Abstract
Loss of peat through increased burning will have major impacts on the global carbon cycle. In a normal hydrological state, the risk of fire propagation is largely controlled by peat bulk density and moisture content. However, where humans have interfered with the moisture status of peat either via drainage, or indirectly via climate change, we hypothesise that its botanical composition will become important to flammability, such that peats from different latitudes might have different compositionally-driven susceptibility to ignition. We use pyrolysis combustion flow calorimetry to determine the temperature of maximum thermal decomposition (Tmax) of peats from different latitudes, and couple this to a botanical composition analysis. We find that tropical peat has higher Tmax than other regions, likely on account of its higher wood content which appears to convey a greater resistance to ignition. This resistance also increases with depth, which means that loss of surface peat in tropical regions may lead to a reduction in the subsequent ignitability of deeper peat layers as they are exposed, potentially resulting in a negative feedback on increased fire occurrence and severity.
Similar content being viewed by others
Introduction
Peat, unlike vegetation, burns almost exclusively by smouldering (non-flaming) combustion1,2,3, resulting in greater emissions of CO and CH₄ and having a higher climate forcing potential4, as well as greater emissions of toxins and particulates5. Despite covering 2‒3% of the Earth’s land surface5,6, peatlands store around a quarter to a third of global soil carbon5,7, probably exceeding the carbon content of global vegetation5,6, and equalling or exceeding that of the pre-industrial atmosphere8,9. Increased burning of peatlands due to climate warming5,10 and anthropogenic peatland degradation6,11,12 thus risks substantial impacts on the carbon cycle and climate system13,14, as well as other harmful environmental impacts15,16.
Despite the short-term climate forcing associated with CH₄ production, peat accumulation has a net cooling effect on the global climate in the long term through carbon sequestration5,9,17. Throughout the Holocene, peatlands have acted as a long-term carbon sink18, but are now increasingly switching from sink to source, due to changes in climate, land use, and fire regimes6. Even though pristine peatlands globally may increase their carbon accumulation potential with climatic warming during this century, any such increase will weaken from c.2100 when enhanced decomposition may exceed enhanced photosynthesis8. The response of peatlands to warming is, however, latitude-dependent and intact tropical peatlands are expected to experience a decrease in carbon accumulation rates in the future due to increased rates of respiration8. However, changes in the balance between productivity and respiration are likely to be negligible in comparison with reductions in the area of intact peatlands, especially in the tropics, where they are subject to extensive deforestation and drainage for agriculture6,19. This is of considerable concern because the stability of peatlands is highly dependent on hydrological conditions, and exposure of the peat to oxygenation by drawdown of the water table, either artificially or as a result of drought, results in peat loss, either by peat mineralisation (decomposition) or by burning6,20. Under peat-forming conditions, high moisture content prevents ignition. Thus undisturbed peatlands are largely free of fire in the tropics5,21, while high-latitude peatlands burn with limited severity under natural conditions5,22,23,24,25. However, lowering the water table, for example, using agricultural drainage schemes, exposes flammable (i.e. non-saturated) peat and this is known to be a strong driver of increased fire occurrence in tropical peatlands6,21,26, and is associated with increased burn severity in high latitude peatlands27 where drying due to climate change may also lead to vegetation changes followed by progressive peat loss through repeated burning at higher fire frequencies28. Therefore, carbon release from peat fires represents an important component of the human-altered carbon cycle29,30.
The smouldering combustion of peat fires1,2,3,31 is markedly different in behaviour compared with aboveground vegetation fires, which are dominated by flaming combustion. The high porosity of peat allows ingress of oxygen and in situ oxidation of the fuel allowing smouldering combustion. When peat burns the fires are slow-moving, with spread rates of 1‒10 cm h−132,33, and of low temperature, with typical peak temperatures of 500‒700 °C32. However, the low thermal conductivity of peat minimises heat loss, so that despite the relatively low temperature, combustion is highly persistent1,32. This enables peat fires to cover extensive areas and penetrate deep into the ground1, burning for many months or even years1. The propagation of smouldering fires in peat is largely controlled by its bulk density and moisture content34,35. However, large peat fires most often occur in areas that have been extensively drained for land use6,21. For example, degraded tropical peat has been found to be as dry as 20% moisture36, and with moisture content no longer high enough to suppress burning, the controls on flammability may be shifted toward variations in peat composition. However relatively little is known about the influence of the botanical constituents within peat on ignition37.
Peat combustion is normally initiated by the heat flux from a flaming vegetation fire, which in tropical regions is often anthropogenic21,38,39. Although flaming and smouldering are distinct—smouldering is a heterogenous reaction of solid fuel with an oxidiser whilst flaming is a homogeneous reaction of gaseous fuel with an oxidiser—both fire types begin with pyrolysis36, which is the thermal decomposition of materials at elevated temperature. The thermal resistance of peat to the energy flux from surface fires is important in understanding the ignitability of peat. The thermal degradation properties of peat have been shown to vary considerably with both botanical composition and elemental composition37. Therefore, differences in peat composition, for example between tropical and boreal environments38, are predicted to result in different ignition responses. For example, it has been suggested that tropical peat, having higher wood content, is associated with higher calorific values and greater flammability than high-latitude peat38. Despite these observations, there has been no attempt to study the variation in resistance to ignition of peat at the global scale, nor contrast in detail differences between tropical and boreal peats. To fill this knowledge gap, we obtained 152 peat samples from 55 sites covering arctic, boreal, temperate and tropical regions, including surface peat from all regions, and subsurface samples from the extreme latitude groups (arctic and tropical regions) (Table S1; Fig. 1) and assessed both their botanical composition and flammability.
Site details are given in Table S1. Map created using ArcGIS® software by Esri. ArcGIS® and ArcMap™ are the intellectual property of Esri and are used herein under license. Copyright © Esri. All rights reserved. Basemap credits: Esri, USGS.
A pyrolysis combustion flow calorimeter is used to heat each sample at the same rate and measure the temperature at which the maximum rate of thermal decomposition (Tmax) occurs in each type of peat. Tmax approximates the ignition temperature and is thus a key parameter in determining the potential of a surface flaming fire to cause ignition of the peat below. Tmax therefore serves as a measure of a material’s thermal recalcitrance, because material that is more resistant to heat will require a greater heat flux and requires a greater temperature to be reached before it will ignite. We couple these data to a composition analysis, to determine the plant constituents forming the peat. We show that boreal and temperate Sphagnum-dominated peat is the least resistant to thermal decomposition whilst tropical peat is considerably more resistant to ignition. Moreover, removal of surface peat in tropical regions will lead to a reduction in material ignitability as deeper peat layers are more thermally recalcitrant. This increase in Tmax with depth in tropical peat should result in a negative feedback on increased smouldering fire activity.
Results and discussion
How variable is peat resistance to thermal decomposition?
Tropical peat displayed higher Tmax (mean 420 °C) than arctic (354 °C), boreal (345 °C) or temperate (351 °C) peat (Fig. 2). Differences between tropical and all other latitude groups were significant (p < 0.001), and differences between all extratropical latitude groups were not significant (p > 0.6) [1-way ANOVA; Tukey’s pairwise]. In the tropical peat, mean Tmax was 403 °C for the upper, aerobic layer, compared to 431 °C for the lower, waterlogged, compacted layer, and this difference was significant (p = 0.004). Extratropical peat showed a mean Tmax of 343 °C for the aerobic layer and 354 °C for the anaerobic layer, and the difference was not significant (p = 0.058). This indicates that tropical peat requires a higher temperature (or greater heat flux) to reach peak pyrolysis rate and therefore ignition, than the temperate, boreal or arctic peats tested. The tropical peats tested therefore have intrinsically greater fire resistance than the peats we tested from higher latitudes, and the resistance of the tropical peats to thermal decomposition increases with depth. This change with depth was not evident in higher latitude peat.
Drivers of resistance to peat ignition
The botanical composition analysis (Fig. 3) shows that tropical peat (n = 20) consisted primarily of wood, roots, and unidentifiable organic matter in varying proportions, with minor components (3‒18%) of undifferentiated aboveground biomass. This is in keeping with the normal forest-based origin of peat in lowland tropical climates40. Sedges (Cyperaceae) were also present in samples from a single site, Oropel Swamp, Panama (48% in aerobic layer, 43% in anaerobic layer), which has the highest absolute latitude of the tropical group, and despite its tropical climate41 may represent a transition to a subtropical peat composition, which is typically sedge- rather than tree-dominated40. Mosses were not present in any of the tropical samples, and no visually discernible differences in the botanical constituents were evident between aerobic and anaerobic samples. Temperate aerobic layer peat (n = 12) had Sphagnum as its primary component, except samples from Slieveanorra, Ireland (sedge peat) and Shestakovo, Russia (wood and root peat) which contained no identifiable Sphagnum or other mosses. Sphagnum mosses are normally the dominant peat-forming vegetation in temperate regions42. The peat from Slieveanorra has previously been characterised as a Sphagnum-containing sedge peat43, and it may be that its high degree of humification43 had obscured some Sphagnum content in our samples. Shestakovo, in southern Siberia, experiences a highly continental climate44 which is associated with the formation of moss-free peat substantially formed from the underground parts of sedges and other Poales40. The Sphagnum-containing peat of the other temperate samples ranged from 41‒94% mosses (26‒94% Sphagnum) with varying additional components of which only roots were present in all samples. Boreal aerobic layer peat (n = 12), which also typically has Sphagnum as its main component42, was more uniformly moss-dominated, with 49‒98% mosses (43‒98% Sphagnum). Arctic peat (n = 9) was variously dominated by Sphagnum, sedges, wood or roots, with some aerobic layer peats having high wood and root contents but their corresponding anaerobic layer samples having a higher sedge content, which may reflect a change in vegetation cover over time45. We found that while peat composition is heterogeneous within each latitude group, there is a clear latitudinal division between generally root/wood-dominated tropical peatlands, and generally Sphagnum-dominated extratropical peatlands, although this may reflect some sampling bias (for example, lack of Papyrus-dominated tropical peatland samples in our collection).
When grouped by peat composition, Tmax was markedly higher in humified (decomposed) peat samples (mean 414 °C) and wood/root peat samples (400 °C) than in sedge (360 °C), Sphagnum/sedge (341 °C), or Sphagnum (340 °C) peat samples (Fig. 4). Tmax differed significantly (p < 0.005) between humified and all other peat categories except wood/root (p = 0.779); wood/root also differed significantly from Sphagnum and Sphagnum/sedge peat samples (p < 0.001) but not from sedge peat samples (p = 0.062). This indicates that humified and wood/root peat requires greater heat fluxes to ignite than sedge- or moss-dominated peat. Across all five compositional categories, the samples with Tmax > 370 °C originate exclusively from tropical climates, and those with Tmax < 370 °C are mostly from higher latitudes, with only 2 of 35 samples being of tropical origin.
