Hi Tyas, When you get negative or superpositive (>1) endmember fractions from unmixing, it usually means that your endmembers are not adequate to accurately model the scene. Maybe there is one or more endmember(s) missing from the unmixing analysis, or maybe the spectra used for some of your endmembers are not the best representations of those endmember materials. I know this is more than you asked for, but I'm going to give you a discussion on using contrained unmixing, because I think it might be useful for you: Using constrained unmixing is popular because it results in endmember fractions for each pixel that seem physically reasonable. In other words, a value of zero means that the endmember covers less than none of the pixel, and that a value of one means that the endmember covers more than all of the pixel. Versions of ENVI before 3.0 did not support constrained unmixing because the developers strongly believe that constrained unmixing is unwise. It forces your results to appear reasonable, even when the endmembers are not reasonable. For example, if you use a sine wave, or a completely random spectrum as an endmember, the resulting endmember fractions can still sum to one. If you always use unconstrained unmixing, then the endmember fractions will give you information about how good your endmembers are. In other words, if the endmember fractions are usually between 0 and 1 when the unmixing is not constrained, then the endmembers are probably appropriate for the scene. Otherwise, they're not. Due to popular demand ENVI 3.0 and later versions do allow constrained unmixing. Nevertheless, the ENVI developers recommend that customers avoid using constrained unmixing so that they will be better able to evaluate whether their chosen endmembers are appropriate for their scene. ENVI supplies many tools that can be used to determine reasonable endmembers for a given scene. If you are finding it difficult to get good unmixing results, perhaps because you can't settle on a good set of endmember spectra, then match filtering can be a better choice than unmixing. It does not require accurate information about all of the endmembers in the scene. It can give information about the abundance of one endmember without information about the other endmembers. It is considered to be more robust for this reason. I hope this information is helpful. Let me know if you have additional questions. Peg