I was wondering if I can get some help on the following issue:
We are trying to use Aqua-MODIS Rhos at 250m over the Chesapeake Bay for deriving algal bloom and hypoxia base on an AI algorithm. Because of the cloud cover a lot of pixels are unusable when colocating with in situ data. Therefore, we want to composite Rhos for 7 days. Even then we may not have sufficient valid data. My question is, has anyone used composite of Terra and Aqua MODIS data - or use Terra MODIS data to fill missing Aqua MODIS images/data?. We realize that the time of day for observation/sun angle are different but in a composite is it OK to use both Terra and Aqua MODIS Rhos without introducing large uncertainties?
- Subject Matter Expert
- Posts: 142
- Joined: Tue Feb 09, 2021 8:19 am America/New_York
I have found that rhos can differ substantially from one day to the next largely because of sun angle and scan angle effects. You can get a sense of this by viewing a what are essentially rhos-based images in Worldview. Here is a comparison of the Chesapeake Bay as seen by Aqua/MODIS on January 12 and 13, 2021.
A comparison between the 13th and 14th is here.
The above three days are in winter. The closer you get to the summer solstice, the more you will also see differences due to sun glint (which very much depends on sun angle and scan angle) from one day to the next.
If you throw Terra/MODIS into the mix, you also had calibration differences to your composite.
Proceed with caution.