A graph that should be front-page news
lyrebirddreaming.com
A graph that should be front-page news
1–10 of 469 posts
Re: A graph that should be front-page news
#2https://www.lyrebirddreaming.com/post/el-ni%C3%B1o-isn-t-an-...
Re: A graph that should be front-page news
#3Re: A graph that should be front-page news
#4Re: A graph that should be front-page news
#5Looking at the graph left me wondering just what it means exactly. I'm not well versed in statistics so "the standard deviation is 3.5°C" doesn't mean much. Also, what's up with that other line going down to -3.5°C? And what do the colors mean? In the sense that I'm not sure whether a darker blue means closer to or further from today.
standard deviation is a measure that informs about the distribution. A high standard deviation means a "wide bell curve". A low standard deviation means that all values are closely clustered around the middle of the curve.
So if your value is 2 x standard deviation (for example) that means it is a relatively rare outlier, since 2 x standard deviation covers 95% of the bell curve. In particle physics I believe they require 5 standard deviations to confirm an observation.
Re: A graph that should be front-page news
#6Looking at the graph left me wondering just what it means exactly. I'm not well versed in statistics so "the standard deviation is 3.5°C" doesn't mean much. Also, what's up with that other line going down to -3.5°C? And what do the colors mean? In the sense that I'm not sure whether a darker blue means closer to or further from today.
Re: A graph that should be front-page news
#7Looking at the graph left me wondering just what it means exactly. I'm not well versed in statistics so "the standard deviation is 3.5°C" doesn't mean much. Also, what's up with that other line going down to -3.5°C? And what do the colors mean? In the sense that I'm not sure whether a darker blue means closer to or further from today.
The baseline of 0.0 represents the average of all years. Anything above / below the baseline is a (standard) deviation from the average. The blue lines are the individual years since 1991 [1] while the red line is the year 2026.
If a line is above the baseline, then the sea-surface temperature was hotter on that day than average. If below, it was cooler than average.
The year 2026 is an outlier, dwarfing all the others starting around June / July. The Nino 3.4 sea-surface temperature is significantly hotter than any previous year during that time. New record, I guess?
[1]: I'm confused about the two date ranges given: 1982-2026 and 1991-2020. I'm assuming this graph is based on measurements from 1982-2026 to calculate the average, but the lines shown are only from 1991-2020, for some statistical reason I don't understand.
Re: A graph that should be front-page news
#8According to one comment on the site, the 3.5 means "3.5 times the SD", which makes much more sense to me.
I initially tried to make sense of "SD being 3.5 on that day of the year", which seems to be a wrong interpretation.
Re: A graph that should be front-page news
#9Looking at the graph left me wondering just what it means exactly. I'm not well versed in statistics so "the standard deviation is 3.5°C" doesn't mean much. Also, what's up with that other line going down to -3.5°C? And what do the colors mean? In the sense that I'm not sure whether a darker blue means closer to or further from today.
There's is an interactive chart that's easier to understand