TheClimateChanger

Number of days with 1"+ snow cover at MFD [dating to early 1950s]:

It's high time the NWS amends the description of climate in Mansfield, Ohio. There is not near continuous snow cover from December through March anymore. And daytime temperatures regularly exceed freezing in the winter. That would be considered absurd today. There's barely been 20 inches the last two winters combined. And 3 of the top 10 least snowy winters are in the last 5 years. I really believe Ohio is ground zero for climate change. Feels like it's warming faster in this region than anywhere on the planet.

Canton-Akron (CAK) Official records, since 1948: Manually expanded dataset back to 1893, including snowfall records from other sites in the threaded record that aren't included in the official record for some reason: Mansfield, OH (MFD) - extended back to 1892-1893, excluding several years with no or significant missing data

Some other snowfall futility notes. Currently, fourth least on record at Erie, PA. Note these values have been manually corrected by me. xMacis and the NWS CLE site [which uses the xMacis] has a bunch of random missing data mainly from the 1920s-1940s, where snowfall actually was observed. Looking at these values, it will be the first season with less than 40" since 1952-1953, and the first season with less than 30" since 1932-1933 [barring some late season magic]. In recent decades, the low to mid 40-inch range was the absolute minima at Erie Airport. In fact, if we narrow the search to the most recent 71 winters [dating back to 1953-1954] at the airport site, it's UGLY. The least amount of snow observed in the 70 years from 1953-1954 to 2022-2023 was 41.2 inches in 1982-83. Last year's 52.0 inches was sixth least in that stretch. This winter is at 24.8 inches. Wow.

Here's Lansing, Michigan snowfall records [earlier years at East Lansing, Michigan], excluding a few years with excessive missing data: Maybe a bit questionable that the very first year with data ranks as lowest. Looking at the monthly data, that looks mostly a function of the fact that precipitation was 25-50 percent of modern norms for the bulk of the cold season.

I said from Toledo to DTW airport. All of the records prior to the mid 1950s are from the city proper. Of course, it's a bit longer drive out to the airport. The snowfall records were retrieved from xMacis, which has nothing prior to 1890-91.

Interesting long-term data from Lake Mendota in Wisconsin.

Same story in Toledo, less than an hour drive from DTW airport. Of course, @michsnowfreaklikes to point out a 6.0" observation from the winter of 1889-1890, but I'm guessing that lacks daily resolution as it is not part of the threaded record, which begins in 1890-1891.

Same in Ohio. Just look at Cleveland the last two years. There hasn't been an above normal season since 2014-2015, and that was actually slightly below the normal in effect at the time [normal snowfall dropped several inches with the 1991-2020 update]. 2013-2014 was the last time a season came in above the existing normals. Note all of the low snow years were from the city office data. The heaviest snow occurs south and especially east of the city (see below), so we can conclude snowfall would have been heavier at the airport location. Of course, snowfall measurement procedure has also changed over that time frame. Of note, Alfred Hyde was a long time cooperative observer for the Smithsonian Institution who kept temperature, precipitation and snowfall records back to 1854-1855. His lowest snowfall winter was 1865-66 with 21.6 inches. Looking at this, we can conclude only one year on record had substantially less snow than last winter dating all the way back to the mid 19th century: 1918-1919. The other two years with less (1931-1932, 1865-1866) were minor differences (1.2 inches, and 0.1 inches, respectively). Also of note, prior to last winter, there had never been less than 30 inches observed in a season since records moved to the airport [around 1940, give or take a year]. The last time less than 30 inches was observed was 1932-1933. There is a good chance that metric is about to be exceeded for the second consecutive winter, barring some late season action.

Seems unlikely that he's had ten years under 24 inches. The official records at Portland, which date to 1881, show the lowest annual total at 27.5 inches.

Interesting. I just checked Pittsburgh, which is still on track for two consecutive seasons of 20” or less. Looks like it’s only happened once dating back to 1879, but it was three seasons in a row (1930-31, 1931-32, and 1932-33). Of course, that was when observations were still downtown at the city office, which is about 400’ lower in elevation and averaged about a foot less than at the airport (although some of that is probably attributable to changes in observation procedure). Prior to last year, there had only been three years since then with less than 20” (1937-38, 1973-74 & 1990-91).

