Guest Post by Bob Tisdale
This is the second in a series of posts about the blog HotWhopper, specifically about the misunderstandings about climate-science basics displayed by its author and proprietor Miriam O’Brien (a.k.a Sou from Bundangawoolarangeera). The first in the series was Open Letter to Miriam O’Brien of HotWhopper (a.k.a. Sou), which was cross posted at WattsUpWithThat here. In this post, we’ll discuss her misunderstandings about a very basic climate metric—one called sea surface temperature.
Keep in mind that Miriam O’Brien’s blog Hot Whopper found a new focus after Miriam, blogging as “Sou”, was permanently banned from WattsUpWithThat for her troll behavior. See Anthony Watts’s post My Blog Spawn. Since her departure from there, Miriam’s new focus has been, of course, the blog posts at WattsUpWithThat. Seemingly, Miriam O’Brien opposes anything and everything presented at WattsUpWithThat.
THE TOPIC OF THIS POST
The paper Kuffner et al. (2014) A Century of Ocean Warming on Florida Keys Coral Reefs: Historic In Situ Observations was getting a good amount of press a month or so ago. See the USGS press release here. For much of the paper, Kuffner et al. focused on two sets of sea surface temperature records for two coral reefs that are part of the Florida Keys. Kuffner et al. used sea surface temperature measurements made by lighthouse keepers back in the late 19th Century to early 20th Century and compared them to the buoy-based samples of sea surface temperatures that started in the late 20th Century. The more-recent buoy-based readings were warmer than the early lighthouse-keeper observations, so Kuffner et al. (2014) concluded (This quote is from their abstract):
Results indicate that the warming observed in the records between 1878 and 2012 can be fully accounted for by the warming observed in recent decades (from 1975 to 2007), documented using in situ thermographs on a mid-shore patch reef. The magnitude of warming revealed here is similar to that found in other SST datasets from the region and to that observed in global mean surface temperature.
As a visual reference, my Figure 1 is Figure 3 from Kuffner et el. 2014.
That paper raised a number of very obvious questions:
- What happened to the sea surface temperatures in that region between the turns of the 20th and 21st Centuries?
- Did they remain flat, increasing slowly until the 1970s, when they skyrocketed? The abstract might lead some people to think that was the case, or,
- Did they cycle, consistent with the variability of the Atlantic Multidecadal Oscillation? or,
- Did they warm early in the 20th Century and cycle since then.
- Long-term data, covering the periods examined by Kuffner et al. and running continuously between them, are available. Kuffner et al. even refer to them in their abstract with, “The magnitude of warming revealed here is similar to that found in other SST datasets from the region…” Why didn’t Kuffner et al. (2014) present that basic background data?
I presented long-term data in my post Data Reveal Florida Keys Sea Surface Temperatures Haven’t Warmed in 80+ Years* to answer those questions. I selected a very small region of the Florida Keys that encompassed the two reefs focused on by Kuffner et al. Using a sea surface temperature dataset with a high resolution (HADISST), I presented the long-term data in raw form here, and then smoothed them with a 121-month filter to highlight the underlying variability. See Figure 2 (which was Figure 3 from my earlier post). The greatest warming in the region took place from around 1910 to the early 1940s, which undermined the claim by Kuffner et al. that “the warming observed in the records between 1878 and 2012 can be fully accounted for by the warming observed in recent decades (from 1975 to 2007)”.
The recent downturn in the data also doesn’t help their cause.
I also examined how far back in time we could go with the long-term sea surface temperature data for that part of the Florida Keys while not showing any warming (based on the linear trend). See Figure 3 (which was Figure 4 from my earlier post). Based on the linear trend, the sea surfaces for that part of the Florida Keys haven’t warmed since 1930, more than 80 years. And as I wrote in my earlier post, Yet, somehow, we’re supposed to believe manmade greenhouse gases are causing harm to the coral in recent years.
Kuffner et al. also set their eyes on August sea surface temperatures, because they are seasonally warmest and would do most harm to the coral reefs. I then presented the August sea surface temperatures for that region, based on the HADISST data. Refer to my Figure 4 (which was Figure 5 from the earlier post). The long-term August data confirmed that coral have had to deal with sea surface temperatures that are said to be “stressful” almost every year, and that sea surface temperatures regularly reached and exceeded levels that are said to be “very stressful” in the 1940s, 50s and 60s…and, if the early data are believable, on occasion, they were above very stressful levels in the late 1800s and early 1900s. In fact, many of the peak August temperatures in the 1940s, 50s and 60s were higher than they have been recently.
