ross dress for less explosion investigation los angeles california 1985 meehan hamilton

7/30/2019 Ross Dress for Less Explosion Investigation Los Angeles California 1985 Meehan Hamilton

1/13

CAUSE OF THE 1985 ROSS STOREEXPLOSION AND OTHER GASVENTINGS, FAIRFAX DISTRICT,LOS ANGELESDouglas H. HamiltonPrincipal GeologistEarth Sciences Associates. Inc.701 Welch RoadPalo Alto. California 94304

Richard L. MeehanConsulting Civil Engineer70 I Welch RoadPalo Alto. California 94304

INTRODUCTIONLate in the afternoon of March 24, 1985, methane gas thathad been accumulating ignited in an auxiliary room of theRoss Dress-For-Less Department Store located on ThirdStreet. in the Wilshire-Failiax District of Los Angeles. Theresulting explosion blew out the windows and pa;tially collapsed the roof of the structure, reduced the store interior to aheap of twisted metal and resulted in injuries requiring hospital treatment of twenty-three people. Police closed off fourblocks around an eerie scene of spouting gas flames that continued through the night .In the following days, a drill rig brought to the site was usedto test for possible gas accumulations in the alluvial soilbeneath the store. A "pocket" of pressurized gas was encountered at a depth of 42 feet beneath the parking lot between the store building and Third Street. Gas was also encountered in several other borings at the site in smaller quantities and at lower pressures. Pressure gauges, control valves,and. on the hole where the high pressure pocket was encountered. a valved flare pipe, were installed. Following abrief period during which gas was flared and bled off intothe air, the anomalous gas condition at the Ross Store sitegradually declined to the norma! gas concentrations characteristic of the local area. In 1989 another venting incidentoccurred. this time at several sites on the north side of ThirdStreet. This second venting fortunately was detected in time,and did not ignite. In this case, water and silt were ejectedfrom outdoor vents along with the gas, in addition to accumulation of dangerous levels of gas in several buildings. Ablow-out crater several feet deep, from which din and smallstones were ejected several feet into the air was formedduring this episode which lasted about 24 hours .

The setting of the accident- an old-world Levantine markplace a few miles from Hollywood; the famed tarry gravyard of the sabre-toothed tigers; pillars of frre dancing in tdarkened streets - these biblical images attracted attentioof the press, the bar, and local politicians. And yet, thrmonths later when a hastily convened panel of experts anounced that the event was caused by digestive rumblingsan ancient and invisible swamp the whole thing had beemostly forgotten, the explanation accepted as yet another prduction of Los Angeles' quirky environment. Outside oflawsuit that was settled quietly in 1990, the possibility ththe accident was caused by the knowing agency of Los Ageles' lesser known industry or that the official report of thexperts, rather than being a serious statement of the scientifcommunity, was a heavily edited script with a happily blamless ending, was not made ! r ~ ' 1 o w n to the public, as we shaproceed to do here.

THE SETTING - TAR PITS AND OILFIELDS IN THE URBAN ENVIRONMENTThe vicinity of the gas ventings lies within a sloping plaibetween the easternmost Santa Monica Mountains and thisolated topographic high of the Baldwin Hills (Fig. I). Tharea drains via Ballona Creek. which cuts through BallonGap to the coastal plain fringing Santa Monica Bay. ThRoss Store is in a neighborhood that has been attracting atention since the mid-Pleistocene on account of its hydroca

Figure 1. Location and Regional Setting


2/13

bon wealth; the La Brea tar pits are about two blocks to thesouth.The old Salt Lake oil field, once the biggest producing fieldin California, directly underlies and extends west and east ofthe i..--nmediate neighborhood (Fig. 2). Methane gas has beena ubiquitous feature of this area so that urban engineering forboth public and private facilities has required considerationof the hazards of flammable gas present in the soil. Examples include the excavation for the L.A. County FloodControl Pan Pacific storm runoff basin and park, the stormdrain along Third Street, and the (formerly planned but sincerelocated) L.A. Metro subway along Fairfax Avenue.Before the Ross Store explosion in 1985, there was generalacceptance of the view that the hydrocarbon presence in thesurface and shallow subsurface environment had its source iii.the underlying reservoir of the Salt Lake oil field. Upwardmigration of oil from parts of this reser"'oii was thought tobe the source of the "breas" or tar accumulations that hadhapped many of the Pleistocene residents of the area (Shawand Quinn, 1986). Gas known to be present at shallow depthwithin the alluvial soil was likewise considered to haveescaped from the reservoir and migrated toward the surface(Converse et al., 1981). The surface accumulation of the residual bitumin "breas" (Spanish for pitch) led to the discovery of the once-prolific Salt Lake oil field in 1902. Asthe field developed, mairJy before 1917, furt..."ter conl .Taunica-tion between the confmed oil, gas, and water in the reservoir

Source: C.D.O.G. Map 118

, ... ., .r ) Approlro>rimate Alea nf Poa1 9$1 Oil Fleld

e Site of Gas Venting E . l

Figure 2. Wilshire-Fairfax area and the Salt Lake Oil Field... ... "'" "

-eo .. "' ... ... ..,5 .....

.. .. ... ... .. .

Figure 3 Abandoned and Active Oil Weltsin the Salt Lake Oil Field

146 C.AUSE OF THE !985 ROSS STORE EXPLOSION AND OTHER GAS VEl\TTJJ--JGS. F A L ~ F A X DISTRICT. LOS A f - - ~ G E L E S


3/13

and the surface was established by drilling more than 450wells (Fig. 3). Nearly 50 million barrels of oil were broughtto the surface through these wells. Most was conveyed offsite to markets, initially in tanks and thereafter mostly bypipeline, but some was spilled and left in sump basins thatwere then a standa...rd feature of the production system. Bythe 1930s, the urban development value of the land overlyingthe Salt Lake oil field exceeded its value as a production sitefor a declining oil field, and the field was largely abandoned.The legacies of a thirty-year oil boom included several largefamily and corporate fortunes, a few square miles of landwhere the natural upward hydrocarbon migration was complicated by the presence of hundreds of abandoned wells plusnumerous filled sumps and covered-over spills, all underlainby a partially depleied hydrocarbon reservoir. Following theabandonment of most oil production, the area was rapidlycovered by the homes, apartments, businesses and industrialfacilities that exist today (Fig. 2). The remaining and mostnotable hydrocarbon presence is manifest in that popular andscientifically important exhibit of the "La Brea Fossil Pits"and Page Museum in Hancock Park on Wilshire Boulevardnear Fairfax Avenue.Meanwhile, beneath this urban environment, t . ~ e underlyingSalt Lake oil field was again tapped, in 1961, through skillfulapplication of modem oil drilling techniques and productiontechnology. This time, however, the entire surface installation was contained in an unobtrusive, screened complexcovering only one acre. This compact facility included the

- ~ I

..,I

IIII

heads of 43 wells which were slant-drilled to exploit sevelease biock hoidings over an an:a of about a square m(Figs. J, 4), together with necessary equipment for separatof the well-head crude oil into oil, gas, and water coponents. The gas was disposed of by injection back into Gilmore block of the field between 1961 and 1971. Swater was initially disposed of by discharge into a lostorm drain, but for environmental reasons was required to injected back into the field, also in the Gilmore block, sin1964 (Crowder and Johnson, 1963; California State Divisiof Oil and Gas [C.D.O.G.], 1974).

AFTERMATH OF THE 1985 ROSS STOREEXPLOSIONAfter the flames subsided and the post-explosion exploratiand gas-control drilling was completed, a special Task Fowas convened at the direction of the Los Angeles City Coucil. The Council wanted to know what had happened ahow future accidents could be prevented. The Task Forwas made up of representatives from both City and Staagencies and from industry, and it had available a GeologTechnical Committee which included experts in engineerigeology and other relevant disciplines. Concurrent with tdeliberations of the Task Force, the Senate of the CalifornState Legislature enacted SB1458, the Roberti Bill, which rquired a study of the location of all abandoned oil and gwells within the State of California.

IIOUUYAAD EXPLANATION

. Concllled fiUII---- Surt.c.Dtceof. . . .uL Jc. . . ...... , .

ecJUI..EVAIIIp

Figure 4. Surface plan of wells, faults and gas ventings, vicinity of Ross Store S1teCAUSE OF THE 1985 ROSS STORE EXPLOSION AND OTHER GAS VENTINGS. FAIRFAX DISTRICT. LOS ANGELES 14


4/13

< O e s p l t ~ i ~ own initial interest in abandoned ' > ~ t e l l s as: a poctential source of the gas and the mandated purpose of theRoberti Bill to study such wells, the Task Force's surprisingconclusion was that neither the existence of dozens of suchabandoned wells, nor even the presence of the underlying oiland gas field which had been venting to the surface for ahundred millennia, had anything to do with the gas ventingaccidents. In fact, the oil field operation was not even mentioned as a possible agent in the Task Force report. Instead,a chemical analysis of a sample of the venting gas led theTask Force to conclude that the gas was of "biogenic" origin,presumably derived from near surface decaying of organicmatter in the alluvial soil, rather than being of petrogenic orthermogenic origin, i.e., derived from rock-source hydrocarbon (oil fieid type) accumulations. The Task Force thenwent on to present a scenario of shallow "biogenic" methanebeing displaced and pressurized by a rising water table in theperched fresh-water aquifer beneath the Ross Store. It wasan imaginative explanation: what school child is not awareof that primordial scene of mastodons thrashing about in agassy swamp (as is artfully represented at full scale in theHancock Park exhibit, only two blocks from the Ross Store).Troublesome legal issues were eliminated by this conclusion,the impiication being that the methane hazard could exist virtually anywhere, so no human agency was at fault for itsworkings. The presence of the abandoned wells, the filledsumps, or anything else having to do with the past and current exploitation of the Salt Lake oil field evidently was considered no more than coincidence.About a year after the issuance of the Task Force report, theresults of two separate studies of the composition and indicated origin of near smface gas samples both at the RossStore and elsewhere in the Los Angeles Basin, became available. One report was prepared under contract to theC.D.O.G., in response to the Roberti Bill (Geoscience Analytical, Inc., 1986). It presented the results of a study ofeight selected urban areas sited over oil and gas fieids wherethere were wells abandoned prior to 1930, and where therewas a history of oil and/or gas seepage. This report concluded that with few exceptions. the methane present in theseareas was, as at the site of the Ross Store building, of biogenic origin and thus unrelated to the underlying oil and gasaccumulations. In their opinion it was essentially swampgas. This remarkable conclusion was all the more interestingsince it turned out that most of the areas sampled are localtopographic highs. In any case, the resujts of this massivestudy assured the public the hazardous conditions identifiedby the Task Force as existing at the Ross Store, as well as"the vast majority of the high concentration gas samples"taken from the surface areas overlying oil and gas depositsthroughout the Los Angeles basin, were also of purely naturalswamp gas origin unrelated to human activities or oil fieldpresence.The other study, based on isotopic analysis of samples of gasat the Ross Store building escaping on the day following the

e ~ p l { ) ) > l t " Y n , c o ~ t d u d e d l!:u


5/13


6/13

the character of the 1989 venting incident led the subsequently reconstituted Task Force to abandon the "risingwater table" mechanism, but to otherwise leave as open questions where the methane actually comes from and the meansby which it is forcefully vented to the surface.

TASK FORCE REPORT II - DISCUSSIONThe Methane Gas Task Force was reestablished by the LosAngeles City Council three days after the second major gasventing in the Fairfax District, which occurred on February7, 1989. Its charge was again to report on the recent incidentand "present recommendations necessary to protect healthand safety within the area". The revived Task Force beganits work with less urgency than had existed in 1985, since noactual damage had occurred in connection with the 1989venting, but with a larger and more confusing data set.Some of the data that had accumulated since 1985 weresimply contradictory, mainly the two different gas-origin studies. One indicated a shallow-source biogenic origin for thegas, the other indicated an oil field origin for the Fairfax gas,but the forceful ejection of gas, water, and mud during the1989 venting clearly required a driving mechanism withmore energy than a "rising water table". Faced with thesecontradictions and uncertainties, the 1989 Task Force electedto, in effect, throw up its hands with the following paragraph ."CONCLUSION ON ORIGIN - Obviously, the origin of themethane gas is complex and not clearly understood. Tworeputable firms with qualifications and experience seem tohave opposing opinions on the origin of the gas. But. i f weintend to shield buildings or vent areas of probable methanegas seepage. the origin of the gas realiy does not matter.The most probable origin of the gas 'may' be a 'combinationor thermogenic and shallow biogenic sources with the highpressure events from thermogenic (deep) sources and the/ow-pressure background gas from a combination of biogenicand thermogenic sources. So, in reality, the origin of themethane has its implications in the total understanding of theproblem, but not in remediation or control. Methane gasproduced within deep fields or in shallow marshes will notphysically function nor react in the same manner. Both arecapable of explosive conditions and both can be safelyvented to the atmosphere."Before going on to its recommendations for dealing with theproblem, by a 3-point program comprising venting by reliefwells, use of an area-wide monitoring survey team, and Cityinspection for Code compliance, the 1989 Task Force pausedto provide assurance that neither past nor present activitiesrelated to exploitation of the underlying Salt Lake oil fieldhad anything to do with the 1985 or 1989 gas venting incidents. With regard to the possibility that the gas might haveescaped via one of the 400 plus abandoned wells thatformerly tapped that field, the Task Force stated the following:

