatment of carbon, uncertainty and climate-driven disturbances. Through this review, three key themes emerged, which are outlined below:

1. Climate-related modeling channels

2. Optimal rotation

3. Policy

“Modeling techniques” refers to how climate change is integrated into the Faustmann model. Across the dataset, three main modeling techniques emerge: carbon accounting, climate-driven hazards and productivity changes and species shifts.

The process through which forests accumulate, or “sequester” carbon in their living biomass is complex and familiar at the same time. Carbon is allocated to the different parts of trees (roots, stems, branches, leaves, etc.) through photosynthesis, which is highly sensitive to environmental conditions (Sands and Landsberg, 2002). Furthermore, relative to other environmental services we receive from forests, carbon sequestration has an interesting characteristic. It begins to decline as trees age, that is, it does not increase monotonically with the age of the stand. As Ning and Sun (2019) mentioned, “the relation between carbon quantity sequestered on the forest and time is nonlinear”. As a result, for carbon sequestration to have any meaning both as a climate change mitigation strategy and as a financial prospect for forest owners, it is necessary to have methods of estimating the amount of carbon stored in living biomass as well as emitted back into the atmosphere after forest stands are harvested. These methods are generally referred to as carbon accounting.

Carbon accounting is based on the carbon cycle of managed forests, which consists of three phases: sequestration in living biomass (growing trees), carbon stored in harvested wood products (HWPs) and the emission of carbon back into the atmosphere at the time of oxidation of the HWPs, or, depending on the carbon accounting method, immediately after harvest (Englin and Callaway, 1993). To mathematically implement this process into the Faustmann model, most research follows the general path of adding a carbon growth function which provides an estimation of the carbon stored in biomass per year (one of the most popular ones being the logistic growth function), and a function or empirical estimation of the moment or rate of release of carbon after harvest (Englin and Callaway, 1993). It is this negative effect of post-harvest emissions (Akao, 2011) that is at the center of much debate around how to account for carbon in HWPs in national greenhouse gas inventories (Akao, 2011).

The default assumption, based on the IPCC 1996 Guidelines (IPCC 1997), is that all carbon in harvested tree biomass is oxidized in the year of removal. This guideline may allow researchers and policymakers to simplify calculations, but it does not paint the full picture of what actually happens to stored carbon after it leaves the forest. For example, Perez-Garcia et al. (2005) calculated that about 50% of harvested wood is turned into lumber, assuming it has a service or half-life of approximately 80 years.

The question of what happens to carbon after harvest is important because in the Faustmann model it can be a significant determining factor for optimal rotation. In fact, the forest’s potential for carbon sequestration leads to longer or shorter rotations mainly based on the path of carbon release post-harvest. This, apart from the type of function or estimation of rate of release, as mentioned earlier, depends to a considerable extent on the discount rate. A common method of incorporating the emissions from the oxidation of HWPs into the Faustmann model is to derive a net discounted emission value by discounting the future release (Murray, 2003). In Englin and Callaway (1993), in a forest managed purely for carbon sequestration, low discount rates make re-emission less important and favor rapid carbon accumulation, whereas higher discount rates make the timing of harvest more strongly affected by the fact that harvest initiates the post-harvest decay and re-emission of stored carbon and can thus favor longer rotations than the classic Faustmann rotation. The numerical implications of this result for optimal rotation are discussed in Section 3.2. In broad terms, however, if the amount of carbon released immediately or soon after harvest is smaller than the amount sequestered during forest growth by a meaningful degree, as in the case of long-lived HWPs, the optimal rotation tends to be shorter (Akao, 2011) since the emissions have less effect on LEV. That of course further depends on the carbon policy of each country, be it taxes or penalties for emissions (Mpekiri and Papaspyropoulos, 2026a).

Other than timber production, the production of biomass is increasingly amassing interest (Mpekiri and Papaspyropoulos, 2026a) both among researchers and forest managers. The utilization of harvested biomass can significantly impact carbon accounting (Thompson et al., 2009). For example, biomass intended as biofuel reduces the net carbon gains of a forest (Mpekiri and Papaspyropoulos, 2026a), since the immediate burning leads to an immediate increase in the net discounted emissions value (Akao, 2011)

At this point, an important distinction should be drawn between the physical and the economic sides of carbon accounting. While the various configurations of the growth function determine the amount of carbon stock in the stand over time, the remuneration of that carbon is the way the additional carbon stock is converted into carbon credits or payments for the forest owner. The way the carbon remuneration scheme is designed can significantly impact the profitability of carbon sequestration for forest owners. For example, Indrajaya et al. (2024) compare the Verified Carbon Standard (VCS), perhaps the most widely used greenhouse gas crediting program, which is based on average additional carbon storage over a rotation, with “current carbon accounting” in which remuneration is based on additional carbon stored in each year, and show that the accounting rule itself affects the incentives faced by forest managers. Under current carbon accounting, for instance, a forest owner would be incentivized to do a 5-year increase (from 7 to 12) in rotation that raises carbon storage by 57% for 1,624 USD/ha, whereas under VCS the same storage outcome requires 2,346 USD/ha (Indrajaya et al., 2024).

Within the forest economic literature, there is growing concern that forest carbon sequestration may not be able to provide us with the full story regarding climate change mitigation. There is a question to be asked about whether carbon accounting, as it is usually done so far, is leaving out important aspects of the forest land ecosystem that contribute significantly to the atmospheric carbon balance of the Earth. One potentially significant factor in this is radiative forcing. IPCC (2001) defined radiative forcing as “an externally imposed perturbation in the radiative energy budget of the Earth’s climate system”. The means by which forests can cause such a perturbation is through the albedo effect (Thompson et al., 2009).

