ChescoWx

Well I finished scrubbing and updating the data. The data looks much better!! I only had to totally delete the Chadds Ford and Honey Brook data. There were many years of good COOP data remaining at all other sites - so only had to delete years at the front or end of the station records. All in all it looks like it may end up with a bit cooler overall look regarding Chester County average temperatures through the years, Having the Phoenixville data ghosted into the Devault data and some of the other estimated data that looks to have skewed us a bit warm may make the difference. Will run some analysis later. Thx

I am having a blast with this!! Fortunately a bit slow at work so I am crunching away on where ghost data is and is not. Some stations have ghost data for decades and then over to actual for decades and then back to ghost! It will be fascinating to see how it all shakes out. I hope to get everything tied up later today or tomorrow. Wish me luck! Thanks again for the heads up!!

Nope - just an excellent VP station with fan aspiration for better accuracy. Plus I am the highest elevation spot reporting, The closest to my elevation in the County is West Grove MADIS 650 ft ASL they have reported a high of 89.5 today with my East Nantmeal station at 685 feet reporting a high of 89.6 so far today. See my station temperature health check for today - all good!

Below is an example of the newest Ghost Station in the Coatesville and Chester County data set. I am working up a list of all real vs ghosted data (and correct dates) to appropriately adjust the Chesco database. I actually have found 2 Ghost Stations in Coatesville one back in the 1940's-1950's and this new one since 2016. Both of these stations are actually NWS Coop precipitation sites that have had post hoc approximate temperatures added to their observations. Below is Coatesville 1E vs. KMQS Airport near same elevation and within 2 NM of the Airport. The month is August 2022. The Ghost Coatesville site reported 11 days over 90 degrees while KMQS and Glenmoore only reported 2 such days. The ghost data increased the average temperature by 1.2 degrees vs KMQS in just that month.

A few spots picked up some showers last evening including 0.14" at Chester Springs...most folks however saw little or no rain. Most spots except the higher locations like Atglen, West Grove and East Nantmeal exceeded 90 degrees. More 90's and high humidity likely through Wednesday....before shower chances increase with the impact of the remnants of Beryl by late week. Chester County records for today: High 101 degrees at Phoenixville (1936) / Low 49 degrees at Coatesville (1954) / Rain 3.44" at Phoenixville (1964)

A few spots picked up some showers last evening including 0.14" at Chester Springs...most folks however saw little or no rain. Most spots except the higher locations like Atglen, West Grove and East Nantmeal exceeded 90 degrees. More 90's and high humidity likely through Wednesday....before shower chances increase with the impact of the remnants of Beryl by late week. Chester County records for today: High 101 degrees at Phoenixville (1936) / Low 49 degrees at Coatesville (1954) / Rain 3.44" at Phoenixville (1964)

As I continue to dig deeper into the analysis for my area of Chester County PA. So far I have found the following additional ghost station data adjustments, these are in addition to Glenmoore, Honey Brook, Chadds Ford, and West Grove (pre 1963) Devault reported ghosted Phoenixville PA readings starting from January 1893 thru June 1, 1951. Devault recorded their own actual observations from June 1951 through January 31, 1988 Devault resumed reporting ghost data from Phoenixville on February 1, 1988 through April 3, 2005 Downingtown PA 2000 thru 2003 all ghosted data looks to be aligned to the Chadds Ford Ghosted Data.... Morgantown PA began ghosting data on February 2, 1986 following actual data from 1951 thru January 1986

Some info on ghost stations - those E's that Charlie mentioned may indeed be these non-existent stations. “They are physically gone—but still report data—like magic,” said Lt. Col. John Shewchuk, a certified consulting meteorologist.“NOAA fabricates temperature data for more than 30 percent of the 1,218 USHCN reporting stations that no longer exist.” He calls them “ghost” stations. Of interest they are still finding more of these stations. In fact one that has been identified is West Chester which last reported real data back in 2017. Do we have more stations we have identified that we can send on over? Mr. Shewchuck said USHCN stations reached a maximum of 1,218 stations in 1957, but after 1990 the number of active stations began declining due to aging equipment and personnel retirements. NOAA still records data from these ghost stations by taking the temperature readings from surrounding stations, and recording their average for the ghost station, followed by an “E,” for estimate. The addition of the ghost station data means NOAA’s “monthly and yearly reports are not representative of reality,” said Anthony Watts, a meteorologist and senior fellow for environment and climate at the Heartland Institute. https://amac.us/blog/health-and-wellness/hidden-behind-climate-policies-data-from-nonexistent-temperature-stations/

