UAV-Based Digital Terrain Model Generation under Leaf-Off Conditions to Support Teak Plantations Inventories in Tropical Dry Forests. A Case of the Coastal Region of Ecuador

Fernando J Aguilar; José R Rivas; Abderrahim Nemmaoui; Alberto Peñalver; Manuel A Aguilar

doi:10.3390/s19081934

UAV-Based Digital Terrain Model Generation under Leaf-Off Conditions to Support Teak Plantations Inventories in Tropical Dry Forests. A Case of the Coastal Region of Ecuador

Sensors (Basel). 2019 Apr 25;19(8):1934. doi: 10.3390/s19081934.

Authors

Fernando J Aguilar¹, José R Rivas², Abderrahim Nemmaoui³, Alberto Peñalver⁴, Manuel A Aguilar⁵

Affiliations

¹ Department of Engineering, University of Almería, Ctra. de Sacramento s/n, La Cañada de San Urbano, 04120 Almería, Spain. faguilar@ual.es.
² Faculty of Technical Education for Development, Santiago de Guayaquil Catholic University, Av. Carlos Julio Arosemena, Guayaquil 090615, Ecuador. jose.rivas02@cu.ucsg.edu.ec.
³ Department of Engineering, University of Almería, Ctra. de Sacramento s/n, La Cañada de San Urbano, 04120 Almería, Spain. an932@ual.es.
⁴ Faculty of Technical Education for Development, Santiago de Guayaquil Catholic University, Av. Carlos Julio Arosemena, Guayaquil 090615, Ecuador. alberto.penalver01@cu.ucsg.edu.ec.
⁵ Department of Engineering, University of Almería, Ctra. de Sacramento s/n, La Cañada de San Urbano, 04120 Almería, Spain. maguilar@ual.es.

Abstract

Remote sensing is revolutionizing the way in which forests studies are conducted, and recent technological advances, such as Structure from Motion (SfM) photogrammetry from Unmanned Aerial Vehicle (UAV), are providing more efficient methods to assist in REDD (Reducing Emissions from Deforestation and forest Degradation) monitoring and forest sustainable management. The aim of this work was to develop and test a methodology based on SfM from UAV to generate high quality Digital Terrain Models (DTMs) on teak plantations (Tectona grandis Linn. F.) situated in the Coastal Region of Ecuador (dry tropical forest). UAV overlapping images were collected using a DJI Phantom 4 Advanced^© quadcopter during the dry season (leaf-off phenological stage) over 58 teak square plots of 36 m side belonging to three different plantations located in the province of Guayas (Ecuador). A workflow consisting of SfM absolute image alignment based on field surveyed ground control points, very dense point cloud generation, ground points filtering and outlier removal, and DTM interpolation from labeled ground points, was accomplished. A very accurate Terrestrial Laser Scanning (TLS) derived ground points were employed as ground reference to estimate the UAV-SfM DTM vertical error in each reference plot. The plot-level obtained DTMs presented low vertical bias and random error (-3.1 cm and 11.9 cm on average, respectively), showing statistically significant greater error in those reference plots with basal area and estimated vegetation coverage above 15 m²/ha and 60%, respectively. To the best of the authors' knowledge, this is the first study aimed at monitoring of teak plantations located in dry tropical forests from UAV images. It provides valuable information that recommends carrying out the UAV image capture during the leaf-off season to obtain UAV-SfM derived DTMs suitable to serve as ground reference in supporting teak plantations inventories.

Keywords: UAV; digital terrain model; forest inventory; remote sensing; structure from motion; teak plantations; tropical dry forest.

MeSH terms

Conservation of Natural Resources
Ecuador
Environmental Monitoring / methods*
Forests*
Remote Sensing Technology / methods*

Grants and funding

no grant number/Research and Development System of the Catholic University of Santiago de Guayaquil, Ecuador