Running spiceinit on Galileo NIMS cubes in ISIS

[cmelda109]

Hi, I want to apply photometric corrections using the photomet function on Galileo NIMS cubes. To do that, I need to run spiceinit first, but I keep encountering this error (see image). I have downloaded both the base and Galileo data. I tried doing the same for SSI, and it worked. Any ideas on how to fix this? Thank you in advance for any help.

I am using data from: [USGS | PDS 1]

[lwellerastro]

H @cmelda109, the file you appear to be trying to work with (I believe it's this one, hard to see in your figure) is not a raw data file that can be ingested into ISIS or run through spiceinit which is why you are encountering that error. The aaredme.txt and volinfo.txt have some additional information that might be (somewhat) helpful if you haven't looked already. Maybe more useful is a readme.pdf I came across.

In addition to the ci_spc.cub you have downloaded, you might also download the ci_loc.cub file and try some of the recommended processing via the gllnims2isis documentation. Since you do not have a raw edr file to work with (*.qub), there is no reason to run the program gllnims2isis, but the information that follows in regard to running cubatt and nocam2map might be useful. I think with the available data the file referred to as core.cub is equivalent to the ci_spc.cub, and suffix.cub is equivalent to the ci_loc.cub.

The current ISIS cube labels on both the image and geometry backplane file does not have all of the relevant Mapping information you will need to try the steps on the gllnims2isis documentation page. Run the isis program origcatlab on either of these *.cub files to get more information, including Group = IMAGE_MAP_PROJECTION which contains the information necessary to run nocam2map. You should have a look at the parameters for nocam2map and what is supplied for the mapping information and be as explicit as possible. The output image will have latitude/longitude information (which can be verified via qview or other programs). I'm honestly not sure everything you need to run photomet will be available on the label though.

Good luck!

[cmelda109]

Hi, thank you so much for your reply. I was able to reproduce all the steps using the cubeatt and nocamp2map functions without any problems. Then, I wanted to use the photomet function. Initially, I tried running the function with ANGLESOURCE: get photometric angles from ELLIPSOID, but it always returned an error stating that some kernels were missing (which is why I thought I needed to run spiceinit first).

However, after manually entering the phase, emission, and incidence angles, it worked. Unfortunately, as I was experimenting with different parameters for the normalization model, the process unexpectedly crashed and started returning the error: segmentation fault (core dumped). I didn't update anything or do anything differently before the crash. I am using Ubuntu 18. Do you have any idea what could have gone wrong?

[lwellerastro]

Hi @cmelda109, could you be more specific about your processing so that we can try and reproduce the error on our end?

Please include the version of isis you are using and precisely what isis commands and settings you used on the downloaded NIMs cubes. Be sure to include everything you passed to photomet including the contents of FROMPVL if you used that parameter and what Models you selected for processing. We might need to start a new post for this or change the title of this one - this is not a spiceinit failure as that program is not expected to work on the type of data you are working with.

I don't believe photomet will work on your processed image as is and the problem is not being handled properly hence the seg fault.

Although it may contain latitude and longitude information after the nocam2map, it probably doesn't have any information in regard to illumination geometry that photomet needs (you can confirm this by opening the image in qnet and using the tracking tool to see if illumination values are displayed when you move cursor over the image). The loc.cub file you have has other bands/backplanes that does contain that information and could be used as input via anglesource=BACKPLANE, but post #5281 in the ticketing system reports that this functionality isn't currently working under the latest versions of isis.

Devs looking at this post can find data in my work user area under /work/users/lweller/Isis3Tests/GLL_NIMS/. These were my processing steps:

cubeatt from=g8g003ci_loc.cub+1 to=latitude.cub
cubeatt from=g8g003ci_loc.cub+2 to=longitude.cub
nocam2map from=g8g003ci_spc.cub.cub latcube=latitude.cub loncube=longitude.cub londir=positivewest to=projected.cub pixres=ppd resolution=13.650 eqradius=2634 polradius=2634

Photomet can be attempted on projected.cub when OP confirms their processing steps and provides more detail on settings that prompted the failure.