## Hysterical About Records

Suppose you broke the Earth’s surface up into a grid of 100 km^2 squares. there would be more than 5 million squares. Now, suppose that the temperature record was 100 years long. On each square, on any given day, the odds of breaking a high temperature record are 100 to 1. So how many high temperature records would we expect in a year if the climate was completely stable?

5,000,000 * 365 / 100 = 18,250,000

That’s right, we expect there to be more than 18 million high temperature records per year in a perfectly stable climate.

People who get hysterical about record temperatures really have no idea what they are doing.

Just having fun
This entry was posted in Uncategorized. Bookmark the permalink.

### 10 Responses to Hysterical About Records

1. Peter Ellis says:

On the other hand, all things being equal you would expect to see record for coldest temperature being broken at the same frequency as record for highest temperature. This is not the case as things stand.

• Sleepalot says:

But the coldest places have the fewest thermometers.

• Given that a large percentage of weather stations are affected by UHI (pavement, snow removal, burning of fuels, etc.), it goes without saying that there will be fewer record lows than there would have been otherwise.

• Sleepalot says:

Ah yes, that an’ all.

• kramer says:

But those temperature stations affected by UHI are ‘adjusted’ in a transparent, open, and honest scientific way and as such, we should trust this ‘adjusted’ UHI temp data from these scientists because their only concern is to save us. /sarc

2. P Gosselin says:

The odds of breaking a record are absolutely nowhere near 100 to 1.
If a record is broken every 20 years on a square, then it is more like 7300 to 1. That means 250,000 new records would be set annually. That is still quite a number, though, and so your point remains valid.

• ?????

In the first year, the odds of breaking a record are 100%. In the second year, they are 50%, in the third year, they are 33%, in the fourth year they are 25%.

If you have a random number generator which generates 1000 numbers, odds are that any particular one of them is the largest are of course one out of a thousand.

• MJB says:

I think you are mixing days and years. I agree that if completely random the odds of the 100th year having a record is 1 in 100. However, the calculation in your post is based on a record being set 1 in every 100 days, not years. To make the math work you had to divide by 36500 (number of days in 100 years) to yield 182,500 records/year. Still a big number, and probably a valid point, but not 18 million.

• If today is the 100th May 17 in your record, the odds of today being a record are one out a 100.

3. suyts says: