To better detect drivers of peatland carbon accumulation rates and patterns


               <jats:p />


Introduction
Climate change is undoubtedly a pressing environmental issue challenging the entire world.To mitigate climate change, various methods have been proposed to reduce net carbon dioxide (CO 2 ) emissions, including taking advantages of the carbon (C) sink capacity of natural ecosystems (Cook-Patton et al 2021).
Peatlands, peat-forming wetlands, are important sink and source components of terrestrial C. Most of the current peatlands were initiated following the last glacial maximum and gradually accumulated thick peat deposits, storing approximately one-third of global soil C (Yu et al 2010).Peat C accumulation, which represents the net balance of C sequestration, emission, and export, is a proxy for evaluating the C sink capacity of peatlands.As peatland C-cycling processes are highly impacted by climatic and environmental conditions, future conditions are likely to cause variations in the C accumulation rates.
To achieve the Paris agreement temperature goals, linking peatland C accumulation to environmental conditions is needed for at least two reasons: (1) predicting future peatland C accumulation rates will provide valuable information for the evaluation of C sink and source capacities of natural systems in the future, and (2) detecting factors that stimulate higher C accumulation rates will provide useful insights for technically increasing the C sink capacity of natural systems.Many studies have tried to link peat C accumulation to environmental conditions, and the overarching aim of the derived relationships is to enable incorporation of peatlands into earth system models.This paper addresses four of the most common aspects that emerge when linking C accumulation to climatic or environmental conditions and proposes some suggestions for future research.

Carbon accumulation rate
For a long time, the 'wrong' way to express accumulation rates has been used to show the temporal patterns of peatland C accumulation.Observed C accumulation rates (so-called apparent C accumulation rates, ACARs) are calculated from bulk density, C content, and chronology.The observed ACAR is commonly used in the literature to reflect the C accumulation rate of a peatland through time (e.g.Heinemeyer et al 2018, Marrs et al 2018).However, the ACAR of a layer is largely dependent on its decomposition level, and thus the ACAR for top layers is likely to be greater than that of bottom layers (figure 1(a)).However, this difference does not mean that the peatland is presently accumulating more C than in the past, because bottom layers have undergone longer decomposition than top layers.This 'ageing' problem makes it challenging to use ACAR to compare the true net rate of C accumulation of a peatland over time (Young et al 2021).Limited recent studies have tackled this issue by applying, for example, the conceptual model of Clymo (1984)   outputs without the long-term decay issue can be better linked to external forcing, i.e. climatic/environmental changes.

Chronology
Chronologies that lack proper datings for top sections have been commonly used to study peatland C accumulation during the recent centuries.In addition to link recent C accumulation to global warming, studying the last ∼200 years is critical for estimating the actual peatland C stocks since the start of industrial global warming.To achieve this, the top layers need to be sufficiently dated to enable a high temporal resolution.Standard dating techniques, such as 14 C ages, are not generally applicable for recent peats, while dating approaches like 210 Pb/ 137 Cs, tephra, and spheroidal carbonaceous particles (SCPs) could provide informative overlapping controls of 14 C-based chronologies, even though they also have limitations (Parry et al 2013).These approaches are either based on short half-live isotopes (e.g. 210Pb) or stratigraphic markers of certain nuclear accident, volcanic eruptions, and atmospheric deposition from industrial sources ( 137 Cs, tephra, and SCPs), which could well constrain the peat chronologies for the past centuries.Yet, many studies have discussed recent C accumulation rates with chronologies established using only 14

Peat characteristics
Detecting the accumulation drivers of peat C accumulation is critical for estimating the future state of peatland C sink capacity.Many studies have linked C accumulation to environmental parameters (Gallego-Sala et al 2018, Zhang et al 2020a).However, so far, most of the studies have not taken into account the peat components/types (generally classified as mosses, herbs, sedges, and shrubs/wood) as drivers on carbon accumulation variations due to, for instance, limited available data.Peat type data can be obtained directly through depth-specific plant macrofossil analysis or indirectly through e.g., plant group-specific chemical compound measurements.It should be noted that modern peatland vegetation cannot always represent the peat types beneath, as peatland vegetation usually undergo successions over time.A few recent studies have compared the observed ACAR for different vegetation phases (Magnan et al 2022, Yang et al 2023).These studies have generally found that past Sphagnumdominated phases have greater ACARs than sedgedominated phases.Different peat types have varying bulk density, C content (Loisel et al 2014), and peat accumulation rate (figure 1(c)), which are direct parameters when calculating C accumulation rates.Therefore, peat component/type variations have great potential to impact peat/C accumulation in addition to climatic and environmental parameters per se.

Environmental drivers
A single climatic or environmental factor (such as temperature, growing degree days, water level) has been used as a potential driver of peat C accumulation.From a statistical point of view, this might cause different (even opposite) correlations between the same parameter and peat C accumulation in different studies, especially if the actual link between the parameter and peat C accumulation is weak (figure 1(d)).For example, both positive and negative linear relationships were detected for C accumulation and mixotrophic microbes in Canadian peat records (Zhang et al 2020b).Natural ecosystems are complex.Modern peatland C flux studies have revealed that the causal links between peatland CO 2 and methane gas exchange processes and external factors are interconnected (Laine et al 2022).Peat C accumulation, the net balance of several C-cycling processes, is therefore highly likely to be simultaneously driven by several factors, even though some factors might have an overriding control (Evans et al 2021).Exploring the causal links between C accumulation and more than one factor might also help to explain the different patterns that are usually observed for different peatland types and regions.

Recommendations
We argue that while studies on climatic and environmental drivers of peatland C accumulation are actively ongoing, they often have clear limitations that prevent full understandings of the causal links between peatland C accumulation and elemental interlinked drivers, thus impeding comprehensive prediction of the role of peatlands as important terrestrial C stocks.Efforts are needed to address these issues, and therefore we propose the following actions: (1) removal of the autogenic trend when comparing recently formed peat with older material that accumulated centuries or even millennia earlier; (2) improvement of the chronologies for young peat sections when focusing on peat C accumulation over recent centuries or decades; (3) incorporation of the impact of peatland type, peat components, and peatland vegetation succession when studying the impacts of climate and environment on peat C accumulation; (4) consideration of the interactive relationships between external factors and peat C accumulation rather than linking only one particular factor to C accumulation rates.
and a peatland carbon flux reconstruction model (Yu 2011) to remove this autogenic peat accumulation trend and to reconstruct past C fluxes (e.g.Zhang et al 2020a, Liu et al 2022, Yang et al 2023).These models could make the different peat layers comparable in terms of temporal carbon accumulation dynamics and the

Figure 1 .
Figure 1.Diagrams presenting the four 'problematic' topics.(A) Two examples showing the differences between the observed apparent carbon accumulation rate (ACAR) patterns and non-autogenic carbon accumulation rate (CAR) patterns; the observed ACAR shows a distinct increase in the top sections, while non-autogenic CAR suggests that both top and bottom sections have higher CAR (Zhang et al 2020a).(B) Two examples showing the differences in ACAR based only on 14 C dates, and 14 C and 210 Pb dates (Sanderson 2016).(C) Two examples showing that the more Sphagnum remains, the faster peat accumulated (see the supplementary materials for details).(D) Example showing the differences of the coefficient (coeff.) between non-autogenic CAR and water-table depth (WTD) investigated using linear regression (LR) and structure equation modelling (SEM); TJJA: June-July-August temperature, peat: peat property, plant: plant type composition (see the supplementary materials for details; Zhang et al submitted).
C dates (e.g.Piilo et al 2019, Liu et al 2022), which might cause uncertainties in the observed C accumulation rates.Studies exist that have tested this by calculating and comparing the ACAR of the same core using chronologies established with and without 210 Pb dating.The results show much lower recent C accumulation rates for those cases that used only 14 C dating and had no 210 Pb dating (Sanderson 2016) (figure 1(b)).
Liu Z and Bai E 2023 Comparison of carbon and nitrogen accumulation rate between bog and fen phases in a pristine peatland with the fen-bog transition Glob.Change Biol.29 6350-66 Young D M, Baird A J, Gallego-Sala A V and Loisel J 2021 A cautionary tale about using the apparent carbon accumulation rate (aCAR) obtained from peat cores Sci.Rep. 11 9547 Yu Z C 2011 Holocene carbon flux histories of the world's peatlands: global carbon-cycle implications Holocene 21 761-74 Yu Z, Loisel J, Brosseau D P, Beilman D W and Hunt S J 2010 Global peatland dynamics since the Last Glacial Maximum Geophys.Res.Lett.37 L13402 Zhang H, Amesbury M J, Piilo S R, Garneau M, Gallego-Sala A and Väliranta M M 2020b Recent changes in peatland testate amoeba functional traits and hydrology within a replicated site network in Northwestern Québec, Canada Front.Ecol.Evol.8228 Zhang H, Piilo S, Guo Z, Zhao Y, Laine A, Korrensalo A and Väliranta M Drivers on carbon accumulation vary along the hydrological gradient of a subarctic patterned peatland submitted Zhang H, Väliranta M, Piilo S, Amesbury M J, Aquino-Lopez M A, Roland T P and Tuittila E S 2020a Decreased carbon accumulation feedback driven by climate-induced drying of two southern boreal bogs over recent centuries Glob.Change Biol.26 2435-48