Tracking the Progression of Burned Areas in Tropical Peat Swamp

Forests by Integrating Sentinel Optical and SAR Imagery: A Case

Study of Binsuluk Forest Reserve in Sabah, Malaysia

Nurul Aina Abdul Aziz

, Mckreddy Yaban

, Muhamad Zulfazli Zakaria

and Siti Atikah Mohamed Hashim

Malaysian Space Agency (MYSA), No. 13, Jalan Tun Ismail, 50480 Kuala Lumpur, Malaysia

Keywords: Forest Fire, Burned Area, Sentinel-1, Sentinel-2, Peatland.

Abstract: Climate change and rising global temperatures are driving forest fires to become more intense and frequent

worldwide, particularly in peat swamp forest. Since the predominant burning mechanism in peatland forest is

smouldering combustion, it causes widespread air pollution and emits massive amounts of carbon due to

prolonged episodes of fire events. Therefore, the development of a unique approach to monitor forest fire

progressions through burned area mapping mainly in persistent cloud cover is vital for the estimation of fire

extent, location, and land cover affected. Thus, this research aims to evaluate the capabilities of Sentinel-1

SAR and Sentinel-2 optical time series in boosting the frequency and accuracy of burn area progression

mapping in peatland areas. Results from the forest fire series in Binsuluk Forest Reserve, which occurred

from February to April 2024, indicated a reduction in the backscatter value of the cross-polarized (VH) signal

in the burned area for Sentinel-1 SAR C band. Despite the cloud cover challenge, Sentinel-2 continues to

deliver essential data on the positioning of active fires and smoke plumes, with burn area detection being more

precise when utilizing the Normalized Difference Moisture Index (NDMI) compared to the Normalized Burn

Ratio (NBR). The integration of Sentinel optical and SAR imagery has effectively facilitated an increased

tracking frequency and precision for the evolution of burned areas.

1 INTRODUCTION

Tropical peat swamp ecosystems are widespread in

Southeast Asia, particularly in Borneo. More than

half of Malaysia's 2.6 million hectares of peat swamp

forest are situated on Malaysian Borneo (Meiling L.,

2016), while Sabah was believed to have 86,000

hectares of peat swamp forest, with roughly 60,000

ha of mixed peat swamp forest on the Klias Peninsula.

Agricultural growth and fires caused by El

Niño/Southern Oscillation have led to the fast

disappearance of peat swamp forests. Kamlun, K. U.,

& Phua, M. H. (2024) indicate that agriculture is the

most influential anthropogenic factor associated with

the fire-affected areas while the distance to settlement

played an increasingly important role in the fire

https://orcid.org/0009-0003-0776-0045

https://orcid.org/0009-0007-5083-327X

https://orcid.org/0009-0002-8965-0722

https://orcid.org/0009-0000-6179-2193

affected areas and contributes to the deforestation of

the peat swamp forest in Klias Peninsula.

Satellite-based earth observation (EO) systems

are able to provide consistent and frequent

measurements over vast remote areas on the earth’s

surface. This allows the monitoring of forest fire

progressions across the globe in a timely and cost-

effective way (Engelbrecht et al., 2017; Chuvieco et

al., 2020). Phua et al. (2007) examined the use of

several vegetation indices in image differencing

technique for detecting burned peat swamp forest.

Phua et al. (2008) has further investigated into a fast

approach for detecting disturbances in multiple

change events.

In previous studies, Sentinel-2 MSI data has been

effectively used to assess burn severity, or the degree

to which an area has been affected by a fire. This is

264

Abdul Aziz, N. A., Yaban, M., Zakaria, M. Z. and Mohamed Hashim, S. A.

Tracking the Progression of Burned Areas in Tropical Peat Swamp Forests by Integrating Sentinel Optical and SAR Imagery: A Case Study of Binsuluk Forest Reserve in Sabah, Malaysia.