The albedo of an object is its tendency to reflect radiation to higher or lower degrees, and it is defined as the “ratio of reflected to incident electromagnetic radiation” (Thompson et al., 2009). Darker surfaces, such as forests, tend to absorb higher amounts of solar radiation than bare, agricultural or snow-covered land, and consequently, it is possible that they have a warming influence on their surrounding region (Thompson et al., 2009). In other words, the albedo of forestland is generally lower than the albedo of non-forested land, since forests reflect less solar radiation back into the atmosphere. Gibbard et al. (2005) reported that average land albedo fell from 0.23 in the bare-ground case to 0.17 when only the direct effect of trees on the land surface was considered, and to 0.15 when the resulting warming was also allowed to reduce snow cover and further lower surface reflectivity.

The climatic effect of differences in albedo can then be evaluated using the standard climate metric global warming potential (GWP), expressed in CO₂-equivalent terms (Rørstad, 2022) which determines the change in the amount of terrestrial carbon stock (the carbon stock of crop land) that would be equivalent to the change in albedo after deforestation. Forests can make positive (cooling) contributions to climate change through carbon sequestration, but they can also make negative (warming) contributions through reducing the surface albedo (Rørstad, 2022). Therefore, when considering climate mitigation strategies involving forestland, it is important to consider the positive (cooling) and negative (warming) contributions of forests to the warming effect of the shortwave solar radiation absorption potential of the Earth (Rørstad, 2022). An argument can be made for moving on from a simplistic greenhouse gas thinking in carbon accounting to a more complete radiative forcing perspective by converting, as Rørstad (2022) suggested, all climate effects (both carbon and albedo) into equivalent CO₂ units for unified pricing. It may be argued that the negative (warming) effect of forest-covered land is larger than the positive (cooling) effect of carbon sequestration following afforestation (Gibbard et al., 2005). Gibbard et al. (2005) estimated that global replacement of current vegetation by trees would produce a global mean warming of 1.3°C, while the potential cooling from storing 1,500 PgC (petagrams of carbon) would be offset by about 40% as a result of the decrease of the albedo. This lower albedo increases the absorption of solar radiation and thus contributes to warming. Moreover, after about 80 years, less than 40% of the initial atmospheric CO₂ reduction would remain, implying net warming when both albedo and carbon-storage are considered. In other words, transitioning from bare or cropland to forested land can be considered as an emission, in CO₂-equivalent terms, both because it, generally, lowers albedo and because of the emissions through harvesting and disturbance events (Thompson et al., 2009). This highlights the immense importance that changes in land-use have in the context of climate change. However, this effect was observed following deforestation (Thompson et al., 2009) and it is unclear how lasting it may be in the long term.

However, more recent research (Rørstad, 2022) points out that the issue is more complicated. There is a difference in the albedo effect of different kinds of forests. For example, while forests in boreal and high-latitude temperate regions exert a much higher warming effect (Gibbard et al., 2005), tropical forests, due to their lower reflectivity and lighter color (Rørstad, 2022), have a more cooling effect. This not only highlights the vast potential of tropical forests in the mitigation of climate change (Canadell and Raupach, 2008) but also informs our understanding of the various policies employed in different countries in boreal regions regarding afforestation projects (Mpekiri and Papaspyropoulos, 2026a). Temperate forests in general occupy a middle ground. In their simulations, Thompson et al. (2009), drawing on Bala et al. (2007), report that large-scale deforestation simulations produced cooling in boreal regions, warming in tropical regions and essentially no net change in temperate regions.

Regarding Mediterranean and similar dryland and semi-arid forest ecosystems, we could not identify, either in the main review dataset or in the wider literature consulted, research that incorporates radiative forcing in general or the albedo effect of forests in particular, into the Faustmann model. There were, however, studies that focused on the inclusion of radiative forcing in carbon accounting for dry and semi-arid forest ecosystems (Rotenberg and Yakir, 2010), including Mediterranean-type dryland regions in Israel (Rohatyn et al., 2023). Hasler et al. (2024) found that, contrary to previous research, there is a greater proportion of areas with a net-negative climate contribution due to the warming influence of the albedo, once tree-cover restoration is expressed in CO₂-equivalent terms. Current research generally suggests that these types of forest ecosystems should not be treated simply as typical temperate-region forests, because the balance between carbon sequestration and the albedo effect can vary strongly depending on aridity, local radiation balance and the comparison between forested and non-forested land (Hasler et al., 2024; Rohatyn et al., 2023; Rotenberg and Yakir, 2010).

Another issue to consider when accounting for radiative forcing within carbon accounting is tree species. Large-scale changes in the distribution of coniferous species in boreal regions are currently taking place with significant economic effects for forest managers (Mpekiri and Papaspyropoulos, 2026b). These findings indicate that increasing the mixture of hardwood species may not only provide protection from unpredictable disturbances (Mpekiri and Papaspyropoulos, 2026b) brought on by climate change but also contribute to the reduction of the warming effect associated with a lower albedo in boreal forests (Thompson et al., 2009).