"ABANDONED WElLS AS A SOURCE OF METHANE -The potential hazard imposed by old abandoned wells wnot considered in the 1985 report because the gas wabelieved to have formed within the shallow biogenic areaabove the production zone. However, the extent of chemicstripping of gas constituents and microbial alte;nationduring migration probably could not occur within a gabubble rising upward in a vertical shaft filled more or leswith water or brine. These could. however. be the sourcelow-volume methane that mixes at the surface with ambiegas at low pressures. However all available chemical anisotopic information exonerates these abandoned wells as thsource of the higher-pressure methane events. (Underlininadded for emphasis).These wells are an entity to be considered in the future or athey are uncovered by area construction. But, even i f thwells could be located. reabandonment to modern standardwould be an expensive program."The writers do not understand how the isotopic informatioreported by Global Geochemistry (1986), which indicated thRoss Store gas to be thermogenic, hence of oil field originin any way exonerates a potential conduit from the oil fielto the surface such as an abandoned well.With regard to the possibility that the problem might be related to disposal by pressure injection back into the subsurface of salt water stripped from the crude oil currently beinextracted from the field, the Task Force stated the following:"PRESSURE INJECTION OPERATIONS - Within the SaLake field aboui 42 wells are curremly producing from th"C" and "D" Production Zones. According to productiorecords of the Division of Oil and Gas. more fluids are beinwithdrawn than are returned to the production zones. Thtype of injection used by McFarland Energy is a water disposal program that does not cause field repressurizationCurrent oil field operations within the Fairfax area, therefore, do not appear to have an adverse affect on the aremethane seepage." (Underlining added for emphasis).We address this latter conclusion in the discussions presentein following sections.Although Task Force Report II explicitly "exonerated" pasand present oil field operations from any role in the 1985 an1989 gas ventings. it did, unlike the 1985 report, at leasmention them. But, even this was evidently regarded astepping outside of some kind of bounds by the Task Forcmember representing the C.D.O.G., since that member wrota letter providing comments regarding the second draft of thTask Force Report II which seemed to generally take his fellow members to task for even mentioning oil field operation(Baker, 1989). This letter was provided as an attachment tTask Force Report II .

150 CAUSE OF THE 1985 ROSS STORE EXPLOSION AND OTHER GAS VENTINGS, FAIRFAX DISTRICT. LOS ANGELES


7/13

At the same time, a contradictory view was se t forth inanother letter, this from Dr. James E. Slosson. fanner California State Geologist and Consultant to the Task Force inboth 1985 and 1989. In this letter, Slosson (1989) details hisdisagreement with virtually every point made in the TaskForce Report II that dissociates the 1985 and 1989 gasventing incidents from the Salt Lake oil field. Unlike the letter from the C.D.O.G. member, Slosson 's letter was not attached to or referenced in Task Force Report II, even thoughit was dated six weeks earlier.

GEOLOGIC SETTING OF THE FAIRFAXAREAThe geology of the Ross Store area has been studied and described in terms variously of surficial geology, emphasizingthe formation of the La Brea tar pits (Shaw and Quinn,1986), hydrogeology (DWR, 1965), neotectonics (Richards,1973; Hill and others, 1979), engineering geology (Converse,1981; L.A. County Aood Control, 1977) and petroleum geology (e.g., Hoots, 1931 ; Soper, 1943; Crowder and Johnson,1963; Wright, 1991). To summarize, the area is situated nearthe northern edge of the Los Angeles Basin, and the centralaxis of the Salt Lake oil field nearly parallels and lies about2 miles south of the Santa Monica fault. This fault forms thelocal boundary between the Western Transverse Ranges andthe Los Angeles Basin. The Santa Monica fault is generallyrepresented as a steeply north-dipping reverse fault, whichhas been active at least as recently as during late Quaternary

A

1001100

1200I 15005 1001100

time (Hill and others, 1979; Crook a.nd .ProctorvoluiT!e). Hill and others (1979) based on data prti:tem by the City of Los Angeles Bureau of Engineerinfollowing Grant and Sheppard (1939), studied the ssubsidence in the region of the Santa Monica fault, incthe Salt Lake oil field area. Although Hill and others viewed the results of analysis of recent micro-earthquativity as suggesting that subsurface faults in the Santaica-Raymond fault zone are undergoing tectonic strainmulation, the pattern of surface subsidence they mwithin a 5-mile wide zone along the inferred tracefault could be interpreted as indicating surface relarather than compression (as well as indicating respoconsolidation in the subsurface resulting from prior extof ground water). So it is not clear that the Santa Mfault is contributing to any near-surface tectonic deforin the area. Within the ground south of the Santa Mfault, the folds and faults that form the structural trapSalt Lake oil field are the principal local geologic struThis area is underlain by about a 200-foot thickness oternary shallow marine and aiiuvial deposits, which olate Miocene and Pliocene age Repetto and Pico Fonn(undivided) and late Miocene Puente Fonnation . Theary strata are both tightly folded and displaced by steeclined reverse faults, but the overlying Quaternary straapparently undeformed (Fig. 5). Therefore, the tecthat formed the compressional structure of the Salt Lafield therefore died out or has been quiescent since apmately Plio-Pleistocene or at least early Pleistocene

