Someone Was Wrong On The Internet

To content | To menu | To search

Tag - global warming

Entries feed - Comments feed

Wednesday 12 April 2017

Tornadoes 101

While going through my newsfeed this morning, I decided to look at an article on Bloomberg from two days ago with the title: “There Was Nothing Normal About America’s Freakish Winter Weather” by one Brian K. Sullivan. Of course, my inner grammar zealot had a snit because “About” was capitalized in a title despite the fact it was a preposition and further noted with disdain that there was no period following the initial in Brian K. Sullivan’s name; however, I am not my inner grammar zealot so we’ll ignore Bloomberg’s grammar sins and move on to what made this article fodder for being wrong on the internet.

When it comes to articles that venture into hypotheses regarding global warming, I am always on the lookout for the usual sins that reporters make on this subject, especially the one that treats the occurrances of one year as proof or disproof of global warming. As I am sure all my readers already know (and if you don’t, just fake it), climate trends are not statistically significant on scales of mere years. Decadal data, as an average or a sliding average, is the smallest timescale that has any grip in demonstrating real world climate trends with statistical significance; and of course, centurial data is preferable. Much to my disappointment with finding something to blog about, the author of the article made none of those mistakes, alas. All things considered, it’s actually a nice piece about the weather this Winter and how that fits into evolving hypotheses of global warming.

The author saved the best for last: there was a tornado this February in Massachusetts, something I vaguely remember from the news this Winter; and there in the very last sentence of his article, he goofed. Here’s what he wrote:

“officials confirmed it was a tornado, the first ever in a state that began documenting its weather patterns back in the 1600s. “

Now, something that every fourth grader in tornado alley knows is that tornadoes can occur anywhere – even in New England. When we moved to Maine from Texas, I know I disappointed a friend from Aroostook County who wanted to shock me with the fact that northern Maine gets a small number of tornadoes every year. I’ve lived in tornado alley twice in my life and I’m quite aware that tornadoes came pop up anywhere when the conditions are right. I’m afraid my friend from Houlton, Maine was rather put off by my lack of surprise over Maine tornadoes. After all, there have been tornadoes reported in every state, and in almost every country in the world too. This is bread and butter knowledge if you take advantage of attending one of the National Weather Service’s free weather spotter classes – which you can take whether or not you plan on becoming a spotter. The class is free regardless. I think everyone should take one – you’ll learn a lot about weather you never knew before, even if you’re a nerd like me. Great classes, good stuff to learn, absolutely no money required: so what are you waiting for? Go grab some free weather education – you won’t be sorry!

So how about tornadoes in Massachusetts, then? Well, our Bloomberg author was wrong on the internet – and he was doing so well with that article too. It almost seems a shame to take him to task for it, but alas, I’ve written enough so far to see this blog post to the end. It turns out that the first-ever report of a tornado in what would become the future United States of American was out of the small hamlet of Lynn in Massachusetts Bay Colony, a scant five miles north of Boston. I found the tornado report from Lynn while doing my research for this blog post. I knew the author of the Bloomberg article had been wrong on the internet, though, because I recalled that the worst tornado to ever visit New England had been the June 9, 1983 F4 tornado that leveled part of the City of Worcester, Massachusetts. It came at the end of a three-day tornado outbreak that left hundreds dead and thousands injured, not to mention the requisite millions of dollars of property damage.

The outbreak started on June 7 as a storm rolled off the rockies and set up a line of supercell convective storms. The first tornado activity was a handful of small F0 through F2 tornadoes that briefly touched down on a Sunday afternoon along the eastmost Colorado-Nebraska border and in northwest Kansas. This was followed that evening by several F2, two F3 and one F4 tornadoes gouging their way across central Nebraska and into north-central Iowa. Sioux City and Fort Dodge lucked out by being just missed by the course of several tornadoes and the death toll was a modest 11, mostly because the storms crossed sparsely-populated farmland and ranchland. The storm front moved on overnight over the Mississippi River and into the eastern half of the Midwest.

