A nonlinear Delayed-Mode Quality Control (DMQC) methodology is developed to estimate monthly mean climatologies from the large Argo dataset (2005-2012) over the North Pacific Boundary Current region. In addition to the commonly-used linear quality control procedure, which checks for instrumental, transmission and gross value errors, we develop and show the impact of nonlinear, iterative statistical checks on the quality of the resulting climatology. Objective analysis is applied to produce progressively statistical climatological checks to flag out-of- standard-deviation ARGO profiles. The optimal method uses horizontal regional climatological averages defined in five regime- oriented subregions in the Kuroshio areas and the Japan Sea. This new method is shown to produce lower standard deviations in the deep water layers and it is thus capable to reject observations with large representativeness/sampling errors. The resulting monthly mean climatology for the period 2005 to 2012 is shown to be consistent with previous estimates for the region, including the WOA13 observed climatology and re-analysis fields.
- Keywords: Kuroshio dynamics, ARGO climatology, non-linear quality control algorithms