The following report will enlighten to which extent LiDAR data could enhance land cover classification. It focuses on the area Lemps (26510) in Southwest France where a land cover classification was made using Sentinel-2 spectral images during the fieldwork. Using the additional LiDAR data, the focus shifts to distinguish coniferous and deciduous trees. Training data from the LiDAR data has been selected in CloudCompare. Using the training data, features for coniferous and deciduous trees were extracted in python. The unique features were used as classifiers. Features based on the Intensity were found to be important. Based on the classifiers, two methods were used to classify the area. Random Forest and Nearest Neighbour were the classification methods. The classification using Random Forest was found to be more accurate. The Random Forest classification map has been compared with previously acquired Sentinel-2 classification maps. The Corine Land Cover Classification and the classification map from the fieldwork were compared to the classification of coniferous and deciduous trees using LiDAR. Lots of overlap was found with the Corine Land Cover, some overlap was present with the map acquired during the fieldwork. The map created during the fieldwork contained less training data, hence the model was not trained enough. If more training data is collected for both LiDAR and Sentinel- 2 classifications, LiDAR data could enhance the general land cover classification. Especially taking the intensity into account as a classifier.