@InProceedings{SantosShiDuaJorGas:2017:MuApEs,
author = "Santos, Erone Ghizoni and Shimabukuro, Yosio Edemir and Duarte,
Valdete and Jorge, Anderson and Gasparini, Kaio Allan Cruz",
affiliation = "{Instituto Nacional de Pesquisas Espaciais (INPE)} and {Instituto
Nacional de Pesquisas Espaciais (INPE)} and {Instituto Nacional de
Pesquisas Espaciais (INPE)} and {} and {Instituto Nacional de
Pesquisas Espaciais (INPE)}",
title = "Multi-stage approach to estimate forest biomass in degraded area
by fire and selective logging",
booktitle = "Proceedings...",
year = "2017",
organization = "AGU Fall Meeting",
abstract = "The Amazon forest has been the target of several threats
throughout the years. Anthropogenic disturbances in the region can
significantly alter this environment, affecting directly the
dynamics and structure of tropical forests. Monitoring these
threats of forest degradation across the Amazon is of paramount to
understand the impacts of disturbances in the tropics. With the
advance of new technologies such as Light Detection and Ranging
(LiDAR) the quantification and development of methodologies to
monitor forest degradation in the Amazon is possible and may bring
considerable contributions to this topic. The objective of this
study was to use remote sensing data to assess and estimate the
aboveground biomass (AGB) across different levels of degradation
(fire and selective logging) using multi-stage approach between
airborne LiDAR and orbital image. The study area is in the
northern part of the state of Mato Grosso, Brazil. It is
predominantly characterized by agricultural land and remnants of
the Amazon Forest intact and degraded by either anthropic or
natural reasons (selective logging and/or fire). More
specifically, the study area corresponds to path/row 226/69 of
OLI/Landsat 8 image. With a forest mask generated from the
multi-resolution segmentation, agriculture and forest areas,
forest biomass was calculated from LiDAR data and correlated with
texture images, vegetation indices and fraction images by Linear
Spectral Unmixing of OLI/Landsat 8 image and extrapolated to the
entire scene 226/69 and validated with field inventories. The
results showed that there is a moderate to strong correlation
between forest biomass and texture data, vegetation indices and
fraction images. With that, it is possible to extract biomass
information and create maps using optical data, specifically by
combining vegetation indices, which contain forest greening
information with texture data that contains forest structure
information. Then it was possible to extrapolate the biomass to
the entire scene (226/69) from the optical data and to obtain an
overview of the biomass distribution throughout the area.",
conference-location = "New Orleans",
conference-year = "11-15 Dec.",
language = "en",
targetfile = "santos_mult.pdf",
urlaccessdate = "05 maio 2024"
}