Jump to content

Western Chester County Weather Analysis 1894-Present


Recommended Posts

2 hours ago, Chubbs said:

Any reputable source of climate information will tell you where we are headed. Warming is pretty much as forecast decades ago.

I wonder if back in the 1950's they forecast the next 3 decades would be coldest in recorded obs history? No one knows the future only God of course....that said I have little doubt climate is cyclical warm - cold - rinse repeat. I am not a climate denier I embrace the fact that climate is and always will be changing....therefore I am confident we will see colder trends in future years.....although if major city airport heat island airports like PHL are included and weighted heavily that will skew the data. Hence why it is important to keep site of rural obs like the Chester County data set to keep us grounded

Link to post
Share on other sites
  • Replies 155
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Popular Posts

Let the temps drop.   

Here is the link I used for KPTW and KQMS (remove quotes) "https://mesonet.agron.iastate.edu/plotting/auto/?q=17"

Paul - Thanks for correcting the data. Below are 90F days for doylestown and quakertown and a plot of all the suburban airports and phl. The data is noisy, and KPTW in particular bounces around more t

Posted Images

1 hour ago, Chubbs said:

Experts use anomalies to combine temperature data from different stations into regional or global records. Here is a simple example using data from PHL and Coatesville Airport. The chart showing below shows annual temperature anomalies for the period where both sites have data: 1984 to 2020. Also below is an example showing how the 2020 anomaly is calculated.

Here are some conclusions:

*) Anomalies facilitate comparision by eliminating baseline differences

*) Year-to-year changes are caused by weather and are often exactly the same at PHL+Coatesville. Because of weather differences, need to use the same period for an accurate comparison.

*) There are a few years when there is a difference between the two stations due to either: 1) bad measurement or 2) some kind of local weather condition. To determine the cause need more data from additional stations. This is the value of having multiple stations in the same area.

*) For the period as a whole there is no difference in the warming at phl and Coatesville.   Trend lines through the two stations are similar with 0.7F warming per decade at PHL and 0.8F per decade at Coatesville.

 

                         1984-2020 Average                  2020 Avg Temp                  2020 Anomaly

Coat                             52.6                                         54.6                                    2.0

PHL                              56.4                                         58.0                                    1.6

phlkqmsanom.png

I agree my charts support the above 100% the last 3 decades have warmed...now when does the next cooling cycle start? This decade next decade? Also, what is the source of the Coatesville data you show. There was a NWS observer Joe Steiner who recorded data for Western Chesco from 1984 to 2007. Then no observers between 2007 and till October 2016??

Link to post
Share on other sites
3 hours ago, chescopawxman said:

I agree my charts support the above 100% the last 3 decades have warmed...now when does the next cooling cycle start? This decade next decade? Also, what is the source of the Coatesville data you show. There was a NWS observer Joe Steiner who recorded data for Western Chesco from 1984 to 2007. Then no observers between 2007 and till October 2016??

Coatesville 2W - for 1984 to 2007 - thought you said previously that was airport data. If not it appears to match KQMS well

KQMS - 2008 to 2020, could also use your data if you want to provide.

 

Again check reputable information - then you will know what to expect. Below is my rough estimate for KQMS extrapolating current trends.

2010-19 53.8
2020-29 54.6
2030-39 55.4

 

Link to post
Share on other sites
53 minutes ago, Chubbs said:

Coatesville 2W - for 1984 to 2007 - thought you said previously that was airport data. If not it appears to match KQMS well

KQMS - 2008 to 2020, could also use your data if you want to provide.

 

Again check reputable information - then you will know what to expect. Below is my rough estimate for KQMS extrapolating current trends.

2010-19 53.8
2020-29 54.6
2030-39 55.4

 

No one of course "knows what to expect" as unfortunately both mid and long range forecasting and longer term climate projections/forecasts are quite far from exact science as we all know. Based on my research and analysis I forecast normal cyclical cooling to start likely during the later half of this current decade 2020-29 - with this decade ending up quite close to temps we saw in the 2010-2019 decade. Then the cooling cycle should kick in during the back half 2025-2029 of this decade. That makes the most sense based on my analysis. Also, I will have to go back at check but KMQS is an AWOS source and none of that is in any of my data. I will check out that non-co op data for 2008-2019 and see how it matches up to my data. Good stuff thanks Charlie!

