Wikipedia:Reference desk/Archives/Science/2023 February 8

= February 8 =

Toaster (not kitchen appliance)
Did the AEM-7 have any sort of dynamic brakes, or did it rely completely on air brakes? 2601:646:8A81:6070:B1AD:1112:B1A9:74A2 (talk) 03:59, 8 February 2023 (UTC)


 * This manual says on pages 1-11 and 1-12 that it had dynamic brakes as well as air brakes, and could blend the two methods automatically to produce uniform braking effort. --174.89.12.187 (talk) 11:37, 8 February 2023 (UTC)


 * Thanks! So, it did have dynamic brakes, but it didn't have a separate control for them (the one brake handle controlling both air brakes and dynamic brakes automatically) -- is that right? 2601:646:8A81:6070:78AA:2859:9D19:FCF4 (talk) 02:16, 9 February 2023 (UTC)


 * See section 3 of the manual. --174.89.12.187 (talk) 05:41, 9 February 2023 (UTC)

Difference between a mean and an average
In Christchurch (and in the articles for many other cities) the climate data table shows both Mean maximum temperatures, and Average high temperatures. They are significantly different. I would have expected them to mean the same thing. What are the meanings of the two terms? HiLo48 (talk) 05:53, 8 February 2023 (UTC)
 * Suppose it's based on data for 10 years. Measure the highest temperature every day in January, 310 days in total. The average high is the average of those 310 temperatures (you get the same result if you first compute the average high of 31 days each year and then take the average of those 10 numbers). Now measure the highest overall temperature in January for 10 years, only one temperature each year. The mean maximum is the average of those 10 temperatures. PrimeHunter (talk) 06:28, 8 February 2023 (UTC)
 * So, suppose we have a two-years period with a 3-days January in each and the highest temperature in corresponding six days of the two months are (10, 12, 16) and (14, 11, 9). Then the maximum January temperature in the two years are 16 and 14. Consequently, the average high would be (10 + 12 + 16 + 14 + 11 + 9)/6 = 12 (an average of daily highs) and the mean maximum would be (16 + 14)/2 = 15 (the average of monthly maxima). Is that right? --CiaPan (talk) 08:33, 9 February 2023 (UTC)
 * Wikipedia has articles on mean and average. Shantavira|feed me 09:19, 8 February 2023 (UTC)
 * I teach mathematics. I know what those words mean in normal situations. The definition being used in our climate tables is a bit specialised. HiLo48 (talk) 10:00, 8 February 2023 (UTC)
 * This is my guess. For any given day of the year, say 13 March, take the highest high ever recorded on that day. Average these over a month, and that is the Mean maximum for that month. The averages of all recorded highs in a given month is the Average high. --Lambiam 13:26, 8 February 2023 (UTC)
 * In the context of weather "Maximum" usually means the record for that day, "High" usually means the greatest temperature on that day specifically. So, for example, where I am right now, in Raleigh, North Carolina, the high temperature today on Feb 8, 2023, is 69 F.  However, according to my research, the maximum temperature on this date was 75 FFeb 8, 2019.  So, the "average high" is the average high temperature (probably a rolling average over a ten year period) for all of the days in a given month, whereas the mean maximum is the mean value of the 28-31 maximums for each day of that month.  That's my best guess.  -- Jayron 32 17:26, 8 February 2023 (UTC)
 * High and low temperature are meteorology terms of art meaning the highest and lowest recorded/predicted temps on a given day. So the "average high" is, take all the high temps over the given time period (stated in the header) and compute their average. "Mean maximum" is the mean of every maximum temp, from every day in the month, for that month. It being the row immediately after "Record high" is meant to be a clue to this. The rows progress from "how hot it can get", to "the typical temperatures", to "how cold it can get".
 * Anyone think high and low temperatures is article-worthy? --47.147.118.55 (talk) 07:25, 9 February 2023 (UTC)
 * Not as an explanation of the problem I brought here. That involves somewhat complex terms. HiLo48 (talk) 07:53, 9 February 2023 (UTC)
 * I was thinking the general topic might have enough to it, either for an article or a section in an existing one to point it at. How are highs/lows predicted in forecasts, how do they interact with climate, how does meteorology record and check them. --47.147.118.55 (talk) 22:16, 9 February 2023 (UTC)