DOI: 10.5220/0013482300003935

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 11th International Conference on Geographical Information Systems Theory, Applications and Management (GISTAM 2025), pages 264-271

ISBN: 978-989-758-741-2; ISSN: 2184-500X

because the MSI data is sensitive to changes in the

chlorophyll content of vegetation, which is typically

reduced in areas that have been burned. In general,

the accuracy of burned area mapping increases with

higher spatial resolution data. Using a lower spatial

resolution image, such as the 20 m spatial resolution

of Sentinel-2 data, may result in less accurate burned

area maps.

In the context of identifying burn areas,

Normalised Burn Ratio (NBR), and Normalised

Difference Moisture Index (NDMI) are among the

indices used in the context of determining burn

regions; each has certain advantages and

disadvantages. Phua et.al (2007) found that NBR is

especially good at identifying burn intensity and

defining burn scars. In dry conditions or when

attempting to distinguish between different kinds of

vegetation, NDMI may be less effective. However, it

is useful for determining the moisture level of

vegetation, which can indirectly suggest disease

susceptibility or recovery.

In tropical rainforests, burned areas may fade

within a few weeks as fresh foliage grows. Some

satellites can detect actively burning places, but may

not detect the entire charred area due to cloud cover

or delays in satellite images. Thus, SAR could serve

as an alternative data source of information since

radar sensors can image day and night, and are

capable of penetrating clouds, smoke, and smog.

Further, SAR is sensitive to changes in vegetation

structure and soil moisture following wildfire

(Bourgeau-Chavez et al., 2007). These characteristics

give SAR unique advantages in monitoring on-going

forest fire event.

Tanase, Mihai A., et al (2010) has analyzed SAR

data at X-, C-, and L-bands to investigate the

relationship between backscatter and forest focusing

on both HH and VV polarizations as well as on cross

polarized (HV). Results obtained in Spain highlighted

that for X- and C-bands, the copolarized (HH and

VV) backscatter increased with burn severity, in

detail: 1) for all frequencies, the cross polarized (HV)

decreased with burn severity; 2) C- and L-bands

cross-polarized backscatter showed better potential

for burn severity; and 3) the small dynamic range

observed for X-band data could prevent its use in

vegetation affected by fires.

Gaveau, D., Descals, A., Salim, M., Sheil, D., &

Sloan, S. (2021) present new and validated 2019

burned-area estimates for Indonesia using a time

series of the atmospherically corrected surface

reflectance multispectral images (level 2A product)

taken by the Sentinel-2A and B satellites. The

frequency–area distribution of the Sentinel-2 burn

scars follows the apparent fractal-like power law or

Pareto pattern often reported in other fire studies,

suggesting good detection over several magnitudes of

scale with 97.9% accuracy.

This research aims are to assess the effectiveness

of both Sentinel-1 SAR and Sentinel-2 optical time

series images in improving the frequency and

precision of burn area progression mapping in

peatland regions. Various approaches for optical and

SAR will be suggested to monitor the size of the

burned area in near real-time.

2 MATERIAL AND METHODS

The primary objective of the suggested methodology

is to utilise image differencing techniques to detect

burned areas in the Binsuluk Forest Reserve through

optical and SAR imagery, hence assessing the

evolution of fire in the affected region. The forest

reserve boundaries provided by the Sabah Forestry

Department is essential for identifying the source of

fire and consistently calculating the area of land

destroyed. Ultimately, we examined the impact of the

fires on the current protective forest reserve.

Figure 1: Workflow of Burned Area Mapping.

2.1 Area of Interest

The research was carried out in Binsuluk Forest

Reserve which is a protected forest reserve on the

Klias Peninsula, in Beaufort District of Interior

Active fire data collection: Hotspot data

analysis to identify location and date of fire

Overlaying with Forest

Reserve Boundary

Sentinel-1 and Sentinel 2 data

collection

Develop techniques of

burned area mapping

using SAR

(Backscatter

ratioVH/VV)Sentinel-

MSI Index: Normalize

Burn Ratio (NBR) and

Normalized Different

Moisture Index (NDMI)

Field Verification

Burned Area Map and Trend of

Burned Area Progression

Tracking the Progression of Burned Areas in Tropical Peat Swamp Forests by Integrating Sentinel Optical and SAR Imagery: A Case Study

of Binsuluk Forest Reserve in Sabah, Malaysia

265

Division, Sabah, the Sabah Forestry Department in

1992. Its area is 12,106 hectares (121.06 km

). The

reserve is mostly flat, consisting mostly of peat

swamp forest, with a small area of mangroves. The

forest type here is peat swamp forest over soils of the

Klias Association. Most of the FR was badly burnt

during the long drought of 1997-1998. Of the

remaining trees, Dryobalanops rappa is the most

dominant canopy tree species. Most of the burnt areas

are dominated by small shrubs. The Binsuluk Forest

Reserve Boundary was overlaid with the images as

shown in Figure 2 to identify the actual AOI of the

study area

Figure 2: Sentinel-2 true colour composite on April 24,

2024, of the study area. Sentinel-2 image available at

https://scihub.copernicus.eu/.

Previously in 2016 large fires in peat bogs

occured, which were caused by fires spread to

Binsuluk and other forest reserves from nearby open

burning had contributed to the 2016 Malaysian haze.

Over half of the reserve were burnt during this event.

Open burning caused yet another forest fire in 2020,

this time burning 274 hectares (2.74 km2). However,

after some action and enforcement from the

government, the trend of hotspots kept decreasing

over the years as a result of mitigation action from

Malaysia via the National Haze and Dry Weather

Committee and ASEAN Agreement on

Transboundary Haze Pollution.

2.2 Active Fire Data Collection

First, peat fires were identified by overlaying active

fire data from from NASA Fire Information for

Resource Management System (FIRMS) data

catalog. FIRMS distributes Near Real-Time (NRT)

active fire data from the Moderate Resolution

Imaging Spectroradiometer (MODIS) aboard the

Aqua and Terra satellites, and the Visible Infrared

Imaging Radiometer Suite (VIIRS) aboard the Suomi

National Polar-orbiting Partnership (Suomi NPP) and

NOAA 20 satellitesTerra/Aqua MODIS hotspots

onto the existing peatland map. Fire and Thermal

Anomalies algorithms are automated pre-processed

utilising Python scripts and ArcGIS software to input

administrative boundary information such as

division, district, and city. Next, the NRT data from

these four sensors from 2020 to 2024 are utilised to

calculate the total number of hotspots, illustrate the

hotspot distribution, and certify the high fire prone

area.

Figure 3: The distribution of hotspot in FR Binsuluk within

January- May 2024.

In this study, the daily hotspot is downloaded

from NASA Fire Information for Resource

Management System (FIRMS) data catalogue.

FIRMS distributes Near Real-Time (NRT) active fire

data from the Moderate Resolution Imaging

spectroradiometer (MODIS) aboard the Aqua and

Terra satellites, and the Visible Infrared Imaging

Radiometer Suite (VIIRS) aboard the Suomi National

Polar-orbiting Partnership (Suomi NPP) and NOAA

20 satellites. Combined (Terra and Aqua) MODIS

NRT active fire products (MCD14DL) are processed

using the standard MOD14/MYD14. Fire and thermal

anomalies algorithms are automated pre-processed

utilising Python scripts and ArcGIS software to input

administrative boundary information such as

division, district, and city, which is then merged into

the ForFIS database for user access. Next, the NRT

data from these four sensors for year 2024 are plotted

to get the distribution as shown in Figure 3.

GISTAM 2025 - 11th International Conference on Geographical Information Systems Theory, Applications and Management

266

2.3 Sentinel Data Collection and

Pre-Processing

Sentinel SAR and optical satellite images of Binsuluk

Forest Reserve, acquired from January to April 2024,

were retrieved from the ESA Copernicus Data Space

Ecosystem portal. The dates picked are based on the

existence of hotspot in continuous cycle of fire event.

Basically, Sentinel-2 and Sentinel-1 are two key

missions in a series of satellite missions initiated by

the ESA to support global environmental monitoring

and resource management. Sentinel-1 mission

consists of a pair of polar-orbiting satellites that

provide high-resolution, all-weather imaging of the

Earth’s surface. It includes C-band imaging operating

in four modes with different resolutions (up to 5 m).

ESA S1 mission provides global coverage of freely

available dual polarization C-band SAR images with

a repeat cycle at 6-days and revisit frequencies at 1–3

days taking into account of ascending and descending

orbits and overlaying. Two bands of Sentinel-1 were

used as SAR features, including VH and VV bands.

The SAR images were exported from Sentinel-1 SAR

GRD (Ground Range Detected) image collection

named “COPERNICUS/ S1_GRD” in the Copernicus

Platform. Major pre-processing step are running using

The Sentinel Application Platform (SNAP). The

software is developed by Brockmann Consult,

Skywatch, Sensar and C-S. SNAP software.

Table 1: Dates of satellite acquired in a descending pass,

by Sentinel-1 (s1) and Sentinel- 2 (s2).

Acquisition

Date

Imagery Acquisition

Date

Imagery

5 Jan 2024 S2 5 March 2024 S2

an 2024 S1 and

2 10

arch 2024 S1 and

an 2024 S2 15

arch 2024 S2

an 2024 S2 20

arch 2024 S2

an 2024 S1 25

arch 2024 S2

an 2024 S2 30

arch 2024 S2

an 2024 S2 3

pri

2024 S1

eb 2024 S1 4

pri

2024 S2

eb 2024 S2 9

pri

2024 S2

eb 2024 S2 14

pri

2024 S2

eb 2024 S2 19

pri

2024 S2

eb 2024 S2 24

pri

2024 S2

eb 2024 S2 2

pri

2024 S1

eb 2024 S1 29

pri

2024 S2

eb 2024 S2

Sentinel-2 is a multi-spectral imaging system that

provides high-resolution imaging of the Earth’s

surface with 13 bands. It also comprises two satellites,

each with a spatial resolution of up to 10 m. These two

satellites can be combined to provide full coverage of

the surface of the earth every five-day interval.

Sentinel-2 images were atmospherically corrected.

Table 1 shows the total images used for tracking the

forest fire. In the normal condition of monitoring

peatland areas in Malaysia, only 12 out of a total of 24

S2 images have less than 30% cloud cover and can be

used to generate the burnt area. The remaining S1

images totally can be utilized to complete the cycle.

2.4 Techniques of Burned Area

Mapping

2.4.1 Multi-Spectral Burned Area Index

Ten bands of Sentinel-2 were selected as spectral

features, including three visible bands, one Near-

Infrared (NIR) band, four Red-edge bands, and two

short-wave infrared (SWIR) bands. The raw image

from band 2,3,4,8 and 12 were selected for displaying

the burned area. The responses of these features in

various spectral bands will shows specific character

during the visualization of burned area.

In the context of identifying burn areas in dense

cloud cover area, we apply various indices like

Normalized Difference Moisture Index (NDMI) and

Normalized Burn Ratio (NBR) are utilized, each

offering unique benefits and facing specific

limitations. NBR generates values ranging from -1 to

1. Intense green vegetation will exhibit a high NBR

value, whereas charred vegetation will have a low

value. Regions characterized by dry, brown vegetation

or exposed soil will have lower NBR values compared

to green vegetation. Otherwise, NDMI is important in

evaluating moisture levels in vegetation, which might

indirectly reflect fire vulnerability or recovery;

nevertheless, its efficacy may diminish in arid

conditions or when distinguishing across vegetation

kinds

2.4.2 Sentinel-1 SAR GRD Backscatter

SAR-based burnt area mapping mainly relies on the

resultant changes in radar backscattering, which

depend on the modification degree caused by fire

events in backscattering mechanisms. The total

amount of energy scattered back to radar sensor can

be influenced by sensor characteristic (signal

wavelength and polarization), target properties

(including vegetation structure, dielectric permittivity,

canopy and water content, soil moisture and dielectric

properties, and surface roughness) and observation

geometry (Imperatore et al., 2017).

2.5 Field Verification

Using a mobile GPS device, we conducted an aerial

survey in Binsuluk FR to validate the burned area.

Tracking the Progression of Burned Areas in Tropical Peat Swamp Forests by Integrating Sentinel Optical and SAR Imagery: A Case Study

of Binsuluk Forest Reserve in Sabah, Malaysia

267

The geographic coordinates and land cover categories

are included in each sample point. In addition to the

burned area, the image captures the nearly identical

features of cleared land as a result of agricultural

activities and flooded wetlands. In addition to aerial

surveys, verification of the hotspot's status as an

active fire was conducted using multispectral higher-

resolution satellites, such as SPOT and Pleiades. We

collaborate with the Sabah Forestry Department to

conduct airborne surveys for field verification at 40

locations throughout the Binsuluk FR zone as shown

in Figure 4.

Figure 4: Location of aerial survey and sample of photos of

burnt areas.

2.6 Multi-Temporal of Burned Area

Progression

Considering the difference in acquisition time of SAR

and optical observation, we assessed the initial until

final stage of the progression maps, which mean

assessing the full progressions of each fire event. This

is deemed as unsuitable for validation since the fire

could progress very fast during that period. In the

normal condition of monitoring peatland areas in

Malaysia, only 12 out of a total of 24 S2 images have

less than 30% cloud cover and can be used to generate

the burnt area. The remaining S1 images totally can

be utilized to complete the cycle.

3 RESULTS AND DISCUSSION

3.1 Multi-Spectral Indices in Burned Area

Mapping

The multi-spectral image was studied with three well-

known band combinations which is true colour

composite, near infrared (NIR) false colour

composite image and shortwave infrared (SWIR)

false colour composite image to decide which

combination best to highlight the burned areas. Figure

5 shows SWIR colour composite that act as the best

band combination to visualize the burned areas.

Figure 5: SWIR false colour composite (a) enhances the

burned area features and active fire location in Sentinel-2

MSI dated 27 April 2024 compared to (b) true colour

composite

NBR excels at assessing burn intensity and

outlining burn scars; however, its efficacy diminishes

in areas with sparse vegetation or in identifying early

post-fire regrowth. Using +0.1 as a criterion produces

false positives on unburned cleared land and forest

areas where foliage has dried prior to the fire. A far

more cautious criterion (+0.3) produces a better

outcome. Rahman et al. (2018) discovered that a

dNBR threshold value of +0.1 is adequate for

distinguishing burnt from unburnt areas using

Sentinel-2.

This study indicates that specific conditions may

result in false positives, since significant smoke in the

post-burn image can distort the dNBR value. Regions

that have had vegetation removal through alternative

methods (logging, harvesting, and landslides) at the

conclusion of the baseline period may erroneously

appear as scorched.

The Normalized Difference Moisture Index

(NDMI) is employed to assess vegetation water

content and monitor drought conditions. Figure 6

indicates that the red and yellow regions are classified

as burned areas. NDMI can distinguish between

GISTAM 2025 - 11th International Conference on Geographical Information Systems Theory, Applications and Management

268

clouds and shadows when identifying burned areas

compare to NBR. The NDMI value range is from -1

to 1. Negative NDMI values (approaching -1)

indicate infertile soil. Values near zero (-0.2 to 0.4)

typically indicate water stress. Elevated, positive

values indicate substantial canopy coverage without

water stress (about 0.4 to 1).

Figure 6: Comparison on Burned Area mapping on S2

imagery dated 24

April 2024 using difference indices

NDMI and NBR.

