Enhance Series-Analysis to read output from TC-Pairs to do cyclone relative verification on the fly.

dwfncar commented 8 years ago

This task is from the SBU project. The current solution for cyclone-relative vx is done via scripts. This task is to enable Series-Analysis to read .tcst to do cyclone relative vx with the forecast centered on the ADECK location and the analysis centered on the BDECK location.

Issues...

Need to let the user define the tile information. (i.e. nx, ny)
  - Should we just round the storm location to the nearest vx grid x, y values or are we actually regridding to a new grid centered on that lat/lon location?
  - What about projection and grid spacing?
  - Is the tile a "new grid" or just a subset of the verification grid? Probably just a subset of the current grid.
  - What grid is used when writing the output of Series-Analysis? [MET-590] created by johnhg

dwfncar commented 7 years ago

We have working capability and considering where we are with 6.0 I'm moving the priority down to minor by jensen

dwfncar commented 7 years ago

SBU if available otherwise AOP2017 VX1 NOAA by jensen

dwfncar commented 7 years ago

During my visit to NCEP on 9/21/2017, Tracey Dorian presented results of running MODE on hurricane Harvey. Would be convenient to have this capability in Series-Analysis to easily extract tiles stacked up on the hurricane center and oriented to the path of the track. by johnhg

dtcenter / MET

Enhance Series-Analysis to read output from TC-Pairs to do cyclone relative verification on the fly. #590