Earthdata Forum

We have been having difficulty generating the OWTD (Moore Optical Water Type) product using SeaDAS and MODIS/Aqua. Since (at least) Seadas-7.5, the product has been coming out with all (non-flagged) pixels with value (water type) 5. We have also tried with SeaDAS 8.1, to no avail.

Is this product still supported? Is there a known bug or development underway? I have searched both this and the old obpg forum without finding any hints. Thankyou for any advice.

Supported? Hmmm..well, it's not not supported

It's not a standard product, so we don't generally pay attention to it. No substantial changes have been made to the algorithm since it was implemented...although it has changed. Perhaps one of those "inconsequential" changes wasn't so inconsequential. I can confirm that even for a SeaWiFS file, the output is overwhelmingly dominated by OWT #5. Had to zoom in to see some speckles that weren't #5....so it's not that ALL pixels are #5, but likely way too many.

So, yes there may be an issue. No idea what that could be, but we'll add it to the things to investigate.

Sean

Thanks Sean. We looked at the C code and there was at least one change to the above/below water reflectance correction that seemed incomprehensible (to us). We have our own Python-based implementation of the algorithm now which works well for both MODIS and VIIRS so we don't have a pressing need for a fix in SeaDAS. I was raising it to see if there might be one coming - but also to make sure it was recorded somewhere as not working.

The Rrs above to below calcluation is based on the Lee approach as used in QAA:
rrs(λ) = Rrs(λ)/(0.52 + 1.7 Rrs(λ))

That likely isn't the issue...but not going to say it definitely isn't

Now that we are aware that there is an issue (thanks!) it will (eventually) be fixed.

Regards,
Sean