Ex-PDVSAs of Caracas Chronicles, Help Me Out

Assuming a 400-meter blast radius, what else got damaged?

At this point, we still don’t have a clear picture of how much damage Amuay Refinery suffered overall from last night’s blast. We still don’t know precisely which facilities were damaged, and how extensively. It takes some insider knowledge of the refinery to piece this together, so I’m calling on former PDVSA workers who know Amuay to write in to [email protected] and help us out.

There’s a lot of speculation right now about what the accident might mean for gasoline production at CRP, but the Complex is huge, and the most seriously affected area looks like it’s a relatively small part of the whole. So what facilities were closest to the blast site? What are we talking about, exactly?

The picture above, taken from a random guy at NoticieroDigital, offers some clues. The National Guard facility that was devastated by the blast was about 400 meters from the primary blast-site. So assuming a 400 meter blast radius, what else can we reasonably assume were damaged? Mostly storage tanks? Actual Cat-Crackers and other gasoline-making facilities? What, exactly?

(Though, in fact, the houses on Calle 1 in Judibana – about 1 km. northeast of the blast site -site, suffered extensive damage too. Perhaps assuming a 400 meter blast radius is too optimistic.)

Do write in.

Caracas Chronicles is 100% reader-supported. Support independent Venezuelan journalism by making a donation.


  1. Puramin plant got blasted. First hand account from a friend that works there. Also 60+ dead bodies at the hospital. She even saw some coworkers bodies.

  2. Francisco, maybe there was some wind going inlands, that could explain the biggest damage to the right of the image. It should had taken some time to get a big gas-air cloud and if did goes to this side, then the explosion will not be centered on the escape site. The metallic structures should be less damage prone than the masonry ones (round shaped and with lots of space in between by design also helps). Maybe they told the truth, maybe. If not, we will sure know in the next days…

    • VP Jaua: An investigation under way by a special task force formed by PDVSA’s PCP division, Intelligence (SEBIN), the Military and the Public Ministry.

  3. Es que el problema no es solo la explosion, sino la extension de la nube de gas.
    Se dice que desde hacia tres dias la gente reportaba el olor; que cantidad de producto escapo, como estaba el viento, era fuerte o debil, hacia que direccion soplaba…?
    Estas variables hacen dificil determinar, sin ver in situ, los daños.

  4. So, after all this, how are you going to get workers to enter any refinery, it being reasonable to assume that they’re all subject to implementation of the same maintenance philosophy?

  5. If the propane gas leak has several hours developing, as several people have claimed, it may have been several blasts, ignited by a vehicle driving by with an exposed ignition wire, a spark caused by a light switch being used, etc. This may explain the widespread damage.

    Remember that propane (if it was the involved gas) is heavier than air, and it may have been driven by wind and accumulated in low areas before been ignited by a random spark.

  6. Researching internet, I found that Mercaptan is used as an odorizer in residential distribution systems , so people can detect natural gas (which has no smell) if there is a leak, but I do not know if this is a standard procedure at a refinery such as the CRP complex.
    The reason why Mercaptan is used with natural gas is due to a terrible accident that ocurred in Texas in 1937: http://en.wikipedia.org/wiki/New_London_School_explosion

  7. Guys and gals, I need help. Trying to write a piece, but I have no luck backing a claim from BBC:

    “Puede resultar un tanto compleja la comparación con las industrias petroleras en otros países. Pero, por ejemplo, en los últimos años en Venezuela murieron 78 personas en accidentes, cifra levemente inferior a la de Estados Unidos en el mismo período, con la diferencia de que Pdvsa emplea a 100.000 personas y EE.UU. a 2,2 millones.”

    The journalist, Juan Paullier, wrote that. I first had assumed it was Jose Toro Hardy. In any case, I cannot find any source confirming those claims. I found an article from the CDC that puts the fatality rate for the US in 30.5 dead per 100000 workers. http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5716a3.htm Which is close to the figures that NYT has also: http://www.nytimes.com/interactive/2012/05/03/us/death-on-the-job.html?ref=us : Between 03 and 06 there were 404 deaths. If we assume a similar distribution of rates, there is no way that the statement from BBC can be true. “cifra levemente inferior”, no way. Also, the number of oil workers seem to be around 1.3 MM.

    If we cross Urru’s database (www.urru.org/papers/2011_varios/CRONOLOGIA_accidentes_pdvsa_2003_a_2011-02-14.pdf) of accidents with the numbers of workers in PDVSA (http://www.pdvsa.com/index.php?tpl=interface.sp/design/readmenuprinc.tpl.html&newsid_temas=26) we actually come with a _lower_ fatality rate for Venezuela.

    I am trying to write a solid piece about this, with solid data, but that thing from BBC is not standing up to scrutiny. Numbers might be more complicated, I might be looking at all workers instead of the subset of workers in plants, and they count all the bomberos in gas pumps. I do not know, so I ask fro help.

    In any case there was/is a maintenance problem, and there’s responsibility from a govt that claims a victory in fencing is due to them, but a tragedy like this is not their fault.