Regardless of the strength of the warming effect produced by the albedo of forestland, especially in boreal and high-latitude temperate regions, it would be erroneous to conclude that increasing deforestation is a viable measure of climate change mitigation (Caldeira, 2007; Thompson et al., 2009). As we will discuss in a later section, forests do not contribute to the mitigation of climate change merely through carbon sequestration or radiative forcing. There are numerous other amenities that forests provide for human society, including water conservation, biodiversity protection, tourism and recreation (Mpekiri and Papaspyropoulos, 2026a), although these broader amenities are not the focus of this survey. Further, it is premature to say that the albedo effect of forests is reason enough to abandon carbon sequestration efforts. The albedo effect may have significant implications when it comes to land-use changes, but within the forested areas that have not been subjected to such a change, the carbon stock at the time of harvest can be more than 40 times that of the albedo when both are expressed in CO₂-equivalent terms (Rørstad, 2022). This leads to the conclusion that while it may be important to move on from a simplistic greenhouse gas thinking in carbon accounting to a more complete perspective that includes radiative forcing, care should be taken to apply these principles where they are most relevant and effective for climate-change mitigation.

The increase in frequency and severity of unpredictable destructive events that has occurred in recent decades is highly connected to climate change. Climate change intensifies forest disturbances by changing the frequency, time of occurrence, the intensity and the duration of hazards like wildfires, storms and subsequent insect infestations (Dale et al., 2001). Climate change, atmospheric carbon and forest disturbances facilitate a destructive feedback loop where the expected rise in temperature promotes an intensified likelihood of disturbance (especially for fire), which then leads to sometimes massive emissions (Kurz and Apps, 1994, 1999; Mpekiri and Papaspyropoulos, 2026b). Between the 1980s and 1990s (Stollery, 2005), the increased frequency of forest fire disturbances led to a steep reduction in the sequestration of carbon in Canada (Mpekiri and Papaspyropoulos, 2026a; Stollery, 2005).

This alarming negative trend was noticed early in the evolution of the Faustmann model and various approaches were developed to address it. While economic and financial systems are not identical (Hummel et al., 2009), there are common steps of risk mitigation that can be applied to forestry. These comprise risk assessment, risk handling and risk control (Hanewinkel et al., 2011).

In the Faustmann literature, for the purpose of risk assessment, earlier approaches focused on modeling the economic loss brought on by catastrophic events by adjusting the discount rate (Buchanan and Kennedy, 2009), which effectively reflects the increased risk from the perspective of the forest owner in investing in the silvicultural use of forestland (Englin et al., 2000; Martell, 1994; Reed, 1984). In a different strand of the earlier literature, destructive events such as fires, windstorms and pest risks were incorporated through either empirical survival probability distributions (for example Dieter, 2001). These methods, however, do not allow the forest hazards to vary based on spatial scale (Hummel et al., 2009) or change over time (Stollery, 2005) as climate change produces higher and higher temperatures. Recently, more sophisticated methods that involve Poisson processes have been employed (Loisel, 2014; Mpekiri and Papaspyropoulos, 2026b; Rakotoarison and Loisel, 2017; Stollery, 2005).

It is interesting that the time variability and unpredictability of forest disturbances, apart from the ecological and climate consequences, produce an economic effect more akin to price and cost uncertainty in the Faustmann model (Stollery, 2005). This is important for risk handling and risk prevention, because not all forest owners may have the same attitude toward risk (Hummel et al., 2009). Indeed, as will be discussed later, forest owners are predominantly risk-averse (Loisel et al., 2020). While risk handling and prevention are outside the scope of this paper, it is interesting to highlight this similarity of hazard risk to price uncertainty as the two influence one another when it comes to policy. Timber price is an extremely volatile variable (Rakotoarison and Loisel, 2017), influenced by many parameters such as location and time, supply and demand (Loisel, 2014; Mpekiri and Papaspyropoulos, 2026b) and climate policy (Buongiorno et al., 2011). When it comes to climate change, timber price uncertainty is perhaps one of the most important concerns for the success of any strategy, since it is the main incentive for forest managers to continue investing in the economic management of forests (Rakotoarison and Loisel, 2017), and it is the most sensitive variable with regard to various risk prevention measures such as a shorter rotation age or regeneration with different tree species (Guo and Costello, 2013).

So far, climate change has been mainly discussed through the lens of disturbance risk – warming raises the probability of damaging events and reduces the expected value of waiting. Stollery (2005) notes that a possible long-run cooling feedback effect could arise if warming accelerates forest growth, allowing forests to absorb more carbon and thereby slowing warming. Regardless of how general that effect is across regions, the modeling implication is clear: if climate affects growth, then climate change enters the Faustmann problem through the growth function, not only through hazard rates.

West et al. (2019), comparing Eucalyptus sp., Acacia sp., Pinus sp. and Tectona sp., also showed that slow-growing and fast-growing species differ markedly in their response to PES (payments for ecosystem services) incentives. For $5–30 Mg CO₂^–1, Acacia sp. had the largest proportional increase in average carbon stocks over the rotation period (13–47%) relative to its no-PES baseline, Eucalyptus sp. and Pinus sp. the largest absolute increases (10–43 Mg CO₂ ha^–1), whereas Tectona sp. increased only by 0.2 Mg CO₂^–1 ha^–1 (1.8%) at the highest payment level.

In this line of research, Buchanan and Kennedy (2009) estimated the economic value of two of the most important species in Australia, Pinus radiata and Eucalyptus globulus, by comparing their effect on LEV with assumed current growing conditions and with prospective accelerated growing conditions due to climate change. Their results showed that climate change affects these two species in a completely different manner. Buchanan and Kennedy (2009) reports expected declines of 0–25% in NPV for Pinus radiata and 0–25% gains for Eucalyptus globulus. This result could be an example of a biome shift with dramatic implications for forest owners, especially in terms of risk aversion. Other researchers have relied on the effect of species shift on NPV to apply portfolio investment theory (Knoke and Wurm, 2006) for the optimal selection of species given the risk attitudes of forest owners.

