Meso-scale studies on the penetration and retention of diluted bitumen in different types of shorelines, Northern British Columbia, Canada

Diluted bitumen (dilbit) is a non-conventional petroleum product produced in Alberta, Canada and transported in large quantities to external markets. The Canadian oil sands contain the world’s third-largest oil reserves after Saudi Arabia and Venezuela (Alberta Energy, 2011). Bitumen is produced from the natural oil sand deposits by a number of processes, including direct mining and in-place extraction (Read and Whiteoak, 2003). In order to move the bitumen to market by pipeline, it is diluted with either condensate or synthetic crude oil to form “dilbit”, with viscosity, density and other properties engineered for pipeline transportation and use by the customer refineries (Environment Canada, 2013).

The need for shoreline response preparedness in the event of an accidental spill of dilbit in the marine environment was recognized in a new initiative by the federal government in 2013. The Enhancing Marine Safety program was subsequently created to address knowledge gaps, to conduct scientific research and studies on the behaviour, fate and appropriate response to possible marine spills of non-conventional petroleum products

The Enhancing Marine Safety program, Phase 1a was a three-year program that finished in March 2016. A major focus of this program was (1) the study of marine shorelines along the coastline of northern British Columbia, and (2) to conduct studies on the fate, behaviour and cleanup of diluted bitumen products on different types of shorelines and under various conditions. The results of the first activity (1) mentioned above will be presented in the other paper presented by this author at IOSC 2017 titled “The Canadian oil spill shoreline research program: establishing a baseline dataset for the marine coast of Northern British Columbia”. The second activity is summarized in this paper and is taken from a three part study to develop decision support information for response to dilbit spills stranded on shorelines (Harper et al., 2016; Sergy, 2016).

Emergencies Science and Technology Section, Environment and Climate Change Canada (ECCC) has a long R&D history in terms of developing knowledge on oil-on-shoreline behavior and response. A few examples include the Baffin Island experimental oil spill program, the Orimulsion shoreline studies program, long term monitoring of the Arrow spill, and a series of lab and meso-scale experiments into oil penetration and retention on coarse sediments. . The meso-scale studies with dilbit are a continuum of a long term series of investigations.

One of the major objectives of the Phase 1a program was twofold:

• To conduct meso-scale studies on the fate, penetration and retention of diluted bitumen on substrates representative of northern BC marine shorelines.

• To provide scientific and operational guidance on shoreline response practices for a diluted bitumen spill in northern British Columbia in order to provide spill response teams with more informed technical support for decisions regarding shoreline treatment options.

ESTS worked with Coastal and Oceans Resources Inc. (CORI) to design and perform the experiments to measure the penetration and retention of dilbit in different types of sediment. The project was carried out at the CORI laboratory in Victoria, Canada. CORI used a methodology derived from a protocol developed during a previous shoreline study project with ECCC to measure the penetration and retention of a former Venezuelan bitumen product called Orimulsion^TM (Harper and Kory 1997). Under the Enhancing Marine Safety program Phase 1a, the region of northern British Columbia was the study site. Consequently, the sediment type used in this meso-scale study would be representative of typical shorelines of northern BC and aligned with the results of the segmentation of the shoreline in the northern BC. In the report by Laforest et al., (Laforest et al., 2014), seven shore types comprise almost all the shoreline and those types can be generalized to four categories: bedrock (73%), coarse sediment (14%), sand & gravel sediment (5%) and salt marsh (8%). This meso-scale study focused on those types of sediments only but excluded the salt marsh.

For this experiment, two dilbits were tested, Access Western Blend (AWB) and Cold Lake Blend (CLB) with three different weathering states: unweathered or fresh dilbits (0% mass loss,W0), moderately-weathered dilbits (~ 16% mass loss; W2) and heavily-weathered dilbits (>20% mass loss; W4). For comparison purposes two fresh fuel oils were tested. These were Bunker C and an Intermediate Fuel Oil (IFO 180). The weathering process was done by an air stripping technique where compressed air was bubbled through the oil while the weight of the oil was monitored to measure the percentage weight loss (Harper et al., 2016). A summary of the physical properties of the oils is presented in Table 1.

A series of 102 column tests were conducted where floating oil was lowered through sediments of different size (Figure 1). The height of each column was 15 cm. Seven sediment types were selected from coarse to very large pebbles for this experiment; sediments were a single clast size and classified as very well sorted. Artificial Ocean™ was used to create “seawater” with an average salinity of 30.9 psu (density = 1.024g/cm³). The control temperature of the seawater was maintained at 10°C. The tides were simulated in each column. The “low tide” exposure period was 3 hr followed by a 1 hr and 24 hr “high tide” submergence.