I've always assumed that the "mean maximum", when given for an entire year, is the mean of the annual maximum temperatures. If that interpretation is correct, it means that only one temperature per year is taken into account. The "average high" will consider every day's high, so you'll have 3650 datapoints when considering ten non-leap years. Conversely, the "mean maximum" will consider every year's high, so you'll have 10 datapoints when considering ten non-leap years. Has anyone else heard such an interpretation? I think it conflicts with Lambiam's and Jayron's guesses.> Nyttend (talk) 19:49, 15 February 2023 (UTC)

HiLo48, I think I've found the answer, and it's complicated :-( The second page of Google search results included a Wikipedia result, Template talk:Weather box/Archive 8, which has a couple of discussions on the subject.  It looks like the template is based on US weather terminology, in which "Monthly Highest Max Temperature" is a kind of data, and "Mean maximum" is the mean of those twelve datapoints.  So at least this should be correct for US locations.   Conversely, we have a different definition here; BOM defines "Mean maximum temperature" as The average daily maximum air temperature, for each month and as an annual statistic, calculated over all years of record.  To me this sounds identical to an average high.  (And who knows what the Kiwi definition is? I've tried restricting a Google search to .nz sites, but didn't find anything official.) If different countries use this term for different figures, probably a lot of Wikipedia articles are using one definition and others are using another.  Nyttend (talk) 20:01, 15 February 2023 (UTC)
 * Thanks Nyttend. As an Australian who once worked at the BOM, that makes a lot more sense to me. HiLo48 (talk) 06:43, 16 February 2023 (UTC)
 * The mean of twelve datapoints, one for each month, is a single value. The Mean maximum row in these tables gives first twelve values, one for each month, and then a year value that is higher than the mean of the month values – even higher than their highest value. --Lambiam 20:06, 16 February 2023 (UTC)

Turkey earthquake
How do animals know in advance when there is an earthquake coming? And why can't scientists predict when an earthquake will occur? 163.202.220.51 (talk) 11:05, 8 February 2023 (UTC)
 * See earthquake prediction, and in particular earthquake prediction. --174.89.12.187 (talk) 11:40, 8 February 2023 (UTC)
 * As it happens, I just saw a NOVA program from 2011 or so, in which they talked about the P-waves and S-waves, and how detection of P-waves can be used to provide a warning about coming S-waves, which the show said are usually more destructive than the P-waves. So it's not quite true that scientists flat-out cannot predict an earthquake. It depends on where they are, and the amount of study that's been done. It's also a bit like predicting tornadoes. When a tornado is sighted, warnings can be issued, but taking cover immediately is essential. ←Baseball Bugs What's up, Doc? carrots→ 12:49, 8 February 2023 (UTC)
 * And, places now have earthquake warning systems that detect the P-waves and, because things like light pulses in optical fiber and electrical signals can "outrun" the S-waves, broadcast emergency warnings. This has been operational in Japan for years, and one is now operating on the U.S. West Coast as well. Also, hey, California earthquake forecast is a blue link. These forecasts are on the time scale of decades however, which not everyone finds satisfying enough. --47.147.118.55 (talk) 07:33, 9 February 2023 (UTC)
 * Also, when certain types of rocks experience a change of pressure or "tidal loading" (i.e. shortly before a quake) they release radon gas; one hypothesis is that some animals can detect this.[citation needed] 136.56.52.157 (talk) 17:15, 8 February 2023 (UTC)
 * The ShakeAlert system is currenty operational in California, Oregon and the State of Washington. It can give up to 45 seconds advance warning. Cullen328 (talk) 22:23, 9 February 2023 (UTC)


 * The prediction of earthquakes is not currently feasible as our article says. Fault systems and their rupture behaviour are just too complex to model successfully. The best that we can do is to forecast the likelihood of an earthquake (or more precisely a particular peak ground acceleration) over a given period of time. Maps are routinely published using this approach, which helps with planning, building codes etc. The presence of radon anomalies associated with earthquakes is just one of the many approaches that have been tried to actually predict them, but again there are just too many false positives and negatives with all the methods that have been used. This BBC report is a good summary. Mikenorton (talk) 09:06, 13 February 2023 (UTC)