Fungal decay of heart wood creates hollows and areas of reduced wood density
within the stems of living trees known as stem rot. Although stem rot is
acknowledged as a source of error in forest aboveground biomass (AGB)
estimates, there are few data sets available to evaluate the controls over
stem rot infection and severity in tropical forests. Using legacy and recent
data from 3180 drilled, felled, and cored stems in mixed dipterocarp forests
in Sarawak, Malaysian Borneo, we quantified the frequency and severity of
stem rot in a total of 339 tree species, and related variation in stem rot
with tree size, wood density, taxonomy, and species' soil association, as
well as edaphic conditions. Predicted stem rot frequency for a 50 cm tree
was 53 % of felled, 39 % of drilled, and 28 % of cored stems,
demonstrating differences among methods in rot detection ability. The percent
stem volume infected by rot, or stem rot severity, ranged widely among trees
with stem rot infection (0.1–82.8 %) and averaged 9 % across all
trees felled. Tree taxonomy explained the greatest proportion of variance in
both stem rot frequency and severity among the predictors evaluated in our
models. Stem rot frequency, but not severity, increased sharply with tree
diameter, ranging from 13 % in trees 10–30 cm DBH to 54 % in stems
Fungal rot of secondary xylem causes hollows and regions of reduced wood density in tree stems. This type of fungal infection, commonly referred to as stem rot, is important for the structure, dynamics, and functioning of forests, given that it may increase tree mortality (Franklin et al., 1987; Ruxton, 2014), facilitate the creation of cavity habitats for a diversity of wood-inhabiting and decaying species (Cockle et al., 2012; Stockland et al., 2012), and may act as a reservoir of nutrients sequestered in stem rot biomass (Janzen, 1976; Dickinson and Tanner, 1978; Boddy and Watkinson, 1995). Moreover, the effect of stem rot on aboveground biomass is of particular importance for efforts to map carbon storage in tropical regions as part of global conservation and climate change mitigation strategies (Saatchi et al., 2011). However, because stem rot is difficult to detect by non-destructive means, we understand little about what controls its frequency and severity, especially in tropical forests.
Most information available on stem rot in tropical forests comes from
forestry studies exploring its influence on the volume and quality of timber.
Among species commonly logged in Old World dipterocarp forests, stem rot
occurs in up to 75 % of large
Understanding how patterns of stem rot infection vary with tree size is
critical to assessing its influence on forest AGB estimation because trees
Quantifying the strength of taxonomic variation in stem rot frequency in trees and determining functional traits that underlie species differences in stem rot infection should provide insight into patterns of carbon storage in tropical forests, which often show large shifts in species composition over spatial gradients. Cornwell et al. (2009) demonstrated that the inclusion of plant traits improved models of global carbon turnover from coarse woody debris pools, however, the traits that influence wood decomposition in living stems are still poorly understood. Among potentially important species traits, wood properties may explain variation among trees in susceptibility to stem rot. Trees with dense wood may be less likely to experience branch and stem breakage due to wind disturbance (Putz et al., 1983) and are thought to be more resistant to termite and fungal infection than trees with softer wood. However, there is conflicting evidence as to whether wood density and decay resistance are strongly correlated (Yoneda, 1975; Bultman and Southwell, 1976; Weedon et al., 2009; van Geffen et al., 2010; Mori et al., 2013). Dense wood has been found to be associated with pathogen resistance in tropical tree species (Augspurger and Kelly, 1984) and slower fungal growth (Romero and Bolker, 2008), perhaps because denser heart wood may retain more water and thereby inhibit fungal growth (Boddy, 2001) or have low porosity that impedes the growth of fungal hyphae (Bjurman and Viitanen, 1996; Schwarze et al., 2000). Within species, fast-grown trees with lower wood density are associated with faster decay rates by saproxylic fungi (Edman et al., 2006; Ranius et al., 2009), although this pattern does not hold across all fungal taxa (Yu et al., 2003). In an Amazonian forest, little or no covariation was found between species' wood density and frequency of stem rot, but the probability of rot was significantly correlated with other wood traits, such as the lumen diameter and density of vessels (Eleuterio, 2011).
