Problem using l2gen to produce VIIRS NIR Rrs

[blesht]

Hi - I am trying to generate VIIRS L2 files using l2gen.  I'd like to add the NIR bands to the standard OC suite of products.  I've been testing the procedure using the GUI.  I can generate the default L2 file with no difficulty.  However, when I used the "Products" tab to include the NIR bands, the output L2 file does not include these bands; it is identical to the "default."  In both cases the "Selected Products" text box shows "Rrs_vvv."    I don't often use the GUI so it is very possible that I'm doing something incorrectly, but this should be pretty simple, right?

Thanks, Barry

[OB.DAAC - SeanBailey]

Barry,

You can't get Rrs for the NIR bands, as these are used in the aerosol selection component of the atmospheric correction process. The assumption is these bands have no water-leaving component (the black pixel assumption).  With the "NIR correction" applied, a model is used to estimate any non-negligible NIR water-leaving contributions to the signal, but as these are not measured data, they are not reported.

Sean

[blesht]

Hi Sean - Hmmm.   Interesting.  I was trying to reproduce a recently published (Yu et al., RSE 2019, 111491) algorithm for SPM that uses VIIRS Rrs_745 and Rrs_862.  In the paper, the authors mention using the combined NIR/SWIR atmospheric correction so I set that option to -9 in my run.  Although they say something about a "pre-computed" turbid water index to indicate where the SWIR correction can be applied, they say nothing about not having Rrs_745 and Rrs_862 elsewhere.  Looking at the paper more closely, I now see that they processed using "NOAA Multi-Sensor Level-1 to Level-2 (MSL12) ocean color processing system."  I guess I was focused on the "MSL12" (which I always think of as synonymous with l2gen) and didn't appreciate that they didn't use SeaDAS.  

Thanks for the explanation.

Barry

[OB.DAAC - SeanBailey]

Ah, well, if you use the SWIR channels for the atmospheric correction, then the black pixel assumption applies to those bands, not the NIR bands.  Doing so will allow Rrs for the NIR bands.
The GUI might be restricting you.  but I was able to select the bands (hit the apply button) and the parameter file showed the bands...

Sean

[blesht]

I'm still not getting it.  Am I right that selecting aer_opt = -9 will get me the SWIR correction?  When I look at the saved parameter file I still get "OC" as the suite and don't see any additional bands.  I hate to trouble you with this.  I'll see if I can get it sorted out myself.   Barry

[OB.DAAC - SeanBailey]

aer_opt=-9 gets you the Shi and Wang NIR/SWIR switching algorithm, which uses a turbidity index to switch between using the NIR bands and the SWIR bands in the aerosol retrieval.
By default, the suite will be OC.  You can remove that line.  Or leave it, as anything in the par file that differs from what is in the OC suite defaults file take precedence.
You won't see any additional bands unless you specifically added them (using the Products tab in the GUI, or manually entering l2prod="the products I care about")

Sean

[blesht]

Thanks - I thought I did the band selection in GUI correctly, but maybe not.  I'll give it another go.   I'll take it from here - maybe go back to my more familiar command line invocation of l2gen.  Have a good weekend.   Barry