The role of spectral vs spatial resolution of satellite data on the accuracy of mapping unburned vegetation within fire scar perimeters
A precise delineation of fire scar perimeters, including the unburned vegetation, is crucial in order to understand the processes that are linked with wildland fires and estimate their effects on landscape and post-fire vegetation recovery. The remote sensing technology is frequently employed in the...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-06-01
|
| Series: | Science of Remote Sensing |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666017225000471 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | A precise delineation of fire scar perimeters, including the unburned vegetation, is crucial in order to understand the processes that are linked with wildland fires and estimate their effects on landscape and post-fire vegetation recovery. The remote sensing technology is frequently employed in the detection and mapping of unburned patches within fire scar perimeters. This study aims to understand the importance of the spectral and spatial resolution of satellite data when mapping the unscorched vegetation islands within burned areas. To achieve this, satellite images at various spectral (VNIR, SWIR) and spatial resolutions (original and spatially degraded datasets up to 512 m) were acquired from LANDSAT, ASTER, and IKONOS satellites shortly after a fire.The supervised maximum likelihood classifier was applied in multiple resolution satellite images to classify them with the utmost accuracy possible. A total of 420 classifications were executed, encompassing various combinations of spectral and spatial characteristics. Linear regression models were employed to capture the relation between the accuracy of the classified images and the characteristics of the satellite images. Our research highlights several key findings including: (a) spectral information content appears to have a significant role when considering the full range of separability values, (b) in combinations featuring high separability values, the primary parameter that influences the accuracy of the mapping the unburned vegetation is the spatial resolution of the satellite images, and (c) there is a discernible disparity in the roles of the spectral and spatial resolution concerning the commission and omission errors of the unburned class. |
|---|---|
| ISSN: | 2666-0172 |