Relationships between Tmax and individual peat components are shown in Fig. 5. Correlations (n = 53) were quantified using Spearman’s ρ due to the non-linear nature of the relations. Tmax was positively correlated with content of wood (ρ = 0.60, p ≪ 0.001), roots (ρ = 0.44, p = 0.001), and undifferentiated aboveground plant remains (ρ = 0.68, p ≪ 0.001), and negatively correlated with mosses (ρ = −0.66, p ≪ 0.001). As shown in Fig. 5, both higher Tmax and higher wood, root and undifferentiated aboveground biomass content, are associated with tropical origin of the peat, which is also associated with zero moss content in all cases. This suggests a latitude or climate-based control on both composition and Tmax. Relationships of Tmax to latitude and climate are shown in Fig. 6 (n = 152). Tmax was negatively correlated with absolute latitude (ρ = −0.38, p ≪ 0.001), and positively correlated with mean temperature (ρ = 0.41, p ≪ 0.001).
a: absolute latitude, b: mean surface temperature 1961‒1990 (ref. 82).
Thermal recalcitrance of the tropical peat samples relative to the temperate, boreal or arctic samples likely results from differences in lignin/holocellulose ratios. Lignin content of plant matter is a determinant of chemical and biochemical recalcitrance46,47 and thermal stability48,49. Tropical peat tends to be dominated by woody material38,50, and therefore will have a high lignin content51,52. High-latitude peat samples are typically dominated by Sphagnum and Cyperaceae50 and should therefore have a higher holocellulose content. As mosses do not contain lignified cell walls53, peat composed almost entirely of Sphagnum will contain negligible lignin. Our composition analysis shows a very strong tendency toward tropical peat samples being wood/root-dominated, and temperate and especially boreal peat samples being Sphagnum-dominated. However, the arctic peat samples are of varying composition, although the majority are dominated by lignin-containing components.
That lignocellulose composition is important in determining Tmax is supported by the correlations of Tmax with different peat components. Across all samples, Tmax has a moderate positive correlation with wood content (ρ = 0.60, p ≪ 0.001) (Fig. 5a) and root content (ρ = 0.44, p = 0.001) (Fig. 5b). Mosses, which are devoid of lignin, show a negative correlation with Tmax (ρ = −0.66, p ≪ 0.001) (Fig. 5d), evidencing a binary relationship in which Tmax exceeds 370 °C only in the absence of moss. The difference in Tmax between tropical and higher latitude peat samples is likely a direct result of the high content of woody material, and thus lignin, in tropical peat samples. This is also supported by the slightly higher Tmax of the woodier arctic peat samples than the boreal or temperate ones. However, peat samples with high wood/root content nonetheless have lower Tmax where they originate from high-latitude sites (Fig. 4), and tropical peat samples generally have the highest Tmax regardless of peat type (Fig. 4). The distribution of our tropical and higher latitude peat samples across the five composition categories suggests that the climatic origin of the samples may be the controlling factor and that the relationship between Tmax and peat type arises from the fact that moss-dominated peatlands do not form in low altitude tropical conditions.
The presence of peatlands in tropical regions, despite higher temperatures leading to faster decomposition, may be due in part to their high woody content inhibiting decomposition50. Greater aromatic content (lignins, tannins, and humic substances) and lower carbohydrate content make tropical peat more chemically recalcitrant than arctic and boreal peat50. This is partly attributable to the high lignin content of woody parent vegetation, but also to higher temperatures enhancing humification so that an initial phase of decomposition leaves the peat in a more recalcitrant state50. Therefore, the high woody content of tropical peat may be due not only to predominantly woody vegetation but to selective incorporation of more ligneous material into the peat, whose chemical recalcitrance counteracts the effect of high temperatures on decomposition rates. While both tropical and some arctic peat may have high woody content, in a tropical environment the more labile components will have been preferentially decomposed, increasing the recalcitrance of the remaining material prior to preservation under anoxic conditions (while ligneous material maintains the structural integrity that allows the material to be identified as wood). Therefore, the high Tmax of tropical peat may largely represent a concentration of (bio)chemically recalcitrant material during decomposition, resulting in peat which also has higher thermal resistance (i.e. reduced ignitability). This interpretation is supported by the fact that Tmax is consistently lower (mean 341 °C) where moss is present in any quantity. As moss is the only peat component free of lignin, and peat samples vary widely in moss content, a direct effect of the lignin/holocellulose ratio on Tmax would be expected to show a more linear relationship as is seen for wood and root content. Instead, Tmax varies from ~320‒370 °C if moss is present (which corresponds entirely to higher latitude sites), and approximately 330‒470 °C where moss is absent (27 out of 33 sites being tropical). This suggests that the relation between Tmax and moss content is likely to be a secondary effect of an underlying relationship between Tmax and tropical or extratropical peat origin.
Implications for tropical forest peatland vulnerability to fire
Our results have important implications for understanding the vulnerability of drained tropical peats to ignition. Tropical peatlands are concentrated in southeast Asia (with 250,000 km² out of 400,000 km² 15,54), where they have been subject to extensive drainage and deforestation in recent decades54,55, and are especially at risk of fire11. In tropical peat swamp forests, peat preservation is dependent on forest cover, which stabilises the peat and maintains a moist microclimate, while the forest cover is itself dependent on the integrity of the peat, including for structural support and hydrological regulation56. Therefore primary peat swamp forest does not easily recover from disturbance, and secondary growth is susceptible to domination by ferns and shrubs with higher fire risk26. Fire may be employed for land clearance, as has occurred widely in the peat swamp forests of insular southeast Asia6, which this century have seen drastic increases in fire frequency and severity6, and lost coverage at a rate of 2.25% yr−1 from 2000 to 20106. Drainage and deforestation also reduce the fire resistance of the surrounding forest due to drawdown of the water table, commonly extending for several hundreds of metres from the forest edge57,58, and effects on microclimate6. Increases in fire frequencies have been exacerbated by climatic changes55.
