Modelling the long-term dynamics of tropical forests: From leaf traits to whole-tree growth patterns
Introduction
Tropical forests provide multiple social, ecological and economical services, represent the most diverse terrestrial ecosystem, and play an important role in the global carbon cycle (Hassan et al., 2005; Heywood and Watson 1995; Malhi and Grace 2000). In addition to high tree diversity, there are even more plant and animal species that directly or indirectly depend on the resources and habitat provided by trees (Erwin, 1988, Nakamura et al., 2017) Almost 9% of all vascular plant species, for instance, live as epiphytes on trees, predominantly in subtropical or tropical regions (Zotz et al., 2021). Ongoing deforestation and potential adverse effects of climate change thus pose a threat to all species associated with tropical forests (Wright 2005). To assess the impact of a changing environment on tropical biodiversity, we thus need a mechanistic understanding of the functioning of these forests, e.g., how they will respond to those changes, and how associated species respond to changing forest dynamics.
A variety of dynamic models are available to predict future changes of tropical forests and to analyze the functioning of these ecosystems. These models differ considerably in the level of detail and in temporal and spatial resolution. Among these models, dynamic global vegetation models focus on large-scale predictions of vegetation dynamics and carbon cycles, but commonly use simplistic forest structures (e.g., Cramer et al., 2001, Purves and Pacala 2008). At small to medium scales (<1 ha to >100 km2), forest gap models and forest landscape models simulate forest dynamics and tree species composition (Bugmann 2001; Petter et al., 2020). Such models represent forest structure in more detail by including stems and crowns of individual trees or cohorts, allowing simulations of within-canopy light attenuation and competition among different species or functional types of trees (Fischer et al., 2016; Köhler and Huth 1998). Even finer structural details can be simulated by functional-structural tree models (FSTMs), in which trees are represented in 3D space by interconnected structural and functional units, such as branches, leaves, or reproductive organs (Godin and Sinoquet 2005; Sievänen et al., 2014). These ‘virtual trees’ explicitly integrate complex, mechanistic interactions between tree architecture and physiological processes, for instance, light-dependent within-tree carbon acquisition and allocation at the meristem level in growing trees (Fourcaud et al., 2008; Sterck et al., 2005). FSTMs are therefore suitable tools to study structural tree growth as an emergent property of lower level ecophysiological processes. Furthermore, integrating FSTMs into forest stand models can improve our understanding of the fundamental principles that interlink tree and forest dynamics. In addition, such detailed 3D forest models could be useful for model-based studies of the spatio-temporal dynamics of canopy-dwelling plants and animals (Cabral et al., 2015; Petter et al., 2021). However, only few attempts have been made to couple FSTMs with forest stand models, and these studies focused on growth of even-age monocultures over short time periods (Feng et al., 2011; Guillemot et al., 2014). So far, species-rich forest models that are based on FSTMs and simulate all crucial demographic processes over long periods of time, i.e., regeneration, growth and mortality, are not available.
Extending FSTMs to tropical forest stands is both computationally and conceptually challenging. On the one hand, stand-scale FSTMs are computer-intensive due to their complexity and thus require efficient modelling techniques to keep simulations feasible. On the other hand, tropical forests pose particular challenges due to their high species diversity (Slik et al., 2015; ter Steege et al. 2013). In contrast to temperate forests, for which the low number of well-studied tree species allows calibrating ecophysiological parameters at the species level (e.g., Seidl et al., 2012), alternative approaches are required for species-rich forests where species-specific physiological information is scarce. Individual-based models for tropical forests typically use distinct functional groups aggregating tree species with similar characteristics (e.g., Fischer et al., 2016; Köhler and Huth 1998). In the simplest case, only light-demanding pioneers and shade-tolerant climax species are distinguished (Swaine and Whitmore 1988), but a classification into more groups has also been proposed (Chazdon et al., 2010; Gourlet-Fleury et al., 2005). While functional group approaches are useful, they retain a simplification of the continuum from fast growing, short-lived pioneer to slow growing, long-lived shade-tolerant species (Denslow 1987; Wright et al., 2003). Similar trade-offs between growth and mortality occur at the leaf scale (Wright et al., 2004): many leaf traits co-vary strongly, and this variation is largely explained by a single principal axis - the leaf economics spectrum (LES). The LES runs from leaves with high photosynthetic capacities and low life spans to leaves with low photosynthetic capacities and long life spans. Hence, a relationship between leaf traits and whole-tree performance can be assumed, and significant relationships were observed for many tropical tree species (Poorter and Bongers 2006; Sterck et al., 2006). A leaf trait-based approach should thus be a promising way to integrate the different life history strategies of trees into forest models. However, we are not aware of any study in which 3D growth over a tree´s entire life span has been modelled as an emergent property of the tree´s set of traits.