The plant traits and environmental constraints associated with variation in the frequency and severity of stem rot may change with the availability of edaphic resources. For instance, low soil fertility and nutrient or water stress may predispose tissues to fungal infection (Boddy and Rayner, 1983; Franklin et al., 1987). There is also a potential indirect effect of age (Ranius et al., 2009), as trees tend to grow slower and live longer on more infertile soils (Russo et al., 2005), and are therefore exposed to chance infection for longer periods. Conversely, trees on more fertile soils not only grow faster, they have less dense and softer wood (Heineman and Russo, unpublished data) and perhaps lower contents of defensive secondary metabolites (Loehle, 1988; Fine et al., 2006). Such wood may be more prone to stem rot (Wagener and Davidson, 1954; Duchesne et al., 1992; Pearce, 1996; Kirker et al., 2013). If stem rot varies along edaphic gradients, then including soil parameters in models estimating forest carbon dynamics (e.g. Yang et al., 2013, 2014) would be important, especially in southeast Asian forests where AGB varies with soil nutrient availability (Lee et al., 2002; Paoli et al., 2008a).
We used legacy and modern data sets to quantify the covariation of taxonomy, tree size, wood density, and soil resource availability with the frequency and severity of stem rot in two Bornean mixed dipterocarp forests. We quantified the impact of stem rot on forest standing biomass, and evaluated the implications of soil-related variation in stem rot for stand-level variation in biomass. Efforts to quantify stem rot in tropical trees are hampered by the difficulty of evaluating rot without compromising the health of trees in long term monitoring plots. We therefore also compared methods for quantifying stem rot frequency, including the direct assessment of stem rot frequency based on destructive harvesting, with two non-destructive methods (coring and drilling).
The data were collected in two locations in Malaysian Borneo: Central Sarawak
and Lambir Hills National Park, Sarawak (Fig. 1). The Central Sarawak tree
drilling and felling data were collected during a timber inventory of lowland
mixed dipterocarp rain forest (1
Location of study sites in Sarawak, Malaysian Borneo. Shaded areas are the Central Sarawak inventory units for stem rot felling and drilling data, with clusters of plots indicated by black dots. The black rectangle (not to scale) indicates the location of the Lambir Hills National Park study site for the stem rot coring data.
Total number of trees and species (including morphospecies) evaluated for the presence of stem rot, and the ranges and sample sizes for predictor variables used in mixed-effect models evaluating the association of ecological covariates with frequency and severity of stem rot in the felling, drilling, and coring data sets. See Sect. 2 for details.
Tree coring data were collected in 2009 in Lambir Hills National Park
(hereafter, Lambir) in northern Sarawak (4
The Central Sarawak felling (1035 trees in 211 species in 31 families) and drilling (1780 trees in 206 species in 34
families) data (Tables 1, S1 in Supplement) were collected from 422 plots
grouped in 69 clusters as part of the Forest Industries Development Project
(FIDP) conducted from 1969–1971 (Fig. 1) (FIDP, 1974b). Each cluster
consisted of nine plots arrayed in a 3
Safety was a primary consideration, and felling crews had on-site autonomy to exclude trees with visible rot, asymmetric crowns or other features indicative of increased risk of the stem splitting during felling. Of trees with desired felling dimensions, 22 % were not felled due to safety concerns. Half of trees excluded for felling were taxa with extremely hard wood, and the remainder were excluded because of excessive slope, severely asymmetric crown, prominent stem or crown damage, or active bee and hornet nests. The exclusion of very large stems with obvious damage and of species with very hard wood means that the felling data are likely conservative in their representation of rot in the forest as a whole. Safety was less of a concern for drilling.
