Perhaps the most common challenge to the systematic and routine comparisons of data from the model output/analysis with NASA remote sensing data is that of CO-locating diverse datasets often…

Perhaps the most common challenge to the systematic and routine comparisons of data from the model output/analysis with NASA remote sensing data is that of CO-locating diverse datasets often obtained from instruments with different observation characteristics for the purposes of detailed comparison e.g. OBS4MIPS (teixeira ET AL 2014) and use e.g. adaptation to climate change. Indeed while a myriad of tools exist to locate access and visualize observational data inter-comparison of disparate data sources requires tedious and often computationally intensive gridding or regridding which is generally implemented in an ad hoc manner by individual users and becomes an obstacle to robust comparisons. In addition to the duplication of effort stemming from the lack of standard gridding tools expediency on the part of individual researchers often results in sub-optimal gridded products that (1) limit accuracy due to failure to adequately account for spatial/temporal sampling bias (2) lack robust gridding uncertainty estimates and (3) omit the provenance all of which limit the value of the products to other researchers. Simplicity and familiarity also drive researchers to apply traditional latitude-longitude (LAT-LON) grids rather than better alternatives. Moreover the trend in the modeling community is to transition to next-generation grid systems such as geodesic and cubed-sphere that possess superior quasi-equiareal scalable characteristics. To maintain numerical stability the CFL criterion for the time step used in the numerical integration of the general circulation models must decrease with the smallest distance between any two grid points. Since the meridians converge to a single point at each pole as the model S spatial resolution increases the smallest grid distance of a LAT-LON grid system quickly approaches zero. Considerable computation will thus be wasted for high-resolution models based on LAT-LON grids for the short time step required near the poles becomes progressively a greater overkill for grid cells toward the equator. Approaches to filter the high-frequency signal to sidestep this concern introduce severe constraints on parallel performance.