![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
kareinaI've known for a while that the fact that none of my samples, from anywhere across Tasmania, has much calcium in it is a bit of a problem in terms of trying to get decent estimates of pressure. Today I've read a paper which indicates that it may be worse than I thought. It is quite common to work out pressure of metamorphism based on the Ca content in garnet (specifically grossular) and feldspar (specifically anorthite). However, if there isn't enough calcium, the errors for the estimates get larger and larger as you have less and less calcium. I knew that this was an issue for my feldspar, and have been concentrating on using a program that models the garnet composition, rather than trying for those methods which require feldspar, garnet +/- other minerals.
It turns out that my instinct to avoid those was good--the above paper gives a rule of thumb about how much calcium one needs in grossular or anorthite in order to trust the pressure estimates. It suggests that I need the mole fraction of grossular to be at least 0.15. Check: of the 1,043 different spot analyses done on my garnets, 7% of them have that much or more grossular. It goes on to say that below that you should use caution when estimating pressure, but if it drops below 0.03 your errors are getting much, much too huge. Alas, fully 55% of my garnet is below 0.03 grossular. It also suggests how much anorthite one needs in one's feldspar. It suggests that ideally you should have better than 0.98 anorthite, but if it goes below 0.13 anorthite the results really can't be trusted. Hah. Not one of my samples is higher than 0.15 anorthite, and only 3 out of 233 feldspar spots analyzed was greater than 0.13!
However, ever the optimist, when it further suggested that if you multiply your grossular by your anorthite that so long as their product was greater than 0.05 you can probably trust the pressure estimates, I decided to check to see if any of my samples happen to do that--after all, some each get into kind of reasonable numbers, perhaps they do so in the same sample. No such luck. of the 23 samples I've analyzed which contain both garnet and feldspar, if you multiply the single highest grosslur content by the single highest anorthite content (without taking into account where these analyses are located within the sample, which, honestly, would matter if the answer were reasonable) the best of the best comes in at 0.025, which is only half as good as it would need to be.
Now I need to decide what can be done despite this so that I have results that are worthy of putting into my thesis. Before the 1st of June.
It turns out that my instinct to avoid those was good--the above paper gives a rule of thumb about how much calcium one needs in grossular or anorthite in order to trust the pressure estimates. It suggests that I need the mole fraction of grossular to be at least 0.15. Check: of the 1,043 different spot analyses done on my garnets, 7% of them have that much or more grossular. It goes on to say that below that you should use caution when estimating pressure, but if it drops below 0.03 your errors are getting much, much too huge. Alas, fully 55% of my garnet is below 0.03 grossular. It also suggests how much anorthite one needs in one's feldspar. It suggests that ideally you should have better than 0.98 anorthite, but if it goes below 0.13 anorthite the results really can't be trusted. Hah. Not one of my samples is higher than 0.15 anorthite, and only 3 out of 233 feldspar spots analyzed was greater than 0.13!
However, ever the optimist, when it further suggested that if you multiply your grossular by your anorthite that so long as their product was greater than 0.05 you can probably trust the pressure estimates, I decided to check to see if any of my samples happen to do that--after all, some each get into kind of reasonable numbers, perhaps they do so in the same sample. No such luck. of the 23 samples I've analyzed which contain both garnet and feldspar, if you multiply the single highest grosslur content by the single highest anorthite content (without taking into account where these analyses are located within the sample, which, honestly, would matter if the answer were reasonable) the best of the best comes in at 0.025, which is only half as good as it would need to be.
Now I need to decide what can be done despite this so that I have results that are worthy of putting into my thesis. Before the 1st of June.