In the peat swamp forests of southeast Asia, there appears to be both positive and negative feedback associated with fire. A greater frequency or severity of burning progressively reduces tree regrowth and shifts species composition toward a more flammable fern- and sedge-dominated community6,59. Therefore, an initial fire also results in greater subsequent ignition risk due to reduced humidity after the loss of tree cover11, and changing fire behaviour due to fuel loading from dead but unburned trunks, and fallen trees resulting from loss of soil integrity11. Negative feedbacks also operate, in which fire frequency or severity may be reduced by the effects of previous fires. Several such feedbacks have been suggested in relation to fuel loading. Depletion of fine surface fuels can reduce surface fire intensity and may limit fire spread60. In peat swamp forests, where almost all aboveground biomass can be lost after repeated fires11, a shift from tree cover to non-woody vegetation is likely to prevent surface fires transitioning to peat fires, which require greater fuel load and temperature11. However, we note that where fires are started deliberately, with the aim to burn vegetation, accelerants will often be used, which may alter the heat flux delivered to fuel components including those of the peat.
It has been noted that in the area of the former Mega Rice Project61 in Kalimantan, Indonesia, degraded peat swamp forest appears to show decreasing depth of burn with subsequent fires, and a range of possible reasons have been suggested6. Firstly, this may be due to progressive reduction of the aboveground fuel load. Secondly, the loss of peat, which can typically be to depths of 30 cm or more62,63, reduces the distance to the water table, thus effectively increasing peat moisture content. Thirdly, the post-fire peat surface is left more recalcitrant as a consequence of selective destruction of more labile forms of carbon such as lignins and polysaccharides, and accumulation of aromatic and aliphatic compounds64. The increase in Tmax with depth found in the present study suggests an additional negative feedback linked to the botanical composition of tropical peat. Our results for the global variation in Tmax indicate that tropical peat requires heating to a higher temperature via a greater flux of heat from a surface fire to ignite. Our results also indicate that fire resistance further increases with depth in tropical peats (Fig. 2). This suggests that if increases in fire severity, or shorter fire return intervals, were to cause the surface layers of peat to be lost at a greater rate than that of peat accumulation, then exposure of more ignition-resistant peat ought to decrease its subsequent vulnerability to fire. Whether this negative feedback influences the prevalence of peat fires will depend on the temperatures attained at ground level due to the heat flux from the burning of overstory vegetation. If these far exceed the ignition temperature of the peat, variation in that temperature should not be relevant to the probability of ignition. Yet where the heat flux delivered means that ground temperatures fall within the range of peat ignition temperatures, variation in the latter may determine ignition.
Ground temperatures attained due to surface fire will vary spatially and temporally, and ignition of the peat will depend on the temperature profile (i.e. temperature as a function of time) as the fire passes, with the required heat input being substantially affected by moisture content34. Since surface fire may or may not ignite the underlying peat21,39, ignition temperature is likely to be a determining factor, with an inverse relationship to ignition probability. This suggests that the loss of upper peat layers (with lower ignition temperatures) does have the potential to leave a peat surface with a lower likelihood of supporting the transition from surface fires to ground fires.
However, the presence of twigs and roots in peat has been reported to promote the propagation of smouldering fires65,66, and it has been suggested that larger wood pieces in tropical peatlands can serve as conduits for conducting surface fires into the ground, assisting flaming surface fires to transition to smouldering ground fires37. Moreover, woody pieces can result in gaps within the peat37 that allow oxygen ingress, enhancing smouldering. Therefore it is likely that wood particles of different size affect flammability in different ways. Small woody particles would tend to lead to overall high lignin contents of peat at the scale measured in this study, reducing ignitability, whilst larger particles may assist with carrying fire and oxygen down into the ground. Hence the vulnerability to smouldering ignition of tropical peat will be decreased by higher wood content, requiring a suitably intense ignition source that allows exposed peat to reach a temperature sufficient to drive pyrolysis and initiate the process of char oxidation (smouldering); but once ignited, larger wood particles may enhance the ability of the fire to propagate through the peat67.
Higher calorific values (19‒23 kJ g−1) have been recorded for peat from Indonesia when compared with boreal peat38. This should increase the energy release per equal peat volume, drying the peat and enhancing pyrolysis at the smouldering front. Moreover, non-Sphagnum based peat tends to have a higher bulk density68,69, which should produce more energy during smouldering due to the oxidation of a greater mass of peat particles70. However, moisture content strongly interacts with bulk density (where higher bulk density peat holds more water) to determine whether or not smouldering propagation can be maintained34,67. High-density peat, with generally higher moisture content, usually either fails to sustain smouldering or tends to carry slower fires34. However, the moisture content of degraded peat in Sumatra has been observed to be as low as ~20% dry mass in the uppermost 10 cm of the peat71, and where there is little water to act as a heat sink, high-density peat will instead provide a large energy source that will support self-sustained propagation of smouldering fires. However, measurements of peat from the former Mega Rice Project area of Kalimantan61 have shown that calorific content appears to decrease with depth72. This coupled to the increase in Tmax in the anaerobic layers of tropical peat ought to lower the risk of ignition and decrease the potential of self-sustained propagation with depth in cases where previous fires have removed the surface peat.
In summary, we conclude that where peat has been drained and degraded, consideration of the botanical composition of peat may have more importance in determining its flammability than has previously been considered. Our findings, which link the botanical constituents of peat and its resistance to ignition, indicate that the composition of tropical peat confers resistance to ignition, but also propensity for more intense burning when in a dry and degraded state. The higher wood content, calorific content and bulk density in tropical peat when compared with peat of other latitude groups will tend to decrease its ignitability but increase the energy release per equal peat volume once smouldering is initiated, allowing self-sustained propagation through the peat. However, we suggest that deeper tropical peat tends to have a lower vulnerability to fire since both the increase in Tmax and decrease in calorific content will have negative feedback on subsequent ignition and self-sustained propagation of smouldering within the peat.
This study of peat types from around the world provides a first attempt at assessing how latitudinal effects and their influence on botanical composition may influence flammability. Further research might seek to determine in more detail how variable ignitability and calorific content are across specific degraded peatland systems. If estimates can be made as to their intrinsic flammability this may serve as a predictor for future fire severity and/or aid in determining which areas might need more focused fire protection and ecosystem restoration than others.
Methods
Peat samples were obtained from 55 different sites: 11 arctic, 16 boreal, 16 temperate and 12 in tropical regions (Table S1; Fig. 1). At each site, samples were taken from between one and four coring locations, depending on existing research designs. Sample depths were determined by total peat depth at each location, with surface (aerobic layer) samples collected from depths of 0‒15 cm, and subsurface (anaerobic layer) samples from 19‒468 cm. 152 samples were obtained for calorimetric analysis, and 53 of these were additionally analysed for composition.
Calorimetric analysis
All samples were analysed by pyrolysis combustion flow calorimetry (PCFC)73, in which the oven-dried sample is pyrolysed in a nitrogen stream, the resulting pyrolysate gases fully combusted in excess oxygen, and the heat release measured by oxygen consumption calorimetry. The sample is thermally decomposed at a constant rate of temperature rise, and the heat release profile from combustion of the pyrolysate used to derive flammability metrics. Whereas flammability parameters obtained from conventional combustion experiments are affected by initial differences (or in-process changes) in the size and shape of the specimens, and by edge effects, PCFC measures intrinsic material properties that are independent of test conditions74, resulting in a high degree of reproducibility. Tmax is the temperature at which the maximum rate of solid mass loss, and thus of pyrolysate generation and heat release, is attained, and is approximately equal to the ignition temperature of the material75. We used an FAA Micro Calorimeter (Fire Testing Technology Ltd, East Grinstead, UK), which is designed for testing the flammability of construction and furnishing materials, but has recently been used to assess combustion properties of wildland fuels76,77,78,79. Each peat sample was analysed in duplicate (sometimes triplicate), using subsamples of 1.2‒29.9 mg. The pyrolyser heating rate was 3 °C s−1, the maximum pyrolysis temperature 750 °C, and the combustor temperature 900 °C. The N₂ flow rate was 80 cm³ min-1 and the O₂ flow rate 20 cm³ min-1. The experimental method follows ASTM D7309-07 Method A80. Tmax values for paired replicates, representing separate calorimetric tests of material from the same peat sample, were highly correlated (r = 0.97), indicating that the test produces replicable results.
Compositional analysis
Peat samples for composition analysis were sieved through a 125 µm sieve using a spray of deionised water. The material retained on the sieve was then analysed for peat components and macrofossils following a standard protocol81. Samples were placed in a petri dish and scanned using a low power (×10–×50) stereo-zoom microscope with a 10 × 10 square grid graticule inserted into one of the eyepieces. The petri dish was moved randomly to 15 different views, plant macrofossil types were estimated as percentages for each view using the graticule, and the results were averaged to represent the whole sample. High power (×200–×400) microscopy was used to confirm identifications. Subsamples of plant macrofossil material were mounted on microscope slides (temporary preparations using water as a mountant) and identified at ×100–×400 magnification. The samples were categorised into broad composition types (Sphagnum peat, sedge peat, Sphagnum and sedge peat, wood and root peat, and humified peat). To avoid bias, the composition analysis was undertaken without knowledge of the sample origins.
Climate (temperature) data82 were obtained via the AQUASTAT Climate Information Tool83. Temperatures obtained for each site are mean values for the period 1961‒1990, interpolated from climate station data at a spatial resolution of 10 min82.
Data availability
The data supporting the findings of this study are available at: https://doi.org/10.6084/m9.figshare.25858402.
References
Rein, G., Cleaver, N., Ashton, C., Pironi, P. & Torero, J. L. The severity of smouldering peat fires and damage to the forest soil. Catena 74, 304–309 (2008).
Hadden, R. M., Rein, G. & Belcher, C. M. Study of the competing chemical reactions in the initiation and spread of smouldering combustion in peat. P Combust. Inst. 34, 2547–2553 (2013).
Lin, S. R., Sun, P. Y. & Huang, X. Y. Can peat soil support a flaming wildfire? Int. J. Wildland Fire 28, 601–613 (2019).
van der Werf, G. R. et al. Global fire emissions and the contribution of deforestation, savanna, forest, agricultural, and peat fires (1997-2009). Atmos. Chem. Phys. 10, 11707–11735 (2010).
Turetsky, M. R. et al. Global vulnerability of peatlands to fire and carbon loss. Nat. Geosci. 8, 11–14 (2015).
Page, S. E. & Hooijer, A. In the line of fire: the peatlands of Southeast Asia. Philos. T. R. Soc. B 371, 20150176 (2016).
