atmospheric correction with episodic strongly absorbing aerosols

Use this Forum to find information on, or ask a question about, NASA Earth Science data.
Post Reply
cmitchell
Posts: 19
Joined: Tue Jun 06, 2017 11:42 am America/New_York
Answers: 0

atmospheric correction with episodic strongly absorbing aerosols

by cmitchell » Thu Dec 19, 2019 3:09 pm America/New_York

Hi,

I have a couple questions related to strongly absorbing aerosols:

1) I'm wondering if there are any recommendations on the best approach on how to deal with atmospheric correction in the case of episodic strongly absorbing aerosols e.g. large dust events. I know I can play around with the aer_opt options in l2gen, but what if the aerosols cover only part of the image and not the whole scene? My understanding is, as things stand currently, there is nothing we can really do about that i.e. if I select an aer_opt model (and don't use the default) it gets applied to the whole scene - correct?

2) Are there instructions anywhere on how to actually select the aerosol model I want (from the aermodels list)? Sean alludes to his method in the following: https://oceancolor.gsfc.nasa.gov/forum/oceancolor/topic_show.pl?pid=30090;hl=aermod - are there more details somewhere? Or is the model list somewhere?

Catherine

Tags:

OB.DAAC - SeanBailey
User Services
User Services
Posts: 1464
Joined: Wed Sep 18, 2019 6:15 pm America/New_York
Answers: 1
Been thanked: 5 times

atmospheric correction with episodic strongly absorbing aerosols

by OB.DAAC - SeanBailey » Thu Dec 19, 2019 7:42 pm America/New_York

Catherine,

Yes, setting the aer_opt defines that option for the entire l2gen run (i.e. the entire scene).

If you want to select a specific model (see attached for a listing that can be used),  aer_opt would be a positive integer between 0 and 80 OR you can set aer_opt=1 and aermodels to be  the single aerosol model name to use, e.g. aer_opt=1 aermodels=r70f10v01

Now for the bad news.

Currently, l2gen cannot handle strongly absorbing aerosols.  While there are 80 aerosol models used in the current implementation, they are all for non- or weakly absorbing aerosols.
Absorbing aerosols are not trivial to handle.  Even if the model set included  absorbing aerosols, selecting them from the non/weakly absorbing ones is not easy (with the current GW94 approach).
Absorbing aerosols also come in many species - even sub species (not all dust is the same), so there would need to be a lot of models to cover the various types.

Then there's the matter of the height of the absorbing aerosols - height of the aerosol layer becomes quite important if the aerosols are absorbing. Determining the height of the aerosol layer with the information available from the current suite of ocean color capable sensor is simply not possible...so an ancillary input would be required. 

Absorbing aerosols do not occur alone, there are going to be non/weakly absorbing aerosols mixed in.  How the absorbing to non-absorbing aerosols mix would need to be known or assumed.

Sean
attachment 1

cmitchell
Posts: 19
Joined: Tue Jun 06, 2017 11:42 am America/New_York
Answers: 0

atmospheric correction with episodic strongly absorbing aerosols

by cmitchell » Fri Dec 20, 2019 9:54 am America/New_York

If you want to select a specific model (see attached for a listing that can be used),  aer_opt would be a positive integer between 0 and 80 OR you can set aer_opt=1 and aermodels to be  the single aerosol model name to use, e.g. aer_opt=1 aermodels=r70f10v01

Great - thanks!

Currently, l2gen cannot handle strongly absorbing aerosols.  While there are 80 aerosol models used in the current implementation, they are all for non- or weakly absorbing aerosols.

Yea, I knew strongly absorbing aerosols were a problem. I guess I was hoping that one of the 80 models could be used...

Then there's the matter of the height of the absorbing aerosols - height of the aerosol layer becomes quite important if the aerosols are absorbing. Determining the height of the aerosol layer with the information available from the current suite of ocean color capable sensor is simply not possible...so an ancillary input would be required. 

So if I did have the height of the aerosol layer and some information about the aerosols (from e.g. MISR), would it be possible to incorporate that (somehow)?

OB.DAAC - SeanBailey
User Services
User Services
Posts: 1464
Joined: Wed Sep 18, 2019 6:15 pm America/New_York
Answers: 1
Been thanked: 5 times

atmospheric correction with episodic strongly absorbing aerosols

by OB.DAAC - SeanBailey » Fri Dec 20, 2019 10:04 am America/New_York

> So if I did have the height of the aerosol layer and some information about the aerosols (from e.g. MISR), would it be possible to incorporate that (somehow)?


That's the rub...the code has no way (yet) of making use of such information.  Knowing what we need is one thing, figuring out how to get what we need AND incorporate that knowledge in the processing is quite another.  I recall a SIMBIOS meeting in 1999 where this was discussed.  Sadly we (as a community) have not made much progress in 20 years...guess it's a hard problem...

Sean

OB.DAAC - SeanBailey
User Services
User Services
Posts: 1464
Joined: Wed Sep 18, 2019 6:15 pm America/New_York
Answers: 1
Been thanked: 5 times

atmospheric correction with episodic strongly absorbing aerosols

by OB.DAAC - SeanBailey » Fri Dec 20, 2019 10:18 am America/New_York

...on the bright side, with the multiangle polarimeters and hyperspectral radiometer that will fly on PACE, we might have a chance of cracking this nut soon :grin:

cmitchell
Posts: 19
Joined: Tue Jun 06, 2017 11:42 am America/New_York
Answers: 0

atmospheric correction with episodic strongly absorbing aerosols

by cmitchell » Fri Dec 20, 2019 10:23 am America/New_York

I had a feeling that would be your response. Until PACE then!

Post Reply