Assessing tropical forest degradation and restoration through lidar remote sensing

Author(s)

Danilo Roberti Alves de Almeida

• Document type: Doctoral Thesis
• Press: Piracicaba.
• Institution: Universidade de São Paulo (USP). Escola Superior de Agricultura Luiz de Queiroz (ESALA/BC)
• Defense date: 2018-12-04

Examining board members

Pedro Henrique Santin Brancalion; Eric Bastos Görgens

• Advisor: Pedro Henrique Santin Brancalion; Juliana Schietti de Almeida

Abstract

The present study investigates new frontiers of lidar technology knowledge assessessing of tropical forest degradation and restoration. The thesis is structured with an introductory chapter, four technical chapters, which explored technical and scientific aspects of the application of lidar technology to the evaluation of forest degradation in the Amazon and restoration of forests in the Atlantic Forest, and a final chapter with considerations and a summary of the main scientific results obtained in this thesis. The results of chapter 2 provided important insights for the correct modeling of leaf area density (LAD) proflies profiles usind airborne lidar. LAD profile is the decomposition of the leaf area index (LAI) along the vertical canopy profile and can be used to answer many ecological questions. The results of chapter 2 provided important insights for the correct modeling of LAD profiles. Chapter 3, using lidar data on aerial and portable ground platforms, in the Biological Dynamics of Forest Fragments Project (BDFFP), demonstrated in an unprecedented way the effect of forest fragmentation in the canopy structure (lidar-derived) and their relationships with the change of the tree community. In this chapter, the results showed that the lidar technology has enormous potential to monitor the impact of forest fragmentation in a high precision scale for large areas. Chapter 4, using data from several forest typologies in the Atlantic Forest biome restoration collect by a portable ground lidar system, demonstrated the potential of canopy structural attributes to distinguish different forest typologies and to estimate above ground woody dry biomass. However, the results were not positive for estimating tree community diversity (richness, Shannon index and species composition). Finally, chapter 5 demonstrated the potential of a novel lidar system on a drone platform (also known as UAV - unmanned aerial vehicle) to monitor forest restoration plantations. Lidar is revolutionizing the way we measure forest landscapes and can be an indispensable tool for the success of forest restoration projects, having the potential to support on planning, monitoring and inspection of forest restoration landscapes. In this thesis, we demonstrate several applications of remote sensing to address the context of forest restoration, and we established methodological bases for other studies to expand the use of this technology for decision making in tropical forest conservation, management and restoration. (AU)

• FAPESP's process: 16/05219-9 - Monitoring forest landscape restoration through Light Detection and Ranging (LiDAR).
• Grantee: Danilo Roberti Alves de Almeida
• Support Opportunities: Scholarships in Brazil - Doctorate