Tuesday, April 14, 2015

Titanic coal

            “A steady roar thundered across the water as everything moveable broke loose. There had never been a mixture like it-–29 boilers . . .the jeweled copy of the Rubaiyat . . . 800 cases of shelled walnuts . . . 15,000 bottles of ale and stout . . . huge anchor chains (each link weighed 175 pounds) . . . 30 cases of golf clubs and tennis rackets for Spalding . . . Eleanor Widener’s trousseau . . . tons of coal . . .”

            Walter Lord, A Night to Remember (1955)
Still from movie Titanic (1997)
        This April is the 103rd anniversary of the sinking of the Titanic. In 2007, I visited "Titanic: the Artifact Exhibition" at the Royal British Columbia Museum in Victoria, British Columbia, Canada, just before the start of that year’s combined meeting of The Society for Organic Petrology (TSOP) and the International Committee on Coal and Organic Petrology (ICCP). The museum is in the first block south of the majestic Empress Hotel on the inner harbor of Victoria. (I followed up my early Friday evening time-ticketed exhibit visit with supper and cocktail on the veranda of the Empress.) The Titanic, of course, is the great British steamship, touted as "unsinkable", that, on its maiden voyage from England to New York, struck an iceberg in the North Atlantic late on Sunday night, April 14, 1912; it sank beneath the ocean surface a few hours later in the early morning of April 15. The traveling exhibit, both through passenger and steamship effects recovered from the seafloor, and through recreations of the ship's interior, told the story of the disaster, life on board, the variety of passengers and their reasons for sailing. The exhibit exited at a gift shop; I purchased a pendant that enclosed a small piece of Titanic coal from the seafloor. I thought that was an appropriate remembrance for me, being 1) a coal petrologist, and 2) born on April 15.

        I had been aware that some of the coal from Titanic had been recovered and sold as part of the fundraising for Titanic recovery and research. Coal lumps, in fact, are the only recovered artifacts that have been ruled legal for sale since they are considered "natural" objects and not man-made (http://law.lclark.edu/live/files/11855-lcb163art8zekalapdf). Although some online photos or details on certificates of authenticity (COA) indicate that coal lumps could be very large (=>3 kg), one photo taken in a Titanic coal bunker before sailing shows mostly fist-size coal. Smaller pieces sold, including those in jewelry or crushed coal mounts, are taken from the larger pieces, and the COA for those specimens are, therefore, derivative. For example, the "object number" for the coal in my pendant is 94/0036: 1994 being the year of collection and 36 the sample number. A photo at http://www.titanicautographs.com/products.php?product=Titanic-coal-recovered-from-the-ship-on-offical-expedition shows the original 94/0036 coal nodule recovered (my estimate of size: 30 cm x 25 cm x 10 cm), plus the COA with that sample number for a small coal fragment. However, a Google image search for just "Titanic coal" produces several offered coal fragments, with COAs of different styles, but the same 94/0036 number! Certainly, many small pieces could be derived from the large original nodule, but is the volume sum of marketed pieces greater than the original whole? The seller at the above link does mention concern with authenticity of the coal being offered. (Addendum, June 11, 2015: Three days ago, I visited the permanent Titanic exhibit at the Luxor casino in Las Vegas, Nevada, USA. The one large lump of coal on display had the sample number, 94/0036.075: same as the numbers mentioned above, but with 3 more digits after a decimal point. My guess is that 94/0036 is a batch number for all coal retrieved in the 1994 expedition, and that ".075" is a nodule/lump number that is not included on samples offered for sale.)

        Titanic could hold 6611 tons of coal in bunkers and an additional 1092 tons in Hold 3 (Hutchings and de Kerbrech, 2011). There are variable reports on how much coal was onboard at sailing. Sheehan and Sickels-Taves (2002) state just under 6000 tons; website http://atlanticliners.com/titanic_home.htm says 5892 tons. Essenhigh (2004) writes that the six bunkers were only half full with 800 tons each (=4800 tons). Smith (2005) cites a website (no longer active) that reported 4427 tons were in the bunkers; Palmer and others (2003) said "more than 4000 metric tons". Steam for the two steam engines and one steam turbine was produced in 29 boilers that contained 159 furnaces. Six hundred tons of coal/day were shoveled into the furnaces around the clock by a total complement of 176 firemen ((Hutchings and de Kerbrech, 2011). There were 73 coal trimmers who handled the coal from loading to maintenance in the bunkers and delivery to the firemen (http://en.wikipedia.org/wiki/Coal_trimmer). One hundred tons of coal ash were disposed at sea each day (Hutchings and de Kerbrech, 2011).  Sheehan and Sickels-Taves (2002) state that the Titanic sank with 2500 tons of coal out of the original load. Coal has been recovered from Titanic ocean bottom debris fields since 1987; coal in various studies and available for sale online is from expeditions in 1994 and 2000.

