Wildly different reflectance values for TM

Prev Go to previous topic

Go to next topic

Last Post 17 Sep 2015 05:20 AM by anon

Wildly different reflectance values for TM - with and without QUAC

3 Replies

Sort:


You are not authorized to post a reply.

Author

Messages

anon

New Member

Posts:16

17 Sep 2015 05:20 AM
Hello, I wanted to share a screen snapshot of this. I have my original TM image as downloaded from glovis.usgs.gov as the bottom image. The values 22, 23, 55 are in DN 0-255. The middle image is the TM image calculated to reflectance values without having been corrected using the QUAC tool. I just took the DN image and used the radiometric calibration. I understand these values are in proportion 0-1 reflected. The top image is where I applied the QUAC tool to the image (using the settings for TM). The output values I understand are integer reflectance scaled to 10,000. My question is why are the values between the middle and top image SO different? For example 0.4% versus 5.9% for the same pixel for band three in each image? Maybe that is truly what the atmosphere does to the reflectance values of imagery. It just seems such a huge difference.

Deleted User

Basic Member

Posts:228

17 Sep 2015 12:18 PM
Surface reflectance can definitely be significantly different than top-of-atmospher reflectance. How different depends on the specifics of the atmosphere that is between the surface and the sensor. The fact that TOA and surface reflectance are usually significantly different is the reason that it is so important to perform an atmospheric correction before strictly interpreting pixel values, or using them in analyses where it is important to have true surface reflectance information. - Peg

Deleted User

New Member

Posts:16

17 Sep 2015 01:41 PM
So the image where I didn't use the QUAC tool would represent top of the atmosphere reflectance values? And the QUAC-processed image is surface? Thank you so much for the help.

Deleted User

Basic Member

Posts:228

21 Sep 2015 12:54 PM
If you used the Radiometric Calibration tool to calculate the first reflectance image, then that result would show Top-of-Atmosphere reflectance. That correction doesn't know about or correct for anything related to the atmosphere. QUAC is an atmospheric correction, so the results from it would show surface reflectance (reflectance without the atmosphere). Also, remember that the scale for reflectance values is 0 to 100. So a value of 0 is not really wildly different than a value of 5. They are both quite small reflectance values (i.e., indicating low reflectance, and a dark object). - Peg

You are not authorized to post a reply.