Ocean Interpolation by Four-Dimensional Weighted Least Squares—Application to the Waters around Australasia

摘要: A new four-dimensional ocean interpolation system based on locally weighted least squares fitting is presented. loess filter used to interpolate irregularly spaced data onto a uniform grid. This involves projecting the quadratic functions of latitude and longitude while simultaneously annual semiannual harmonics by squares. The smoothness scale mapping method adapts match density, thus producing gridded estimates with maximum resolution. has vertical dimension, such that adjacent levels are included in computation. greatly reduces effects discontinuities distributions between levels, since at each level no longer independent. scheme been further modified so weighting points adjusted allow for influence both bathymetry land barriers. allows mapped fields natural way, leakage structure deep shallow regions produces far more realistic coastal gradients. flexibility approach allowed adjustments compensate irregularities spatial temporal sampling. shown be statistically consistent an objective measure priori noise dataset. Departures from independent surface temperature climatologies mean sections derived withheld expendable bathythermograph (XBT) within error limits. applied major seas around Australia, New Zealand, Papua Guinea, Indonesia (508S‐108N, 1008E;‐1808) form high-resolution seasonal climatology temperature, salinity, oxygen, nitrate, phosphate, silicate, referred as CSIRO (Commonwealth Scientific Industrial Research Organisation) Atlas Regional Seas (CARS). Stringent quality control procedures have comprehensive dataset assembled all known sources. resulting maps successfully resolve large-scale narrow features illustrate how influences property distributions.