        Based on testimony of surviving crewmen, there is good evidence for a coal fire in one of the bunkers at the time of sailing. Bunker fires were not uncommon on steamships, caused by spontaneous combustion within the piled coal. The usual solution to quell the fire was to sail at full speed to quickly shovel down the coal pile until the smoldering coal could be removed by simply adding it to the furnaces. However, Titanic survivor reports are not consistent regarding 1) whether the fire started around the time of sailing from Southampton or earlier during sea trials in early April; 2) if the fire was out by Saturday, April 13, or still smoldering at collision; 3) if heat from the fire damaged the adjacent watertight bulkhead; and 4) in what bunker was the fire. Essenhigh (2004), calculating rate of fire spread versus coal pile drawdown, assumed the fire was probably in the top half of the bunker pile, however, Fireman John Dilley testified in 1912 that the fire was at the bottom of the coal pile. There are some, including Essenhigh (2004), who speculate that running full speed through the ice field on April 14 was not to break an Atlantic-crossing speed record but to get rid of the burning bunkered coal, assuming that the fire had not been extinguished the day before. The website http://titanic-model.com/db/db-03/CoalBunkerFire.htm discusses inconsistencies in quoted original testimony on the bunker fire from surviving crewmembers.

        I was curious if any analysis of recovered Titanic coal had been undertaken. A 2012 discussion in an online Titanic forum mentions that an analysis of coal recovered in 1994 concluded there were 5 or more geographic coal sources including Pennsylvania anthracite. The "chat" also says that the technical analysis is no longer online (perhaps it is the same as the now-defunct website cited by Smith (2005)). I also could not find any such study that matches those results. However, there are available published scientific studies, using modest-size sample sets, by two groups of researchers. The earliest is by Michael Sheehan and Lauren Sickels-Taves (in 2002 at Eastern Michigan University) from their presentation at a symposium on materials issues in archaeology. The second group of authors (Palmer et al., 2003) presented their findings at both the 2002 Pittsburgh Coal conference and the 2003 TSOP annual meeting in Washington, DC.

        Sheehan and Sickels-Taves, importantly, detail the historic background of coal labor and supply issues in the United Kingdom in spring 1912 that affected the availability and sources of coal loaded on Titanic. From February 22-April 6, 1912, ending just before Titanic's sailing from Southampton, England, on April 10, there was a major UK coal strike that limited the national supply of coal. Some ship sailings were cancelled due to the strike. However, in preparation for the celebrated maiden voyage of Titanic, White Star Line, owner of the ship, scavenged leftover coal in Southampton from other ships of the International Mercantile Marine trust, of which White Star was a member, or had ships already at sea take on extra coal in non-British ports. Sheehan and Sickels-Taves mention that extra coal sacks were even stockpiled in the Second Class public rooms of Titanic's sister ship, Olympic. It is important to note that coal had been already loaded on Titanic at Belfast for the sea trials (Smith, 2005); those coals, I assume, are British since there are few Irish coals and Belfast is close to major British coalfields across the Irish Sea. 

        Sheehan and Sickels-Taves examined 19 coal nodules that were recovered in 2000. They used samples made available to them. They state there was no scientific sampling plan in seafloor recovery, and, therefore, samples may not be representative of the actual range and distribution of coal types onboard. Fifty grams taken from each nodule were used for petrographic, palynological, and trace element analysis (latter data not reported). Testing was done by TES Bretby (formerly Scientific Services Divison of British Coal), UK, and Virginia Polytechnic Institute (VPI), USA.  

        The range of mean vitrinite reflectance for the 19 samples is 0.99% to 2.28% ((British Standard 6127=ISO 7404-5). (See my first blog post in October 2014 for description of 'vitrinite reflectance'.) Fifteen samples are "medium or low volatile bituminous coals with reflectance values between 1.19-1.99%"; eight have reflectances between 1.65 and 1.76%. Only one sample has reflectance greater than 2%, the 2.28%Ro semi-anthracite sample. This range of maturity is consistent with the rank variation within the South Wales Coalfield, which it seems the authors assume is the primary British source of Titanic coal. One other reflectance data point is available for Titanic coal: a medium volatile bituminous 1.15%Ro from the only sample obtained by Smith (2005).

