Aquarius Level-3 Binning and Mapping Fred Patt

Definitions Projection - any process which transforms a spatially organized data set from one coordinate system to another. Mapping - a process of transforming a data set from an arbitrary spatial organization to a uniform (rectangular, row-by-column) organization, by processes of projection and resampling. Binning - a process of projecting and aggregating data from an arbitrary spatial and temporal organization, to a uniform spatial scale over a defined time range. Ideally the binning process will preserve both the central tendency (e.g., average) and the variation in the data points that contribute to a bin.

NASA Ocean Product Projections Equal-area: sinusoidal, with equally space rows and number of bins per row proportional to sine of latitude. Equal-angle: rectangular (Plate Carrée) with rows and columns equally spaced in latitude and longitude. Ocean equal-area and equal-angle projections are equivalent at the equator.

Sinusoidal Equal-Area Projection

Why Use Equal-Area Bins? When generating large-area (global or regional) averages, equal-area bins provide correct areal weighting without correction. Other uses (e.g., models) may also require equal-area inputs. Ultimately the decision about a binning projection is driven by science needs.

Binned and Mapped Products Binned products are generated daily from Level-2, and then aggregated to weekly, monthly, seasonal, annual and mission. Standard mapped image (SMI) products are generated at each temporal resolution by projecting binned files to the equal-angle grid.

Binned Product Format Metadata indicates product type, start/end dates, and geographic extent. Bin geometry parameters describe projection and resolution. Binned data are stored in record format: –Common fields for all parameters (e.g, number of observations, temporal information, weighting). –For each parameter, central tendency (e.g., average) and variability (e.g., std dev). –Only bins with observations are stored.

Mapped Product Format One product for each geophysical parameter Metadata similar to binned product. Map parameters describe projection and resolution. Mapped data are stored in array format: –Current Ocean mapped products use scaled integers for data. –Fill values for unfilled points

Questions What is the optimum method for averaging? What information should be saved about the variability? Will additional post-processing (i.e., smoothing) be required? How can testing with various kinds of noise be incorporated into the mission simulation? Should temporal sampling information (e.g., days of the month that contributed to a data point) be preserved? How should the beam footprint-to-bin registration be handled?