I think this will work, but ....1) What samplexy will do in this situation is to bilinearly interpolate the velocities to the chosen points. It looks like your points are very close together. How many gridboxes do they span? (If less than 1, it seems unlikely that the result will be meaningful.)
2) You can check the result by choosing a line that begins and ends at existing gridpoints. Then do the calculation along the two x and y lines that complete the triangle. Is mass conserved in the triangle? (Or is it conserved when any sea level variation is taken into account?)
3) A similar check is to do the calculation along latitude-longitude steps that are along gridboxes.
A possibly better way would be the following:* define your section line as a succession of, say y(x), points, namely the latitudes of the line as it crosses each gridpoint longitude * interpolate to find the velocity perpendicular to the section line in each gridbox * multiply that perpendicular component by the distance spanned within each gridbox to get the perpendicular transport (also multiply by z)
* sum those along the line=> again a check is to do the same thing using switched components (i.e. find the line as a series of x(y) points)
Billy K On 04 May 2010, at 12:40 PM, Praveen V K wrote:
Hi All, I am wondering whether ferret can do this. I have a 3D-regular grid data of 240 X 240 X 24 (x,y,z) dimension.Can we calculate the Mass transport normal to a Section line which is not parallel to Longitude or Latitude in ferret?I tried the following method. use uv.nc let vtrans = v*xbox*zbox let utrans = u*ybox*zbox ! Transport Section line (approximately diagonal to lat-lon grid) let xpoints = {-42.2090,-41.7973,-41.3856} let ypoints = {62.3584, 62.1288,61.8992} let utrans_sec = samplexy(utrans,xpoints,ypoints) let vtrans_sec = samplexy(vtrans,xpoints,ypoints) let trans_mag = (utrans_sec^2+vtrans_sec^2)^0.5let trans = trans_mag*vtrans_sec/abs(vtrans_sec) ! This is to get the direction, since I know they are going south.I am not sure whether I am getting the transport normal to the section or not!If not, can we get normal transport for such sections in ferret. Thanks Praveen
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ William S. Kessler NOAA / Pacific Marine Environmental Laboratory 7600 Sand Point Way NE Seattle WA 98115 USA william.s.kessler@xxxxxxxx Tel: 206-526-6221 Fax: 206-526-6744 Web: http://www.pmel.noaa.gov/people/kessler