The Aqua satellite platform carries a MODIS sensor that observes sunlight reflected from within the ocean surface layer at multiple wavelengths. These multi-spectral measurements are used to infer the concentration of chlorophyll-a (Chl-a), most typically due to phytoplankton, present in the water.
The data are produced from the near real time (nrt) data stream formed by combining data from all the available direct broadcast reception stations in Australia (Alice Springs, Melbourne, Townsville, Perth, Hobart) supplemented by delayed-mode data from NASA in the US. The granules have been remapped from satellite projection into a geographic (Latitude/Longitude axes) projection and are presented as a sequence of daily mosaics covering the region (80 <= Longitude <= 180, -60 <= Latitude <= +10) formatted as CF-compliant netCDF files. It should be noted that the data are not processed until the definitive spacecraft ephemeris becomes available, usually 12-24 hours after the overpass. This means that the geolocation should be of a uniformly high standard.
There are multiple retrieval algorithms for estimating Chl-a. These data use the Garver-Siegel-Maritorena (GSM) method implemented in the SeaDAS processing software l2gen and described in “Chapter 11, and references therein, of IOCCG Report 5, 2006, (http://ioccg.org/wp-content/uploads/2015/10/ioccg-report-05.pdf). The radiometric sensitivity of the MODIS sensor is evolving continuously during its mission and is monitored regularly by NASA. The SeaDAS software uses tables of calibration coefficients that are updated periodically. From time to time upgrades to the algorithms and/or the format of the calibration tables are required, in which case a new version of SeaDAS is released. These data were initially being produced using SeaDAS 6.4 and more recently, with SeaDAS 7.x.
The filenames are of the form Ayyyymmdd.vv.aust.chl_gsm01.nc4, where 'A' denotes MODIS/Aqua, 'yyyymmdd' is the GMT date of the mosaic, 'vv' is the SeaDAS processing version, 'aust' indicates a whole-of-Australia mosaic and the 'nc4' suffix is for netCDF4 format data files.
****NETCDF FILENAMING CONVENTION FOR AQUA FILES****
The filenaming scheme puts a lot of useful metadata into the filename with the aim of making it easy to parse by machine and eye.
Folder D-20120802.G-0720.P-aqua.C-20120802082919.T-d263047n000000.S-m.E-definitive.Z-ok.R-20120802185051/ --
Folder D-20120802.G-0715.P-aqua.C-20120802111359.T-d549724n000000.S-mn.E-definitive.Z-ok.R-20120802185343/ --
Folder D-20120802.G-0710.P-aqua.C-20120802111224.T-d549724n000000.S-mn.E-definitive.Z-ok.R-20120802185344/ --
Folder D-20120802.G-0545.P-aqua.C-20120802101905.T-d549724n000000.S-na.E-definitive.Z-ok.R-20120802185437/ --
Folder D-20120802.G-0540.P-aqua.C-20120802101900.T-d549724n000000.S-cmna.E-definitive.Z-ok.R-20120802185314/ --
Folder D-20120802.G-0535.P-aqua.C-20120802101855.T-d549724n000000.S-cmna.E-definitive.Z-ok.R-20120802185436/ --
Folder D-20120802.G-0405.P-aqua.C-20120802084036.T-d549724n000000.S-qna.E-definitive.Z-ok.R-20120802185313/ --
Folder D-20120802.G-0400.P-aqua.C-20120802084028.T-d549724n000000.S-cqna.E-definitive.Z-ok.R-20120802185314/
Split the names on ‘.’, and then you have NAME-VALUE pairs where
D = GMT Date
G = GMT Acquisition
P = Platform
C = Creation date/time (yyyymmddhhmmss)
T = number of modis packet types (d=day packets, n=night packets)
S = contributing reception stations (a=Alice Springs, c=Crib Pt, m=Murdoch, q=AIMS, n=NASA DAAC)
E = Ephemeris (predicted or definitive)
Z = L1B processing status (should always be ok for these data)
R = date/time of processing of L2 Chl granule (but I forget what R stands for)
