bioremediation of diesel from a rocky shoreline in an arid tropical climate

Marine Pollution Bulletin xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Marine Pollution Bulletin

journal homepage: www.elsevier .com/locate /marpolbul

Bioremediation of diesel from a rocky shoreline in an arid tropicalclimate

http://dx.doi.org/10.1016/j.marpolbul.2015.07.0590025-326X/� 2015 Elsevier Ltd. All rights reserved.

E-mail address: [email protected]

Please cite this article in press as: Guerin, T.F. Bioremediation of diesel from a rocky shoreline in an arid tropical climate. Mar. Pollut. Bull. (2015),dx.doi.org/10.1016/j.marpolbul.2015.07.059

Turlough F. GuerinClimate Alliance Ltd, c/o 5 Retreat Crescent, Sunbury 3429, Victoria, Australia

a r t i c l e i n f o

Article history:Received 17 February 2015Revised 22 July 2015Accepted 25 July 2015Available online xxxx

Keywords:SpillMarineMaintenanceResourcesShorelineFlushingFuel

a b s t r a c t

A non invasive sampling and remediation strategy was developed and implemented at shoreline contam-inated with spilt diesel. To treat the contamination, in a practical, cost-effective, and safe manner (to per-sonnel working on the stockpiles and their ship loading activity), a non-invasive sampling andremediation strategy was designed and implemented since the location and nature of the impacted geol-ogy (rock fill) and sediment, precluded conventional ex-situ and any in-situ treatment where drilling isrequired. A bioremediation process using surfactant, and added N & P and increased aeration, increasedthe degradation rate allowing the site owner to meet their regulatory obligations. Petroleum hydrocar-bons decreased from saturation concentrations to less than detectable amounts at the completion oftreatment.

� 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Resource construction, as well as operational mine sites, forboth new and expansion of existing projects, has been a majorcontributor to the growth of Australia’s economy in the past twodecades, deploying large numbers of plant and equipment, andnecessitating the transfer and use of large volumes of chemicalsincluding fuels and lubricants. Despite the strong culture ofcompliance of Australian resource companies to the standards gov-erning the safe handling of hazardous and dangerous goods, loss ofcontainment occurs on mining and resource construction opera-tions (Guerin, 2014), in some cases with potentially significantimpacts on aquatic and marine ecosystems, or even human health(Colombo et al., 2005; Hewstone, 1994; Ismail and Karim, 2013;Reed et al., 2005; Smith, 1995; Vazquez-Duhalt, 1989). Habitat lossand its effects on biodiversity are a growing global concern and is amajor cause of the decline of coastal species. Changes in habitat asa result of petroleum spills is of particular concern (Margesin andSchinner, 2001; Olagbende et al., 1999), and the region wherethe current spill occurred is an important marine conservation areaand is under consideration by regulatory authorities for furtherincreasing the conservation value of the region.

Mining operations and resource construction with activitiesadjoining marine waters, present a particularly high potential risk

to marine pollution, and mine owners and plant operators must bevigilant to ensure integrity of chemical (fuel, lubricant, hydraulicfluid) storage tanks, lines, pumps, secondary containment andwash down facilities (Guerin, 2002, 2009). Operational personnelmust also be trained in the risks and the specific controls to pre-vent fluid releases and emergency procedures where preventativemeasures fail. Problems associated with working close to marinewaters are the risks from loss of containment from such facilitiesand the resulting loss of aquatic life and negative impacts onmarine ecosystems. While biodegradable refined products areincreasingly gaining popularity in marine and related applications(Mercurio et al., 2004), spills into marine waters (regardless of pur-ported biodegradability), attract heavy fines and pose unacceptablerisks to operator (owner) companies.

Diesel is a ‘‘light oil’’ and small spills of 2–20 kL will usuallyevaporate and disperse within a day or less. For larger spills, a resi-due of up to one-third of the amount spilled will usually remainafter a few days. Diesel will not sink or accumulate on the seafloor,except as a result of adsorption to sediment, but is considered to beone of the most toxic oils to water-column organisms and fish killsare possible if large spills occur in non-readily mixing water. Lightoils contain moderate concentrations of soluble toxic compounds,and can leave a film or layer in the intertidal zone with the poten-tial to cause long-term contamination (Anon., 2011).

Heavy fuel oils are typically not very toxic to water columnorganisms compared to lighter oil. Diesel on the other hand is

http://

http://dx.doi.org/10.1016/j.marpolbul.2015.07.059

mailto:[email protected]


http://www.sciencedirect.com/science/journal/0025326X

http://www.elsevier.com/locate/marpolbul



Fig. 1. Overview of the location of the spill. (Top Left) Primary (feeder) stockpile of mineral salt in foreground that feeds the stockpile on the island where the spill occurredand the remediation treatment was setup and showing proximity to the marine environment. (Top Right) Secondary stockpile that is used to load ships located on the islandwhere the spill occurred and was the source of salt contamination in the spill zone showing dozers in operation. (Middle Left) Secondary stockpile being added to from feeder(primary) stockpile via a conveyor belt along causeway. (Middle Right) Close view of dozer moving secondary stockpile of salt to hoppers for ship loading. (Bottom Left)Refueling bay area where spill occurred. (Bottom Right) Refueling hose and crystalline salt on the catchment drain [within the refueling bay area] where salt and fuel mixedprior to (overflow) release into rockfill area immediately below.