It appears that species and biome shifts will continue to affect the economic value of forests as climate change progresses. Thuiller (2007) estimates that with every 1 °C increase in temperature, the ecological zones will shift by approximately 160 km from low to high latitudes. This magnitude helps explain why climate-driven species re-evaluation is not only hypothetical. In central Europe, for example, active forest transformation from Norway spruce (Picea abies) to European beech (Fagus sylvatica) is described as already occurring on a large scale (Spiecker et al., 2004). The modeling of the NPV gains and losses of species in the Faustmann model allows us to derive important conclusions about climate change adaptation and mitigation strategies because climate-driven species switching is not evaluated only on the basis of expected growth gains; it is constrained by the economic weight of regeneration and replanting expenditures that must be paid up front.

One implication from the examination of these commonly identified modeling techniques is that there is more than one aspect to the integration of carbon sequestration into the Faustmann model. To provide a complete picture of how carbon sequestration affects the value of the forest in terms of climate change, one needs to consider the physical aspect of how carbon accumulates in biomass, the remuneration method through which carbon sequestration is given a monetary value in carbon markets as well as the contribution of forests to the warming or cooling of the climate through the albedo effect, which becomes especially important in afforestation projects.

Another implication is that stochastic methods of modeling natural hazards, while more complex than changes in the discount rate, better reflect the unpredictable timing and intensity of fires, storms and insect outbreaks.

Lastly, changes in the productivity and distribution of different species seem to also function as a form of economic risk, since it can make currently preferred species less valuable and force forest owners to consider costly species switching, which incurs considerable regeneration and replanting costs that carry greater weight in the present than the future through discounting.

A rather challenging result one can conclude from the scientific literature around climate change and the Faustmann model is that there is no general implication about the optimal length of the rotation period. Adapting to the myriad of risks, uncertainties and adverse environmental consequences of climate change for forests is not possible through simple policies that can always be applied in all locations and settings. What can be done, however, through the review of the literature is to determine in broad strokes which aspects of climate change tend to lead to shorter rotations, and which to longer ones. Although, as some cases indicate, the net result may produce an effect where opposing forces cancel out (Ning and Sun, 2019).

Initially, carbon sequestration entered the Faustmann literature at a time when carbon markets were not yet widely developed. One of the first and most influential efforts at doing so was by Englin and Callaway (1993) and Englin and Callaway (1995) when carbon was still a non-market commodity. They used various shadow-price scenarios ($10, $100 and $200 per ton of CO₂) which represented the range of values under discussion at the time (Englin and Callaway, 1993, 1995), which were more reflective of the social cost of carbon (Mpekiri and Papaspyropoulos, 2026a). In an early work by Englin and Callaway (1993), accounting for carbon sequestration in the Faustmann model has noteworthy implications for optimal rotation. Incorporating the full carbon life cycle, that is accounting not only for carbon sequestration in growing trees but also for the carbon stored in HWPs and its subsequent re-emission post-harvest, produces different results from the classic timber-profit maximizing Faustmann formula. Under the classic Faustmann rule, higher discount rates shorten the optimal rotation, as expected, since an increase in the discount rate also increases the opportunity cost of capital. When the objective is purely to maximize carbon sequestration, a higher discount rate can lead to longer rotations because, in that setting, delaying harvest postpones the post-harvest release of carbon. In their Douglas-fir case study, optimal rotation increased from 20 years at a 2% discount rate to 49 years at a 10% discount rate, while the timber-only rotation conversely, fell from 48 to 26 years. As Akao (2011) pointed out, a forest has the dual function of “sequestering atmospheric carbon and postponing sequestered carbon release”. If more importance is placed by policymakers and stakeholders on sequestration, rotation becomes longer and vice versa. Hence, policymakers may choose to affect the discount rate.

However, this discount-rate result should be distinguished from the effect of carbon prices. Additionally, the rate of release from the different HWPs plays an important role in carbon accounting and therefore directly influences optimal rotation. Timber products that deteriorate quickly, such as wood fuel, contribute to longer optimal rotations, because a smaller share of harvested carbon remains stored over the long term. In Van Kooten et al.'s (1995) model, the shadow price of carbon is used to calculate an annually paid subsidy for the carbon stored in each additional m³ of timber added to the growing stock and a carbon tax forest owners pay based on the fraction of harvested carbon that is released rather than stored long-term in HWPs (i.e. the “pickling” factor). Their results show that higher carbon prices generally lengthen rotation, but the size of this effect depends on timber prices and the pickling factor. Specifically, when none or only half of the carbon is stored long-term in HWPs and timber production produces no profit, optimal rotation can become infinite, that is, it becomes more economically advantageous not to manage the forest for timber production, irrespective of the different discount rates considered in their coastal and boreal forest case study.

On the other hand, when all carbon is assumed to be stored in HWPs in the long term, higher carbon prices still tend to lengthen rotation. In the presence of profit from timber production, however, if even when timber prices are low, the optimal rotation becomes finite and, generally, timber profit tends to lead to earlier harvests, even in the cases with a lower pickling factor (Van Kooten et al., 1995). Susaeta et al. (2014) later made this link more explicit in their generalization of the Van Kooten et al. (1995) model by assigning different pickling factors to different harvested wood products – 0.1 for pulpwood, 0.6 for chip-and-saw and 0.9 for sawtimber – so that products with shorter expected lifespans store a smaller share of harvested carbon over the long term. Another factor that makes shorter rotations favorable is the incentive to maintain the activity of industries surrounding forest management (for example sawmills, paper production and furniture production) on which local economies may heavily rely (Akao, 2011). As a general rule though, when all other factors are equal, increased carbon prices tend to increase forest rotation (Murray, 2003).