Perhaps most insane is the 47.9F mean at Jefferson County Airport, at more than 1800 feet elevation, would be third warmest on record at low-elevation Harrisburg in the southern part of the State!

At DuBois, the first 17 days of March have been 2F warmer than the same period in any of the year since records began in 1963: At Bradford, the first 17 days of March have been 2.7F warmer than the same period in any other year dating back to 1958:

In what has become a common refrain, yet another scorching start to the month in the Commonwealth. Harrisburg Williamsport Erie Philadelphia Pittsburgh

Second warmest start to the month, through yesterday.

This is a good discussion for the banter thread. Much more topical than most posts that are made in the banter threads.

The so-called hottest summer on record was downright pleasant in Detroit outside of a single week. Even including the 7-day heat wave, it was a moderate summer and fairly mild by modern standards. The mean temperature of 71.2F places 57th in the threaded record. In the past 9 years, only 2015 & 2023 have been cooler (and both by very small margins). By contrast, 2016 was nearly 4F warmer than 1936! And to be honest, it makes much more sense to compare this to records observed at City Airport (DET) since that it was where the official temperatures were observed in 1936. On that metric, it was the 27th COLDEST summer out of 90 years of records. Incredibly, removing one 7-day stretch would drop the summertime mean to 69.9F. I know you can't do this, but it gives a much better reading of the overall character of the summer. And that was one of low humidity and generally mild weather, with cool overnight lows, punctuated by a few episodes of extreme heat, including one colossal 7-day stretch in early to mid-July. At these levels, you'd be looking at the 12th coolest of 90 years at DET and 44th coolest of 150 years in the threaded Detroit area records. The 59.5F mean summer minimum is 5th lowest on record, out of 90 years' worth of observations at Detroit City Airport. That value has not been exceeded in recent decades since the volcanically cooled summer of 1992, and even then only by a negligible amount [59.3F]. In the threaded record, it's only 26th coldest of 150 years, as the coldest minima are dominated by the 1960s-1980s records from DTW when it radiated like a freezer at night [developed in low density suburb]. Still pretty cold either way. Ignoring the 7-day inferno [July 8-July 14], the mean minima the rest of the summer was just 58.4F, which would easily be the coldest on record at DET and would tie 1875 for 10th coldest in the threaded record. Regarding the heat wave, you can pull up the old records and see the humidity and dry bulb temperatures at 3 periods each day. The mean dewpoint was probably around 55-55.5F, based on those observations. This is less than even the "frigid" July of 2009 when the average dewpoint at Detroit was a comfortable 56.3F. During the worst of the heat wave, humidity was generally very low. Under these conditions, the heat index would not greatly vary from the dry bulb temperature. Despite the extreme temperatures, there have been more dangerous conditions in recent summers when factoring in the cumulative effect of high dewpoints and temperatures. The most extreme heat indices in Detroit were experienced not in 1936, but in 1995, when triple digit temperatures were combined with dewpoints nearing 80F - rather than in the 50s. The combination of generally moderate temperatures and very low humidity would have made the summer of 1936 more comfortable than the majority of recent summers in Detroit. If those same conditions repeated this year, and I tried to point out the upper 90s/100s as the result of climate change, you would no doubt change tune and attempt to rebut my assertion by going on about how cool and dry it was for the majority of the summer.

Looks like you guys are doing alright. 4 of the top 16 years in the past 13 summers [with a 150 year period of record]. At least it's not the 19th century. I don't see a single year before 1916.

