Not just semantics: CO2 fertilization can be a disturbance leading to worldwide forest degradation

Abstract

Dampak fisiologis dari peningkatan CO₂ atmosfer (pemupukan CO₂) pada hutan utuh umumnya dilihat sebagai suatu proses yang dapat melindungi hutan dari dampak perubahan iklim. Akan tetapi, pemupukan CO₂ juga dapat menyebabkan perubahan yang merugikan pada kondisi hutan, seperti perubahan komposisi spesies dan berkurangnya ketersediaan kelembapan ekosistem. Oleh karena itu, kami berpendapat bahwa dampak pemupukan CO₂ merupakan gangguan yang menyebabkan perubahan pada fungsi hutan di seluruh dunia dan menyebabkan degradasi ekosistem ini dalam intensitas rendah. Hal ini menunjukkan bahwa bahkan hutan yang paling terpencil di dunia telah diubah oleh tindakan manusia melalui pemupukan CO₂, dan tanggung jawab atas integritasnya harus dipikul bersama secara global.

INTRODUCTION

The direct physiological effect of increased atmospheric CO₂ concentration, also known as CO₂ fertilization effect, is long known to potentially enhance carbon assimilation and lead to increased water-use efficiency by plants, as revealed by a number of laboratory studies with potted plants, field-based open-top chambers, ecosystem-scale experiments using free-air CO₂ enrichment (FACE) technology and remote sensing (Ainsworth & Long, 2005; Fortirer et al., 2023; Walker et al., 2021). One of the possible cascading effects of increased carbon assimilation is the accumulation of that additional carbon in plant biomass, particularly wood and derived litter, which can increase ecosystem carbon storage due to their longer residence times.

Forests worldwide have sequestered carbon at constant rate over the last three decades (3.6 ± 0.4 PgC year⁻¹) (Pan et al., 2024). Statistical analyses on long-term field plots indicate CO₂ fertilization as the most likely explanation for that carbon sink observed for several decades in worldwide forests (Hubau et al., 2020; Pan et al., 2024; Walker et al., 2021). Nevertheless, the effects of CO₂ fertilization on net primary productivity and carbon uptake are, in general, stronger in tropical rather than in boreal forests due the temperature dependence of Rubisco kinetics (Hickler et al., 2008). The strength of the carbon sink in intact tropical forests has been reduced by 31% since the 1990s apparently due to increased temperature, more frequent droughts, and losses in intact forest area (Hubau et al., 2020; Pan et al., 2024). European boreal forests have also been supposedly affected by CO₂ fertilization effect, as well as longer growing seasons (Kauppi et al., 2022; Pan et al., 2024). The same seems to hold true also for forests in eastern United States, where climatic changes, namely droughts, have not been severe (Davis et al., 2022; Hogan et al., 2024).

Such an enhanced carbon sink caused by CO₂ fertilization effect has been observed, although temporarily, in planted temperate forest FACE experiments (Norby et al., 2010), in a mature temperate forest FACE (Norby et al., 2024), but not in a savanna woodland in Australia (Yang et al., 2020). No direct experiment-based evidence of such enhanced carbon storage driven by CO₂ fertilization exists for tropical or boreal forests yet (Norby et al., 2016).

The balance between tree recruitment/growth and tree mortality/turnover is determinant for the evolution of that carbon sink. Tree recruitment/growth is consistently rising in Amazon and Central African forests, apparently due to CO₂ fertilization, but in the Amazon, it is increasingly being outpaced by tree mortality/turnover supposedly as a reaction to increasing temperature (Hubau et al., 2020). Also, CO₂ fertilization has long been suggested—though not properly proved—as the cause of increased tree turnover in tropical forests (Phillips & Gentry, 1994), as well as favoring the occurrence of lianas in these ecosystems (Phillips et al., 2002). Indirect (radiative) effects of increased CO₂ such as increased temperature and more frequent droughts may also play a role in such altered turnover.

Inventory data and model data integration show that enhanced growth driven by CO₂ fertilization might be offset by decreased longevity in temperate forest trees (Bugmann & Bigler, 2011; Büntgen et al., 2019), though exceptions to that tradeoff are common across all forest types (Cailleret et al., 2017; Rüger et al., 2020). More recent and robust observational evidence suggest that CO₂ fertilization is driving a shift within Amazonian tree communities to large-statured species, even though such large-statured species tend to be the most affected by droughts (i.e., CO₂ fertilization effects are apparently overwhelming the effects of more frequent droughts in the region in that specific ecological process) (Esquivel-Muelbert et al., 2019).

All in all, the role of CO₂ fertilization on forest tree mortality, turnover, and the growth–survivorship tradeoff is still uncertain (Maschler et al., 2022; Walker et al., 2021), with the indirect (radiative) effects of increased atmospheric CO₂ likely playing an important role too (Hubau et al., 2020). Because of that, the attribution of the long-term observed carbon sink in worldwide forests majorly to CO₂ fertilization (Pan et al., 2024) is permeated by uncertainties as well, especially given the discrepancies found so far in ecosystem-scale FACE experiments and long-term inventories, tree rings, and remote sensing data (Maschler et al., 2022). Notwithstanding, taking the available evidence into account, it seems reasonable to at least hypothesize that CO₂ fertilization can have a key role in altering carbon fluxes, community dynamics, and the structure of tropical, temperate, and boreal forests (Bugmann & Bigler, 2011; Esquivel-Muelbert et al., 2019; Hubau et al., 2020; Laurance et al., 2004; Pan et al., 2024).

On the other hand, the increased water-use efficiency usually resulting from CO₂ fertilization is related to reductions in stomatal conductance (De Kauwe et al., 2013; Gimeno et al., 2016; Norby & Zak, 2011) and to potential decrease in canopy-level transpiration of forests (De Kauwe et al., 2013). Such an increased water-use efficiency is highly variable among different experiments in temperate forests, sometimes species-specific, smaller in needle-leaf trees, and can potentially be offset by increases in total leaf area (Norby & Zak, 2011; Ward et al., 2018). Modeling exercises indicate that such a potential reduction in the flux of moisture from the vegetation to the atmosphere over large expanses of tropical forests, namely the Amazon forest, can substantially modify the hydrological cycle and cascade into basin-wide reduced precipitation totals (Kooperman et al., 2018; Langenbrunner et al., 2019; Sampaio et al., 2021). In fact, the physiological effects of increased CO₂ (CO₂ fertilization) on tree transpiration dominate multimodel mean precipitation projections over the Amazon in the future (Richardson et al., 2018).

Reduced precipitation is long known to have low-intensity but a large-scale effect on tree mortality and carbon emissions to the atmosphere (Lewis et al., 2011). Long-term reductions in soil water, such as represented in the above-mentioned modelling exercises could potentially, have an even worse effect on forest functions and structure, increasing the mortality of selected functional groups of trees, especially large trees (Rowland et al., 2015). This sort of widespread changes in forest community composition is already being observed in several sites in the Amazon forest, a process that is attributed to CO₂ fertilization and the concomitant lengthening of the dry season, with dry-affiliated taxa replacing wet-affiliated taxa (Esquivel-Muelbert et al., 2019).

The body of evidence so far suggests that CO₂ fertilization is a process that can possibly lead to considerable alteration of forest water cycle. In the case explored above, CO₂ fertilization leads to reduced moisture flux from the vegetation to the atmosphere, with consequent reduction in large-scale precipitation, which feedbacks to the forest in the form of selective mortality of certain functional groups of trees, ultimately leading to changes in tree community composition (Esquivel-Muelbert et al., 2019).

In that sense, the CO₂ fertilization effect could be considered a low-intensity disturbance leading to degradation of worldwide forests, namely due to its impacts on the water cycle and likely on carbon balance and community composition of these ecosystems. A recent review paper on Amazon forest degradation defines degradation as:

(…) a transitory or long-term (10 to 1000-year time scale) deleterious change in forest condition. Condition includes functions, properties, or services such as, but not restricted to, carbon storage, biological productivity, species composition, forest structure, local atmospheric moisture or uses or values of the forest to humans. (Lapola et al., 2023).