The stem rot data from Lambir were collected from 365 trees (116 species in
35 families; Tables 1, S1) with a 5 mm increment hand borer (Haglöf
Sweden AB, Sweden), bored to half of the DBH at one point on the stem, in
contrast to the drilling method which tested for rot at two points on each
stem. Trees
DBH was recorded for all individuals in the drilling, felling, and coring
data sets. Wood density (oven dry mass/fresh volume; g cm
Each tree species in the three data sets was categorized according to its soil association. Generalists were species that are similarly abundant on all soil types. Species with distinct soil associations were categorized as specialists of clay/fine loam, fine loam/loam, loam/sandy loam, in order of decreasing fertility and water retention. Assignments were based on analyses of species' distributions within the 52 ha forest dynamics plot at Lambir (Davies et al., 2005) and across a network of plots in Sarawak (Potts et al., 2002). For species not in these studies, classifications were assigned by P.S. Ashton from his extensive studies of tree species distributions in the region (Ashton, 1964, 1973, 2015). The density of sound wood was assigned to stems in the felling and drilling data sets from timber group values (FIDP, 1974b) and from species average densities in the coring data.
Edaphic data (Table S2) were collected for each plot in the Central Sarawak
data. Soil morphology was described in shallow profile pits at plot centers
(Baillie et al., 1987). The profiles and augerings, located 2.5 m from the
center, were sampled at 0–10 and 45–55 cm, bulked by depth (topsoil and
subsoil, respectively), and the soils analyzed for pH electrometrically,
organic carbon by Walkley Black acid dichromate oxidation, and total nitrogen
by micro-Kjeldahl distillation. Exchangeable cations were extracted with 1M
NH
For the central Sarawak data, we used principal component analysis (PCA) to
create a reduced number of orthogonal axes of soil variation, using the
function
Linear mixed effect models were used to evaluate (1) if the detection of
stem rot presence and/or absence differed among data sets, (2) if variance in stem
rot frequency and severity is explained by taxonomic levels, and (3) if stem
rot frequency and severity varied with species, tree, and edaphic covariates.
In models testing variation in the frequency of stem rot, we used generalized
linear mixed models (GLMM) with a binomial probability distribution and logit
link function using the Laplace method and Cholesky root algorithm for
parameter estimation (Bolker et al., 2009). Linear mixed models (LMM) with
restricted maximum likelihood parameter estimation were used to evaluate
variation in stem rot severity (percent stem volume lost to stem rot), which
was log transformed to meet the assumption of normality of residuals. GLMM
and LMM models were fit using the package
Stem rot detection rate of drilling and felling could be compared directly in
the validation data set of stems sampled by both methods. To compare the
detection rate of coring relative to drilling and felling, we used a GLMM to
test the interaction between DBH and data set as fixed effects on the
aggregated stem rot frequency data. We included only individuals
To estimate the variance in stem rot frequency and severity explained due to
taxonomy, we fit LMM with DBH as a fixed effect and a nested taxonomic random
effect (Family/Genus/Species) separately for the drilling, felling, and
coring data sets. We used the normal approximation for a binomial error
distribution for models of stem rot frequency in order to be able to
calculate the variance partitioning coefficient (VPC) for each random effect
and the residual variance (Goldstein et al., 2002). VPC for a model with one
random effect is calculated as
For each data set, we fit two models testing the effects of ecological
covariates on stem rot frequency and severity. Model 1: main effects: DBH,
wood density, soil PC1, soil PC2, soil PC3, soil PC4; 2-way interaction: wood
density
We examined the variance in stem rot probability and severity explained by
random and fixed effects using pseudo-R squared (
Stem rot influences carbon storage and flux in forests, and the ability to
predict variation in stem rot due to environmental factors, such as soil
resource availability, is important for improving global carbon models. To
evaluate how variation in stem rot frequency and severity due to soil
properties influenced stand-level carbon stocks, we estimated the maximum
percent of stem biomass lost to stem rot at the stand level
(Loss
There was substantial variation between data sets in the frequency of stem rot (Fig. 2), which occurred in 9 % of cored, 41 % of drilled, and 53 % of felled stems. The classification error of the drilling method was 18 % for the 419 trees scored for rot by first drilling and then felling, where felling observations were taken to be correct. Of the stems misclassified by the drilling method, 58 of the 77 errors (75 %) were false negatives, in which drilled stems were scored as having no rot, but rot was later observed when the stem was felled. The average percent rot was substantially less severe in rotted stems misclassified by the drilling method (5 %) than in rotted stems trees correctly categorized by drilling (19 %), indicating that the drilling method was effective for scoring stems with extensive rot.
Stem diameter-related variation in the frequency of stem rot found
in trees in mixed dipterocarp rain forest of Central Sarawak, Bornean
Malaysia. Pie charts show the percentage of trees with stem rot (shaded),
with pies sized according to the total number of sampled stems (
The discrepancy in the prevalence of stem rot between the central Sarawak and
Lambir data sets may be caused in part by differences in the tree sizes
sampled: at Lambir 78 % of the cored trees were
Variance partitioning analysis using nested taxonomic random effects revealed
that family identity explained a negligible proportion of variance in stem
rot frequency, whereas the genus-level effect was retained in all three
data sets (Table S6). Among the three most well-sampled genera (all
Dipterocarpaceae) in the felling data set, stem rot occurred significantly
more frequently in
Variation in the frequency and severity of stem rot among three
important dipterocarp genera (representing a subset of the 65 genera from the
felling data) in the rain forest of Central Sarawak, Borneo, Malaysia. The
number of trees (
In the GLMMs fit for each data set, the following predictor variables were not retained in the final models either as main effects or interactions, and so had little effect on variation in stem rot frequency (Table S7): felling Model 1 – species wood density and soil PC1, PC2, and PC4; Felling Model 2 – species wood density; Drilling Model 1 – species wood density, soil PC1 and PC4; Drilling Model 2 – species wood density, soil association; Coring Model 1 – tree wood density, soil type; Coring Model 2 – tree wood density, soil association.
Stem rot significantly increased in probability with tree diameter in all
data sets (Tables 2, S5; Figs. 2, S5). Indeed, DBH was the only significant
predictor in the coring data set. Across all data sets, the frequency of rot
increased drastically with larger DBH: stem rot was present in 13 % of
stems 10–30 cm DBH, 37 % of stems 30–50 cm DBH, and 54 % of stems
Changes in model fit based on Akaike's Information Criterion (AIC)
with removal of single factors from the most-supported generalized linear
mixed-model testing associations of the probability of stem rot in trees of
mixed dipterocarp rain forest in Sarawak, Borneo, in the drilling, felling,
and coring data sets. The
Stem rot frequency varied significantly with edaphic variables, represented by soil PCs. In the drilling and felling data, the probability of stem rot increased significantly with soil PC3 (Tables 2, S5; Fig. S6a, c), which had a strong negative association with reserve and exchangeable Ca in the topsoil (Table S2). In the drilling data set, the probability of stem rot also decreased with increasing values of soil PC2, which were associated with high pH and high exchangeable Mg in both topsoil and subsoil (Tables 2, S5; Fig. S6b). Overall, these results suggest that stem rot in Central Sarawak was found more frequently on lower fertility soils with reduced cation availability.
The effect of DBH on the incidence of stem rot varied significantly among species with different soil associations, for the felling, but not the drilling or coring data (Tables 2, S5). For species associated with high-fertility clay and fine loam soil, the probability of rot did not increase with DBH, whereas the probability of rot significantly increased with DBH in all other soil association groups (Fig. S7). Thus, for large diameter trees, stem rot frequency was lower in stems of species associated with more fertile, finer textured soil, than for species in the other three soil association groups.
