Which demographic processes control competitive equilibria? Bayesian calibration of a size‐structured forest population model

Heiland, Lukas; Künstler, Georges; Šebeň, Vladimír; Hülsmann, Lisa

doi:10.1002/ece3.10232

Cited by 2 publications

(21 citation statements)

References 75 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our simulations largely support the prediction that Fagus predominates in mesic environmental conditions while other species become more prevalent at the extremes of soil pH and water gradients (Section 4.1). Moreover, we substantiate the theory with demographic analysis (Section 4.2): In a previous study, Heiland et al (2023) showed that di erences in mean demographic rates between Fagus and others, especially the shadetolerance of its saplings, explain Fagus' predominance in its core range (these results are largely consistent with the estimates presented in Table 2). Expanding on this approach, here we introduce a model with environmentally-responsive demographic rates to explain the environmental variation of Fagus' relative abundance and why other species prevail towards more stressful environmental conditions.…”

Section: Discussionsupporting

confidence: 88%

“…The JAB model (Heiland et al, 2023) is a simple dynamic forest population model that includes three size stages including the sapling stage. The model concentrates on competition between multiple species and on the species-specific sapling dynamics by assuming a simple two-layer size structure (see also Valladares & Niinemets, 2008; Cordonnier et al, 2019)…”

Section: Methodsmentioning

confidence: 99%

“…Plots with missing sample trees for "unknown reasons" (category only available in the third survey) were also excluded. But unlike Heiland et al (2023), we did not select the plots based on whether Fagus was present or not to sample a greater environmental range of plots.…”

Section: Short Time Series From National Forest Inventory Datamentioning

confidence: 99%

“…The simulated total equilibrium basal area of around 57 m 2 ha 1 (Table S7) is only slightly higher than expected from observations. Stillhard et al, 2022 report total basal areas of ⇠36 m 2 ha 1 from pristine beech-dominated forests in Uholka-Shyrokyi Luh, Ukraine, while Moreno et al (2018) project an upper limit of European forests around 50 m 2 ha 1 (see also Heiland et al, 2023 where the JAB model calibrated with a di erent selection of plots projected maximum carrying capacities below 50 m 2 ha 1 ).…”

Section: Fagus Predominates At the Center Of Environmental Spacementioning

confidence: 99%

“…In particular, Fagus is assumed to outcompete other tree species at mesic soil conditions, so that their ecological optima are shifted towards more extreme conditions (Ellenberg, 1963; Keddy, 2001). Fagus ’ competitive demographics compared to other species include better shade-tolerance of seedlings and saplings (Wagner et al, 2010; Käber et al, 2021), which gives rise to a positive feedback mechanism between sapling recruitment and the shading from the overstory, so that at late successional stages Fagus predominates at the competitive equilibrium (see also Peters, 1997; Leuschner & Ellenberg, 2017b; Petrovska et al, 2021a; Heiland et al, 2023). However, through environmental variation of forest demography, Fagus ceases to be predominant towards the extremes of the soil gradients.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

How do sapling and adult demography explain beech predominance along environmental gradients in Central European forests?

Heiland,

Kunstler,

Hülsmann

2023

Preprint

Self Cite

View full text Add to dashboard Cite

Understanding how species abundances are driven by biotic interactions along environmental gradients is a fundamental question in ecology. For abundances at competitive equilibria in Central European forests, a classical ecological theory formulated by Ellenberg (1963) predicts that beech (Fagus sylvaticaL.) outcompetes other tree species within a mesic range of soil pH and water levels, while other species prevail under less favorable conditions. While the theory is generally accepted in forest ecology, only certain aspects of it have been substantiated by empirical evidence. Moreover, the demographic processes driving the turnover from beech to other tree species at the extremes of the soil gradients remain mostly unexplained.To address this, we inversely calibrated a parsimonious forest model (JAB model) with a sapling stage and interacting populations with short time series of observed tree abundances from the German national forest inventory. By modelling how demographic rates vary along pH and soil water gradients, we were able to test the prediction that beech naturally predominates only at favorable soil conditions. Moreover, we tested with simulations how the environmental response of demographic rates explainsFagus’ changing relative abundance along the soil gradients.Our results largely confirm thatFagusout competes other species in a central environmental range. Environmental change ofFagus’ relative abundance is primarily explained by environmental variation of its net basal area increment, followed by its competition response at the overstory and at the sapling stage. We found that even though sapling tolerance to shading is the primary mechanism forFaguspredominance, it only plays a secondary role for the environmental variation of its relative abundance.Synthesis:By inverse calibration of a forest population model with demographic rates that respond to the environment, we confirm the predictions of Ellenberg’s classical, albeitonlypartially-evidenced, theory onF. sylvatica’s predominance in Central European forests. Furthermore, for thefirsttime, we substantiate the theory by elucidating how the environmental variation in species composition is based in demographic processes. This demonstrates that our approach can be utilized to predict distributions of interacting species and to explain the dynamics between species, as influenced by their environment.

show abstract

Section: Discussionsupporting

confidence: 88%

Section: Methodsmentioning

confidence: 99%

Section: Short Time Series From National Forest Inventory Datamentioning

confidence: 99%

Section: Fagus Predominates At the Center Of Environmental Spacementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

How do sapling and adult demography explain beech predominance along environmental gradients in Central European forests?

Heiland,

Kunstler,

Hülsmann

2023

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Partitioning time series to improve process-based models with machine learning

Boussange

Aceituno²,

Pellissier

2022

Preprint

View full text Add to dashboard Cite

Ecosystems are involved in global biogeochemical cycles that regulate climate and provide essential services to human societies. Mechanistic models are required to describe ecosystem dynamics and anticipate their response to anthropogenic pressure, but their adoption has been limited in practice because of issues with parameter identification and because of model inaccuracies. While observations could be used to directly estimate parameters and improve models, model nonlinearities as well as shallow, incomplete and noisy datasets complicate this process. Here, we propose a machine learning (ML) framework relying on a mini-batch method combined with automatic differentiation and state-of-the-art optimizers. By splitting the data into mini-batches with a short time horizon, we show both analytically and numerically that the mini-batch method regularizes the learning problem. When combined with the proposed numerical implementation, the resulting ML framework can efficiently learn the parameter of complex dynamical models and is a workhorse for model selection. We evaluate the performance of the ML framework in recovering the dynamics of a simulated food-web. We show that it can efficiently learn from noisy, incomplete and independent time series, accurately estimating the model parameters and providing reliable short-term forecasts. We further show that the ML framework can provide statistical support for the true generating model among several candidates. In summary, the proposed ML framework can efficiently learn from data and elucidate mechanistic pathways to improve our understanding and predictions of ecosystem dynamics.

show abstract

Which demographic processes control competitive equilibria? Bayesian calibration of a size‐structured forest population model

Cited by 2 publications

References 75 publications

How do sapling and adult demography explain beech predominance along environmental gradients in Central European forests?

How do sapling and adult demography explain beech predominance along environmental gradients in Central European forests?

Partitioning time series to improve process-based models with machine learning

Contact Info

Product

Resources

About