ABSTRACT # 1141233
The cargo of a double-tank truck carrying diesel and gasoline was released directly into a fast-flowing upland stream following an accident on a mountainous section of road in British Columbia (BC), Canada. High concentrations of the product were trapped in the interstitial spaces of coarse (cobble-boulder) sediments during a period of rising water levels. Almost the entire river backshore in the affected area was characterized by steep wooded slopes so that access everywhere was very difficult. These constraints for the SCAT program largely were overcome where direct backshore access was not possible using river rafts, boats (on the reservoir above the dam) and small Unmanned Aerial System (sUASs). Based on the survey results, a 4x4 Spider Walking Excavator equipped with a Universal grab on the hydraulic arm was deployed over a 2.5 km section of river immediately downstream of the accident site over a 9-day period. The grab rotated to mix the sediment or lifted and moved cobbles and boulders along the channel margin and in river bed sediments to release the oil. Swift Water Rescue personnel and river rescue response equipment were positioned with the Spider operations and the SCAT river bank surveys throughout the project, and used to scout river conditions ahead of SCAT rafting operations. Air monitoring was maintained throughout the response during all operations both along river banks as well as in the cab of the Spider while working in the river. A small UAS quadcopter was deployed to monitor the mixing activity in real time where the excavator could operate but ground access was unsafe or physically not possible. Standard SCAT practices were followed to provide the Unified Command (UC) with Shoreline Treatment Recommendation (STR) forms to guide the operations activities and once the treatment criteria were achieved STR Inspection Reports (SIRs) were submitted for approval by the UC. A downstream daily water sampling program monitored for PHs, VOCs and PAHs in the river waters during the mixing operations downstream of the operations area. At no time during the mechanical mixing activities (April 3 – 12) did the results of the analyses exceed Canadian and BC Water Quality Guidelines standards downstream past the confluence with the Salmo River and standards only were exceeded for the first few days of mechanical mixing (April 3 – April 5) during the period that the Spider was working on the upper reaches of the South Salmo.
The incident occurred at 20:45 on 27 March, 2019 when a double-tank truck carrying diesel and gasoline left the road on a mountainous section near the intersection of Highways 3 and 6 (Figure 1) approximately 14 km south of Salmo, British Columbia (BC), Canada. The entire cargo was released directly into the South Salmo River, a fast-flowing upland stream adjacent to the road (Figure 2).
The purpose of this case study is to illustrate how (a) a SCAT survey adapts to the environmental setting of a spill and (b) directly supports Operations during treatment activities.
A Unified Command (UC) of the British Columbia Environment Ministry and the Responsible Party was established at Creston, BC, approximately 60 km east of the incident location through the Kootenay Pass. The major features of the response time line are summarized in Table 1.
Although the incident site initially was closed to Operations due to the fatality that resulted from the accident, boom was placed immediately downstream of the Point of Entry (PoE) (Figure 2) and adjacent to an access point at the Highway 6 bridge on the South Salmo (Figure 1). Residual oil was released during removal of the tank trailers at 14:00 on the 29th. The response area was divided into four operational Divisions (Figure 3) from the PoE downstream 25 kilometers to the Seven Mile Dam on the Pend-d'Oreille (spelled Pend Oreille in the USA) River Reservoir. Downstream distances (Kilometer Post – KP) were measured from the PoE and used to identify the Division and Segment boundaries.
Each Division had a distinctly different river hydrology and morphology.
Division A on the South Salmo River extended from the PoE downstream 2 kilometers to the confluence with the Salmo River. This section of the river system has predominantly coarse sediments with narrow banks and shallow fast flowing water
Division B extended for 5 kilometers downstream along the Salmo River from the confluence of the South Salmo River, following Highway 6 south toward the US international border. This section of the Salmo river is wider and deeper, however still had swift water flow with sections of rapids. Shorelines are commonly mixed sediment and vegetated banks.
Division C extended for 7 kilometers downstream along the Salmo River where it turns East, ending at the confluence with the Pend-d'Oreille River. This section of the river channel is narrower, predominantly bedrock with fast flowing water and class 4 rapids. Shoreline access is very restricted, with the only safe access at the downstream end of the division near the bridge at the confluence.
Division D extended along the Pend-d'Oreille River, 11 kilometers downstream ending at the Seven Mile dam. This section of the river system is a reservoir formed by the dam and is wide (300–400 meters) and deep. The shorelines along the reservoir are variable with steep sediment banks, low mixed sediment and vegetated banks.
