PosterPDF Available

Remote sensing-based burn severity analysis of Amazon wildfire disaster in 2019: The case of Candeias do Jamari municipality in Rondonia State, Brazil

Authors:

Abstract

The wildfire of August 2019 in Brazilian Amazon gathered worldwide attention. It is important to analyze the burn pattern in order to investigate the causes behind this forest blaze such as climate change, deforestation, invasion, illegal logging, or other human activities. The objective of this research is to classify burn severity across different vegetation types using Landsat 8 imagery at Jamari municipality in State of Rondonia of Brazil affected by Amazon wildfire disaster in 2019. The study developed a burn severity map in order to assess areas affected by wildfires. The Normalized Burn Ratio (NBR) is calculated from the pre-fire and post-fire satellite images and differenced NBR (dNBR) image was then used to classify according to the burn severity ranges proposed by United States Geological Survey (USGS). The classified images of burned area were recoded into four classes of burn severity: Low Severity, Moderate-low Severity, Moderate-high Severity, and High Severity. The next step was to identify normalized difference vegetation index (NDVI) for the area and assign burn severity classes according to the pre-fire vegetation type. The study finds that total burned area covers 3715.56 hectares where areas with low severity, moderate-low severity, moderate-high severity, high severity covers 85.0, 255.1, 638.3 and 2737.3 hectares respectively. Different classes of burned areas show significant difference in types of Vegetation cover. 80.4% of area with High severity show no vegetation (NDVI values close to .1 or less). On the other hand, 66.9% of areas with low severity represents dense vegetation (approximately 0.6 to 0.9). The paper further discusses the causes behind this amazon wildfire and recommends polices. Keywords: Amazon wildfire disaster, Remote sensing, Normalized Burn Ratio, Climate change
RESEARCH POSTER PRESENTATION DESIGN © 2015
www.PosterPresentations.com
The wildfire of August 2019 in Brazilian Amazon gathered worldwide attention. It is important to analyze the burn pattern in
order to investigate the causes behind this forest blaze such as climate change, deforestation, invasion, illegal logging, or
other human activities. The objective of this research is to classify burn severity across different vegetation types using
Landsat 8 imagery at Jamari municipality in State of Rondonia of Brazil affected by Amazon wildfire disaster in 2019. The
study developed a burn severity map in order to assess areas affected by wildfires. The Normalized Burn Ratio (NBR) is
calculated from the pre-fire and post-fire satellite images and differenced NBR (dNBR) image was then used to classify
according to the burn severity ranges proposed by United States Geological Survey (USGS). The classified images of burned
area were recoded into four classes of burn severity: Low Severity, Moderate-low Severity, Moderate-high Severity, and High
Severity. The next step was to identify normalized difference vegetation index (NDVI) for the area and assign burn severity
classes according to the pre-fire vegetation type. The study finds that total burned area covers 3715.56 hectares where areas
with low severity, moderate-low severity, moderate-high severity, high severity covers 85.0, 255.1, 638.3 and 2737.3 hectares
respectively. Different classes of burned areas show significant difference in types of Vegetation cover. 80.4% of area with
High severity show no vegetation (NDVI values close to .1 or less). On the other hand, 66.9% of areas with low severity
represents dense vegetation (approximately 0.6 to 0.9). The paper further discusses the causes behind this amazon wildfire
and recommends polices.
Keywords: Amazon wildfire disaster, Remote sensing, Normalized Burn Ratio, Climate change
Abstract
Candeias do Jamari, the study area is a municipality located in the Brazilian state of Rondonia. It contains 76% of the fully
protected 71,161 hectares Samuel Ecological Station. It also contains part of the 221,218 hectares Jacunda National Forest.
Study Area
Conclusion
The study developed a burn severity map using Landsat 8 satellite data in order to
assess the areas affected by wildfires. The Normalized Burn Ratio (NBR) is calculated
from the pre-fire and post-fire satellite images and differenced NBR (dNBR) image
was then used to classify according to the burn severity ranges proposed by United