Huang, Y. Y. et al. Tradeoff of CO2 and CH4 emissions from global peatlands under water-table drawdown. Nat. Clim. Change 11, 618–622 (2021).
Gallego-Sala, A. V. et al. Latitudinal limits to the predicted increase of the peatland carbon sink with warming. Nat. Clim. Change 8, 907–913 (2018).
Evans, C. D. et al. Overriding water table control on managed peatland greenhouse gas emissions. Nature 593, 548–552 (2021).
Turetsky, M. R. et al. Recent acceleration of biomass burning and carbon losses in Alaskan forests and peatlands. Nat. Geosci. 4, 27–31 (2011).
Hoscilo, A., Page, S. E., Tansey, K. J. & Rieley, J. O. Effect of repeated fires on land-cover change on peatland in southern Central Kalimantan, Indonesia, from 1973 to 2005. Int. J. Wildland Fire 20, 578–588 (2011).
Miettinen, J., Shi, C. H. & Liew, S. C. Fire distribution in peninsular Malaysia, Sumatra and Borneo in 2015 with special emphasis on peatland fires. Environ. Manag. 60, 747–757 (2017).
Lasslop, G., Coppola, A. I., Voulgarakis, A., Yue, C. & Veraverbeke, S. Influence of fire on the carbon cycle and climate. Curr. Clim. Change Rep. 5, 112–123 (2019).
Nelson, K., Thompson, D., Hopkinson, C., Petrone, R. & Chasmer, L. Peatland-fire interactions: a review of wildland fire feedbacks and interactions in Canadian boreal peatlands. Sci. Total Environ. 769, 145212 (2021).
Hu, Y. Q., Fernandez-Anez, N., Smith, T. E. L. & Rein, G. Review of emissions from smouldering peat fires and their contribution to regional haze episodes. Int. J. Wildland Fire 27, 293–312 (2018).
Latif, M. T. et al. Impact of regional haze towards air quality in Malaysia: a review. Atmos. Environ. 177, 28–44 (2018).
Treat, C. C. et al. Widespread global peatland establishment and persistence over the last 130,000 y. Proc. Natl Acad. Sci. USA 116, 4822–4827 (2019).
Yu, Z. C., Loisel, J., Brosseau, D. P., Beilman, D. W. & Hunt, S. J. Global peatland dynamics since the Last Glacial Maximum. Geophys. Res. Lett. 37, L13402 (2010).
Loisel, J. et al. Expert assessment of future vulnerability of the global peatland carbon sink. Nat. Clim. Change 11, 70–77 (2021).
Wilkinson, S. L. et al. Wildfire and degradation accelerate northern peatland carbon release. Nat. Clim. Change 13, 456–461 (2023).
Goldstein, J. E. et al. Beyond slash-and-burn: The roles of human activities, altered hydrology and fuels in peat fires in Central Kalimantan, Indonesia. Singap. J. Trop. Geo. 41, 190–208 (2020).
Kuosmanen, N. et al. Repeated fires in forested peatlands in sporadic permafrost zone in Western Canada. Environ. Res. Lett. 18, 094051 (2023).
Kuhry, P. The role of fire in the development of Sphagnum-dominated peatlands in Western Boreal Canada. J. Ecol. 82, 899–910 (1994).
Magnan, G., Lavoie, M. & Payette, S. Impact of fire on long-term vegetation dynamics of ombrotrophic peatlands in northwestern Québec, Canada. Quat. Res. 77, 110–121 (2012).
Gałka, M. et al. Palaeoenvironmental changes in Central Europe (NE Poland) during the last 6200 years reconstructed from a high-resolution multi-proxy peat archive. Holocene 25, 421–434 (2015).
Miettinen, J., Shi, C. H. & Liew, S. C. Land cover distribution in the peatlands of Peninsular Malaysia, Sumatra and Borneo in 2015 with changes since 1990. Glob. Ecol. Conserv 6, 67–78 (2016).
Lukenbach, M. C. et al. Hydrological controls on deep burning in a northern forested peatland. Hydrol. Process 29, 4114–4124 (2015).
Kettridge, N. et al. Moderate drop in water table increases peatland vulnerability to post-fire regime shift. Sci. Rep. 5, 8063 (2015).
Page, S. E. et al. The amount of carbon released from peat and forest fires in Indonesia during 1997. Nature 420, 61–65 (2002).
Ballhorn, U., Siegert, F., Mason, M. & Limin, S. Derivation of burn scar depths and estimation of carbon emissions with LIDAR in Indonesian peatlands. P Natl Acad. Sci. USA 106, 21213–21218 (2009).
New, S. L., Belcher, C. M., Hudspith, V. A. & Gallego-Sala, A. V. Holocene fire history: can evidence of peat burning be found in the palaeo-archive? Mires Peat 18, 26 (2016).
Rein, G. in Fire Phenomena and the Earth System: An Interdisciplinary Guide to Fire Science (ed Belcher, C. M.) 15–33 (Wiley Blackwell, 2013).
Hayasaka, H., Takahashi, H., Limin, S. H., Yulianti, N. & Usup, A. in Tropical Peatland Ecosystems (eds Osaki, M. & Tsuji, N.) 377–395 (Springer Japan, 2016).
Prat-Guitart, N., Rein, G., Hadden, R. M., Belcher, C. M. & Yearsley, J. M. Propagation probability and spread rates of self-sustained smouldering fires under controlled moisture content and bulk density conditions. Int. J. Wildland Fire 25, 456–465 (2016).
