UPDATED: Cantarell and Questions Regarding Mexico's Oil Infrastructure
Posted by Khebab on August 21, 2007 - 8:00am
Topic: Supply/Production
Tags: cantarell, google earth, hurricane dean, KMZ, mexico, oil, oil prices, oil rig, peak oil, pipelines, refining, rig, shipping [list all tags]
Scroll down for the 5:00p and 11:50p EDT updates.
Hurricane Dean became a Category 5 storm last night with winds reaching 165 mph and reaching a low pressure of 909 mb (as of 2:15a EDT; Katrina was 920 mb and Camille 909 mb). Landfall occurred early yesterday morning with a second landfall occurring some time today. This is an historic hurricane by any standard.
Why this matters: If there were Cat2 winds in that area, we could have been talking about around 2.5 million barrels per day of Mexico's supply capacity being shut in for a while, and some of that shut in for an extended amount of time. Around 1.5 mbpd of that capacity is exported to the US (of the 20.5 mbpd the US uses, and the 85mbpd the world uses, each day). There are also some questions about the resilience of refineries and flow lines in the area of the second landfall.
Update (Khebab, 11:50 EDT):
HURRICANE DEAN ADVISORY NUMBER 35A
...DEAN IS MOVING BETWEEN THE WEST AND WEST-NORTHWEST NEAR 20 MPH...32KM/HR...AND THIS GENERAL MOTION IS EXPECTED DURING THE NEXT 24 HOURS. ON THE FORECAST TRACK...DEAN IS EXPECTED TO BE VERY NEAR THE COAST OF CENTRAL MEXICO DURING THE DAY WEDNESDAY. MAXIMUM SUSTAINED WINDS REMAIN NEAR 80 MPH...130 KM/HR...WITH HIGHER GUSTS. DEAN IS A CATEGORY ONE HURRICANE ON THE SAFFIR-SIMPSON SCALE. AN AIR FORCE PLANE IS CURRENTLY APPROACHING DEAN. SOME RE-STRENGTHENING IS FORECAST DURING THE NEXT 24 HOURS.HURRICANE FORCE WINDS EXTEND OUTWARD UP TO 35 MILES...55 KM...FROM THE CENTER...AND TROPICAL STORM FORCE WINDS EXTEND OUTWARD UP TO 140 MILES...220 KM. ESTIMATED MINIMUM CENTRAL PRESSURE IS 970 MB...

Cantarell and KMZ oil complex, 50-knots wind speed probabilities (NHC, forecast #35). Click to Enlarge.

Refinery position, 50-knots wind speed probabilities (NHC, forecast #35). Click to Enlarge.
More under the fold.
UPDATE FROM CHUCK WATSON at 5p EDT, 8/21
Chuck Watson of KAC/UCF has just released his damage estimates. Here's what he has to say:

NHC has dropped their intensity more in line with our estimates. I'm sticking with the earlier damage forecast - Southwestern about 10 days, Northwestern (Cantarell etc) 15 to 18 days, unless something unlucky happens like a critical component failure. So our total shut in forecast from this event is around 30 million BBL, which was at the low end of my earlier model. The storm decayed more than the NHC estimates.
Under the fold (click "there's more" below), we are going to try to bring together some of our information we have gleaned to this point. We ask that you post all stories, insights, links, maps, and other tools in this comment thread. If you have stories or insights you wish to share, please also feel free to use the editors' box to contact us.

Cantarell and KMZ oil complex, forecast track and wind strength distribution (the purple area is at least tropical storm winds, blue is category 1 winds) from last night's run (21:00 UTC). Click to Enlarge.

A category 5 can make this kind of damage on landfall:
Hurricane Andrew (1992), category 5, a whole Miami neighborhood is
leveled..

Forecast track possibility along with "spaghetti models" from yesterday night run (21:00 UTC). Click to Enlarge.

Cantarell and KMZ oil complex and 50 Knot (58 mph) Wind Speed Probabilities - 120 Hours from yesterday night run (red to purple means probability > 70%). Click to Enlarge.

Wind distribution. Click to Enlarge.

