Tropical tree allometry and crown allocation, and their relationship with species traits in central Africa

https://doi.org/10.1016/j.foreco.2021.119262Get rights and content

Highlights

  • Aboveground biomass scales with tree diameter and the scaling coefficient varies between species.

  • The average value of scaling coefficients across species is close to the prediction of metabolic theory.

  • Wood density explains interspecific variation in the allometric relationships between aboveground biomass and tree diameter.

  • The percentage of biomass allocated to crown is linearly related to tree diameter and the intercept varies between species.

  • Interspecific variation in biomass allocation to crown depends on species dispersal mode and adult stature.

Abstract

Common allometric patterns have been reported across the tropics and good performance on independent data was retrieved for the most recent pantropical model predicting tree aboveground biomass (AGB) from stem diameter, wood density and total height. General models are undoubtedly useful for the estimation and monitoring of biomass and carbon stocks in tropical forests, however specific allometry, allocation, and traits, are at the core of many models of vegetation dynamics, and there is lack of such information for some regions and species. In this study, we specifically evaluated how size-dependent changes in above-ground biomass and biomass allocation to crown relate to other allometric and life-history traits for tropical tree species. We gathered destructive data available in eight terra firme forest sites across central Africa and the combined dataset consisted of 1,023 trees belonging to 54 tropical tree species phylogenetically dispersed, with only two congeneric species. A huge body of field and laboratory measurements was used for computing AGB and crown mass ratio (CMR) at the tree level, and to derive key allometric traits at the species level. For the latter, species-specific relationships between tree diameter and total height, crown exposure to light, wood density, and bark thickness were fitted for 50 species. Our results show interspecific variation in the relationships relating tree diameter to both AGB and CMR, and including species traits in a multi-specific AGB model confirmed that interspecific variation in biomass allometry is primarily determined by species wood density. We also showed that the allocation of biomass to crown increases linearly with tree diameter for most species, and that interspecific variation in the CMR model is associated with the species dispersal mode and maximum height. Trait covariations among our set of tropical tree species widespread and/or locally abundant in central Africa, revealed a continuum between large-statured species, which tended to be light-demanding, deciduous and wind-dispersed, and species with opposite attributes. Information on allometry, allocation, and traits provided here could further be used in comparative ecology and for parameterizing dynamic and succession models. Also importantly, the species-specific AGB models fitted for major tree species, including most timber species of central Africa, will help improve biomass estimates.

Introduction

Accurate assessments of biomass and carbon stocks in tropical forests underpin policies that aim to mitigate carbon dioxide emissions such as the UN-REDD + program and the recommendations of the Intergovernmental Panel on Climate Change (Gibbs et al., 2007) and using allometric equations to predict the aboveground biomass (AGB) of tropical trees is a cost-effective and accurate approach (Chave et al., 2004, 2014). In the last decades, general multi-specific allometric models have been updated, with the pantropical model of Chave et al. (2014) refining the previously developed models (Brown et al., 1989, Chave et al., 2005). The most recent pantropical model of Chave et al. (2014) has been shown to perform well at the tree level, on independent dataset of destructive AGB in a peat swamp forest in Indonesia (Manuri et al., 2014) and in six terra firme forest sites distributed across central Africa (Fayolle et al., 2018).

After tree diameter, species wood density and total tree height are both important AGB predictors for tropical trees (Brown et al., 1989, Chave et al., 2014, Chave et al., 2005, Feldpausch et al., 2011, Van Breugel et al., 2011), respectively expected to explain between-species and between-site variations (Ketterings et al., 2001). The inclusion of height has indeed been shown to offset allometric variation across forest types (Chave et al., 2005), and when not available locally, height can possibly be predicted from tree diameter using a local or a regional equation (Feldpausch et al., 2011) or using a general equation including an environmental stress variable (Chave et al., 2014). In addition to the classical AGB predictors, the use of crown information has recently received attention (Goodman et al., 2014, Fayolle et al., 2018), and its importance to predict AGB has been notably explained by ontogenic and/or size-related changes in biomass allocation to crown (Ploton et al., 2016). Besides wood density averaged at species (Molto et al., 2013) or even at genus level (Slik, 2005), interspecific variation in tropical tree allometry and allocation have not been deeply explored. Because it is extremely costly, time-consuming, and difficult to collect and assemble biomass data at the tree and species levels (Picard et al., 2012), many questions about the usefulness of generic models remain unanswered. Still, it is unclear to what extent data on biomass allometry should be pooled or separated according to morphological, phylogenetic and/or phenological characteristics of species (Paul et al., 2016). In the latter study, specific allometric equations have been developed for species grouped into a few plant functional types covering contrasted biomes and ecoregions across Australia. It has also been argued that species-specific models are especially needed when the gains in accuracy at the stand level are high, which is true for high-value monocultures (Paul et al., 2016) or for monodominant species such as Gilbertiodendron dewevrei in central Africa, and for which a specific allometry has been developed (Umunay et al., 2017). Specific allometric relationships have been developed for species grouped at genus level in a Dipterocarp forest of south-eastern Asia (Basuki et al., 2009) or by family and wood density in peat swamp forests in Indonesia (Manuri et al., 2014). In the latter study, it has to be noted, however, that the multi-specific pantropical model showed similar performance as the local models.

