Tidal Flat and Sand Beach Remediation: Choosing Remediation Techniques to Speed Ecological Recovery of Habitats Still Impacted 20 Years after the Gulf War Oil Spill

During the First Gulf War an estimated 10 million barrels of crude oil were released into the Arabian Gulf (Tawfiq and Olsen 1993) with an estimated 0.9-1.9 million barrels of oil deposited intertidally (Jones et al. 2008) impacting approximately 800 kilometers (km) of coastline along the Eastern Province of the Kingdom of Saudi Arabia. In 2002-2003 a comprehensive oiled shoreline survey (OSS) was conducted to support claims for damages to the coast environment (Pandion Technology Ltd. and RPI 2003). Figure 1 depicts the southern oiling extent identified in 2003. Based on these surveys, it was determined that nearly 8,100,000 m³ of sediments were oiled as a result of the spill to some degree: nearly 45 % of oiled sediments were identified within sheltered muddy tidal flats, 23 % within marshes, 26 % within sandy tidal flats, and 11 % on sand beaches. The focus of this paper is on sheltered muddy tidal flats, sandy exposed tidal flats, and sand beaches still impacted more than 20 years after the Gulf War oil spill.

Damages to coastal habitats were a direct result of oil smothering (Barth 2003) (Jones et al. 1998), deep penetration into the sediment via crab burrows (Barth 1995), and limited spill response. Watt (1996) described a range of cleanup techniques used during the response, noted their ineffectiveness in soft sediments environments, and concluded that natural recovery was limited. After the initial die-off, tidal flats were covered by an expansion of algal mats that prevented re-colonization, trapped liquid or emulsified oil within sediments or directly beneath, and disrupted hydrology (Barth 2003) (Al-Thukair & Al-Hinai 1993). As a result, sheltered muddy and exposed sandy tidal flats were targeted for remediation activities to remove stressors and encourage ecological recovery.

Twenty years after the spill, the Coastal Ecosystem Restoration and Remediation Program (CE-RRP) was established by the Presidency of Meteorology and Environment (PME) of the Kingdom of Saudi Arabia (KSA) under the guidance of the United Nations Compensation Commission (UNCC) and the panel of independent reviewers. The CE-RRP aims to restore the integrity and heath of intertidal ecosystems. During 2009-2011 a validation survey was conducted to identify where oiling has persisted, to what degree, and where ecological recovery has been slow. In addition, ecological stressors inhibiting recovery were identified.

The three major stressors identified as limiting ecological recovery within sheltered and exposed tidal flats are as follows:

1. 1)

   Oiled sediments and physical oil barriers;

2. 2)

   Laminated algal mat; and

3. 3)

   disrupted hydrology

By removing stressors, CE-RRP aimed to achieve the following three goals:

1. 1)

   Increase suitable habitat for grazers and burrowing infauna;

2. 2)

   Reduce total petroleum hydrocarbon levels (TPH); and

3. 3)

   Improve physical processes

The purpose of this paper is to present remediation techniques utilized in tidal flats and sand beaches 20 years after the Gulf War oil spill. Results of each technique are presented, as well as short-term ecological responses.

Between 2011 and 2013, fifteen large-scale (238 million USD) coastal remediation contracts targeted over 2,100 ha of impacted coastline. Marsh habitats made up the majority of the work, as the marsh habitats were deemed highest priority during initial planning. Following the concept of landscape connectivity many of the adjacent habitats were also targeted for remediation including three contracts within sheltered and exposed tidal flats and sand beaches along the eastern province of Saudi Arabia (Figure 2). The CE-RRP identified and utilized three oil removal techniques within these impacted environments in order to achieve goals 1-3 listed above. Additional coastal remediation contracts (approximately 72 million USD budgeted) will be implemented in spring 2014 using lessons learned from these projects. Techniques are as follows:

1. 1)

   Dry tilling of oiled sediments using tines (rake) or disc harrow attachment, followed by manual removal of remaining surface oil;

2. 2)

   Complete physical removal and disposal of the surface or cohesive subsurface oiling layers; and

3. 3)

   Submerged tilling contained by physical barriers to liberate liquid oil, followed by skimming/vacuum pumping.