QuickSCAT wind field. Click to Enlarge.
What is the area we're worrying about? Cantarell and KMZ, as said above, PEMEX is shutting down ~140 rigs and moving 13,000+ workers on land. Cantarell + Ku-Maloob-Zaap account for 66% of the total oil production of PEMEX in June 2007. Cantarell alone is 47% of the total....and here's the best pic we have (right click and view image to see full size):

***
The United States imported approximately 620,000,000 barrels from Mexico last year. (http://tonto.eia.doe.gov/dnav/pet/hist/mttimusmx1m.htm)
More on importers (Mexico is the 2nd ranked importer (at 1.5 MBPD) to the US behind Canada in 2007, then comes Saudi Arabia): http://www.eia.doe.gov/pub/oil_gas/petroleum/data_publications/company_l...
The IEA on Mexico: http://www.iea.org/Textbase/country/m_country.asp?COUNTRY_CODE=MX
Google maps of the Bay of Campeche: http://maps.google.com/maps?ie=UTF8&om=1&z=7&ll=19.849394,-94.163818&spn...
***
Khebab posted "An Update on Mexico's Oil Production--The Rapid Collapse of Cantarell by the Numbers"
Last year, I expressed my concerns about the eventual impact of a rapid collapse of Cantarell on Mexico's oil production (story here). The last production numbers from PEMEX seems to confirm the rapid decline of Cantarell as well as the inability of the Mexican to rapidly bring new production online. The Wall Street Journal (thanks to Jerôme) published an article on Cantarell last week:
The virtual collapse at Cantarell -- the world's second-biggest oilfield in terms of output at the start of last year -- is unfolding much faster than projections from Mexico's state-run oil giant Petroleos Mexicanos, or Pemex. Cantarell's daily output fell to 1.5 million barrels in December compared to 1.99 million barrels in January, according to figures from the Mexican Energy Ministry.
***
Euan Mearns put some flesh on the bones of Mexican oil production here (discusses the KMZ area very nicely and has a couple of nice maps too.)
Following on from Khebab's posts (Jan 2007 and July 2006) I wanted to put some production geology flesh on the bones of Mexican oil production. The main points I want to make are:
1. Forecast production decline of 14% per annum in Cantarell sounds alarming but it is in fact the result of planned reservoir management.
2. The forecast decline of Cantarell is due in part to the diversion of nitrogen injection from Cantarell to the neighbouring Ku-Maloob-Zaap (KMZ) complex of fields. Production at KMZ is forecast to rise to around 800 MBD and this will partly offset production falls at Cantarell.
3. Cantarell / Mexican production is predominantly heavy crude, and it is postulated that any production declines in Mexico may be met by additoinal production of Saudi Arabian heavy crude forward to 2012.
4. Notwithstanding point 3, Mexican oil production decline means that 4 out of 5 major OECD producers are now in decline (Norway, UK, USA and Mexico), leaving only Canada with growing production and this presents the OECD with a growing problem of energy security.
5. The Hubbert Linearisation (HL) for Mexico reflects reservoir management (gas lift and nitrogen injection) and new field developments but the interpretation remains equivocal. A brief description is given of why Pemex have used gas lift and nitrogen injection to boost production at Cantarell.

