Spatial autocorrelation in multiscale land use models

K.P. Overmars, G.H.J. de Koning, A. Veldkamp

Research output: Contribution to journalArticleAcademicpeer-review

251 Citations (Scopus)

Abstract

Various modelling approaches exist for the simulation and exploration of land use change. Until recently often ordinary statistics were used in studies dealing with spatial data, although several techniques are available to deal with spatial autocorrelation. This article presents the spatial autocorrelation technique
In several land use models statistical methods are being used to analyse spatial data. Land use drivers that best describe land use patterns quantitatively are often selected through (logistic) regression analysis. A problem using conventional statistical methods, like (logistic) regression, in spatial land use analysis is that these methods assume the data to be statistically independent. But, spatial land use data have the tendency to be dependent, a phenomenon known as spatial autocorrelation. Values over distance are more similar or less similar than expected for randomly associated pairs of observations. In this paper correlograms of the Moran's I are used to describe spatial autocorrelation for a data set of Ecuador. Positive spatial autocorrelation was detected in both dependent and independent variables, and it is shown that the occurrence of spatial autocorrelation is highly dependent on the aggregation level. The residuals of the original regression model also show positive autocorrelation, which indicates that the standard multiple linear regression model cannot capture all spatial dependency in the land use data. To overcome this, mixed regressive-spatial autoregressive models, which incorporate both regression and spatial autocorrelation, were constructed. These models yield residuals without spatial autocorrelation and have a better goodness-of-fit. The mixed regressive-spatial autoregressive model is statistically sound in the presence of spatially dependent data, in contrast with the standard linear model which is not. By using spatial models a part of the variance is explained by neighbouring values. This is a way to incorporate spatial interactions that cannot be captured by the independent variables. These interactions are caused by unknown spatial processes such as social relations and market effects. (C) 2003 Elsevier Science B.V. All rights reserved.
Original languageEnglish
Pages (from-to)257-270
JournalEcological Modelling
Volume164
DOIs
Publication statusPublished - 2003

Fingerprint

autocorrelation
land use
spatial data
logistics
land use change
regression analysis
market
modeling
simulation
method

Keywords

  • land use
  • soil suitability
  • spatial variation
  • models
  • pattern

Cite this

Overmars, K. P., de Koning, G. H. J., & Veldkamp, A. (2003). Spatial autocorrelation in multiscale land use models. Ecological Modelling, 164, 257-270. https://doi.org/10.1016/S0304-3800(03)00070-X
Overmars, K.P. ; de Koning, G.H.J. ; Veldkamp, A. / Spatial autocorrelation in multiscale land use models. In: Ecological Modelling. 2003 ; Vol. 164. pp. 257-270.
@article{7953a831be114fdeb68a604b51470da3,
title = "Spatial autocorrelation in multiscale land use models",
abstract = "Various modelling approaches exist for the simulation and exploration of land use change. Until recently often ordinary statistics were used in studies dealing with spatial data, although several techniques are available to deal with spatial autocorrelation. This article presents the spatial autocorrelation techniqueIn several land use models statistical methods are being used to analyse spatial data. Land use drivers that best describe land use patterns quantitatively are often selected through (logistic) regression analysis. A problem using conventional statistical methods, like (logistic) regression, in spatial land use analysis is that these methods assume the data to be statistically independent. But, spatial land use data have the tendency to be dependent, a phenomenon known as spatial autocorrelation. Values over distance are more similar or less similar than expected for randomly associated pairs of observations. In this paper correlograms of the Moran's I are used to describe spatial autocorrelation for a data set of Ecuador. Positive spatial autocorrelation was detected in both dependent and independent variables, and it is shown that the occurrence of spatial autocorrelation is highly dependent on the aggregation level. The residuals of the original regression model also show positive autocorrelation, which indicates that the standard multiple linear regression model cannot capture all spatial dependency in the land use data. To overcome this, mixed regressive-spatial autoregressive models, which incorporate both regression and spatial autocorrelation, were constructed. These models yield residuals without spatial autocorrelation and have a better goodness-of-fit. The mixed regressive-spatial autoregressive model is statistically sound in the presence of spatially dependent data, in contrast with the standard linear model which is not. By using spatial models a part of the variance is explained by neighbouring values. This is a way to incorporate spatial interactions that cannot be captured by the independent variables. These interactions are caused by unknown spatial processes such as social relations and market effects. (C) 2003 Elsevier Science B.V. All rights reserved.",
keywords = "landgebruik, ruimtelijke variatie, modellen, bodemgeschiktheid, land use, soil suitability, spatial variation, models, pattern",
author = "K.P. Overmars and {de Koning}, G.H.J. and A. Veldkamp",
year = "2003",
doi = "10.1016/S0304-3800(03)00070-X",
language = "English",
volume = "164",
pages = "257--270",
journal = "Ecological Modelling",
issn = "0304-3800",
publisher = "Elsevier",

}

Spatial autocorrelation in multiscale land use models. / Overmars, K.P.; de Koning, G.H.J.; Veldkamp, A.