2 T.F. Guerin / Marine Pollution Bulletin xxx (2015) xxx–xxx

considered to be one of the most toxic oils to water column organ-isms, and direct mortalities are possible for fish, invertebrates andseaweed in direct contact with a spill (Hansen et al., 2013), as wellas algae and plankton (Romero-Lopez et al., 2012; Seuront, 2011).However, small spills in open water are very rapidly diluted and fishkills have not been reported in these scenarios. Bioremediationprocesses reduce the toxicity of diesel (Hohener et al., 1998;Molina-Barahona et al., 2005).

Mortality rates for seabirds can be high where populations areconcentrated (for example, during bird migrations), and shorebirdsfeeding in intertidal habitats are at risk of sub-lethal effects fromcontaminated or reduced prey populations. Direct contact withdiesel can affect marine birds by ingestion during preening andhypothermia from matted feathers, although the oil evaporatesso rapidly that the number of birds affected is usually small.Small spills have the potential for serious impact if they occur closeto a large nesting colony or are transported into an area of largebird population (Brown et al., 2011).

A very effective clean-up is possible for spills of light oil such asdiesel. Diesel is not sticky or viscous in comparison to heavier oils,

Please cite this article in press as: Guerin, T.F. Bioremediation of diesel from adx.doi.org/10.1016/j.marpolbul.2015.07.059

and tends to penetrate porous sediments and rocks on the shore-line but be washed off quickly by waves and tidal flushing, mean-ing that shoreline clean-up is usually not required. Completedegradation (of available fractions) by naturally occurringmicrobes takes one to several months.

The site on which the current spill study is focused is a smallisland off the coast of North West Western Australia which hasbeen joined by a causeway to another island, which in turn is alsoconnected to the mainland (Fig. 1). The region is unique withinAustralia in that it is the only marine environment adjacent to anarid tropical terrestrial environment. Rocky shores are the mostcommon habitat in the region. The coastline is largelyPrecambrian igneous rock, but in some areas there is an overlayof Pleistocene limestone (Jones, 2004).

2. Purpose

Remediation criteria for the pollution of the marine watersadjoining the spill site, was determined in consultation with theenvironmental regulator. The remediation criteria primarily was

rocky shoreline in an arid tropical climate. Mar. Pollut. Bull. (2015), http://



Table 2Sources of petroleum hydrocarbon contamination at the mining and ship loading site.

Activity and area Source Types ofcontaminants

Product handling and shiploading activities

Wastewater discharge,spills from bulk handlingequipment

Fuels, oils,equipment fluids(mixed with salt)

Wastewater dischargefrom conveyer transferpoint

Fuels, oils,equipment fluids,greases (mixedwith salt)

Vehicle and plantmaintenance workshop(on adjacent productionfacility)

Oily water separatordischarge

Fuels, oils, metals,detergent,equipment fluids

Refueling bays Oily water separatordischarge (includingproduct from spillevents)

Fuels, oils,equipment fluids

Bulk fuel storage Tank and line failure orfilling spill

Fuel (diesel andpetrol)

Vehicle wash-down bays Oily water separatoroverflow discharge

Fuels, oils, metals,detergent,equipment fluids

T.F. Guerin / Marine Pollution Bulletin xxx (2015) xxx–xxx 3

that there would be ‘‘no visible sheen of petroleum hydrocarbon onthe surface of the marine water body’’. The purpose of the studywas to design, construct and operate a process to remove the spiltdiesel entrapped in the rockfill.

3. Materials and methods

3.1. Site location and assessment

The site was located on a small island connected via a causewayto the mainland, on a rocky outcrop on the North West WesternAustralian coastline, used for loading mined product onto ships(Fig. 1). The principal product being mined was salt (sodium chlo-ride) and the operation was undergoing earthworks to expand theevaporation area of the mine, and to improve roads across the mineand port loading facilities (Table 1) at the time of the spill. The spilllocation was on the top of a rock fill area approximately 8 m abovethe low tide mark. The coastline was exposed to a large daily tidalmovement of approximately 5–6 m. During king tide events (1–2per year) the tidal variation increased to approximately 7–8 m.

3.2. Plant, equipment and spill source

A refueling bay (within concrete secondary containment) waslocated on the eastern edge of the island, on top of the rock fill area(Fig. 1). An above ground diesel storage tank (30 kL) was located inthe refueling area. During the peak of the construction expansionworks, the diesel tank was refilled from a land-based tanker fort-nightly. A grader and excavators (for road construction), and bull-dozers (for stockpile movement) were being refueled from thesame refueling bay. Table 2 describes the types of activities acrossthe entire operation that are potential sources of contamination,including refueling. Internal investigations (by the site owner)identified that the cause of the spill was the release of 7000 L ofdiesel during the refueling operation of a grader undertaking roadconstruction activity near the stockpile. It was noted that only at18–24 months post spill was petroleum hydrocarbons observedin the water in the lower intertidal area, and only a sheen (not freeproduct).

3.3. Shoreline sediment and rock fill sampling

After a consideration of the intractability of the shoreline andabove lying rock fill (from an initial site visit and detailed siteinspection), a purposive sampling plan was developed. This planrequired collection of samples from the area immediately belowthe spill location (or spill zone), a 40 m2 area including representa-tive samples from the upper, middle and lower intertidal zones(Fig. 2). Samples were collected in a directed (purposive) mannerso as to represent the size fraction of �50 mm, and to ensure themost visibly contaminated material was collected, reflecting thepractical objectives of this study (to ensure cost-effective remedi-ation of the diesel spill). This sampling (in the depth range of

Table 1Production of mineral salt at the nearby mine and bulk handling at the ship loadingstockpile and treatment site.a

Production parameter Value/description

Production mechanism Solar evaporationSalt production rates (annual) 4 million tonnesNo. of evaporation ponds 6Salt production (daily) 40,000 tonnesSalt harvesting frequency AnnualSalt washing by-product Gypsum

a Gypsum is used on site for road repairs and other uses.


0–15 cm) included small rocks, stones, sand, fines and the residualliquor. Liquor samples were collected to assess petroleum hydro-carbons as a check for the presence of a sheen and were assessedseparately to the solid samples. All samples were collected duringlow tide on each sampling day. Sediment and rock fill sampleswere collected from the intertidal area up to a distance of 200 maway from the center of the spill zone.

3.4. Petroleum hydrocarbon analysis

Once the objectives of the clean-up were determined (i.e. nosheen on marine waters and the source of contamination dispersedand removed), an analysis regime was developed. Confirmatory(fingerprinting) analysis showed that the spilled product was die-sel. The n-alkane/biomarkers ratios, C17/Pristane and C18/Phytane,were used to identify the extent of weathering and biodegradation,under the action of natural attenuation (for the period of0–112 weeks post spill), and throughout the treatment process,as has been adopted in other bioremediation assessmentsconducted by the author (Guerin, 2000). Since contaminationconcentrations immediately after the spill were not assessed, rep-resentative rock fill and sediment samples were saturated withdiesel and free drained, to estimate ‘diesel holding capacity’ (after1 week from dosing), using conventional instruments for deter-mining moisture and volumetric properties of soil (Guerin, 1993).Maximum petroleum hydrocarbon concentrations at week 0(i.e. at the time of the spill) were extrapolated from this assess-ment and represented an effective ‘‘field capacity’’ or free drainedconcentrations of diesel in samples of the contaminated rock fillimmediately under the spill zone.

3.5. Physico-chemical analysis

Due to the diesel contamination occurring in a marine environ-ment, with the added contribution of salt from an adjacent saltstockpile, electrical conductivity (EC) was considered an importantconstraint to the likely effectiveness of any microbiological treat-ment. Given that inorganic nutrients were likely to be limiting inthe intertidal zones, the macro nutrients for biological activity, N &P (in their available forms), were also assessed throughout the study.The analyses of pH, EC and available N & P were conducted on solidsand liquors using the conventional methods described previouslyand which provide explanatory methodology (Guerin, 1999a,b,




Fig. 2. Salt-imapcted, intractable rock fill, sediment and shoreline. (Top Left) Rockfill from an adjacent area without any fuel or salt contamination. (Top Right) Image showingcontamination source area at the top of rockfill area. (Middle Left) Immediately below source zone showing salt overflow onto rockfill from bulk handling of stockpile.(Middle Right) Middle intertidal zone showing fuel contamination. (Bottom Left) Lower intertidal zone showing boom in place to capture free product entering the marinewaters. (Bottom Right) A pre-spill photograph (no concrete refueling bay in place) used here to illustrate the treatment and monitoring zone of the spill including steeprockfill face (see the bottom right hand corner of this image).


2001). Particle size fractionation was also conducted using conven-tional fractionation techniques as described by the particle size anal-ysis was the Bouyoucos method described in Day (1953). Dissolvedoxygen was measured by an Orion A329 DO meter (Thrermofisher,Sydney Australia) and calibrated on a daily basis.

3.6. Microbial analysis

Total heterotrophic and petroleum hydrocarbon degradingmicrobial populations were determined using a most probablenumber (MPN) technique. An increase in the petroleum hydrocar-bon degrading population (using the MPN method) was consideredto be a key indicator of the success of any bioremediation strategyselected. This study deployed an enhanced method of petroleumhydrocarbon MPN which uses resazurin (dye) to make positive(i.e. growing) wells more clearly observable to the human eye,


overcoming the constraints associated with low turbidity, butactive degrading populations (Guerin et al., 2001).

3.7. Design of treatment methodology

A contamination model (or conceptual model of the contamina-tion at the site) was developed based on all the data available,including the limited subsurface data, and was used to informthe design and construction of the treatment strategy.Acceptance of the model by the site owner and governmentregulator meant that a non-invasive clean-up strategy, would beacceptable as a practical solution for the spill’s remediation. Thedesign would also need to prevent further discharges to sea,remove residual diesel from the rock fill and enhance naturalbiodegradation processes already present (albeit limited) in theshoreline rock fill and sediments.




Fig. 3. Establishment of the infiltration treatment system. (Top Left) Agricultural drain pipe entering into the base of the refueling bay for evenly distributing the aerated mixof surfactant, water and soluble nutrients to the contaminated zone in the rockfill immediately below. (Top right) Large 600L plastic tubs for mixing surfactant and solublenutrients used in the flushing process. (Bottom Left) View of the southernmost area of the refueling bay showing proximately to marine water. (Bottom Right) Surfactant andsoluble nutrient mix undergoing aeration to saturation prior to flushing events. This aeration was conducted 24–48 h prior to each flushing event (to ensure super-saturationwith oxygen).


The design involved shallow drilling (15–20 cm) and creation ofchannels through the concrete secondary containment where thediesel spill occurred, on top of the rock fill, and placing agricultural(perforated) drain pipe in place in the top of the subsurface, toallow the delivery of a mixture of surfactant and water solublenutrients to infiltrate through contaminated rock fill and the lowerintertidal sediments (Fig. 3).

