Territorial cohesion impacts of high-speed rail under different zoning systems

Abstract

Large-scale transport infrastructure projects such as high-speed rail (HSR) produce significant effects on
the spatial distribution of accessibility. These effects, commonly known as territorial cohesion effects, are
receiving increasing attention in the research literature. However, there is little empirical research into
the sensitivity of these cohesion results to methodological issues such as the definition of the limits of
the study area or the zoning system. In a previous paper (Ortega et al., 2012), we investigated the influence
of scale issues, comparing the cohesion results obtained at four different planning levels. This paper
makes an additional contribution to our research with the investigation of the influence of zoning issues.
We analyze the extent to which changes in the size of the units of analysis influence the measurement of
spatial inequalities.
The methodology is tested by application to the Galician (north-western) HSR corridor, with a length of
nearly 670 km, included in the Spanish PEIT (Strategic Transport and Infrastructure Plan) 2005-2020. We
calculated the accessibility indicators for the Galician HSR corridor and assessed their corresponding territorial
distribution. We used five alternative zoning systems depending on the method of data representation
used (vector or raster), and the level of detail (cartographic accuracy or cell size). Our results
suggest that the choice between a vector-based and raster-based system has important implications.
The vector system produces a higher mean accessibility value and a more polarized accessibility distribution
than raster systems. The increased pixel size of raster-based systems tends to give rise to higher
mean accessibility values and a more balanced accessibility distribution. Our findings strongly encourage
spatial analysts to acknowledge that the results of their analyses may vary widely according to the definition
of the units of analysis.