4 ms·

So, a clueless Luddite like me went and pulled the temperature reading database since 1903 and did the following: - search for all United States - excluded all

by egberts1 10mo ago

So, a clueless Luddite like me went and pulled the temperature reading database since 1903 and did the following: - search for all United States - excluded all 14,200 US stations installed since 1973 - average/median them the rest Trend? Largely flat. Did I jump the gun? Were excluded stations like number 80238 (Arcadia, FL) being installed near heat-producing objects (air conditioning condenser unit, nuclear cooling towers, new parking lots) the cause? A valid cause?

renewiltord 10mo ago
The other thing is that people use the classic chart trick of not starting the Y axis from zero Kelvin (the only scientific scale). If you do that you see that even including all these stations the trend is imperceptible.
- trehalose 10mo ago
  On that scale, the difference between normal human body temperature and dangerous hyperthermia is just about imperceptible too. Even the difference between summer and winter is pretty small. Dunno why we bother with heating and air conditioning and having two sets of clothing.
chaps 10mo ago
Wait so, in response to an article about the current year being hot, you excluded the past 50 years and made a conclusion that things are flat. Am I missing something from your post?
- renewiltord 10mo ago
  Not excluding the 50 years. Excluding stations installed then. Older stations still report. You have to tune the number right. If you choose 40 years it doesn’t work. So something must have happened in 1973.
polotics 10mo ago
May i suggest you consult the website realclimate.org? this and many other classic denialist tropes are well addressed there. Thank You!
- ctenb 10mo ago
  Care to paraphrase the relevant explanation?
  - rcxdude 10mo ago
    https://www.realclimate.org/index.php/archives/2007/07/no-man-is-an-urban-heat-island/ https://www.realclimate.org/index.php/archives/2007/07/no-ma... Basically, it's something that's taken into account. The two main ways are calibrating urban stations relative to nearby rural stations, and by looking at the variation between windy and calm days, since the effects are larger on calm days (so if things are corrected well, there shouldn't be a difference)
  - polotics 10mo ago
    Please no, because this would be feeding the sea lions.
rcxdude 10mo ago
It's not something that climatologists and meteorologists are unaware of. Even for longer-running weather stations you can get a heat island effect if they're in a city and it's built up over time. So while I can't say exactly why you got the result you did with this specific query and dataset it is something that is taken into account when analyzing this kind of data.
sam_lowry_ 10mo ago
Show me the source, Luke!
gamerdonkey 10mo ago
I have many questions about your methodology: - How many stations were you left with? - Did that number decline over time (as you excluded replacement stations)? - What was your scale? - Are you willing to share your results? - How is this still the top comment after 30 minutes? But your comment touches on a common misconception, which is that heat islands must be excluded to accurately measure the overall temperature. You refer to the idea as "heat-producing objects", but I would argue that a parking lot is more of a heat reflecting object. More to the point, even heat islands must be considered as part of the worldwide climate, simply because they are part of this wide world. Their heat does not simply disappear (I hope you agree that would violate physics). Imagine we want to measure the average temperature inside a single 30-foot by 10-foot room during winter. We have two probes: one near a burning fireplace on one end of the room, and one near a window on the other. If we excluded data from one probe or the other, do you believe we would get an accurate average reading? Of course, when scientists are calculating a global temperature, they have to handle special cases in the data (like heat islands). This has been known for some time, and you can read more about it here: https://www.realclimate.org/index.php/archives/2007/07/no-man-is-an-urban-heat-island/ https://www.realclimate.org/index.php/archives/2007/07/no-ma... I fear that I've spent too much time responding to this, but I wanted to take it on in earnest.

renewiltord 10mo ago

The other thing is that people use the classic chart trick of not starting the Y axis from zero Kelvin (the only scientific scale). If you do that you see that even including all these stations the trend is imperceptible.

trehalose 10mo ago
On that scale, the difference between normal human body temperature and dangerous hyperthermia is just about imperceptible too. Even the difference between summer and winter is pretty small. Dunno why we bother with heating and air conditioning and having two sets of clothing.

chaps 10mo ago

Wait so, in response to an article about the current year being hot, you excluded the past 50 years and made a conclusion that things are flat. Am I missing something from your post?

renewiltord 10mo ago
Not excluding the 50 years. Excluding stations installed then. Older stations still report. You have to tune the number right. If you choose 40 years it doesn’t work. So something must have happened in 1973.

polotics 10mo ago

May i suggest you consult the website realclimate.org? this and many other classic denialist tropes are well addressed there. Thank You!

ctenb 10mo ago
Care to paraphrase the relevant explanation?
- rcxdude 10mo ago
  https://www.realclimate.org/index.php/archives/2007/07/no-man-is-an-urban-heat-island/ https://www.realclimate.org/index.php/archives/2007/07/no-ma... Basically, it's something that's taken into account. The two main ways are calibrating urban stations relative to nearby rural stations, and by looking at the variation between windy and calm days, since the effects are larger on calm days (so if things are corrected well, there shouldn't be a difference)
- polotics 10mo ago
  Please no, because this would be feeding the sea lions.

rcxdude 10mo ago

It's not something that climatologists and meteorologists are unaware of. Even for longer-running weather stations you can get a heat island effect if they're in a city and it's built up over time. So while I can't say exactly why you got the result you did with this specific query and dataset it is something that is taken into account when analyzing this kind of data.

sam_lowry_ 10mo ago

Show me the source, Luke!

gamerdonkey 10mo ago

I have many questions about your methodology: - How many stations were you left with? - Did that number decline over time (as you excluded replacement stations)? - What was your scale? - Are you willing to share your results? - How is this still the top comment after 30 minutes? But your comment touches on a common misconception, which is that heat islands must be excluded to accurately measure the overall temperature. You refer to the idea as "heat-producing objects", but I would argue that a parking lot is more of a heat reflecting object. More to the point, even heat islands must be considered as part of the worldwide climate, simply because they are part of this wide world. Their heat does not simply disappear (I hope you agree that would violate physics). Imagine we want to measure the average temperature inside a single 30-foot by 10-foot room during winter. We have two probes: one near a burning fireplace on one end of the room, and one near a window on the other. If we excluded data from one probe or the other, do you believe we would get an accurate average reading? Of course, when scientists are calculating a global temperature, they have to handle special cases in the data (like heat islands). This has been known for some time, and you can read more about it here: https://www.realclimate.org/index.php/archives/2007/07/no-man-is-an-urban-heat-island/ https://www.realclimate.org/index.php/archives/2007/07/no-ma... I fear that I've spent too much time responding to this, but I wanted to take it on in earnest.