Hi, I am trying to interpolate an atmopsheric model to an altitude grid, and I am totally confused by the interpolation methods, and just can't work out how to do it. I have: lon = fltarr(144) ; 0 to 360 lat = fltarr(96) ; -90 to 90 alt = fltarr(144,96,66) ; calculated from pressure grid and so different at all points. I want to interpolate other property arrays (144,96,66) onto the same lat/lon, and a specific altitude grid (alt_i = fltarr(66)). Currently, I am simply looping over all lats and lons, and using prop_i[lon,lat,*] = interpol(reform(prop[lon,lat,*]), reform(alt[lon,lat,*]), alt_i). I've had to start using a new computer on which this has become extremely slow, and I guess it's probably not as accurate as other methods. I've tried using griddata, but I didn't understand the output, and when I tried to get triangles, it said that my x and y arrays had to be the same length. I also tried grid3, but this takes significantly longer than the simple method before it fails with the nodes co-planar error. I considered interpolate, but this would involved the same amount of use of interpol to get the indices anyway as far as I can tell. I'm sure there must be a way to interpolate this properly, i.e. something with 3 dimensions of different length, where one of the location arrays is 3d, to something with 3 dimensions of different length, where all the location arrays are 1d. Is there anything that can be done? Thanks very much for your help, James