An online calculator input format for multiple-nuclide data
If you’re reading this, it’s hard to imagine that you’re not already familiar with the online exposure age calculator that has been serving up exposure ages since 2006. The current version of the portion of this calculator that computes surface exposure ages, which I believe is something like 2.2.1 but will henceforth be referred to as version 2, computes exposure ages from Be-10 and Al-26 measurements in quartz. This fairly limited selection of nuclide-mineral options is just because i) the Be-10-in-quartz pair is by far the most common exposure-dating technique, and ii) Be-10 and Al-26 can be simultaneously extracted from quartz samples with just a little bit more work than just measuring Be-10, so these two nuclides are commonly both measured.
However, there are lots of other nuclide-mineral pairs out there that can be used for exposure-dating. Among the reasonably routinely measured are Ne-21 in quartz, He-3 in olivine and pyroxene (and quartz), and carbon-14 in quartz. It would be rather useful if the present online exposure age calculator would do calculations for all of these nuclides. It would also be rather useful if computations for all nuclides measured in a particular sample could be done at once, because the comparison of concentrations of multiple nuclides with different half-lives (or diffusion kinetics) has a lot of information content that can be used to estimate things like whether samples have experienced a complex exposure history or what temperature history they’ve experienced during surface exposure.
So this is the goal of a prototype version 3 of the online exposure age calculator. It ingests data for all of the reasonably routinely measured nuclide-mineral pairs where production is primarily due to spallation and secondarily due to muon interactions — this includes He-3 in quartz, olivine, and pyroxene; Be-10 in quartz; C-14 in quartz; Ne-21 in quartz, and Al-26 in quartz.
For this to work, however, requires a different data entry format than in version 2. The version 2 input format is a single line of text for each sample that includes information about the sample location and other field observations; Be-10 concentration and standardization information; and Al-26 concentration and standardization information. Here is an example of an input data line:
PH-1 41.3567 -70.7348 91 std 4.5 2.65 1 0.00008 123454 3717 KNSTD 712408 31239 KNSTD
The details of what all the fields are is here. Basically, there are 9 fields describing the sample and its field characteristics, then 3 fields describing the Be-10 measurement, then 3 more describing the Al-26 measurement.
The upside of this is that it is pretty simple. You can have a line of data for each sample in a spreadsheet and then just cut and paste into the online calculator.
One downside of this is that you have to carry around a lot of zeros if you didn’t make any Al-26 measurements, which is common. Example:
CF_207_08 -80.11722 -25.78078 474 ant 4.0 2.30 0.9859 0 2.175e+06 4.880e+04 NIST_27900 0 0 KNSTD CF_204_08 -80.11717 -25.77962 474 ant 3.5 2.65 0.9859 0 5.195e+06 1.150e+05 NIST_27900 0 0 KNSTD CF_205_08 -80.11717 -25.77962 474 ant 3.5 2.30 0.9859 0 5.694e+06 1.210e+05 NIST_27900 0 0 KNSTD CF_208_08 -80.11722 -25.78078 474 ant 5.0 2.65 0.9859 0 4.820e+06 1.070e+05 NIST_27900 0 0 KNSTD CF_206_08 -80.11722 -25.78078 474 ant 4.5 2.65 0.9859 0 2.095e+06 6.980e+04 NIST_27900 0 0 KNSTD CF_209_08 -80.11571 -25.78073 429 ant 3.5 2.30 0.9859 0 1.535e+06 3.390e+04 NIST_27900 0 0 KNSTD CF_211_08 -80.11574 -25.78122 427 ant 4.0 2.65 0.9859 0 4.017e+06 9.000e+04 NIST_27900 0 0 KNSTD
The requirement that each line have the same number of fields exactly makes the job of the online calculator interface a little easier, because you can just count 15 fields and know that the next field begins a new sample; no explicit separation of the sample records is needed. But then you have to carry around all the unused fields for each sample. This isn’t that big a deal if you are only carrying around a couple of columns of zeros for missing Al-26 measurements, but what if the calculator is redesigned to handle 7 different potential nuclide-mineral pairs? Then if you maintain a single-line-for-each-sample, fixed-number-of-fields input form, you would have to carry at least another 5 x 3 = 15 colums of zeros, which is a) annoying, and b) likely to start causing errors.
A more serious downside is that you can only enter one Al-26 and one Be-10 measurement for each sample. Other nuclides, not possible.
So the prototype version 3 will still accept version 2 format input data. However, there is now a new input format that solves the above problems for samples with (or without) multiple measurements of the same or different nuclide-mineral pairs. Here is an example:
PH-1 41.3567 -70.7348 91 std 4.5 2.65 1 0.00008 1999; PH-1 Be-10 quartz 123453 3717 KNSTD; PH-1 Al-26 quartz 712408 31238 KNSTD;
What is going on here is that it takes multiple lines to describe the data for a single sample. There is a line containing the information associated with the sample — things like the location, sample thickness, appropriate atmosphere model, etc. Then any nuclide measurement you make on this sample generates an additional line. In this case there are two lines describing one Be-10 and one Al-26 measurement. The lines describing a nuclide measurement contain the sample name (so the nuclide measurement line can be linked up with the sample description line), some information to identify what nuclide-mineral pair was measured, the nuclide concentration, and some standardization info that varies by nuclide.
A couple of things:
One is that different types of lines have different numbers of fields. So the code can’t just count fields any more to tell when the next line starts; some sort of a separator is needed. In this case I’m using a semicolon; thus each line must end with a semicolon.
Two is that multiline input is a bit hard to keep track of in a spreadsheet; life is easiest when all the information for a single sample is on a single line. This shouldn’t be a problem because the separation between input data lines is defined by the placement of semicolons, and it doesn’t matter whether different “lines” are on the same line or not. In other words, this:
PH-1 41.3567 -70.7348 91 std 4.5 2.65 1 0.00008 1999; PH-1 Be-10 quartz 123453 3717 KNSTD;
is the same as this:
PH-1 41.3567 -70.73483333 91 std 4.5 2.65 1 0.00008 1999; PH-1 Be-10 quartz 123453 3717 KNSTD;
The latter will often be easier to lay out in spreadsheets.
Three is that this can handle any combination of nuclide measurements for a sample. Example:
10-MPS-046-NNS -83.27825 -58.16226 499 ant 10 2.6 0.9958 0 2010 ; 10-MPS-046-NNS Be-10 quartz 2.43E+06 3.29E+04 07KNSTD ; 10-MPS-046-NNS He-3 quartz 2.50E+06 2.73E+05 CRONUS-P 5.20E+09 ; 10-MPS-046-NNS He-3 quartz 2.34E+06 2.59E+05 CRONUS-P 5.20E+09 ; 10-MPS-046-NNS Ne-21 quartz 2.97E+07 2.88E+06 CRONUS-A 3.38E+08 ; 10-MPS-046-NNS C-14 quartz 4.71E+05 1.20E+04 ;
That sample had five different measurements made on it, some of which were replicates of the same nuclide-mineral pair. This input form allows all that information to be entered for samples that have this level of data, while not having to carry around a lot of blank fields for samples that don’t. That’s really the task that the new input form is designed for.
The prototype version 3 calculator is here. It is highly simplified: there is only one scaling scheme and it is mainly to test the new I/O formatting. However, there is some neat code to make all applicable pairs of multiple-nuclide diagrams, not just Al-26/Be-10 diagrams.
Detailed details about exactly what the fields in the new input form are are here.
If there is some fatal flaw with the new input format that I am too dumb to notice, send me an email.