Jump to content
PDS Geosciences Node Community
June Wang

Questions about CRISM MTRDR Data.

Recommended Posts

User Bobo Xi
August 30, 2017 9:06 AM


Dear Prof.

I'm a postgraduate in xidian university,china. I'm research in the hyperspectral imge of mars at present and have downloaded the CRISM data on the PDS Geosciences Node. The datatype of the data is MTRDR and  ID is “frt0000474a_07_if164j_mtr3”. Then,I have two problems with it :

1.First, I have read the "CRISM Data Product Specification", it said that the TER datas is "A spatially reconciled, full spectral range 

I/F targeted observation central scan image cube in the IR (L-detector) sensor space that has been corrected for geometric, photometric, atmospheric, and instrumental effects." and "MTRDRs are map-projected versions of TER data products." Then,I want to confim that does that mean that the MTRDR datas have calibrated for geometric, photometric, atmospheric,  instrumental effects and map-projected ? 

2. Second,the attachment of this email is the Browse Image of the MTRDR data  I have dowmloaded . My question is that what is the meaning 

of the curve below ?In addition, I have open the MTRDR data by envi5.1 and why the spectral curve almost stay the same for each pixel in the image ?

I'm feeling very confused and looking forward to your reply!

Best regards!

 frt0000474a_07_if164j_mtr3_composite.png
 

 
User Seelos, Frank P.
August 30, 2017 1:38 PM
 


> I want to confim that does that mean that the MTRDR datas have calibrated for geometric, photometric, atmospheric,  instrumental effects and map-projected?

This is correct. The only structural difference between the sensor space TER spectral image cube (the IF product) and the corresponding map-projected MTR spectral image cube is that radiometrically questionable bands have been removed from the MTR. The TER image cube has 545 bands and the MTR has 489.

> Browse Image of the MTRDR ... My question is that what is the meaning of the curve below

The plot shows the scene median and [5, 95] percentile envelope spectra.

> the spectral curve almost stay the same for each pixel in the image

The spectral variability in this scene is subdued by a relatively dusty atmosphere. The TER/MTRDR processing accounts for atmospheric gas absorptions and the relative geometric/photometric effects of aerosol scattering, but the data are top-of-atmosphere corrected I/F, not model ground reflectance.

User Bobo Xi
August 30, 2017 11:14 PM


Hello,

Thanks for your reply. but exuse me, I'm still not clearly understand my last question. In other words,if I want to make mineral classification in the image by using the material's spectral variability, does that mean that I have to use the TRDR datas and calibrate them myself by using some ground reflectance model rather than directly using the MTRDRs?

User Seelos, Frank P
2017-09-01 03:40:48


If you’re interested in quantitative mineral abundance modeling then the retrieval of surface spectral reflectance through some type of surface/atmosphere separation or atmospheric correction will be required – either the TRDRs or the TERs/MTRDRs could be the starting point in this case. However, mineral identification, classification, and investigation of scene spectral variability are definitely possible with corrected I/F data – the majority of the CRISM-derived results in the literature used this type of workflow.

Thanks,

-Frank

User Bobo Xi
September 03, 2017 9:13 PM


Sorry, I have been busy with something else and forgotten to write back. After seeing your reply I have totally understood, thank you very much!

 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×