Here, we present a dynamic forest stand model in which trees are represented as 3D functionally- and structurally-explicit individuals. This model simulates the long-term forest dynamics (500–1000 years) at the plot scale (∼1 ha) with a high level of structural detail and functional diversity. Branches are considered up to the second order and leaf biomass development is modelled at a resolution of 1 m3, which allows detailed consideration of competition for light and space. Tree species are characterized by a set of leaf traits under consideration of the trade-offs and correlations between traits (LES; Wright et al., 2004). Using the principles of the pipe model theory (Shinozaki et al., 1964), the light-driven carbon assimilation at the leaf level and the within-tree carbon allocation are coupled. We hypothesize that this continuous, trait-based approach captures essential life history variations across species regarding their growth, survival, and light demand. In addition, we hypothesize that the long-term dynamics of diverse, tropical forest communities can be reproduced by coupling the FSTM with a forest stand model, in which key demographic processes and between-tree competition are simulated (e.g., competition for light and space, neutral recruitment). To test these hypotheses, we contrast emerging growth patterns at the tree level (e.g., diameter growth rates, maximum height and life span) against observations. In addition, we simulate forest stands from bare soil over 500 years and test if the model can successfully reproduce multiple attributes of Neotropical forests. Specifically, we test if a dynamic equilibrium is reached and if 12 forest attributes (e.g., basal area, net primary production, mortality rate) and several ecological patterns (e.g., diameter and height distribution, vertical leaf area distribution) resemble field observations. Our main purpose here is to present the structure, parameterization and validation of the model, and to explore how trait trade-offs at the leaf level are functionally linked to emergent patterns of tree growth and forest stand dynamics. Our model can form the basis for more detailed future studies of (successional) dynamics, distribution of functional and structural traits and tree diversity in detailed 3D forests, or spatiotemporal dynamics of canopy dwellers, making it an important tool for forest and canopy ecologists.
Section snippets
Methods
The model developed in this study generates 3D structural growth of individual trees and (long-term) dynamics of forest stands at the plot level. We term this model a functional-structural forest model (FSFM). This ecophysiological model was implemented using the open-source 3D modelling platform GroIMP (Growth Grammar Interactive Modeling Platform; available at https://sourceforge.net/projects/groimp). In GroIMP, relational growth grammars are implemented in the programming language XL, which
Results
We found an appropriate combination of parameter values that reproduced the ecological pattern at the tree and forest level simultaneously (Table 2; Appendix A: Table A.6). The emergent patterns using this best combination of parameter values are described below. Additionally, the effects of variations of the free model parameters and the two main traits (SLA, wood density) on forest model results are shown in Table 4.
Discussion
When comparing our model with other individual- or cohort-based forest models (e.g., Huth and Ditzer, 2000, Köhler and Huth, 1998, Liu and Ashton, 1998, Phillips et al., 2004), two main differences emerge. First, our model simulates 3D tree structures in detail, including branch segments up to the second order and within-tree leaf distribution at high resolution (here, at 1 m3). This allows three-dimensional light and space competition within and between individuals. Second, tree species are
Conclusion and perspectives
The position of leaf traits along the leaf economics spectrum determined the maximum height and age, size-dependent growth rates and shade tolerance, as well as a smooth continuous transition from fast-growing, short-lived pioneers to slow-growing, long-lived climax species as seen in the tree-level patterns when varying SLA. These emergent results were remarkably consistent with well-known functional tree types. Therefore, our model reveals a fundamental role of leaf traits in determining (a)
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.P. and H.K. were funded by the DFG Initiative of Excellence Free Floater Program at the University of Göttingen. J.S.C. acknowledges financial support by the German Research Foundation (DFG SA-21331). We thank Nadja Rüger, Volker Grimm and three anonymous reviewers for helpful comments on the manuscript.
References (155)
- et al.
Detecting trends in tree growth: not so simple
Trends Plant Sci.
(2013) Research in large, long-term tropical forest plots
Trends Ecol. Evol. (Amst.)
(1995)- et al.
Lessons learned from applying a forest gap model to understand ecosystem and carbon dynamics of complex tropical forests
Ecol. Modell.
(2016) - et al.
A standard protocol for describing individual-based and agent-based models
Ecol. Modell.
(2006) - et al.
The ODD protocol: a review and first update
Ecol. Modell.
(2010) - et al.
Simulation of the growth of a lowland Dipterocarp rain forest with FORMIX3
Ecol. Modell.
(2000) - et al.
The effects of tree species grouping in tropical rainforest modelling: simulations with the individual-based model FORMIND
Ecol. Modell.
(1998) Mathematical models for cellular interactions in development. II. Simple and branching filaments with two-sided inputs
J. Theor. Biol.
(1968)Mathematical models for cellular interactions in development. I. Filaments with one-sided inputs
J. Theor. Biol.
(1968)- et al.
FORMOSAIC: an individual-based spatially explicit model for simulating forest dynamics in landscape mosaics
Ecol. Modell.
(1998)
Tropical forests and atmospheric carbon dioxide
Trends Ecol. Evol. (Amst.)
Forests and their canopies: achievements and horizons in canopy science
Trends Ecol. Evol.
Tree species diversity and structure of a Nigerian strict nature reserve
Trop. Ecol.
Tree allometry and crown shape of four tree species in Atlantic rain forest, south-east Brazil
J. Trop. Ecol.
The role of wood mass density and mechanical constraints in the economy of tree architecture
Am. Nat.
Above-and below-ground net primary productivity across ten Amazonian forests on contrasting soils
Biogeoscience
Crown damage and the mortality of tropical trees
New Phytol.
Forest canopy gap distributions in the southern Peruvian Amazon
PLoS ONE
Variation in wood density determines spatial patterns in Amazonian forest biomass
Glob. Chang. Biol.
What controls tropical forest architecture? Testing environmental, structural and floristic drivers
Global Ecol. Biogeogr.
Tropical forest wood production: a cross-continental comparison
J. Ecol.
Plant architecture: a dynamic, multilevel and comprehensive approach to plant form, structure and ontogeny
Ann. Bot.
Leaf δ13C in Pinus resinosa trees and understory plants: variation associated with light and CO2 gradients
Oecologia
Long-term stem inventory data from tropical rain forest plots in Australia
Ecology
Gap-phase regeneration in a tropical forest
Ecology
Toward a metabolic theory of ecology
Ecology
A review of forest gap models
Clim. Change
Branchfall as a demographic filter for epiphyte communities: lessons from forest floor-based sampling
PLoS ONE
Inferring Amazon leaf demography from satellite observations of leaf area index
Biogeosciences
Allometric relationships predicting foliar biomass and leaf area:sapwood area ratio from tree height in five Costa Rican rain forest species
Tree Physiol.
Quantitative ecological inventory of terra firme and várzea tropical forest on the Rio Xingu, Brazilian Amazon
Brittonia
The steady-state mosaic of disturbance and succession across an old-growth Central Amazon forest landscape
Proc. Natl. Acad. Sci.
Growth and wood density predict tree mortality in Amazon forests
J. Ecol.
Spatial and temporal variation of biomass in a tropical forest: results from a large census plot in Panama
J. Ecol.
Regional and seasonal patterns of litterfall in tropical South America
Biogeosciences
Photosynthetic light environments in a lowland tropical rain forest in Costa Rica
J. Ecol.
Composition and dynamics of functional groups of trees during tropical forest succession in northeastern Costa Rica
Biotropica
Assessing the growth of tropical rain forest trees: issues for forest modeling and management
Ecol. Appl.
Getting to the canopy: tree height growth in a Neotropical rain forest
Ecology
Herbivory and defensive characteristics of tree species in a lowland tropical forest
Ecol Monogr.
Global response of terrestrial ecosystem structure and function to CO2 and climate change: results from six dynamic global vegetation models
Glob. Chang. Biol.
Why do trees die? Characterizing the drivers of background tree mortality
Ecology
Tree mortality and growth in 11 sympatric Macaranga species in Borneo
Ecology
Tropical rainforest gaps and tree species diversity
Annu. Rev. Ecol. Syst.
Structure and biomass of four lowland Neotropical forests
Biotropica
Seasonal patterns of tropical forest leaf area index and CO 2 exchange
J. Geophys. Res.
Extensions and evaluations of a general quantitative theory of forest structure and dynamics
Proc. Natl. Acad. Sci. U.S.A.
The tropical forest canopy: The heart of biotic diversity
Photosynthetic acclimation of plants to growth irradiance: the relative importance of specific leaf area and nitrogen partitioning in maximizing carbon gain
Plant Cell Environ.
Theories of optimization, form and function in branching architecture in plants
Funct. Ecol.
Cited by (4)
Drivers of wildfire burn severity in the montane rainforests of northern Vietnam
2023, International Journal of Wildland FireModelling and Design of Habitat Features: Will Manufactured Poles Replace Living Trees as Perch Sites for Birds?
2023, Sustainability (Switzerland)The road to integrate climate change projections with regional land-use–biodiversity models
2023, People and Nature