  8. Last but not least, the Puramin plant which was damaged in the blast made dielectric oil used in electric transformers and also oil used in motor vehicles:


    Lets hope that there are others sources for the dilectric oil, because transformers need this oil to be replaced periodically (it gets contaminated with water, main reason for “exploding transformers”).

  9. Guido, I hope this will be of some assistance.

    Census of Fatal Occupational Injuries
    Original Data Value

    Series Id: FIU00X32411X80N00
    Area: All U.S.
    Case Type: Fatalities in all sectors
    Category: Ownership– All
    Industry: Petroleum Refineries
    Years: 2006 to 2010

    Year Annual
    2006 4
    2007 5
    2008 6
    2009 5
    2010 7

    This does not include 2005, when 15 died at a BP refinery in Texas City. Yes, the same BP that had the Macondo blowout.[Thanks, Gabo]. Pardon the phrase, but it is no accident that BP was involved in both.


    • Yup, that was it. Then it becomes much more complex to compare PDVSA with the US oil industry, but the fatality rate in refining seems to be way lower than in drilling.

      Thank you!

  10. Guido, as far as I can tell, the CDC article you linked to refers to drilling personnel, not to refining personnel. I will look for data for number of refinery employees.

  11. Series Id: ENUUS000105324110
    State: U.S. TOTAL
    Area: U.S. TOTAL
    Industry: NAICS 324110 Petroleum refineries
    Owner: Private
    Size: All establishment sizes
    Type: All Employees

    Year Annual
    2001 74977
    2002 73911
    2003 71437
    2004 69162
    2005 68427
    2006 69124
    2007 72337
    2008 75099
    2009 75588
    2010 72689
    2011 71278 [tentative]

    Guido: “NAICS 324110” is the BLS code for petroleum refineries.


  12. Guido, one way to compare might be to look at refinery production/capacity in the US versus Venezuela, and come up with comparable figures for deaths/ million BBL s refined/year. We have data for refinery deaths in the US from 2006-2010. Of the 78 PDVSA deaths from 2003-2011 that the BBC articles cites, I wonder if one can find an itemized list to find other refinery deaths among the 78- which does not include the 41(?) killed this weekend.

    One problem is that while we have production and capacity figures for the US, I have found only capacity figures for Venezuela. A further question: do we include the heavy oil, which is nonstandard petroleum. May as well include them- won’t make a lot of difference.

    http://www.eia.gov/dnav/pet/pet_pnp_wiup_dcu_nus_w.htm US Refinery production. Around 15 million BBL/day refined, with about 17 million BBL/day capcity.

    http://www.eia.gov/cabs/Venezuela/Oil.html Venezuela 1.28 million BBL/day capacity. While it doesn’t include heavy oil in this figure, it has heavy oil refining capacity further down.

    http://en.wikipedia.org/wiki/List_of_oil_refineries#Venezuela See the heavy oil

  13. Guido:
    I went and published the article without that bit of data.

    Given the difficulty in sorting out the data cited in the BBC article you quoted, you made a good decision. Get out what you know: the meta-points behind the explosion. Your article did a good job of summing up the meta-points behind the explosion. The 5 points are also a good summary of how PDVSA under Chavista control has operated the last decade. The small picture reflects the big picture.

    The BBC article is difficult to sort, especially since there are no links to specify what they are citing. I found this most telling:

    Eddie Ramírez, quien fue directivo de Pdvsa entre 1998 y 2002 y ahora es coordinador de la organización Gente del Petróleo, asegura que en su época el Índice de Frecuencia Bruta (IFB) –que mide el número de lesiones de trabajo en un millón de horas/hombre– se ubicaba entre 0,5 y 0,8. El año pasado el IFB fue de 9,40.

    Mientras tanto, el Índice de Severidad (días perdidos por cada millón de horas/hombre), alcanzó en 2011 los 530,92 puntos, un aumento de 37% con respecto a 2010.

    But then all that would need to be cross-checked. Not easy. Which is why I never tried to earn any money as a journalist.


    • I sent him a message asking for clarification, I also sent an info request to Dr. José Toro. No luck yet, but they are busy.

      Glad that you liked the article, Boludo.

  14. Last try, two posts were halted as spam? So, you have to search for El País: “Una de las válvulas de la esfera número 206 se había dañado tres días antes. “La mandaron a reparar, pero la fuga continuó”, dice un trabajador de Pdvsa que por temor a ser despedido mantiene su nombre en reserva. “El viernes por la tarde llovió y la humedad contribuyó a que el gas se acumulara en el patio de las esferas””
    R. Ramírez said something very, very, similar, that the weather conditions allowed this to happen (as if was something already going on, but not fatal until the weather changed…) Also they said one “esfera” was empty at the time, how comes that?

  15. Thanks Moses,
    at first you would think, how can this be, gas isn’t invisible? Well, the gas was pressurized, as it leaks it expands and cools the air, water vapor normally present then condenses as in a cloud (the same effect as with dry ice). So actually you don’t see the gas, but his cooling effect on a humid atmosphere.
    I hope for a legal process, but now the families had received a new house…


Please enter your comment!
Please enter your name here