# [a popular science] Why may occur once in "N" in every year?

Release time：2015/12/11 11:17:28

After watching many newspapers and TV can not help but sigh, this years may occur once in \"N\" of the year! Just entered the north came in 2010 for decades in a blizzard, then southwest to encounter \"fifty years in a drought, across the country are now in\" once-in-a-lifetime \"torrential rain, media reported several rivers from south to north, the highest flood peak\" millennium \"... Some netizens complained in recent years, appeared frequently occur once in \"N\", people see the aesthetic fatigue, even doubt it is avoid responsibility \"hat\" in silhouette. May occur once in \"N\" is calculated or racquet head wants to come out? If they can use occur once in \"N\"? N can be infinite amplification to thousands of tens of thousands of? ...

May occur once in \"N\" is not cycle is probability

May occur once in hydrology, \"N\" has a more scientific name \"return period\". According to long-term hydrological records somewhere, can work out how much one order of magnitude of the average years of time, namely the flood return period. The flood return period of one hundred years, there is one order of magnitude flood said local probability of 1%, that is commonly known as a once-in-a-century flood. Different places a 1% chance of flooding is not the same size, \"hundred-year flood\" perennial drought region, moved to a tributary of Yangtze river valley is likely to become \"once every 50 years\". The government when determining flood control construction scale and level, to refer to the local return period. So that we can effective flood control at the same time, avoid excessive construction. It is worth mentioning that \"return period\" to assume that \"history will be stable again\". If climate change make a dry place rain increased year by year, in the past, a 1% chance of flooding will slowly become 2%, namely \"once every 50 years\". Therefore, in the process of flood control and drought relief decision-making is not only to calculate \"return period\", also should have the trend of future climate change into consideration.

May occur once in \"N\" is not to say that every N years disasters will happen once, it only represents the degree of rare disasters in history. If somewhere for successive years reports \"N > 50 years a meet\" the same disaster, either the local climate change, let originally rare disasters occur frequently, or media abuse words. Therefore it is strongly recommended that the media before the may occur once in \"N\" is put forward, to verify the data source and the calculation process.

N is calculated, not racket head

The college entrance examination in 2007, hunan liberal arts examinee in the face of such a mathematical fills up the topic: according to the following some rivers probability distribution, choose the river, on average, at least it only happens once in one hundred flood the lowest water level. In fact, \"a once-in-a-century\" flood is calculated in this way. A river in the past 400 years, the water level of 50 meters there have been four times, probability is 1%, is the \"hundred-year flood\". Flood water level rose to 50 m this year, so next year statistical probability of 50 meters when the water level is 1.25%, so 50 meters water became \"in 80 by a flood. But statistical samples and the method itself is error, so N take a large Numbers, only commonly 50 years, for example, in one hundred.

Nt enlarge at random

May occur once in \"N\" is the historical statistics, need enough number of samples, N is smaller than fixed number of year of the data is reliable. Hydrologic records of only one hundred rivers, for example, calculate it \"once every 50 years\" of the flood level has grudgingly, let alone \"once-in-a-lifetime\" and \"millennium\"... Hydrology history is very long in our country, some areas of the recorded history of the floods and droughts to thousands of years. But factors such as precipitation, wind speed, temperature of meteorological disasters in the modern observation instruments were invented it is difficult to obtain. Beginning in the 1950 s in China, the only comprehensive building unified specification of meteorological observation network. The so-called \"fifty years once\" or even \"a once-in-a-century\" meteorological disasters is very suspicious. The conventional meteorological observation record is five hundred and sixty, and how to determine the average in the history of meteorological disasters will happen in one hundred? Actually meteorological expert reports of the commonly used term is \"precipitation (temperature) high (low) values in N years, receiving Williams became\" once every 50 years \", \"a once-in-a-century\".

The worst may occur once in N and N years

Have sent a word, \"N years of the worst\" like a shanzhai occur once in \"N\". It can be understood as the meaning of the last 50 years there was no damage, can also be a probability of less than 2% of the disaster. Its frequent let is inherently good and evil people mixed up occur once in \"N\" become more complicated. If only to describe how much of a disaster, not involving the disaster degree of rare. In order to avoid confusing, suggest to switch to \"nearly N years in\". Such media to be rest assured, the audience also see more understand.

(source: science squirrels author: snow song)