Sets all instances (i.e., time) of a grid point to missing if a user-prescribed percentage of points is missing.
load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/contributed.ncl" function rmInsufData ( x : numeric, percent : float ) return_val [dimsizes(x)] : typeof(x)
An array of any dimensionality. Time is assumed to be the rightmost dimension.percent
The threshold percentage to compare against, in decimal form (value between 0 and 1).
The results are returned in an array of the same type and dimensionality as x.
For each index of dimensions 0...n-2, this function sets the n-1th (rightmost) dimension of input array x to missing (_FillValue) if a user prescribed percentage of data is missing.
For example, taking example array y, dimensioned (lat,lon,time):
z = rmInsufData(y,.5)
If in y(0,0,:), 41% of the data is set to _FillValue, then y(0,0,:) will equal z(0,0,:). However, if in y(0,1,:), 52% of the data is set to _FillValue, then z(0,1,:) will be set to missing (=z@_FillValue).
This function can be useful if one is looking at data with a large number of missing data, and one does not want to include those grid points that have too much missing data in upcoming calculations.
Note that you can use it to overwrite the original array:
y = rmInsufData(y,0.75)
No examples are currently available for this function.