Link to post
Share on other sites
1 hour ago, chescopawxman said:

No one of course "knows what to expect" as unfortunately both mid and long range forecasting and longer term climate projections/forecasts are quite far from exact science as we all know. Based on my research and analysis I forecast normal cyclical cooling to start likely during the later half of this current decade 2020-29 - with this decade ending up quite close to temps we saw in the 2010-2019 decade. Then the cooling cycle should kick in during the back half 2025-2029 of this decade. That makes the most sense based on my analysis. Also, I will have to go back at check but KMQS is an AWOS source and none of that is in any of my data. I will check out that non-co op data for 2008-2019 and see how it matches up to my data. Good stuff thanks Charlie!

Climate is not the same as weather, much more predictable and scientists know exactly what too expect - more warming in the coming decades until we get our act together.

Below are our two predictions for the next decade. We will see who is right.

Paul:  2020-29 = 2010-19; for KQMS this is 53.8

Charlie - 2020-29 = 2010-19 plus 0.8F; for KQMS this is 54.6

 

 

 

 

Link to post
Share on other sites
5 hours ago, Chubbs said:

Climate is not the same as weather, much more predictable and scientists know exactly what too expect - more warming in the coming decades until we get our act together.

Below are our two predictions for the next decade. We will see who is right.

Paul:  2020-29 = 2010-19; for KQMS this is 53.8

Charlie - 2020-29 = 2010-19 plus 0.8F; for KQMS this is 54.6

 

 

 

 

How and what does "get our act together" mean? I have not put out a forecast yet - one quick note you keep saying KQMS it is KMQS that is the GO Carlson Airport in Chester County. Also to be clear 2010-2019 was 53.06 for Chester County the delta from PHL has never been greater then it was in this last decade with a difference of +4.29 degrees...just take a peek at the analysis I have produced and you can see the variance will only get greater as we allow the PHL heat island obs to run away from the field. I certainly expect the PHL heat island to continue it's acceleration to even greater and more frequent heat records as the runways and river fill projects continue to make the variance between major heat island spots like the swamps of PHL and where most folks live to be even greater. All of that said while PHL will likely never show a cooling ever again in our lifetimes based on the obstacles at PHL....I expect suburban locales to show cooling over the time between 2026 and 2056.....based on my detailed climate analytics.....unfortunately most here on this board won't be here by the time we validate these forecasts!! unless I live to be 93 yrs old in 2056!!

Link to post
Share on other sites

No surprises here ( at least to me) but likely not going to be embraced by some on this board (hello Tom)....but detailed analysis of KMQS and East Nantmeal Township yield a statistically significant P-value (meaning it is the exactly the same and a statistical impossibility by chance that any readings between these sites and observations can be anything but the same and valid according to statistical probabilities) so readings are scientifically proven as valid for both East Nantmeal Twp in Chester County PA and the NWS COOP observers from 1983 to 2007 and now adding the KMQS AWOS obs from 2008 to Present  analyzed against the East Nantmeal detailed data now clearly support this finding. So the data set I analyze and report for the climate period from 1983 to Present is once again proven as scientific fact

 

Link to post
Share on other sites
10 hours ago, chescopawxman said:

No surprises here ( at least to me) but likely not going to be embraced by some on this board (hello Tom)....but detailed analysis of KMQS and East Nantmeal Township yield a statistically significant P-value (meaning it is the exactly the same and a statistical impossibility by chance that any readings between these sites and observations can be anything but the same and valid according to statistical probabilities) so readings are scientifically proven as valid for both East Nantmeal Twp in Chester County PA and the NWS COOP observers from 1983 to 2007 and now adding the KMQS AWOS obs from 2008 to Present  analyzed against the East Nantmeal detailed data now clearly support this finding. So the data set I analyze and report for the climate period from 1983 to Present is once again proven as scientific fact

 

I honestly could give two hoots about your climo nwsheco data hence why I never respond to it

  • Like 1
Link to post
Share on other sites
7 hours ago, tombo82685 said:

I honestly could give two hoots about your climo nwsheco data hence why I never respond to it

I know.....just havin a little fun with you!