Overall, the variation in the probability of stem rot explained by the fixed
effects in our models differed among the data sets, ranging from the lowest
marginal
For the 53 % of felled trees showing stem rot, the percent of stem volume lost (stem rot severity) averaged 16 %, but ranged widely, from 0.1–82 %. Similar to stem rot frequency, stem rot severity in felled stems did not differ among tree families, and genus had a higher VPC (0.10) than did species (0.04) (Table S6). Stem rot severity did not vary significantly with DBH, nor any soil PC, as none of these predictors was retained in the final model (Tables 3, S7). In Model 2, there was a significant interaction between wood density and soil habitat association (Table 3), whereby stem rot severity declined with increasing wood density in species associated with low-fertility soils but increased with wood density for generalist species (Table S5).
Changes in model fit based on Akaike's Information Criterion (AIC)
with removal of single factors from the most-supported linear mixed-model
testing associations of the stem rot severity (percent stem volume lost to
rot) in trees of mixed dipterocarp rain forest, Central Sarawak, Borneo, in
the felling data set. The
Alone, the fixed effects explained a very small proportion of variance in
stem rot severity (range of marginal
Our estimate of the maximum percent of stand-level stem biomass lost to rot,
Loss
A significant proportion of this variation was explained by soil variables.
Loss
The relationship between the stand-level percent stem biomass lost
to stem rot (Loss
Understanding the prevalence and ecology of stem rot in tropical forests is limited by the scarcity of data on stem rot and its potential explanatory correlates. Our analysis of extensive legacy and recently collected data from a total of 339 tree species of mixed-dipterocarp forests is the first to highlight the importance of edaphic properties, along with tree size, taxonomy, and functional traits, in explaining the frequency and severity of stem rot among trees. Together, these factors generated spatial variation among forest stands in the maximum percent of AGB lost to stem rot, and this variation was correlated with soil properties. Our finding that 7 % of forest AGB is in some stage of wood decay in these Bornean forests not only justifies greater consideration of stem rot and its soil-related variation in the estimation of carbon storage by tropical forests, but also underscores the need for standardized methods of stem rot detection to be applied across tropical forest regions.
The relationship between the stand-level percent stem biomass lost
to stem rot (Loss
Our results affirm the few existing previous findings that stem rot infection
is frequent and often severe in dipterocarp forests (Bakshi, 1960; Bagchee,
1961). Stem rot occurred in 41 % of drilled, 53 % of felled stems,
and encompassed 9 % of stem volume on average in our study of central
Sarawak. These estimates exceed observations from neotropical forests, where
30–38 % of trees among Amazonian species were found to have hollow stems
(Apolinário and Martius, 2004; Eleuterio, 2011), and volume losses to
stem rot ranged 0.7–4 % (Brown et al., 1995; Clark and Clark, 2000;
Nogueira et al., 2006). There may be several reasons for these discrepancies,
many related to methodological differences. First, most studies from
neotropical forests measured the hollow fraction of stems, whereas our study
quantified the volume of wood in any stage of decay visible in the field.
Therefore, our methods would inherently generate larger estimates of volume
and biomass loss. Second, approaches that quantify stem rot in decomposing
stems, such as stumps cut along access roads (Brown et al., 1995) or in
naturally fallen logs in coarse woody debris censuses (Clark and Clark,
2000), may confound rotting that occurs
While logging concessions may be opportunistically exploited for detailed evaluation of stem rot, accurate non-destructive measures are still needed to estimate stem rot where destructive harvest is impossible. Drilling proved to be an accurate means of scoring trees for stem rot, correctly characterizing stem rot presence/absence in 80 % of trees. The majority of misclassifications likely occurred because drilling tested for areas of wood decay only at breast height, missing rot occurring higher or lower in the stem. However, classification error diminished to 8 % for stems that had lost more than 10 % of their volume to rot, indicating that drilling at breast height is a reliable means of identifying trees containing large sections of rot. Moreover, the drilling data sampled larger trees than the felling data. Hence, results of felling and drilling analyses require subtly different interpretations, with the felling data set exploring the correlates of stem rot infection overall, and the drilling data set exploring the correlates of severe stem rot infection. There was no way to validate the accuracy of the coring method; however, the probability of detecting rot was lower at a given stem diameter in the coring data set relative to the drilling data set. Assuming that the true dependence of stem rot frequency on DBH was the same in Lambir and the Central Sarawak sites, then we suspect the drilling method may have been more effective at detecting rot because it was conducted in two perpendicular directions at breast height, whereas trees in the coring data set were bored only once. Inclusion of all three data sets not only allowed assessment of non-destructive methodologies for estimation of stem rot, but also improved inferences about correlates of stem rot by increasing size range to span the smaller and larger trees included in the coring and drilling data sets, respectively. Sonic tomography has been applied non-destructively to evaluate both the frequency and severity of hollows in tree stems (Nicolotti et al., 2003; Deflorio et al., 2008; Wunder et al., 2013). In the absence of access to this costly equipment, drilling may be a viable, less expensive alternative for assessing the presence of stem rot in remote tropical forests.
