This page is under construction. A short summary of MSP-Tool (including an optimised workflow for MSP experiments) is also shown on this poster (IUGG Prague, 2015).
If you come across any bugs in the software or if you have other questions, please leave a comment or send me an e-mail. Thank you!
1. Selecting the right temperature
For a succesful MSP experiment, it is important to select an appropriate temperature for your experiment. Ideally, this temperature should be high enough to unblock a significant fraction of the sample’s NRM (preferably > 20%), while also being below the specimen’s alteration temperature. At Fort Hoofddijk, we use the following three steps to select a temperature:
- Susceptibility-versus-temperature diagram. We measure the specimen’s susceptibility as a function of temperature while cycling to progressively higher temperatures to check for irreversible behaviour indicative of chemical alteration.
- Thermal NRM decay curve. Based on the susceptibility-versus-temperature plot and the thermal NRM decay curve, we select a temperature at which no alteration has occured and a significant fraction of the NRM has been unblocked. For sites that have high Curie temperatures, this is often not possible. (See site TD in the associated paper, which retains > 90% of its NRM at the MSP temperature.)
- ARM test. The ARM test (de Groot et al., 2012) provides an additional way to test for subtle alteration at the selected MSP temperature. If the amount of ARM gained by a heated sample is equal to the amount gained by a pristine sample, the MSP protocol should produce the correct palaeointensity (PI). ARM test data were analysed using ARMTester.
For sites PI from the associated paper, a temperature of 300 ℃, safely below its alteration temperature and unblocking about 40% of its NRM, was selected:
2. Sample alignment
To minimise tail effects (and to enable scalar instead of vector calculations) the specimens should be aligned parallel to the field in the oven. MSP-Tool can detect and (partially) correct for alignment errors, but it does not take into account tail effects etc. It is therefore still important to carefully align your samples. The alignment procedure is illustrated in the figure below. We use a sample holder similar to the one used by Böhnel et al (2009).
1. Data input
Data can be imported in two different formats: using the Cartesian components of the remanence or using the intensity, declination and inclination of the remanence. The latter type will automatically be converted to Cartesian format. Errors (in %) and exponents (multiplication factor for e.g. the JR6 spinner magnetometer) are optional. After clicking ‘import data’ you can choose between these two input types:
[sample name], [lab field] [step name], [mx], [my], [mz], [error], [exponent]
E.g. for site PI using the DSC protocol and a lab field of 12 µT (including errors and exponents):
[sample name], [lab field] [step name], [intensity], [declination], [inclination], [error]
E.g. for site MT using the DB protocol and a lab field of 36 µT (no errors or exponents provided):
Example input files:
- DB protocol; Cartesian (download)
- DSC protocol; Cartesian (download)
- Mix of DB, FC and DSC protocol; Cartesian (download)
- Mix of DB, FC and DSC; int/dec/inc (download)
These files are all from site PI (shown in the associated paper). Half of the specimens were misaligned on purpose to test and showcase the alignment correction, resulting in a large amount of scatter in the uncorrected plots. For a ‘real’ MSP plot, see e.g. this site from La Palma (Monster et al., 2015; DSC protocol).
N.B.: Please don’t forget to separate the specimens by one empty line! MSP-Tool will not import your file and show a message box (“Please check your input file. It appears that your file includes duplicate measurements.”) if more than six consecutive lines (header + 5 DSC measurements) contain data. If your data are imported incorrectly, please go to the trouble shooting page.
MSP-Tool also checks for the following types of error, which are indicated in message boxes and are marked in the input sheet:
- Missing laboratory field.
- Possible missing header.
- Possible missing data.
- Possible duplicate lines.
1. The ‘input’ sheet
The ‘input’ sheet shows the imported data (column A-F) as well as the following calculations:
- Columns G-I: The isolated pTRM components, using the NRM remaining as estimated from the vectors m1 and m2. (NRMremaining = 0.5 · (m1 + m2))
- Column J: The scalar intensities m0 to m4.
- Column K: The alignment-corrected scalar intensities m1 to m4, obtained by adding up the intensities (lengths of the vectors) of the isolated NRM remaining and the isolated pTRMs.
- Columns L-M: The declination and inclination of m0 to m4.
- Columns N-O: The difference in declination and inclination, respectively, between the NRM remaining and the pTRMs. Using Δdec and Δinc makes it easier to spot systematic alignment errors (e.g. rotating the specimens the wrong direction when aligning them to the field in the oven) than when a single angle would be used.
- Column P: The ‘m2 factor’. This factor is +1 when m2 and m0 are parallel and -1 when they are anti-parallel. It is calculated by taking the normalised dot product of these two vectors.
- Column Q: The percentage NRM lost, as estimated from the isolated NRM remaining.
- Column R: The overprint check. If the angle between the NRM remaining and the NRM lost exceeds the accepted angular deviation (AAD), a warning is shown in this column.