Kuffner et al. show similar excursions into “very stressful” temperatures in the late-19th and early-20th centuries in their Figure 2. They even provided dashed lines to highlight those temperatures, but they failed to point them out. Those “very stressful” temperatures disappeared when Kuffner et al. used multidecadal averages for their Figure 3 (shown above as my Figure 1).
Bottom line, Kuffner et al. referred to local long-term sea surface temperature datasets in their abstract—“The magnitude of warming revealed here is similar to that found in other SST datasets from the region…”—but failed to present that data. The basic reason they did not present it appears to be that it does not support their claim that “the warming observed in the records between 1878 and 2012 can be fully accounted for by the warming observed in recent decades (from 1975 to 2007)”.
For some reason, Miriam O’Brien of HotWhopper was displeased with my post and published Perennially Puzzled Bob Tisdale surfs the surface at Florida Keys. (An archived version is attached to link. My thanks to blogger 7DaBrooklynKnight7 for the archived version.) Her post is quite humorous. I think you’ll enjoy it.
Unsuccessfully, Miriam O’Brien tried a number of different tactics to counter my data presentations, without ever addressing the fact that my data presentations were correct.
Miriam’s Misdirection 1 – Miriam claimed that the region I selected was too large. But, of course, she failed to show that there was any difference between the sea surface temperature data I presented and the sea surface temperature data presented by Kuffner et al.
A blogger called 7DaBrooklynKnight7 noted in a comment:
i checked that knmi website and the average august sst from the fowey rocks and carysfort reef buoys (for 1991 to 2012) is the same (29.9 deg c) as the august sst from the hadisst for the coordinates used by tisdale. please check my work.
That wasn’t well received by Miriam.
I did confirm 7DaBrooklynKnight7’s statement. For the period of 1991 to 2012, the average of the August sea surface temperatures for the Fowey Rocks and Carysfort Reef buoys (29.9 deg C) is the same as HADISST for the coordinates of 24N-25N, 81W-80W that I used in my post. The data for the Fowey Rocks and Carysfort Reef buoys were provided by Kuffner et al. through the webpage here. The Excel spreadsheet with the Carysfort Reef buoys data are here, and the Fowey Rocks data are in the spreadsheet here. And, of course, the HADISST data are available at the KNMI Climate Explorer.
Miriam’s Erroneous Claim 1 – After stating correctly that I had presented “gridded data estimated from combining observations from ships, from buoys, from satellites, with data gaps filled by interpolation”, Miriam then incorrectly claimed that I was “only interested in the temperature of the thin skin of the sea surface”. She repeated that claim of “skin” temperature in the post and in her comments on the thread.
It turns out, the only “thin skin” is Miriam’s, who responded with insults when a blogger point out her errors. More on that later.
HADISST is a sea surface temperature dataset, not a sea skin temperature dataset as Miriam claims. The metric presented by Kuffner et al. (2014) in their Figures 2 and 3 was sea surface temperature—same metric I presented. Miriam is correct that HADISST includes the skin temperature observations from satellites, but she fails to acknowledge:
- that satellite data only extend back in time to the early 1980s,
- that there are also “surface temperature” measurements from buoys and ship inlets in that time, and,
- more importantly, that adjustments have been made to account for satellite biases. The make-up and corrections applied to the HADISST dataset were discussed in the paper Rayner et al. (2003) Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century.
Miriam even provided a link to Rayner et al. at the end of her blog post. She either failed to comprehend the paper or she missed the part of the paper where Rayner et al. state:
We adjusted the satellite SSTs to be unbiased relative to the in situ data (Appendix C).
This was pointed out to Miriam by blogger 7DaBrooklynKnight7:
according to the rayner paper you provided they also adjust the satellite data with data from ship inlets and buoys, like the buoys at the reefs, to take care any biases from the satellites sensing the skin.
That also was not well received by Miriam.
Miriam’s Misdirection 2 – Miriam wrote:
But in this case, because deniers want to pretend that coral bleaching doesn’t happen, the world isn’t warming etc etc, they decide to ignore the careful measurements taken on site over the years. Data that is much more appropriate when considering the actual reef. Deniers decide that this time around they prefer data that’s been “tampered with”. Data from multiple sources, with gaps interpolated.
I never claimed that coral wasn’t stressed by high sea surface temperatures. That’s an outright fabrication. In fact, I presented a graph that showed that sea surface temperatures in that region reached levels that are said to be “stressful” to coral almost each and every year, and that sea surface temperatures regularly reached and exceeded levels that are said to be “very stressful” in 1990s and 2000s….and in the 1940s, 50s and 60s, too. Her post is riddled with off-the-cuff remarks that have no basis in fact. I present data, Miriam fabricates.