Dry tilling on sheltered tidal flats was identified as a method of remediation to reduce stressors when there was a well-defined gradient between hardened oiled residue (crust) and unoiled substrate, and when laminated algal mat exists on the surface of the cohesive oiled layer. Sediments within sheltered and exposed tidal flats exhibiting these oiling characteristics tend to be sandy with greater water permeability, thus more stable than sheltered muddy tidal flats. Machinery such as a tractor with a rake attachment (Figure 3a) or walk-behind tiller can maneuver the surface with ease.

Sixty-five hectares (ha) of exposed and sheltered tidal flats were dry tilled using a combination of mechanical tilling means to a depth of 30 centimeters (cm). These include: walk-behind tiller, tractor with a rake or tine attachment, or a tractor with a disc harrow. Contractors executing the activity decided on mechanical methods based on cost, size of tilling area, site accessibility, and sediment stability. In some cases, sediments would not support machinery and equipment modifications were developed. Figure 3b illustrates a modified walk-behind tiller. This modification utilized concepts from rice paddy cultivation to increase weight distribution across the soft sediments.

The endpoint for dry tilling was the complete breakup of all barriers (algal mat and hardened oil layers). In addition, any compaction or depressions as a result of the activity was corrected so that new stressors were not developed. In most cases, an area required multiple passes with a tiller or rake to break up all barriers.

Complete removal of oiled sediments was considered when a dry cohesive surface oiling layer or dry cohesive subsurface layer beneath a layer of clean sand was identified within sandy tidal flats, or along sand beach shorelines. The CE-RRP targeted 3.6 km of oiled shoreline for complete removal of oiled sediments by a combination of manual and mechanical methods. Delineation of surface oiling was straightforward using visual cues; however, buried cohesive oiled sediments were identified using observation pits. Once a length, width, and depth of oiling were identified along a given shoreline, removal methods were chosen. Depth of oiling tended to be the limiting factor when deciding between manual and mechanical removal techniques. Where oiling was greater than 20 cm thick, or too cohesive to break-up with a shovel, an excavator was used. Conservation of clean sediment was a priority, thus, manual techniques were more efficient when only a thin layer existed. Clean sediments (overburden) were side cast to access buried, cohesive oiled sediments. Side-casts were returned to the site after removal of the contaminated material. In most cases, a combination of mechanical and manual removal methods were utilized. Multiple laborers equipped with shovels and rakes worked an area following mechanical removal to increase effectiveness.

Verification of this oil removal technique required both visual surveys and analytical analysis. Following the removal activity, surface and subsurface sediments must meet a “no oil observed” (NOO) endpoint. The subsurface sediments were assessed using a density of three subsurface observation pits per 1,000 m² worked. Once visual endpoints were achieved, one composite sample (three sub-samples) per 1000 m² was collected and sent to the program laboratory for total petroleum hydrocarbon (TPH) analysis. TPH was measured by gravimetric analysis based on method 5520G of the 29^th edition of standard methods. If the sample contained less than 1,500 parts per million (ppm) TPH, then treatment of the area was considered complete (i.e. approaching background oiling levels).

All areas designated for complete removal were tilled and re-graded so that topographic lows and compacted sediments from operations were avoided. Due to the volume of oiled sediments removed in some areas, replacement with clean sand (of similar grain-sized sediments) was required in order to avoid leaving any depressions.

Submerged tilling was assigned to sheltered and exposed tidal flats exhibiting a high level of subsurface cohesive and liquid oiling. Laminated algal mat was also prevalent throughout these impacted environments. Grain size within sheltered tidal flats ranged from muddy to sandy. Muddy flats tended to be very soft, and as expected, they proved to be an operational challenge to work with heavy machinery.

Submerged tilling was achieved by flooding a contained area, either artificially (by pumping) or via tides, followed by towing a rake/disc harrow or running a power roto-tiller through oiled sediment. The tilling exposed subsurface sediments to water. Liquid oil within the pore spaces was liberated, and floated to the surface.

