[Thread Prev][Thread Next][Index]

Re: [ferret_users] a possible bug in the high resolution time series- depth section



Hi all,
Golla and I talked outside the list a bit.  The grid of the data is 24 points in Z and 51300 in the T direction. Ferret's contouring algorithm just doesn't work on such a grid.

And in fact there are not enough pixels in the plot page to show anything near 51000 points. A plot of a subset of the data in the T direction should work fine, for instance,

  fill/L=1:5000 temp

To plot the data for the whole time range, the right answer here is to regrid to a coarser grid in time, or even just decimate the data for plotting.

Ansley

On 3/15/2013 9:12 AM, Ansley Manke wrote:
Hi Golla,
To follow up, an easy way to make the check that Steve is suggesting is to do a SHADE plot with exactly the same qualifiers and region as your FILL command.  That should show whether this is the issue.

Ansley


On 3/15/2013 7:16 AM, Steve Hankin wrote:
Hi Golla,

Check and see if there are lots of missing values in your data.  The FILL command requires adjacent valid points in order to trace the boundaries of the color changes, so it is not suitable when there are lots of embedded missing values.

If you regrid using an @ave regridding transformation, then the coarser grid will have valid data as long as there is at least one valid point from the fine grid within the coarse grid cell.  So the number of missing values is greatly reduced.

    - Steve


On 3/15/2013 12:21 AM, golla nageswararao wrote:
Hi all,
 Now-a-days I am facing some problem with the high resolution time series data (5 second).
Suppose I am having data in z axis and t axis (5 sec) for suppose 5 days. When I try to plot the depth-time section using fill command, it is not plotting anything (blank plot) but scale bar will be there.The same data when I plotted with shade command, only second half of the data is plotted. Then, I checked whether data is there inside the .nc file or not. Suprising data is there inside. If I regrid this data to a coarser resolution in t-axis, again full data is plotting both either with shade and fill commands. I think this a bug or ferret is not able plot high resolution datasets. I didnt understand this weird behaviour. Any help in this regard is appreciated.

Thanks in advance.

--
With Best regards,
G.NageswaraRao.




[Thread Prev][Thread Next][Index]
Contact Us
Dept of Commerce / NOAA / OAR / PMEL / Ferret

Privacy Policy | Disclaimer | Accessibility Statement