Other definitions are similar to that and include:

(…) a negative trend in land condition, caused by direct or indirect human-induced processes including anthropogenic climate change, expressed as long-term reduction or loss of at least one of the following: biological productivity, ecological integrity or value to humans (Olsson et al., 2019).

Degradation implies a persistent reduction of some attribute relative to a preferred (nondegraded) condition. (…) forest degradation involves changes in the structure, composition, or function of a forest. The outcomes of these changes are subject to alternative interpretations and values, which might be based on productive functions (e.g., goods), biodiversity, carbon storage, or other ecosystem services. The preferred condition might reflect a desire to return to some perceived historical condition, often assumed to be relatively undisturbed by humans. Perceptions of degradation and recovery are, however, social constructs subject to cultural and ethical norms (Ghazoul & Chazdon, 2017).

Whether such human-driven ecological changes linked to forest degradation are “deleterious” or “negative” is therefore a matter of “social constructs subject to cultural and ethical norms”. In this analysis, we follow a conservationist perspective, considering the pre-industrial state of these forests, when they were little affected by the direct and indirect effects of increased atmospheric CO₂. Surely some of these changes, like large-scale intentional shifts in species composition, are not seen as deleterious or negative by some actors such as indigenous people (Clement et al., 2015). The latter interpretation however should not be extended to actors undertaking illegal activities like as timber extraction.

Following the aforementioned conservationist perspective, CO₂ fertilization is a potential disturbance and its ultimate effect on the water cycle is the negative change in forest function, compared to a pre-industrial condition (when atmospheric CO₂ levels were relatively stable). The same potentially applies to other changes driven at least partially by CO₂ fertilization, such as the probable changes in community composition referred above and its cascading effects on the forest carbon balance (Esquivel-Muelbert et al., 2019; Hubau et al., 2020; Laurance et al., 2004). CO₂ fertilization has a lower intensity of impacts compared to fire or edge effects (Lapola et al., 2023). But it induces a feedback loop by enhancing carbon emissions to the atmosphere (which opposes the negative feedback normally associated with CO₂ fertilization; Friedlingstein, 2015), increasing the CO₂ fertilization disturbance over these forests (Figure 1). Moreover, such changes in forest properties have the potential to also alter the provision of ecosystem services such as food, wood, and fiber, thus making it possible that forest inhabitants themselves classify the forest as “degraded,” as predicted in the definitions of degradation given above.

[IMAGE OMITTED. SEE PDF]

One can argue that the press of CO₂ over tropical forest is not short-lived and that formal definitions account that (ecological) disturbance is relatively discrete in time (Newman, 2019). However, we understand that it can indeed be called a disturbance given that it disrupts certain aspects of the functioning of tropical forests in a time horizon that is relatively short in ecological terms, i.e., a few to several decades, which is way shorter than the upper time limit (1000 years) used in the definition of degradation above. Moreover, as a process causing a permanent disturbance, it has the potential to lead the system to a steady state different from the original (Rykiel, 1985). And, of course, the CO₂ fertilization effect is undoubtedly human-caused.

The effects of CO₂ fertilization on worldwide forests are, nevertheless, still permeated by uncertainties (Maschler et al., 2022; Walker et al., 2021). For example, short-term experimental evidence suggests that leaf-level reductions in transpiration in tropical forest understory can be compensated by an increase in leaf area index (Damasceno et al., 2024), resulting in potentially no changes in the integrated flux of moisture from the forest canopy to the atmosphere. The attribution of the causes of the worldwide forests carbon sink, increased turnover and tree mortality, and compositional changes to CO₂ fertilization is so far the result of probabilistic analyses, still lacking mechanistic explanations and more direct evidence (Esquivel-Muelbert et al., 2019; Hubau et al., 2020; Pan et al., 2024). One logical way of obtaining such direct evidence would be through FACE experiments run over meaningful ecological scales (>80 years) that could properly capture tree recruitment and mortality dynamics and potential community shifts, but seems financially impractical today and certainly too long for guiding short-term policy.

