Assessment of Burned Forest Area Severity and Postfire Regrowth in Chapada Diamantina National Park (Bahia, Brazil) Using dNBR and RdNBR Spectral Indices

Sarah Moura B Dos Santos,Gustavo Baptista,António Bento-Gonçalves,Washington Franca-Rocha

doi:10.3390/geosciences10030106

Abstract

Fire scar detection through orbital data can be done using specific techniques, such as the use of spectral indices like the normalized burn ratio (NBR), which are designed to help identify burnt areas as they have typical spectral responses. This paper aims to characterize burn severity and regrowth in areas hit by three fires in the Chapada Diamantina National Park (Bahia, Brazil) and its surrounding area through the differenced normalized burn ratio (dNBR) and relative differenced normalized burn ratio (RdNBR) spectral indices. The data acquired were pretreated and prepared adequately to calculate the indices. We conclude that for the study area, considering the limitations of fieldwork, the multitemporal index dNBR and the relative index RdNBR are important tools for classifying burnt areas and can be used to assess the regrowth of vegetation.

Highlights

In 2015, the Chapada Diamantina National Park (Bahia, Brazil) was ravaged by several fires that lasted four months from September to December, a typical fire period in the region, as demonstrated by previous studies [1,2,3]. To manage such an environment, we need to look at the impact of fires therein since their increase in frequency and intensity has been documented in several ecosystems and represents a significant impact on global warming, given that fires often entail a great loss of biomass and release of carbon, which can contribute to changes in local climate [4,5,6,7,8,9]
The normalized burn ratio (NBR) multitemporal difference, that is, the differenced normalized burn ratio (dNBR) index and the relative differenced normalized burn ratio (RdNBR) relative index, have recently become the standard fire severity measurements, using data from the Landsat satellite because of their ease of implementation, since they generally have a large spectral separation that can be achieved between the near-infrared and the short-wave infrared bands, and because of the unique combination of moderate spatial resolution data captured in these regions [6,9,10,12]
The NBR index proved to be sensitive to pre and postfire offsets of the pixels affected by the fires, within the short-wave infrared reflectance (SWIR)–near infrared reflectance (NIR) space

Summary

Introduction

In 2015, the Chapada Diamantina National Park (Bahia, Brazil) was ravaged by several fires that lasted four months from September to December, a typical fire period in the region, as demonstrated by previous studies [1,2,3] To manage such an environment, we need to look at the impact of fires therein since their increase in frequency and intensity has been documented in several ecosystems and represents a significant impact on global warming, given that fires often entail a great loss of biomass and release of carbon, which can contribute to changes in local climate [4,5,6,7,8,9]. From a broad perspective, fire severity can be defined as the degree of changes in soil and vegetation caused by the fire [16], as it quantifies the short term effects of the fire [12]

Objectives

Results

Discussion

Conclusion