I'm recently doing the research on the impacts of smouldering in the arctic, but due to the travel restriction, I'm doing this remotely. My current issue is to identify the smouldering from satellite data, distinguish it from general wildfires. Does anyone know any method which can allow me doing this?
Use any platform with thermal channels and enough spatial resolution.
Worth trying no?
Cheers,
Frank
Well, perhaps you could use the wildfire data to establish the candidate areas in your region of interest, when the start and stop. You could then extend a buffer around those areas, and then collect the signatures for cells n those areas before the fire started, over say, for each of the 12 months pre-fire. You could then compare those by the 12 corresponding months post fire signatures to detect any smoke effects from the smouldering. And maybe use any wind direction information at the sample areas at those times to confirm the plume drift. Seasonal variations are going to add some complexity, and it would be helpful if you had a known test case to confirm the analysis process. https://www.nasa.gov/image-feature/goddard/2020/nasas-aqua-satellite-shows-siberian-fires-filling-skies-with-smoke/
There are a few platforms with IR sensors that have data available from night time passes that might be leveraged also, like Aster, which could perhaps detect the much lower heat output than an active fire: Article The application of day and night time ASTER satellite imager...
You might want to try a principal component analysis of the derivatives of the radiation spectrum and the total energy detected to look for differences. Good Luck
Preprint A Model of Color Vision for Remote Sensing 1
10.1109/ACCESS.2021.3053605
Article Bio-Inspired Approaches for Energy-Efficient Localization an...
- Badges
- Science topic
- Similar topics
- Geoscience
- Remote Sensing