TY - JOUR

T1 - Using Principal Component Analysis for information-rich socio-ecological vulnerability mapping in Southern Africa

AU - Abson, David

AU - Dougill, Andrew

AU - Stringer, Lindsay

PY - 2012/11/20

Y1 - 2012/11/20

N2 - Aggregate measures that capture multiple aspects of socio-ecological vulnerability in a single or small number of vulnerability indices can produce vulnerability maps that act as powerful visual tools to identify those areas most susceptible to future environmental changes. Such indices are easily communicable and offer valuable guidance to policymakers and investors, providing insights as to where more targeted research or policy interventions can address current challenges and reduce future risks. However, such aggregation inevitably reduces the richness of information provided by the suites of individual vulnerability indicators on which the maps are based. This trade-off between information richness and information communicability is a challenge in the quantification and communication of complex phenomena such as socio-ecological vulnerability. This paper investigates the use of Principal Component Analysis (PCA) techniques as a means of creating information-rich spatially-explicit aggregate indices of socio-ecological vulnerability.We present a 'proof of concept' analysis of socio-ecological vulnerability for the Southern Africa Development Community (SADC) region using both PCA and traditional normalization based techniques for generating spatially explicit, aggregated socio-ecological vulnerability indices. The vulnerability indices are based on published biophysical and socio-economic data and mapped at a 10 arc minute resolution. The resulting PCA based vulnerability maps indicate the regional spatial variability of four statistically independent, unique components of socio-ecological vulnerability, providing more information than the single index produced using a normalization/summation approach. Such uncorrelated, information-rich vulnerability indices represent a potentially useful policy tool for identifying areas of greatest concern in terms of both the relative level, and the underlying causes and impacts of, socio-ecological vulnerability to environmental changes across broad spatial scales.

AB - Aggregate measures that capture multiple aspects of socio-ecological vulnerability in a single or small number of vulnerability indices can produce vulnerability maps that act as powerful visual tools to identify those areas most susceptible to future environmental changes. Such indices are easily communicable and offer valuable guidance to policymakers and investors, providing insights as to where more targeted research or policy interventions can address current challenges and reduce future risks. However, such aggregation inevitably reduces the richness of information provided by the suites of individual vulnerability indicators on which the maps are based. This trade-off between information richness and information communicability is a challenge in the quantification and communication of complex phenomena such as socio-ecological vulnerability. This paper investigates the use of Principal Component Analysis (PCA) techniques as a means of creating information-rich spatially-explicit aggregate indices of socio-ecological vulnerability.We present a 'proof of concept' analysis of socio-ecological vulnerability for the Southern Africa Development Community (SADC) region using both PCA and traditional normalization based techniques for generating spatially explicit, aggregated socio-ecological vulnerability indices. The vulnerability indices are based on published biophysical and socio-economic data and mapped at a 10 arc minute resolution. The resulting PCA based vulnerability maps indicate the regional spatial variability of four statistically independent, unique components of socio-ecological vulnerability, providing more information than the single index produced using a normalization/summation approach. Such uncorrelated, information-rich vulnerability indices represent a potentially useful policy tool for identifying areas of greatest concern in terms of both the relative level, and the underlying causes and impacts of, socio-ecological vulnerability to environmental changes across broad spatial scales.

KW - Environmental informatics

KW - Climate change

KW - Mapping

KW - PCA

KW - SADC

KW - Trade-offs

KW - Vulnerability indices

KW - Geography

KW - Environmental planning

KW - Sustainability Science

KW - Sustainability sciences, Management & Economics

UR - http://www.scopus.com/inward/record.url?scp=84870250975&partnerID=8YFLogxK

U2 - 10.1016/j.apgeog.2012.08.004

DO - 10.1016/j.apgeog.2012.08.004

M3 - Journal articles

VL - 35

SP - 515

EP - 524

JO - Applied Geography

JF - Applied Geography

SN - 0143-6228

IS - 1-2

ER -