7

I am interested in mapping vegetation stress utilizing the blue shift to the red edge phenomenon in remote sensing. The figure below shows the expected change in reflectance of green leaves under stress. I would ideally like to produce a raster-based change map utilizing commonly available multispectral imagery (e.g. 4-band NAIP, Landsat, etc). Are there any well developed methods to put this phenomenon into practice within a GIS or digital image processing software? Any step-wise advice to accomplish this would be of great help.

More details:

enter image description here

|improve this question
  • Do you mean in addition to the (often abused) NDVI and FAPAR? – whuber♦ Apr 24 '13 at 20:28
  • 1
    @whuber I'm especially curious about the red-edge band (~690-720nm) and what methods are being used to characterize veg stress utilizing this wavelength (e.g. Time series analysis of red-edge inflection points (REIP)). I know WorldView-2 has a Red-edge band and we are likely to see a lot more research into veg stress utilizing this band. – Aaron♦ Apr 24 '13 at 21:04
1

How is the stress expressed in your data? Is it over time or spatially? If it's over time then a simple raster calculator algorithm should work, if spatially then there is a kernel tool that would allow a moving window matrix operation over the data. Both would require band ratios and thresholds to be calculated but all this is standard RS and can be done in Arc (if you insist), ENVI, and even in QGIS, though the kernel function may not be available as a standard tool (I haven't found it myself yet).

|improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.