        Three high volatile bituminous coals (0.99-1.09%Ro) of Sheehan and Sickels-Taves were submitted for palynological analysis. Results indicate all are Upper Carboniferous, but stratigraphic location within the British Coal Measures could not be determined. One sample, 7B (0.99%Ro) had a type of spore rare in British coals. The authors also say the rare spore type, not identified, is not common in the eastern USA, but speculate it is still possible the sample could be from an eastern US coal transported to Southampton by the Olympic.

        Palmer and others (2003) examined 20 samples chosen by the RMS Titanic, Inc. curator because "each piece appeared to be different". Their fixed carbon and volatile matter results indicated an equal distribution of low, medium, and high volatile bituminous coals among the samples. Ash yields are more consistent with British coals than US coals. Trace elements were analyzed by neutron activation, and results were compared to a data base of 24 British coals and 1450 US coals from beds that had been exploited in 1912. Rare earth element data "suggest. . . five distinct sources". Iron, potassium, and arsenic concentration range and median values are more similar to British coals. For most elements, however, the Titanic values were less than both US and UK coals, which the authors suggest is due to differences in the Titanic coal, mined in 1912, versus data base samples collected decades later from the same mine or seam.

        Two of Palmer and others’ samples yielded spores for palynological study. The authors list identified species of the spore assemblages which indicate a Carboniferous Langsettian (Westphalian A) age. They report that nearly all British coalfields contain such beds, but few US coals mined in 1912 were that age.

        One research goal of Palmer and others was to determine the environmental effect of shipwrecked coal in a deep marine setting. In the Titanic samples, higher iodine and bromine than either US or UK coals suggests absorption of those elements from seawater; chlorine data indicates some leaching of that element from the coal. Otherwise the coals are unaltered, and there appears to be “minimal environmental impact” since 1912.

        Both scientific studies conclude most coal samples examined were sourced in Britain. Sheehan and Sickels-Taves write that the limited Titanic sample set shows "considerable uniformity" consistent with usual British sources of coal used by White Star. Although it is known that ships like the Olympic onloaded extra coal from non-British ports during the 1912 strike, it does not appear, with the possible exception of their low-rank sample 7B, that non-typical coals were a significant portion of Titanic fuel. Palmer and others, using two different statistical methods, concluded that 12 samples are probably from the UK, three from US, but provenance of the other four cannot be clearly assigned.

        With results indicating that some of Titanic’s coal may be from the United States, what were the typical US sources of coal for White Star Line ships? The Coal and Coal Trade Journal (vol. 22) wrote in May 23, 1883, that White Star Line renewed a contract with New Central Coal Company of Maryland to fuel their ships in American ports. That company mined Carboniferous low volatile bituminous coal (http://pubs.usgs.gov/sir/2010/5152/pdf/sir2010-5152_fig2.pdf) in the western Maryland Georges Creek basin. I could not find if a contract continued to 1912. Ten years later, the Colliery Engineer (volume 14, August 1893) mentioned a White Star US contract alluding to a Pocahontas coalfield source (West Virginia/Virginia; medium to low volatile bituminous). Gas World (v. 20, March 31, 1894) also mentioned that White Star Line used Pocahontas coal. Gas World continued that Pocahontas was the preferred coal of the US Navy, and some steamships had set Atlantic crossing records using the same. They wrote that Pocahontas coal "is declared by a Newcastle analyst to be 'equal to the best Welsh steam coal, and excellent coal for steam-raising purposes.'" This still leaves questions, however, about the source of high volatile bituminous coals identified by both Sheehan and Sickels-Taves (their possible-US sample 7B) and Palmer and others.

        The results of the scientific studies do call into question colliery source data or coal rank given on some of the Titanic coal COA and labels currently found online. As mentioned earlier, some labels state the rank is anthracite; one COA, without rank assignment, claims that the mine of origin is Six Bells Mine, South Wales Coal Fields, UK (closed since 1988). The study by Sheehan and Sickels-Taves clearly indicates that the exact mine of origin for a particular lump of coal cannot be ascertained with certainty even if the rank is determined. An included table by TES Bretby shows possible early 20th-century South Wales colliery associations for sample reflectance clusters. Most reflectances are correlated with more than one mine, although only the Lake Windsor colliery (Ynysybwl, Wales) is listed for the rank range of the eight 1.65-1.76%Ro samples. In addition, none of the studied coal, recovered in 2000, is, in fact, anthracite, but the limited number of samples does not rule out anthracite onboard. 