Best results were achieved when the liquid oil was vacuumed or skimmed off the surface of the water immediately prior to draining the contained area (Figure 4). Industrial vacuum trucks were utilized for large-scale tilling efforts while manual skimming was used in smaller areas. Containment (hard) booms surrounded all tilling areas to prevent any oil release during a high tide. Passive oil management (sorbent booms) was utilized in all areas. If vacuum trucks or skimmers could not be procured, then teams of laborers equipped with sorbet pads, buckets and shovels removed surface oiling. All oiled waste was disposed of or reused in compliance with environmental regulations in Saudi Arabia.

Degrees of subsurface liquid oiling varied across project areas. Sediments that were heavily oiled required multiple events of submerged tilling. One event was defined as a complete pass perpendicular to shore and then parallel to shore, while contained and flooded by at least 20 cm of water. Oil liberation was maximized when oiled sediments were completely submerged. Rake or disc harrow attachments often lifted subsurface sediments above the surface by a maximum of 15 cm. Each area (usually 1 ha) was contained either by sand bags or berms at least 50 cm above the surface. Contained areas were filled with water artificially via pumps, or by trapping water during an ebbing tide. Subsequent removal of liberated oil from the surface completed an event. Best results were achieved when an area containing moderately to heavily oiled sediments was treated multiple times. The remediation endpoint set by the CE-RRP required four tilling events.

The selection of machinery used to pull rakes or disc harrows depended on the grain size of sediments. Muddy, or finer, oiled sediments, required a track vehicle to stay above the surface. In some areas only smaller power roto-tillers were able to function in the soft sediments using a team of laborers to keep it moving. Multiple vehicle prototypes were tested. Wheeled vehicles tended to sink when working in soft sediments and could not be used. Figure 5 depicts the most effective track vehicle for maneuvering within soft sediments. Wheeled vehicles were utilized within sandy tidal flats with heavily oiled sediments. Figure 6 illustrates wheeled vehicles and rake attachments utilized within sandy, oiled tidal flats.

In all, 65 ha of impacted tidal flats were remediated via dry tilling. In order to meet remediation endpoints, all three stressors must be eliminated. Barriers in the form of cohesive oiled sediments and laminated algal mat, were physically broken up during the tilling activity, and thus, eliminated as a stressor to the environment (Figure 7). Persistent ponding was eliminated as a stressor during de-compaction/re-grading, which was an essential process associated with the dry tilling technique.

Laminated algal mat or a thin layer of accreted clean sand covered the surface of most dry tilling areas (Figure 8a). Dry tilling initially resulted in an average 20% cohesive surface oiling cover (Figure 8b). Rakes or disc harrows broke up and exposed subsurface cohesive oiled sediments to the surface. Cohesive surface oiling pieces were termed surface residue balls (SRBs) (Michel et al. 2013). Individual SRBs ranged from 1.0-15 cm in width and averaged 2 cm in thickness. Thus, an average SRB volume of 40 m³ per hectare resulted. Production of SRBs as a result of dry tilling was identified during a small-scale demonstration contract and lessoned learned were applied to larger scale efforts. Removal of all SRBs greater than 2.5 cm was required from all subsequent remediation contracts conducting dry tilling. Removal efforts were manual, and consistent with complete removal methods described above. A total of 2,600 m³ of SRBs were removed as a result of dry tilling activities. Percent barrier was calculated using percent surface cover estimates of algal mat and oiling from ten .25m x .25m quadrats within thirty-two 10 x 10m plots (n=320).

Dry tilling reduced TPH from subsurface sediments significantly as shown in Figure 9. The reduction of TPH was a result of the following:

1. Physical removal (removal of SRBs);

2. Mixing cohesive subsurface oiled sediments with clean sediments (dilution);

3. Exposure of oiled sediments to the sun and oxygen; and

4. Liberation (and mechanical recovery) of any liquid oil trapped within subsurface sediments during high tide.

In all, 45 ha of exposed oiled tidal flat and sand beach habitats were remediated using the complete removal technique, with 40,000 m³ of oiled sediments along 3.6 km of shoreline removed and, disposed of, or reused according to the CE-RRP waste management plan. Verification of this activity required that all three stressors be eliminated. Barriers, in the form of cohesive oiled sediments, were physically removed, and thus, eliminated as a stressor. Persistent ponding was eliminated as a stressor during the de-compaction/re-grading process that was associated with complete removal. Sediment replacement by clean sediments was required within areas of deep excavation in order to increase elevations for re-grading and to avoid relapsing ponding.

