A model‐based meta‐analysis for estimating species‐specific wood density and identifying potential sources of variation

Ogle, Kiona; Pathikonda, Sharmila; Sartor, Karla; Lichstein, Jeremy W.; Osnas, Jeanne L. D.; Pacala, Stephen W.

doi:10.1111/1365-2745.12178

Cited by 25 publications

(34 citation statements)

References 68 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, the American Forest's National Register of Big Trees (http://www.americanforests.org/our-programs/bigtree/) reports heights of about 28 m and 50 m for “champion trees” of red maple and loblolly pine, respectively; the posteriors are reasonable given that we should not expect “champion tree” H max values from our sample of the FIA data. Surprisingly, the FIA data also appear to contain some information about ρ , resulting in posterior estimates of ρ that are slightly higher than the prior means but in agreement with a recent meta‐analysis of published values of ρ (Ogle et al ).…”

Section: Discussionsupporting

confidence: 85%

Reversible jump MCMC for inference in a deterministic individual–based model of tree growth for studying forest dynamics

2013

Self Cite

View full text Add to dashboard Cite

Scientists use deterministic models to study and forecast the behavior of complex environmental processes, with increasing emphasis on incorporating data to inform model input parameters and accounting for parameter uncertainty. We work with a deterministic, individual‐based model (IBM) of tree growth and mortality, which is under development to explore forest dynamics. Some values of IBM input parameters cause premature virtual tree mortality relative to the actual mortality status of an observed tree. This discordance in mortality causes dimension changes in the state of a stochastic implementation of IBM outputs and leads us to address trans‐dimensional moves among states with a novel formulation of reversible jump Markov chain Monte Carlo (RJMCMC). In particular, we present an RJMCMC algorithm that uses a continuously supported, multidimensional index—the IBM input parameter—instead of a discrete index typical of model determination applications. We use both synthetic data and data from the Forest Inventory and Analysis database representing two tree species. We compare results for each dataset and species between our reversible jump (RJ) specification and an alternative, non‐RJ specification. The RJ formulation compares favorably to the non‐RJ formulation with regard to achieving convergence and yielding biologically realistic IBM input parameter estimates. Copyright © 2013 John Wiley & Sons, Ltd.

show abstract

Section: Discussionsupporting

confidence: 85%

Reversible jump MCMC for inference in a deterministic individual–based model of tree growth for studying forest dynamics

2013

Self Cite

View full text Add to dashboard Cite

show abstract

“…, Ogle et al. , ). Below, we first summarize the ACGCA model; then, we provide an overview of the data sources used to inform the ACGCA model, followed by a description of the Bayesian approach used to fit the ACGCA model to the data, including a description of the custom MH algorithm and how the Bayesian results were analyzed.…”

Section: Methodsmentioning

confidence: 99%

“…, Ogle et al. , ; see Tree functional traits database ). The FIA database provided repeated measurements for radius and height data for 965,003 individual trees occurring in approximately 100,779 plots that are an unbiased sample of forested areas in the United States (Bechtold and Patterson ).…”

Section: Methodsmentioning

confidence: 99%

“…, Ogle et al. , ) to help constrain parameter values (traits) to realistic ranges. To achieve this, we employed a Bayesian framework that utilized a fast, custom stochastic algorithm for fitting the ACGCA model to the aggregated FIA data and literature information, which produced a 32‐dimensional (joint) posterior distribution of parameters (traits) that lead to realistic tree growth.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Multidimensional trait space informed by a mechanistic model of tree growth and carbon allocation

