Weather records have tumbled across North America, with freezing temperatures even in the southern US.
The most extreme arctic blasts, blamed on a weather pattern known as the polar vortex, were said to have affected nearly 190 million people.
In Kentucky, an escaped prisoner turned himself in to get out of the cold.
Some parts of the Midwest hit -26C (-14F), as low as the Antarctic coast in winter, and much colder than the inside of a domestic freezer.
Temperature records were shattered in states across the US, including Alabama, Georgia, Tennessee, Arkansas, Michigan, Maryland, Ohio, Pennsylvania and New York.
Remember that record breaking cold spell which large parts of North America suffered through last month? The worst affected parts were the East coast and Midwest.
It might surprise citizens in those parts to learn that the temperature during January was little colder than average. In the Northeast for instance, the region that includes New England along with NY, PA, NJ, DE and MD, temperatures look positively mild.
Of course, the record cold did not last all month, and there were a couple of mild spells later in the month.
But how does the NOAA summary above compare with the actual temperature data on the ground?
Last month, I analysed one climate division in New York State, “Central Lakes”. The analysis showed how the 1940s had been cooled by about 3F at every single station in the division with data records for then and now, with the exception of the international airport at Syracuse.
So let’s take this same group of stations, which are all high quality USHCN ones, and see how they compare with NOAA’s declared version for last month.
According to NOAA, last month’s mean temperature was just slightly less than the 1901-2000 average, similar to the overall Northeast trend.
In my previous work, I compared current temperatures with January 1943. According to NOAA, the average temperature for Central Lakes was 20.8F last month, compared with 18.7F in January 1943.
In other words, it was 2.1F warmer last month than in 1943.
The actual temperatures recorded in January 1943 are below, and come from the Monthly Climatology Report for NY State at the time.
Take a close look at Ithaca (2), which the Monthly Report clarifies is Cornell University (see here) – the 1943 temperature was 20.6F. (Note that the temperature at the other Ithaca site is even higher!)
The official Northeast Regional Climate Center have a separate page for Ithaca Cornell, as it is long running, high quality, well maintained, and representative of the area.
So the average mean temperature last month was 19.6F, which means it was 1.0F colder than in January 1943.
Recall that NOAA say Central Lakes was 2.1F warmer last month than in January 1943. That makes a total difference of 3.1F.
We can check last month’s temperature against the original station record, shown below.
This shows an average of 20F, slightly higher than NRCC. However, note that the entry for 1st Jan is missing. It was exceptionally cold that day. If we add that back, we end up with an average of 19.4F. (The daily numbers all seem to agree, so I presume there is a small rounding error here)
In my post last month, I listed five other stations in the division. Unfortunately, three have missing data last month, Avon, Dansville and Hemlock. The latter for instance has 9 days missing:
But if we look at the two with full data, Auburn and Geneva, we find the same pattern as Ithaca. That is, last month was colder than January 1943:
- Auburn - Jan 2018 = 20.5F : Jan 1943 = 22.2F : Diff –1.7F
- Geneva – Jan 2018 = 21.0F : Jan 1943 = 22.3F : Diff –1.3F
So at the three sites of Ithaca, Auburn and Geneva, we find that January 2018 was colder than January 1943 by 1.0, 1.7 and 1.3F respectively.
Yet NOAA say that the division was 2.1F warmer last month. NOAA’s figure makes last month at least 3.1F warmer in comparison with 1943 than the actual station data warrants.
Could there possibly be such a large difference within the Central lakes division? It seems impossible.
New York is split into ten climate divisions all of a similar size, so it is a relatively small area. The divisions are also structured to be of similar climatic nature. (They would avoid mixing coastal and mountain areas, for instance).
It is quite inconceivable that there could be such large variations within such a small area.
Clearly NOAA’s highly homogenised and adjusted version of the Central Lakes temperature record bears no resemblance at all the the actual station data.
And if this one division is so badly in error, what confidence can there be that the rest of the US is any better?
Station data for January 2018 is here: