Effects of Spatial Interpolation Algorithm Choice on Regional Climate Model Data Analysis

The analysis of regional climate model (RCM) outputs frequently requires spatial interpolation of the data from the model's native grid to another set of locations: a different grid is needed for comparison with other models, a set of station locations for modeling of dependent processes or comparison with raw observations, specific points of interest for impacts studies, and so on. Different interpolation algorithms will produce results with different spatial characteristics, such as smoothness, synoptic patterning, and distribution of extremes. To explore the importance of these differences in the NARCCAP context, we regrid model output from six different RCMs driven with NCEP boundary conditions using several interpolation methods of varying mathematical sophistication: nearest-neighbor, bilinear, inverse-distance weighting, and thin-plate spline interpolation. For each algorithm, the results are compared with observations, driving data, and source model data to determine what the magnitude of the artifacts due to interpolation is and whether these effects are likely to be significant for intermodel comparison, impacts modeling and analysis, and other uses popular in the NARCCAP community.