Jump to content

TheClimateChanger

Members
  • Posts

    4,011
  • Joined

  • Last visited

Everything posted by TheClimateChanger

  1. Today was both my warmest and coolest readings of the past 7 days.
  2. Yes, I have the same thing in Pennsylvania. I don't know what it is. First time I have seen it. I thought it was some sort of insect nest.
  3. *Correction: 1F below the observation from June 29, 2012, at 5 am, and 1F below the 8am observation today.
  4. July 17, 1988 is also a candidate. It was 82F at 5 am and 8 am, 1F cooler than the observations from June 29, 2012 for those hours. I don’t have the hourly observations for 6, 7 or 9.
  5. Most likely, the actual highest temperature for those hours occurred in 2012, or perhaps this morning.
  6. The 83F reading from 7 am on September 9, 1989 also appears to be erroneous. It was 69F at 5 & 8 on that date. And likewise the 83F from 6 am on June 23, 1989 also appears to be erroneous. Readings from 5 & 8 am were 67 and 66, respectively.
  7. Yeah, it’s an error. I pulled up the climate data from July 1989 for KPIT and it has daily observations for each three hours starting at 1 am Local Standard Time (i.e. 2 am EDT) and it was 74F at 8 am, 89F at 11 am, and 93F at 1 pm. While the 9 am observation wasn’t listed, obviously wasn’t 93F. Looks like there was some weird time offset in the way the data got uploaded to the IEM database.
  8. Looks like that urban heat island effect is creeping into East Nantmeal Township with the low temperature readings being significantly elevated in comparison to the high temperatures - which is especially surprising given the lack of precipitation [which would ordinarily lead to heightened diurnal ranges]. Have they been converting a lot of nearby woods and farmland to subdivisions? Maybe you need to add a UHI correction to your analysis.
  9. I'm not satisfied with the current ranking system. Is this really as hot as it can get in Reading, Pennsylvania in the current climate? I don't think so. We should be comparing summers to 19th & 20th century summers observed in locations hundreds of miles south. So, yes, indeed, the current mean temperature of 77.1F for the summer to date is the highest on record for Reading, Pennsylvania. However, it's only ninth warmest on record for Richmond, Virginia, for the period ending in 2007 [111 years]. I excluded recent years since RIC has been putting up video game numbers. There were only 4 warmer starts to summer in Richmond prior through 1980. It's only 22nd warmest on record for Knoxville, Tennessee, for the period ending in 2001 [131 years]. Again, I excluded recent years due to temperatures being played on easy mode in Madden type numbers. It's only 26th warmest on record for Raleigh, for the period ending in 2001 [115 years]. It's only equal to or warmer than two years on record in Tallahassee, with 3 more within 1/2 of a degree. And it's second coldest on record for Orlando, nearly a degree warmer than the bone-chilling Orlandonian summer of 1930.
  10. Looks like 92 was the official high again. Pretty impressive actually considering the cloud cover. Morgantown made it up to 95F, but there was more sun there this afternoon.
  11. Another scorcher in the Research Triangle. Up to 97 at the Southeast Regional Climate Center today, with a peak heat index of 112F.
  12. 8 straight 5-minute observations of 90. Got to think even with the rounding at some point, it ticked up sufficiently to reach a true 90F reading. But yeah, cloud cover was much too heavy for to see 94+. Not sure where the NWS was expecting to see heat advisory criteria met today, let alone heat index values up to 104F.
  13. It was 98 yesterday at the Southeast Regional Climate Center in a shady, tree-filled, low-density part of Chapel Hill, but carry on kvetching about the RDU temperatures. Pretty sure the SERCC knows how to reliably measure temperature.
  14. Edit: Confirmed. We can see the extreme thermometers were installed on a roof of the Administration Building through December 18, 1957, then on a rooftop of the International Arrival Building through at least May 26, 1961. A hygrothermometer was first installed 3400' south of the International Arrival Building at an appropriate 4' AGL on May 26, 1961. Prior to that installation, all of the readings were noncompliant, rooftop readings.
  15. It's not possible for JFK to have had so many 100+ readings in the early decades. I suspect if you looked at the station metadata, it was installed on a rooftop. Without some sort of correction or adjustment, hard to draw any sort of conclusion about the trend in 100+ heat at JFK.
  16. Up to 95F at Central Park so far, which is the hottest day since August 9, 2022.
  17. Technically 2017-2018 was 0.1F above the 1981-2010 normal then in effect, by my calculation. The so-called "normal" keeps being raised.
  18. Can't wait for someone to "ackshually" me. The 1966 reading is bogus - the park was 1.2F warmer than the EWR tarmac? GTFO. Yes, it's cooler than 2010, but not by a margin that anyone can discern. And, of course, the 1994 readings are absolute garbage measured by garbage equipment that are not comparable to readings taken before or after that era.
  19. Like I said, people "feel" it's cooler because they are so acclimated to the heat. I told the Harrisburg crew they should be comparing their summertime temperatures to Richmond, Virginia. It's one of the hottest summers on record at Harrisburg, but compared to historical, 20th century summers in Richmond, it's still a very hot summer just not among the record hottest. I think the New York City crew should do the same. 2024 is among the hottest on record at NYC and EWR, but if you compare NYC to 20th century Richmond and EWR to 20th century RDU then it's still well above the typical summer in those locations during that era, but not among the hottest on record. The DC crew can use Tallahassee or Savannah, Georgia.
×
×
  • Create New...