Basics for CRISM processing with ENVi 5.3

[Wladimir Acevedo]

Hi dear community,

I'm starting with the CRISM data processing with ENVI 5.3 + IDL + CAT 7.4 but for now i feel a few lost, altough i keep the adventure. In any case, i have a couple of questions (very basics, probably) for the TRDR images from the CRISM map data center.

1.- Why when i open the trr3 images with CAT it appears as a rectangle (img_1) and not compressed or deformed  like the images  that i've seen in papers or documents? additionally it opens inverted so i have to apply a 180° rotation.

2.- When i open the images the band 1 has a value of 65535.0 (img_2)! and that carries problems when i want to visualize the spectral plot. So, it is recommended to eliminar this band or i'm doing something wrong with the images treatment?

Any kind of help will be very useful, since i'm really new on this.

Regards,

-Wladimir

 

[Feng]

Hi Wladimir,

 

For your first question, the trr3 is not map projected. So you will see the rectangle instead of the ones you saw on the paper. To map project the trr3, you can use the CAT map projection function.

 

Or you can download the corresponding MTRDR (which is map projected TRDR) if it is available.

I don't understand " additionally it opens inverted so i have to apply a 180° rotation. ". I think the original one is correct and you don't need rotate 180 degree.

For your second question, the band 1 with value of 65535 is the non-data band. There are other bands which is bad band, these bad band data can not be used. For the bad band list you shall ignore these bands. For example, the following screenshot shows a symbol before the bad bands:

Please let me know if you have any other questions.

 

Thanks,

 

Feng

[Wladimir Acevedo]

Hi Feng,

First of all, thank you for your answer, it solved my first question. Nevertheless, i'm facing a trouble with the spectral profile view (img_1) due to a supposed high value of the data, so i have to manually change the Y axis range to see the spectra correctly (img_2). Why this happens?. because i'm understand that the value of 65535 is for the "wavelenghts" but it seems to interfere on de "data" values even being ignored oh the metadata of the image. Also i could not find a way no ignore some bands like you told me. Or am i misunderstanding?

Another question that i have is when a use a "browse product" (i.e. PHY: R=D2300, G=BD2210, B=BD1900), i check the spectral profile appearing like these (img_3) and i don't know how to put the values of the wavelenghts on the X axis to get a coherent profile.

If you could help me with these it will be really great.

Thank you,

-Wladimir
 

 

[Feng]

Hi Wladimir,

Your adjustment of the x and y in img_2 is correct. This is also what I did to ignore the non-data or bad bands. The number 65535 is also assigned to the bad bands. That's why you see the two straight lines between 2.5 to 3.0 um.

For the problem you mentioned in img_3, you can do the following step to make it work:

1, Add wavelength through Edit ENVI header

2, edit the wavelength of your selected band, for example for D2300: tyep in 2.300 then hit Enter, it will update D2300 wavelength

3, after update wavelength of your selected bands, plot the spectral profile, and change the X range:

 

Please let me know if you have other questions.

 

Thanks,

 

Feng

 

[Wladimir]

Hi Feng,

Do you know what this field named "filter" in header_form is used for?

 

Wladimir

[Feng]

Hi Wladimir,

The CAT_WAVELENGTH_FILTER in the header files is a numeric code from PDS  label (in PDS3 label:  "MRO WAVELENGTH_FILTER"). It is used when reading original CRISM data into CAT. When parsing PDS label,  the default wavelength array to apply to the data (i.e., the standard re-sampling wavelengths for the appropriate wavelength filter used for the observation). It is most likely a book-keeping value.

 

Thanks,

 

Feng

[Wladimir Acevedo]

Hi again Feng,

I have tried the method that you showed for the wavelenght problem, but when i display the results the whole new image appears in black (img_1) and i don't know how to manage this in order get a right image again. Is this method correct for a browse product image (i.e. R: OLINDEX3. G: D2300,  B: BD2500_2)? and how i know what it's the right wavelenght for any of these parameters?. I have used 1.0, 2.3 and 2.5 μm, respectively, based on yout example and looking on the Viviano-Beck et al., 2014 paper. It is right?. Also, when i display the new spectral plot it shows like img_2. What am i doing wrong?

In addition to this, i have a question about the atmospheric and photometric correction. Are the TRDR and MTRDR images already corrected?.

Thank you again,

-Wladimir

[Feng]

Hi Wladimir,

I processed the same data as you did and there is no problem at all (the black img_1 and img_2). Would you please redo it?

For your additional question, the answer from the team scientist is:

 

The user will need to look at both the product type and data type to figure out the state of the data – and there are also keywords in the label that indicate the corrections that have been applied. A quick summary –

 

 

IF TRR3 – top-of-atmosphere apparent spectral reflectance (no corrections)

AL TRR3 – surface spectral lambert albedo (photometric and atmospheric corrected via RT modeling)

IF TER3 – top-of-atmosphere corrected I/F – CAT-like PHT and ATM corrections (and some other steps) have been applied

IF MTR3 – map projected version of the corresponding TER

 

 

For these correction, you can use the ATP correction function in CAT.

 

Thanks,

 

Feng

[Wladimir Acevedo]

Hi Feng,

I have reprocessed the image and still happening the same, even with others products that i have tested too. I have noticed that the data ignore value has changed to -1.0000 but i can't reset the 65535.0 default value. I'll leave you the information of the metadata of the image in case you see something different from yours.

Thanks,

Wladimir

[Feng]

hi Wladimir,

 

I think the only difference between yours and mine is the data ignore value. Since you cannot change -1 to 65535. How about you directly open the .hdr file in NotePad and change -1 to 65535 then save the file? You need close the .img in ENVI otherwise, you are unable to save the .hdr file.

 

Thanks,

Feng