To date, 450 sediment samples taken from within removal areas have been analyzed for TPH. All results were less than 1,500 ppm TPH (average 114 ppm), which was considered as “background level”. Results of TPH analysis conducted on in situ asphalt pavement (AP) and subsurface cohesive oiled sediment ranged up to 45,000 ppm prior to excavation. All areas met NOO endpoints on the surface and subsurface. Figure10 depicts before and after surface and sub-surface conditions.

In all, 45 ha of sheltered and exposed tidal flats have been remediated using the submerged tilling technique. During liberation, liquid oil mixes with water, air and sediments and becomes emulsified. Total volume of emulsified oil collected via vacuum amounts to 1,025 m³ , thus, a recovery rate of 26 m³/ha. A sample of the liberated oil was analyzed for TPH. The sample had 234,940 ppm TPH, or 23.5% oil. Total ‘oil’ collected amounted to 240 m³ at a rate of 6.5 m³/ha or 40 barrels/ha. Thus a total of 1,480 barrels of hydrocarbons were collected. In 2014, the CE-RRP will remediate an additional 100 hectares of impacted sheltered tidal flat via submerged tilling. Using similar methods, 2,562 m³ of liberated emulsified oil are expected to be released from the sediments.

Verification of this activity required that all three identified stressors be eliminated. Liquid oil was released and actively collected. Barriers in the form of cohesive oiled sediments and algal mat, were removed via tilling. Persistent ponding was eliminated as a stressor during the decompaction/regarding process that was associated with the submerged tilling technique. Figure 11 illustrates that subsurface oiling concentrations were significantly reduced following submerged tilling.

While oil liberation and removal was high, a level of oiling still existed within sediments following four submerged tilling events. However, analytical results showed that oil concentrations were significantly reduced. This trend is consistent with visual pit observations before and after remediation. A visual comparison of pit oiling conditions from seventy-three pits across thirty-five hectares indicates a reduction from heavily oiled residues (HOR) to moderately oiled residues (MOR), HOR to lightly oiled residues (LOR) and MOR to LOR (Figure 12). Descriptions of these oiling conditions can be found within the Oiled shoreline survey in support of the marine and coastal damage assessment (Pandion Technology Ltd. and RPI, 2003)., In all instances, a reduction in visual oiling conditions was observed.

Table 1 presents monitoring metrics from: 1) before submerged tilling; 2) after submerged tilling; 3) a comparison tidal flat (reference); and 4) an impacted tidal flat (set aside). Monitoring metrics include two stressors, percent algal mat and average TPH, and one recovery metric, average abundance. Stressor metrics post-remediation are lower than those pre-remediation and at the impacted site. TPH after remediation is significantly lower than before remediation based on a t-Test: Two-sample assuming equal variances (P= 9.8E-11). In addition, post-remediation stressor metrics are similar to the comparison tidal flat stressor metrics. Total species abundance, an ecological recovery metric, is greater following remediation. Species abundance was measured across nine 10 x 10m monitoring plots in the lower, middle and upper stratum. Within each plot ten .25m x .25m quadrats were thrown and individual species were counted.

Dry tilling was most effective at meeting goals when there was a well-defined boundary between hardened oiled residue (crust or AP) and lightly-oiled substrate. Dry tilling also limited sediment disposal, which may be a priority if budgets are limited and disposal costs are high. Unnecessary sediment removal can be a stressor if sediment in the area either is not replenished, or is re-nourished using sediments of differing grain sizes.

Sixty-five hectares of sheltered and exposed tidal flats were dry tilled, met contract endpoints, and successfully achieved all 3 program goals. Goal 1: increase suitable habitat for grazers and burrowing infauna, was met by removing barriers for burrowing infauna or grazers. Insect, amphipods, crabs and snails have space to colonize and the extent to which this happens is being monitored. Goal 2: Reduce TPH, was achieved by exposing oiled sediments to aerobic conditions, dilution, and limited physical removal. Oil toxicity reduction and degradation is expected over time, and will be indentified using biomarker and polycyclic aromatic hydrocarbon (PAH) analyses. Goal 3: Improve physical processes and reduce associated stressors: was achieved via de-compaction and re-grading to restore hydrology and reduce persistent ponding that encourages algal mat growth.