3.2 SAR-Based Burned Area Mapping

Figure 7 illustrate that VV and VH polarizations can

be display into a false colour visualization. It uses the

VV polarization in the red channel, the VH

polarization in the green channel, and a ratio of

VH/VV in the blue channel. It shows water areas in

dark red (black), urban areas in yellow, vegetated

areas in turquoise, and bare ground and burned area

in dark purple The grayscale visualization of the

gamma0 of the VH polarization also can be displayed.

The values for the cross polarization (VH) are

generally lower (darker visualization) than for the co-

polarization (HH, VV). The VH polarization has

higher values for surfaces characterized by volume

scattering, e.g., branches, dry coil bodies, or canopies

(lighter color in the visualization) and lower for

surfaces

Figure 7: Visualisation of Sentinel-1 SAR false colour

composite VV:VH:VH/VV enhance the burned area

features and grayscale VH.

3.3 Tracking Burned Area Progression by

Combination of Optical and SAR

Taking

advantage of the equivalent spatial and

temporal resolutions of radar and optical information

is facilitated by the availability of near-concurrent

active (Sentinel-1) and passive (Sentinel-2) datasets.

Figure 8 shows the comparison of trends of SAR false

colour composite VV:VH:VH/VV and SWIR colour

composite in monitoring forest fire progression

within four month periods. Sentinel-2 images were

matched to the Sentinel-1 dates for each detection

period as follows when there was not any temporally

coincident image: for the pre-fire date, the closest

Sentinel-2 image acquired before was selected. Based

on observation, SWIR colour composite shows

higher percentage of cloud cover that distract burned

area delineation. While the delineation of burned

areas using SAR images can provide more

information on the area due to cloud penetration but

still limited for some landscapes, since the radar

signal reflected from the burned surface may be

similar in intensity to the signal from other

components of the landscape for example, areas of

open ground.

NDMI

NBR

Tracking the Progression of Burned Areas in Tropical Peat Swamp Forests by Integrating Sentinel Optical and SAR Imagery: A Case Study

of Binsuluk Forest Reserve in Sabah, Malaysia

269

Figure 8: Comparing Burn Area Propagation visualized by SAR and SWIR false colour composite within January-April 2024.

3.4 Incorporation of Burned Area Map

into WebGIS Database

The validated burned area maps will be used in

iForSABAH system which is an innovative

application webGIS system that aims to address these

challenges by using space technology and remote

sensing to monitor the forest areas in Sabah,

Malaysia. iForSABAH stands for Integrated Forest

Resource Information System for Sabah, and it is a

collaboration between the Malaysian Space Agency

(MYSA) and the Sabah Forestry Department. The

system uses Geographic Information System (GIS)

technology and high-resolution satellite images to

detect any changes in permanent forest reserves, and

provide information on forest cover, forest type,

forest degradation, forest fire, forest restoration,

forest carbon stock, and forest biodiversity.

The system also supports the empowerment of

indigenous peoples and local communities who

depend on the forest resources for their livelihoods.

iForSABAH is a cutting-edge solution that leverages

the power of space technology to enhance the

management of forest resources, support the

implementation of the Sabah Forest Management

Plan, and contribute to the national and international

commitments on forest conservation and climate

change mitigation. By using iForSABAH, users can

access reliable, timely, and accurate information on

the status and trends of the forest areas in Sabah, and

make informed decisions for the benefit of the

environment and the society.

4 CONCLUSION

In conclusion, the combining of Sentinel-1 and

Sentinel-2 facilitates a higher frequency of data

collecting and enhances the detection of burned areas,

particularly in regions with significant cloud cover.

The Sentinel-2 SWIR and NIR bands have

demonstrated effectiveness in defining burned areas,

in conjunction with the temporal backscatter patterns

using Sentinel-1 SAR for forest fire propagation

analysis. The multi-date of prefire and postfire data are

crucial due to the characteristics of burned areas,

which are often linked to agricultural clearance

activities that precede the fire event, potentially

resulting in misclassification between burned land and

cleared land. Future research may demonstrate that

NBR and NDMI differencing can yield a more

effective methodology. Additionally, noise reduction

techniques, such as cloud masking, are essential for

improving detection outcomes. Information regarding

the progression of burned areas, including segmented

areas and hectares, can also be obtained from the final

map.

