Cycle 4 (2015 Deadline)
Lidar remote sensing of Brazilian Amazon forests: Analysis of forest biomass, forest degradation, and secondary regrowth
PI: Jean Ometto (email@example.com), Instituto Nacional de Pesquisas Espaciais (INPE)
U.S. Partner: Michael Keller, USDA - Forest Service
Project Dates: November 2015 - October 2018
|Georeferencing of field plots.|
Recently, the Earth System Science Center at INPE received substantial funding from the Amazon Fund-BNDES to improve knowledge of land use change and carbon budgets in the Brazilian Amazon. As part of this project, INPE plans to contract Lidar (light detection and ranging or laser scanning) remote sensing flights to acquire forest structure information over the Amazon that can be used to improve our knowledge of forest biomass. These data will substantially overcome the current limitations of insufficient and biased sampling of the Brazilian Amazon forest and provide the first large-scale, statistically balanced characterization of forest carbon stocks across the Amazon region. The experimental design will not only estimate carbon stocks across the region but also identify the proportions of forest that are currently secondary or degraded and thus are potentially large carbon sinks in the future. The Lidar data calibrated by a network of ground-based forest inventories will be used to achieve three objectives:
1. Reduce the uncertainty in the quantification of the above-ground carbon stocks of the forests of the Brazilian Amazon
2. Provide improved estimates of carbon emissions from deforestation and avoided carbon loss from reducing deforestation policies in the Brazilian Amazon region
3. Improve our ability to predict future carbon fluxes in the Brazilian Amazon by quantifying the current biomass status of secondary and degraded forests.
This project should improve the estimation of carbon emissions from deforestation in the Brazilian Amazon and construct a sound basis for future emissions scenarios, considering different options of land use and land cover change. Several strategies, policies, and compensation mechanisms have been proposed to minimize the impact of human actions on the natural forest. Among those, effective implementation of REDD+, soy and beef moratoria, and establishment of conservation units can benefit from better calculation of the balance between carbon emission by deforestation and forest degradation, and uptake, by secondary vegetation and mature forest growth. Through capacity building activities, this project will train young researchers in Lidar estimates of biomass, statistical biomass modeling and mapping, and emissions modeling. The data gathered will be a resource for the Brazilian Ministry of Science, Technology, and Innovation to produce the National Inventories on GHG emissions. The Brazilian Earth System Model, currently under development at INPE and associated universities and research institutes, is also an obvious client of the planned database. Finally, through its leadership in forest monitoring activities, Brazil has the potential to share its knowledge and technologies with neighboring countries, especially those of the Amazonian Cooperation Treaty Organization.
Summary of Recent Activities
In the most recent quarter (April - July 2017), Dr. Ometto and his team collected, validated, and processed a total of 625 transects during the first LiDAR campaign. Because of the complexity of the big data process, his team is developing a database that will store and process the collected LiDAR data using Postgresq, an open-source database, to potentially reducing the processing time. In the next month, the LiDAR provider will start to collect a total of 381 new transects as part of the second contract. As presented in the last report, 1006 transects will be collected over forest, secondary vegetation, and wetland forest in the Amazon region. In parallel, the team is applying the equation proposed by Longo et al. (2016) to estimate the forest biomass in Amazon region using LiDAR data.
A member of Dr. Ometto’s team attended the PEER Brazil Forum in Brasilia and Dr. Ometto attended a meeting in Piracicaba. At these meetings, new partnerships were established with different institutions to create a database for field data, such as a partnership with Dr. Yosio Edemir Shimabukuro of the National Institute for Space Research (INPE) and partnerships with Dr. Thiago S. F. Silva at the Geography Department of the São Paulo State University (UNESP). Other collaboration partners identified at the PEER Forum were Dr. Tomas Ferreira Domingues of University of São Paulo and Dr. Sabina Cerruto Ribeiro of the Federal University of Acre. Dr. Ometto’s team initially plans to share with these partners their knowledge and experience in improving biomass estimation methods for the Amazon.
Dr. Ometto’s team organized another field campaign to georeference field plots using Differential Global Positioning System (DGPS). The field work was done in Alta Floresta region in Mato Grosso state. His team continues collecting these data to increase the field data samples that will be used to calibrate the forest biomass map that will be produced with LiDAR data.
Also, his team organized and gave a presentation at the thematic session at the XVIII Brazilian Symposium on Remote Sensing. This session focused on presenting some of the major initiatives and potentials related to LiDAR, especially for biomass estimation of forests.
|The team conducts extensive data collection activities in the Amazon (Photos courtesy of Dr. Ometto).|
Back to PEER Cycle 4 Grant Recipients