Identifying the environmental and tree-related correlates of stem rot provides insight into potential mechanisms governing stem rot and its implications for forest ecosystem processes. Tree size was the only factor significantly associated with stem rot frequency in all three data sets, consistent with previous observations of substantial increases in stem rot frequency with diameter in tropical forests (Nogueira et al., 2006; Eleuterio, 2011). To the extent that diameter correlates with tree age, these results suggest that trees experience accumulating risks of becoming infected with stem rot with time. Future studies should ensure that large trees are not under-sampled, especially considering that large, dominant canopy trees predominantly structure carbon dynamics in tropical forests (Slik et al., 2013; Bastin et al., 2015). Given that tree size was not a significant predictor of the percent stem volume lost to rot in infected trees, the advancement of stem rot infection may depend less on tree age than on the identity of the fungi involved and the physical and chemical properties of the heart wood (Wagener and Davidson, 1954; Pearce, 1996; Schilling et al., 2015).
Species wood density was not a significant predictor of stem rot frequency, consistent with findings from an Amazonian forest (Eleuterio, 2011). This result is somewhat surprising, as high wood density has been associated with pathogen protection in tropical tree species (Augspurger and Kelly, 1984). However, any pathogen protection conferred by higher wood density may not result in lower incidence of stem rot, because of the inverse relationship between wood density and mortality rates (King et al., 2006): at the same size, trees with high wood density are expected to be older than trees with low wood density and so may have longer exposure time to incur stem rot infection.
Stem rot frequency was higher in trees on lower fertility soils, and to a lesser extent, soils with lower pH, in Central Sarawak, but did not differ between edaphic habitats in the smaller trees cored at Lambir. Our results cannot identify how soil properties affect stem rot, but soil PC axes included in final models were correlated with soil Ca and Mg which have also been found to associate strongly with Bornean tree species distributions (Baillie et al., 1987) and explained significant variation in fine root growth at Lambir (Kochsiek et al., 2013). In the Lambir coring data set, the smaller sample size of larger trees may have prevented us from detecting differences in stem rot frequency among soil types. The association of stem rot with lower fertility soils in Central Sarawak is in some ways counter-intuitive, as forests on high-fertility sites generally have more frequent canopy disturbance (Coomes and Grubb, 2000), potentially causing more wounds and opportunities for infection. In addition to the longer exposure times presumably experienced by tree species characteristic of more dystrophic and drought-prone soils (Russo et al., 2005), trees under nutrient stress may be more prone to infection, if resources to produce secondary compounds are limited (Bryant et al., 1983). Soil chemistry may also influence the composition of the soil microbial community, from which wood decay fungi often originate (Boddy, 2001). At Lambir, the community composition of soil bacteria (Russo et al., 2012) and root-associated mycorrhizal fungi (Peay et al., 2009) differ between clay and sandy loam soil types, which vary in many properties including texture, nutrient supply, and pH (Table S3), indicating that the taxa and functional groups present to colonize wood may differ among edaphic habitats. Furthermore, the relative availability of nutrients in soil vs. wood may influence the infection and growth of wood decay of nutrient-limited fungi in trees, however, this interaction is likely to be complex (Merrill and Cowling, 1966; Donnelly and Boddy, 1998).
Variation among soil habitats in stem rot may also be driven by the change in species composition of the tree community combined with differences among taxa in susceptibility to stem rot. Yet, the effect of tree species' soil habitat association on stem rot frequency and severity was less consistent among data sets than the effects of soil properties. In the felling data set, stem rot frequency increased more slowly with DBH in species adapted to high-fertility soil than in species from other association groups: at 80 cm in diameter, tree species associated with sandy loam/loam soil were 8.6 times more likely to have stem rot than species associated with clay/fine loam and 4.1 times more likely to have stem rot than species associated with loam/fine loam. These results are consistent with the notion that trees specializing on more fertile soils may be younger at the same size compared to species adapted to resource-depleted soils, as their growth and mortality rates are higher (Russo et al., 2005). This pattern may not have been detected by drilling, which was less effective than felling at detecting minor stem rot infections. The significant interaction between soil association and wood density explaining the severity of stem rot remained difficult to parse, aside from the indication that the relationship of stem rot with species traits could be highly multi-dimensional and driven by a complex combination of interacting factors.
While tree size and edaphic factors were significantly associated with stem rot infection, these factors alone explained a relatively small fraction of the variance in the frequency of stem rot, and even less for stem rot severity. The explanatory power of models improved dramatically when fixed effects were conditioned on the species random effect, meaning that the occurrence of stem rot had a strong taxonomic component due to species properties other than wood density and soil association. Perhaps surprisingly, stem rot frequency varied among genera but not among families, suggesting that dipterocarp taxa do not differ systematically in susceptibility to rot from non-dipterocarp taxa in MDF forests, despite showing broad differences in other traits such as mycorrhizal association (Wang and Qiu, 2006).
Among species traits not evaluated in this study, wood anatomical properties and wood chemical content with respect to nutrient stoichiometry and secondary defense may be particularly important in understanding taxonomic variation in stem rot. Lumen diameter and vessel density are significantly correlated stem rot frequency in Amazonian tree species (Eleuterio, 2011) and have been hypothesized to influence the growth of fungal hyphae (Schwarze et al., 2000). Chemical properties of wood may also influence rates of fungal colonization in stems, as tree species wood N and P concentrations correlate with wood decomposition rates in angiosperms (Weedon et al., 2009), and wood decomposition rates decline with initial wood pH among Costa Rican tree species (Schilling et al., 2015). Furthermore, secondary defensive chemistry has been shown to vary among species and generate differences in the microbial colonization of woody debris (Cornwell et al., 2009). Most dipterocarps are known for copious resin production (Appanah and Turnbull, 1998) and are likely to differ widely in the composition and mycostatic effectiveness of these compounds (Bisset et al., 1966, 1971; Norhayati et al., 2013).
In addition to constitutive chemical defences, some of the taxonomic variation in the susceptibility to stem rot may be due to differences in induced defenses against fungal or insect pathogens (Pearce, 1996; Kovalchuk et al., 2013). When wounding allows exposure to pathogens, anatomical modification of xylem in the living sapwood, including compartmentalization, limits the spread of infection (Shigo, 1984; Pearce, 1996), and the extent and effectiveness of this response likely differs among species (Guariguata and Gilbert, 1996; Romero and Bolker, 2008). During formation, heart wood is suffused with secondary metabolites considered inimical to fungal growth (Yamada, 2001; Taylor et al., 2002; Kirker et al., 2013), and interspecific variation in this process may also affect susceptibility to fungal rot. Even after accounting for species identity, most variation in stem rot frequency and severity remained unexplained. Stem rot infection may be highly stochastic because it appears to require both wounding and subsequent colonization by fungal spores or their insect vectors, which have varying dispersal capacities (Peay and Bruns, 2014) and host requirements (Gilbert, 2002).
