Hacker News Reader

Comment by ams6110 | Hacker News Reader

What is magical about 1880-1910 as the implied "normal" average baseline? Why not 1600? 1400? 2500 BC?

Beyond a certain time, data collection wasn't common enough, so my understanding is that people pick a baseline from a period of time when people were already collecting data in substantial amounts. I guess 2nd half of 19th century, with all the industrialization, was when people both had enough resources, and were also interested in trying to actually collect it. I think this is similar to how the baselines for all satellite-measured data (like arctic sea ice, I think) start in the 70s.

But the other comment is right too, which baseline doesn't matter so much, so long as the choice is reasonable, and used consistently. This is because, in the end, absolute values aren't meaningful on their own, only differences between them (similar to how one always reads about global temperature anomalies, rather than global temperature).

You shouldn't care about what reference is taken. The point is that compared to any reference, the curve is increasing.

The point is that compared to any reference, the curve is increasing

But that in and of itself doesn't mean anything - the Earth has warmed from two ice ages that we know of. The baseline absolutely determines what conclusions can be drawn from the data.

That's ridiculous, of course you need a baseline for any sort of evaluation. Your comment implies any sort of warming, regardless of absolute levels, is bad.

Ok, then, how about this. The baseline is 0 degrees, because that is the lowest that temperature can be.

Would that be a fair baseline?

sorry for not being clear enough. I meant that the speed of warming is of interest here. That speed is independent of the baseline chosen. And that speed is what is worrying.

That doesn't make sense. What if the reference is 5 minutes ago?

Take the reference that is 5min ago, call it T0. You will still observe that temperatures (i.e the sequence of measured temperatures minus T0) are increasing. Changing the reference changes the y-scale of a graph, but does not change the derivative (the speed at which it changes, i.e. increases or decreases).

Did you look at the graph?

The local derivative is most definitely not indicative of a general trend.

For comparison, I throw a ball down at the ground, and for the first 0.1 seconds it's moving down - can I conclude that the ball is going to keep moving down, even when it hits the ground? That's your logic.

I don't think you know what baseline means.

Baseline is just the starting point of the x-axis. It doesn't matter.

Personally I'd set the baseline to 0 degrees. Because thats how temperature works.

But you can pick whatever you want. It does not matter what you x-axis starts at. It doesn't change the data.

Or you could just have no baseline. Just remove the X-axis line from the graph.

What? Just because it's daytime (in the US) right now? In a few hours when it's dark, the temperature will definitely be going down when compared with a baseline of 5 minutes in the past. I don't think you've thought this through...

These aren't "temperatures at a specific point in the planet at a specific instant". Rather temperature averaged globally, the mean temperature of the whole planet.

Ah, I see what you meant now.

Then you could still look at the graph from 10 years ago, and compare it to your baseline of 5 minutes ago.

"baseline" just means the line that defines the 0 point on the X-axis.

It is just a flat line across the bottom graph It doesn't matter.

If we're talking about annual average temperature, what doesn't make sense is thinking about 5min ago.

800 million years ago is my reference point

see another comment because it is a period that lends itself to showing a stronger increase than exist. the simple truth is that nearly every model still shows more warming than has occurred.

there were some wonderful events in the 1880s, including one of the most amazing snow storms the NY area ever saw, can you imagine fifty foot drifts?

still there is a lot of warming but when we focus on per capita it allows many of the true violators to escape notice (Read Asia (china/india/etc)). plus looking at many sources the impact of C02 ranges from 10 to 30% so how can you factor in somethings whose range is so great. I still think a lot of industrialized raising of livestock isn't fairly accounted for as methane is more dangerous than CO2

just for fun https://en.wikipedia.org/wiki/Winter_of_1886%E2%80%9387

https://en.wikipedia.org/wiki/January_1886_blizzard

https://en.wikipedia.org/wiki/Great_Blizzard_of_1888

though Buffalo would again see thirty to forty foot drifts in 1977

Reply on news.ycombinator.com