***
Luis de Souza put together this analysis of world oil exports for Mexico
Mexico is here assessed to be capable of producing a total of 50 Gb to 2075, giving a midpoint of depletion in 1999, some fourteen years after what appears to be a premature actual peak in 1985. Production now stands at about 3.2 Mb/d, being subject to a fairly high depletion rate of 5% a year.
Mexico seems to have peaked only in 2004, but the future decline rate is maintained.
***
Ace also does some work on Mexico on his Updated World Oil Forecast.
***
All TOD Work on Cantarell and Mexico
http://www.theoildrum.com/tag/cantarell
http://www.theoildrum.com/tag/mexico
***
Mexican Oil reserves:
http://www.ruf.rice.edu/~leeman/MexOilReserves.gif
Mexican refineries:
http://www2.nrcan.gc.ca/es/es/NA-enrgpic2006/rpImages/5-04-e.gif
***
Some PEMEX background:
http://en.wikipedia.org/wiki/Pemex
http://en.wikipedia.org/wiki/Cantarell_Field
Here's a list of refineries:
* Minatitlan Refinery (Pemex) 170,000 bpd
* Cadereyta Refinery (Pemex) 292,000 bpd
* Tula Refinery (Pemex) 320,000 bpd
* Salamanca Refinery (Pemex) 236,000 bpd
* Ciudad Madero Refinery (Pemex) 190,000 bpd
* Salina Cruz Refinery (Pemex) 320,000 bpd
***
More from PEMEX on KMZ: http://www.pemex.com/files/content/ACFWVNY7kO2v.pdf
***
Almost Total Cantarell Evacuation
Mexico clears oil rigs as hurricane nears
Might order total well closureMexican state oil company Pemex on Sunday evacuated thousands of oil workers from the Gulf of Mexico and warned it might close up to 2.2m barrels a day of crude oil production as the powerful hurricane approached.
Pemex said it was evacuating 13,360 workers – most of its workforce in the area - and that it might order the ”total closure of the oil wells” in the Cantarell oil field and other fields.
http://www.ft.com/cms/s/0/96593284-4e46-11dc-85e7-0000779fd2ac.html
***
This is from the EIA Hurricane analysis for 2007, showing historical production of oil and gas from the Outer Continental Shelf of the Gulf of Mexico.

Oil production seems to be about 1.4 million BPD; natural gas appears to be about 7.5 billion cu. ft. per day.
***
And then some charts and maps and such:





Update (Khebab, 11:55 EDT)
Hurricane Dean is a clone of Hurricane Janet (1955) who emerged in the gulf as a category 2:

Cantarell and KMZ oil complex, trajectories of Hurricane Dean and Janet (colored circles). Click to Enlarge.



Latest(10PM CDT) forcast track shows projected direct hit on Cantarell 7PM Tuesday.
http://www.nhc.noaa.gov/refresh/graphics_at4+shtml/203526.shtml?3day#con...
it seems that earlier predictions that land would slow the hurricane to below hurricane level winds may have been mistaken. This will be the FIRST landfalling Cat 5 in 15 years (Andrew) and a strengthening one at that. As of 12:45 EST NWS is stating it will re-enter Bay of Campechi as Cat2, which certainly will damage some of those old rigs...methinks the market has now overreacted in its underreaction
It'll hit Cozumel head on.
http://en.wikipedia.org/wiki/Cozumel
Cozumel has 75k people on it, its 12 miles to the coast, and the high point on the island is 15m. How much storm surge is expected with a 914mb core pressure? Is the tide in or out?
I am suddenly feeling a little sick and wishing I hadn't Googled that ...
It's going to hit far south of Cozumel. It is actually aimed directly at the capital of Quintana Roo, Chetumal.
I was looking at an older path projection. Sorry for the excitement ...
Winds down to 80 mph, making it a very low strength hurricane. Details can be found at http://www.weather.com/newscenter/tropical/?from=wxcenter_news
Again, as I mentioned on all the other threads, the sheer distance the hurricane had to cross over land would knock this hurricane down tremendously. It'll be lucky to reach a category 2 hurricane before hit makes landfall a second time. Seems like the worst is over.
What are the odds of a DIRECT HIT on Canterell at the worst possible time for the US?? Canterell may not recover. Why rebuild it if its in terminal decline? Just let whatever's left get the oil out at whatever rate it can....
Let's see here....2005 GOM oil production smashed, yet to recover fully....now Mex production smashed......Me thinks the odds are tooooo great for this to be coincidence..
I am trying not to think about the fact that there's no way the folks built those rigs to withstand Cat 2 winds in the Bay of Campeche.
But I am also trying not to think about the human toll of this storm. It absolutely wrecks me.
Professor,
I'm digging now trying to get current information on Dean. Did it actually cross Cozumel? Do we know exactly when? If the tide was in and its a 914mb storm the island is only 45' tall at its highest. How much storm surge comes with pressure like that?
If this came to pass at high tide Cozumel could very well be Mexico's New Orleans. There are 71,000+ permanent residents and if they can't have evacuated all of them.
-SCT
It hasn't made landfall yet. The recon just measured 905 mb.
Where can we get current position and speed, preferably with a map of the area? The tracks I've seen so far put it going right over the top of San Miguel.
Oh, and Wikipedia says they took a direct hit from Wilma in 2005 - took the top 5' off every tree, ruined the sewers, basically scrubbed the island bare. And now they'll be getting it just as bad, if not worse, and that is before knowing if they'll be taking storm surge, too.
Lets hope their government is more competent than ours at dealing with such events.
Best place to look is easternuswx.com or wunderground.com (Jeff Masters' blog). They're the real weather geeks and they're talking about this all night, guaranteed.
Fly the flag a little whereever you guys go. We need news tips on this story, especially on those rigs and infrastructure.
Cozumel 20.30' 86.57'
Dean 2100 GMT 18.2' 84.2'
Dean 0300 GMT 18.4' 86.0'
Dean 0900 GMT 18.6' 87.8' (estimate)
Closest approach estimate 0800 GMT (as I post) - 1.75' x 1.25'- roughly 2 degrees or 138 miles? No surge, no eye, but still quite the mess.
It is possible that many of the older rigs (drilling rigs) have not been designed to withstand a high category hurricane. However most of the newer production facilities (production platforms and FPSO) on the KMZ field have been designed to withstand cat. 5 hurricanes (100 year hurricane).
http://www.offshore-mag.com/display_article/295832/120/ARTCL/none/VESSL/...
Similarly moored installations remained in operation through both hurricanes Rita and Katrina. Although production will be shut in for a few days as a precautionary measure, production should be up again within 24 hours of the hurricane passing.
Older installations on the Cantarell field will clearly be affected to a greater degree.
I've found once a hurricane is big that the best predictor of its future intensity over water is the TCHP:
http://www.aoml.noaa.gov/phod/cyclone/data/go.html
While it's a foregone conclusion that it will wallop the Yucatan, once reduced over land it may not strengthen much over the Bay of Campeche, based on the TCHP. Of course that's not to say that it won't be strong by the time it reaches Cantarell...
_FT, LEX column, Aug 10, 2007
I would think that the government has a pretty good incentive to fix what it can.
http://science.reddit.com/info/2h52g/comments
http://digg.com/business_finance/Cantarell_Now_Directly_in_Hurricane_Dea...
We need your help folks. Spread this around, please.
From the NHC forecast update (02:25 EDT):
DROPSONDE DATA FROM THE AIR FORCE RESERVE HURRICANE HUNTER AIRCRAFT
INDICATE THAT THE MINIMUM CENTRAL PRESSURE IN THE EYE OF HURRICANE
DEAN IS NOW 909 MB...26.84 INCHES.
As if things could not get worse...
same as Camille. Yeesh.
Looking at the satellite loops it's uncanny how both arms of Dean remain over water - the northern one slotting through the gap. So even though it's got to navigate through Yucatan it might be hardly diminished. Surely the most severe hurricane that could possibly hit Cantarell.
"Waves On The River Of Mind" -- graffito outside local hamburger joint.
From the 2 am NHC advisory:
REPEATING THE 200 AM EDT POSITION...18.5 N...86.8 W. MOVEMENT
TOWARD...WEST NEAR 20 MPH. MAXIMUM SUSTAINED WINDS...160 MPH.
MINIMUM CENTRAL PRESSURE...911 MB.
Since then, Dean has continued to intensify, with a pressure of 909 mb just measured - the 9th lowest ever measured in the Atlantic basin. Landfall is imminent.
If we assume the worst and there is heavy damage on the rigs (I sincerely hope that there won't be any damage), what kind of repair timeframe could we be looking at? I mean how long do repairs on these things usually take?
Ignore me if this is too speculatory.
Hurricane Dean is really something. It made landfall a little before 8am GMT as a cat 5 – the first Atlantic basin storm to do so in 15 years since Andrew in 1992. The minimum pressures are likely to make this a “10 ten” storm. Who knows what’ll happen over the next 12 hours as it transverses the peninsula but seeing as the oil infrastructure is right on the forecast path and the storm made landfall as powerful as it did and is moving quite fast I’d say this is as bad as it could be for Cantarell and KMZ.
Hello TODers,
I sure hope they keep flying those Hurricane Hunter planes into Dean to fully instrument the 'cane crossing the Yucatan. I am insanely curious to see if Dean intensifys over land like Andrew '92 did in FL, and how far it gets before it starts to lose strength and intensity. Then if it does any re-intensification before clearing the Yucatan, then finally how much and how quickly it re-intensifies before it hits Cantarell & KMZ. Time will tell soon.
If the eastern landfall is just North of Chetumal: the storm surge of 12-18 ft, plus waves on top, plus 5,10,20 inches of rain could really saturate this area with warm water to further power the 'cane. As can be seen by this maplink:
http://images.google.com/imgres?imgurl=http://www.vivanatura.org/Yucmap3...
and a better geological relief map here:
http://ambergriscaye.com/pages/town/mapreliefbay.html
On the Eastern side: there is a huge estuary or bay that can can be overrun by this surge and then the seawater can be pushed quite a bit further inland if the surrounding land is under 20 ft above mean sea-level. Could this be miles inland? The cane rotational winds will fill this bio-reserve with stormsurge and the escape route for this water will probably down thru Chetumal Bay, instead of a quick flowback to the sea. My guess is that this storm surge will also be encountering huge floods of warm rainwater flowing down from higher locations. All potential fuel for the storm as the winds whip the heat out of these frothy, foaming waters. Not as good as hot, deepwater ocean, but it might be sufficient enough to keep Dean from degrading in intensity and size quickly on the overland traverse.
Notice from the map that a similar bay or estuary exists on the western side where Dean will be exiting into BoC. This can be another source of heat to help keep Dean strong over its land traverse if the feeder bands or eyeball bands can extract any energy from this area before finally clearing into the GoM proper.
Of course, the biggest factor is Dean's areal size. Does it have the 'oceanic suckage' to keep sufficiently sized feeder bands and eyeball bands pumping energy into the 'cane as it traverses land? From the latest satellite-animation maps showing Dean just offshore: by my eyeballing Dean appears big enough to basically fill the land traverse with the eyewall outer bands, and the feeder bands swirling in from the GoM, Yucatan Slot, and the Caribbean seem to be long and thick. All good signs to help Dean slowly degrade instead of petering out over the Yucatan. I am still speculating CAT 3 at Cantarell/KMZ, but I sure wish I could find a highly detailed topographical map of Dean's Yucatan traverse path.
Feel free to refute or embellish as desired, my fellow TODers.
Bob Shaw in Phx,Az Are Humans Smarter than Yeast?
Bob, the reason that Andrew, and some other Florida landfalling hurricanes retain much of their strength as they cross the penninsula is because of the numerous warm fresh water lakes and rivers in Florida. Also, storms maintain strength in Fl. after a rainy episode across the state that causes large amounts of standing water. My wife and I have been victims of these phenomena on numerous occasions.
Hello TODers,
To add further proof to my theory: please examine the infrared bloom that explodes off from the small bay in the western Yuc, and also the rapid expansion of the southside and northside infrared bloom when the eyewall of Dean first starts sucking the heat out of Chetumal Bay and the bio-reserve [northside of Dean] estuary south of Tulum.
http://www.weather.com/maps/news/atlstorm4/atl4satellite_large_animated....
This was from 6:00 to about 8:00 GMT if the animation auto-updates. Also notice if the eyewall tightens and shrinks as Andrew did when it got over land. Recall my earlier NOAA documented posting on this phenomena please.
Bob Shaw in Phx,Az Are Humans Smarter than Yeast?
If I recall correctly, hurricane hunters will not fly into a hurricane that is over land. There is too much turbulence. They can still measure the intensity of the hurricane accurately enough using land-based measurements.
Hello Chris,
Thxs for the history book footnotes. I just read the 4am NHC update:
----------------------
REPORTS FROM AN AIR FORCE HURRICANE HUNTER PLANE INDICATE THAT THE EYE OF DEAN MADE LANDFALL ON THE EAST COAST OF THE YUCATAN PENINSULA NEAR COSTA MAYA OR MAJAHUAL AROUND 330 AM CDT. THIS LOCATION IS ABOUT 40 MILES...65 KM...EAST-NORTHEAST OF CHETUMAL, MEXICO.
----------------------
If Dean keeps its 20 mph speed, just how much time will it take on the overland traverse? I figure the distance from Majahual to Champoton [south from city of Campeche and on the shoreline] as 200 to 220 miles; ten or eleven hours overland. Thus: 3:30 + 10 = 1:30 pm or 3:30 + 11 = 2:30 pm when it finally reaches the first waters of Campeche Bay. Does that sound about right, or does someone have a better calculation than me?
Bob Shaw in Phx,Az Are Humans Smarter than Yeast?
Per NOAA 4AM CDT Report central pressure just before landfall was 906 mb.
Just browsed the mainstream media reports on Dean. They are reassuring investors that Dean will do minimal damage to Mexico's oil fields but say nothing of the lost oil production off the world market.
This is what the BBC are saying this morning:
I don't get it, 2.7mbpd is shut in for at least a week, the most powerful hurricane to ever hit the the area is approaching... seeing as the oil infrastructure is right on the forecast path and the storm made landfall as powerful as it did and is moving quite fast I’d say this is as bad as it could be for Cantarell and KMZ. Can the old Mexican infrastructure withstand the likely Cat 2 hurricane tomorrow?
Even without any damage I would have expected more response from the market from the lost production alone.
As mentioned in an earlier post, much of the infrastructure on KMZ certainly can withstand a Cat. 2 hurricane as it is designed for Cat. 5. And as Mexico now is shifting most of its production from Cantarell to KMZ by transferring injection wells, KMZ is actually more important than Cantarell for future oil production. This is the reason for the calm oil markets. The infrastructure is far newer than that off the South US coast line that was heavily affected by Katrina and Rita. I predict that Mexican oil production will only be disrupted for a short period of time. However, I may of course be mistaken. My greatest worry is not the oil-production but the lives of the people in the path of the hurricane.
Those poor people in Quintana Roo. Its one of the very poorest parts of Mexico, very Indian, subsistance farmers and fishermen.
www.weatherunderground.com predicts that it will emerge as a category 2. The older platforms likely will have some damage, the most destructive may be to the pipelines. i doubt they are buried very deep in the bay mud, and if they are broken all the wells further out will be offline reguardless of whether the rigs are damaged
Most of the offshore workers likely live in areas which are being heaily affected with rain and winds near the bay of Campeche, the roads may not be passable for them to get to the boats to get them back to the rigs before Thursday or Friday, and the same with the seamen on the crewboats. We won't have any real idea about the true damage until early next week.
After Rita we had shortages of gasoline for a couple of days in the Houston area. There was no damage or electricity shortages, but all the tank truck drivers had evacuated with their families in the runaway scrape of an evacuation, plus the refinery workers to put the gasoline in the trucks. And I expect they will have the same in the Campeche area. You can expect those men to take care of their families firstbefore they get back to work, so it will be early next week before they get out there and survey the damage.
I seem to recall that, after Katrina, there were quite a number of offshore platforms that were not rebuilt, because it was not economical (too little oil remaining vs. cost of repair).
So, if older facilities are both more susceptible to damage and less likely to be repaired, that begs the question as to how much oil is being produced by older rigs.
Does anyone know whether this info is available?
Critical Mind
Those were mostly gas production platforms that't weren't rebuilt after Katrina, rather than oil. Gas declines a lot quicker. As I noted, i think it will be next week before the reports of permanent damage filter in. Pemex won't be able to even do a visual survey by helicopter until at least Wed. Afternoon and probably Thursday, too rough. Bob Ebersole
Bob,
Is there a good primer out there on what goes on in an oil/gas field for those of us who've not worked in the industry. My guess regarding the above statement is that natural gas is more easily recovered simply because it isa gas - no need for water or nitrogen injection to raise it, but it would be nice to have a resource that explains that stuff for we laymen.
-SCT
Note up under the header graphic "Tech Talk"...all of that is HO's sets of primers on extraction and such.