Thanks to Charlie for the great heads up!! While we need to do a bit more analysis.... but see below for an analysis of the IEM data set vs. actual obs. These estimates at least in this one month are running warmer than the actual obs by over 1 degree in August 2005. It is starting to look like some of these IEM stations and their data might be what some in the climate world have coined "ghost stations" where older stations are either replaced with these estimates - like we clearly see in near 20 years of the Glenmoore Data. If this is true I will of course strip out these "ghost" stations and only use real data. While the ghost data is similar between these ghost stations....there is some subtle differences. The blue is where the actual real obs are the same as the estimate for Glenmoore - with the yellow differences from the Glenmoore ghost data. It will be interesting to see the impact of stripping out these non-actual data points.

With all the above up there said I am not sure exactly what is going on and will go back and do some further analytics. I just love these kind of puzzles!!. I am after all a recovering accountant/finance guy! I am sure it will become clear if NCEI simply copies certain stations over - as I found in the past with old Devault data. But the big question comes if they are using averaged data from 1 station or some sort of derived estimate and then paste it across additional stations that would of course impact any calculations done....thus weighting more heavily on estimated data or adjusted data - if in fact this data is actually used as part of NCEI's adjustment to the actual or calculation of their adjusted averages. Should be fun!

Plenty of inconsistencies in the above Glenmoore PA comments. First we are told that there was not any temperature data at all for this station and only rain. Later...sorry no actually it looks like the temperature sensor was installed in 2009. But as we see below we have the evidence that the observer has been reporting temperatures to the NWS consistently at least as far back as nearly 20 years! (see below August 2005 handwritten submitted obs). So Then we see that the NCEI Historical Observing Metadata Repository site has begun to erase the fact that data was observed at least that far back.... but has not yet apparently updated for 2009 (still listing temp obs). NCEI true to their word is well on the way to revising the past data as we can see they have "cleaned up" August 2005 below with their updated computer output eliminating the Glenmoore handwritten obs. Call me stunned NCEI adjusting or eliminating data after the fact - I never would have thought such a thing LOL!! Now we need to ask well where is the analysis to support the deletion of this station data? Well they do mention above these supposed reliability concerns....sounds fair but where is the analysis against a control station or 2 to show how unreliable this station supposedly is? Is that station running "statistically" higher or lower than surrounding "valid" stations to invalidate the readings? Well, no details have been provided of that analysis.... just we feel like there are "reliability concerns" and it is not NWS approved. A little analysis and comparison to the closest surrounding stations does not show any issues with the data. In fact in the final year that Glenmoore/Coatesville and East Nantmeal (2007) all reported they were within 0.1 degree of each other in average temperature. East Nantmeal is within 2 nautical miles of the Glenmoore COOP observer within 60 feet of elevation. So even though not NWS approved that data sure passes the statistical p-value to validate and remains extremely consistent with the current Glenmoore MADIS station and of course East Nantmeal.

All spots across the County exceeded the 90 degree mark with the oh so close exception of East Nantmeal where we could "only" reach 89.9 degrees!. The next coolest spot was West Grove at 90.9. Some of the hotter spots included Chester Springs 95 degrees and. KOQN Brandywine Airport also 95. The hot temps looks to continue for the next few days before we drop back into the 80's by Wednesday. Along with the temp drop will come increasing shower chances by Tuesday night. Each day from Wednesday thru Saturday will feature a chance at some showers. Chester County records for today: High 103 degrees at Phoenixville (2010)/ Low was 47 degrees at West Chester (2001) / Rain 2.23" at Coatesville (1907)

