NCL Home > Documentation > Functions > Regridding

triple2grid2d

Places unstructured (randomly-spaced) data onto the nearest locations of a curvilinear grid.

Prototype

	function triple2grid2d (
		x        [*] : numeric,  
		y        [*] : numeric,  
		data     [*] : numeric,  
		xgrid [*][*] : numeric,  
		ygrid [*][*] : numeric,  
		option   [1] : logical   
	)

	return_val [dimsizes(xgrid)] :  typeof(data)

Arguments

x
y

One-dimensional arrays containing the coordinates associated with the data values. For geophysical variables, x and y correspond to longitude and latitude respectively.

data

A one-dimensional array containing the values associated with the x and y coordinates. Missing values may be present (data@_FillValue), but they are ignored.

xgrid

A two-dimensional array containing the X coordinates to be associated with the returned two-dimensional grid. For geophysical variables, these are longitudes.

ygrid

A two-dimensional array of the same size as xgrid containing the Y coordinates to be associated with the returned two-dimensional grid. For geophysical variables, these are latitudes.

option

If option=False, the function will operate under default mode (that is, all grid points will be assigned a value).

If option=True, then this variable may have associated with it the attributes mopt and/or distmx.

option@mopt - an integer value of 0 or 1

option@mopt=0 (default) - use a quick approximation to determine the distance of an x/y location to a grid point.

option@mopt=1 - calculate the distance using the great circle distance formula. This will result in slower execution times.

option@distmx

Any x/y/data observation greater than option@distmx from a grid point will not be used. Note: If this option is used, it is possible for some grid points to be returned as missing (data@_FillValue).

Return value

The returned two-dimensional array will be the same size as the xgrid/ygrid arrays and the same type as data.

Description

This function puts unstructured data (randomly-spaced) onto the nearest locations of a curvilinear grid.

This function does not perform interpolation; rather, each individual data point is assigned to the nearest grid point. It is possible that upon return, grid will contain grid points set to data@_FillValue if the option@distmx is set.

The units of distmx are "grid-units" if the default distance formula is used, and kilometers if the great circle distance formula is used, as indicated by option@mopt.

For more robust regridding, see the ESMF regridding examples, which show how to regrid data from and to rectilinear, curvilinear, or unstructured grids. Available in version 6.1.0 and later.

Examples

Example 1

Assume xlon, ylat, and zVal are one-dimensional (1D) arrays of length KPTS. Assume lon and lat are 2D arrays specifying grid locations. Then:

   grid = triple2grid2d(xlon,ylat,zVal, lon,lat, False)

will use default behavior and return a 2D array of the same size as xgrid.

Setting option=True and option@mopt=1 will result in the function using the more accurate great circle distance formula. The execution time will be longer, however:

  opt       = True       ; setting options
  opt@mopt  = 1    
  grid      = triple2grid2d(xlon,ylat,zVal, lon,lat, opt)

Setting the option@distmx option will result in any observation greater than distmx to be not used: