AUTHOR=Medeiros Thaís Pereira de , Morellato Leonor Patrícia Cerdeira , Silva Thiago Sanna Freire 

TITLE=Spatial distribution and temporal variation of tropical mountaintop vegetation through images obtained by drones

JOURNAL=Frontiers in Environmental Science

VOLUME=Volume 11 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/environmental-science/articles/10.3389/fenvs.2023.1083328

DOI=10.3389/fenvs.2023.1083328

ISSN=2296-665X

ABSTRACT=Modern UAS (Unmanned Aerial Vehicles) or just drones have emerged with the primary goal of producing maps and imagery with extremely high spatial resolution. The refined information provides a good opportunity to quantify the distribution of vegetation across heterogeneous landscapes, revealing an essential strategy for biodiversity conservation. Some approaches have been applied to produce a consistent classification result, e.g., the use of the Object-Based Image Analysis (OBIA) and Random Forests algorithm. We investigate whether the above techniques are efficient to process the ultrahigh-resolution data generated by UAS images and whether using imagery from different temporal times improves the performance and results of the proposed vegetation descriptors. We focus our fieldwork on a highly diverse, seasonally dry, complex mountaintop vegetation system, the campo rupestre or rupestrian grassland, located at Serra do Cipó, Southeastern Brazil. According to our results, the use of the Random Forests classifier reduced the misclassified pixels and the stack with two dates was important for identifying vegetation types phenophases. The color change identified in the images represented the phenological status of the vegetation and indicated the leafing dynamics. Therefore, ultrahigh spatial resolution UAS images provided valuable data for the classification of complex vegetation systems such as the campo rupestre.