Are you suggesting that atmospheric science should use laboratory quality measurements for it's research? How much accuracy do you think is necessary to compute the global mean temperature on the order of 0.01C? How much accuracy do you want from atmospheric measurements? And are you willing to apply the same standard to all fields of practice/research or do you just want the burden to be on atmospheric science?
depends on what the purpose of the study is and intended use of the data. Also on something of this magnitude I expect both accuracy and precision. Do you know the difference?
Yes I do actually. Do you how climate scientists handle the accuracy problem? Also what kind of precision is required on an individual measurement basis to compute a global mean temperature on the order of 0.01C from many measurements?
Obviously yes, as the cross-checking demonstrates it. Why do you think it's not? Show all your numbers and calculations. Don't just say "Well, it's not like what I do at my own workbench, so it must be wrong."
Yeah, I mean we collect anywhere from 10-100 million observations a day. There is no cell in the grid meshes covering the whole Earth in our reanalysis datasets in which only 1 measurement is reported. None. There is always a check and balance on each measurement and usually there are multiple measurements reported up from each grid cell sometimes hundreds or thousands. Edit: I just went to NOAA's data monitoring page. The observation count for 12 GMT (7am EST) this morning was over 25,000,000. All of these observations make into reanalysis which is run 4x per day and that's just the count for the 12 GMT snapshot.
Oh yeah, we have had two snows this year, which is rare. But I got rid of it, melted it down, in my entire county. Know how I did it, within 30 seconds? I walked outside and yelled, "trump is a great man and will MAGA" The snow was gone before I finished the sentence....the snowflakes melted. I triggered them and the heat from that trigger did the job.
you are asking why calibration done once a year isnt ok and why using temperature at one instrument to crosscheck temperature at another is not ok? Before I answer, are there any people not in the sciences that can think of why that is not ok?
You may be proud of your specialty. 90% of people (including those in science) do not know what you are talking about and do not want to know and hardly would be able to understand the simplest text of my middle school level https://www.mathsisfun.com/measure/error-measurement.html if all over suddenly they decided to learn. Of course, any educated person ( not necessarily in science) knows that an ideal thermometer of a typical weather station for a moment after calibratioin in an ideal laboratory conditions has no less than +- 1C error. What you don’t know is that the thermometers measure temperature of the atmosphere, but not how warm it is. This simple fact is not accessible to a mind of a typical college graduate and 90% of those who are in science. What you don’t know is If one ideal thermometer shows 20+-1C in Canada and another ideal thermometer shows 24C+- in Germany it does not mean the atmosphere in Germany is hotter (has more heat content) than the atmosphere in Canada. This fact is not accessible to a mind of a typical college graduate, a scientist or not. Have you ever read weather forecast and can tell what you have read? Can any of your coworkers in your occupation? http://politicalforum.com/index.php...year-on-record.517168/page-26#post-1068515025 Will air be the same hot in Germany as in Canada if both forecasts the same 30C temperature while 80% humidity in Germany and 40% in Canada? You can ask anybody you know, does matter an alarmist or denialist and see that nobody knows what I am talking about. That’s why people in science are successful in promoting the total hoax of GW/CC, no different from Flying Spaghetti Monster among the population brainwashed for 12 years in schools and then for 4 years in colleges, and PhDs are not an exclusion from that population. Can you calculate the mean of readings (different values) and the error if in Canada it is 20+-1C, 24C+-1c in Germany, 40C+-1 in Sahara, 1C-+1 in Sweden? Can you calculate the mean, average and the error of the same value of Temperature in Canada when 1 Thermometer is showing 20C+-1, another is showing 21C+1, the third 19C+-1, the fourth 22C-+1? Do you see the difference? http://www.radford.edu/~biol-web/stats/standarderrorcalc.pdf http://politicalforum.com/index.php...year-on-record.517168/page-26#post-1068509777 Nobody does, that’s why NASA, NAS, IPCC and 2432 academies of science can claim any kind of idiocy including the idiotic claim that the mean temperature of weather stations* has increased 1,51C with+-.01C error during the last century and it means the air warmed up and have it accepted as reality by both alarmist and denialists. (*The 1st weather station in Antarctica appeared in 1956 or so, but who cares.)
Yes it will be the same "hot". However, it won't have the same heat content. The dry bulb temperature is what mostly determines the phase changes of water. However, if you want a heat content or entropy based metric you need to use equivalent potential temperature (theta-e). The global mean theta-e is rising faster than the global mean temperature though so this probably isn't a good argument to work from.
