Updated: Actual 90+ Days 37 year Comparison PHL Airport vs KMQS Airport not Chescowx data - amazing stats

[chescowx]

[Chubbs]

Here's West Chester. I don't have a problem with the data you are collecting, but don't think it is very representative.

[Chubbs]

Here's Coatesville 1983-2007. PHL looks like a better match than E. Nant or the Chesco airport?

[chescowx]

Below is some incredible data that really highlights the Heat Island problem that is the PHL Airport and of course many other airports that folks point to as examples of warming. Just look at these numbers comparing KPHL to KMQS (FYI this is not my data site) for the last 37 years. What really stunned me was the fact that while rural locations like KMQS airport have seen less 90 degree days with each successive decade the PHL Heat Island effect gets worse with each decade. The KMQS airport has seen the number of 90+ days per decade decline as follows;

17 in the 1980's/ 15 in the 1990's / 9 in the 2000's and now only 6 in a average year in this current decade of the 2010's - conversely PHL Airport has gone in a total different direction with an average of 27 ninety plus days in the 1980's / 31 in the 1990's /22 in the 2000's and now a jaw dropping 32 such days in the 2010's - if you only followed urban airport data you would be inclined to believe that we are warming at an alarming rate....however when viewing a rural location we are seeing the opposite. I am not saying one or the other is correct but to just point to an observation at a heat island airport along the I95 corridor is not a true representation of the region. That being said not many folks live above 650 ft in the PHL suburbs like where KMQS is situated. But if nothing else closer analysis of non-airport locations should be taken into account for what is a very complex situation and not fall prey to analysis only based on official airport observations sites that are sited in locations like PHL in a swamp at near sea level where most folks in the PHL area do not live. 

[Chubbs]

Here is a plot of the data in the table and the Coatesville 2W data from NOAA. The airport is roughly 2 miles W of coatesville. NOAA only has Coatesville 2W data from 1983 through 2007. Coatesville 2 and KQMS have  exactly the same# of 90F days from 1983-2003 but in 2004 kqms drops lower and continues low through 2018.  Coatesville2 and phl are well aligned throughout the 1983 to 2007 period, if anything coatesville 2 is getting closer to phl at the end. My conclusion is that there is a discontinuity in the kqms data. The location and/or the equipment changed in 2004.

 

 

[colonel_kurtz]

11 hours ago, chescopawxman said:

Below is some incredible data that really highlights the Heat Island problem that is the PHL Airport and of course many other airports that folks point to as examples of warming. Just look at these numbers comparing KPHL to KMQS (FYI this is not my data site) for the last 37 years. What really stunned me was the fact that while rural locations like KMQS airport have seen less 90 degree days with each successive decade the PHL Heat Island effect gets worse with each decade. The KMQS airport has seen the number of 90+ days per decade decline as follows;

17 in the 1980's/ 15 in the 1990's / 9 in the 2000's and now only 6 in a average year in this current decade of the 2010's - conversely PHL Airport has gone in a total different direction with an average of 27 ninety plus days in the 1980's / 31 in the 1990's /22 in the 2000's and now a jaw dropping 32 such days in the 2010's - if you only followed urban airport data you would be inclined to believe that we are warming at an alarming rate....however when viewing a rural location we are seeing the opposite. I am not saying one or the other is correct but to just point to an observation at a heat island airport along the I95 corridor is not a true representation of the region. That being said not many folks live above 650 ft in the PHL suburbs like where KMQS is situated. But if nothing else closer analysis of non-airport locations should be taken into account for what is a very complex situation and not fall prey to analysis only based on official airport observations sites that are sited in locations like PHL in a swamp at near sea level where most folks in the PHL area do not live. 

PHL is a representation of the city of Philadelphia not the region. Center City may well average 2 degrees warmer than the airport so overall PHL is a good snapshot for the city as a whole.

Charlie's awesome graph sums up perfectly this nonsense with regard to the lack of 90 degree days at KMQS since. I mean seriously PHL records (37) 90 days in 2015 while there's none at KMQS - LOL. Reality check - when it's 95-96 degrees at PHL it's at least 90 degrees throughout Chester County.

 

[tombo82685]

1 hour ago, Chubbs said:

Here is a plot of the data in the table and the Coatesville 2W data from NOAA. The airport is roughly 2 miles W of coatesville. NOAA only has Coatesville 2W data from 1983 through 2007. Coatesville 2 and KQMS have  exactly the same# of 90F days from 1983-2003 but in 2004 kqms drops lower and continues low through 2018.  Coatesville2 and phl are well aligned throughout the 1983 to 2007 period, if anything coatesville 2 is getting closer to phl at the end. My conclusion is that there is a discontinuity in the kqms data. The location and/or the equipment changed in 2004.

 

 

Yea thats pretty vast in how it goes from more 90s to less. Especially in a time period that was generally hotter too. 