Two primary measurements were made during the sediment column experiments:

1. Penetration depth as observed through the plexiglass walls of the column,

2. Oil Retention following oiling, a 3-hr “low-tide” exposure and “high-tide” submergence determined by mass balance of oil initially added minus oil refloated after 1hr and 24 hr.

The columns were eight-sided and an observation of penetration was made in each face of the column so the depth of penetration is actually the mean of eight measurements. In some of the finer sediment columns, the sediments were excavated to determine if edge effects were important in the penetration. Column replicates indicate the methodology is repeatable with the average difference of around 8% (relative standard error).

The amount of oil retained in sediment within each column was estimated by comparing the amount that was original loaded on the column to the amount recovered off the water surface of the column at the end of the experiment. Oil was layered onto the water surface, the surface lowered for 3hr to simulate oil stranding on the sediment at low tide, then the water level was raised to simulate flood tide and oil on the water surface collected after 1hr and 24 hours of submergence. A comparison of 88 pairs of replicates showed a mean difference of 3.2% in estimating oil retention.

Cobble/rock adhesion tests were conducted on 34 cobbles with both mineral surfaces and biological substrates (the seaweed Fucus). The cobbles were collected from intertidal zone of the Gorge Waterway in BC and they likely included a biofilm on their surface. The cobbles were oiled by lowering the water level such that cobbles were coated with oil, then left exposed for 3 or 24 hours and finally submerged for 1hr so that the tenacity of oil adhesion to the surface could be observed. The retention was visually and photographically recorded before both cold and hot water washing to further evaluate coating adhesion.

The primary results are based on two observations: (1) the observed depth of penetration measurements of different oils (oils listed in Table 1) into different sediments and (2) total oil retention of different oils in different sediments after 3 hours of exposure and 1 or 24 hours of submergence.

Penetration depths are summarized in Figures 2 and 3 and show the differences between the two dilbits as well as comparison of dilbits to conventional oils. The results for the two unweathered dilbits with all types of sediments show that there is no significant difference in penetration between the two dilbits. For moderate weathering, penetration of the AWB dilbit is slightly greater than the CLB. The difference is most obvious in the medium pebble tests where AWB freely penetrated to 15 but the CLB only penetrated to around 11 cm. For the heavily weathered dilbits, the CLB showed greater penetrations up to 4.4 cm in very large pebbles, whereas the AWB showed < 1 cm penetration in all cases. Overall the differences in penetration depths are not significantly different between the two dilbits (Figure 2). The two moderately weathered dilbits have similar penetration potential to unweathered Bunker C oil whereas unweathered dilbits perform similarly to unweathered IFO 180 fuel oil (Table 2).

The comparison of the 1 hour and 24 hours post-oiling submergence penetrations (Figures 2–3) shows little difference between the two submergence periods indicating minimal oil mobility at the 10°C test temperatures. That is, penetrations will not increase while sediments are submerged.

The penetration potential provides responders with a relative index of how different oils will penetrate into shoreline sediments. The penetration potential is grouped in terms of three general classes, where the high penetration potential oils freely penetrated all sediments (>14cm) and the low penetration oils did not freely penetrate any of the sediments. Table 3 shows the ranking of the test oils based on observed penetration values.

Oil retention is expressed as per cent of initial loading and is summarized in Table 4. Generally, highest oil retention is in granular and small pebble range. The sediment of maximum retention is coarser for more viscous oils than for less viscous oils (Table 4). The ranking in the Table 4 is based on the observations after 1 hour and 24 hours submergence.

In general, the two dilbits performed similarly, although the actual retention varied widely depending on degree of weathering. Moderately weathered dilbits were similar in retention characteristics to unweathered Bunker C oils. There are large differences between the 1 hour and 24 hours submergence times, indicating that moderately weathered dilbits are quite mobile with retention decreasing by ~20% between the two test durations.

The results for the penetration and retention can be impacted by the weathering state, the type of the sediments and also by the time of emergence/submergence during the tides. The results showed that the two dilbits used in the experiment performed similarly. There are no major differences between penetration and retention at the 10°C test conditions. The tests also indicate that dilbit penetration and retention of dilbits are within the range of results for conventional oils (Bunker C and IFO180).

For the tests of the cobble/rock adherence, thirty-three tests were conducted with rock surfaces, hard-biological substrates and soft biological substrates. The 3 hours drying tests showed that all visible oil was removed upon subsequent emergence. The 24-hr drying test showed that virtually no oil was removed upon submergence and coatings were very resistant to removal by cold and warm-water flushing. The tests indicate that there is a “set time” between 3hours and 24hours; once the oil coating is “set”, it very difficult to remove. There are no significant differences in adherence between AWB [18] (W2) and CLB [15] (W2) on rock surfaces, The moderately weathered dilbits were more difficult to remove than IFO[0].