Interspecific variation in tropical tree allometry has been demonstrated in the early stages (seedlings and saplings) and related to life-history strategy and traits in Panama (King, 1990, King, 1996). The allometric specialization of light-demanding species was found to be associated with biomass allocation toward efficient height growth, and that of understory species with the maximization of light interception and the persistence in the shaded understory (King, 1990). Light-demanding species are indeed generally more exposed to light at 10 cm diameter (Sheil et al., 2006) because of greater height, as reported in Bolivia (Poorter et al., 2006) and in northern Congo (Loubota Panzou et al., 2018). It is thus important to both consider allometry and allocation in evaluating the adaptive significance of interspecific variation. In addition, grouping tropical tree species into meaningful groups of tree species is not straightforward (Swaine and Whitmore, 1988) and only a few plant functional types are generally recognized in Dynamic Global Vegetation Models (DGVM, Fisher et al., 2018), generally two groups, evergreen vs deciduous tropical trees. In the successional model TROLL (Maréchaux and Chave, 2017), up to 12 tropical plant functional types are recognized, and for each of them, the tree geometry is modelled explicitly with specific allometric relationships relating tree diameter, to height, crown radius and depth. Though plant functional types are at the core of many models of vegetation dynamics or of forest succession, the lack of information on allometry, allocation and traits, for some species and regions hampers models predictions and this is particularly true for central Africa which is a largely under-sampled region.

In this study, we evaluated how size-dependent changes in above-ground biomass and biomass allocation to crown relate to other allometric and life-history traits for tropical tree species. We addressed three questions. First, to what extent biomass allometry and allocation vary among a set of tropical tree species? Using published data of destructive biomass in central Africa, we were able to provide new information on biomass allometry and allocation for 54 tropical tree species widespread and/or locally abundant, and covering a variety of genera and families. Second, is interspecific variation in allometry and allocation related to key traits? We notably tested the hypothesis of contrasted biomass allometry and allocation, between evergreen and deciduous species, but besides these two groups, we further investigated interspecific variation in response to life-history traits extracted from literature and to allometric traits newly developed for 50 species, and characterizing species architecture, light requirement, the wood economics spectrum, and adaptation to fire (Table 1). Third, which allometric or life-history traits can be used in multi-specific allometry and allocation models as a surrogate of interspecific variation? We notably tested the hypothesis that wood density explains interspecific variation in biomass allometry (Ketterings et al., 2001) and that light requirement explains interspecific variation in biomass allocation to crown (King, 1990).

Section snippets

Study sites

Destructive biomass data were available for more than 1,000 trees sampled across central Africa in eight sites (Fig. 1) which includes the six sites sampled in the PREREDD+ project (sites #1–2 and #5–8 on the map, Fayolle et al., 2018), and two sites sampled earlier, the Zadié site in Gabon (#3, Ngomanda et al., 2014) and the Mindourou site in Cameroon (#4, Fayolle et al., 2013). In each site, the sampling covered a large number of trees (≥100) and a vast range of diameters and wood density, as

Tree allometry

Benefitting from existing destructive AGB data in eight sites across central Africa, we were able to fit species-specific allometric models for 54 tropical tree species (5–69 trees per species, mean of 19). As expected, we found strong interspecific variation in tree allometry when only considering the bivariate relationships between tree AGB and D (Fig. 2). The coefficients of AGB models varied among the 54 tropical tree species with a mean of 2.45 (± 0.24 standard deviation) found for the