3.8. Treatment implementation and operation

The flushing of surfactant and nutrients was initiated at week 112(i.e. just over 2 years after the initial spill event). A floating absorbentboom was put in place to minimize the escape of mobilized diesel(from the rock fill) to the sea. Surfactant and soluble N & P nutrientformula were also sprayed over the rocks and sediments in the inter-tidal zone with a knapsack, and in addition, MaxBac Slow Release(5–6 months) N & P granules (Scotts, Sydney Australia), were dis-tributed to all accessible areas of sediment, an area representingthe contaminated zone (Fig. 3) approximately 8 m � 25 m, as wellas the smaller spill zone (40 m2). Rock fill flushing was repeated atweeks 121, 131, 177 and 183, with shoreline spraying of thesurfactant and nutrient placement (soluble and granular N & Pformulations) at week 132, 173, 177 and 183. Treatments wereinterspersed with sampling and analyses to monitor progress ofdiesel removal in the rock fill and intertidal sediments.

4. Results and discussion

4.1. Assessment of the site and regional characteristics impacting theenvironmental sensitivity and nature of contamination and treatment

The geology and other geographic information on the site andthe surrounding region was reviewed in order to develop the


model for the site’s contamination and subsequent treatmentstrategy. The contaminated zone was represented by large igneousrocks and smaller rocks down to the size of silt (Fig. 2). The geologyand the severe climatic conditions are the most significant con-straint to establishing and implementing a successful remediationprogram at the site.

4.2. Development of the contamination model

The spill occurred along the coastline such that the diesel wasreleased slowly and intermittently into the lower sand and sedi-ment (at low tide) when high and king tides occurred, as observedby the presence of a slight sheen as seen by operators who wereworking at the contaminated site area (prior to treatment inter-vention) during the period of 18–24 months post-spill. The highand king tides caused the residual diesel in the rock fill to becomemobile, causing a distinct sheen in the lower lying marine water.This bleed or release behavior (of the entrapped diesel) can beexplained by the presence of perched diesel in free product form(Fig. 4). This slow release of diesel over the 2 years leading up tothe initiation of the remediation strategy was contributing to theongoing observation of a sheen on the rock fill/sediment surfaceat mid and low tide, as well as the presence of a sheen on themarine waters at the initiation of the remediation works and the23 weeks immediately after the surfactant flushing events(Figs. 4 and 5). Salt may also have had the effect of preservingthe diesel in situ for the two years where no degradation wasobserved, in addition to the oxygen-limited environment, andlow concentrations of available N & P.

A proportion of diesel (approximately 10% or 700 L) wasestimated to have moved through the rock fill all the way to thebottom of the rock fill, exiting into the sand and sediment at thelow tide mark into the ocean, as evidenced from the observations




Refuelling

Spill zone

Rocky outcropLow �de

Contaminated (vadose) zone

High �dePerched

diesel plume

Sheen

(A) Prac�cal exclusion zone

Refuelling Infiltra�on system ac�vated with surfactant and nutrients

(B)

Sheen

Refuelling

Diesel released, migrates toward

ocean

(C)

Soap/Foam Traces

Fig. 4. Model of diesel plume migration during treatment with surfactant andnutrients (A) diesel remains in situ in perched location in rock fill 2 years after spill.Minor sheen evident at the shoreline in the period 18–24 months post spill; (B)oxygenated solution of surfactant, N & P infiltrate into the source (perched diesel),dispersing the plume and making it more susceptible to wave, tidal and microbi-ological processes at 112–115 weeks post spill; (C) mobilized diesel furtherdisperses and biodegrades under influences of wave and tidal action both of whichin turn further increase aeration and aerobic biodegradation at 180 weeks post spill.


provided by the operators and environmental personnel on the sitewho had been monitoring the spill since it occurred. An additional10% (or 700 L) was estimated to have migrated directly into thesubsurface below the rock fill without migration into the ocean.A further 10% (or 700 L) was estimated to have dissipated intothe marine water and lost via evaporative processes. This left anestimated residual volume of approximately 4800 L in the rock filland sediment (including in any underlying perched diesel) andthroughout the vadose zone, including the amount that could have

0

5000

10000

15000

20000

25000

30000

0 20 40 60 80

Petr

oleu

m h

ydro

carb

ons (

mg/

kg)

Time a�er spi

(a)

Fig. 5. Diesel removal from rockfill after spill and surfactant infiltration. Three distinbiodegradation, and (c) slow, asymptotic decay. Arrows show dates of interventions.


been expected to have been sorbed onto the rock and sediment inthe contaminated zone.

4.3. An assessment of remedial options

The main options reviewed for the remediation were removaland treatment ex-situ, bioventing and sparging (combined), andflushing with a surfactant to enhance dispersion and biodegrada-tion. Removal and treatment ex-situ was eliminated because ofthe extremely high cost of excavation due to the extensivedistribution of diesel into the underlying rock fill. Bioventing andsparging were technically feasible options, both introducingoxygen to enhance the natural biodegradation processes in thesubsurface, however the main constraint was the costs of estab-lishing the required wells in a safe manner. Finally flushing usinga surfactant was considered the most technically feasible and costeffective as it allowed required oxygen to be introduced (via a sat-urated solution and by increasing the surface area of diesel held inthe rock fill and sediment as free product) (Khalladi et al., 2009;Lee et al., 2005; Mena et al., 2015), as well as dissolved nutrients,and surfactant to release the sorbed and entrapped diesel (Maoet al., 2015; Martienssen and Schirmer, 2007; Mulligan et al.,2001; Zhou et al., 2005). The flushing action would also transferthe sorbed diesel into the lower intertidal areas, exposing theentrapped diesel to further wave action and tidal activity, whichin time would increase aeration, which in turn increases biodegra-dation. Such an intervention would avert the need for extensiveearthworks and drilling which was unlikely to have been effectivein such a heterogeneous subsurface.