Climate change also affects optimal rotation through changes in productivity and stand growth. Yang et al. (2015) found that increases in tree volume shorten the optimal rotation age, albeit marginally (Yang et al., 2015). This intuition is rather straightforward: if volume accumulates faster, the timber and carbon harvest threshold can be reached earlier.

The increased risk of hazard resulting from climate change generally shortens optimal rotation (Mpekiri and Papaspyropoulos, 2026a). When the risk of fire is allowed to change over time, there is a steep reduction in both the socially and commercially optimal harvest time. This is owed mainly to the “forward-looking nature of the model” on the stand level (Stollery, 2005). Additionally, risk handling measures such as increased investment in fire suppression have little effect on optimal rotation (Stollery, 2005). Conceptually, increasing fire risk over time functions like a discount rate that rises over time, reducing the value of waiting and inducing shorter profit-maximizing rotations (Stollery, 2005) and although carbon sequestration generally benefits from longer rotations both socially and commercially, the increasing occurrence and intensity of wildfires has a much more dominant effect. Storm and insect risk similarly reduce the optimal rotation of forest stands (Rakotoarison and Loisel, 2017).

Beyond biophysical risk, economic uncertainty – especially price uncertainty – also reshapes rotation outcomes and expected land value. While the classic Faustmann rotation and earlier extensions of the model, like that of Samuelson (1976), depended on perfect foresight (i.e. known future timber prices), over the years the Faustmann model has been extended many times and in many different ways to account for price volatility in timber – and more recently – carbon markets, as discussed in a recent review of the literature (Mpekiri and Papaspyropoulos, 2026a).

Timber price uncertainty has a significant effect on optimal rotation and LEV, greater even than that of storm risk (Rakotoarison and Loisel, 2017) and other types of risk. As a result, stochastic prices have been incorporated into the model to an extended degree (Mpekiri and Papaspyropoulos, 2026a). In broad terms, price uncertainty makes optimal rotation shorter (Ning and Sun, 2019). Brazee and Mendelsohn (1988) show why stochastic formulations can produce different rotation results than deterministic models. Assuming perfect foresight does not reflect the real-world variability of prices. Instead, stochastic models, like the reservation-price, are better representations of realized prices. The reason for that is that stochastic prices allow the Faustmann model to consider the fact that under stochastic prices landowners are given more speculation opportunities and can make the decision to harvest earlier when a more advantageous timber price presents itself. Stochastic prices also make joint production management (production of more than one output such as timber, carbon credits and biofuel) more profitable by increasing LEV (Ning and Sun, 2019). That being said, the opposite effect is also possible: higher price uncertainty can incur higher risk and higher financial loss (Ning and Sun, 2019). Insurance coverage can help mitigate this (Ning and Sun, 2019), especially in the context of most forest owners being generally risk-averse (Loisel et al., 2020).

Lastly, it is worth mentioning that the majority of Faustmann extensions in our dataset concern optimization on the stand level. While optimization on the landscape level is not a theme present in our dataset, a recent study by Paarsch and Rust (2020) proposed a dynamic programming solution where harvesting occurs across spatially heterogeneous forest blocks at random intervals determined by the expected rate of change in the current lumber market price (spot price) and timber volume on the block. Similarly, Lauer et al. (2020) noted that the effect of fire risk on optimal rotation in a single stand is really a function of forest conditions across the entire landscape, and especially when ownership is distributed among many parties, risk is compounded by each owner increasing risk for their neighbors.

Overall, the literature does not support a general optimal-rotation response to climate change. Rotation outcomes are contingent on the dominant channel and its parameterization – carbon incentives and accounting rules, growth and productivity shifts, species choice, stand structure, evolving disturbance risk, and timber and carbon price uncertainty – and these modeling techniques can reinforce or offset one another. Parajuli and Chang (2012) found that adding carbon sequestration benefits in uneven-aged loblolly pine stands did not significantly alter the optimal cutting cycle and residual basal area, even though financial returns increased as carbon prices rose. Another example comes from joint production models, where carbon-credit incentives generally lengthen rotation while biofuel revenues tend to shorten it, with the net effect depending on the specific parameter values adopted (Ning and Sun, 2019). The practical implication is that policy on optimal rotation, as we will discuss later, must be context and location specific.

The third theme emerging from the reviewed literature is the issue of policy, and what conclusions can be drawn about it in the context of climate change.

The development of public policy proves to be a thorny issue because policymakers need to factor in the interests of forest industries, forest owners and the public at the same time (Ning and Sun, 2019). A straightforward example is bioenergy policy: subsidies for bioenergy products may look attractive when framed as a climate tool, but they still need to be assessed in terms of who gains, who bears costs and how incentives change across the supply chain (Ning and Sun, 2019). The goals and needs of all those stakeholders may at times be at odds (Ning and Sun, 2019). Yousefpour and Augustynczik (2019) argue that transparent communication of uncertainty with stakeholders and policymakers is crucial to avoid unrealistic expectations about forest ecosystem performance in terms of climate change mitigation.