Miriam apparently wants to dismiss decades of research into the adjustments required to correct for different methods used for sampling sea surface temperatures. Granted, many persons are concerned about the adjustments, especially when, globally, they suppressed the warming during the late-1930s and early-1940s that existed in the source ICOADS data, which impacts the amount of warming during the early warming period of the 20th Century. See Figure 5, which is Figure 15 from the post Multidecadal Variations and Sea Surface Temperature Reconstructions.
The best part: Miriam obviously forgot that Kuffner et al. presented sea surface temperature data from “multiple sources”. Kuffner et al. presented thermometer readings taken by lighthouse keepers during the late 19th and early 20th centuries, using one technology with biases. And they presented temperature measurements taken from buoys in the late 20th and early 21st centuries.
Miriam’s Misdirection 3 – Miriam wrote (my boldface):
He was only interested in the temperature of the thin skin of the sea surface – averaged over a wide area well beyond the corals the scientists were researching. Yet leaving out some of the area that the researchers did cover.
The long-term sea surface temperature data that I presented was for coordinates that included the two reefs presented in Figures 2 and 3 of Kuffner et al. (Fowey Rocks and Carysfort Reef). The only other temperature data presented by Kuffner et al. in a time-series graph was the sporadic late 20th century subsurface temperatures for the Hen and Chickens Reef. (See their Figure 4.) It wasn’t a long-term sea surface temperature comparison, like those at Fowey Rocks and Carysfort Reef, and was not applicable to my post.
A Few More Comments by A Blogger on the HotWhopper Thread – Blogger 7DaBrooklynKnight7 presented many of the realities I’ve discussed in this post. But he also made a couple of other observations. 7DaBrooklynKnight7 wrote:
one last thing. i also checked that knmi website and they have more sst data. there’s one from noaa that does not use satellites in recent years. the noaa ersstv3b, for the florida keys, the sea surfaces have cooled sinse 1930.
Cooled, he wrote? From a dataset that excludes satellite data? It turns out that the negative trend is so slight that it’s basically flat.
Keep in mind that the minimum grid size of the NOAA ERSST.v3b data is 2-deg latitude by 2-deg longitude. So this is a larger area than what was presented in my first post about Kuffner et al. ERSST.v3b also extends further back in time, and the further back one goes, the less realistic the data become. (The hump that peaks in the 1880s looks suspicious.) With that in mind, Figure 6 presents the sea surface temperature anomalies for the coordinates of 23N-25N, 81W-79W. Also shown is the linear trend for the period of January 1930 to September 2014. A cooling rate of -0.008 deg C/decade is essentially flat. Sorry, 7DaBrooklynKnight7, you got a little carried away with “cooled” since 1930.
But, along with the similarities in the buoy temperatures and the HADISST-based readings discussed above, the ERSST.v3b dataset (which does not use satellite data) do help to dismiss Miriam’s complaints about satellite “skin” temperatures. (NOAA actually removed the satellite data from their ERSST data.)
Blogger 7DaBrooklynKnight7 also asked a very basic question of Miriam:
something else you’ve overlooked. are the data from the lighthouses and from the buoys included in the hadisst data?
Miriam ignored the question.
And that’s a question that I cannot answer with any certainty. The buoy data should be included (key word “should”) in the ICOADS sea surface temperature dataset, which is the source data for the NOAA and UKMO datasets. The lighthouse data might be, key phase “might be”.
Miriam’s Response When Confronted with Data-Based Questions – First Miriam insulted blogger 7DaBrooklynKnight7. Then when he responded to her incivility and exposed more problems with her post, she deleted his comment. The exchanges between Miriam (who poses as blogger “Sou”), 7DaBrooklynKnight7, and the HotWhooper denizens are worth a read. They start with 7DaBrooklynKnight7’s initial question at September 13, 2014 at 1:33 AM. They extend down to his response to her insults at September 15, 2014 at 8:07 PM. (Miriam deleted his final comment, as visible at a direct link to where that comment once existed.) After 7DaBrooklynKnight7 was obviously banned, bloggers continued to insult him at HotWhopper.
Blogger 7DaBrooklynKnight7 then wound up at my blog, alerting me to the exchange at HotWhopper.
Thanks, 7DaBrooklynKnight7. Sorry it took so long for me to respond to Miriam’s post.
I really do enjoy reading Miriam’s posts at HotWhopper. She is willing and able to expose her misunderstandings about the topics she addresses, along with her want and need to misinform her readers. I also enjoy replying to her posts. You can look forward to more of my posts about Miriam O’Brien and her escapades at HotWhopper.