Huang, X. Y. & Rein, G. Downward spread of smouldering peat fire: the role of moisture, density and oxygen supply. Int. J. Wildland Fire 26, 907–918 (2017).
Santoso, M. A., Christensen, E. G., Yang, J. & Rein, G. Review of the transition from smouldering to flaming combustion in wildfires. Front. Mech. Eng. 5, 49 (2019).
Cancellieri, D. et al. Kinetic investigation on the smouldering combustion of boreal peat. Fuel 93, 479–485 (2012).
Usup, A. Combustion and thermal characteristics of peat fire in tropical peatland in Central Kalimantan, Indonesia. Tropics 14, 1–19 (2004).
Ryan et al. Ignitions for peat fires in Indonesia: a critical look. Proc. Fire Continuum Conference, (U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, 2020).
UNEP. Global Peatlands Assessment—The State of the World’s Peatlands: Evidence for action toward the conservation, restoration, and sustainable management of peatlands. (Global Peatlands Initiative/United Nations Environment Programme, Nairobi, 2022).
Swindles, G. T., Baird, A. J., Kilbride, E., Low, R. & Lopez, O. Testing the relationship between testate amoeba community composition and environmental variables in a coastal tropical peatland. Ecol. Indic. 91, 636–644 (2018).
Pacheco-Cancino, P. A., Carrillo-López, R. F., Sepulveda-Jauregui, A. & Somos-Valenzuela, M. A. Sphagnum mosses, the impact of disturbances and anthropogenic management actions on their ecological role in CO2 fluxes generated in peatland ecosystems. Glob. Change Biol.30, e16972 (2024).
Swindles, G. T., Blundell, A., Roe, H. M. & Hall, V. A. A 4500-year proxy climate record from peatlands in the North of Ireland: the identification of widespread summer’drought phases’? Quat. Sci. Rev. 29, 1577–1589 (2010).
Dergacheva, M., Fedeneva, I., Bazhina, N., Nekrasova, O. & Zenin, V. Shestakovo site of Western Siberia (Russia): pedogenic features, humic substances and paleoenvironment reconstructions for last 20‒25 ka. Quat. Int. 420, 199–207 (2016).
Gałka, M., Swindles, G. T., Szal, M., Fulweber, R. & Feurdean, A. Response of plant communities to climate change during the late Holocene: palaeoecological insights from peatlands in the Alaskan Arctic. Ecol. Indic. 85, 525–536 (2018).
Talbot, J. M., Yelle, D. J., Nowick, J. & Treseder, K. K. Litter decay rates are determined by lignin chemistry. Biogeochemistry 108, 279–295 (2012).
Cesarino, I., Araujo, P., Domingues, A. P. & Mazzafera, P. An overview of lignin metabolism and its effect on biomass recalcitrance. Braz. J. Bot. 35, 303–311 (2012).
Page, W. G., Jenkins, M. J. & Runyon, J. B. Mountain pine beetle attack alters the chemistry and flammability of lodgepole pine foliage. Can. J. For. Res 42, 1631–1647 (2012).
Alam, M. A. et al. Shoot flammability is decoupled from leaf flammability, but controlled by leaf functional traits. J. Ecol. 108, 641–653 (2020).
Hodgkins, S. B. et al. Tropical peatland carbon storage linked to global latitudinal trends in peat recalcitrance. Nat. Commun. 9, 3640 (2018).
Bacon, K. L. et al. Questioning ten common assumptions about peatlands. Mires Peat 19, 12 (2017).
Lawson, I. T., Jones, T. D., Kelly, T. J., Coronado, E. N. H. & Roucoux, K. H. The geochemistry of Amazonian peats. Wetlands 34, 905–915 (2014).
Ligrone, R., Duckett, J. G. & Renzaglia, K. S. Conducting tissues and phyletic relationships of bryophytes. Philos. T R. Soc. B 355, 795–813 (2000).
Mitchard, E. T. A. The tropical forest carbon cycle and climate change. Nature 559, 527–534 (2018).
Page, S. et al in Tropical Fire Ecology (ed Cochrane, M. A.) 263–287 (Springer Praxis Books, 2009).
Graham, L. L. B., Giesen, W. & Page, S. E. A common-sense approach to tropical peat swamp forest restoration in Southeast Asia. Restor. Ecol. 25, 312–321 (2017).
Evans, C. D. et al. Rates and spatial variability of peat subsidence in Acacia plantation and forest landscapes in Sumatra, Indonesia. Geoderma 338, 410–421 (2019).
Hooijer, A. et al. Subsidence and carbon loss in drained tropical peatlands. Biogeosciences 9, 1053–1071 (2012).
Page, S. et al. Restoration ecology of lowland tropical peatlands in southeast Asia: current knowledge and future research directions. Ecosystems 12, 888–905 (2009).
Balch, J. K. et al. Negative fire feedback in a transitional forest of southeastern Amazonia. Glob. Change Biol. 14, 2276–2287 (2008).
Medrilzam, M., Smith, C., Aziz, A. A., Herbohn, J. & Dargusch, P. Smallholder farmers and the dynamics of degradation of peatland ecosystems in Central Kalimantan, Indonesia. Ecol. Econ. 136, 101–113 (2017).
Kusin, K., Jagau, Y., Ricardo, J., Saman, T. N. & Aguswan, Y. Peat lost by fire in Kalampangan area, Central Kalimantan, Indonesia. Proc. IOP Conference Series: Earth and Environmental Science 504, 012009 (2020).
Wilkinson, S. L., Tekatch, A. M., Markle, C. E., Moore, P. A. & Waddington, J. M. Shallow peat is most vulnerable to high peat burn severity during wildfire. Environ. Res. Lett. 15, 104032 (2020).