As discussed earlier, forest climate policy can mean different things depending on whether the goal is framed as reducing radiative forcing or reducing atmospheric greenhouse gas concentrations (Thompson et al., 2009). This would suggest that afforestation incentives or the expansion of forest area for the purpose of generating carbon credits should not be treated as uniformly beneficial across locations and forest types. In boreal regions where the ground can be covered with snow for extended periods of time, forestation incentives (which may include afforestation projects and subsidies or expansion of forestland) may not be the ideal policy, despite the increased economic potential for carbon sequestration (Thompson et al., 2009).

As intuitive as the thought might be, in simulations, a hypothetical global afforestation or reforestation was projected to lead to increased mean temperatures. Gibbard et al. (2005) suggest that establishing plantations in non-tropical regions could have adverse results, and so we may conclude that incentivizing new plantations in boreal regions through policy should be limited. Conversely, in places where bare ground contributes to warming, such as tropical and subtropical regions (IPCC, 2023), afforestation can deliver substantial climate benefits (Thompson et al., 2009). This location-dependent effect can even extend to silvicultural choices, not only land-use change. Thompson et al. (2009) suggest that, rather than relying on uneven-aged single-tree selection, patch cutting within even-aged management could be beneficial because it exposes bare soil that may create a cooling effect due to increased albedo – especially where snowfall is high or soils are particularly light (Thompson et al., 2009).

On the other hand, tropical forests store more than 50% of the entire Earth’s forest carbon and produce a stronger albedo effect than boreal forests (Thompson et al., 2009), but are suffering extended and intense deforestation (Curtis et al., 2018; Hansen et al., 2013; Laso Bayas et al., 2022) which suggests that carbon sequestration schemes better suited to tropical forests may be advantageous. A more immediate policy implication emerging from the reviewed literature, however, is that carbon trading policies may need to shift from a carbon accounting philosophy that focuses only on the offset of greenhouse gases to a radiative forcing perspective which accounts for albedo (Canadell and Raupach, 2008, p. 1457) using “equivalent carbon” (Thompson et al., 2009). The data and modeling demands might make such accounting too costly in practice (Thompson et al., 2009) for the time being, but advances in the Faustmann-based modeling approaches and technology could help us adopt a more complete method of carbon accounting and, additionally, reinforce the climate mitigation potential of tropical forests (Canadell and Raupach, 2008, p. 1457).

Pricing policy – both for timber and carbon – shapes rotation, harvest timing and the willingness of landowners to invest. Policies that help protect forest owners against price uncertainty and secure some level of assurance against risk have been incorporated into the Faustmann model from an early stage. One such policy is the Reservation Price Strategy (RPS), which Brazee and Mendelsohn (1988) found to increase both forest value (LEV) and optimal rotation age relative to the deterministic Faustmann model. Gong and Löfgren (2007) found that RPS models produce a net LEV gain of up to 80% (depending on growth conditions and discount rate). However, this strategy only affects these values in the short run and cannot influence consumer surplus in the context of natural hazards (Rakotoarison and Loisel, 2017). For example, in France, after beech sales dropped significantly in the last two decades – from approximately 25.5% of ONF (Office National des Forêts) annual turnover in 1999 to 12% in 2015, representing an accumulated loss of approximately €1 billion over sixteen years (about €62 million annually) – (Rakotoarison and Loisel, 2017) argued that more fundamental policy measures are needed, such as the development of new wood products, the reduction of industrial and transportation costs, and the provision of subsidies for forest owners.

On the side of carbon, it has been noted time and again in the Faustmann literature that, all else being equal, increased carbon prices are necessary to incentivize the storing of carbon in living trees (Mpekiri and Papaspyropoulos, 2026a). It has even been said that “increasing carbon prices will thus have an opposite effect to the increased disturbance risk” (Stollery, 2005, p. 108). So far, harvest ages have only been extended when carbon prices are consistently high, implying that landowners bear an opportunity cost when delaying harvest and will do so only when they expect a dependable financial payoff from sequestration (Thompson et al., 2009). For example, under the tax-and-subsidy scheme examined by Thompson et al. (2009) (5% discount rate, $25/m³ timber price), the optimal rotation age increased from 45 years at $10/tC to 77 years at $100/tC; when timber price was $15/m³, it increased from 47 to 148 years. Other policy instruments have also been employed and evaluated. One approach is a carbon tax-and-subsidy scheme in which landowners receive payments for periodic carbon uptake in biomass and face charges at carbon release (harvest and subsequent decay) (Thompson et al., 2009). Another option is to treat landowners as liable for carbon emissions from harvest. Van Kooten et al. (1995) show that rotation ages shorten as landowner liability decreases, because the tax burden at harvest is smaller (van Kooten et al., 1995, as discussed in Thompson et al., 2009). What these studies show is that the effect of carbon pricing and policy on rotation depends not only on the increased profit landowners can make from higher carbon prices, which tends to lengthen rotations, but also on how much landowners are charged when carbon is released at harvest and when those charges are imposed, which tend to shorten rotations when the harvest penalty is lower.

Price credibility problems can undermine the effectiveness of these instruments. Lang (2015) notes that a sharp fall in carbon prices occurred when supply began to exceed demand, linked to rapid growth in the number of established carbon credit projects. In countries with newly emerging but intensely developing carbon markets, such as China (Mpekiri and Papaspyropoulos, 2026a), it seems that even when carbon sequestration appears more profitable than timber-only production, small forest owners can show little interest in changing their management goals (Mpekiri and Papaspyropoulos, 2026a; Zhu et al., 2017), due not only to price uncertainty but also to the additional uncertainty of the costs of sequestration (Mpekiri and Papaspyropoulos, 2026a). This is especially relevant because small-scale private ownership and rights-based management have expanded in many countries (Harrison et al., 2002). This implies that carbon policy based on the assumption of a large, diversified and risk-neutral representative owner may not translate well to settings dominated by small-scale forest owners, and it may fail precisely where adoption potential is largest.

