Home Forums Code snippets Smoothing spikes in ratio data

This topic contains 3 replies, has 2 voices, and was last updated by  Dr Tephy Marillo-Sialer 1 year, 6 months ago.

  • Creator
  • #9870

    Sarah Gilbert


    I’ve recently created a DRS for Pb-Pb isotope ratios for CPS data, modified from the ‘Laser_Pb_Faraday’ DRS. It’s fairly straightforward: correcting for Hg interference on 204Pb; calculating isotope ratios from the cps data; applying a mask to cut low noisy signals; and correcting to a reference material. That all works well.

    However for some analyses in the Output there are large spikes in the final ratios (caused by abnormally high CPS for 1 isotope, rather than regular signal noise).
    The only smoothing code I have found in other DRS relate to smoothing the standard data for calculating down hole ratiocination curves (which isn’t applicable to this Pb-Pb ratio data).

    Smooth/M=(abs(FitOutlierTolerance*(ThisWave[p]+ThisWave[p+1]+ThisWave[p-1])/3))/R=(NaN) 9, ThisWave //aiming at only removing spikes. set here at 1.5 times the average value for this portion of the wave (9 point range centered on this point)

    I haven’t had any success at modifying this to apply to the Output ratios…
    Is there a way to smooth these spikes or apply an outlier rejection for the final Output ratios?


Viewing 3 replies - 1 through 3 (of 3 total)
  • Author
  • #9881

    Hi Sarah,

    There is an option to perform outlier rejection on your selections. This option is called ‘Stats for normal selections’ and it is found in the DRS tab under the General Settings section. You can choose from a list of predetermined options for outlier rejections.

    If this does not have the filtering effect you are looking for, I would recommend using linked selections in order to exclude the spikes from your data rather than performing smoothing on a channel or selection. For more information about Linked Selections, please refer to section 6.4 of our user’s manual. You can find a copy of this manual within your iolite folder or by clicking on this link.



    Sarah Gilbert

    Hi Tephy

    Thanks for the reply. That helps to clarify thing for me.
    Does the smoothing from the ‘stats for normal selections’ just gets applied to the final results (ie effects the reported average ratio & 2SE) but not to the Wave itself?
    My ‘Final’ ratio wave (versus time) still has spikes regardless of the outlier rejection applied. I think that’s what was confusing me. I didn’t think it was being applied at all.
    Is there a way to determine which spikes have been removed? To help determine if using Linked Selections is required



    Hi Sarah,

    Yes, the outlier rejection is applied when averaging the data within each selection and affects the reported average and uncertainty. It does not change the original data and for that reason the changes are not reflected in the Time Series windows.

    Unfortunately iolite does not store the info on which outliers have been removed as this calculation is done on a selection by selection basis.

Viewing 3 replies - 1 through 3 (of 3 total)

You must be logged in to reply to this topic.