Some of the co-op data has also, unfortunately, been manipulated. If we compare the raw data from Saginaw to the "official" records, we see numerous instances of mistakes and/or temperatures just completely made up and not borne out by the actual observation. It looks like there's potential issues on the 14th, 23rd and 29th from July 1936. The high on the 14th on the record was observed as 92F, curiously below the set max of 98F on the previous day. I've seen this in the past and the temperature is often bumped up the high temperature, assuming the observer somehow reset the temperature device afterwards. Of course, I would surmise that the set max was simply incorrectly reported and it is showing the max set on the prior day, but bumping it up one day to the day it impacts. In this case, they simply made up a 104F high on that date! Unbelievable - that's the sixth hottest temperature observed at Saginaw, and it wasn't even observed. The high was clearly reported as 92F! I mean maybe you could make a case for adjusting to the set max, but simply making up a value - what?!? And not only making up a value, but generating one of the hottest readings in the entire threaded record from thin air. Same thing on the 23rd. High reported as 79F, but goes in the "official" books as 85F. Not only 6F above the actual observation, but 5F above the reported set max from the prior day. Again, it's one thing to bump it up to the set max. But why make up a reading?!? On the 29th, they bumped the observed high of 70F up to the 77F set max from the prior day. Reasonable, but again relies on the assumption that the observer was reporting the set max from that day - and not the prior day's set maximum. That's how it was supposed to be reported, but you can see how that could be screwed up. Easier to mess that up than somehow reseting the max thermometer a second time without reporting it, IMO. Also, while not a big error, the high on the 10th is clearly reported as 106F, not 107F as shown in the "official" records. Admittedly, the 104F on the 14th is more in line with surrounding observations, but it's worth noting that a sharp cold front passed through that day, and would have made it to Saginaw earlier. It could have generated a wind shift off of the bay into the city. Either way, there shouldn't be record temperatures made up from thin air without justification. The effect of all these errors, irregularities and/or adjustments is to bump up the monthly mean high by 0.8F and the monthly mean average by 0.4F. The "adjustment" to the 29th also has the effect of deleting what would be the third coolest maximum temperature on record for July 29th in Saginaw. So we're going to delete the 70F high because it was maybe (possibly) 77F the previous day when the observer reset the thermometer? Not even certain about that, but maybe. I mean I can see making those adjustments to the climate record [USHCN, GHCN] since there's an explicit adjustment for TOBs, so you want to make sure it's being applied properly. But these are unadjusted values presented for informational purposes only - not for any sort of climate research. What in the world is the justification for adjusting the ThreadEx values?

Just for the record, when you compare the raw data from Flint and Saginaw to Detroit, a good chunk of the difference (admittedly, not all - but a good chunk) is completely explainable by the fact that the old Co-op data was not gathered on a midnight-to-midnight time frame, but rather late afternoon. A 5 pm observation time adds about 1F to the mean temperature relative to a day ending at midnight. Obviously, this is not a problem in the Detroit threaded record since it's always been a manned weather bureau station [with a midnight to midnight day]. Understanding Time of Observation Bias | Climate Etc. (judithcurry.com)

Here are Detroit's hottest summers on record. Just look how unbelievably hot it was in 1933! I guess I was wrong! Cue rant about Urban Heat Island Effect in 5, 4, 3, 2, 1!

It's important to note when people complain about adjustments to the temperature record, what they are really saying is that they want to use flawed data because it gives them a more favored result. Time of observation makes a HUGE difference, especially in the summertime when weather variance tends to be at an annual minimum. We know this, and we correct for it.

I don’t understand this obsession with using flawed data. Why would you use the unadjusted numbers when they have known biases? How does that make more sense than adjusting for the biases?

Summer is almost certain to be a scorcher. Just looking over the numbers for some recent summers, and it's insane. At Toledo, 3 of the top 5 hottest summers have occurred this decade [and it's only 4 years old]. At Mansfield, Ohio, the top 2 hottest summers have occurred in the past 4 years. At Akron-Canton, ALL of the top 5 hottest summers have occurred in the past 8 years. Cleveland has lucked out a bit on the summer heat front, but still 4 of the top 8 have occurred in the past 14 years, and 5 of the top 8 have occurred in the past 19 years. These are all long period of record sites. Despite all this, we'll still hear about how much hotter it was in the 1930s!