START Conference Manager    

Examining the applicability of Sentinel-1 data for identifying fires and burn scars on the Irish Landscape

Calvin McCoy

(Submission #166)


Abstract

In Ireland, fires and fire-related damage place considerable pressure on fire services and local authority resources. They also have numerous detrimental impacts on the environment. Fires in Ireland rarely occur spontaneously or due solely to weather conditions. Rather, they are primarily the result of either unforeseen or careless human actions. Using remote sensing resources and techniques to map such fires and the scarring on the landscape produced by them is an important endeavour in trying to better manage institutional and local responses. However, such endeavours pose a particular challenge in the Irish context. Not only are many of these fire events relatively small in size, but using optical remote sensing techniques to identify and map them is further complicated by the extent of cloud cover over the country, which regularly obscures many of the affected areas over long periods of time.

SAR remote sensing technology for fire and burn scar detection is a relatively underexplored area of study. However, the potential benefits of SAR in this area are significant, given the penetration of clouds by microwave signals, and the ability to capture data during both the day and night. Furthermore, some success has recently been achieved in using SAR data for identifying burn scars on the landscape.

This presentation will focus on some preliminary results obtained by utilising and analysing Sentinel-1 data to identify burn scarring in Ireland. A series of SLC images have been processed using the SNAP toolbox and such metrics as intensity, coherence, polarimetric decomposition and texture analysis have been considered to see whether areas scarred or affected by fires can be identified. While the results may be far more complex than typical optical data, various burn scars have been identified. This points towards the potential usefulness of SAR Sentinel-1 data for fire and burn scar detection, and suggests a need to expand and refine the analysis of SAR data in this area.

Sentinel 1 data is INSPIRE compliant. The products produced in relation to fires and burn scarring could be used as ancillary data sources to assist with datasets used in INSPIRE reporting under the various required themes.

Categories

Topic Area:  [1.2] Environmental monitoring and assessments
Abstract Type:  Oral Presentation

START Conference Manager (V2.61.0 - Rev. 4840)