This paper presents the results of extended testing of a long-term mapping algorithm for mobile robots that was previously developed by the authors. The algorithm merges multiple 3D point clouds used for localization, based on when the point clouds were recorded, in order to create an updated map that more accurately represents the current environment. The algorithm was originally tested for only a few months on a Boston Dynamics Spot robot. This paper presents the results of extended testing of the algorithm over 16 months and 182.5 km of travel by having Spot autonomously complete an 837 m circuit, outdoors on the paths and sidewalks around the Ontario Tech University campus, including traversing multiple staircases. This additional testing further proves the ability of the long-term mapping algorithm to adapt its navigation maps to changing environments providing better localization over long periods of time.