An Environment and Climate Change Canada (ECCC) gauge (# 08NE074) in the Salmo River at KP 5.8 (Division B: ~2.0 km downstream of sample station SW-250; Figure 1) provided data on the water levels and discharge during the project. No hydrometric data were available for the South Salmo River and surrogate data for a stream of similar size and character at Hidden Creek (ECCC # 08NE114) near Salmo just upstream of the confluence with the Salmo River was used to develop advice and recommendations for the UC. During the operational period the water level for the Hidden Creek gauge increased from +0.53 m on 27/29 March to +0.65 m on 6/7th April and the discharge increased from 1.8 m3/s on 27/29 March to 3.3 m3/s on 6/7th April (Figure 4: the solid bar indicates the period of the river treatment operations). The increase in water levels observed in the South Salmo Rive was similar to the data from the Hidden Creek gauge following the release, so that the oiled river banks were quickly submerged. Water levels and discharge then slowly decreased until the freshet in late April (see discussion below).
SCAT PROGRAM AND RIVER SURVEYS
The SCAT program was directed by the Environmental Unit Leader who also was the SCAT Coordinator. A full-time data manager initially was assigned in the Incident Command Post (ICP) but then worked remotely after the ICP was demobilized. Shoreline Cleanup Assessment (SCA) surveys were constrained by poor river access and unsafe water conditions in the affected area. The river channel margins were surveyed and oiling conditions documented wherever safe land access was achievable and intervening areas were surveyed in Division A using an sUAS, in Divisions B and C by “river SCAT”, and in Division D by boat.
SCAT surveys by boat and raft were conducted to fill in ground access gaps; however, these were restricted to safe landing sites. For the “river SCAT” surveys the SCAT Team included a certified river raft guide and a swift water rescue specialist. A Cataraft™ and Necky™ kayak were used to transit through Divisions B and C to assess portions of segments for oiling where the channel margin was accessible from the water. Prior to beginning the survey, a detailed safety briefing was conducted that included a review of hand-signals, swimming skills, and hazard avoidance. As this river had not been rafted before by any of the team members, the swift water rescue specialist used a kayak to scout river conditions ahead of the raft and indicated using hand-signals the safest route for the raft to take (Figure 5a). Channel margins with mixed sediment substrate and natural collection areas (e.g., eddies, mid-channel bars, outside channels) were specifically targeted for oiling condition “spot checks” (Figure 5b). When on the shoreline, the SCAT Team Lead reviewed the next downstream river section within sight with the raft guide and communicated the next location to safely land the raft for a “spot check”.
SCAT support for the operations in sections of Division A where ground access was not possible, but which were accessible to the operations equipment, included real-time video feeds from UASs. During the operations, at least one Team Lead was assigned to support the treatment activities.
Safe river operations throughout the project in this remote area included ambulance and medical services as well as river rescue response equipment that were on site during all activities. Swift water rescue specialists were positioned with Spider operations and SCAT shoreline surveys as well as with the downstream raft surveys. Air monitoring was maintained throughout the response during all operations both along shorelines as well as inside the equipment cab while working in the river
Within the first few days, the SCAT team developed a set of oiling categories appropriate to the oil character on this incident and to the treatment criteria developed by the UC (Table 2). These Oiling Categories were based on past experience with similar products in rivers and streams. As part of the calibration process a Job Aid was developed to illustrate the oiling categories (Figure 6).
The majority of the product observed was in the South Salmo River from the PoE to the confluence with the Salmo River 2 km downstream (Division A; Figure 1 and 3). This entire section of the river system was categorized as having “Moderate” oiling based on the observations at assessable locations. Sections of “Moderate” oiling were observed in Segments B-001 and B-002 in Division B. The remaining channel margin oiling observations downstream of River Segment B-002 were “Light”, “Very Light” or “Trace”, which were below the treatment criteria for this incident (Table 2). Shoreline spot checks by boat or raft also found No Oil Observed (NOO) on channel margins at multiple locations. The length of the maximum shoreline (channel margin) oiling observed is summarized by Division in Table 3.
The SCAT team evaluated the Hidden Creek and the Salmo River water levels and reviewed the BC River Forecast Center's “10-day forecast of discharge” for the Salmo River gaging station (forecasts@http://bcrfc.env.gov.bc.ca/freshet/map_clever.html). The forecast at that time indicated that the spring freshet would not be expected for several weeks. Based on the SCAT data and on this discharge information, the UC determined that allowing natural attenuation was not acceptable and that treatment actions would be required.
The water level in the Salmo River varied over a 30 cm range (+1.0 m to +1.3 m) during the response and the spring freshet maximum water level of +1.72 m with a maximum discharge of 88 m3/s occurred between 21–24 April, almost four weeks after the incident and ten (10) days after the treatment operation was demobilized (Figure 7: solid line = 2nd to 12th April operational period). Similarly, the freshet at Hidden Creek peaked at +0.87 m water elevation and 8.3 m3/s discharge on 19/20 April (Figure 4; solid line = 2nd to 12th April operational period).
Based on prior experience of fuel oil spills into small streams and rivers (1999 Whatcom Creek WA and 2013 Lemon Creek BC incidents), the SCAT team recommended a mechanical mixing technique using a “Spider Walking Excavator” to release oil trapped within the channel margin and submerged sediments. A Menzi-Muck A91E 4x4 Plus (Spider”) equipped with an articulating “Grabber” attachment was contracted to agitate the coarse river margin and stream bed sediments. This equipment is ideally suited to operations on very steep slopes, such as river access points, and in very large sediments such as cobbles and boulders. Each of the four hydraulic legs has all-wheel drive and steering. The hydraulic boom arm provides a fifth point of contact so that the machine can “walk” in a crab- or spider-like movement (Figure 8 - left). The Universal grab on the hydraulic arm is illustrated in Figure 8 (right).
Two generic Shoreline Treatment Recommendations (STRs) were generated for Division A:
STR # 001: The PoE area (“Earth Bank; Boulder, Cobble River Bank”)
STR # 002: Streams and Shallow Rivers (“Boulder, Cobble, Pebble sediments”)
An additional section of river downstream of the confluence (Segments B-001 and B-002) along the Left Descending Bank (LDB) of the Salmo River was observed to have zones of “Moderate” oiling on mixed sediments that required treatment to meet the treatment criteria. An STR was generated for this section of shoreline:
STR # 003: Streams and Shallow Rivers (“Mixed Boulder, Cobble, Pebble sediments”).
Following treatment, STR Inspection Report (SIR) recommendations were completed and submitted to the UC for individual segments identified for treatment in Divisions A and B and for the remaining untreated downstream areas. Each SIR recommendation was for “No Further Treatment” (NFT) and was signed by the SCAT Team Lead and the RP's Qualified Person (QP), and then submitted to the UC. A total of fifteen (15) SIRs were approved by the UC.
A SCAT survey of Division A was conducted on foot on 30 March and in Division A, B and C on 31 March. Based on these observations an STR (# 001) for the 100 m river bank section at the PoE was prepared and low-pressure flushing commenced on 2nd April. This tactic was not effective and was ended due to the risk of destabilizing the man-made embankment.
As recommended in STRs # 002 and # 003, the Spider mixed the oiled channel margin and river bed sediments along a 2,500 m length of the South Salmo and Salmo Rivers between 3rd and 12th April, with one day of down time (7th) due to mechanical issues. The operator initially was given guidance from a SCAT Team Lead when sufficient agitation had been accomplished to reduce the released oil to less than a continuous rainbow sheen (Table 2; Figure 6). This guidance became unnecessary as the operator quickly appreciated the treatment target. The SCAT Team Lead continued to provide guidance on which locations required agitation. Access was limited immediately downstream of the PoE past the first river bend (Figure 2) and in this 250-m section an sUAS operated by Stantec was used to observe and support the mixing.
Water samples were collected to obtain background data upstream of the PoE (Station BG-100; Figure 1) and throughout the affected area during the period 29th March through 2nd April as part of the monitoring program. A sample plan was developed and implemented during the mixing operations which involved water collection from 2nd through 12th April at station SW-250, which was specifically located for this operation near KP 3.8 (Figure 1). The mechanical mixing operations ran 3rd through 12th April, with the exception of 7th April due to maintenance that day (Table 1). Samples were collected from 2nd through 12th April at SW-250. Four (4) samples were collected at different times on 3rd April and two (2) on the 4th; otherwise one (1) sample was collected per day. In addition, closer to the operations in the South Salmo River, station SW-200 at approximately KP 1.8 below a boom location at a bridge access site was sampled five (5) times on the 3rd and subsequently once (1) each day on the 4th, 5th, 6th and 12th April.
