It sounds like you might have incorrect wavelength units. What units are the wavelengths? Can you provide a snippet of the header file that contains the wavelengths and/or the wavelength units? Versions prior to 5.0 assumed micrometers if no wavelength units were provided in the header. Providing the units should allow ENVI to convert as needed. Hope that helps Josh

I did specify the units when I set the wavelengths - I set it to nanometers. Below is the header file in its entirety: ---------------------------------------------- ENVI description = { Create Layer File Result [Wed Apr 09 23:33:27 2014]} samples = 7671 lines = 7841 bands = 11 header offset = 0 file type = ENVI Standard data type = 12 interleave = bsq sensor type = Unknown byte order = 0 map info = {UTM, 1.000, 1.000, 141285.000, 2196015.000, 3.0000000000e+001, 3.0000000000e+001, 19, North, WGS-84, units=Meters} coordinate system string = {PROJCS["UTM_Zone_19N",GEOGCS["GCS_WGS_1984",DATUM["D_WGS_1984",SPHEROID["WGS_1984",6378137.0,298.257223563]],PRIMEM["Greenwich",0.0],UNIT["Degree",0.0174532925199433]],PROJECTION["Transverse_Mercator"],PARAMETER["False_Easting",500000.0],PARAMETER["False_Northing",0.0],PARAMETER["Central_Meridian",-69.0],PARAMETER["Scale_Factor",0.9996],PARAMETER["Latitude_Of_Origin",0.0],UNIT["Meter",1.0]]} wavelength units = Nanometers band names = { Layer (Band 1:LC80080472014049LGN00_B1.TIF), Layer (Band 1:LC80080472014049LGN00_B2.TIF), Layer (Band 1:LC80080472014049LGN00_B3.TIF), Layer (Band 1:LC80080472014049LGN00_B4.TIF), Layer (Band 1:LC80080472014049LGN00_B5.TIF), Layer (Band 1:LC80080472014049LGN00_B6.TIF), Layer (Band 1:LC80080472014049LGN00_B7.TIF), Layer (Band 1:LC80080472014049LGN00_B8.TIF), Layer (Band 1:LC80080472014049LGN00_B9.TIF), Layer (Band 1:LC80080472014049LGN00_B10.TIF), Layer (Band 1:LC80080472014049LGN00_B11.TIF)} wavelength = { 440.000000, 480.000000, 560.000000, 655.000000, 865.000000, 1610.000000, 2200.000000, 590.000000, 1370.000000, 10895.000000, 12005.000000} ----------------------------------------------------

My guess is that two of your bands are outside the range 400nm to 2500nm (http://www.exelisvis.com/...tationindices.html). Try removing bands 10 and 11 from your band stack. Cheers Josh

What's the best way to do that? I can't seem to find any place that lets me remove bands.

Sorry about my choice of words there. You can't really remove bands from a file without actually re-creating the file. So you could just re-create the file, or create an ENVI metafile (http://www.exelisvis.com/...terData.html#Create) but select a spectral subset in the creation process. Anyway, I tried a test data set using only the first 9 of you 11 wavelengths and ENVI returned the same result that you were getting. So, I created another test data set with 11 bands, but substituted the last two wavelengths with 830nm and 660nm (from Landsat 5). The indices calculator worked and returned 4 indices, NDVI, SRI, EVI, & ARVI. All part of the broadband greenness. Unfortunately almost all of the contribution came from the 2 substituted wavelengths, with a minor contribution from the the blue band. I could be completely wrong but my guess is that the calculator has a predefined set of spectral wavelengths (which makes sense in a way), with a tolerance for wavelengths that are 'close by' that can be applied to a given index. Someone from Exelis might be able to shed more light on this, but it looks like for the moment you won't get anything out of the index calculator with that set of wavelengths. Good luck :) Cheers Josh

Wow, thank you for helping me troubleshoot this. I would not have expected that Landsat 8 data would be useless for this analysis. Especially with the 655 and 865 nm bands. Given the fact that the bands are so close to the ones you had success with, as well as the fact that ENVI is only looking at the center wavelength rather than the actual range - what's your opinion on tweaking the center frequency designation in the header file until the indices analysis runs. What's your opinion on the quality of the resulting analysis? Edit: I was able to keep Band 4 set to the original 655 nm. The only change needed was to redesignate Band 5 as centered at 850 nm.

Seeing as the intention of LS8 is to continue on from its predecessors, it wouldn't hurt to change the center wavelengths of bands 2,3,4,5,6,7 to the corresponding bands of Landsat 5/7 (ie 1,2,3,4,5,7). Those bands are spectrally quite similar. No analysis is being done in the index calculator, merely it returns a list of potentially useful indices that you might want to calculate. Or just eyeball the spectral ranges and determine the indices yourself, ie bands 5 & 4 are NIR & RED, therefore NDVI can be calculated as well as the simple veg ratio. EVI by incorporating the blue (band 2) etc. Use ENVI's band math to calculate your own ratios with other bands and determine their usefulness as vegetation descriptors. Hope that helps Josh