If You Put Junk Science In, You’ll Get Junk Science Out

By Chris Martz | September 16, 2019

There are plenty of climate scientists in the world that I highly respect, many of whom I don’t share the same views with on climate change. However, these scientists are respectful towards others, they’re pretty honest with their data, and still have scientific integrity.

There are a select few scientists out there, however, whom I have lost all respect for - Dr. Michael Mann being one of them.

I haven’t lost my respect for Dr. Mann because I disagree with his claims that there is a climate crisis (that I could care less about; that’s his opinion and he’s entitled to it), but rather for the way HE treats other scientists and his poor judgement and representation of data.

His true colors really shined last Friday, September 13 when he posted this Tweet:

The chief meteorologist at the CBS affiliate in Boston, Eric Fisher, posted a Tweet noting the record heat that was occurring in the Southeastern United States Friday afternoon.

On Friday, temperatures as high as 100°F (37.8°C) were recorded in Alabama (Huntsville), and many locations in Alabama and nearby states neared, tied, or set new daily record highs.

Just to the east in Georgia, Atlanta recorded record high temperatures on four consecutive days (Figure 1).¹ The high of 96°F (35.6°C) on September 11 was a tie with 2010, thus it was not a new record.¹

This image has an empty alt attribute; its file name is image.png
Figure 1. Highest maximum temperatures recorded in Atlanta, Georgia by day between September 10 and September 13. – Threaded Extremes.

Of course, Michael Mann decided to quote Eric’s Tweet and use this heatwave as an opportunity to ridicule University of Alabama Huntsville‘s (UAH) Dr. Roy Spencer, a meteorologist, and Dr. John Christy, Alabama’s state climatologist, who monitor satellite-based global temperatures.

Mann stated “Hey John Christy & Roy Spencer: hope you’re not finding that record century mark heat too uncomfortable down there in Huntsville.”

Aside from Mann making himself look like a fool, he couldn’t get his facts straight either.

The supposed record high of 100°F (37.8°C) on Friday in Huntsville that Dr. Mann alludes to wasn’t even a record. Had Mann actually looked at temperature data from NOAA, he’d have realized that September 13th’s record high actually still stands at 101°F (38.3°C) from 1927.¹

This image has an empty alt attribute; its file name is image-1.png
Figure 2. Highest maximum temperatures (degrees F) in Huntsville, Alabama by day between September 1 and September 13. – Threaded Extremes.

Dr. Spencer stumbled upon Mann’s Tweet and replied stating that Alabama has around 100 years of temperature data showing no long-term warming trend, in addition to the very simple fact that we can not confuse weather and climate; something both most climate activists AND quite a few skeptics have a hard time comprehending.

Instead of cherry picking a handful of days to declare a climate emergency, I’m going to look at the long-term trends, because that is what climate is.

Raw NOAA USHCN data from Tony Heller‘s UNHIDING THE DECLINE software shows little to no warming in Alabama over the last 124 years. Two takeaways from this are a.) there is a lot of year-to-year variation, and b.) there’s a lot of multidecadal variability.²

Alabama’s warmest year on record was 1927, with an average statewide temperature of 77.2°F (25.1°C). Conversely, Alabama’s coolest year on record was 1903 with an average temperature just above 67°F (19.4°C) (Figure 3).

This image has an empty alt attribute; its file name is average-temperature-at-all-alabama-ushcn-stations.png
Figure 3. Average temperature at all Alabama USHCN stations. – UNHIDING THE DECLINE software.

In 1954, all USHCN stations across the state averaged around 109 days (~30% of the year) with maximum temperatures of at least 90°F (32.2°C) (Figure 4).

This image has an empty alt attribute; its file name is percent-of-days-at-or-above-90c2b0f-at-all-alabama-ushcn-stations.png
Figure 4. Percent of days at or above 90F at all Alabama USHCN stations. – UNHIDING THE DECLINE software.

So, instead of apologizing to Roy, Michael decided to dig his hole deeper and cherry pick data. He responded with “Who’s confused Roy?”

The map that Dr. Mann posted is from a 5-year-old Climate Central article. Climate Central is an organization that is made up of scientists and journalists who research and report facts [and opinions] about climate change. The map shows meteorological summer temperature trends in the U.S. from 1970 to 2013 (Figure 5).³

So, right off the bat, this map is essentially useless now since it hasn’t been updated in six years.

This image has an empty alt attribute; its file name is 6_3_14_Brian_SummerTemperatureTrends_720_405_s_c1_c_c.jpg
Figure 5. Summer temperature trends. – Climate Central.

Even so, I would like to address two other issues I have with this map.

The first issue I have is that the data used in this map is not only NOT up-to-date, but also, the beginning of the time period used is 1970.

The U.S. has by far the most coherent surface temperature record in the world with many station databases extending as far back as the 1890s. Globally, only 762 stations have data spanning from January 1905 to January 2019 (Figure 6), and that number is significantly less the farther back you go.⁴ Most of these stations are in the U.S. Since 1880, only 106 stations have been continuously active globally.⁴

This image has an empty alt attribute; its file name is image-3.png
Figure 6. GHCN V3 station locations in January 1905 and January 2019. – NASA GISS.

Because the U.S. has such a good temperature record, I find it rather peculiar that Climate Central chose to start their data in 1970.

In fact, starting graphs in 1970 is quite a common move by Climate Central. I have pointed out numerous times on Twitter that their graphs need to cover the entire period of record, or at least back to 1900 to see the full trend.

The graph below (Figure 7), titled “U.S. Summers Are Getting Hotter,” was published by the organization back in June of this year.⁵ As usual, they started the x-axis in 1970 and they completely obliterated the period 1895 to 1969.

Summer Temperature Trends in the Contiguous U.S.
Figure 7. U.S. summers are getting hotter. – Climate Central.

Why did they do this? Answer: because they want to fool their followers into thinking the U.S. is burning up. Had they included pre-1970 data, their argument wouldn’t hold water.

The average summer temperature across the U.S. has increased slightly over the last 124 years (Figure 8).⁶ The Dust Bowl era had summers just as warm as recent summers, the only difference being the 1930s had warmer days and cooler nights while more recent years have had cooler days, but warmer nights.

This image has an empty alt attribute; its file name is image-4.png
Figure 8. Contiguous U.S. average summer temperature. – NOAA Climate At A Glance.

In terms of describing how hot a summer has been, I don’t like using the average temperature (mean of the maximums and minimums). Rather, I like to use the average maximum temperature, because overnight lows are not generally “hot.”

The average summer maximum temperature has seen even less change since 1895 (Figure 9).⁶ The 1930s by far had hotter summers than anything that we’ve seen in the last four decades.

This image has an empty alt attribute; its file name is image-5.png
Figure 9. Contiguous U.S. average summer maximum temperature. – NOAA Climate At A Glance.

Another metric I like looking at are heat wave indices.

Heatwaves are very typical of summer and thus the frequency and intensity of them are very important statistics for long-term trends.

According to the Fourth U.S. National Climate Assessment, the average duration of heat waves has declined from eleven days during the 1930s to 6.5 days during the 2000s (Figure 10).⁷ That’s a 41% drop!

This image has an empty alt attribute; its file name is 9e7cf-warm2bspells.png
Figure 10. Warm spell duration. – U.S. Fourth National Climate Assessment.

Moreover, the average maximum temperature during a heat wave has also declined from 101°F in the 1930s to 99°F since the 1980s (Figure 11).⁷

This image has an empty alt attribute; its file name is baac8-warmest2btemperature.png
Figure 11. Average warmest temperature each year. – U.S. Fourth National Climate Assessment.

The other issue I have with the Climate Central map doesn’t have to do with the map itself, but rather with the way it was being used.

In Mann’s original Tweet, he attempted to blame the heatwave on global warming. This is probably because he has no training in meteorology, and thus can’t come up with a meteorological explanation for this. If he just did a simple Google search, he’d find some really good articles on heatwave mechanisms… just sayin’.

