Jump to content

Wikipedia:Reference desk/Archives/Science/2017 September 6

From Wikipedia, the free encyclopedia
Science desk
< September 5 << Aug | September | Oct >> September 7 >
Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is an 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 6

[edit]

Objectively, when has the US been lucky with tropical cyclone damage?

[edit]

Has anyone simulated enough hurricane seasons in a computer that each place gets the worst plausible storm many times and found out what the expected value of damage per year is? (either assuming the world is always like when the study was done or adjusted for less development and/or inflation in the past) What is this number? Then one could pick a date to start adding up damage totals (I'd pick end of WW2, beyond all but the old-timers' memory) and see when we've been ahead. Perhaps we were ahead right before Hugo (because of the below average early 60s-'94 Atlantic multidecadal oscillation period). Then Andrew came, then the above average 1995-future AMO period and 2004-5 and 2012 and 2017 and I wonder what year America stopped being overdue for a correction (like California is with earthquakes). Could we even still be due for a correction now? New Orleans has been hit but Miami, Tampa and Houston haven't been hit hard for a long time (if Harvey was bad then imagine a Cat 4 or Cat 5). I would be interested in the expected value of tropical cyclone damage/year without global warming (even though it would probably make a big difference in the future) Sagittarian Milky Way (talk) 01:12, 6 September 2017 (UTC)[reply]

Now That's an interesting question! There are a lot of models by climatologist that predict a long term (over 50 to 1000 years) increase, and a bunch of short-term models by meteorologists that predict what will happen tomorrow, next week, or maybe a bit longer, but I have seen nothing like what you describe. Doing that would have to predict things like El Nino, the temperature and path of the gulf stream. the path of the high altitude jet stream, etc. The climatology models can ignore all that because it averages out over those long time frames. The meteorologists can ignore all of that because they can measure the current gulf stream temperatures, etc. Also there is a lot of randomness -- the last hurricane got stuck in one place because of high pressure systems on both sides, and we cannot predict the location high or low pressure systems even a few months from now. Another oft-ignored aspect is this; gulf coast cyclones tend to hit the US and thus get a lot of study. Cyclones hitting other countries, not so much, and cyclones that stay over the ocean even less. --Guy Macon (talk) 07:02, 6 September 2017 (UTC)[reply]
You're looking for a type of ensemble forecasting. Check out Atlantic Hurricanes over the past 1500 years [1], Impact of Atlantic multidecadal oscillations on India/Sahel rainfall and Atlantic hurricanes [2], Whither Hurricane Activity [3], Normalized Hurricane Damages in the United States: 1925–95 [4]. SemanticMantis (talk) 15:07, 6 September 2017 (UTC)[reply]
  • Simulations of this sort are never very convincing, because lots of weather phenomena show a distribution resembling pink noise, also known as 1/f noise. As a consequence, events that are expected to be rare, such as "100 year storms", actually occur a lot more frequently than predicted. Looie496 (talk) 00:14, 7 September 2017 (UTC)[reply]
    Estimating the rates of rare events is difficult, but your criticism is too simplistic. If you know that your weather phenomena had a simple pink noise distribution then you would use that distribution in determining the 100-year recurrence rates, and the stats would be just fine. Often the tail of a distribution is not something that can by simply extrapolated from the common events (as fitting a pink noise distribution would imply). In some cases, the underlying physical processes are simply different. For example, the frequency of certain rain amounts during common rain storms provides next to no information about the possible rainfall during a hurricane. Much of challenge is in estimating the shape of the tail of distribution. Also, when it comes to things like floods, accurately determining a 100-year flood plain means knowing not just the amount of rain but also how it will interact with the ground. Many flooding problems have occurred due to inaccurate groundwater flow models even when the rainfall risk was predicted correctly. Dragons flight (talk) 13:52, 7 September 2017 (UTC)[reply]
With all due respect, aren't you a neuroscientist? My point being, maybe extend a little good faith to the scientists who have devoted their lives to study of atmosphere, climate, and weather. I suspect they may hold some uncharitable opinions on whatever your specific area of expertise, but in science, not all opinions are equally valid. SemanticMantis (talk) 15:34, 7 September 2017 (UTC)[reply]
Well, before I went into neuroscience I was a math graduate student focusing on dynamical systems, so I'm not totally ignorant here. But I don't think I was actually criticizing climate scientists -- the ones who do simulations know perfectly well that weather parameters often have non-Gaussian statistics, with 1/f-like distributions being surprisingly common. Typical of the uncertainties take for example the bullet-summary of the Vecchi paper you cited: "Alternative interpretations of the relationship between sea surface temperature and hurricane activity imply vastly different future Atlantic hurricane activity." Looie496 (talk) 20:57, 7 September 2017 (UTC)[reply]
  • Note that, unlike seismic stress, the energy which drives hurricanes (water temperature) does not simply accumulate until it is released in a hurricane. While this is somewhat true within a single hurricane season, each winter cools the water and "resets" the energy level. Thus, several years with no major hurricanes does not mean they becomes more likely, or less likely, than the long term average, on the following year. Global climate change, on the other hand, may do just that. StuRat (talk) 00:31, 7 September 2017 (UTC)[reply]
Right, this would not really due but the gambler's fallacy. Like heads or tails being ahead with coins. Perhaps the Atlantic multidecadal oscillation is not insignificant compared to the expected amount of random chance but I don't know the relative importances. Sagittarian Milky Way (talk) 01:08, 7 September 2017 (UTC)[reply]
Relevant articles are grey swan and black swan theory. 38.88.99.222 (talk) 13:44, 10 September 2017 (UTC)[reply]

Do rattlesnakes charge?

[edit]

Looking around the web for advice on how to handle a rattlesnake that is blocking your trail, it seems that quite a few commentators advise not to throw rocks at it or poke it, even from a distance, on the grounds that you might provoke it to attack.

Are there actual documented cases of a rattlesnake (let's say one with an easy escape option, like gliding off the trail) charging someone who gently tried to get its attention with pebbles, or by nudging it with a trekking pole? By "charging" I mean not just striking, but running after a person farther than its striking distance.

The commentators mostly say to just wait, or go around. But sometimes there's no way around, and if the snake is enjoying the sun, it might be a long wait. --Trovatore (talk) 10:07, 6 September 2017 (UTC)[reply]

Most sites I can find say that rattlesnakes are NOT aggressive (that is, attacking unprovoked), but will take defensive action when they perceive a threat. These sites all state that in nearly all cases, attacks on people are provoked, and that defensive actions are often misinterpreted as aggression; that is people may not intend to provoke a rattlesnake but may accidentally do so, and in doing so may misinterpret defensive action by the rattlesnake as aggression (i.e. passing very close to a snake, stepping on it in leaf litter, backing it into an inescapable situation, etc.) See This from the US Department of Agriculture, This from National Geographic, Here from Brittanica.com, etc. --Jayron32 12:28, 6 September 2017 (UTC)[reply]
A rattlesnake can be surprisingly good at disappearing quickly and quietly when you look away. If you throw a bunch of rocks at a rattlesnake to make it go away ... then lose track of where it ended up ... then walking through the area could be interesting. Wnt (talk) 19:52, 6 September 2017 (UTC)[reply]
As a kid we used to catch them with my uncle (he made belts from their hide). Having personally caught several dozen of them by hand my observation is that they are most certainly not inherently aggressive animals. But they can also be very stubborn and will often prefer to stay put ("stand their ground") rather than flee. If provoked they can definitely be quite defensive, although typically using "feigned" charges/strikes. The best advice I can give you is to simply go around them. They seem to be fairly perceptive insofar as intentions go, so as long as you pass by them in a non-threatening way you really shouldn't have anything to worry about. Just don't get too close, and use common sense. And by all means, DO NOT THROW OBJECTS OR POKE AT THEM! I was bitten once as I grabbed one (badly aimed swipe) and it was extremely painful to say the least (some people have severe reactions to the venom, which can sometimes be fatal) so for safety's sake always exercise caution with these creatures. 73.232.241.1 (talk) 20:16, 6 September 2017 (UTC)[reply]
Well, your experience is interesting but doesn't seem directly relevant, since your hand was in striking range. I was asking whether they will pursue beyond their striking distance. --Trovatore (talk) 20:44, 6 September 2017 (UTC)[reply]
From my experience, no, they just aren't prone to chasing people around. But again, intention is the most important thing. If they sense that they're being actively stalked or if you try to interact with them even from a distance, they will often take up a defensive posture, make striking motions, or perhaps appear to chase after you (although I've never actually seen them do much of that myself). Otherwise, you could probably pass by it even within striking range and it wouldn't do a thing. They just tend to avoid confrontation. 73.232.241.1 (talk) 20:56, 6 September 2017 (UTC)[reply]
The nature of snakes is such that they really need to face their perceived attacker to be able to strike, and they can't really back up, so they either need to turn tail and run for it, or stand their ground. A defensive retreat isn't an option. Professional snake handlers seem to like to use a very long pole, not to poke them, but to put it under their coils and push them off the path. StuRat (talk) 20:26, 6 September 2017 (UTC)[reply]
Well, not exactly. If several people surround the snake, for example, their awareness seems to become rather heightened and they can readily shift around to strike from different directions. For that very reason our approach was to have just one person approach to catch the snake. That way, they leave their guard down somewhat and thus much easier (and safer) to capture. Also, while they don't move backwards per se, they can however undulate their bodies to easily move around in just about any direction, so just keep that in mind. 73.232.241.1 (talk) 20:46, 6 September 2017 (UTC)[reply]

OK, so here's the backstory. I ran into this little guy on a hike, going up to Mount Wilson. I would guess about two feet long, but could have been three, I suppose. He was sitting very still; I couldn't see that he was moving at all. There was no way around that didn't involve getting into striking range. Oh, I suppose we could have tried to bushwhack above or below, but the hill was steep and crumbly and full of poison oak.
Eventually I started to consider the idea that he might be a former rattlesnake, pining for the fjords. I tossed rocks (really pebbles) at him to test that theory, and hopefully to gently induce him to reconsider his location if he was alive. He flicked his tongue but made no other sign of having noticed.
Finally I reached out as far as I conveniently could with my right pole and gave him a gentle nudge or two. With no sign of hurry or distress, he glided off into the brush and left us alone.
Obviously this anecdote is not meant as advice to anyone; it's just my personal experience. I never felt that I was in danger. I was trying to find out if that perception was accurate. --Trovatore (talk) 21:25, 6 September 2017 (UTC)[reply]

Yes, they're generally very easy-going creatures (although their prey might beg to differ about that!). And probably due to the very fact that they are well-aware of their own defensive capabilities, they don't seem to be easily spooked by non-threatening gestures either, which paradoxically makes them much safer to be around then say a wild boar or something; when they retreat, they do so confidently and without much fanfare. Thanks for the pic, by the way, beautiful colors that one!73.232.241.1 (talk) 21:43, 6 September 2017 (UTC)[reply]
Yes, a good comparison might be a cat. Both are absolutely deadly to mice, but will usually run from people, or watch from a safe distance, unless cornered, then watch out, and be prepared to feel the fangs (and claws in the case of the cat). Another commonality is that both might cuddle up to you while you sleep, for your body heat (which might be quite alarming in the case of the rattlesnake, and even in the case of the cat, if you're a dog person). StuRat (talk) 21:55, 6 September 2017 (UTC)[reply]
  • The crucial fact here is that a rattlesnake is almost defenseless unless coiled, so when threatened it coils up and is very difficult to persuade to uncoil except by giving it something to strike at. Consequently throwing rocks is not useful unless you can kill it, which is not easy. Poking a coiled rattlesnake with a trekking pole is dangerous -- you're too close. The proper solution is to find a tree branch at least ten feet long and use it to push the snake off the trail. It absolutely will not charge you, because as I said when "charging" it is uncoiled and therefore quite vulnerable. I have used this approach myself when I encountered a very large coiled rattlesnake in the middle of a trail on Mt. San Jacinto. Looie496 (talk) 23:53, 6 September 2017 (UTC)[reply]
    So as you can see from the picture, this one was not coiled, just "folded" once, as it were. --Trovatore (talk) 00:16, 7 September 2017 (UTC)[reply]
Would throwing cold water on it make it go away to seek out a warm place? Count Iblis (talk) 01:47, 7 September 2017 (UTC)[reply]
Possibly, but you would need a large supply of cold water, which seems unlikely on the trail. Also, you would need a way to spray the snake from out of range. A squeezable water bottle with a small hole on the top (for a straw) might work. Of course, you don't want to use up your water supply, either. StuRat (talk) 02:43, 7 September 2017 (UTC)[reply]
No, rattlesnakes don't charge, they bite for free (if the snakes are lawyers, they call it "pro bono"). StuRat (talk) 02:56, 7 September 2017 (UTC) [reply]
"I've never had an accident." "Weren't you bitten by a snake once?" "That was no accident - he did it on purpose." ←Baseball Bugs What's up, Doc? carrots03:07, 7 September 2017 (UTC)[reply]