It is important to consider subsurface oiling conditions prior to assigning a dry till treatment. Dry tilling is not an appropriate method when medium to heavy subsurface liquid oiling exists (MOR/HOR). Uncontained releases of liquid oil may persist for weeks or liquid oil may be buried deeper within sediments. During scoping, if MOR or HOR exists within observation pits in a tidal flat, submerged tilling should be considered, because dry tilling will not achieve all three goals. If dry tilling is conducted within tidal flats containing liquid oil, adequate oil containment must be provided over the period of release. In addition, a level of subsurface oiling should be expected. Nonetheless, dry tilling does remove barriers and aerate oiled sediment, which is essential for oil degradation (Rhykerd et al. 1999). Thus, dry tilling can be a cost effective tool to achieve goals and facilitate long-term degradation of exposed oiled sediments which can promote ecological recovery.

Complete removal was most effective at meeting goals when there was a dry cohesive surface oiling layer or dry cohesive subsurface layer beneath a layer of clean sand, within sandy tidal flat or exposed sand beaches, and when sediment conservation was not a priority. Forty five hectares of tidal flats and sand beaches were remediated by complete removal and successfully achieved all three program goals.

Goal 1 was met by completely removing barriers for burrowing infauna or grazers. Insect, amphipods, crabs and snails have space to colonize. Goal 2 was achieved by completely removing all visible surface and subsurface oiling. Goal 3 was achieved via de-compaction and re-grading to restore hydrology and reduce persistent ponding which aids algal mat growth. In areas with thick layers of cohesive subsurface oiling, sediment replacement with clean sediments was required to meet goal 3. Re-nourishment must ensure the donor material is of comparable grain size and does not lead to excessive sediment/silt deposition.

It is important to consider subsurface oiling conditions prior to selecting complete removal as a remediation technique. Complete removal requires excavation, transportation, and disposal / treatment of large quantities of contaminated material followed by re-grading and possible replacement of material; all of which increase the cost. Therefore, complete removal may not be an appropriate method when subsurface liquid oiling is patchy or has penetrated deep into the sediments where removal will result in deep excavations and large volumes of sediment. It is most appropriate where removal of only the highly contaminated (discrete) layers of sediment can be targeted, either at the surface, or where the uncontaminated overburden can be side-cast and managed successfully for replacement after removal of the oiled layers.

During scoping, if moderately or heavily oiled sediment exists below an asphalt pavement, crust, or cohesive subsurface layers, a combination of techniques may be required to meet goals. In this case, the CE-RRP recommends complete removal of AP and subsurface cohesive oiling followed by submerged tilling of moderately or heavily oiled sediment below.

Submerged tilling was most effective at meeting goals when there was a high level of subsurface liquid oiling within tidal flats, and if sediment conservation was a priority. Forty hectares of tidal flats were remediated by submerged tilling and successfully achieved all three program goals.

Goal 1 was met by completely removing barriers for burrowing infauna or grazers. Insect, amphipods, crabs and snails now have space to colonize. Goal 2 was achieved by liberating and removing liquid oil. Goal 3 was achieved via de-compaction and re-grading to restore hydrology and reduce persistent ponding which aids algal mat growth.

Complete removal produced approximately 295,000 m3 sediment removed across 2,100 ha of intertidal habitat; this material required off-site disposal. Dry tilling removed approximately 40 m3 cohesive oil per ha, and reduced surface/sub-surface oiling by an average of 10,000 ppm TPH. Submerged tilling removed approximately 1025 m3 emulsified oil per ha, and reduced sub-surface oiling by an average of 6,700 ppm TPH. The latter two methods produced a more concentrated waste stream (either solid or liquid) which required off-site disposal. In the case of submerged tilling, a greater volume of oil can be released and recovered if more than four passes are conducted while conserving the same volume of sediment.