Vegetation Indices Derived from Spectral Bands for Monitoring Environmental Restoration Areas Using the Parrot Sequoia

ABSTRACT

Abstract

Vegetation Indices Derived from Spectral Bands for Monitoring Environmental Restoration Areas Using the Parrot Sequoia

ABSTRACT

Remote sensing performed by sensors attached to Remotely Piloted Aircrafts (RPAs) represents an effective alternative compared to orbital satellites. This study aimed to analyze, among the vegetation indices NDVI, NDRE, MCARI, GNDVI, and SIPI, obtained through the Parrot Sequoia multispectral sensor, which best represented the evolution of vegetation cover in a Permanent Preservation Area undergoing forest restoration over a period of one and a half years (April 2021 to October 2022). The RPA used was the DJI Phantom 4 Pro, equipped with a Parrot Sequoia multispectral scanning system. Image processing was carried out using Pix4Dmapper software, from which vegetation indices (VIs) were generated. Although statistical analyses did not indicate significant temporal differences, some indices better illustrated the dynamics of transformations in the area. NDVI, for instance, allowed the identification of typical Cerrado vegetation behavior, with values close to 1 in both analyzed periods, indicating areas with high vegetative activity. On the other hand, SIPI and NDRE indices showed saturation in the images, making it difficult to distinguish vegetation from exposed soil. The vegetation indicators proved useful for visually detecting variations in soil cover and changes in vegetation structure due to its growth.