Back to search

South Dakota State University

Using New and Long-Term Multi-Scale Remotely Sensed Data to Detect Recurrent Fires and Quantify Their Relationship to Land Cover/Use in Indonesian Peatlands

Abstract

dc:description.abstract

Indonesia has committed to reducing its greenhouse gases emissions by 29% (potentially up to 41% with international assistance) by 2030. Achieving those targets requires many efforts but, in particular, controlling the fire problem in Indonesia’s peatlands is paramount, since it is unlikely to diminish on its own in the coming decades. This study was conducted in Sumatra and Kalimantan peatlands in Indonesia. Four MODIS-derived products (MCD45A1 collection 5.1, MCD64A1 (collection 5.1 and 6), FireCCI51) were initially assessed to explore long-term fire frequency and land use/cover change relationships. The results indicated the product(s) could only detect half of the fires accurately. A further study was conducted using additional moderate spatial resolution data to compare two years of different severity (2014 and 2015) (Landsat, Sentinel 2, Sentinel 1, VIIRS 375 m). The results showed that MODIS BA products poorly discriminated small fires and failed to detect many burned areas due to persistent interference from clouds and smoke that often worsens as fire seasons progress. Although there are unique fire detection capabilities associated with each sensor (MODIS, VIIRS, Landsat, Sentinel 2, Sentinel 1), no single sensor was ideal for accurate detection of peatland fires under all conditions. Multisensor approaches could advance biomass-burning detection in peatlands, improving the accuracy and comprehensive coverage of burned area maps, thereby enabling better estimation of associated fire emissions. Despite missing many burned areas, MODIS BA (MCD64A1 C6) provides the best available data for evaluating longer term (2001-2018) associations between the frequency of fire occurrence and land use/cover change across large areas. Results showed that Sumatra and Kalimantan have both experienced frequent fires since 2001. Although extensive burning was present across the entire landscape, burning in peatlands was ~5- times more frequent and strongly associated with changes of forest to other land use/cover classes. If fire frequencies since 2001 remain unchanged, remnant peat swamp forests of Sumatra and Kalimantan will likely disappear over the next few decades. The findings reported in this dissertation provide critical insights for Indonesian stakeholders that can help them to minimize impacts of environmental change, manage ecological restoration efforts, and improve fire monitoring systems within Indonesia.

Degree

thesis:*
Name thesis:degree_name
Doctor of Philosophy (PhD)
Level thesis:degree_level
Dissertation - Open Access
Discipline thesis:degree_discipline
Geospatial Science and Engineering

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Vetrita, Yenni
Contributors dc:contributor
  • Xiaoyang Zhang
  • Mark Cochrane

Subjects

dc:subject × 11

Rights

dc:rights

Identifiers

dc:identifier.*
Repository record dc:identifier
https://openprairie.sdstate.edu/etd/5205

Chain of custody

source
Harvested from
South Dakota State University
Base URL
openprairie.sdstate.edu/do/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Vetrita, Yenni. Using New and Long-Term Multi-Scale Remotely Sensed Data to Detect Recurrent Fires and Quantify Their Relationship to Land Cover/Use in Indonesian Peatlands. Dissertation - Open Access thesis, https://openprairie.sdstate.edu/etd/5205