Abstract

ESTRADA ZUNIGA, Andrés C.  and  NAUPARI VASQUEZ, Javier. Detection and identification of high Andean plant communities, Wetlands and Tolar de Puna Seca by means of RGB and NDVI orthophotos in “Unmanned Aerial Systems” drones. Scientia Agropecuaria [online]. 2021, vol.12, n.3, pp.291-301. ISSN 2077-9917.  http://dx.doi.org/10.17268/sci.agropecu.2021.032.

Remote sensing and geographic information systems are tools that in the last decade have been widely used in the management of natural resources, however, they have presented deficiencies for precision livestock studies due to the quality of spatial resolutions, spectral and temporal. Faced with this limitation, microsensors appear as an alternative in Unmanned Aerial Systems (UAS) that allow obtaining orthophotographs with better resolutions. Considering these advantages, a study was developed to determine the best flight height in the detection and identification of the tolar and bofedal plant communities of the dry puna. For the study, RGB and NDVI photographs were collected with ZENMUSE X3 DJI RGB-NDVI sensors in UAS with flight heights of 25, 50, 75 and 100 m. In the field, tola plants and DIMU cushions were counted in quadrants of 10 m x 10 m (100 m2). The preparation of orthophotographs was carried out in the Pix 4D software and to analyze the information an algorithm was developed with the ability to identify a segmented element (tola Plant and / or DIMU cushion) using Python. The study found that the NDVI range for the identification of tolares of Parastrephia lepidophilla is from 0.20 to 0.45 and for Distichia muscoides bogs is from 0.68 to 0.95; finally, using RGB and NDVI orthophotographs, it was determined that the best flight height to identify the Tola and DIMU segmented species is 25 m followed by 50 m.

Keywords : Plant community; Remote sensors; Tolar; Bofedal; UAS; UAV.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )