The paper describes processing methods for portraying forest areas on maps utilizing point symbols. The forest map design is based on the use of individual tree data, which is detected from airborne laser scanning data and colour-infrared orthophotos. Several generalisation and symbolisation approaches have been tested in order to use the tree data set for cartographic purposes. Our generalisation method differentiates between trees in dense forests and tree structures in open areas, such as isolated trees, tree groups, tree rows and sparsely scattered trees. The tree symbols are integrated visually with the other map content. The results show that the production of attractive and useful maps requires an advanced generalisation method due to the massive amount of tree objects extracted from the laser scanning data as well as a generalisation level and a symbol type for the trees that are carefully chosen according to the map scale. Furthermore, sophisticated visualisation methods are needed for composing the maps. The created maps are part of a project that aims at supporting outdoor activities using multi-scale maps as part of a multi-publishing service.