Ul BI'N Tr PitSLagoon A'0300

}:150011002100

2400270030003300

J


8/13


9/13

plot shows the gradual decline in oil production noted above.Of particular interest in the context of the present discussion,however, is the history the production (and disposal) ofwater, brought to the surface along with the crude oil and gashydrocarbon fluids. This is plotted separately in Fig. 7.showing monthly production and injection back into the fieldof the salt water byproduct of oil production. The indicatedpattern is one of steadily increasing water production as oilproduction decreased, Fig. 6, from the beginning of 1980through mid-1984. Between mid-1984 and about mid-1985,however, reported water production declined precipitouslyfrom around 14,000 to around 6,000 and less bbVmo, the lowbeing about 3,000 bbl in January 1986. A gradual upwardtrend in water production was re-established in 1985, so thataverage water production had increased to about 8,000bbVmo. by late 1988.Taken at face value, the water production data shown in Figs.6 and 7 would seem to indicate either a radical change in theoil-water ratio of the field's production, or a remarkablysuccessful exercise in selective control of water production(for example, by shutting in certain wells or grouting off certain producing zones). In fact, however, well records on filewith the C.D.O.G. show no evidence that anything like thisoccurred. What did happen, however, was that a secondwell, Gilmore No. 9, was reconditioned and converted from aproduction to an injection disposal well, in 1984. Meanwhilethe operator, McFarland Energy, Inc., sought and receivedpermission from C.D.O.G. to raise well head in ection res-

sures fourfold from 200 psi to 770 psi. The worlcc o r d ~ show that considerable effort was expended in 9 conversion, including acid treatment with 5,000 galhydrochloric acid and 2,500 gallons of hydrofluoric January, 1985. The well was then tested by injectininto the C zone of the field at 475-500 psi. ThereafC.D.O.G.-tabulated records of production for the Saoil field show Gilmore No. 9 as a disposal well, but notion disposal is recorded for it prior to March, 1989.ever the dramatic decrease in the quantity of water dof by injection into the original disposal well, Gilmo16, coincided with the time Gilmore No. 9 was convea disposal well. Since the production of oil did not corresponding decrease, but instead actually increa3,000 to 4,000 bbl/mo. from early 1985, when the din water disposal was most dramatic, to early 1986, planation for the pattern of water disposal must lie elsGiven the generally constant (or at least only grchanging) ratio of oil to water for production from thiwe suggest that the explanation may involve omissionfrom the record rather than implausibly effective conwater production. According to this scenario, by miwater production would have exceeded the injection cof the only disposal well in the field, Gilmore No. work was begun to convert a second well, Gilmore Ncomplete.d in the same lease and zone as No. 16, to fas a second disposal well; this was initiaUy accomplislate 1984, and an increasing percentage of the field'sproduction was diverted into the new disposal weU t

Barrels (Thousands)1 8 ~ - - - - ~ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ,

1 " 7 10 , " 7 10 1 " 7 10 1 " 7 10 1 " 7 10 1 " 7 10 1 4 7 10 1 4 7 10 1 4 7 10I eo I a1 I 82 I 83 I u I 85 I 88 I 87 I 88 I

Fli!;clure 7.. RePQ rted m t l ' l : l ~ producti.on/injec. on o! water


10/13

m i < l Thereafter, we suggest that the water productionmay have been disposed of in both wells, probably in aboutequal quantities between them. For some reason, however,none of the (inferred) injection into Gilmore No. 9 wasentered into the C.D.O.G. statistics, so C.D.O.G. records showonly the production-injection volumes plotted in Figs. 6 and 7.It is of interest to note that the gas venting that caused theRoss Store explosion (on March 24) occurred about twothirds of the way down the sharp decline in Well No. 16'srecorded disposal history (Fig. 7). Declining injection intoWell No . 16 seemingly would tend to decrease a tendencytoward formation destabilization, i f the factors of it1jectionpressure and interference from recently established injectionthrough Well No. 9 are neglected. However, it seemspossible that one or both of these factors may have caused orcontributed to an excursion or incremental buildup of formation pressure in the Gilmore Block "C" zone adjacent to theThird Street fault. I f the injection performance of Well No.16 is considered alone, the prior history of that well showedthat the operator attempted to compensate for declining injection (due to clogging) by increasing the well head injectionpressure as early as 1980. This would result in a corresponding increase in formation pressure, even if the actual amountof injection take did not increase (or continued to decline).The effect of transferring some of the injection to Well No.9, located on the opposite side of Well No. 16 reiative to theThird Street fault (Figs. 4, 5), would be to create a new pressure bulb which would form a barrier to the radial dissipationof pressure around Well No. 16, and would result in furtherconcentration of pressure in the zone between Well No. 16and the fault.

ALTERNATIVE SCENARIOS FOR EPISODICGAS VENTINGSThere appear to be two basic scenarios that have been proposed, and one we propose, to explain the 1985 and 1989 gasventings in the vicinity of the Ross Store near Third andFairfax:

Scenario 1:Scenario 2:Scenario 3:

The 1985 Task Force biogenic gas/risingground-water scenario;The Roberti Bill oil field gas/abandonedwell conduit scenario; andThe oil field gas/intermittent fractureconduit scenario.