Discussion

In accordance with the predictions of the metabolic theory, under certain hypotheses of biomechanical constraints of tree stability and hydraulic resistance in conductive cell networks, tree AGB scales with tree diameter with an exponent of 8/3 (West et al., 1997). The metabolic theory has been hotly debated, specifically its general nature and its underlying assumption of invariant scaling coefficient (Zianis and Mencuccini, 2004, Muller-Landau et al., 2006). Here, the scaling coefficient of

Conclusions

Using published destructive biomass data of tropical trees in eight sites representative of terra firme forests across central Africa we were able to provide new information on biomass allometry and allocation, for a set of 54 tropical tree species widespread and/or locally abundant, and phylogenetically dispersed. New allometric traits were derived for 50 species and were both related to the coefficients of the allometry and allocation models, and were also included in multi-specific models in

CRediT authorship contribution statement

Géraud Sidoine Mankou: Conceptualization, Formal analysis, Investigation, Writing - original draft, Writing - review & editing. Gauthier Ligot: Conceptualization, Formal analysis, Investigation, Writing - review & editing. Grace Jopaul Loubota Panzou: Conceptualization, Investigation, Writing - review & editing. Faustin Boyemba: Data curation, Funding acquisition, Project administration, Writing - review & editing. Jean-Joël Loumeto: Data curation, Funding acquisition, Project administration,

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

G.M. benefited from a fellowship of the International Tropical Timber Organization [ITTO grant #084/17A]. The data collection was supported by Gembloux Agro-Bio Tech, Université de Liège through the EBALAC project, by the ACP Secretariat and the European Commission under the EUACP ‘‘Establishment of a forestry research network for ACP countries’’ [9 ACP RPR 91#1–FORENET], and by a gift from the Global Environment Fund [TF010038] administered by the World Bank to the COMIFAC which funded the

References (53)

  • J. Chave et al.

    Error propagation and scaling for tropical forest biomass estimates

    Philos. Trans. R. Soc. Lond. B. Biol. Sci.

    (2004)
  • J. Chave et al.

    Towards a worldwide wood economics spectrum

    Ecol. Lett.

    (2009)
  • J. Chave et al.

    Improved allometric models to estimate the aboveground biomass of tropical trees

    Glob. Change Biol.

    (2014)
  • Fayolle, A., Ngomanda, A., Mbasi, M., Barbier, N., Bocko, Y., Boyemba, F., Couteron, P., Fonton, N., Kamdem, N.,...
  • A. Fayolle et al.

    Patterns of tree species composition across tropical African forests

    J. Biogeogr.

    (2014)
  • T.R. Feldpausch et al.

    Height-diameter allometry of tropical forest trees

    Biogeosciences

    (2011)
  • S.E. Fick et al.

    WorldClim 2: new 1-km spatial resolution climate surfaces for global land areas

    Int. J. Climatol.

    (2017)
  • R.A. Fisher et al.

    Vegetation demographics in Earth System Models: A review of progress and priorities

    Glob. Change Biol.

    (2018)
  • H.K. Gibbs et al.

    Monitoring and estimating tropical forest carbon stocks: making REDD a reality

    Environ. Res. Lett.

    (2007)
  • J.-F. Gillet et al.

    A commented checklist of woody plants in the Northern Republic of Congo

    Plant Ecol. Evol.

    (2012)
  • R.C. Goodman et al.

    The importance of crown dimensions to improve tropical tree biomass estimates

    Ecol. Appl.

    (2014)
  • K. Guan et al.

    Photosynthetic seasonality of global tropical forests constrained by hydroclimate

    Nat. Geosci.

    (2015)
  • Hawthorne, W.D., 1995. Ecological profiles of Ghanaian forest trees, Tropical Forestry Papers. Oxford Forestry...
  • Jarvis, A., Reuter, H.I., Nelson, A., Guevara, E., 2008. Hole-filled SRTM for the globe Version 4. Available CGIAR-SXI...
  • T. Jucker et al.

    Competition for light and water play contrasting roles in driving diversity–productivity relationships in Iberian forests

    J. Ecol.

    (2014)
  • T. Jucker et al.

    Allometric equations for integrating remote sensing imagery into forest monitoring programmes

    Glob. Change Biol.

    (2017)
  • Cited by (0)

    View full text