An alternative approach for mapping burn scars using Landsat imagery, Google Earth Engine, and Deep Learning in the Brazilian Savanna

Arruda, Vera; Piontekowski, Valderli Jorge; Alencar, Ane; Pereira, Reginaldo Sérgio; Matricardi, Eraldo Aparecido Trondoli

doi:10.1016/j.rsase.2021.100472

Cited by 32 publications

(32 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Thus, the smaller area detected by AQM30m in 2015 is very close to that obtained by MCD64 (10,954,230 ha), which is known for its underestimation of burned areas in Cerrado [10,75,76]. This low accuracy is generally attributed to MODIS coarse spatial resolution (500 m), which challenges the detection of small and highly fragmented fires, in particular, those associated with agricultural burns [11,40,76]. The comparison of Landsat medium-and MODIS coarseresolution BA products at the continental scale of Cerrado not only confirms the very conservative behavior of MCD64 but also the strong underestimation of savanna fire emissions, as highlighted by previous studies in Africa [77].…”

Section: Discussionsupporting

confidence: 62%

“…Other authors [37] developed an annual BA product using a random forest algorithm and [38] presented a machine learning approach to map burned areas over Asia, using composites of differential spectral-indices and three different classifiers, Classification and Regression Tree, Random Forest, and Support Vector Machine (SVM), both in the Google Earth Engine (GEE) platform [39]. Although a variety of studies applying Landsat image composites for burned area mapping have been successfully conducted worldwide, to date, there are few initiatives over the Cerrado region [40,41].…”

Section: Introductionmentioning

confidence: 99%

“…Another notable limitation of most BA mapping algorithms is the need for human intervention in training sample acquisition [28,[40][41][42][43], making it a challenge to maintain operational products over large areas and long periods. This limitation may be circumvented by exploring two specific aspects of the problem: the relationship between active fires and burned areas and the possibility of envisaging BA mapping under the framework of one-class classification [44].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Multi-Sensor, Active Fire-Supervised, One-Class Burned Area Mapping in the Brazilian Savanna

et al. 2021

View full text Add to dashboard Cite

Increasing efforts are being devoted to understanding fire patterns and changes highlighting the need for a consistent database about the location and extension of burned areas (BA). Satellite-derived BA mapping accuracy in the Brazilian savannas is limited by the underestimation of burn scars from small, fragmented fires and high cloudiness. Moreover, systematic mapping of BA is challenged by the need for human intervention in training sample acquisition, which precludes the development of automatic-generated products over large areas and long periods. Here, we developed a multi-sensor, active fire-supervised, one-class BA mapping algorithm to address several of these limitations. Our main objective is to generate a long-term, detailed BA atlas suitable to improve fire regime characterization and validation of coarse resolution products. We use composite images derived from the Landsat satellite to generate end-of-season maps of fire-affected areas for the entire Cerrado. Validation exercises and intercomparison with BA maps from a semi-automatic algorithm and visual photo interpretation were conducted for the year 2015. Our results improve the BA mapping by reducing omission errors, especially where there is high cloud frequency, few active fires are detected, and burned areas are small and fragmented. Finally, our approach represents at least a 45% increase in BA mapped in the Cerrado, in comparison to the annual extent detected by the current coarse global product from MODIS satellite (MCD64), and thus, it is capable of supporting improved regional emissions estimates.

show abstract

Section: Discussionsupporting

confidence: 62%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Multi-Sensor, Active Fire-Supervised, One-Class Burned Area Mapping in the Brazilian Savanna

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Os incêndios florestais alteram historicamente a vegetação natural. Assim sendo, a frequência de incêndios vem mudando nosúltimos anos, o que contribui para a perda da biodiversidade e afeta cada vez mais a vegetação nativa, os habitats naturais e o ecossistema nas regiões tropicais [Jr. et al 2014, Arruda et al 2021 O Instituto Nacional de Pesquisas Espacias (INPE), por meio do Programa Queimadas, monitora e modela a ocorrência e a propagação de incêndios na vegetação, utilizando técnicas de sensoriamento remoto, geoprocessamento e modelagem numérica. Em seu Portal 2 são disponibilizados dados abertos sobre focos de calor, atualizados diariamente e relacionados aos continentes americano, africano e europeu.…”

Section: A Questão Ambiental E Focos De Incêndio No Brasilunclassified

Análise de Dados de Focos de Calor no Brasil Através de Técnicas de Visualização

Costa

Henriques

Belloze

2021

Anais Do XV Brazilian E-Science Workshop (BRESCI 2021)

View full text Add to dashboard Cite

Este trabalho analisa dados de focos de incêndio nos Estados brasileiros entre os anos de 1998 e 2020. Técnicas de visualização de dados são usadas para realizar uma análise exploratória dos dados, de modo a gerar informações úteis, tais como os meses e anos com maior quantidade de focos. Além disso, um modelo de predição foi usado, de modo a avaliar a tendência da quantidade de focos, para três Estados específicos. Os resultados mostraram que as técnicas de visualização, em conjunto com a predição, formam um conjunto válido de ferramentas para a análise de dados e tomada de decisões.

show abstract

“…Although some review studies involving wildfires in conjunction with the use of remote sensing techniques have been proposed [20,25,26,[36][37][38][39], it was considered important to list and map the main conceptual and methodological trends, identifying significant theoretical contributions and describing their applications through case studies developed in different parts of the world. Considering the need for a structured revision to be carried out of scientific production addressing the research topics, the objective of this paper is to analyse the scientific publications indexed in the Scopus database, simultaneously encompassing the subjects of forest fires and remote sensing over a 30-year timescale.…”

Section: Introductionmentioning

confidence: 99%

Research on Wildfires and Remote Sensing in the Last Three Decades: A Bibliometric Analysis

Santos

Bento‐Gonçalves

Vieira

2021

Forests

View full text Add to dashboard Cite

Evaluating the impact of wildland fires on landscapes, a pursuit increasingly supported by remote sensing techniques, requires an understanding of wildfire dynamics. This research highlights the main insights from the literature related to “wildfires” and “remote sensing” published between 1991 and 2020. The Scopus database was used as a source of information regarding scientific production on these topics, after which bibliometric tools were employed as a means through which to reveal patterns in this network of journals, terms, countries, and authors. The results suggest that these subject areas have undergone significant developments in the last three decades, having been the focus of growing interest among the scientific community. The most relevant contributions to the literature available have been made by researchers working in the areas of earth and environmental sciences (54% of the publications), primarily in the United States, China, Spain, and Canada. Research trends in this field have undergone a significant evolution in recent decades, explained by the strong relationship between the technological evolution of detection methods and remote sensing data acquisition.

show abstract

An alternative approach for mapping burn scars using Landsat imagery, Google Earth Engine, and Deep Learning in the Brazilian Savanna

Cited by 32 publications

References 27 publications

Multi-Sensor, Active Fire-Supervised, One-Class Burned Area Mapping in the Brazilian Savanna

Multi-Sensor, Active Fire-Supervised, One-Class Burned Area Mapping in the Brazilian Savanna

Análise de Dados de Focos de Calor no Brasil Através de Técnicas de Visualização

Research on Wildfires and Remote Sensing in the Last Three Decades: A Bibliometric Analysis

Contact Info

Product

Resources

About