4.4. Physico-chemical characteristics of the rock fill and intertidalzones at start of treatment strategy

Numerous physico-chemical parameters, particularly those rel-evant to supporting biodegradation, were assessed as part of thetreatment program. These results are presented and summarizedin this section.

Analysis of macro nutrients required for biological degradationindicated their status was limiting, with measurements reporting<3.4 mg available N per kg and <0.8 mg available P per kg.Typically for biodegradation processes to be effective, these valuesshould be in the minimum range of at least 10–50 mg/kg, depend-ing on the corresponding quantities of petroleum hydrocarbon tobe biodegraded. The measured concentrations were reflective ofthe characteristics of an intertidal zone where soluble chemicalswere susceptible to leaching leaving little or no soluble nutrients.Ratios of C:N:P of 100:10:1 provide a useful rule of thumb for

100 120 140 160 180 200

ll (weeks)

Individual Samples

Sampling Time Means

(b)

(c)

ct degradation phases (a) slow natural attenuation post-spill, (b) fast enhanced




Table 3Summary of Microbiological Monitoring in sediment and rock fill solid material samples pre-Infiltration and post Infiltration with surfactant and nutrients.

Sample No. Petroleum degraders (cfu rangeper gram at weeks post spill)

Sample descriptions Biostimulationresponse

112 121 180

1 ? 3 104–105 107–108 105–106 Representative of high tide in the contamination zone; wide range of particle sizes; visiblycontaminated with diesel fuel at Week 112, 121, and 180 post-spill

Strong

4 104–105 108–109 104–105 Representative of mid-tide in the contamination zone; wide range of particle sizes; visiblycontaminated with diesel fuel at Week 112, 121, and 180 post-spill

Strong

5 104–105 109–1010 102–103 Representative of shoreline; predominantly sand and rocks to �20 mm; no contaminationevident (though sheen had been observed by site personnel in period 6–18 months post-spill

Strong

6 103–104 <102 <102 Background sample (taken 100 m away from source of contamination); predominantly sandand rocks to �20 mm

No response

7 102–103 <102 <102 Background sample (taken 150 m away from source of contamination); predominantly sandand rocks to �20 mm

No response

8 109–1010 108–109 109–1010 Background sample (taken 200 m away from source of contamination) + spiked with diesel oiland BioSolve; predominantly sand & rocks to �20 mm

Strong

(a) Biostimulation tests were conducted in the laboratory to determine response of samples to aeration and nutrient addition (80 ml diesel/500 ml of soil sample) and 1% (v/v)BioSolve solution.(b) ANOVA analyses showed that there were significant differences between Sample 8 and each of the other samples 1–7 at weeks 112 & 180.(c) The data ranges represented the data from 3 separate measurements. The standard errors were relatively high due the heterogeneous nature of the sediment particles ofthe means varied from 18% to 45%.(d) Scale of responses: None, weak, moderate and strong.


ensuring nutrients are not limiting, though wider ratios may stillallow an effective bioremediation process.

The pH of sediments was alkaline, reporting values between 8.5and 9.2. Biological degradation processes are usually optimized inthe neutral pH zone. It is likely that pH has limited biodegradationof the spilt diesel since the time of the spill. No further modifica-tion of the pH of the contaminated zone was attempted. The addi-tion of soluble nutrients had the effect of lowering pH toapproximately 7.5.

The EC varied markedly with the sediment particle size andtype, but were typically in the range 1100–5000 lS/cm. The rela-tively high EC values reflected the fact that stockpiling of the saltfor shipment had progressively lead to salt being pushed over ontothe rock fill as a result of normal loading operations, thus con-tributing to the already saline liquors in the intertidal zone.Highly saline environments are known to limit biodegradationhowever the adaptive potential of indigenous microorganisms ishigh are previously reported (Margesin and Schinner, 2001).

Oxygen in the liquors collected (prior to any flushing events wereinitiated), as measured by dissolved oxygen, was relatively low atranging from <2 to 5 ppm. The low levels of dissolved oxygen wereconsistent with the odors that emanated from the rock fill whichwas a sulfur-based smell of a ‘‘rotten egg gas’’ (hydrogen sulfide)smell. These odors of hydrogen sulfide and petroleum product ema-nating from the spill area indicated that anaerobic processes wereoccurring at the site, and were likely to be influencing the biodegra-dation processes degrading the residual diesel, at least to someextent.

Table 4Phases of the petroleum hydrocarbons removal in the contaminated rock fill spill model w

Phase of spill remediation Degradationrate (mg/kg/day)

Qualitativeremovalrate

Proposedmechanisms

Evid

Pre-treatment (0–2 years postspill)

0 Slow,asymptoticdecay

Naturalattenuation

Lowdegodo

Initial treatment interventions(surfactant, nutrientflushing)

170 Fast Enhancedaerobicbiodegradation

Higpetsho

Late treatment interventions(ongoing surfactant,nutrient flushing)

12 Slow,asymptoticdecay

Enhancedaerobicbiodegradation

Hig


4.5. Distribution of petroleum hydrocarbons across the intertidal zonesat start of treatment strategy

At week 112 post-spill, petroleum hydrocarbon concentrationsat the high tide mark were in the range of 15,000–27,500 mg/kg.These fractions were in the range of C10–C36, reflecting the compo-sition of slightly weathered diesel. These petroleum hydrocarbonconcentrations had not decreased over 112 weeks, indicating die-sel was still being released from the rock fill into the high tide zonesediments. The period from 0 to 112 weeks post-spill representedthe effect of natural attenuation processes in a predominantlyanaerobic (or at least an anoxic) environment.

4.6. Changes in degradation rates in response to addition of surfactantand macro-nutrients

The progress of petroleum hydrocarbon degradation over the190 weeks since the spill, is represented in Fig. 5. The degradationcurve followed a predictable tri-phasic pattern. This patternillustrates how the inherent (initial) slow rate of degradation,attributable to natural attenuation (under anaerobic/anoxic condi-tions), was improved by the treatment strategy (intervention) offlushing the contaminated zone with surfactant and nutrients.Microbiological changes in the sediment (from the rock fill areaand intertidal zone) provided verification that a biostimulationprocess was occurring as a result of the added surfactant andnutrients described in the following paragraphs.

ith perched diesel.

ence for proposed mechanism Biodegradationconditions

DO; absence of a measurable petroleum hydrocarbonradation rate; only traces of sheen observed on shoreline; H2Srs mixed with diesel

Anoxic,potentiallyanaerobic

h DO; high petroleum hydrocarbon degradation rate; highroleum hydrocarbon degrader populations; Distinct sheen onreline

Aerobicenhanced bysurfactant

h DO; high petroleum hydrocarbon degrader populations Aerobicenhanced bysurfactant




Table 5Petroleum hydrocarbon sheen and odors in the intertidal zone.

Observations in intertidal zone Weeks post-spill

0 115 122 169 180

Presence of sheen � ++ + � �Odor (H2S + diesel) ++ +++ + � �Soap discharge (foam) � � ++ + �

Table 6Licence conditions for the site relevant to petroleum hydrocarbon release andmanagement developed as a result of the spill.a

Regulated environmentalactivity

Specific site requirement

Pollution controlequipment

Be maintained and operated to manufacturer’sspecification or to an effective internalmanagement system

Hazardous goodsmanagement

All such chemicals to be stored in accordancewith the WA Code of Practice for Storage andHandling of Dangerous Goods

Spill management The Licensee shall immediately recover orremove and dispose of spills ofenvironmentally hazardous chemicals (if spilloccurs outside of an engineered secondarycontainment area)

Stormwater management The licensee shall implement all practicalmeasures to prevent stormwater run-off frombecoming contaminated by site activitiesThe licensee shall treat contaminated orpotentially contaminated stormwater asnecessary prior to being discharged (to theenvironment, offsite)Discharge limits for total petroleumhydrocarbons (US EPA 3550) is set at 15 mg/Lfor the designated wastewater dischargepoints

Information and training The licensee shall ensure that any person whoperforms tasks on the site is informed of all theconditions of the licence that relate to the taskswhich that person is performing

Auditing, reporting andgovernmentnotifications

The Licensee shall complete an annualcompliance audit indicating the extent towhich the licence conditions have been metincludingContaminated sites investigations underwayand related monitoring or remediation[treatment] works underway (during theprevious 12 month period)Breaches of any limits specified in the licenceor failure of any pollution control equipmentor any incident that may cause pollution

a There were no discharge limits set on contaminants entering groundwater.


In the initial phase, the total heterotrophs, reflective of theintrinsic total aerobic microflora in the sediment and rock fill2 years after the spill, and petroleum hydrocarbon degraders, wereestimated to be in the order of 104–105 cfu per gram at 112 weekspost-spill, that is, before the remediation strategy was imple-mented. It was noted that there was no significant differencebetween the total heterotrophic and petroleum hydrocarbon degra-ders that were enumerated. This range (i.e. 104–105 cfu per gram) islow for soil and sediments that are contaminated with petroleumhydrocarbons in aerobic environments (Guerin, 1999c; Guerinet al., 2001) this nevertheless reflects a positive presence for petro-leum hydrocarbon biodegradation capability in the intertidal zonesamples tested. The low oxygen levels in the rock fill are consistentwith relatively low numbers of aerobic microflora. It appears fromthe degradation rate in this initial phase, where natural processeswere the predominant ones at work, that the rate was severely lim-ited by the depleted nutrient status of the sediments, including oxy-gen, and their near saturation with contaminant.

During the second stage of degradation of the diesel (Fig. 5), therate was considerably higher, mostly due to dispersal, coupledwith enhanced biodegradation. This rate was calculated at170 mg/kg/day. This is fast when considered relative to petroleumhydrocarbon biodegradation on other projects such as in bioventedprocesses where rates can reach 250 mg/kg/day or more. The sur-factant is likely to have successfully dispersed the petroleumhydrocarbons more uniformly throughout the surrounding sedi-ments, making them more available to the weathering action ofthe waves and rising (and falling) tides and intrinsic microbial pop-ulations, stimulated by greater nutrient availability and oxygen.This second and most rapid degradation stage was mostly due todispersal processes that released entrapped diesel to undergoenhanced degradation in the intertidal zone. The surfactant suc-cessfully dispersed the petroleum hydrocarbons more uniformlythroughout the surrounding sediments, making the petroleumhydrocarbons more available to the weathering action of the risingtides and intrinsic microbial populations, stimulated by greaternutrient availability and oxygen. The microbial count at this stagewas increased up to 106–107 and 108–109 cfu per gram of sediment(solids), respectively for total heterotrophs and petroleum hydro-carbon degraders (Table 3). These changes indicated, when corrob-orated with other parameters assessed in this study including DOreadings, changes in pristine and phytane ratios, and high contam-inant degradation rates, that a very effective biostimulation pro-cess had occurred (Tables 4). The dispersive influence of thesurfactant was verified by the presence of traces of soap (foam)on the low tide mark, beginning several weeks after the first flush(of surfactant) (Table 5).