Density-dependent population mortality caused by the differential susceptibility of tree species to pathogens has been hypothesized to explain the relative abundance of tree species in forest communities (Comita et al., 2010; Mangan et al., 2010). Given that stem rot affects dead tissue, it is unclear whether this hypothesis also applies to wood decay fungi. Stem rot, often when combined with other stressors, has been implicated in tree death, as it is thought to make trees more vulnerable to sources of mortality (Franklin et al., 1987). The high stem rot frequency and severity among these Bornean species is surprising in this light, and their great longevity (Whitmore, 1984) suggests a large capacity to tolerate stem rot. Stem rot may structurally weaken trees and predispose them to buckling from wind-throw or other disturbances, which often vary in frequency in relation to soil properties and topography (Gale and Hall, 2001; Ohkubo, 2007). The severity of stem rot required for biomechanical failure may be high: only when the radius of the hollow region is ca. 70 % of the total stem radius, which would constitute a loss of stem volume due to stem rot of 49 %, is structural failure viewed to become considerably more likely (Mattheck et al., 2006; Ruxton, 2014). Among Bornean trees with any stem rot in the felling data, 6 % had stem rot of this severity or greater. Whether stem rot contributes to tree death and if so, how, are topics that merit more investigation in tropical forests.
When tree-level stem rot was scaled to the stand-level, we found large
spatial variation in the potential ecosystem stem biomass lost to rot in
central Sarawak. Stems of trees
The effect of stem rot on standing biomass showed strong spatial variation,
and was significantly greater for stands growing on less fertile soil. An
analysis in a lowland Bornean rainforest found that AGB positively correlated
with surface soil nutrient concentrations, including P, K, and Mg, due to the
increased stem density of trees
It is difficult to determine if current methods of biomass estimation adequately account for stem rot in tropical trees. In principle, stem rot may be implicitly incorporated into allometric equations used to estimate AGB from forest inventories (e.g., Chave et al., 2005), which are empirically derived from destructive harvest data sets likely to include trees with stem rot. In practice, however, large trees are often severely under-represented in such allometric data sets. Moreover, the strong variability in biomass loss among species and edaphic habitats in this study indicate that site-specific corrections for stem rot may be needed. Thus, greater consideration of soil conditions and broader-scale quantification of stem rot using standardized methods are critical to improving the estimation of carbon storage in tropical forests.
Stem rot is a poorly quantified source of error in aboveground biomass estimation throughout the tropics. Our study of stem rot frequency and severity in mixed-dipterocarp forests in Sarawak Borneo indicates that spatial variation among forest stands in biomass losses to stem rot coincidences with variation in soil-related factors, which may influence patterns of tropical forest carbon storage across edaphically heterogeneous landscapes. Moreover, the considerable taxonomic variation in heart rot susceptibility observed here could potentially underlie differences in the ecosystem consequences of stem rot among tropical regions with contrasting biogeographic histories. Consequently, using standardized, nondestructive methods to quantify stem rot across tropical regions and environmental gradients would help better constrain estimates of carbon dynamics in tropical forests.
We thank Sarawak Conservator of Forests for permission to use Central Sarawak inventory and soils data and photographs for research, and Sarawak Forest Department and the Protected Areas and Biodiversity Conservation unit of Sarawak Forestry Corporation for permission to conduct research in Lambir Hills National Park. We are grateful to many colleagues in the FIDP inventory, which was a substantial team effort involving scores of people working for several field seasons in difficult conditions. We thank two anonymous reviewers, James Dalling, and Amy Zanne for providing comments that improved this manuscript. Data collection at Lambir was funded by the Center for Tropical Forest Science, NSF DUE-0531920, the University of Nebraska – Lincoln (UNL) UCARE undergraduate research program, and the UNL School of Biological Sciences. Edited by: J. Schöngart