et al. 2018

Self Cite

View full text Add to dashboard Cite

Citation: Fell, M., J. Barber, J. W. Lichstein, and K. Ogle. 2018. Multidimensional trait space informed by a mechanistic model of tree growth and carbon allocation. Ecosphere 9(1):e02060. 10. 1002/ecs2.2060 Abstract. Plant functional traits research has revealed many interesting and important patterns among morphological, physiological, and life-history traits and the environment. These are exemplified in trade-offs between groups of traits such as those embodied in the leaf and wood economics spectra. Inferences from empirical studies are often constrained by the correlative nature of the analyses, availability of trait data, and a focus on easily measured traits. However, empirical studies have been fundamental to modeling endeavors aiming to enhance our understanding of how functional traits scale up to affect, for example, community dynamics and ecosystem productivity. Here, we take a complementary approach utilizing an individual-based model of tree growth and mortality (the allometrically constrained growth and carbon allocation [ACGCA] model) to investigate the theoretical trait space (TTS) of North American trees. The model includes 32 parameters representing allometric, physiological, and anatomical traits, some overlapping leaf and wood economics spectra traits. Using a Bayesian approach, we fit the ACGCA model to individual tree heights and diameters from the USFS Forest Inventory and Analysis (FIA) dataset, with further constraints by literature-based priors. Fitting the model to 1.3 million FIA records-aggregated across individuals, species, and sites-produced a posterior distribution of traits leading to realistic growth. We explored this multidimensional posterior distribution (the TTS) to evaluate trait-trait relationships emerging from the ACGCA model, and compare these against empirical patterns reported in the literature. Only three notable bivariate correlations, among 496 possible trait pairs, were contained in the TTS. However, stepwise regressions uncovered a complicated structure; only a subset of traits-related to photosynthesis (e.g., radiation-use efficiency and maintenance respiration)-exhibited strong multivariate trade-offs with each other, while half of the traits-mostly related to allometries and construction costs-varied independently of other traits. Interestingly, specific leaf area was related to several rarely measured root traits. The trade-offs contained in the TTS generally reflect mass-balance (related to carbon allocation) and engineering (mostly related to allometries) trade-offs represented in the ACGCA model and point to potentially important traits that are under-explored in field studies (e.g., root traits and branch senescence rates).

show abstract

“…, Ogle et al. , ) and/or values used to develop and test the ACGCA model (Ogle and Pacala ), as described in Fell et al. ().…”

Section: Methodsmentioning

confidence: 99%

Refinement of a theoretical trait space for North American trees via environmental filtering

Fell

Ogle

2018

Ecological Monographs

Self Cite

View full text Add to dashboard Cite

We refer to a theoretical trait space (TTS) as an n‐dimensional hypervolume (hypercube) characterizing the range of values and covariations among multiple functional traits, in the absence of explicit filtering mechanisms. We previously constructed a 32‐dimensional TTS for North American trees by fitting the Allometrically Constrained Growth and Carbon Allocation (ACGCA) model to USFS Forest Inventory and Analysis (FIA) data. Here, we sampled traits from this TTS, representing different individual “trees,” and subjected these trees to a series of gap dynamics simulations resulting in different annual light levels to explore the impact of environmental filtering (light stress) on the trait space. Variation in light limitation led to non‐random mortality and a refinement of the TTS. We investigated potential mechanisms underlying such filtering processes by exploring how traits and the environment relate to mortality rates at the tree, phenotype (a specific set of trait values), and stand (a specific gap scenario) levels. The average light level at the forest floor explained 42% of the stand‐level mortality, while phenotype‐ and tree‐level mortality were best explained by six functional traits, especially radiation‐use efficiency, maximum tree height, and xylem conducting area to sapwood area ratio (γX). These six “mortality” traits and six traits related to the leaf and wood economics spectra were used to construct trait hypercubes represented by trees that died or survived each gap scenario. For trees that survived, the volume of their refined trait space decreased linearly with increasing stand‐level mortality (up to ~50% mortality); the location also shifted, as indicated by non‐zero distances between the hypercube centroids of surviving trees compared to dead trees and the original TTS. Overall, the patterns were consistent with empirical studies of functional traits, in terms of which traits predict mortality and the direction of the relationships. This work, however, also identified potentially important functional traits that are not commonly measured in empirical studies, such as γX and senescence rates of relatively long‐lived tissues.

show abstract

A model‐based meta‐analysis for estimating species‐specific wood density and identifying potential sources of variation

Cited by 25 publications

References 68 publications

Reversible jump MCMC for inference in a deterministic individual–based model of tree growth for studying forest dynamics

Reversible jump MCMC for inference in a deterministic individual–based model of tree growth for studying forest dynamics

Multidimensional trait space informed by a mechanistic model of tree growth and carbon allocation

Refinement of a theoretical trait space for North American trees via environmental filtering

Contact Info

Product

Resources

About