Mapping experience on the forest typological diversity in the Moscow Region
Geobotanical mapping · 2024
DOI: 10.31111/geobotmap/2024.25 ↗Kurz gefasst
Fachartikel in der Zeitschrift Geobotanical mapping, 2024. Verfasst von I. P. Kotlov Kotlov und 2 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 46 Literaturangaben.
Abstract (Originalsprache)
Quantitative assessment of forest area and forest structure is crucial for monitoring and landscape planning in urban regions. Despite the long history of forest research in the Moscow Region, there is no unified basis for one-step assessment of the background condition and monitoring of forest biodiversity. One of the most promising aspects of biodiversity is the typological diversity of forests, assessed within the framework of the ecological-phytocoenotic approach. At the same time, the most effective way to assess the spatial structure of typological diversity is through the combination of remote sensing data, field measurements and cartographic modelling. This study proposes the integration of ecological and phytocoenotic approaches together with the use of remote sensing for transformed forest area. The aim of the work is to assess the background condition of forests in the Moscow Region to create a basis for regular monitoring of natural resource and ecological potential. Field geobotanical data collection was carried out in 2006–2019 in forest areas without obvious traces of logging and other anthropogenic disturbances. The localization of points is related to the coverage of habitats, heterogeneous in terms of vegetation composition and genesis of underlying rocks, by ground surveys. Sample plots were made according to the standard methodology on the area of 400–625 m2 and 100 m2 when describing plantations. The total number of sample plots is 1694. Full species composition and vertical structure were recorded at sample plots. Species coverage is given in percent. The remote sensing data used was a completely cloud-free multispectral Sentinel-2 survey taken over two days (20 and 23 June 2021) and merged into a seamless mosaic (Fig. 2). In addition to 11 spectral channels, 41 spectral indices were calculated, including indices assessed as sensitive to vegetation stress (Abdullah et al., 2019). SRTM digital elevation model and 10 morphometric features calculated from it were used to improve modelling accuracy. Palsar-2 radar data (Shimada et al., 2014) were obtained also. A total of 63 raster layers were got. UTM coordinate system, zone 37 north, WGS datum 84 was used. To remove the multicollinearity of the rasters, the method of removing highly correlated layers by the cut-off boundary of 0.5 (Graham, 2003) was applied. As a result, 7 raster layers that have pairwise correlations not exceeding 0.5 were left: the blue (2) and red edge (6) channels, NDWI2, BNDWI, GLI, elevation, and HH coordinated polarization (Table 2). The random forest method was applied to model the spatial structure of forests (Breiman, 2001; Gislason et al., 2004; Grabska et al., 2020). This algorithm requires no special data preparation (variance equalization, distribution transformation, etc.) and is sensitive to correlation between variables (Brosofske et al., 2014). OrfeoToolbox software (Inglada, Christophe, 2009) was used, with a test sample size of 30 %. The resulting maps were synchronized with the Global Forest Watch Forest mask (Hansen et al., 2013). Layers of farmland, water bodies and settlements were prepared using OpenStreetMap data (Haklay, Weber, 2008). The overall convergence across association groups was 59 %. The convergence of association groups varies from 13% to 100 (Table 3). The lowest convergence is observed for the following association groups 7 — 13 %, 4 — 18 %, 8 — 21 %; 3 — 23 %, 5 — 30 %, 31 — 36 %. The spatial distribution of forests corresponds to both small-scale sources, such as the map of forest zones, and large-scale sources, such as the vegetation map of the Moscow region. Analysis of the vegetation composition of spruce and spruce-small-leaved forests demonstrates a transition from boreal to nemoral species composition. At the landscape level, sub-latitudinal differentiation of spruce and spruce-small-leaved communities is observed within the region. Most of the forest cover (30–40 %) of spruce and spruce-small-leaved communities was observed in the Smolensk and Moscow physiographic provinces (north and north-west of the region), to a lesser extent — in the eastern and southern parts of the Moscow Region. The minimum projective cover of spruce forests (less than 10 %) is recorded in the Zaokskaya and Srednerusskaya provinces, where there is a transition to the zone of broad-leaved forests. Mixed stand composition of trees can also be formed as a result of active birch regeneration in spruce crops with inadequate quality of care. Spruce and birch (with a small admixture of aspen) are present in equal proportions in spruce-small-leaved communities, while spruce forests are dominated by spruce with a small admixture of small-leaved tree species (Fig. 3).
Zitieren
APA
Kotlov, I. P. K., Chernenkova, T. V., & Belyaeva, N. G. (2024). Mapping experience on the forest typological diversity in the Moscow Region. Geobotanical mapping. https://doi.org/10.31111/geobotmap/2024.25
Vancouver
Kotlov IPK, Chernenkova TV, Belyaeva NG. Mapping experience on the forest typological diversity in the Moscow Region. Geobotanical mapping. 2024. doi:10.31111/geobotmap/2024.25
BibTeX
@article{Kotlov2024,
author = {Kotlov, I. P. Kotlov and Chernenkova, T. V. and Belyaeva, N. G.},
title = {Mapping experience on the forest typological diversity in the Moscow Region},
journal = {Geobotanical mapping},
year = {2024},
doi = {10.31111/geobotmap/2024.25}
}
EUVIMED