On Monday afternoon, June 8, 1953, this same storm spawned a line of nine tornadoes 300 miles long north-to-south, crossing the northern Midwest from the top of the Michigan mitten to the middle of Ohio. The worst of the tornadoes was an F5 that cut through northern half of the City of Flint, Michigan, killing 116 and injuring 844. In addition, an F4 started some 30 miles southwest of Toledo which cut up to Lake Erie at Sandusky and then traveled along the lake shore into Cleveland, leaving a 120 mile path of destruction with 17 dead and 379 injured. Overnight, the storm continued east over the mountains of the Appalachian Orogeny.

The early afternoon of Tuesday, June 9 was extremely hot. The stormfront crossed the Hudson River and started its first actions with three inch hail falling on the Connecticut River at the Vermont-Massachusetts border. The National Weather Service recognized that southern New England might be struck by some unprecedented weather – but in the 1940s and 1950s, it was policy not to announce that tornadoes might be on their way for fear of creating a panic in the general public. In a era before the severe storm watch system was put into place in 1972, the National Weather Service announced instead the first severe thunderstorm warning ever issued in New England.

At 2:25 PM, a mile-wide F4 tornado developed just east of the huge Quabbin Reservoir that the late great H. P. Lovecraft often invoked in his famous horror stories placed in Massachusetts. It cut a 40 mile long path of devastation into the City of Worcester, killing 94 people in the hour and a half it was on the ground and injuring 1228 others. As it died, an F3 developed 6 miles south of Worcester in Milbury, It traveled some thirty miles to where it died out in Foxboro on the Massachusetts-Rhode Island border. The last two spots of activity of the outbreak were two tornadoes that briefly touched down just to the west of the City of Portsmouth, New Hampshire before the squall line of storm cells moved out over the Atlantic Ocean.

In total, the 1953 “Flint-Worcester” Outbreak spawned 50 tornadoes over 3 days, killing a total of 247 people, injuring 2562, and causing over 2 billion dollars of property damage. The Flint tornado is currently rated the tenth most deadly tornado in the history of the USA and the Worcester tornado is currently rated as the twenty-second most deadly. Of the 25 most deadly tornadoes in the country, the Worcester tornado of 1953 is the most easterly and the only one in New England; all the others are in the South or the Midwest.

All of the references today deserve a visit if you want to learn more on this subject. The Stormstalker blog is highly recommended and the Tornado History Project website is a marvel to cruise around. If you click on the tornado symbols, it will show you death and injury statistics. The table feature will summarized outbreak statistics for you. There’s also some vintage footage from the Worcester tornado out on YouTube, if you’re into that kind of thing.


  • "There Was Nothing Normal About America’s Freakish Winter Weather,” Brian K.Sullivan, Aprile 10, 2017, Bloomberg,
  • “25 Deadliest U.S. Tornadoes,” NOAA,
  • “84 Minutes, 94 Lines: the 1953 Worcester Tornado,” New Worcester Spy,
  • “June 7-9, 1953 – The Flint-Worcester Outbreak,” Stormstalker Blog,
  • “History of Tornado Forecasting,” NOAA,
  • “U.S. Tornado Climatology,” NCEI/NOAA,


Thursday 14 March 2013

The Keystone XL Pipeline: Do Journalists Read What They Write About?

(Correction added 3/20/13)

I missed an editorial a few days ago but managed to catch up on my reading this morning, whereupon I spotted a new piece on the Keystone XL Pipeline. Just in case you don't know the details, the proposed pipeline will pump liquefied oil sands bitumen across the Alberta-Montana border and then head southeast across the Midwest toward Steele City, Nebraska. The article in question is "When to Say No," an New York Times editorial dated March 11, 2013. The subject of the editorial is the Environmental Impact Statement ("EIS") issued a little over a week ago by the US State Department.