Milner, L. E. Influence of fire on peat organic matter from Indonesian tropical peatlands. PhD thesis, University of Leicester, (2013).
Miyanishi, K. & Johnson, E. A. Process and patterns of duff consumption in the mixedwood boreal forest. Can. J. For. Res. 32, 1285–1295 (2002).
Davies, G. M., Gray, A., Rein, G. & Legg, C. J. Peat consumption and carbon loss due to smouldering wildfire in a temperate peatland. For. Ecol. Manag. 308, 169–177 (2013).
Wuquan C., Yuqi H. & Rein, G. Experimental study of the ignition conditions for self-sustained smouldering in peat. Proc. Combust. Inst. 39, 4125–4133 (2023).
Loisel, J. et al. A database and synthesis of northern peatland soil properties and Holocene carbon and nitrogen accumulation. Holocene 24, 1028–1042 (2014).
Liu, H. & Lennartz, B. Hydraulic properties of peat soils along a bulk density gradient—a meta study. Hydrol. Process 33, 101‒114 (2019).
Ohlemiller, T. J. Modeling of smoldering combustion propagation. Prog. Energy Combust. Sci. 11, 277–310 (1985).
Santoso, M. A. et al. GAMBUT field experiment of peatland wildfires in Sumatra: from ignition to spread and suppression. Int. J. Wildland Fire 31, 949–966 (2022).
Yulianti, et al. Situation of peat fire combustion on Southern Kalimantan, Indonesia. in Wetland Systems: Ecology, Functioning and Management. (eds Borin, M. Malagoli, M., Salvato, M., Tanis, B.) 163–164. (2013).
Lyon, R. E. & Walters, R. N. Pyrolysis combustion flow calorimetry. J. Anal. Appl. Pyrol. 71, 27–46 (2004).
Lyon, R. E., Walters, R. N., Stoliarov, S. I., Safronava, N. Principles and practice of microscale combustion calorimetry. Report No. DOT/FAA/TC-12/53, R1, 95. Department of Transportation, Federal Aviation Administration (ed William, J.) (Hughes Technical Center, Atlantic City International Airport, NJ 08405, 2014).
Lyon, R. E., Walters, R. N. & Stoliarov, S. I. Thermal analysis of flammability. J. Therm. Anal. Calorim. 89, 441–448 (2007).
Haworth, M. et al. Impaired photosynthesis and increased leaf construction costs may induce floral stress during episodes of global warming over macroevolutionary timescales. Sci. Rep. 8, 6206 (2018).
Dewhirst, R. A., Smirnoff, N. & Belcher, C. M. Pine species that support crown fire regimes have lower leaf-level terpene contents than those native to surface fire regimes. Fire 3, 17 (2020).
Solofondranohatra, C. L. et al. Shade alters the growth and architecture of tropical grasses by reducing root biomass. Biotropica 53, 1052–1062 (2021).
Crawford, A. J., Feldpausch, T. R., Marimon, B. H. Jr, de Oliveira, E. A. & Belcher, C. M. Effect of tree wood density on energy release and charcoal reflectance under constant heat exposure. Int. J. Wildland Fire 32, 1788–1797 (2023).
ASTM International. ASTM D7309-07: Standard test method for determining flammability characteristics of plastics and other solid materials using microscale combustion calorimetry (2007).
Mauquoy, D., Hughes, P. D. M. & van Geel, B. A protocol for plant macrofossil analysis of peat deposits. Mires Peat 7, 6 (2010).
New, M., Lister, D., Hulme, M. & Makin, I. A high-resolution data set of surface climate over global land areas. Clim. Res. 21, 1–25 (2002).
Food and Agriculture Organization of the United Nations. AQUASTAT Climate Information Tool https://aquastat.fao.org/climate-information-tool/ (2021).
Acknowledgements
We thank Matthew J. Amesbury, Michelle Garneau, Mark J. Grosvenor, Victoria A. Hudspith, Markku Mäkilä, Lisa Orme and Nicole K. Sanderson for providing peat samples. C.M.B., S.P. and A.G-S. acknowledge funding from the KaLi Project, funded by the UKRI Global Challenges Research Fund, Grant No. NE/T010401/1. A.G-S. has also received funding from NERC (NE/1012915 and NE/S001166/1) and from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 865403). This work reflects only the author’s view and the European Commission/Agency is not responsible for any use that may be made of the information it contains. G.T.S. has received funding from the Dutch Foundation for the Conservation of Irish Bogs, The Quaternary Research Association and Leverhulme Trust RPG-2021-354. T.A.B. acknowledges funding from Russian Science Foundation Grant N 23-27-00217.
Author information
Authors and Affiliations
Author notes
Unaffiliated: Outi Lähteenoja.
- Outi Lähteenoja
Contributions
C.M.B. and A.G.-S. designed the study. A.G-S., G.T.S., T.A.B., H.C-Q., D.J.C., M.G., P.D.M.H., O.L., D.M., S.P., T.P.R. and M.V. obtained peat samples. C.M.B. and S.N. conducted calorimetry experiments. G.T.S. analysed peat composition. A.J.C. analysed data with input from C.M.B. and A.G.-S. A.J.C. and C.M.B. wrote the text with input from all authors.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Crawford, A.J., Belcher, C.M., New, S. et al. Tropical peat composition may provide a negative feedback on fire occurrence and severity. Nat Commun 15, 7363 (2024). https://doi.org/10.1038/s41467-024-50916-7
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-024-50916-7
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.