Display of L3m files - statistical adjustments

[jcfischer]

Dear all,
I am working with Kd_490 level 3 mapped data and face the problem that my maps barely show the spatial differences of the (relatively low) differing values. The area of interest I am looking at basically looks "all the same".
Here the statistics of an example file:
Minimum: 0.0166
Maximum: 4.3290
Mean: 0.0253
Sigma: 0.0082
Median: 0.0242
Looking at the pixel value plot I only have a thin line at the very left side. How can I adjust that and make it look more like a Gaussian distribution? Ideally the mean would be the center of my plot and the very few outliers would not distort the whole graph.
Does anybody know a way to better visualize the differences in the Kd 490 value magnitudes?
Thank you very much for helping! Greetings, JC

[OB SeaDAS - knowles]

SeaDAS 7.5 will feature a major upgrade to the Statistics Tool with "better" default plotting behavior and more user controls.  Of course regarding defaults, "better" is in the eye of the beholder.  The current default shows "all" the data, which in your specific case, you don't want.

For now, if all you are talking about is the domain span of the x axis, just right click on the image and adjust it within the SeaDAS GUI.

If outliers are too high, you can always consider masking them.

It is not clear what you mean by "spatial differences" and it is hard to comment on a better way to visualize the differences without knowing the specifics of actually what you are trying to do.  If you are comparing two separate readings of Kd_490 from separate times or separate instruments, you can collocate them together, and do a scatterplot.   Math band can of course be used to compare data and then produce subsequent histogram plots or anomaly images. 

Danny

[jcfischer]

Dear Danny, thanks for the advice!
Good to know that the new version will upgrade the statistical options.
What I did to solve my problem: I used the color manager and adjusted the range by lowering the max from 6.0 to 0.5. That way the differences in the Kd 490 values got way better depicted. Is that what you also proposed? Or what do you mean by "adjust it within the SeaDAS GUI"?
What I meant with spatial differences was a clear display of for example a plume or front of turbid water in comparison to the surrounding waters within one file. What I am trying to do is to visualize monthly mean turbidity and finally compare this with annual means and other months. All advice is more than welcome.

Is there a way to detect turbidity fronts with SeaDAS?

Thanks for your help! Greetings, JC

[OB SeaDAS - knowles]

I believe most of the answers to your questions can be found in
The Case Study on SeaDAS Video tutorials which gwiii referred to in another of your question threads.  It compares a 3 month SST with an SST climatology.  Also watch the other videos.  They are helpful.

Yes, you can adjust the dynamic range of the color bar to target your specific range of interest.  This topic is touched upon in the video.  There is an internal SeaDAS help which discusses this in detail.

I was referring to the fact that you can mask out anomalous values for doing statistics and image display which seemed to be what you were looking for.  Again a topic of the videos.

Please also take a look at the slides here:
https://seadas.gsfc.nasa.gov/tutorials/presentations/

The slides are mostly self-explanatory.

[jcfischer]

Thank you very much for answering. I adjusted the color bar to display the data and therefore visualize the differences in a better way.
Thanks for the links to all those useful resources! Greetings, JC