It’s also important to note that the heatwave Dr. Mann is referring to has occurred in September. Because of this, the Climate Central map he posted is an invalid scientific argument against Dr. Spencer. He’d have a much stronger case if he were to post September temperature trends, specifically for Huntsville, Alabama, which is the location he mentioned in his initial Tweet.

So, let’s take a look at the average maximum temperature in Huntsville for the first 15 days of September throughout the entire period of record.

So far this month, the average maximum temperature in Huntsville has been exactly 95°F (35°C).⁸ 2019’s month-to-date average maximum temperature for the city is so far ranked as 7th warmest.⁸ That is statistically significant, no doubt about it (Figure 12).

Another thing to note about the table below is that of the top ten warmest first 15 days of September for Huntsville, eight occurred prior to 2000, seven occurred prior to 1960, and six occurred just in the span of 17 years between 1922 and 1939. ⁸ That’s even more statistically significant than this year ranking 7th place.

This image has an empty alt attribute; its file name is image-6.png
Figure 12. Maximum 15-day mean max temperature for Huntsville, Alabama. – xmACIS2.

The average maximum temperature for September 1-15 in Huntsville, using the standard period 1981-2010, is 86.8°F (30.4°C).⁸ Since 2000, nine Septembers have had average maximum temperatures at or below average for the first 15 days (Figures 13 and 14).⁸

This image has an empty alt attribute; its file name is image-7.png
Figure 13. Average September 1-15 maximum temperature in Huntsville, AL. – xmACIS2.
This image has an empty alt attribute; its file name is image-9.png
Figure 14. Average September 1-15 maximum temperature in Huntsville, AL. – xmACIS2.

During the period 1920-1939, only seven September 1-15ths were cooler than average by maximum temperature (Figure 15).⁸

This image has an empty alt attribute; its file name is image-8.png
Figure 15. September 1-15 average maximum temperature in Huntsville, AL. – xmACIS2.

As for the leftover Septembers that were above average for the first 15 days, the average temperature departure was 6.6°F during 1920 to 1939 and 3.9° from 2000 to 2019, nearly 3° cooler than 1920-1939.

These statistics are very easy to find and take little time to evaluate. Instead of Dr. Mann being honest and/or credible with his data, he cherry picked one hot day of weather and ignored the long-term trend, which is what climate is.

Here’s an analogy: weather is like the atmosphere’s mood, climate is the atmosphere’s personality. Both are prone to change, but on much different time scales.

If you put junk science in, you’ll get junk science out.


[1] Threaded Extremes. Accessed September 16, 2019. http://threadex.rcc-acis.org/.

[2] Heller, Tony. “UNHIDING THE DECLINE For Windows.” August 14, 2017. Accessed September 16, 2019. https://realclimatescience.com/unhiding-the-decline-for-windows/.

[3] Kahn, Brian. “Here’s How Much U.S. Summers Have Warmed Since 1970.” Climate Central. June 3, 2014. Accessed September 16, 2019. https://www.climatecentral.org/news/us-summer-temperatures-climate-change-17510.

[4] “GISS Surface Temperature Analysis.” NASA GISS: NASA Goddard Institute for Space Studies. Accessed September 16, 2019. https://data.giss.nasa.gov/gistemp/stdata/.

[5] “Summer Temperature Trends in the Contiguous U.S.” Climate Central. June 5, 2019. Accessed September 16, 2019. https://www.climatecentral.org/gallery/graphics/summer-temperature-trends-in-the-contiguous-us.

[6] “National Time Series.” Climate At A Glance | National Centers For Environmental Information (NCEI). 2018. Accessed September 16, 2019. https://www.ncdc.noaa.gov/cag/national/time-series/.

[7] “Chapter 6: Temperature Changes in the United States.” CSSR. 2017. Accessed September 16, 2019. https://science2017.globalchange.gov/chapter/6/.

[8] xmACIS2. Accessed September 16, 2019. https://xmacis.rcc-acis.org/.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s