The forum is locked.
The Ocean Color Forum has transitioned over to the Earthdata Forum (https://forum.earthdata.nasa.gov/). The information existing below will be retained for historical reference. Please sign into the Earthdata Forum for active user support.
Which geographical coordinate system is used for the LEVEL 2 DATA of MODIS aqua? Is it same for ocean colour data from all sensors? I am trying to correlate reflectance data from MODIS aqua and in-situ chlorophyll-a data in the Chesapeake Bay. Chesapeake Bay program uses the NAD83 coordinate system. Do I need to reproject MODIS reflectance data?
Florida International University
Our level-2 data sets have not been projected to any particular coordinate system.
Rather, the data are stored roughly in the order in which they were collected by
the various instruments such as MODIS. To work with these data you will need
to read the latitude and longitude data sets that are stored in the same file (one
coordinate pair marking the center of each pixel). Our latitudes and longitudes
are based on GPS measurements and reference the WGS84 datum.
If you use the SeaDAS software that we distribute (https://seadas.gsfc.nasa.gov/),
this is all handled automatically.
Note that MODIS pixels are ~1 kilometer wide and larger (or 0.25 km at best if you
are using certain bands), so differences between WGS84 and NAD83 are probably not
Thank you for your explanatory reply. It helped me a lot.
I have another question related to this topic. As I mentioned, I am interested in correlating satellite reflectance data with the in-situ measurement of chlorophyll-a. If in situ measurement is taken at three depth(at 0.5 m from the top, 1 m from top and 1 m from bottom), which of the following will be a better way of addressing the scenario?
1. Compare chlorophyll concentration of the top layer(0.5m) with reflectance data
2. compare MEAN of the measurements at three-layers with reflectance data
Our satellite ocean-color sensors "see" different amounts of light reflected
from the water column depending on the optical characteristics of the water
(turbidity, presence of chlorophyll, CDOM, etc.). The number of photons
returned from the water column decreases exponentially as depth increases.
Also note that your in situ measurements are collected in a very small volume
of water while the satellite senses a square kilometer or more of ocean which
is often quite spatially variable.
Given the above, I would not use your meter-from-the-bottom measurement.
Our algorithms are designed to work in optically deep waters where no photons
are coming from the bottom, so either your near-bottom measurement is not
contributing to the satellite signal, or the area is too shallow for our standard
algorithms to work because of interference from bottom reflections.
The difference between the top two measurements won't be significant unless
you are in high chlorophyll or turbid waters. In the first case spatial variability
will probably swamp any difference in the two measurements; in the second case,
our standard algorithms do not work well.
There are plenty of other issues to be aware of when comparing in situ and satellite
o What is the accuracy of the in situ values themselves?
o How much time separates the satellite and in situ measurement?
o How close are you to a coastline (stray light from land becomes an issue)?
o What is the state of the overlying atmosphere (aerosols, partial clouds)?
... et cetera.
This is not to discourage you from comparing satellite and in situ values; such work
is important, and we of the Ocean Biology Processing Group spend a lot of effort on
just such activities. It's important that you are aware of the challenges involved,
When deciding how to average multiple near-surface in situ measurements for satellite match-ups, generally speaking, you should calculate an optically-weighted average rather than picking a single value or taking the simple mean. This is because the NASA OC Chl algorithms (in principle) return weighted averages from the first optical (attenuation) depth. In your particular case (i.e., just two measurements in total, from 0.5 and 1.0 m), weighting probably won't make a big difference. However, it is important when dealing with more varied depths or greater numbers of measurements.
Optical weighting requires additional information about the underwater light field (e.g., Kd measurements, or estimates, or else there are other approaches) so you can calculate which and how much the various measurements contribute to the signal detected by the satellite. For more details on the weighting approach and algorithm development data sets, see Gordon and Clark 1980 (Appl. Opt. 19, 3428-3430) and Werdell and Bailey 2005 (Rem. Sens. Environ. 98, 122-140), or other weighting approaches like Zaneveld, Barnard, and Boss 2005 (OSA, Vol 13 (22), 9052-61).
Powered by mwForum 2.29.7 © 1999-2015 Markus Wichitill