The following paragraph caught my eye:

To its credit, the State Department acknowledges that extracting, refining and burning the oil from the tar-laden sands is a dirtier process than it had previously stated, yielding annual greenhouse gas emissions roughly 17 percent higher than the average crude oil used in the United States. But its dry language understates the environmental damage involved: the destruction of the forests that lie atop the sands and are themselves an important storehouse for carbon, and the streams that flow through them. And by focusing on the annual figure, it fails to consider the cumulative year-after-year effect of steadily increasing production from a deposit that is estimated to hold 170 billion barrels of oil that can be recovered with today’s technology and may hold 10 times that amount altogether.

There are several things wrong with this paragraph but I'll start with the one that I knew was incorrect even without researching. What's wrong is the statement that oil sand petroleum has an "annual greenhouse gas emissions roughly 17 percent higher than the average crude oil used in the United States." That 17% rang a bell since it parallels a similar 17% mistake made in a Wall Street Journal blog from 2009 see my blog post from March 6 for m.... Seventeen percent also echoes numbers from various published sources discussing Canadian oil sand greenhouse gases ("GHG") including the State Dept. EIS itself:

WCSB (Western Canadian Sedimentary Basin) crudes are more GHG-intensive than the other heavy crudes they would replace or displace in U.S. refineries, and emit an estimated 17 percent more GHGs on a life-cycle basis than the average barrel of crude oil refined in the United States in 2005.

The important words here are "life-cycle basis." Modeling GHG emissions for any distinct source is usually done over the lifetime of all of the fossil fuel at that source in one lump regardless of how long it takes to deplete all the fuel there (e.g. Skone and Gerdes, 2009). The words "life-cycle" always refer to estimating all the GHG released "from cradle to grave," i.e. from untouched resource in the ground to final product for the entire life of a resource (Lattanzio, 2012). It is not an average with respect to a unit of time. A well-to-wheel ("WTW") GHG life-cycle estimate accounts not only for the GHG contribution from burning the fuel, but also accounts for the energy sources used to mine or extract the resource, pump it in pipelines or ship it by sea or rail, process it at a refinery and deliver it to the end-user customer. Using the life-cycle approach neatly dodges the problem of variable rates of production which every oil and gas field experiences due to the supply and demand conditions in the global marketplace for fuels.

The New York Times editorial made a mistake in calling the 17% GHG increase an annual rate of emissions. Whoever wrote it didn't understand the concept of life-cycle estimates for GHG. This becomes even more evident when the text goes on to say about the EIS:

by focusing on the annual figure, it fails to consider the cumulative year-after-year effect of steadily increasing production from a deposit that is estimated to hold 170 billion barrels of oil

The author of this editorial doesn't seem to know that the GRG life-cycle estimate of 17% was indeed based on the total extractable oil sand reserves for all four of the big Canadian oil sand deposits regardless of production rates or time. This leaves me wondering if the editorial author bothered to read the EIS at all since several EIS sections and one appendix were devoted to the discussion and comparison of well-to-wheel life-cycle GHG emissions for oil sand petroleum pumped through the proposed pipeline.

Correction and commentary added 3/20/13: I made a mistake which is why the paragraph above is struck out. I confused GHG footprint as discussed in the State Dept. EIS with life-cycle GHG estimates. But the NY Times quip about using an annual GHG figure vs. a total GHG footprint for total lifetime of the Alberta oil sands is still irrelevant. Why? Because the GHG life-cycle estimates are not annual figures. They are normalized, usually on either a per unit energy release or unit volume basis. Time is not a parameter when reporting GHG life-cycle estimates. One can convert GHG well-to-wheel or well-to-tank life-cycle estimates to annual emissions easily. It's simple if one starts with a life-cycle number reported on a per volume basis and then multiplies by the total amount of the fuel in question used annually. Alternatively, one could also estimate the total GHG footprint for a fossil fuel resource or any given industrial complex and then divide by the number of years that fuel source or industrial complex is expected to last, thus deriving an annual rate. Either way is a legitimate means to calculate an annual GHG emissions estimate.