ACKNOWLEDGEMENT

The authors would like to thank the Malaysian Space

Agency (MYSA) and Sabah Forestry Department for

supporting and facilitating this research which part of

the collaboration project IForSabah and Forest Fire

GISTAM 2025 - 11th International Conference on Geographical Information Systems Theory, Applications and Management

270

Information System (ForFIS). We acknowledge the

use of data and/or imagery from NASA's Fire

Information for Resource Management System

(FIRMS) (https://earthdata.nasa.gov/firms), part of

NASA's Earth Observing System Data and

Information System (EOSDIS) and The Copernicus

Data Space Ecosystem Browser.

REFERENCES

Bourgeau-Chavez, L. L., Kasischke, E. S., Brunzell, S.,

Mudd, J. P., & Tukman, M. (2002). Mapping fire scars

in global boreal forests using imaging radar data.

International Journal of Remote Sensing, 23(20), 4211-

4234.

Gaveau, D., Descals, A., Salim, M., Sheil, D., & Sloan, S.

(2021). Refined burned-area mapping protocol using

Sentinel-2 data increases estimate of 2019 Indonesian

burning. Earth System Science Data Discussions, 2021,

1-23.

Imperatore, P. et al. Effect of the Vegetation Fire on

Backscattering: An Investigation Based on Sentinel-1

Observations. IEEE J. Sel. Top. Appl. Earth Obs.

Remote. Sens. 10, 4478–4492, https://doi.org/10.1109/

JSTARS.2017.2717039 (2017).

Kamlun, K. U., & Phua, M. H. (2024). Anthropogenic

influences on deforestation of a peat swamp forest in

Northern Borneo using remote sensing and GIS. Forest

Systems, 33(1), eSC02-eSC02.

M. A. Tanase, F. Pèrez-Cabello, J. de la Riva, and M.

Santoro, “TerraSAR-X data for burn severity evaluation

in Mediterranean forests on sloped terrain,” IEEE Trans.

Geosci. Remote Sens., vol. 48, no. 2, pp. 917–929, Feb.

2010

Melling, L. (2016). Peatland in Malaysia. Tropical Peatland

Ecosystems. Springer

Phua, M. H., Tsuyuki, S., Lee, J. S., & Sasakawa, H. (2007).

Detection of burned peat swamp forest in a

heterogeneous tropical landscape: A case study of the

Klias Peninsula, Sabah, Malaysia. Landscape and urban

planning, 82(3), 103-116.

Rein, G., & Huang, X. (2021). Smouldering wildfires in

peatlands, forests and the arctic: Challenges and

perspectives. Current Opinion in Environmental

Science & Health, 24, 100296.

Sirin, A., & Medvedeva, M. (2022, January 2). Remote

sensing mapping of peat-fire-burnt areas: Identification

among other wildfires. MDPI. https://www.mdpi.com/

2072-4292/14/1/194

Sloan, S., Tacconi, L., and Cattau, M.: Fire prevention in

managed landscapes: Recent success and challenges in

Indonesia, Mitig. Adapt. Strat. Gl., 26, 32,

https://doi.org/10.1007/s11027-021-09965-2, 2021

Suwanprasit, Chanida, and Shahnawaz. "Mapping burned

areas in Thailand using Sentinel-2 imagery and OBIA

techniques." Scientific Reports 14.1 (2024): 9609.

Tanase, Mihai A., et al. "Sensitivity of X-, C-, and L-band

SAR backscatter to burn severity in Mediterranean pine

forests." IEEE Transactions on Geoscience and Remote

Sensing 48.10 (2010): 3663-3675.

Tracking the Progression of Burned Areas in Tropical Peat Swamp Forests by Integrating Sentinel Optical and SAR Imagery: A Case Study

of Binsuluk Forest Reserve in Sabah, Malaysia

271