Link to post
Share on other sites
18 hours ago, chescopawxman said:

image.png.77d3f5a520f8eaa1484754b2b887359f.png

Here are the 1998-2020 anomalies for enant, kmqs and phl. All 3 move together for the most past. Phl and kmqs are very close for most years. Interestingly, in cases where kmqs and phl disagree, Enant can often help flag the inaccurate data. KMQS was probably too cold in 2008+2009 and too warm in 2004, 2011 and 2012. Compared to the other 2 sites, ENant is warmer to start, colder in the middle, and a little warmer to end. Were there any changes in siting or equipment to explain the differences?

phlkmqsenant.png

Link to post
Share on other sites
34 minutes ago, Chubbs said:

Here are the 1998-2020 anomalies for enant, kmqs and phl. All 3 move together for the most past. Phl and kmqs are very close for most years. Interestingly, in cases where kmqs and phl disagree, Enant can often help flag the inaccurate data. KMQS was probably too cold in 2008+2009 and too warm in 2004, 2011 and 2012. Compared to the other 2 sites, ENant is warmer to start, colder in the middle, and a little warmer to end. Were there any changes in siting or equipment to explain the differences?

phlkmqsenant.png

Can you layer in the actual PHL temp (not anomaly)on top of both the ENANT and KMQS data? Thanks!

Link to post
Share on other sites

I went back and updated the analysis for the last 12 decades now including the Philadelphia data from 1900-1949 (this represents raw unadjusted data with no after the fact adjustments) comparing data for the last 120 years from 1900 thru 2019. The key finding is that the suspected rapid PHL heat island impact and problems are growing at an even greater rate than ever imagined....the difference between the heat island near Philadelphia and the burbs using actual NWS observed temps is telling....Just look at how quickly the lines are diverging and you will see where this is going.  Just look at that % variance difference column and you can see that the disparity between average temperatures has never been higher and is only getting more pronounced.

image.png

Link to post
Share on other sites
6 hours ago, chescopawxman said:

I went back and updated the analysis for the last 12 decades now including the Philadelphia data from 1900-1949 (this represents raw unadjusted data with no after the fact adjustments) comparing data for the last 120 years from 1900 thru 2019. The key finding is that the suspected rapid PHL heat island impact and problems are growing at an even greater rate than ever imagined....the difference between the heat island near Philadelphia and the burbs using actual NWS observed temps is telling....Just look at how quickly the lines are diverging and you will see where this is going.  Just look at that % variance difference column and you can see that the disparity between average temperatures has never been higher and is only getting more pronounced.

image.png

PHL is fine, very stable. Agrees perfectly with the Coatesville2/KMQS data since 1984.  In comparison your data is not stable:  multiple sites, different elevations, different equipment, shelters etc. 

phlkqmsenantemp.png

Link to post
Share on other sites

Here is the old Coatesville data compared to the regional temperature series.  The oldest Coatesville data was collected with a mercury thermometer in an inadequate shelter. Daily max/min depend on when the thermometer was checked. That is the way things were done back then. Not directly comparable to modern digital measurements. 

sepacoats.png

Link to post
Share on other sites
3 hours ago, Chubbs said:

Here is the old Coatesville data compared to the regional temperature series.  The oldest Coatesville data was collected with a mercury thermometer in an inadequate shelter. Daily max/min depend on when the thermometer was checked. That is the way things were done back then. Not directly comparable to modern digital measurements. 

sepacoats.png

Charlie can you answer me this - how many and exactly which sites in your pooled data were made on mercury thermometers before "modern digital measurements - what year did these "modern" measurements begin? How did you determine that WT Gordon or Howard Pyle or the Philadelphia Electric company used inadequate shelters in the Coatesville NWS data set? The data was at the time clearly verified and stamped as such on the handwritten obs sheet by whatever branch of the weather service was in charge at the time. My analysis includes the "Old" Coatesville NWS COOP data and no KMQS AWOS data. The divergence is clearly growing....I will check the KMQS data alone and see if we can identify if any inconsistencies since the AWOS began to be used back in 2008

Link to post
Share on other sites

I  have gone back and analyzed long running data now for both long term Chester County sites West Chester PA and Coatesville PA. Clearly West Chester and Coatesville are well aligned for 12 decades....and almost exactly the same over the last 2 decades....however, PHL is just as clearly demonstrating the heat island problem as they begin to accelerate away - not at all consistent.image.png.2a171da9246295180ff0551558d3c11a.png

Link to post
Share on other sites
8 hours ago, chescopawxman said:

