Atmospheric correction for NIR-Red bands of MODIS

[manzar1428]

Hello,
I want to utilize NIR-Red bands of MODIS to develop chl-a retrieving algorithm. From the paper G. Dall’Olmo et al. / Remote Sensing of Environment 96 (2005), I came to know that there is no atmospheric correction method for NIR-Red bands of MODIS. What could be done to obtain atmospherically correct reflectance data for NIR-Red band of MODIS. Kindly answer.

Thank you,
Mohd Manzar Abbas
Florida International University

[OB.DAAC - SeanBailey]

Mohd,
The bands *are* atmospherically corrected.
The default atmospheric correction approach uses the NIR bands (748 and 869nm for MODIS), and so we do not report Rrs for the bands used in the aerosol determination in our standard products.  However, you can process the data to return the Rrs values for the NIR bands.  Be aware, however, that if you use the iterative NIR correction when doing the aerosol selection, the Rrs values in the NIR bands will be model derived, not "measured".  The reason for this is that a model is used to estimate the water-leaving contribution to be removed to allow the black-pixel assumption of the GW atmospheric correction to work.

Sean

[manzar1428]

Thank you Sean for the reply. I understood that Rrs data for 748 band is not available. Could you kindly guide me about the following:

1. What is raw data (Also source) that I will use to obtain Atmospherically correct Rrs data at 748 band.
2. I want to derive  chlorophyll-a concentration using Rrs at 667 and 748. For this, I want to analyse performance of different atmospheric correction method. I am interested in trying NIR method and SWIR method. What platform I can use for the task. I use Seadas on window platform. Could it be done using Seadas?
3. I have noticed that default chl-a retrieving algo for MODIS do not work properly for chlorophyll-a concentration above 10mg/m3. This is the reason that I am trying NIR-Red based algo for concentration above 10mg/m3. Could you please suggest me any other approach.

Being a master student my questions might be basic. Kindly guide me.
Thank you,
Mohd

[OB.DAAC - SeanBailey]

The source data are available from our L1/2 browser.  For MODIS, you would need to process the L1A data to produce an L1B and geolocation file to input into l2gen for processing.   SeaDAS is the software of choice :grin:  However, we do not yet support data processing under the Windows platform... It is on the near horizon, but currently you cannot do the processing under Windows, you would need a Mac or Linux machine.

Sean

[manzar1428]

Thank you Sean.