The samples were analyzed for Petroleum Hydrocarbons (PH), Volatile Organic Compounds (VOC), and Polycyclic Aromatic Hydrocarbons (PAH) to address the federal and provincial standards (Table 4).
At station SW 250, below the treated sections of river (Figure 1), no concentrations exceeded the standards. At station SW 200, within the treated area, concentrations exceeded one or more of the VOC and PAH standards in three (3) of the five (5) samples collected on the 3rd and in the two (2) single samples collected on the 4th and 5th April. No results exceeded the standards for the samples collected on 6th and 12th April (Figure 9). All twenty five (25) samples except one had PH concentrations <0.20 mg/L or < 35 μg/L.
Table 5 identifies the specific PHs and VOCs that were analyzed and those PAH compounds that provides the highest single concentration from all 25 samples. Table 6 summarizes the results for the specific analytical results that exceeded the standards; for the 3rd April results the worst case value has been selected. (Note for Tables 5 and 6, (a) a value of “<x.x” indicates that the analytic was not detected at a concentration greater than the laboratory reporting limit, and (b) the superscript in the value “xA” indicates the standard that was exceeded by the concentration).
Following a spill of diesel and gasoline from a double tank truck accident, the light products were not fully remobilized and dispersed downstream of the PoE despite the turbulent nature of the flow in the very shallow South Salmo River (Figure 2). Water levels increased immediately following the spill, so that the oiled river banks were quickly submerged. SCAT surveys were conducted on foot and by boat and raft. An sUAS was deployed where ground access was not safe in areas of highest oil concentrations immediately downstream of the PoE. The primary treatment technique involved a Spider Walking Excavator equipped with a Universal grab for mechanical agitation of sediments to release the trapped oils. This technique proved highly effective as the Spider was able to easily and quickly agitate or move even the largest boulders in the stream. On the order of 2,500 m of river was treated during a 9-day working period including river bed sediments and channel margins.
Analytical results that exceeded standards were only from water samples collected on the first few days following the release at sample sites 100 and 200 near the Highway 6 bridge (1.8 km from PoE), and at sample site 300 (14.1 km from the PoE). At no time during the mechanical mixing activities (April 3 – April 12) did the results of the analyses exceed standards downstream past the confluence with the Salmo River. Analyses that exceeded standards were only for the first few days of mechanical mixing (April 3 – April 5), during the period that the spider was working on the upper reaches of the South Salmo River near the PoE, which would have had the heaviest instream oiling conditions.
In a similar spill response in which mechanical equipment (a backhoe and a Spider) were used to agitate gasoline from steam bed sediments the analyses of interstitial water samples immediately downstream after mixing detected hydrocarbons of 0.13 μg/L for gasolines and Non-Detect for BTEX (Owens et al. 2001). The sample data collected during this incident and from the Whatcom Creek response indicate that the most probable transfer pathway for the light products released by mixing in streams and rivers is evaporation, with only small a proportion dissolving (into the water). This conclusion is important as it highlights (a) the potential and real safety issues associated with mixing for spilled light products and the need for air monitoring, as well as (b) the low potential risk for aquatic life in the water column. Sediment mixing has been recommended as a viable and effective treatment option for a range of spilled oil scenarios (API 2016a, b) and this study reinforces that option for light products in streams and rivers, as well as other freshwater bodies and marine environments.
SCAT surveys adapt to the environmental setting of a spill and in this instance boat and raft surveys supplemented the traditional systematic ground survey approach where access from the backshore was limited. In addition, an sUAS was deployed where ground access was not safe in areas of highest oil concentrations immediately downstream of the PoE and was used to observe and support the mixing activities in that area. The project highlighted the value of close cooperation between the SCAT and Operations program.
Permission to publish the results of the surveys and the sample analyses was provided by WestCan Bulk Transport. The water samples were collected and analyzed by Stantec. Polaris Applied Sciences, Inc. (Elliott Taylor and Andy Graham) and Triox Environmental Emergencies (Shannon MacDonald and Leanne Zrum) provided the SCAT Team Leads.
1 FOOTNOTE – page 7: Water-level scale (left) is +0.50 m to- +0.95 m and discharge (right) is 1.0 to 9.0 m3/s
2 FOOTNOTE – page 12: Water-level scale (left) is +0.9 m to- +1.85 m and discharge (right) is 20 to 100 m3/s