Encouraging greater levels of physical activity is a key policy focus in many urban areas, given the link between this and improved physical and mental health outcomes. The majority of previous studies on walkability however, have focussed on walking for transport; and generally consisted of creating desktop based walkability indices using variables such as residential density, street connectivity, and land use mix. This paper reports on research which investigated a lesser studied aspect of walkability; walking for recreation, in the Gold Coast City Council, Queensland, Australia. A Geographical Information Systems (GIS)-based recreation walkability layer was created, and used to define the relative walkability of the Gold Coast. Unlike many such indices, this was based on raster rather than vector data and used different variables, distances, and weightings than the majority of walking for transport indices. The distances were based on the minimum recommended distance of 30 minutes exercise per day, and the variables, such as parks, dog parks, proximity to water, and fitness stations, were chosen and weighted according to their appeal for both routes and destinations. Other variables used as cost factors in Network Analysis, included slope, road parameters (traffic volume, speed, footpaths and cycleways). The vector datasets were a small scale (SA1) and rasterized to 5m square cells. The results of this indicated that as expected, the denser coastal areas were more walkable than the steeper, lesser populated hinterland. Moreover, the raster surface was superior to the vector scores (SA1 level) with a far greater level of detail, and the 5m raster cell size could be aggregated to any size spatial unit. Further research aims to groundtruth this index.