Hi,
I work with ocean data and have never found a good solution to this problem. I download ocean current data as a NetCDF file which has the currents shown in vectors (northward (u) and eastward (v)) component for each time step (e.g. data files from Copernicus Marine Service). I would like to create a new multidimensional raster that has the current velocity (magnitude but no direction) for each time step such that I can calculate the average, 95th percentile, frequency above a certain threshold, etc. I have very basic coding skills with R and Python, enough to adjust a script but not much more.
In the past, I've managed this by sub-sampling days out of the data such that I can create u and v rasters, calculate velocity in the raster calculator (sqrt (u^2 + v^2)), and then work with those few files. However, to characterize an environment well, I really should be using 3+ years of data (over 1000 files) which makes this manual method unviable.
Does anyone have a more streamlined method to create a multidimensional file (NetCDF or just a raster) with directionless velocity for each time step?
Thanks
