Jump to content

Wikipedia:Reference desk/Archives/Science/2023 September 21

From Wikipedia, the free encyclopedia
Science desk
< September 20 << Aug | September | Oct >> September 22 >
Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.


September 21

[edit]

Becquerels released in nuclear explosions

[edit]

https://en.wikipedia.org/wiki/Becquerel#Examples states

whereas https://www.sciencedirect.com/topics/earth-and-planetary-sciences/fission-weapon states

How is it possible that a small nuclear explosion release over 30000 times more becquerels than all atmospheric tests combined?

Thanks,
cmɢʟeeτaʟκ 00:04, 21 September 2023 (UTC)[reply]

According to Nuclear weapons testing, "Nuclear explosions close enough to the ground to draw dirt and debris into their mushroom cloud can generate large amounts of nuclear fallout due to irradiation of the debris. This definition of atmospheric is used in the [1963] Limited Test Ban Treaty, which banned this class of testing along with exoatmospheric and underwater." That would explain at least part of it. Clarityfiend (talk) 13:18, 21 September 2023 (UTC)[reply]
No, it's some sort of dumb error. Abductive (reasoning) 15:17, 21 September 2023 (UTC)[reply]
Which part do you think is in error, @Abductive:? cmɢʟeeτaʟκ 16:02, 21 September 2023 (UTC)[reply]
Lessee, the article on says Little Boy had 64 kg of highly enriched uranium, and added 8 YBq to the environment. The article on the Chernobyl disaster says "8.5×1016 Bq equals 24 kilograms of caesium-137". These are, what, 7 orders of magnitude apart? Something's wrong. And it looks like it might be Wikipedia. Abductive (reasoning) 18:06, 21 September 2023 (UTC)[reply]
Even "highly enriched" uranium is not very radioactive. U-235 has a half-life of around 700 million years, whereas U-238 has a half-life of 4 billion years, and HEU is still mostly U-238. On the other hand cesium-137 has a half-life of about 30 years, so it's something like 8 orders of magnitude more radioactive than HEU per unit mass.
But of course the radioactivity from Little Boy didn't come (mainly) from the HEU, but from its fission products, which were immensely more radioactive. Direct comparison is challenging based just on this. --Trovatore (talk) 19:00, 21 September 2023 (UTC)[reply]
7 orders of magnitude. Abductive (reasoning) 19:47, 21 September 2023 (UTC)[reply]
7 orders of magnitude sounds too small to me. Why do you seem to think that it is too big? (If I understand your position correctly.)Liberté2 (talk) 20:31, 21 September 2023 (UTC)[reply]
As can be seen below, the huge number for Hiroshima in the Becquerel article is grotesquely inflated by counting all the energy released as if it was radioactive fallout. Abductive (reasoning) 02:08, 22 September 2023 (UTC)[reply]
According to our article, Little Boy's uranium was 80% U-235.
At what time you measure this radioactivity also matters. Little Boy injected quite a lot of radioactivity into the atmosphere one second after detonation. Within minutes, much of that had already decayed to stable isotopes. If you measure fallout, you're talking about the radioactive materials that make it back to the ground in the weeks after detonation. You can get the number for the Hiroshima explosion almost as high as you want by counting the very short lived fission products. Which makes the number meaningless. PiusImpavidus (talk) 20:01, 21 September 2023 (UTC)[reply]
Oh, I bet that's it. Sure, if you look at the first second after detonation, then you've got all those nuclides with half-lives on the order of a second, disintegrating at a furious rate. Doesn't mean much for humans compared to the direct radiation from the bomb, but technically it's ultra-high radioactivity for that short time. --Trovatore (talk) 20:39, 21 September 2023 (UTC)[reply]
@Trovatore,@PiusImpavidus,@Liberté2,@Clarityfiend,@Abductive: Thanks for your explanations. Does that mean Becquerel#Examples should be changed in any way? Thanks, cmɢʟeeτaʟκ 09:10, 24 September 2023 (UTC)[reply]
Absolutely, yes. While you could theoretically estimate the distribution of isotopes at the moment of the Hiroshima explosion and estimate the activity of each one, and then make a (horribly complicated) table of all the daughter products and their quantities at any given time after, that's NOT a useful example for understanding the concept of the unit, even if the number given is accurate. I'll change it. PianoDan (talk) 23:12, 25 September 2023 (UTC)[reply]
Thanks, @PianoDan:. cmɢʟeeτaʟκ 11:08, 28 September 2023 (UTC)[reply]

Intelligence and Biases

[edit]

Which biases are correlated with intelligence (g)? Which are positive, which are negative, and which are independent? 2A02:8071:60A0:92E0:418B:548F:263F:56D7 (talk) 18:31, 21 September 2023 (UTC)[reply]

Your question is way too vague. ←Baseball Bugs What's up, Doc? carrots19:06, 21 September 2023 (UTC)[reply]
There is a very long list of biases at bias and bias (disambiguation). I suggest you consult those articles or clarify your question. Shantavira|feed me 08:39, 22 September 2023 (UTC)[reply]
From what I've seen there is very little connection between intelligence and the amount or type of biases. More intelligence just makes for better rationalization. And unfortunately as far as AI sytems are concerned it seems the better ones are also getting better at not only incorporating biases but also at having the same types of failure of reasoning as people. As to whether they are good or bad - that's what I'd call a moral or policy question rather than a science one. NadVolum (talk) 13:41, 22 September 2023 (UTC)[reply]

RESCUE REQUEST (Sorry, I misspoke last time ; addition of a sentence)

[edit]
Trolling
The following discussion has been closed. Please do not modify it.

🆘 I CAN'T STAND IT ANYMORE, cryonics is said to be too neuron-destroying for the technologies of the future to do anything. I did come up with a process to do something for all these people, but given the previous recital, we need to focus on preserving neurons in good conditions. I need a process (cold or otherwise) that works. Could you please send a message "Extreme urgency: current Cryonics destroy too many neurons and vitrifixation does not preserve the excitability threshold of synapses and is said to destroy neurons too, could you please immediately launch research into finding a satisfactory process" to the authorities, public services and non-profit organisations.

What we are going to do is, instead of using classic cryonics followed by dismantling of the brain into individually excited neurons by computer (thanks to a computer simulation (based on the data collected during dismantling)) - to do this we would also have to dismantle each neuron into each of its parts in order to recover information about (among other things) the excitability threshold of the synapses - use vitrifixation with preservation of the excitability threshold of the synapses and with non-destruction of the synapses and neurons (a process yet to be invented); This will allow us, thanks to the downloading of data from the mind, to simply dismantle the brain by removing the neurons one by one without having to dismantle them themselves, which makes this operation much easier.

Could someone here please take the administrative steps to organise a march in Washington DC ? 2A01:CB0C:C45:E000:94D6:C59A:5856:535F (talk) 18:58, 21 September 2023 (UTC)[reply]

combinatorics of chromosome

[edit]

In a discussion with friends, the topic of how some conditions, such as autism and ADHD, are more common for males. For example, Sex and gender differences in autism.

Someone brought up the XX vs XY issue:

Women have two X chromosomes, and men have one X and one Y chromosome. This changes the combinatorics of recombination (meaning: the likelihood of various combinations of genes appearing together over generations of reproduction). 

I don't know much about genetics. Is this actually a factor?

On the one-hand, it sounds perfectly reasonable and there's no reason to doubt it. On the other hand, it also sounds very "pop-sci". Liberté2 (talk) 20:30, 21 September 2023 (UTC)[reply]

This is the idea that having only one X chromosome unmasks recessive genes responsible for autism, as seen in color blindness, right? I think that there isn't much evidence for it. Abductive (reasoning) 01:58, 22 September 2023 (UTC)[reply]
The reasoning is sound, and it works because each chromosome has its own allele for each gene, meaning that in the case of X and Y chromosomes, there are effectively four in total. Since there are four gene alleles, there is a greater chance that one of them will be affected. Another consequence of having four alleles is that the severity of the autism is variable depending on how many alleles are affected. With the Y chromosome being as small as it is, there is little opportunity for meiotic exchange of material and thus it tends to be affected more frequently in a heterozygous state. Hope this helps. Buckrune (talk) 04:02, 22 September 2023 (UTC)[reply]
I am not sure where Buckrune gets the idea of FOUR alleles, nor the argument about a greater chance of being affected the more copies there are. In a diploid species like humans each individual has TWO alleles per locus (if the gene lies on an autosome, i.e. not a sex chromosome). If one chromosome has a defective allele, usually the allele on the sister chromosome will not be defective and that one copy will be sufficient to maintain function, hence masking the defect. But in the case of a defect in a gene on the X chromosome, in men the Y chromosome would not have this "backup" copy to maintain function. It is more complicated in women, because in each cell one X chromosome is inactivated and which it is tends to be inherited by daughter cells; this causes the patchiness of colouration in tortoiseshell cats. That is the principle behind sex-linked genes. But whether any of the genes involved in autism are sex-linked is another issue that I cannot answer. Mostly the reasons for correlations between between phenotype and sex are not because the genes involved are on the sex chromosomes. JMCHutchinson (talk) 08:58, 22 September 2023 (UTC)[reply]
The thing is, the genes on the X being fully expressed in males means that natural selection takes a cold, hard look at them every time. Aside from fragile X syndrome, there aren't many genes implicated in autism on the X. And there are hardly any genes at all on the Y. As to the idea that recombination itself has anything to do with an increased likelihood of autism or any other disease or condition, that's not how recombination works. Abductive (reasoning) 09:57, 22 September 2023 (UTC)[reply]
Good points, but I would add that although selection indeed ought to weed our defects on the X chromosome quicker than those on autosomes, a number of X-linked diseases are not so rare, including haemophilia and Duchenne muscular dystrophy. JMCHutchinson (talk) 06:40, 23 September 2023 (UTC)[reply]
There's an argument that particularly devastating conditions that affect children are not selected against as strongly as one might suppose. Instead, the selection pressure is a wasted pregnancy – a stillbirth that lives for a while. Abductive (reasoning) 12:44, 23 September 2023 (UTC)[reply]

Blacklight/UV light

[edit]

Browsing through Amazon, it seems like there are two main types of UV/black lights: those in the 395-400 nm band and those at 365-370 nm. What is the functional difference between the two? The ad copy seems largely the same. Am I right in thinking the 365 nm one is not functionally visible unless it's fluorescing something, while the 395 nm will appear purple? Anything else? Matt Deres (talk) 21:33, 21 September 2023 (UTC)[reply]

Here are a couple of pages from lighting manufacturers discussing the differences: 1 2 CodeTalker (talk) 23:47, 21 September 2023 (UTC)[reply]
Thank you - very helpful! Matt Deres (talk) 11:10, 22 September 2023 (UTC)[reply]