| Popis: |
In the central semiarid region of Argentina, the alien woody invasive species modify the water cycle, reduce biodiversity and alter the nutrient cycles. One of the most important alien woody invasive species is the glossy privet (Ligustrum lucidum W.T. Aiton) with an agressive invasive ability. The objective of this research is to evaluate the NDVI and SAR interferometric coherence time series to discriminate L. lucidum within the native forest matrix in Central Cordoba, Argentina. The time series of interferometric coherence in VH and in VV polarizations and NDVI were plotted and analyzed statistically. Pixel-based classifications with the Random Forest classifier were made. The results show that L. lucidum invaded areas have statistically different differences with the native forest matrix when NDVI and interferometric coherence of C band SAR images, with polarizations VV and VH are analyzed. Also, the Random Forest classifier applied individually on the time series or the combination of them allowed the differentiation between invaded and uninvaded areas. More research should be conducted to evaluate the possibility to integrate temporal series of backscattering, L-band and X-band SAR sensors data, and to test other classifiers. |
