newdrift

Replacement for Grumbine, 1998 drift model -- https://github.com/NOAA-EMC/drift_grumbine

To be runnable as

• Standalone program ('offline')
• Subroutine to CICE ('inline')

Inputs:

• .nc file of points to track ** reformat_points.py in ush to take text file of locations and create .nc file ** skiles.nc has just the skiles points ** merged.nc has skiles + ~25 km grid based on RTOFS points
• u, v of ice -- RG: rotation? grid?
• metric (needs input lat,lon of points, computes dlatdi, etc.; add rotation info here?)

Outputs:

• .nc file of buoy starting point, drift distance and direction w.r.t. time lead
• kml output option

Notes:

• Parallelized -- handle drifter passing out of domain of a processor (inline) -- -- > reference processor number that handles re-partitioning, knows tiling

#Auxiliary:

• Blender of ensemble outputs
• RG: ?how to represent uncertainty
• Sidfex-compatible output?