Common ground among beech forest development stages: Matrix versus stage-typical live tree structure

Jeri Lynn E. Peck, Eric Zenner

Research output: Contribution to journalArticle

Abstract

Questions: Doubt has recently been shed on the validity of the European patch mosaic-forest life cycle model involving the assignment of stages of development to natural forests. To partially explain the inconsistency in development stage assignment, we ask: does the high neighborhood-scale structural heterogeneity inherent to old-growth beech forests subject to small-scale gap dynamics transcend development stages?. Location: Natural Oriental beech (Fagus orientalis Lipsky) forests in four regions of the Elborz Mountain range in northern Iran. Methods: Natural tree neighborhoods at three different scales (ca. 112, 508, and 1,228 m2) were identified within 1.0 ha plots in each of the three main development stages (Initial, Optimum, Decay) in each region using the spatially explicit Delaunay-triangulation-based floating neighborhood approach. The diameter distribution across all neighborhoods in all three stages was summarized using Principal Components Analysis. Neighborhoods exhibiting shared structure among stages were identified as those within one standard deviation of the centroid of the 2D ordination space. Results: Shared neighborhoods, which were highly heterogeneous with a weakly rotated sigmoid size class structure, were found to consistently occur in ca. 10%–20% of neighborhoods in all three development stages. Only neighborhoods more than one, and particularly more than two, standard deviations from the centroid differed among development stages in size class structure, with stage-typical distributions observed in each development stage (i.e., negative exponential in the Initial stage, normal in the Optimum, and bimodal in the Decay). Conclusions: Regardless of development stage, forested beech stands subject to a small-scale disturbance regime share a common core of tree neighborhoods with similar, heterogeneous live tree structure. We conjecture that this structure reflects temporal and spatial variation in gap dynamics, with variously aged but relatively newer gap patches with relatively high within-patch homogeneity nested within a matrix of the highly heterogeneous neighborhoods that develop in the absence of recent disturbance.

Original languageEnglish (US)
Pages (from-to)893-904
Number of pages12
JournalJournal of Vegetation Science
Volume30
Issue number5
DOIs
StatePublished - Sep 1 2019

Fingerprint

Fagus
matrix
deterioration
Fagus sylvatica subsp. orientalis
gap dynamics
temporal variation
spatial variation
life cycle (organisms)
Iran
principal component analysis
mountains
disturbance
triangulation
old-growth forest
ordination
homogeneity
life cycle
methodology

All Science Journal Classification (ASJC) codes

  • Ecology
  • Plant Science

Cite this

@article{5cc7372dfcc14e1897609c38abe1a35b,
title = "Common ground among beech forest development stages: Matrix versus stage-typical live tree structure",
abstract = "Questions: Doubt has recently been shed on the validity of the European patch mosaic-forest life cycle model involving the assignment of stages of development to natural forests. To partially explain the inconsistency in development stage assignment, we ask: does the high neighborhood-scale structural heterogeneity inherent to old-growth beech forests subject to small-scale gap dynamics transcend development stages?. Location: Natural Oriental beech (Fagus orientalis Lipsky) forests in four regions of the Elborz Mountain range in northern Iran. Methods: Natural tree neighborhoods at three different scales (ca. 112, 508, and 1,228 m2) were identified within 1.0 ha plots in each of the three main development stages (Initial, Optimum, Decay) in each region using the spatially explicit Delaunay-triangulation-based floating neighborhood approach. The diameter distribution across all neighborhoods in all three stages was summarized using Principal Components Analysis. Neighborhoods exhibiting shared structure among stages were identified as those within one standard deviation of the centroid of the 2D ordination space. Results: Shared neighborhoods, which were highly heterogeneous with a weakly rotated sigmoid size class structure, were found to consistently occur in ca. 10{\%}–20{\%} of neighborhoods in all three development stages. Only neighborhoods more than one, and particularly more than two, standard deviations from the centroid differed among development stages in size class structure, with stage-typical distributions observed in each development stage (i.e., negative exponential in the Initial stage, normal in the Optimum, and bimodal in the Decay). Conclusions: Regardless of development stage, forested beech stands subject to a small-scale disturbance regime share a common core of tree neighborhoods with similar, heterogeneous live tree structure. We conjecture that this structure reflects temporal and spatial variation in gap dynamics, with variously aged but relatively newer gap patches with relatively high within-patch homogeneity nested within a matrix of the highly heterogeneous neighborhoods that develop in the absence of recent disturbance.",
author = "Peck, {Jeri Lynn E.} and Eric Zenner",
year = "2019",
month = "9",
day = "1",
doi = "10.1111/jvs.12791",
language = "English (US)",
volume = "30",
pages = "893--904",
journal = "Journal of Vegetation Science",
issn = "1100-9233",
publisher = "Wiley-Blackwell",
number = "5",

}

Common ground among beech forest development stages : Matrix versus stage-typical live tree structure. / Peck, Jeri Lynn E.; Zenner, Eric.

In: Journal of Vegetation Science, Vol. 30, No. 5, 01.09.2019, p. 893-904.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Common ground among beech forest development stages

T2 - Matrix versus stage-typical live tree structure

AU - Peck, Jeri Lynn E.

AU - Zenner, Eric

PY - 2019/9/1

Y1 - 2019/9/1

N2 - Questions: Doubt has recently been shed on the validity of the European patch mosaic-forest life cycle model involving the assignment of stages of development to natural forests. To partially explain the inconsistency in development stage assignment, we ask: does the high neighborhood-scale structural heterogeneity inherent to old-growth beech forests subject to small-scale gap dynamics transcend development stages?. Location: Natural Oriental beech (Fagus orientalis Lipsky) forests in four regions of the Elborz Mountain range in northern Iran. Methods: Natural tree neighborhoods at three different scales (ca. 112, 508, and 1,228 m2) were identified within 1.0 ha plots in each of the three main development stages (Initial, Optimum, Decay) in each region using the spatially explicit Delaunay-triangulation-based floating neighborhood approach. The diameter distribution across all neighborhoods in all three stages was summarized using Principal Components Analysis. Neighborhoods exhibiting shared structure among stages were identified as those within one standard deviation of the centroid of the 2D ordination space. Results: Shared neighborhoods, which were highly heterogeneous with a weakly rotated sigmoid size class structure, were found to consistently occur in ca. 10%–20% of neighborhoods in all three development stages. Only neighborhoods more than one, and particularly more than two, standard deviations from the centroid differed among development stages in size class structure, with stage-typical distributions observed in each development stage (i.e., negative exponential in the Initial stage, normal in the Optimum, and bimodal in the Decay). Conclusions: Regardless of development stage, forested beech stands subject to a small-scale disturbance regime share a common core of tree neighborhoods with similar, heterogeneous live tree structure. We conjecture that this structure reflects temporal and spatial variation in gap dynamics, with variously aged but relatively newer gap patches with relatively high within-patch homogeneity nested within a matrix of the highly heterogeneous neighborhoods that develop in the absence of recent disturbance.

AB - Questions: Doubt has recently been shed on the validity of the European patch mosaic-forest life cycle model involving the assignment of stages of development to natural forests. To partially explain the inconsistency in development stage assignment, we ask: does the high neighborhood-scale structural heterogeneity inherent to old-growth beech forests subject to small-scale gap dynamics transcend development stages?. Location: Natural Oriental beech (Fagus orientalis Lipsky) forests in four regions of the Elborz Mountain range in northern Iran. Methods: Natural tree neighborhoods at three different scales (ca. 112, 508, and 1,228 m2) were identified within 1.0 ha plots in each of the three main development stages (Initial, Optimum, Decay) in each region using the spatially explicit Delaunay-triangulation-based floating neighborhood approach. The diameter distribution across all neighborhoods in all three stages was summarized using Principal Components Analysis. Neighborhoods exhibiting shared structure among stages were identified as those within one standard deviation of the centroid of the 2D ordination space. Results: Shared neighborhoods, which were highly heterogeneous with a weakly rotated sigmoid size class structure, were found to consistently occur in ca. 10%–20% of neighborhoods in all three development stages. Only neighborhoods more than one, and particularly more than two, standard deviations from the centroid differed among development stages in size class structure, with stage-typical distributions observed in each development stage (i.e., negative exponential in the Initial stage, normal in the Optimum, and bimodal in the Decay). Conclusions: Regardless of development stage, forested beech stands subject to a small-scale disturbance regime share a common core of tree neighborhoods with similar, heterogeneous live tree structure. We conjecture that this structure reflects temporal and spatial variation in gap dynamics, with variously aged but relatively newer gap patches with relatively high within-patch homogeneity nested within a matrix of the highly heterogeneous neighborhoods that develop in the absence of recent disturbance.

UR - http://www.scopus.com/inward/record.url?scp=85071891856&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071891856&partnerID=8YFLogxK

U2 - 10.1111/jvs.12791

DO - 10.1111/jvs.12791

M3 - Article

AN - SCOPUS:85071891856

VL - 30

SP - 893

EP - 904

JO - Journal of Vegetation Science

JF - Journal of Vegetation Science

SN - 1100-9233

IS - 5

ER -