It has nothing to do with a specific study. Proxy records typically have an error of about 0.05C for the global mean temperature even though surface station RMS error is about 1.0C (give or take). Reanalysis datasets are even better. Their error is about half that or even lower. The RMS error on the individual cells in the grid mesh of a reanalysis is about 2.0C (note that it's higher than the RMSE of a typical temperature instrument). Pure statistical theory says the error of the mean should be 2.0/sqrt(500,000) where 2.0 is the cited RMSE above and 500,000 is the number of grid cells that are typically averaged together. So pure theory says the error should be less than 0.01C, but in practice it's usually more due to various other factors, but still pretty close to 0.01C. The accuracy problem is handled by reframing the problem as one of a comparison between two times or an anomaly. For example, if at time Ta you get a reading of (Ra+B) and at time Tb you get (Rb+B) where R is the reading and B is the unknown bias then the anomaly is (Rb+B) - (Ra+B) = Rb - Ra. Notice how the bias B vanishes by algebraic simplification. Note that for this to work the bias B must be constant through time (or close to it). In practice it is very close to being constant because the exact same algorithm is ran on the data at all times so the Earth-temperature-measuring aggregate instrument that is created has no wear-and-tear or other temporal dependencies. This leaves only the inputs that could be biased. The trick here is that there are millions of inputs. The more inputs you have the higher the probability that biases (positive and negative) cancel each other out. With millions of inputs the probability that the biases don't cancel out (or least approach zero) is astronomical. This leaves only the bias in the aggregate instrument which as I already explain is constant and cancels when doing anomaly comparisons.
Weasel your way through and out. Depends on definition of "Yes". Hot, but not "hot". Or "hot", but not hot. Thank you for proving my point that it is so easy for your ilk to deceive 90% of typical college graduates, scientists or not. Will a a typical college graduate and 90% of those who are in science understand a this in the following: Enthalpy is the heat content of a system https://en.wikipedia.org/wiki/Enthalpy As I stated: "This simple fact is not accessible to a mind of a typical college graduate and 90% of those who are in science." As I stated: striving on these simple facts you, "NASA, NAS, IPCC and 2432 academies of science can claim any kind of idiocy including the idiotic claim that the mean temperature of weather stations* has increased 1,51C with+-.01C error during the last century and it means the air warmed up and have it accepted as reality by both alarmist and denialists". Thanks again for confirming each and every point I made: http://politicalforum.com/index.php...s-all-50-states.523983/page-6#post-1068600950 Any other comments objections to the above? theta-e....
mmmmmkayyyyyy....its their study, they are calibrating once per year and using one temp reading at one location to verify another. Anything after that isnt worth the skin cells to type about.
mmmmmkay...so you are still calibrating once per year and using one once per year calibrated instrument reading to verify another. In your science what day was the once per year calibration done on the instruments used?
It's more complicated than that. Temperature readings may be calibrated by a technician once per year, but the data assimilation system itself is a calibration mechanism. So all observations are actually calibrated and quality controlled by the automated system every single time. It actually doesn't matter if the temperature instruments are calibrated by a technician at all. There are 2 reasons for this. First, the error on our temperature readings could be 10C (or higher) and we'd still be able to compute an accurate global average because with so many other observations the signal-to-noise ratio is very high. Second, you actually don't need any temperature readings at all to know the global mean temperature. I mean literally none. In fact, our reanalysis systems are so robust that we can compute a very accurate global mean temperature (plus all of the other atmospheric properties) with nothing but barometric pressure readings. And that's just the tip of the iceberg. Google for 3DVAR and 4DVAR assimilation for more information.
sooooo...you calibrate your instruments with the instrument data....from the instrument that is calibrated once per year? But you actually dont even need real measurements now?
Yes, correct. You calibrate one instrument using another instrument exactly like it would be done by a technician or in a laboratory. I didn't say you don't need real measurements. I said you don't need real temperature measurements. In fact, electrical resistance, interferometer, etc. instruments aren't the most useful for measuring the temperature of Earth. It's easier and more accurate to do it with barometric and geopotential height readings. Proxy datasets (like GISS, NOAAGlobalTemp, HadCRUT, Berkley Land+Ocean) use surface station data (electric resistance, interferometer, etc.) instruments. Reanalysis uses hundreds of different kinds of instruments yielding millions of observations. They are the most useful for climate research. Proxy datasets are just another line of evidence that use completely different techniques as a check and balance on reanalysis. All of our datasets agree with each other within a reasonable margin. The UAH dataset is the one outlier right now, but even it shows that the Earth is warming. I'm actually not aware of any dataset that does not show a warming Earth...not a single one.
wrong, you calibrate an instrument with a certified instrument, not another uncalibrated instrument. Additionally the earth always watms and cools. The question is, is man causing it. There is no equipment on the planet that can accurately determine that.
I am unsure that you are properly educated on this topic, and quite sure you have a position you do not wish to abandon regardless of verified data. The speed with which our current warming trend is occurring is the biggest concern and indicator as it directly coincides with changes in atmospheric composition attributed to human industry. Analysis of chemical composition verifies the increases in greenhouse gas over many decades....this is confirmed data.
So is every certified piece of certified equipment until it doesnt work right, then you need to re certify it. Now, are any of your measurement devices located near any artificial or natural heat sources?
That is why it must shouted down by the corporate state and your reality president, and really now, the entire economic and political system.
My position is to ensure your analyses are in order. So far not a single one of you has shown the calibration, training and logbooks of even your latest measurement much less the first. For example, show me your first measurement. Lets look at that science and make sure its accurate. Surely you would agree that. Now, what was the first data point you are citing? Do any of you even know?