[chescowx]

1 hour ago, Chubbs said:

Here is a plot of the data in the table and the Coatesville 2W data from NOAA. The airport is roughly 2 miles W of coatesville. NOAA only has Coatesville 2W data from 1983 through 2007. Coatesville 2 and KQMS have  exactly the same# of 90F days from 1983-2003 but in 2004 kqms drops lower and continues low through 2018.  Coatesville2 and phl are well aligned throughout the 1983 to 2007 period, if anything coatesville 2 is getting closer to phl at the end. My conclusion is that there is a discontinuity in the kqms data. The location and/or the equipment changed in 2004.

 

 

Nice summary but not sure I follow - you say Coatesville2 and KMQS are the same thru 2004 and then you conclude without data that it must be equipment....good luck with that analysis. Plus we do indeed have my data which as you can see almost mirrors what you view as suspect data from 2004 to present for 90 degree days - see my earlier analysis. If you have better data to run up against KMQS then mine for Chester County at almost the same elevation. So in reality you say equipment failure to explain away something that an observer 8 miles away at the same elevation confirms??

[chescowx]

57 minutes ago, colonel_kurtz said:

PHL is a representation of the city of Philadelphia not the region. Center City may well average 2 degrees warmer than the airport so overall PHL is a good snapshot for the city as a whole.

Charlie's awesome graph sums up perfectly this nonsense with regard to the lack of 90 degree days at KMQS since. I mean seriously PHL records (37) 90 days in 2015 while there's none at KMQS - LOL. Reality check - when it's 95-96 degrees at PHL it's at least 90 degrees throughout Chester County.

 

Hi Colonel - it would be nonsense but how the heck (unless you use try to explain away the data as both KMQS and myself having the same exact equipment failure) for both my data and KMQS  are almost exactly the same regarding 90+ days for the last 16 years of data. Now, I know you will say but you are 8 miles from KMQS - true but at the same elevation. Is there better data then mine to at least say it sure does seem to support what KMQS is reporting and another observer in the same county at a similar elevation?

[chescowx]

3 minutes ago, tombo82685 said:

Dude, it's clear as crystal that something changed after 2004. How does qms go from 1983 to 2004 with 12 years of 10 or more 90 days. Then all of sudden after 2004 they have 3, which was a hotter period than prior. Some thing changed. Why on earth is this so difficult to see. You put that graph up to any weather person and they can see something changed

But Tom we do have data - mine that supports those exact numbers....can you produce better data than mine for comparison? Just saying equipment failure is not scientific - at least I have daily data for every day of the same 16 year reporting period and as you know KMQS and here are almost always the same? Show this data to any analytics person and they will say - well certainly the data from 8 miles away at near the same elevation is better than comparing it to an airport 40 miles away and 630 feet lower on a swamp on a river.

[tombo82685]

2 minutes ago, chescopawxman said:

But Tom we do have data - mine that supports those exact numbers....can you produce better data than mine for comparison? Just saying equipment failure is not scientific - at least I have daily data for every day of the same 16 year reporting period and as you know KMQS and here are almost always the same? Show this data to any analytics person and they will say - well certainly the data from 8 miles away at near the same elevation is better than comparing it to an airport 40 miles away and 630 feet lower on a swamp on a river.

It has nothing to do with elevation. I proved that to you last year. Either you live in some wild micro climate where temps don't rise in the afternoon, or there is some local effect, or it's sensor failure. Elevation though is not why you don't get that many 90 degree days 

[chescowx]

1 minute ago, tombo82685 said:

It has nothing to do with elevation. I proved that to you last year. Either you live in some wild micro climate where temps don't rise in the afternoon, or there is some local effect, or it's sensor failure. Elevation though is not why you don't get that many 90 degree days 

Tom - then why is my "wild micro climate" exactly the same as KMQS data for the last 16 years? You bring forth zero data while I have and analyzed 16 years of data for both locations - do you have anything else or in your mind it can only be sensor failure at both KMQS and my equipment that incredibly happened at the same time and has persisted for 16 years....data and facts don't lie here in Chester County

[chescowx]

7 minutes ago, tombo82685 said:

It has nothing to do with elevation. I proved that to you last year. Either you live in some wild micro climate where temps don't rise in the afternoon, or there is some local effect, or it's sensor failure. Elevation though is not why you don't get that many 90 degree days 

Sorry I missed your last point so why then do KMQS and East Nantmeal get so few 90+ days - you say it's not elevation then what have you concluded from the facts on this??

[tombo82685]

Because there are numerous airports in the northeast that are above 1000’ that have had more 90 days than you. If it was solely elevation driven then they wouldn’t have any 90 days

[chescowx]

5 minutes ago, tombo82685 said:

Because there are numerous airports in the northeast that are above 1000’ that have had more 90 days than you. If it was solely elevation driven then they wouldn’t have any 90 days

Ok so another airport further away and/or in another state invalidates data from 2 independent sites within 8 miles of each other with similar elevation and rural locales? Not sure that works real well....

