Introduction  The Sea-Viewing Wide-Field Sensor (SeaWiFS) on the OrbView-2 platform, launched in fall, 1997, provides a time series of phytoplankton chlorophyll a maps of the global ocean with unprecedented spatial and temporal coverage (McClain et al., 1998). Starting with 8-day global composite SeaWiFS chlorophyll a imagery, we averaged the data in space and time (to reduce the number of cloud-obscured pixels) to build a 7-year time series at 1 degree spatial resolution with nominal 8-day temporal resolution. 

This data set has been prepared to help understand temporal and spatial variability in the global ocean and how chlorophyll a variability may be related to other fundamental biogeochemical measurements. 

Methods  Global (9-km and 8-day resolution) SeaWiFS chlorophyll concentration images for the period September 1997 through December 2004 were acquired from NASA's archive (http://eosdata.gsfc.nasa.gov/data/dataset/SEAWIFS). NASA designates these data, which are image representations of binned data products, as "Level 3 Standard Mapped Image (SMI) files". "Bins" correspond to grid cells on a global grid, each cell approximately 81 square kilometers in size. All the data for each grid cell is accumulated for 8 days and placed in the same "bin". Methods describing NASA's data processing are discussed in McClain et al., 1998, Robinson et al., 2000, and O'Reilly et al., 2000. 

We averaged the NASA data within each 8-day array onto a 0.25 x 0.25 degree grid using a maximum likelihood estimator (MLE) as described by Campbell et al. 1995. We then applied a 1 x 1 degree running box median filter to remove small-scale variability and noise. The resultant images were then sub-sampled to reduce the spatial resolution of the arrays to 1 x 1 degree. 

As satellite chlorophyll data can be approximated with a log-normal distribution, we log-transformed the data and then used a 3-point (24-day) running mean (and when necessary, used 1 of the 3 points to represent the mean) to smooth the time series in each pixel. This process resulted in a comparatively gap-free time series between 50 N and 50 S, with each of 21,352 ocean grid points having a 332-element time series (7.25-years with nominal 8-day resolution). At grid points poleward of either 50N or 50S, there is a lot of missing data during winter months owing to cloud cover or to low incident sunlight (and thus too low water-leaving radiance for calculating ocean chlorophyll). 

After the smoothing procedures discussed above, an additional 4,726 grid points were missing less than 5% data in the 332 "week" time series. To maximize spatial coverage, we filled these missing values using "monthly" (mean of 4, 8-day periods) mean values centered on the week of missing data. In no case did we use the monthly mean fields to fill values missing more than 3 consecutive elements in the time series. Our final space-time cube (x,y,time) consisted of 332 arrays (maps) of ocean chlorophyll coverage. With the exception of a few persistently cloudy regions, we had almost complete coverage of the ocean between 50N and 50S during each 8-day period. 

Data Set Description The data set archived here consists of a 1 degree x 1 degree chlorophyll data cube [360 x 180 x 332] for the period September 1997 through December 2004. Chlorophyll are in units of mg/m^3. Land has been filled with the value -999.9 and missing data have the value -99.0. "Time" in the data set corresponds to the start of each 8-day period. An animation of this time series can be viewed at: http://www.po.gso.uri.edu/~maureen/sm_seawifs.html

Data Analysis An EOF analysis of the 4-yr (1998-2001) data set between 50N and 50S, which consisted of 184 arrays, each consisting of 25,551 grid points of ocean chlorophyll coverage, was carried out to study the variability at different time scales. The results of this analysis are described in Yoder and Kennelly, 2003.

Campbell, J.W., J.M. Blaisdell and M. Darzi. 1995. Level-3 SeaWiFS data products: spatial and temporal binning algorithms. SeaWiFS Technical Report Series, NASA Technical Memorandum 104566, Vol. 32. Goddard Space Flight Center, Greenbelt, Maryland. 

McClain, C.R., M.L. Cleave, G.C. Feldman, W.W. Gregg, S.B. Hooker and N. Kuring, 1998. Science quality SeaWiFS data for global biosphere research. Sea Technology, 39: 10-16. 

O'Reilly, J.E., S. Maritorena, D.A. Siegel, M.C. O'Brien, D.Toole, F.P. Chavez, P. Strutton, G.F. Cota, S.B. Hooker, C.R. McClain, K.L. Carder, F. Muller-Karger, L. Harding, A. Magnuson, D. Phinney, G.F. Moore, J. Aiken, K.R. Arrigo, R.Letelier, and M. Culver, 2000. Ocean Chlorophyll a Algorithms for SeaWiFS, OC2, and OC4: Version 4. In O'Reilly, J.E., and 24 Coauthors, SeaWiFS Postlaunch Calibration and Validation Analyses, Part 3. NASA Tech. Memo. 2000-206892, Vol. 11, 9-19. 

Robinson, W.D., G.M. Schmidt, C.R. McClain, and P.J. Werdell, 2000. Changes made in the Operational SeaWiFS Processing. In McClain, C.R., R.A. Barnes, R.E. Eplee, Jr., B.A. Franz, N.C. Hsu, F.S. Patt, C.M. Pietras, W.D. Robinson, B.D. Schieber, G.M. Schmidt, M. Wang, S.W. Bailey, and P.J. Werdell, SeaWiFS Postlaunch Calibration and Validation Analyses, Part 2, NASA Tech. Memo. 2000-206892, Vol. 10, pp 12-28.

James A. Yoder
Graduate School of
Oceanography 
University of Rhode Island 
South Ferry Road 
Narragansett, RI 02882
401-874-6864
jyoder@gso.uri.edu