|
|
Impact of forest understory on Sentinel-2 satellite data time series
Antošová, Kristýna ; Laštovička, Josef (advisor) ; Štych, Přemysl (referee)
In this paper, time series of the Sentinel-2 satellite mission were compared to observe different types of forest understory. The aim was to determine the influence of understory type and physical-geographical characteristics of the selected localities on the time series curves. During the field research, selected sites were precisely located using GNSS, therefore, their botanical and physical-geographical characteristics were explored. Subsequently, the spectral reflectance curves for individual localities at different times of the year were analysed in GEE, and time series of vegetation indices NDVI and NDMI, temperature and precipitation from 2017 to 2024 were also created. In the none-growing season, the spectral reflectance curves of single localities showed higher variability compared to the growing season. The most appropriate bands for observing understory appeared to be SWIR 1 and SWIR 2. The time series of vegetation indices were statistically tested to delineate differences within sites. In most cases, for the pairs of sites being compared based on the NDVI and NDMI time series the hypothesis of dissimilarity was confirmed. The NDMI index appeared to be more appropriate than NDVI for distinguishing between sites.
|
guest ::
RSS feed.