Assuming the measurements are independent samples from a normal distribution. Which they of course aren't, as measurements of adjacent days are obviously correlated (if they were independent, a 1-in-7000 event could be expected to happen on about 2 days within a 44-year span). Now the question is what the nature of the deviation is.
- How independent are measurements of different years?
- Has there been a systematic change in the distribution mean?
- Has there been a systematic change in the distribution variance?
- Was there a good reason to assume that the temperature distribution would be normally distributed to begin with? (Maybe there are strong non-additive effects.)
In any case, it's clear that assuming the observed temperatures in the 1991-2020 range follow a normal distribution and temperatures outside that date range will follow the same distribution is a bad model of reality.
Yes there are. Scientists compare the measurements from both methods in the time when there's data from both. For example ice cores formed last decade should match direct temperature measurements from last decade. It's the same way the oldest rings from living trees are matched against the newest rings from tree fossils, and radiocarbon dating is checked against all of that.
That does not look very reliable to me, because that implies certain things are only affected by ambient temperature.
Btw, Can you tell me how ancient temperature is measured from ice cores? My lookup only says we can detect atmospheric composition, and not temperatures from the ice cores.
>Past precipitation can be used to reconstruct past palaeoclimatic temperatures. δD and δ18O is related to surface temperature at middle and high latitudes. The relationship is consistent and linear over Antarctica[9]...
I think this is also flawed, because this computation requires the ratio of the isotopes in the atmosphere at the period of interest, which we don't have. And the people making this measurement just seem to assume it would certain value derived from the present, which is also unreliable...
So yea, all things considered, pretty unreliable..
See, since it seems that you are convinced yourselves, can you tell me how accurately the past temperature can be assessed by using this method? I mean +/- how many degrees?
Also share if this accuracy requires any implicit assumptions. For example, "If we assume that X today is true then it is true back then as well" where X is not a physical constant...
See, since it seems that you are convinced yourselves, can you tell me how accurately the past temperature can be assessed by using this method? I mean +/- how many degrees?
Can you tell me how many research papers on this specific subject you have read, before coming to that conclusion?
Hint: chances are you're correct (!). Size of the error bars, timeframe(s) talked about, how data was gathered, how results were processed etc, that is what such papers are about. Educate yourself! (Wikipedia is usually a good start)
None. It goes against my common sense. That is why I am asking what accuracy can this method give. I am asking that to people who are already seemingly convinced that the methods listed (fossil records, ice-cores) can be accurate..
> That applies to things you're experiencing on a daily basis.
That we cannot accurately look into the past unless we explicitly record it, is something that I experience on a daily basis.
Seriously, if you are so convinced, then why don't you just give me an accuracy measure and reasons to support it? I am not interested in anymore articles that does not contain an explicit accuracy claims and reasons to support the same.
So please understand this. It is not that I am questioning the existence of such techniques. I am questioning their accuracy/reliability..
> So please understand this. It is not that I am questioning the existence of such techniques. I am questioning their accuracy/reliability..
That is fine. Scientists do the same, all the time.
> we cannot accurately look into the past unless we explicitly record it
Please allow me an opportunity to enlighten you:
We cannot look into the past unless it is recorded somehow.
Note the absence of "we" in the recording part. Also note "somehow" may happen in many different ways.
Sensor data, satellite measurements, ice cores, fossilized creatures captured in stone, slices of tree trunks, samples of minerals whose radioactivity decayed over millenia, history books or landscape features, are just different types of records. Each with their own set of properties (accuracy being 1 of them).
So I am aware. The "we" is important because I said "accurately". When "we" set up the recording, we can make it to record with the required precision. Inferring measurements from incidental things can give the required precision, but it should be justified.
It seems like you're on the trajectory to refuting all historical measurements altogether. In that case, I can't help you. Good luck. I suggest the philosophy of Last Tuesdayism, it may interest you.
I only reject using them for certain things that require vastly more reliable data. And I am not making the mistake of relying on unreliable data just because it is the best we can manage.
Also why don't you answer my question about ice cores?
I'm not an ice core scientist. I suggest you ask them for detailed specifications. But the way you have posed the question strongly suggests that you will not be satisfied with any possible answer. Anything short of your eyes laying upon a thermometer will always be "not reliable enough", and thermometers only record the current temperature.
> I'm not sure how you can make that claim with only 29 years of data without making some pretty big assumptions about the underlying distribution.
Under what circumstances would 29 years be an insufficient mean to compare to the last few years? Precisely those circumstances in which the climate had changed.
Under any sane statistical model. If we assume that this is a naturally occurring once in century El Nino under an equal probability model we'd expect not to see it before we have 70 years of records.
Within the 1982-2026 span there is an equally negative 3.5 sigmas deviation somewhere (no year labels on the graph). The article doesn't touch on it at all so I have no context as to what it could be. But it definitely suggests 3.5 sigmas is not really 1 in 7000 years.