        "Titanic: the Artifact Exhibition" at Victoria was one of the itinerant exhibitions of RMS Titanic, Inc.; they have a permanent artifact exhibit now at the Luxor (pyramid-shaped casino) in Las Vegas. Another excellent exhibit of Titanic effects, mostly flotsam, is the permanent one at the Maritime Museum of the Atlantic in Halifax, Nova Scotia (https://maritimemuseum.novascotia.ca/what-see-do/titanic-unsinkable-ship-and-halifax). The flotsam was picked up at sea by both rescue-and-recovery and commercial boats soon after the sinking. Copies of some items were incorporated into James Cameron's 1997 film, Titanic: a stairway newel post ornament like the one shot off the grand staircase, and an entryway entablature panel similar to what Kate Winslet floats on after the ship goes down (the real one is about half the size of the movie version). One hundred fifty victims of the Titanic sinking are buried in Halifax.

     Essenhigh, R.H., 2004, What sank the Titanic? The possible contribution of the bunker fire (abs.): Geological Society of America Abstracts with Program, Vol. 36, No. 5, p. 42.

     Hutchings, D. and de Kerbrech, R., 2011, RMS Titanic Manual: 1909-1912 (Olympic Class): Haynes Owners’ Workshop Manual Series, Zenith Press, 160 pages.

     Palmer, C. A., Finkelman, R. B., Luttrell, G. H., Zhang, C., and Eble, C., 2003, The source of the coal in the Titanic and effects of exposure to seawater: Program and Abstracts for the 20th Annual Meeting of The Society for Organic Petrology, v. 20, p. 54-58.

     Sheehan, M.S., and Sickels-Taves, L.B, 2002, The technological analysis of RMS Titanic’s Coal: The enhancement of archaeological research: Material Research Society Symposium Proceedings, vol. 712, p. 525-532 (Materials Issues in Art and Archaeology VI: Symposium held November 26-30, 2001, Boston, Massachusetts, USA).

     Smith, A.H.V., 2005, Coal microscopy in the service of archeology: International Journal of Coal Geology, v. 62, p. 49-59. (http://his.library.nenu.edu.cn/upload/soft/haoli/112/122.pdf)

(This blog post, with very minor changes, appeared earlier as an article in the March 2015 edition of the TSOP newsletter (http://tsop.org/newsl/nl321pr.pdf))

Thursday, April 9, 2015

Grace Hopper, software pioneer

In early 1974, as I was entering the building of my first apartment, River House in Arlington, Virginia, by the Pentagon, I saw across the lobby the elevator closing on what seemed like an incongruity: a petite older gray-haired lady in a blue Navy captain's uniform with the four wide gold stripes on the cuff. Unthreatening sweetness vs. authority.

River House, S. Joyce Street, Arlington, Virginia, and Grace Hopper Park (2011)
 Several weeks later, I, a US Department of State employee at the time, took a two-week Introduction to Computing course at the Department of Defense Computer Institute (DODCI) at the Navy Yard in neighboring Washington, DC. Speaking the first afternoon was the elevator woman, Navy Captain Grace Hopper (December 6, 1906- January 1, 1992), a pioneer in computing software. Memorable that afternoon was her brilliant (and well-known) analogy of a "nanosecond" and the speed of computing and electronic communication: an 11.8 inch (29.97 cm) length of thin colored telephone wire representing the maximum distance electrons travel in one-billionth of a second at the speed of light (in a vacuum). Captain, and eventually Rear Admiral, Hopper always carried with her a bundle of "nanoseconds", and we were each gifted with our own. A short video of her discussing this in a class, not unlike one I attended and of the same vintage, can be found at https://www.youtube.com/watch?v=JEpsKnWZrJ8; she explains nanoseconds to David Letterman at https://www.youtube.com/watch?v=1-vcErOPofQ.

The next day, on my way to catch a bus to the Navy Yard, I ran into Hopper in the lobby of our apartment building. I introduced myself, telling her I was in the computer class. As her Navy driver pulled up, she generously offered me a ride in her government car. I wish I remembered the details of our conversation! I only lived in the building a month or two more, and our subsequent interactions were just passing greetings.

Hopper received her undergraduate degree in mathematics from Vassar College, one of the pre-eminent women's colleges of the 20th century before admitting men in 1969. (I went to Smith College, another "Seven Sisters" school and still all women; maybe we talked about women's higher education in her limo?) After getting a PhD in math from Yale, Hopper taught at Vassar until joining the Navy during World War II.  She was a programmer on the Navy's Mark I electro-mechanical computer at Harvard; after the war, she worked on UNIVAC (the second general-purpose electronic computer according to Wikipedia) at Remington Rand. In 1967, she returned again to active duty with the Navy (https://www.history.navy.mil/research/library/research-guides/modern-biographical-files-ndl/modern-bios-h/hopper-grace-murray-text.html).

Hopper's life and achievements were chronicled this January (2015) in the short (16 minutes) ESPN FiveEightThirty documentary, The Queen of Code (http://fivethirtyeight.com/features/the-queen-of-code/). During Women's History Month (http://www.nwhp.org/) in March 2015, National Public Radio interviewed the documentary's director (http://www.npr.org/blogs/alltechconsidered/2015/03/07/390247203/grace-hopper-the-queen-of-code-would-have-hated-that-title). Also during WHM, the National Science Foundation highlighted several pioneering women in STEM careers (science, technology, engineering and math), connecting these "names you should know" with current female scientists supported by NSF (http://www.nsf.gov/discoveries/disc_summ.jsp?cntn_id=134386). They included seven women who worked on the ENIAC computer (the first general-purpose electronic computer) during World War II. While Hopper's Mark I was Navy-funded housed at Harvard, ENIAC was supported by the US Army and situated at the University of Pennsylvania. Both computers were used in conventional ballistic and atomic bomb research calculations. The women working on ENIAC were not named in official documents or photographs at the time; they have since been inducted into the Women in Technology Hall of Fame. If you have seen the movie, Imitation Game, about British WWII cryptography, you will be aware of the difficulty and prejudice that talented women in mathematics faced for a long time in being accepted seriously and having access to the same job opportunities as men.

Hopper's best-known achievement is the programming language, COBOL (COmmon Business Oriented Language). It was designed to be a language that could be shared among different computers and be more like regular English, therefore more accessible to non-scientists. I personally see it as the beginning of making computing more user-friendly. I never had to use COBOL. In a previous blog, I mentioned that my short experience with FORTRAN, in 1969, was one factor influencing my decision to be an undergraduate history major. (The DODCI course in 1974 did not teach coding, but covered systems analysis, applications, flow charts.) My serious foray into coding was during my geology Master's degree at Dartmouth College, the birthplace of BASIC programming language. Without the cumbersome formatting statements of FORTRAN and, like COBOL, more like regular English, BASIC was much more transparent to me and easier to use. I learned to enjoy "debugging" programs (Hopper is credited with originating the computing terms "bug" and "debugging"). Dartmouth required geology grad students to be able to write a moderately-complex computer program and that was easily solved by taking either Aqueous Geochemistry or Clay Mineralogy whose final semester projects were programs. Subsequently, my first introduction to coal petrology was programming, in BASIC, the collection of liptinite epifluorescence spectral data through an analog/digital converter that connected a microscope to an Apple 2 computer (this was 1982 at Southern Illinois University; lipid-rich liptinites do glow in UV and blue light). On desktop computers at that time, one had to be very thrifty with program size. I was limited to 512K (that's KILObytes) for the program steps and data collected within the program. And now I complain that my iPhone does not have enough GIGAbytes for my photos!

The Apple 2 did not have the "desktop" display and file folder management system we know today, just a line entry typing screen. Word processing at Dartmouth in the early 1980's was through commands in BASIC on time-sharing terminals. We got our first home Macintosh, with MacWrite, in about 1986. The current 21st-century user interface on a computer, smart phone, computer game, and in word processing, graphics, spreadsheet and other applications is layers above the programming language that creates what we see on the screen. This makes digital technology accessible to all; Grace Hopper's programming vision was an essential early step.

Plaque, Grace Murray Hopper Park, So. Joyce Street, Arlington, VA (2011)
 Addedum (1/12/16): Admiral Hopper mentioned in the State of the Union address. Additional biographical information: https://www.whitehouse.gov/blog/2014/12/22/honoring-grace-hopper