Scenario 1 would be favored by the interpretation that thechemical composition of gas san1p1es obtained following theMarch 24th explosion indicate the methane gas could nothave originated under petrogenic or thermogenic (oil field)conditions, hence could not have had its source in the underlying oil field. This interpretation, of course, is contrary tothe indication from the isotopic analyses that the gas in fact

did originate under thermogenic conditions. Negleclh!g thiswe believe that beyond this particular coincidence of a gablowout occurring directly over a pressurized oil field .bubeing completely unrelated to that field, the dubious conclusion of the Roberti Investigation, that the "vast majority" ogas above oil-gas fields in the Los Angeles Basin is unrelateto the underlying fields, effectively destroys the credibility othese "biogenic source" interpretations. Note that the 198Task Force abandoned the rising water table as a drivingmechanism, while preserving the option that the gas mighstill be of shallow biogenic origin.Scenario 2. The Roberti Bill "leaking abandoned wellscenario, corresponds to a recognized phenomenon, whereinpressurized gas and/or oil, water and entrained sedimenescape via an abandoned well (or other conduit, includingnatural fractures and fault zones), either after the source irepressured, after the blockage or seal in the well fails, oafter the fracture conduit is opened by some means. Thiscenario was dismissed by the 1985 Task Force on the basisof the "biogenic gas" analysis, since gas escaping from theunderlying oil field would presumably be of "petrogenic" o"thermogenic" character. The 1989 Task force explicitly "exonerated" leaking abandoned wells as the source of either the1985 of 1989 venting incidents even though the 1986 GlobaGeochemistry report had at least cast doubt on, if not, in theview of the Task Force, altogether disproved, a shallow biogenic origin for the gas. Endres and others (1991) on theother hand, cite leakage from naturally repressurized uppezones in the underlying oil field, most probably via old wellsor well bores, as the probable source of the vente-d gas. Weconsider there to be three objections to the argument thaabandoned wells are the cause, or at least the exclusivecause, of gas migration during the 1985 and 1989 ventingevents at Fairfax:a) There is no known abandoned well within about 300 feet othe Ross Store building (nor within 100 feet of either ofthe 1989 venting sites), according to C.D.O.G. Map 118(C.D.O.G., 1984);b) The ventings were of brief duration- 24 hours or less,whereas an abandoned well blowout or fracture ventingusually lasts longer and may continue indefinitely if notcontained (a point also recognized in the Task Force Report II); andc) The Ross Store site overlies Lease Block U-55 of the SaltLake oil field. This block is isolated in the subsurfacefrom the main part of the field to the north by the ThirdStreet fault, but is undergoing active pumping extractionvia wells slant drilled from the Gilmore Drilling Island.The abandoned original vertical wells in this lease blocktherefore bottom in a part of the field where oil, water anddissolved gas are being pumped out and pressure isthereby maintained below hydrostatic. Even in case theseal in one of these abandoned wells should fail, the ambient reservoir fluids should not have enough pressure tovent to the surface.


11/13

Endres (pers. comm., 1991) suggests that gravity drainage ofoil down the steep north flank of the anticline in theMetropolitan lease block would leave a gas cap at the crestof the structure, and this together with a process of naturalrepressuring by access to high pressures from deeper withinthe structure means reservoir sands beneath the Ross Storewould form a preferred source of gas. We believe, however,that in case repressuring (or some relict overpressuring) ofthe oil and gas-bearing zone beneath the site existed, themost likely route of fluid (especially gas) escape would bevia the Sixth Street fault. This fault breaches the top of thetight anticlinal fold in the Metropolitan Lease Block (southof the Third Street fault - Fig. 5). Another route for gaswould be via abandoned wells with inadequate or failingseals located in the area over the crest of this anticline. Thecrestal part of this structure, however, trends parallel to ThirdStreet and lies about a thousand feet south of the Ross Storessite (Fig. 5). There would be required, therefore, for considerable lateral migration of pressurized gas (and water) for escaping oil field fluids to reach the Ross Store and GilmoreBank venting sites.Scenario 3. We propose a third alternative scenario, whichwe call the oil field gas/intermittent fracture conduit scenario.This scenario, not mentioned in the Task Force report nor setforth in the Roberti Bill charge, envisions a mechanism ofintermittent releases of pressurized gas from the Salt Lake oilfield, via passages temporarily opened along the Third Streetfault, through the process of hydraulic fracturing. The fracturing would be triggered by buildup of fluid pressure causedby injection disposal of salt water in the Lease Block U-20fault-isolated compartment of the field. This mechanism issimilar to the fault activation process that occurred in the Inglewood Oil Field in response to pressure injection for secondary oil recovery, which in that case was instrumental inthe breaching and ultimate failure of Baldwin Hills Dam in1963 (Hamilton and Meehan, 1971). In that case, there wasboth escape of injected fluid to the surface along one or morefault conduits. and differential movement across faults. Inthe Ross Store explosion scenario proposed here, the ThirdStreet fault fractured enough to allow upwardly migratingpulses of gas-pressurized water and free gas discharging intoand through the overlying alluvial section, without detectibledifferential movement of the fauit itself. With migration ofthe water upward, gas would be released either into the soil,or could vent directly to the surface.

