Time interpolation of directional data - avoid interpolating over 360-0.

[jsmariegaard]

Currently, when working with directional data, there is not stopping from interpolating over 360-0 jump, which could lead to some wrong comparisons. One way to handle this would be to convert both obs and model to x- and y-components upon interpolation. For now a check could be made to ensure no interpolation is going on.

[ecomodeller]

We need to be able align model results and observations to a common time axis.

Is method='nearest' what you are after?