Forest cover types map and vegetation classification made using remote and terrestrial methods (case study for Valuevo forest park, Moscow)статьяИсследовательская статья
Аннотация:Mapping forest cover types in temperate broadleaf and mixed forests using remote methods has proven to be problematic. One key challenge is understanding how to deal with species diversity and spatial variability to create accurate and precise maps, as satellite imagery can only provide data about canopies. Image processing techniques such as supervised classification allows to increase the accuracy of mapping forest cover types. Nonetheless, survey data proved to be essential for creating training areas. Moreover, field surveys of specific sites containing species lists of the entire community makes it possible to infer not only about forest cover types, but also about the types of vegetation communities. In this paper, both terrestrial and remote sensing methods were combined to create a forest cover type map and vegetation classification. Furthermore, multiple types of 2006 and 2016 data were used to infer changes in vegetation structure. The study site is the Valuevo forest park (Moscow, Russia). An ecological-dynamic approach was used to create the classification of vegetation. Then for each forest cover type shown on the map, a list of different communities is shown in the classification. An ecological-dynamic approach considers each community as a particular succession stage of a climax community. Notably, an extent of monocultures is also shown both on the map and in the classification.