Local indicators for categorical data: Impacts of scaling decisions

Jed A. Long, Trisalyn Nelson, Michael A. Wulder

Research output: Contribution to specialist publicationArticle

12 Citations (Scopus)

Abstract

When the geographic distribution of landscape pattern varies, global indices fail to capture the spatial nonstationarity within the dataset. Methods that measure landscape pattern at a spatially local scale are advantageous, as an index is computed at each point in the dataset. The geographic distribution of local indices is used to discover spatial trends. Local indicators for categorical data (LICD) can be used to statistically quantify local spatial patterns in binary geographic datasets. LICD, like other spatially local methods, are impacted by decisions relating to the spatial scale of the data, such as the data grain (p), and analysis parameters such as the size of the local neighbourhood (m). The goal of this article is to demonstrate how the choice of the m and p parameters impacts LICD analysis. We also briefly discuss the impacts spatial extent can have on analysis; specifically the local composition measure. An example using 2006 forest cover data for a region in British Columbia, Canada where mountain pine beetle mitigation and salvage harvesting has occurred is used to show the impacts of changing m and p. Selection of local window size (m = 3,5,7) impacts the prevalence and interpretation of significant results. Increasing data grain (p) had varying effects on significant LICD results. When implementing LICD the choice of m and p impacts results. Exploring multiple combinations of m and p will provide insight into selection of ideal parameters for analysis.

Original languageEnglish (US)
Pages15-28
Number of pages14
Volume54
No1
Specialist publicationCanadian Geographer
DOIs
StatePublished - 2010
Externally publishedYes

Fingerprint

scaling
agricultural product
indicator
decision
forest cover
beetle
data analysis
mitigation
Canada
interpretation
mountain
trend
index
analysis
parameter

Keywords

  • Composition
  • Configuration
  • Dendroctonus ponderosae
  • Fragmentation
  • Mountain pine beetle
  • Spatial analysis
  • Spatial pattern

ASJC Scopus subject areas

  • Geography, Planning and Development
  • Earth-Surface Processes

Cite this

Local indicators for categorical data : Impacts of scaling decisions. / Long, Jed A.; Nelson, Trisalyn; Wulder, Michael A.

In: Canadian Geographer, Vol. 54, No. 1, 2010, p. 15-28.

Research output: Contribution to specialist publicationArticle

Long, Jed A. ; Nelson, Trisalyn ; Wulder, Michael A. / Local indicators for categorical data : Impacts of scaling decisions. In: Canadian Geographer. 2010 ; Vol. 54, No. 1. pp. 15-28.
@misc{215cafbb815f4a809ef39839daf7eb88,
title = "Local indicators for categorical data: Impacts of scaling decisions",
abstract = "When the geographic distribution of landscape pattern varies, global indices fail to capture the spatial nonstationarity within the dataset. Methods that measure landscape pattern at a spatially local scale are advantageous, as an index is computed at each point in the dataset. The geographic distribution of local indices is used to discover spatial trends. Local indicators for categorical data (LICD) can be used to statistically quantify local spatial patterns in binary geographic datasets. LICD, like other spatially local methods, are impacted by decisions relating to the spatial scale of the data, such as the data grain (p), and analysis parameters such as the size of the local neighbourhood (m). The goal of this article is to demonstrate how the choice of the m and p parameters impacts LICD analysis. We also briefly discuss the impacts spatial extent can have on analysis; specifically the local composition measure. An example using 2006 forest cover data for a region in British Columbia, Canada where mountain pine beetle mitigation and salvage harvesting has occurred is used to show the impacts of changing m and p. Selection of local window size (m = 3,5,7) impacts the prevalence and interpretation of significant results. Increasing data grain (p) had varying effects on significant LICD results. When implementing LICD the choice of m and p impacts results. Exploring multiple combinations of m and p will provide insight into selection of ideal parameters for analysis.",
keywords = "Composition, Configuration, Dendroctonus ponderosae, Fragmentation, Mountain pine beetle, Spatial analysis, Spatial pattern",
author = "Long, {Jed A.} and Trisalyn Nelson and Wulder, {Michael A.}",
year = "2010",
doi = "10.1111/j.1541-0064.2009.00300.x",
language = "English (US)",
volume = "54",
pages = "15--28",
journal = "Canadian Geographer / Geographie Canadien",
issn = "0008-3658",
publisher = "Wiley-Blackwell",

}

TY - GEN

T1 - Local indicators for categorical data

T2 - Impacts of scaling decisions

AU - Long, Jed A.

AU - Nelson, Trisalyn

AU - Wulder, Michael A.

PY - 2010

Y1 - 2010

N2 - When the geographic distribution of landscape pattern varies, global indices fail to capture the spatial nonstationarity within the dataset. Methods that measure landscape pattern at a spatially local scale are advantageous, as an index is computed at each point in the dataset. The geographic distribution of local indices is used to discover spatial trends. Local indicators for categorical data (LICD) can be used to statistically quantify local spatial patterns in binary geographic datasets. LICD, like other spatially local methods, are impacted by decisions relating to the spatial scale of the data, such as the data grain (p), and analysis parameters such as the size of the local neighbourhood (m). The goal of this article is to demonstrate how the choice of the m and p parameters impacts LICD analysis. We also briefly discuss the impacts spatial extent can have on analysis; specifically the local composition measure. An example using 2006 forest cover data for a region in British Columbia, Canada where mountain pine beetle mitigation and salvage harvesting has occurred is used to show the impacts of changing m and p. Selection of local window size (m = 3,5,7) impacts the prevalence and interpretation of significant results. Increasing data grain (p) had varying effects on significant LICD results. When implementing LICD the choice of m and p impacts results. Exploring multiple combinations of m and p will provide insight into selection of ideal parameters for analysis.

AB - When the geographic distribution of landscape pattern varies, global indices fail to capture the spatial nonstationarity within the dataset. Methods that measure landscape pattern at a spatially local scale are advantageous, as an index is computed at each point in the dataset. The geographic distribution of local indices is used to discover spatial trends. Local indicators for categorical data (LICD) can be used to statistically quantify local spatial patterns in binary geographic datasets. LICD, like other spatially local methods, are impacted by decisions relating to the spatial scale of the data, such as the data grain (p), and analysis parameters such as the size of the local neighbourhood (m). The goal of this article is to demonstrate how the choice of the m and p parameters impacts LICD analysis. We also briefly discuss the impacts spatial extent can have on analysis; specifically the local composition measure. An example using 2006 forest cover data for a region in British Columbia, Canada where mountain pine beetle mitigation and salvage harvesting has occurred is used to show the impacts of changing m and p. Selection of local window size (m = 3,5,7) impacts the prevalence and interpretation of significant results. Increasing data grain (p) had varying effects on significant LICD results. When implementing LICD the choice of m and p impacts results. Exploring multiple combinations of m and p will provide insight into selection of ideal parameters for analysis.

KW - Composition

KW - Configuration

KW - Dendroctonus ponderosae

KW - Fragmentation

KW - Mountain pine beetle

KW - Spatial analysis

KW - Spatial pattern

UR - http://www.scopus.com/inward/record.url?scp=77749279908&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77749279908&partnerID=8YFLogxK

U2 - 10.1111/j.1541-0064.2009.00300.x

DO - 10.1111/j.1541-0064.2009.00300.x

M3 - Article

AN - SCOPUS:77749279908

VL - 54

SP - 15

EP - 28

JO - Canadian Geographer / Geographie Canadien

JF - Canadian Geographer / Geographie Canadien

SN - 0008-3658

ER -