THE OIL FIELD GAS/INTERMITTENTFRACTURE CONDUIT SCENARIOThe actual mechanism of fracturing and migration is indicated by the following calculations, in which the process ismodelled by considering a fault which acts as a membraneseparating adjoining fluid bearing zones within ProductionZone C. We place our calculations at 3000-feet depth,roughly representative of the injection depth. This condition

.-------------------- -- .. ~

t HCftEAIIINOPMSSUIIEAT FAULT

Figure 8. Diagram showing hypothelical evolution and distributionpressure from injection wells .is shown on Fig. 8. We assume that the producmg Zone has the following history: fluid pressure initially hydrostatior 3000 x 0.45 =1350 psi, with resulting partial buoyancy sediments. Reduction of fluid pressure to about 200 psi, folowing extended periods of production causes compressio(consolidation) of sediments with increases in both verticand horizontal effective intergranular stresses. Assuminduring this process zero horizontal strain, (the so-called "rest" or Ko condition, in which the ratio Ko of horizontalvertical intergranular stress is about 0.5), and a unit weight sediments corresponding to 1 psi per foot depth (144 pcresults in horizontal total pressure of (3000-200) x 0.5 + 2=1600 psi. Under this condition, any pressure, Px, from tinjection side of the block a c t i n g . ~ n the fault membrane thexceeds the 1600 psi may cause incipient displacement paning of the fault.Now, with injection at X, fluid pressure will increase. Rstoration of hydrostatic pressure in the X side of the produing zone will create original fluid pressure of 3000 x .451350 psi. This is 250 psi (at ground level) short of the flupressure that would equal the 1600 psi necessary to displathe fault membrane slightly. This incipient displacemecondition would require an injector well head pressure of least 150 psi. but in the short-term it would need more th150 psi because of pressure loss in the well bore and timdependent pressure loss between the well and the fault. Thpressure decay process including both geometric and ter


12/13

poral decay factors can be simulated by modelling, as isroutinely done by petroleum engineers. Even without doingso, however, it appears likely, based in part on other studiesof the field , that the permitted 770 psi wellhead injectionpressure would suffice to raise the formation pressure enoughto displace the fault membrane. (Note that fracturing occursat the fault discontinuity at some significant time after thewell is pressurized, not at the well, therefore short-term hydrofracture tests of the well, which are sometimes required tocheck for hydrofracture hazards, would not indicate a potential for later hydrofracture at nearby faults).Displacement of the fault membrane would permit openingof a parting lenticular fracture transmitting a mobile pulse offluid which would mig rate rapidly and upward (Fig. 9).Escape of gas from the mobile and increasingly less confinedfluid into the alluvium (or into abandoned wells) would follow. The fracture opening, however, would collapse as thefluid pressure was released, and would remain closed until anadequate pressure gradient again built up in consequence ofcontinuing injection into the formation. This would accountfor the episodic and brief character of the surface ventingevents.

r - B r i n e ~ OilandGaIJil!looalWoU......... I vPracLdon W. l

p""

- K,-0 .5o-lic Gradient 1 psU!lDllpoaal Woll P"'""'"" l,CI Tllroughoulln;octlonZono

Py . K, {3000-200) + 200 1800 polP1 3000 K 0.43 + Pwh 1290 + Pwn

H P1 Lo. lndpienlF - . d l o np..,. 1800. 1210.310 polF . . . - r e O - ~ .0.53

Figure 9. Fault injection pressure for case of long term injection

CONCLUSIONOur analysis indicates that the raise of well head presGilmore #16, sanctioned by C.D.O.G., from 200 to 7was sufficient to fracture the formation at thefault. This possibility is not recognized in current Creview of oil field operations, though the analyticaniques necessary for its consideration are certainly ain the oil industry. Moreover, there is evidence sugthat additional injection, unreported in C.D.O.G. recortaking place at the time of the disaster. It seems to future studies of the Los Angeles gas problem can haconsidered complete without full investigation of thsues. Finally, our analysis suggests that the criteria cuemployed by C.D.O.G. and industry may not adequateteet against formation fracturing and escape of formatiinjected fluids and gas to the surface envirorunent, rin disastrous consequences comparable to the 198Store explosion.

ACKNOWLEDGEMENTSThe writers benefitted from assistance in analysis of dfonnulation of geologic interpretations provided by PhFrame and John Wakabayashi. A petroleum geologwishes to remain anonymous provided helpful insighseveral aspects of the new development and exploitathe Salt Lake oil field . Bernard Endres, equaUy at hthe law and the sciences, shared his views on the prinvolved in the methane venting in 1985, and also prreports and data not available directly from the public the C.D.O.G. The Long Beach office of the C.D.O.Gfully responded to our request for watertlood testing intion. Dr. James E. Slosson provided a copy of hicommenting on the 1989 Task Force Report II.Each of the authors was involved in technical analysescauses of the methane gas ventings in the Fairfax Distconnection with the disputes that arose following thexplosion. The views and interpretation presented in ticle, however, are purely those of the writers and do nessarily correspond to the views and interpretationsparty to these disputes, or to any other entity.

REFERENCESBaker, 1989,lenerprovided as attadunent to Task Force Report II .California Division of Oil and Gas, 1974, California Oil and Gas FieldVolume II.-----------=- 1984, Las Cienegas. Lake, W. Ponion, Logeles, City. Beverly Hills East Area, San Vicinte E. Ponion, Sout.lt Sal!Map 118.California Depanrnent of Water Resources. 1965, Water well standardCentral, Hollywood. Santa Monica basins, Los Angeles Coumy, CalifoDepamnent of Water Resources Bulletin 74-4.Converse Consu!ta!lts, Earth Sciences Associates. and GeoResourcc Cants. 1981. Metro Rail Project Geotechnical Investigations Report, VoRcpon submitted to the Southern California Rapid Transit District.

156 CAUSE OF THE i985 ROSS STORE EXPLOSION AND OTHERUAS V"ENrh"GS. E.-\IR!FAX .DlSTI


13/13

Crowder. R.E., and Jolmson, R.A ., 1963, Recent developments in Jade-Buttram area of the SAlt Lake Oil Field: California Division of Oil and Gas Summary ofOpenuions. California Oil Fields, Annual Report, v. 49, p. 53-59.E m l r ~ s . B., Chilingarian, G.V., and Yen , T.F., 1991, Environmental hazard oiurban oil field operations: Jour. Petroleum Science and Engineering, vol. 6 (inpress).Geoscience Analytical, Inc ., 1986. A study of abandoned oil and gas wells andmedtane and other hazardous gas accumulations: Report to California Department of Conservation, Division of Oil and Gas.Global Geochemistry, 1986, Unpublished repon referenced in Task ForceReponllGrant, U.S., and Sheppard, W.E., 1939 , Some recent changes of elevation inthe Los Angeles basin of southern California and their possible significance:Bulletin-Seismological Society of America, val. 29, p. 299-326.Hamilton, D.H. , and Meehan. R.L.,1971, Ground rup1ure in the BaldwinHills: Science, vol. 172, p. 333-344.Hill, R.L., Sproue, E.C., BeiUiett, Real, C.R and Slade. R.C 1979,Earthquake hazards associated with faults in the greater Los Angelesmetropolitan area, Los Angeles County, California. including faults in theSanta Monica-Raymond, Verdugo-Eagle Rock. and Benedict Canyon faultzones: U.S. Geological Survey Open File Repon 79-16.Hoots, 1930, Geology of the eastern pan of the Santa Monica Mountains. LosAngeles County, Calif=ia: U.S. G-eological Survey Professional Paper !65-C, p. 83-134.Los Angeles County Flood Control District. 1977, Proposed Pan Pacific Retention Basin, Repon of Investigation, Summary of Geotechnical Investigation.

Richards, C.A . 1973, Engineering geology aspects of petrok trm in tlle urUenvirotunent: in Moran, D.E., Slosson, J.E., Stone, R.O., and ~ i v ~ m m ,(eds.), Geology, Seismicity and Enviroomental Impact: Associarion of Engineering Geologists Special Publication,Los Angeles, p. 391-399.Shaw, C. A and Quinn, J.P., 1986. Rancho La Brea: a look u coutalsouthern California's past: California Geology, vol.39, p. 123-133.Slosson, J.E., 1989, letter regarding conclusioos ofTask Force Report II.Stille, H., 1936, Present tectonic state of the eanh, Am. Assoc. Pt:tro. GeolBull. 20, p. 849-880.Soper, 1943, Salt Lake Oil Field in Jenkins, O.P., (ed.). Geologic fonnariooand economic development of the oil and gas fields of California: CalifornDivision ofMi.l!es and Geology, Bulletin ! 18, p. 284-2!!6.Task Force Repon. 1985, Task Force Repon on the March 24. 1985 methagas explosion and fire in the Fairfax area: City ofLos Angeles.Task Force Repon. 1989. Task Force Repon II on the methane gas incursioCity of Los A,,gcles.Wright, T.L., 1987, Geological Setting of the RancholaBrea TarPits in T.Wright and R. Heck, (eds.) Pt:trolcum Geology of Coastal Soulhem Cali fonia: Pacific Section AAPG Guidebook 60, p. 47-91.Wright, T.L i 99i , Structurai Geology and Tectonic Evoiution of he LosAgeles Basin, California, in K.T. Biddle, (ed), Active Margin Basins; Am.Assoc. Petro. Geol. Memoir 52, p. 35-134

:

ross dress for less explosion investigation los angeles california 1985 meehan hamilton

Documents