I'm trying to sort out the different chl products offered in the L3 mapped MODIS and VIIRS data and have a couple of questions:
1) Are these definitions correct?
chlor_a: ? chl-a calculated using Hu et al. (2012) doi:10.1029/2011JC007395 for lower values and OC3 (see O'Reilly et al. NASA Tech. Memo. 2000-206892) for higher values? (Only Hu et al. is cited within the data files but I see this description of the combined algorithm elsewhere.)
chl_ocx: ? chl calculated using the OC3 algorithm (see O'Reilly et al. NASA Tech. Memo. 2000-206892)?
chl_gsm: ? chl calculated using the GSM algorithm (see Maritorena et al. (2002) doi:10.1364/AO.41.002705)?
2) Do the chl_ocx and chl_gsm values represent chl a or total chl?
3) Are there any general conclusions about the differences between these 3 products and conditions under which one should be preferred or avoided? Given there is only an ATBD for chlor_a, and chl_ocx is incorporated into its calculation, do you in general recommend chlor_a over the others?
Thanks in advance for any feedback!
definitions of chl from MODIS and VIIRS
-
OB SeaBASS - cwproctor
- Subject Matter Expert
- Posts: 20
- Joined: Mon Feb 05, 2007 12:20 pm America/New_York
definitions of chl from MODIS and VIIRS
Regarding your questions:
1) Refer to the Chlor_a algorithm description for more details and reasoning, but yes, you mostly have the right definitions. The chlor_a product is made using a combination of 2 algorithms, the Hu Color Index (CI) and O’Reilly band ratio (OCx, i.e., OC3V for VIIRS and OC3M for MODIS). A slightly modified version of Hu et al. 2012 is used where those algorithms are blended between chlorophyll concentrations of 0.15 to 0.2 mg m^-3.
2) CI and OCx (and chlor_a) are empirical algorithms tuned to Chl a pigments, using the NOMAD data set. Where HPLC pigments were used, they are the sum of the Chl-a related pigments.
3) It depends on the nature of your study, but chlor_a is currently the best choice as a global algorithm; it performs well over a large dynamic range of chlorophyll concentrations. However, it is impacted by other optically active water constituents. For example, in waters with high CDM (usually CDOM) concentrations, it will likely yield an overestimation of Chl. Semi-analytic models (e.g., GSM as you mentioned) have the capability and advantage to retrieve IOPs separated from Chl, if properly parameterized.
Chris
1) Refer to the Chlor_a algorithm description for more details and reasoning, but yes, you mostly have the right definitions. The chlor_a product is made using a combination of 2 algorithms, the Hu Color Index (CI) and O’Reilly band ratio (OCx, i.e., OC3V for VIIRS and OC3M for MODIS). A slightly modified version of Hu et al. 2012 is used where those algorithms are blended between chlorophyll concentrations of 0.15 to 0.2 mg m^-3.
2) CI and OCx (and chlor_a) are empirical algorithms tuned to Chl a pigments, using the NOMAD data set. Where HPLC pigments were used, they are the sum of the Chl-a related pigments.
3) It depends on the nature of your study, but chlor_a is currently the best choice as a global algorithm; it performs well over a large dynamic range of chlorophyll concentrations. However, it is impacted by other optically active water constituents. For example, in waters with high CDM (usually CDOM) concentrations, it will likely yield an overestimation of Chl. Semi-analytic models (e.g., GSM as you mentioned) have the capability and advantage to retrieve IOPs separated from Chl, if properly parameterized.
Chris