But in case we can indeed call CO₂ fertilization a disturbance and its impacts—for example on the water cycle or on community composition—as changes attributable to forest degradation; then, it is reasonable to think that something close to 100% of the Amazon forest is somehow degraded, and not the 38% previously estimated analyzing other disturbances (fire, edge effects, logging, and extreme droughts) (Lapola et al., 2023). The same line or argument applies to other “intact” forests and raises the question whether the “wild woodlands” anthrome category is appropriate (Ellis et al., 2021). While on the one hand this seems only a matter of semantics, it has a key relevance from the political perspective. Defining CO₂ fertilization as a disturbance leading to forest degradation (in addition to the already recognized impacts of the indirect effects of increased CO₂ on forest degradation; Lewis et al., 2011) puts in evidence the shared responsibility of other countries (other than forest countries), namely top emitters, over the integrity of worldwide forests. Curbing deforestation and certain disturbances (e.g. forest fires or illegal timber extraction) that lead to forest degradation is certainly a major task for the countries hosting these forests. But the curbing of others disturbances that lead to degradation and which are related to global climate change, like extreme droughts (Lapola et al., 2023) or CO₂ fertilization, is something that forest countries cannot attain alone, but depends on a concerted effort of other countries in curbing greenhouse gas emissions.

AUTHOR CONTRIBUTIONS

David M. Lapola was responsible for the conceptualization of this manuscript. All authors wrote the original draft, revised subsequent drafts and contributed additional ideas. Barbara R. Cardeli produced the manuscript figure. David M. Lapola and Carlos A. N. Quesada acquired funding that made this study possible.

ACKNOWLEDGMENTS

This manuscript derived from the “Anthromes, CO₂, and terrestrial carbon – from deep past to net zero” symposium, held in Potomac, MD, USA, between March 27 and 30, 2023, and funded by the New Phytologist Foundation. We are grateful to the organizing committee of that symposium. We also thank C. Körner, W. Hubau, and other colleagues who provided comments at the presentation of this work in session BG3.31 of the European Geosciences Union General Assembly 2024. We are also grateful to the two anonymous reviewers for taking their time to attentively review this study. Your comments surely improved this manuscript. DML thanks the support of Brazil's Conselho Nacional de Desenvolvimento Científico e Tecnológico—CNPq (grant 309074/2021-5) and Fundação de Amparo à Pesquisa do Estado de São Paulo—FAPESP (grant 2022/14271-5). AE-M is supported by the UKRI TreeScapes MEMBRA (NE/V021346/1), the Royal Society (RGS\R1\221115), the NERC-NSF Gigante project (758873), and the CESAB Syntreesys Project. BRC and BFR acknowledge the funding provided by FAPESP (grants 2019/04223-0 and 2022/00194-9), and JVM acknowledges the funding granted by Brazil's Coordenação de Aperfeiçoamento de Pessoal de Nível Superior—CAPES (grant 001). CANQ is funded by CNPq (grant 309917/2017-4). CHLS-J was also supported by CNPq under the project YBYRÁ-BR: Space-Time Quantification of CO₂ Emissions and Removals by Brazilian Forests (grant 401741/2023-0). All authors acknowledge Brazil's Ministry of Science, Technology and Innovation—MCTI and UK's Foreign, Commonwealth and Development Office—FCDO for their support to the AmazonFACE Programme.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

DATA AVAILABILITY STATEMENT

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

References

Ainsworth, E. A., & Long, S. P. (2005). What have we learned from 15 years of free‐air CO2 enrichment (FACE)? A meta‐analytic review of the responses of photosynthesis, canopy properties and plant production to rising CO2. New Phytologist, 165(2), 351–372. [DOI: https://dx.doi.org/10.1111/j.1469-8137.2004.01224.x]