Spatial pattern analysis in forest dynamics: Deviation from power law and direction of regeneration waves

Robert Schlicht, Yoh Iwasa

Research output: Contribution to journalReview article

8 Citations (Scopus)

Abstract

Grid-based models have been used to understand spatial heterogeneity of the vegetation height in forests and to analyze spatio-temporal dynamics of the forest regeneration process. In this report, we present two methods of identifying lattice models when spatio-temporal data are given. The first method detects directionality of regeneration waves based on the timing of local disturbance events. The second evaluates the forest pattern by recording the fraction of high and low vegetation areas at multiple spatial scales. We illustrate these methods by applying them to patterns generated using three simple stochastic lattice models: (1) two-state model, distinguishing sites with high and low vegetation, (2) three-state model, in which sites can be in an additional disturbed state, and (3) Shimagare model, which considers a continuous range of states. The combination of the two methods provides efficient means of distinguishing the models. The first method has a more direct ecological meaning, while the second is useful when forest data are limited in time.

Original languageEnglish
Pages (from-to)197-203
Number of pages7
JournalEcological Research
Volume22
Issue number2
DOIs
Publication statusPublished - Mar 1 2007

Fingerprint

forest dynamics
power law
regeneration
vegetation
methodology
forest regeneration
analysis
method
disturbance

All Science Journal Classification (ASJC) codes

  • Ecology

Cite this

Spatial pattern analysis in forest dynamics : Deviation from power law and direction of regeneration waves. / Schlicht, Robert; Iwasa, Yoh.

In: Ecological Research, Vol. 22, No. 2, 01.03.2007, p. 197-203.

Research output: Contribution to journalReview article

@article{c088e9dc0ca741df9d93c46b0211656e,
title = "Spatial pattern analysis in forest dynamics: Deviation from power law and direction of regeneration waves",
abstract = "Grid-based models have been used to understand spatial heterogeneity of the vegetation height in forests and to analyze spatio-temporal dynamics of the forest regeneration process. In this report, we present two methods of identifying lattice models when spatio-temporal data are given. The first method detects directionality of regeneration waves based on the timing of local disturbance events. The second evaluates the forest pattern by recording the fraction of high and low vegetation areas at multiple spatial scales. We illustrate these methods by applying them to patterns generated using three simple stochastic lattice models: (1) two-state model, distinguishing sites with high and low vegetation, (2) three-state model, in which sites can be in an additional disturbed state, and (3) Shimagare model, which considers a continuous range of states. The combination of the two methods provides efficient means of distinguishing the models. The first method has a more direct ecological meaning, while the second is useful when forest data are limited in time.",
author = "Robert Schlicht and Yoh Iwasa",
year = "2007",
month = "3",
day = "1",
doi = "10.1007/s11284-006-0016-x",
language = "English",
volume = "22",
pages = "197--203",
journal = "Ecological Research",
issn = "0912-3814",
publisher = "Springer Japan",
number = "2",

}

TY - JOUR

T1 - Spatial pattern analysis in forest dynamics

T2 - Deviation from power law and direction of regeneration waves

AU - Schlicht, Robert

AU - Iwasa, Yoh

PY - 2007/3/1

Y1 - 2007/3/1

N2 - Grid-based models have been used to understand spatial heterogeneity of the vegetation height in forests and to analyze spatio-temporal dynamics of the forest regeneration process. In this report, we present two methods of identifying lattice models when spatio-temporal data are given. The first method detects directionality of regeneration waves based on the timing of local disturbance events. The second evaluates the forest pattern by recording the fraction of high and low vegetation areas at multiple spatial scales. We illustrate these methods by applying them to patterns generated using three simple stochastic lattice models: (1) two-state model, distinguishing sites with high and low vegetation, (2) three-state model, in which sites can be in an additional disturbed state, and (3) Shimagare model, which considers a continuous range of states. The combination of the two methods provides efficient means of distinguishing the models. The first method has a more direct ecological meaning, while the second is useful when forest data are limited in time.

AB - Grid-based models have been used to understand spatial heterogeneity of the vegetation height in forests and to analyze spatio-temporal dynamics of the forest regeneration process. In this report, we present two methods of identifying lattice models when spatio-temporal data are given. The first method detects directionality of regeneration waves based on the timing of local disturbance events. The second evaluates the forest pattern by recording the fraction of high and low vegetation areas at multiple spatial scales. We illustrate these methods by applying them to patterns generated using three simple stochastic lattice models: (1) two-state model, distinguishing sites with high and low vegetation, (2) three-state model, in which sites can be in an additional disturbed state, and (3) Shimagare model, which considers a continuous range of states. The combination of the two methods provides efficient means of distinguishing the models. The first method has a more direct ecological meaning, while the second is useful when forest data are limited in time.

UR - http://www.scopus.com/inward/record.url?scp=33847402302&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33847402302&partnerID=8YFLogxK

U2 - 10.1007/s11284-006-0016-x

DO - 10.1007/s11284-006-0016-x

M3 - Review article

AN - SCOPUS:33847402302

VL - 22

SP - 197

EP - 203

JO - Ecological Research

JF - Ecological Research

SN - 0912-3814

IS - 2

ER -