Because many forest owners are risk-averse, risk-transfer instruments appear repeatedly as policy recommendations under climate change (OECD, 2015; United Nations, 1992, 1998; World Economic Forum, 2014) and in the Faustmann literature. Bernetti et al. (2011) have shown that the cost of adaptation strategies can even be higher than the expected damages from various types of risk, which implies that forest owners may need assistance through financial and governmental policies to adopt climate change adaptation measures, especially on the extensive margin which tends to be more costly (Guo and Costello, 2013).

One policy response proposed in the reviewed studies to increase insurance uptake is to subsidize insurance itself. Loisel et al. (2020) propose a public transfer type of insurance scheme, where governments subsidize part of the insurance indemnification of the insurer, which reduces the premium paid by the owner. Zhu et al. (2013) similarly suggest that governments should consider subsidizing small forest owners seeking to purchase insurance for managing their forests for carbon sequestration. This policy not only benefits the various stakeholders, but, if adopted widely (Loisel et al., 2020), could also have climate mitigation effects (OECD, 2015). As more forest areas are covered by insurance, risk becomes more diversified, and so the ability to withstand unpredictable destructive events increases (Loisel et al., 2020) both on the market and individual level.

Forestry predictions rely on large time-scales (decades or more) and inevitably include a lot of uncertainty, which makes the consequences of decisions – such as changing species mixtures – difficult to assess (Hanewinkel et al., 2010). That difficulty is not merely technical. Behavioral economics contains many examples of decision-making under uncertainty that relies on inappropriate heuristics and produces inconsistent or counterproductive choices (Patt and Dessai, 2004).

Commonly, there is a distinction made in forest economics between adaptation on the intensive vs extensive margin. There are two ways in which humans tend to adapt: reallocating inputs with existing capital (which is called the “intensive margin”) and investing in new capital (which is referred to as the “extensive margin”) (Guo and Costello, 2013). From the forest owner’s perspective, the intensive margin usually involves changing the harvest age, while the extensive margin means harvesting the current stock in order to replant an entirely different species (Guo and Costello, 2013). While adaptation on the extensive margin can sound like an extreme measure, the choice of this new species could act as a risk-management policy (Hanewinkel et al., 2010). Transformation strategies – such as shifting from conifers to hardwoods – “have to be designed such that the economic losses that are linked to these activities are minimised” (Hanewinkel et al., 2010, p. 718). As mentioned earlier, the cost of adaptation can be very high (Bernetti et al., 2011) which again reinforces the idea that assistance is needed.

Finally, the adaptation discussion also extends to phytosanitary risk. Since the predominant type of forest in many countries where carbon markets are emerging is plantation (Mpekiri and Papaspyropoulos, 2026a), one implication for policy design is that the particular risks associated with plantation systems need to be taken into account. Brunette and Caurla (2016) describe monocultures as management systems with high sensitivity to phytosanitary problems and list possible responses, beyond conventional chemical treatments, including the selection of genetically modified tolerant or resistant families (Jactel and Brockerhoff, 2007), stump removal for the eradication of infestations and conversion to mixed-species forests. All of these fall under the category of adaptation on the extensive margin, and as such, require increased investment of capital. Another measure is implementing a fallow period at the end of a rotation (Brunette and Caurla, 2016; Jactel and Brockerhoff, 2007), which can be considered an intensive-margin adaptation.

With the intense focus on carbon sequestration, it is easy to forget that forests provide an immeasurable number of other benefits and amenities both for the climate and society. Soil stabilization, biodiversity and habitat conservation, reduction in sedimentation (Englin and Callaway, 1995), water conservation (Creedy and Wurzbacher, 2001) are among the many amenities that forests produce and which impact the environment and climate change in a substantial way. However, they generally do not have a market value, which often makes private timber firms disregard them (Englin and Callaway, 1995).

Rotation length is one management lever that connects directly to both carbon outcomes and co-benefits. Rotation length can significantly influence carbon storage (Harmon and Marks, 2002) but also improve the production of forest amenities. Longer rotations allow more soil organic matter and litter to accumulate (Krankina and Harmon, 2006) which improves soil fertility (Smaill and Garrett, 2016), water regulation (Chen et al., 2018), biodiversity (Bujoczek et al., 2021), erosion control (Rodrigues et al., 2020) and overland flow (Chen et al., 2018). Further, forests with limited profitability under carbon markets or carbon-credit schemes, such as old-growth forests whose carbon increments are negligible (Englin and Callaway, 1995), can be maintained and produce economic value through amenity production. For example, some recreational activities like mushroom picking could allow a forest manager to maintain some over-mature forest stands, despite their little timber or carbon production value (Rakotoarison and Loisel, 2017).

