Advancing Biodiversity Exploration and Habitat Monitoring: Utilizing the NaturaSat Software for ecosystems from temperate lowland wetlands to Arctic tundra vegetation.

Remote sensing plays a crucial role in ecology and nature conservation by allowing for effective monitoring of spatio-temporal changes in ecosystems. The NaturaSat software, developed by a multidisciplinary team of botany fieldwork scientists, nature protection managers, mathematicians, and software developers (Mikula et al., 2021), is a recent tool that focuses on vegetation exploration. This software offers solutions to challenging questions, such as accurately estimating the areas and boundaries of Natura 2000 habitats and tracking their spatio-temporal changes using an evolving curves approach. It also enables the monitoring of biodiversity and habitat quality through the use of a graph-Laplacian function. Additionally, the software utilizes a novel deep learning method called the Natural Numerical Network to classify habitats on the most detailed scale represented by vegetation units.

 

Within this talk, we will present how NaturaSat software has been extensively tested in various habitats, ranging from temperate lowland wetlands along the Danube River to riparian forests, broadleaved deciduous forests, ancient montane woodlands of the Carpathian Mountains, and even the arctic tundra vegetation in the Alaska region. The results have demonstrated the software's capability to accurately identify habitat borders and detect shifts in these borders due to changes in water regimes or climate. Furthermore, it has proven effective in distinguishing between species-rich natural forests and planted forests dominated by the same tree species. The software also enables the classification of habitats and the automatic detection of new habitats in previously undiscovered areas. In conclusion, the NaturaSat software complements detailed ground-based approaches in biodiversity exploration and habitat monitoring. The presented methods can be repeated over long time periods, ensuring temporal consistency, and they offer a cost-effective means of identifying trends in vegetation changes and biodiversity.