In: Ecological Modelling, Vol. 164, 2003, p. 257-270.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Spatial autocorrelation in multiscale land use models

AU - Overmars, K.P.

AU - de Koning, G.H.J.

AU - Veldkamp, A.

PY - 2003

Y1 - 2003

N2 - Various modelling approaches exist for the simulation and exploration of land use change. Until recently often ordinary statistics were used in studies dealing with spatial data, although several techniques are available to deal with spatial autocorrelation. This article presents the spatial autocorrelation techniqueIn several land use models statistical methods are being used to analyse spatial data. Land use drivers that best describe land use patterns quantitatively are often selected through (logistic) regression analysis. A problem using conventional statistical methods, like (logistic) regression, in spatial land use analysis is that these methods assume the data to be statistically independent. But, spatial land use data have the tendency to be dependent, a phenomenon known as spatial autocorrelation. Values over distance are more similar or less similar than expected for randomly associated pairs of observations. In this paper correlograms of the Moran's I are used to describe spatial autocorrelation for a data set of Ecuador. Positive spatial autocorrelation was detected in both dependent and independent variables, and it is shown that the occurrence of spatial autocorrelation is highly dependent on the aggregation level. The residuals of the original regression model also show positive autocorrelation, which indicates that the standard multiple linear regression model cannot capture all spatial dependency in the land use data. To overcome this, mixed regressive-spatial autoregressive models, which incorporate both regression and spatial autocorrelation, were constructed. These models yield residuals without spatial autocorrelation and have a better goodness-of-fit. The mixed regressive-spatial autoregressive model is statistically sound in the presence of spatially dependent data, in contrast with the standard linear model which is not. By using spatial models a part of the variance is explained by neighbouring values. This is a way to incorporate spatial interactions that cannot be captured by the independent variables. These interactions are caused by unknown spatial processes such as social relations and market effects. (C) 2003 Elsevier Science B.V. All rights reserved.

AB - Various modelling approaches exist for the simulation and exploration of land use change. Until recently often ordinary statistics were used in studies dealing with spatial data, although several techniques are available to deal with spatial autocorrelation. This article presents the spatial autocorrelation techniqueIn several land use models statistical methods are being used to analyse spatial data. Land use drivers that best describe land use patterns quantitatively are often selected through (logistic) regression analysis. A problem using conventional statistical methods, like (logistic) regression, in spatial land use analysis is that these methods assume the data to be statistically independent. But, spatial land use data have the tendency to be dependent, a phenomenon known as spatial autocorrelation. Values over distance are more similar or less similar than expected for randomly associated pairs of observations. In this paper correlograms of the Moran's I are used to describe spatial autocorrelation for a data set of Ecuador. Positive spatial autocorrelation was detected in both dependent and independent variables, and it is shown that the occurrence of spatial autocorrelation is highly dependent on the aggregation level. The residuals of the original regression model also show positive autocorrelation, which indicates that the standard multiple linear regression model cannot capture all spatial dependency in the land use data. To overcome this, mixed regressive-spatial autoregressive models, which incorporate both regression and spatial autocorrelation, were constructed. These models yield residuals without spatial autocorrelation and have a better goodness-of-fit. The mixed regressive-spatial autoregressive model is statistically sound in the presence of spatially dependent data, in contrast with the standard linear model which is not. By using spatial models a part of the variance is explained by neighbouring values. This is a way to incorporate spatial interactions that cannot be captured by the independent variables. These interactions are caused by unknown spatial processes such as social relations and market effects. (C) 2003 Elsevier Science B.V. All rights reserved.

KW - landgebruik

KW - ruimtelijke variatie

KW - modellen

KW - bodemgeschiktheid

KW - land use

KW - soil suitability

KW - spatial variation

KW - models

KW - pattern

U2 - 10.1016/S0304-3800(03)00070-X

DO - 10.1016/S0304-3800(03)00070-X

M3 - Article

VL - 164

SP - 257

EP - 270

JO - Ecological Modelling

JF - Ecological Modelling

SN - 0304-3800

ER -