States Geological Survey (USGS). The classified images of burned area were recoded
into four classes of burn severity: Low Severity, Moderate-low Severity, Moderate-
high Severity, and High Severity. The next step was to identify normalized difference
vegetation index (NDVI) for the area and assign burn severity classes according to the
pre-fire vegetation type. (Explained in the flow chart).
NBR = (NIR SWIR2) / (NIR + SWIR2)
NDVI = (NIR Red) / (NIR + Red)
Methods Results
Department of Geography and Environment, University of Dhaka.
Sakib Imtiaz
Remote sensing based burn severity analysis of Amazon wildfire disaster in 2019:
The case of Candeias do Jamari municipality in Rondonia State, Brazil
References
Petropoulos, G. P., Griffiths, H. M., & Kalivas, D. P. (2014). Quantifying spatial and temporal vegetation recovery dynamics following a wildfire event in a Mediterranean landscape using EO data and GIS. Applied Geography, 50, 120131.
U.S. Geological Survey. (2016). LandsatEarth Observation Satellites; Version 1.1; U.S. Geological Survey Fact Sheet20153081; U.S. Geological Survey: Washingotn, DC, US.
Background
The 2019 Amazon rainforest wildfires season occurred in the Amazon rainforest and Amazon biome within
Brazil, Bolivia, Paraguay, and Peru during tropical dry season. Brazil has seen a high number of fires in
2019. The National Institute for Space Research says its satellite data shows a 76% increase on the same
period in 2018 (BBC, 2019). Slash-and-burn methods are often used to clear the forest for agriculture,
livestock, logging, and mining, leading to deforestation of the Amazon rainforest. Wildfires are common in
the Amazon during the dry season. However, this year most are believed to have been caused by farmers
and loggers clearing land for crops or grazing. The increased rates of fire counts in 2019 led to international
concern about the future of Amazon rainforest, which plays a significant role in mitigating Global
Warming. Apart from climate change and illegal activities in the forest, environmentalists also criticized
Brazil’s pro-business environmental policies behind this wildfire disaster.
The burn severity pattern in different types of land cover reveals that
the wildfire was mainly caused by anthropogenic activities.
However, that dry weather, wind, and extreme heat caused the fires
to spread so widely. Further study is needed to explore the indepth
cause behind this massive wildfire.
The study finds that total burned area covers 3715.56 hectares where areas with
low severity, moderate-low severity, moderate-high severity, high severity covers
85.0, 255.1, 638.3 and 2737.3 hectares respectively. Different classes of burned
areas show significant difference in types of Vegetation cover. 80.4% of area with
High severity show no vegetation (NDVI values close to .1 or less). On the other
hand, 66.9% of areas with low severity represents dense vegetation (approximately
0.6 to 0.9).
Contact Info:
sakibimage@gmail.com
www.sakibimtiaz.com
DRR Exposure & Opportunities 2019
Organized By: Climate & Disaster Resilience Club
Department of Disaster Science and Management
Supported By: American Geophysical Union(AGU)
Types of Vegetation NDVI Value
Classification of Burn Severity (Hectares)
High
Severity
Moderate-
High
Severity
Moderate-
Low
Severity
Low
Severity
No vegetation (barren
rock, sand, or snow) 0.1 or less 68.3 78.5 78.8 6.4
Sparse vegetation such
as shrubs and grasslands
or senescing crops
approximately
0.2 to 0.5 0.0 118.9 559.4 900.1
Dense vegetation
(temperate and tropical
forests or crops at their
peak growth stage
approximately
0.6 to 0.9 16.7 57.7 0.0 1830.8
Total Area (Hectares) 85.0 255.1 638.3 2737.3
Figure 1: Aerial view of some of the burned areas in the city of Candeiras do Jamari in the state of Rondônia, Brazil.
(Photo Courtesy: AFP, Greenpeace)
Table 2: Types of vegetation in different classes of burn severity
Candeias do Jamari
Figure 2: Candeiras do Jamari in the state of Rondônia,
Brazil. (Source: Open Street Map) Figure 4: Burn severity Map, Candeiras do Jamari in the
state of Rondônia, Brazil. (Source: Open Street Map)
Satellite Bands Date of Acquisition
Landsat 8 1 to 7 and 9 27 July, 2019 (Pre-Fire)
Landsat 8 1 to 7 and 9 28 Aug, 2019 (Post-Fire)
Source :U.S. Geological Survey Archive
Table 1. Data characteristics of satellite imageries
Figure 3 : Flowchart
ResearchGate has not been able to resolve any citations for this publication.
ResearchGate has not been able to resolve any references for this publication.