The use of a geodatabase to carry out a multivariate analysis of coastline variations at various time and space scales

Abstract : This study proposes a method, based on the use of spatial data and aerial photographs to analyse long term evolution of the coastline along the Gâvres - Penthièvre beach system extending over more than 50 km (Southern Brittany; France). Spatial data layers include: air photograph measurements of coastline position through time, sediment cover, coastline morphology and bathymetry. These were combined with current data derived from both field and modelled data to digitise the limits of sediment cells. Aerial photographs analysis also provides data about the location and the density of anthropogenic features established along the coastline through time. All these data were compiled in a personal geodatabase and a series of both spatial and geostatistical analyses were carried out to assess their spatial relationships. Spatial analyses aim at identifying spatial associations between the type of site evolution (erosion polygon vs accretion polygon) and the characteristics of the coastline (type of sediment cell sites / low density of human pressures / high density of human pressures). Statistical analyses aim at assessing the statistical dependency between variables defining the spatial associations. The results enable identification of at least two different dynamics controlling coastline movements. At a regional scale coastal dynamics are mainly related to incident wave propagation. Over more local areas, interrelationships between coastal dynamics and increasing concentration of anthropogenic features partly explain coastline variations. Through these results, this study highlights benefits associated with GIS facilities to analyse the role of various factors on coastal dynamics control over a regional scale.