The State Dept. EIS calculated the annual GHG emissions that would be introduced by the Keystone XL pipeline based on the assumption that the pipeline would be operating at capacity (830,000/barrels per day). Despite the fact that Canadian oil sands production is expected to double between 2012 and 2015, the capacity of the pipeline is a constant. Since the editorial is arguing against the pipeline, the lifetime of the Canadian oil sands is really moot. A recent Congressional Research Service study (Lattanzio, 2012) put this in a perspective that I believe most environmental activists would rather avoid. The numbers speak for themselves as to whether GHG-based arguments against the pipeline have any real substance. I will leave it to the reader to decide if the less than 1 percent annual GHG contribution of the Keystone XL pipeline to the annual US GHG emissions is meaningful:

The 2013 Draft SEIS analysis found that the potential range of incremental GHG emissions contributed by the pipeline would be 3.7 to 20.7 MMTCO2e annually. As the United States reported a total domestic GHG inventory of 6,865.5 MMTCO2e in 2010, the incremental pipeline emissions would represent an increase of 0.06%-0.3% in total annual GHG emissions for the United States.

The comment in the editorial about the destruction of forests appears to have been made in ignorance of Alberta's aggressive surface-mine reclamation laws. Many of the mines that were in production in the 60s and 70s are already on their way back to being forest. Also, only about 20% of the oil sands are mineable (Humphries, 2008); the remaining 80% can only be extracted through the use of down-hole in situ methods to convert the bitumen into a pumpible liquid. Since well fields are much less invasive than surface mines, preservation of Alberta's boreal forest will improve over time as oil sand production shifts from mining to pumping. There is a downside to oil sand mining and its subsequent reclamation. No reclamation method is capable of restoring the extensive peatlands that the surface mines must destroy to get at the oil sands

I find the whole Keystone XL Pipeline issue somewhat ridiculous, to be frank. The rest of the New York Times editorial makes further protests over the pipeline on the basis of the impacts it will have on groundwater aquifers and soils in the event of a spill. You would think from reading this editorial that no one had ever built a pipeline before or that there had never been a pipeline leak. The reality is that there are numerous large capacity pipelines already in existence - including two that are already pumping liquified oil sands bitumen from Alberta into the US - and their environmental impact to date is quite low. I've worked on pipeline leaks. They're bread-and-butter remediation projects. The thought that a spill might reach one of the big Midwest production aquifers is a bit silly since hundreds of feet and the presence of aquicludes like the Pierre Shale separate them. The Keystone XL pipeline proposal exists because the two pipeline from Canada currently in use are at full capacity and have been for several years now.

I'm really tempted to wonder out loud how people like the author of this editorial get paid to write nonsense like this instead of me? But since I gave up temptation for Lent, I will refrain from doing so.

Here's a NASA pic of the Athabasca oil sands mine in Northeast Alberta. Not exactly a garden spot.



Gerdes, K., and Skone, T. J. (2009), Consideration of Crude Oil Source in Evaluating Transportation Fuel GHG Emissions, National Energy Technology Laboratory (Report) DOE/NETL-2009/1346, US Dept. of Energy, ( ; accessed 3/6/13).

Humphries, M (2008), North American Oil Sands: History of Development, Prospects for the Future, Congressional Research Service No. RL34258. (, accessed 3/14/13).

Lattanzio, R. (2012), Canadian Oil Sands: Life-Cycle Assessments of Greenhouse Gas Emissions, Congressional Research Service No. R42537 ., accessed 3/14/13).

Levi, M. A. (2009), The Canadian Oil Sands Energy Security vs. Climate Change, Council Special Report No. 47, Council on Foreign Relations, New York, ( ; accessed 3/6/13).