Advanced features | Department of Earth Sciences

FORCinel: Advanced features

Data Masking and Elimination of Artifacts

Because of its nearest neighbor approach to smoothing, FORCinel allows FORC distributions to be extrapolated across gaps in the data. By contrast, conventional FORC processing requires a regular grid of points. Thus, FORCinel allows users to remove anomalous curves and points from the original data set. This additional flexibility to mask outlying data points caused by instrumental instabilities allows users to reduce artifacts in the final FORC diagram. Alternatively, specific regions of interest in the FORC measurement can be masked out in order to investigate their contribution to the final FORC diagram.

a set of FORCs from a micromagnetic simulation of a 150 nm cube of magnetite the resulting FORC diagram

Example:

Above left is a set of FORCs from a micromagnetic simulation of a 150nm cube of magnetite. Many of the curves show systematic changes as afunction of the reversal field. Three curves, however appear to dosomething different. We wish to explore the impact of these threecurves on the resulting FORC diagram (above right) by removing themfrom the analysis.

FORCinel Data Masking panel

Step 1

From the FORCinel menu, select Data Masking to bring up the Data Masking panel.

This panel allows you to plot the raw magnetisation matrix M(H_a, Hb) in one of three modes. M plots the raw magnetisation data, dM/dHa and dM/dHb plots the first derivative of M with respect to Ha or Hb. The latter two options are very useful for identifying outlying points caused by instrumental instabilities. Outlying points may sometimes appear in one view but not the other, so be sure to check all three to see which suits your case best. You can change the preferred view at any time to see how the masking is progressing.

Drag a marquee selection over the region of the raw data you would like to mask out

Step 2

Press the M button to bring up an image of the raw M(H_a, Hb) matrix. Drag a marquee selection over the region of the raw data you would like to mask out, the press the Mask button on the panel to mask that area. The masked area will turn white. If you have made a mistake then press the Undo button to return to the previous state. You can mask as many different regions as you wish, although please note that FORCinel currently only supports one level of undo.

After masking an area, enter the desired smoothing factor in the box and press the Smooth button

Step 3

After masking an area, enter the desired smoothing factor in the box and press the Smooth button to see the effect on the processed FORC diagram. If the mask has not had the desired effect, press the Undo button and then press the Smooth button again to return to the prior state. If you wish to return to the original FORC diagram (i.e. without any masking at all), simply choose Process FORCs from the FORCinel menu.

Eliminating the 'First-point artefact'

It has often been observed that the very first point of each FORC curve is anomalous. In the figure below you can see that the very first point in each curve lies far off the trend established by the rest of the data. This may be an artefact of the instrumentation or an effect due to time-dependent magnetisation processes. Either way, it leads to a very obvious signal in the processed FORC diagram that can obscure real information that you need to extract. This 'first-point artefact' can be effectively removed from the processed FORC diagram by selecting the "Remove First Point Artefact" option in the FORCinextras menu. This removes the very first point from each FORC curve and reprocesses the data using the last smoothing factor. In this example, removing the artefact allows us to see the true nature of the PSD signal at low Hc and -ve Hu.