The biomarker ratios changed during the treatment processfrom 2–3:1 in samples from 112 weeks post-spill, to <1:1 at week121 and 165, further indicating a biological process was occurring.

The third stage, between week 121 and 165 post spill, exhibiteda slow removal rate of 12 mg/kg/day. No treatments occurredbetween week 132 and 173 due to a labor shortage (because ofthe remote location) to undertake the flushing or sample collec-tion. Activity between weeks 173 and 183, featured further rock fillflushing with surfactant, leading to mobilization of diesel by,which further enhanced biodegradation.


After 170–190 weeks post-spill, petroleum hydrocarbon con-centrations reduced to 23–190 mg/kg, with many of the samplingsites reaching ‘not detected’ limits in the sediment. The remainingpetroleum hydrocarbons were determined to be within the C15–C36

n-alkane fraction, the least biodegradable fraction of diesel, andwas typical of a residual quantity remaining after an effectivebioremediation process.

4.7. Review of the effectiveness of the treatment

Table 5 summarizes the changes in the intertidal zone duringthe course of treatment. There was a clear biostimulation responseafter the flushing intervention reflected in the increase in thenumber of petroleum hydrocarbon degraders and in the reductionof petroleum hydrocarbon concentrations in the rock fill andsediment in the intertidal zone. The regulatory requirements toeliminate the sheen on the marine waters at the shoreline andremove the source of ongoing diesel were met through the appliedinterventions. There was evidence of a reduced sheen observedafter the second flushing event and the diesel had been flushedfrom the upper rock fill areas.

4.8. Remediation cost analysis

Tight budgetary constraints at the operation precluded readyavailability of funds to remove the contaminated rock fill, andany residual sources, and treat these materials ex-situ. Such anextensive excavation which would have to be conducted using civil





earthworks plant and equipment including a specialized long armexcavator, over several weeks, which would have cost up to anestimated AUD500,000, excluding any surfactant, nutrients, ormonitoring (and analyses). Comparison with other researcherssuggests remediation costs can vary from at least $50–200/t for soilwashing and or flushing. The total labor and materials cost for theclean-up of this contaminated zone and intertidal shoreline, was inthe order of AUD25,000 and as such, this biostimulation strategywas a cost effective solution to what would otherwise have beena major disruption to the refueling bay and large expense itemfor the operation.

5. Conclusions

The clean-up of the spill was achieved with evidence that bios-timulation was a key underlying process responsible for dieseldegradation. A pronounced increase in the petroleum degradingmicroflora, as a result of the flushing of surfactant, nutrients, andincreased oxygen, verified that the biodegradation processes hadbeen increased in the contaminated rock fill and intertidal zone.A demonstration of the evidence for biostimulation from the flush-ing treatment interventions and reduction in petroleum hydrocar-bons, met the regulator’s requirements for the clean-up endpoints.A modified license was subsequently put in place that now pro-vides greater protection for this sensitive tidal marine environment(Table 6).

The selection of remediation technology and treatment designwas fit for purpose with mine production and ship loading ableto continue during the entire treatment period. The practical,cost-effective treatment design also kept costs to a minimum,making the clean-up project acceptable to the management teamresponsible for meeting the legal obligations for the site. Thedesign also allowed for a minimum exposure of unsafe conditionsand behaviors to site personnel involved in the process or workingin the area, as an intrusive civil earthworks program was avoided.

This practical, surfactant-enhanced biostimulation approachcan be applied to other intractable spill sites where access to thespill and contaminated zone is restricted and unsafe for traffickingwith excavation and earthworks equipment.

References

Anon., 2011. Dampier Marine Services Facility Assessment on ProponentInformation WorleyParsons Services Pty Ltd, p. 250.

Brown, C.E., Fieldhouse, B., Lumley, T.C., Lambert, P., Hollebone, B.P., 2011.Environment Canada’s methods for assessing oil spill treating agents. In:Fingas, M. (Ed.), Oil Spill Science and Technology. Gulf Professional Publishing,Boston, pp. 643–671 (Chapter 19).

Colombo, J.C., Barreda, A., Bilos, C., Cappelletti, N., Demichelis, S., Lombardi, P.,Migoya, M.C., Skorupka, C., Suárez, G., 2005. Oil spill in the Rio de la Plataestuary, Argentina: 1. Biogeochemical assessment of waters, sediments, soilsand biota. Environ. Pollut. 134, 277–289.

Day, P.R., 1953. Experimental confirmation of hydrometer theory. Soil Sci. 74, 181–186.

Guerin, T.F., 1993. The Relative Significance of Biodegradation and Physico-Chemical Dissipation of Endosulfan from Water and Soil. Department ofAgricultural Chemistry and Soil Science, University of Sydney, Sydney, NewSouth Wales, Australia, p. 240.

Guerin, T.F., 1999a. The anaerobic degradation of endosulfan by indigenousmicroorganisms from low-oxygen soils and sediments. Environ. Pollut. 106,13–21.

Guerin, T.F., 1999b. Bioremediation of phenols and polycyclic aromatichydrocarbons in creosote contaminated soil using ex-situ landtreatment. J.Hazard. Mater. 65, 305–315.


Guerin, T.F., 1999c. A comparison of mesophilic composting and unamended landtreatment for the bioremediation of aged tar residues in soil. Remediation 9,59–77.

Guerin, T.F., 2000. Long term performance of a land treatment facility forbioremediation of non-volatile oily wastes. Resour., Conservat. Recycl. 28,105–120.

Guerin, T.F., 2001. Co-composting of residual fuel contamination in soil. Soil Sedim.Contam. 10, 659–673.

Guerin, T.F., 2002. Heavy equipment maintenance wastes and environmentalmanagement in the mining industry. J. Environ. Manage. 66, 185–199.

Guerin, T.F., 2009. Improving environmental performance in the minerals supplychain using a life-cycle approach: the role of fuel and lubricant suppliers inenabling sustainable development. In: Richards, J.P. (Ed.), Mining, Society, and aSustainable World. Springer-Verlag, Berlin, Heidelberg, pp. 205–250.

Guerin, T.F., 2014. Root causes of fluid spills from earthmoving plant andequipment: implications for reducing environmental and safety impacts. Eng.Fail. Anal. 45, 128–141.

Guerin, T.F., Mondido, M., McClenn, B., Peasley, B., 2001. Application of resazurin forestimating abundance of contaminant-degrading micro-organisms. Lett. Appl.Microbiol. 32, 340–345.

Hansen, B.H., Altin, D., Overjordet, I.B., Jager, T., Nordtug, T., 2013. Acute exposure ofwater soluble fractions of marine diesel on Arctic Calanus glacialis and borealCalanus finmarchicus: effects on survival and biomarker response. Sci. TotalEnviron. 449, 276–284.

Hewstone, R.K., 1994. Health, safety and environmental aspects of used crankcaselubricating oils. Sci. Total Environ. 156, 255–268.

Hohener, P., Hunkeler, D., Hess, A., Bregnard, T., Zeyer, J., 1998. Methodology for theevaluation of engineered in-situ bioremediation: lessons from a case study. J.Microbiol. Methods 32, 179–191.

Ismail, Z., Karim, R., 2013. Some technical aspects of spills in the transportation ofpetroleum materials by tankers. Saf. Sci. 51, 202–208.

Jones, D., 2004. The Burrup Peninsula and Dampier Archipelago, Western Australia:an introduction to the history of its discovery and study, marine habitats andtheir flora and fauna. Rec. West. Austral. Mus. Supplement 66, 27–49.

Khalladi, R., Benhabiles, O., Bentahar, F., Moulai-Mostefa, N., 2009. Surfactantremediation of diesel fuel polluted soil. J. Hazard. Mater. 164, 1179–1184.

Lee, M., Kang, H., Do, W., 2005. Application of nonionic surfactant-enhanced in situflushing to a diesel contaminated site. Water Res. (Oxford) 39, 139–146.

Mao, X., Jiang, R., Xiao, W., Yu, J., 2015. Use of surfactants for the remediation ofcontaminated soils: a review. J. Hazard. Mater. 285C, 419–435.

Margesin, R., Schinner, F., 2001. Biodegradation and bioremediation ofhydrocarbons in extreme environments. Appl. Microbiol. Biotechnol. 56, 650–663.

Martienssen, M., Schirmer, M., 2007. Use of surfactants to improve the biologicaldegradation of petroleum hydrocarbons in a field site study. Environ. Technol.28, 573–582.

Mena, E., Ruiz, C., Villaseñor, J., Rodrigo, M.A., Cañizares, P., 2015. Biologicalpermeable reactive barriers coupled with electrokinetic soil flushing for thetreatment of diesel-polluted clay soil. J. Hazard. Mater. 283, 131–139.

Mercurio, P., Burns, K., Negri, A., 2004. Testing the ecotoxicology of vegetable versusmineral based lubricating oils: 1. Degradation rates using tropical marinemicrobes. Environ. Pollut. 129, 165–173.

Molina-Barahona, L., Vega-Loyo, L., Guerrero, M., Ramírez, S., Romero, I., Vega-Jarquín, C., Albores, A., 2005. Ecotoxicological evaluation of diesel-contaminatedsoil before and after a bioremediation process. Environ. Toxicol. 20, 100–109.

Mulligan, C.N., Yong, R.N., Gibbs, B.F., 2001. Surfactant-enhanced remediation ofcontaminated soil: a review. Eng. Geol. 60, 371–380.

Olagbende, O.T., Ede, G.O., Inyang, L.E.D., Gundlach, E.R., Gilfillan, E.S., Page, D.S.,1999. Scientific and cleanup response to the IDOHO-QIT oil spill, Nigeria.Environ. Technol. 20, 1213–1222.

Reed, M.D., Blair, L.F., Burling, K., Daly, I., Gigliotti, A.P., Gudi, R., Mercieca, M.D.,McDonald, J.D., Naas, D.J., O’Callaghan, J.P., Seilkop, S.K., Ronsko, N.L., Wagner,V.O., Kraska, R.C., 2005. Health effects of subchronic exposure to diesel-wateremulsion emission. Inhal. Toxicol. 17, 851–870.

Romero-Lopez, J., Lopez-Rodas, V., Costas, E., 2012. Estimating the capability ofmicroalgae to physiological acclimatization and genetic adaptation topetroleum and diesel oil contamination. Aquat. Toxicol. 124–125, 227–237.

Seuront, L., 2011. Hydrocarbon contamination decreases mating success in a marineplanktonic copepod. PloS One 6, e26283.

Smith, A.L., 1995. Securing the Future – in Partnership with the Environment.Offshore Eur Conf Proc, Society of Petroleum Engineers (SPE), Richardson, TX,USA, pp. 639–653.

Vazquez-Duhalt, R., 1989. Environmental impact of used motor oil. Sci. TotalEnviron. 79, 1–23.

Zhou, Q., Sun, F., Liu, R., 2005. Joint chemical flushing of soils contaminated withpetroleum hydrocarbons. Environ. Int. 31, 835–839.


http://refhub.elsevier.com/S0025-326X(15)00492-0/h0010
































































