[chescowx]

Just now, chescopawxman said:

Ok so another airport further away and/or in another state invalidates data from 2 independent sites within 8 miles of each other with similar elevation and rural locales? Not sure that works real well....

and Tom I am not saying elevation is the only reason.....but wouldn't most analytical minds say jeeze why the heck are 2 weather observation sites so similar in their reporting for 16 straight years - saying that both sensors are defective or incorrectly sited seems to be just lazily making up a solution to arrive at your desired outcome....right?

[chescowx]

So I updated the analysis to show the last 16 years of reporting comparing KMQS, East Nantmeal and PHL ...now unless you believe both the KMQS (AWOS) and East Nantmeal (VP2) stations both malfunctioned at around the same time 16 years ago and have continued to be very consistent in their errant temperature reporting and 90 degree days....these facts and analytics stand on their own without trying to make up reasons why they must be wrong

 

[hstorm]

I don't really have a strong view on either side of this debate, but wouldn't it be important to know the actual temperature at PHL and MQS (and Paul's house) on each >90 day?  If there are a lot of 90-92 degree temps in the PHL data set (which makes sense given that, in a normal distribution, temps would skew closer to the average high), then it would not be surprising for MQS, given its elevation and more rural location, to be below 90 on those days.  And if the number of 90-92 days per year at PHL is increasing -- for whatever reason, climate, heat island, it doesn't really matter for this purpose -- then the spread of >90 days between PHL and MQS would also logically increase.

The better point of comparison would seem to be the actual temperature spread on each day, regardless of high temperature.  The impact of a more intense heat island would presumably be shown in a growing spread (because MQS would not be as affected), whereas a general warmer climate impact would likely affect PHL and MQS similarly.

That all being said, I am certainly not a scientist.  So perhaps I'm missing something.  But it seems like both sides of this debate could be partially right and partially wrong. 

[hstorm]

An interesting and somewhat parallel point of comparison would be to look at data for BOS (Logan), Blue Hill, and Worcester, MA.  Like PHL, BOS is at/near sea level, Blue Hill is just outside Boston, but at 635 feet, and ORH is at 1000 feet.  All are at essentially the same latitude.

[Chubbs]

1 hour ago, chescopawxman said:

Nice summary but not sure I follow - you say Coatesville2 and KMQS are the same thru 2004 and then you conclude without data that it must be equipment....good luck with that analysis. Plus we do indeed have my data which as you can see almost mirrors what you view as suspect data from 2004 to present for 90 degree days - see my earlier analysis. If you have better data to run up against KMQS then mine for Chester County at almost the same elevation. So in reality you say equipment failure to explain away something that an observer 8 miles away at the same elevation confirms??

Paul - KQMS is Coatesville 2W until 2004. Then something happened and the series diverge.

[chescowx]

14 minutes ago, Chubbs said:

Paul - KQMS is Coatesville 2W until 2004. Then something happened and the series diverge.

I get that but what diverged? it diverged right in line with my station the most appropriate station for comparison - unless you have another one. The fact both KMQS and my station are trending exactly the same is telling - certainly much more of an explanation then a machine error - right??

[snowlurker]

1 minute ago, chescopawxman said:

I get that but what diverged? it diverged right in line with my station the most appropriate station for comparison - unless you have another one. The fact both KMQS and my station are trending exactly the same is telling - certainly much more of an explanation then a machine error - right??

Paul, have you had the same exact equipment pre-2004 and post-2004?

[Chubbs]

If you are looking at individual stations then urbanization needs to be considered. However in regional climate analysis, urban heat island and other siting effects are easily corrected for.  Note that trends at individual sites aren't very meaningful due to siting, equipment and microclimate issues. Many sites have downward trends even though the average global and regional trend is up.

https://climatechange.procon.org/sourcefiles/influence-of-urban-heating-on-global-temperature-land-average.pdf

 

 

[hstorm]

1 minute ago, Chubbs said:

If you are looking at individual stations then urbanization needs to be considered. However in regional climate analysis, urban heat island and other siting effects are easily corrected for.  Note that trends at individual sites aren't very meaningful due to siting, equipment and microclimate issues. Many sites have downward trends even though the average global and regional trend is up.

https://climatechange.procon.org/sourcefiles/influence-of-urban-heating-on-global-temperature-land-average.pdf

 

 

Certainly. But this thread started with, and appears to be focused entirely on, an analysis of three individual stations. 

[chescowx]

6 minutes ago, snowlurker said:

Paul, have you had the same exact equipment pre-2004 and post-2004?

yes - VP2