Earthdata Forum

Hii :)

I am analysing MODIS-Aqua derived OC and SST parameters for the Cape Verde region. I applied statistical analyses to see the correlation between some of these parameters and I realized that if I use Kd490 as independent variable and PIC as dependent variable (or vice versa), there is no significant correlation among the two variables. This is strange to me because if there was a major coccolithophores bloom in some regions, I would expect the Kd value to lower accordingly. I have a high correlation between Chl a and Kd and this is expected. But, PIC has no relation with Kd490 at all? Why?

Thank you

Hey,

If you came across data related questions or issues to report, our experts will be able to help; but for data analysis related scientific questions for a particular region, you got to figure it out by yourself.

There may be several reasons for a lack of correlation. The empirical algorithm for Kd490 was not developed with data that includes high concentrations of PIC (i.e. coccolithophore blooms). The conditions present during an intense coccolithophore bloom likely violate the assumptions used in the development of the Kd490 algorithm. If you are doing your analysis on Level 3 data, intense coccolithophore blooms are excluded from the binning of the Kd490 product (well, alot of reflectance based products for the same reasons).

Hope this helps,
Sean