All spots across the County exceeded the 90 degree mark with the oh so close exception of East Nantmeal where we could "only" reach 89.9 degrees!. The next coolest spot was West Grove at 90.9. Some of the hotter spots included Chester Springs 95 degrees and. KOQN Brandywine Airport also 95. The hot temps looks to continue for the next few days before we drop back into the 80's by Wednesday. Along with the temp drop will come increasing shower chances by Tuesday night. Each day from Wednesday thru Saturday will feature a chance at some showers. Chester County records for today: High 103 degrees at Phoenixville (2010)/ Low was 47 degrees at West Chester (2001) / Rain 2.23" at Coatesville (1907)

Even though it has an E flag on the Data Calendar below there actually is submitted NWS Coop forms (see below) that show the actual observations taken. The dataset I am using for the analysis matches the below official coop data form and does not use the estimated figures on the data calendar. You can see the differences below. Not sure why they would put estimates in when we have actual data?

Hey Charlie looks like quite a bit of missing data under the IEM sites. To your point they list Glenmoore as having no temperature data for March 2009 (see below) but there were temperature obs taken per obs sheet. They do correctly record the precipitation however. These reading align with my dataset for Glenmoore for that month. The metadata sourced from the Penn State Climate site also shows no temps and only rain data....strange!

The only issue I found during my QC was that they had simply mirrored Devault data with Phoenixville from 1893 thru 1951. Devault did not begin daily temperature obs till 1951....so I deleted all of those years for Devault. Otherwise the data is consistent but with clear obs variations between the various stations see below for example.

Most locations exceeded 90 degrees yesterday with the only exceptions being the higher spots like West Grove (88.2) and East Nantmeal (88.5). The heat advisory will continue through tomorrow with temps again in the 90's at most spots. There is a chance of some scattered storms today with a better chance by Tuesday and through mid-week. While year to date we are still over 2" above normal rainfall the last 2 months we are running about 2.5" below normal. Chester County records for today: High 103 degrees at Phoenixville (2010) / Low 45 degrees at Morgantown/Elverson (1979)/ Rain 4.15" at Chadds Ford (1989)

Most locations exceeded 90 degrees yesterday with the only exceptions being the higher spots like West Grove (88.2) and East Nantmeal (88.5). The heat advisory will continue through tomorrow with temps again in the 90's at most spots. There is a chance of some scattered storms today with a better chance by Tuesday and through mid-week. While year to date we are still over 2" above normal rainfall the last 2 months we are running about 2.5" below normal. Chester County records for today: High 103 degrees at Phoenixville (2010) / Low 45 degrees at Morgantown/Elverson (1979)/ Rain 4.15" at Chadds Ford (1989)

No worries Don....I have made a few of those myself! Even though we may disagree on some of these big issues I still enjoy your posts!! take care!!

And no matter how many machine produced non-Chester County obs you put forth.....we know it is not real or factual data. The facts are not supportive....I feel your pain!!

Fortunately we have plenty more stations that say.....sorry Charlie!!

Not true again Charlie there are more stations similar to the old lower elevation sites than the relative higher and we can split them out and analyze them see below summer only lower elevation (warmer) sub 410 ft. stations. Same story cyclical changes and minor unalarming gentle warming.

A great point again from Chris Martz. "You’d think hearing the good news that the planet isn’t facing some sort of climate catastrophe would bring a sigh of relief to climate alarmists. Instead, they cling to irrational fear and get angry when presented with evidence that extreme weather isn’t increasing (generally) and that human welfare has never been better than it is today across all metrics that have at least some link to climate (e.g., crop yields, life expectancy, nourishment and to an extent, disaster-related deaths). Until the data on these metrics begins to trend in the opposite direction, I remain unconvinced that we’re facing a “climate crisis.” People who suffer from “anxiety” over the weather suffer from a mental health issue. It’s not a “different opinion,” it’s not normal. If it has to be normalized, it isn’t normal. Period." I will do my small part by continuing to share the good news that at least here in Chester County PA our climate is not warming at an unprecedented or alarming rate.....see our unscary summer warming above!!

Actually Charlie when we exclude the Phoenixville Station.....the 1930's were still the warmest decade in County history. Summer temps are basically flat....nothing alarming here at all!

I thought those that he answered were well thought out.....but as always with Don plenty of conjecture and some fearful prognostications (3 meter or more sea level rise over the next 75 years?)