Charlie can you answer me this - how many and exactly which sites in your pooled data were made on mercury thermometers before "modern digital measurements - what year did these "modern" measurements begin? How did you determine that WT Gordon or Howard Pyle or the Philadelphia Electric company used inadequate shelters in the Coatesville NWS data set? The data was at the time clearly verified and stamped as such on the handwritten obs sheet by whatever branch of the weather service was in charge at the time. My analysis includes the "Old" Coatesville NWS COOP data and no KMQS AWOS data. The divergence is clearly growing....I will check the KMQS data alone and see if we can identify if any inconsistencies since the AWOS began to be used back in 2008

I can't answer your question in detail, but we can assume that all measurements were non-digital until after WW2 and there are a range of other station changes that can impact measurements.  The blog article describes a couple of systematic cooling biases in US measurements. Note that automated software works well to identify and correct any individual site issues.

https://judithcurry.com/2015/02/22/understanding-time-of-observation-bias/

 

 

Link to post
Share on other sites
18 hours ago, chescopawxman said:

I  have gone back and analyzed long running data now for both long term Chester County sites West Chester PA and Coatesville PA. Clearly West Chester and Coatesville are well aligned for 12 decades....and almost exactly the same over the last 2 decades....however, PHL is just as clearly demonstrating the heat island problem as they begin to accelerate away - not at all consistent.image.png.2a171da9246295180ff0551558d3c11a.png

You are attributing every bump and wiggle in the single station data to heat island effects. The reality is that s**t  happens. Here are the 4 main airports in our area. For the most past they agree well, but there are problem areas. PHL was too cold in the early+mid 60s - no heat island just a measurement issue. Similarly phl was too warm for 3 years in the early 90s. This is why the regional dataset is superior, bad data, including heat island effects, is readily identified and corrected.

airports.png

Link to post
Share on other sites

West Chester and Coatesville have their own problems. Looks like both stations were changed in the early 1970s leading to a rather large cooling bias, relative to other stations in our region. Note that the Coatesville spike to 3F in the chart below is an incomplete data flag that I am using, not real data. There are a few holes in the Coatesville data that I was able to obtain.

westchestcoatscool.png

Link to post
Share on other sites

Fortunately in my data set used for the analysis there are no changes or holes in any of the data. Both Coatesville and West Chester are complete data sets with 365 or 366 daily data sets for every year since 1894. Seems to be a statistical improbability that both stations could be malfunctioning at exactly the same level thru over 120 years of data....I will layer in some additional counties with long term data to see if there is a similar significant reduction in warming when compared to the PHL heat island.

Link to post
Share on other sites

 

58 minutes ago, chescopawxman said:

Fortunately in my data set used for the analysis there are no changes or holes in any of the data. Both Coatesville and West Chester are complete data sets with 365 or 366 daily data sets for every year since 1894. Seems to be a statistical improbability that both stations could be malfunctioning at exactly the same level thru over 120 years of data....I will layer in some additional counties with long term data to see if there is a similar significant reduction in warming when compared to the PHL heat island.

Didn't say they were malfunctioning - something changed, equipment/procedures - who knows. There are other periods periods when they are not as closely aligned.

westchestcoatscool.png

Link to post
Share on other sites

Hi Charlie - I found another relatively long term Chester County site closer to your area of Chester County....Octoraro Lake with 4 complete decades of data (1979 to Present)....again as you can see in the analysis data between Coatesville / West Chester and now Octoraro are in incredible agreement on both actual readings and on the degree of warming - which is significantly less than the PHL data is showing image.png.4f70de3d64402a65c4b6ce5a11c96e38.png

Link to post
Share on other sites
4 hours ago, chescopawxman said:

Hi Charlie - I found another relatively long term Chester County site closer to your area of Chester County....Octoraro Lake with 4 complete decades of data (1979 to Present)....again as you can see in the analysis data between Coatesville / West Chester and now Octoraro are in incredible agreement on both actual readings and on the degree of warming - which is significantly less than the PHL data is showing image.png.4f70de3d64402a65c4b6ce5a11c96e38.png

Meh. We keep covering old ground. Your conclusions rest on older data which has problems when looked at carefully. Per chart below, with proper adjustment the Coatesville data lines up perfectly with the other regional data.  All the data we are talking about: Coatesville, West Chester, Octararo Lake, Phl, ABE, plus many others is all included in the regional series.

 

 

coatesville.png

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    No registered users viewing this page.


×
×
  • Create New...