Does an 'exagram' mean anything to you? Doesn't to me. Those weights are meaningless in that format, except to geologists I suppose. (Please no-one explain to me what 'exa-' means, I don't mean that!)
Anyway, it says the median is 43±30ng/g, the mean 62±35ng/g. Why wouldn't they use the mean in the main results? That's what I imagine would give the 'concentration of mercury', the average (mean) concentration.
Reading the method, it's a huge guesstimate in a dozen ways, even much further than I would have thought. Guess that's how they get 13-73ng/g. I wonder how they have any confidence in the error range.
You just need the density of permafrost and then you can compute the volume from the exagram value, and depending on how thick you make it you get back either the surface of northern Canada and Siberia or you get a really really tall tower of permafrost that reaches into low earth orbit :-).
> I wonder how they have any confidence in the error range.
It is my understanding that there are many many core samples at various locations in the permafrost to understand its nature and make up. My guess would be that they derive the error ranges using the change in concentrations between all of the sample points leading them to a median and mean difference in concentration samples as their base set.
The mean is only a good measure of central tendency for Gaussian distributions. It is very sensitive to outliers. The median is better when you don't know what the distribution is.
It's best to just roll with it, I think. Follow the guidelines and post as substantively as you can. The population of HN is large and diverse and you're unlikely to be able to figure out why each and every downvote (or upvote, for that matter) is handed out, and even if you could, you may not agree with why they did so or think it’s fair. They can be useful as a guide and an opportunity to imagine how the comment may come across differently than you intend. But at the end of the day, it's the nature of the beast. Use what you find useful and let the rest slide.
Anyway, it says the median is 43±30ng/g, the mean 62±35ng/g. Why wouldn't they use the mean in the main results? That's what I imagine would give the 'concentration of mercury', the average (mean) concentration.
Reading the method, it's a huge guesstimate in a dozen ways, even much further than I would have thought. Guess that's how they get 13-73ng/g. I wonder how they have any confidence in the error range.