An important goal of the ‘Enhancing Marine Safety’ program was to support effective preparedness and response actions and help stakeholders involved in a spill of diluted bitumen with special consideration to the shoreline characteristics of northern British Columbia. This was realized by preparing a report to summarize scientific and operational knowledge and provide guidance on diluted bitumen spills (Sergy 2016). The reports highlights relevant scientific findings as well as updating and consolidating existing information and practices from conventional shoreline spill response that have bearing on marine shoreline response in BC. Some highlights of the report follow. The initial composition and properties of any single diluted bitumen product are different from any other single conventional crude or fuel oil. Following release to the marine environment, the important difference in dilbit when compared to conventional oils, is the accelerated rate of weathering, primarily due to the rapid loss of the light diluent, and the relatively rapid increase in density, viscosity and adhesion properties associated with that weathering. The dramatic change can be significant in terms of fate and behaviour and has response implications.

Dilbit adhesion characteristics appear to be of high significance to response priorities and cleanup techniques. Both low adhesion and high adhesion forms of dilbit in different weathering states are possible. We can predict scenarios where stranded dilbit could have initial low adhesion / high mobility, and conversely, dilbit that penetrates and exhibits high adhesion characteristics that will be very tenacious. Upper intertidal zone, storm berms and supratidal zone will be more problematic as the sediments are drier, and there is more exposure to air allowing for the oil to become sticky and adhere.

Fresh to lightly weathered dilbits will readily penetrate sediments of very coarse sand to very large pebbles, with maximum retentions in the coarse sand/granule size materials, Penetration and a relatively high retention of moderately weathered dilbits will occur in small pebbles or larger sediment sizes. The response in case of the dilbit spills will be challenging if we want to minimize the impact on the shorelines. The sediment structure of segments of shoreline will have to be well known by the SCAT teams in order to address the problem of subsurface contamination.

Dilbit which has advanced to a stage of moderate weathering or greater should be considered persistent type of oil. Long term persistence of weathered dilbit oil residues on shorelines is clearly possible on the northern British Columbia coastline for example where oil has been sequestered within coarse grained sediments or deposited above the zone of normal wave action. Most of the north coast is a protected/semi-protected wave environment. This implies a reduced natural self-cleaning potential from mechanical wave energy and increased potential oil persistence, which will have a bearing on protection priorities and cleanup options. As a substantial amount of the BC shoreline is also remote, response time is inherently slower and logistics made more difficult. Windows of opportunity for some response techniques may be narrower and applications of techniques may be limited by access and waste disposal issues.

Responders should also be aware that negatively buoyant weathered dilbit can be created under some conditions at the shoreline itself, as it could likewise be for heavy conventional oil. This would become a source of sinking dilbit or submerged dilbit being deposited in the immediate adjacent nearshore or residing on the shoreline surface sediments. Pilot scale evaluation of some cleanup techniques (e.g. washing and mixing) and SCAT monitoring during shoreline operations would be appropriate measures to assess the potential for creation of sinking oil and address the high profile concern about such an event.

Although dilbit is non-conventional oil, it is recommended that the selection of shoreline response objectives, strategies, and techniques follow the same principles, process and options for a dilbit spill as for a conventional crude or fuel oil spill. There is no strong evidence for development of new or innovative shoreline cleanup techniques for dilbit stranded on shorelines. That being said, there will be a need for modifications and fine tuning of approach - how, when and where they are best applicable - keyed to the real time dilbit oil properties and oiling conditions.

The standards and procedures for current and evolving SCAT in Canada (Environment and Climate Change Canada, 2017  in press) should be followed for a spill of diluted bitumen or conventional oil in northern BC. However, it will be important to immediately deploy knowledgeable and experienced SCAT specialists and begin SCAT assessment as soon as we have a dilbit spill due to its rapidly changing properties and in order to try and reduce problematic oiling scenarios. It will be important for SCAT recon and ground surveys to identify locations and oiling scenarios where immediate protection/treatment/response can reduce problematic scenarios, e.g. where the dilbit adhesion is low, in order to take action before adhesion becomes high.

In conclusion, the focus or goal of the Enhancing Marine Safety Phase 1a was reached. These were the study of marine shorelines of northern BC and studies to better understand the fate, behavior and cleanup of non-convention crude oil on shorelines typical or northern BC. We have more information on the characteristics of the northern BC shorelines. The Bit_Ex experiment provided new knowledge on the fate and behavior of the non-conventional dilbit products on marine shorelines and both operational and scientific guidance has been given for a potential dilbit spill that impact the coastline of norther BC. These studies provide good support for a future dilbit spill for decisions regarding shoreline treatment options for the northern British Columbia coastline but also for shorelines with similar conditions.