International climate policy has explicitly treated forests as part of mitigation strategy, but the literature raises conditions and limits that matter for how this role should be implemented. Under the Kyoto Protocol of the United Nations Framework Convention on Climate Change (UNFCCC), countries that formed the so-called Umbrella Group (Stollery, 2005) were allowed to count a portion of the carbon sequestered in their forests as an offset to the emission of greenhouse gases. In addition, they had the opportunity to use not only their own forests but also forests located in other countries for this purpose. The same logic also holds in the Paris Agreement, where forests and land use are treated as a mitigation component within national contributions (Grassi et al., 2017). At the same time, emission reduction measures (an adaptation on the extensive margin) could be implemented within a less intense time framework (Stollery, 2005). Additionally, at the time, the Intergovernmental Panel on Climate Change (IPCC) estimated that the potential climate change mitigation measures such as afforestation and reforestation could produce a significant offset in terms of carbon sequestration, approximately the equivalent of 10–20% of estimated fossil fuel-related emissions (Akao, 2011).

As evident in the Faustmann literature, these predictions may have been a bit premature. Caution is repeatedly emphasized regarding how forest-based mitigation is calculated and where it is implemented. Thompson et al. (2009) argue that more research is needed before forest carbon sequestration is deployed as a mitigation strategy at scale, especially if equivalent measures on the front of greenhouse-gas emissions from fossil fuels and industrial expansion are not implemented contemporaneously.

To the best of our knowledge, this is the first qualitative synthesis that focuses specifically on how climate change is addressed in the Faustmann model and how modeling choices translate into implications for optimal rotation, forest value and policy. We highlight an important research and policy sensitivity: common carbon-accounting approaches may omit land-use change implications (notably the albedo effect) relevant to climate impact, highlighting a need for more complete accounting frameworks and careful attention to biome- and latitude-specific effects when evaluating forest-based mitigation or adaptation strategies. Overall, the results of this paper show that there is a diverse array of methods and variables involved when addressing climate change through the Faustmann model. The demands of climate change have pushed research on the Faustmann model beyond the classical version of a timber-profit maximization model.

The nonlinear relationship that characterizes time and carbon sequestration means that we cannot derive universal guidelines for key forestry decisions such as optimal rotation. Therefore, more attention and further research should be dedicated to exploring methods of carbon accounting that encompass all the natural processes involved in forest ecosystems that contribute to global warming. Additionally, the focus of policy should shift from the number of emissions offset by carbon sequestration to policies that address the many and often incompatible needs of the various stakeholders involved.

The discount rate emerged as a central factor affecting the practicality of climate mitigation policies. It affects the importance of post-harvest emissions, the intensity of financial damage done by the increasing natural hazards associated with climate change and the difficulty in realizing extensive margin adaptation strategies which generally put much higher financial strain on the forest owner.

Another key factor was location and forest type. The differences in the albedo effect between boreal and tropical forests have significant implications for policy. Afforestation projects, while widely advocated in recent years, may not be the optimal strategy for greenhouse-gas reduction in high-latitude regions. At the same time, little attention is given in the literature to tropical forests, despite the fact that they contain the largest share of stored carbon and suffer from intense deforestation, which leads to more emissions.

The effectiveness of carbon markets is still the subject of much debate. It seems that there is a gap in policy design, where the theoretical framework of CO₂ payments does not translate to longer rotations and the expansion of carbon sequestration management. Carbon prices are inconsistent and, in many cases, not high enough to provide security to the largely risk-averse body of forest owners, especially the small-scale non-industrial ones. Consequently, there is also reduced incentive to invest in carbon sequestration when timber profitability remains the main driver. It might merit consideration to avoid policies that “punish” stakeholders for emissions, and focus, instead, on providing economic relief through subsidies for extensive-margin adaptation strategies which will become more and more necessary in the future, as well as making the acquisition of insurance more affordable.

Lastly, it is becoming increasingly evident in the literature that forest-based mitigation cannot be evaluated on carbon terms alone, nor treated as a uniform substitute for fossil fuel emissions reduction. International efforts should focus more on addressing the individual characteristics of the forest economy of each country and how those affect the efficacy of the proposed mitigation measures.

Despite the breadth of modeling techniques and policies covered, our review is subject to limitations related to cross-study comparability. Among the dataset of papers selected, comparability is constrained by heterogeneous parameter choices relating to the ecology and economy that form the frame of reference for each paper. Several modeling techniques are demonstrated in region-specific settings, such as Canadian fire regimes, European storm risk, Chinese carbon market conditions and Australian species responses. In addition, the reviewed models often rely on stylized hazard processes and simplified growth functions, and they rarely represent ownership heterogeneity, liquidity constraints or costs that shape adoption. Care was taken to prioritize conclusions that appear robust across settings, but the findings should still be interpreted as context-dependent.

Future research should develop Faustmann extensions that integrate carbon and radiative forcing in a unified metric, and test whether our current understanding of the best climate change mitigation policies changes, as a result, across latitude and forest type. Another avenue for future research would be to quantify how discount rates interact with alternative carbon accounting rules. Emerging research directions in the Faustmann literature that also deserve further attention include dynamic carbon markets and stochastic carbon-price modeling (Ning and Sun, 2019; Zhang et al., 2023), spatial disturbance modeling for regions where forests suffer from frequent and intense fires (Lauer et al., 2020), stochastic dynamic optimization (Buongiorno, 2001; Huang et al., 2022; Tahvonen et al., 2022) and stochastic dynamic programming with geographical heterogeneity on the landscape level (Paarsch and Rust, 2020). These approaches may help address the increasingly complex and multidimensional challenges posed by climate change. Finally, policy modeling should move beyond a representative landowner by incorporating heterogeneous, risk-averse, small-scale owners and evaluate to what degree financial assistance for high-cost, extensive-margin mitigations strategies can enhance and hasten the positive effects of international climate agreements.