Application of a spatial dataset for monitoring invasive woody plant species in the forests of Transcarpathia, Ukraine

The paper describes and provides examples of the application of the developed spatial dataset on the spread of invasive woody plant species in the forests of Transcarpathia, Ukraine. The dataset was developed based on forest inventory data using the open source QGIS program, which performed digitisation of forest plantation plans of permanent forest users and overlay analysis. The created dataset contains 4212 records of the distribution of the following invasive woody plant species: black locust (Robinia pseudoacacia L.), northern red oak (Quercus rubra L.), ash maple (Acer negundo L.), and brittle willow (Salix fragilis L.). The dataset contains information on the location of forest plots, silvicultural and taxonomic characteristics of stands with predominance and participation of invasive woody plant species. The created spatial dataset was used to study the distribution of invasive woody species of black locust and northern red oak in Transcarpathia by geobotanical and floristic zones. The results of the study showed that the most favourable conditions for the distribution of black locust are the area of beech-oak and oak-beech foothill forests of the Volcanic Range of the Ukrainian Carpathians within Transcarpathia, as well as the Transcarpathian Plain forest area. Unlike the artificially created northern red oak forests, the distribution of the black locust in the study area is primarily due to its biological properties, specifics of environmental relationships and of forestry in the past. Regarding the distribution of the northern red oak, the results of using our spatial dataset show that such forests have the largest portion in the areas of hornbeam-beech and beech forests (39.4%) and in the forests of the Volcanic Ridge (24.2%). The created spatial dataset can be used to study the distribution of invasive woody plant species within the forest cover of Transcarpathia. The spatial dataset can also be utilised as a source of training samples for machine learning, which is involved in the processing of satellite images to identify new habitats of invasive woody plant species.