Explosion:
Seismometer data suggest charge sizes with a relative effectiveness of 500 kg TNT-equivalent each were used at the four explosion sites (Fig. 2). The north-eastern explosions near Nord-Stream 1 (NS1) took place within the deepest part of the Bornholm Basin, known as the Bornholm Deep. The south-eastern explosions near Nord-Stream 2 (NS2) took place just south of the munitions dump site. The Bornholm CWA dumpsite is located approximately 22 km north of Explosion 1 and 21 km south of the explosions. The escaping gas from the sabotaged pipelines resulted in a pressure drop from 115 to 7 bars in all pipes indicating a significant jet of gas for six days before the pipes were emptied [1]. This in combination with the explosions were a major disturbance for local hydrodynamic generating currents that exceeded the resuspension thresholds for the sediments.
We found no evidence that the natural gas release alone would cause any lasting significant marine impacts. We also assessed the relevance of explosive residues from the TNT, but assuming complete detonation, explosive chemical residues would be minimal and represent negligible toxic risk. Therefore, these two impacts are only covered in Supporting Information (SI 1).
Description of site:
The Bornholm Deep is the deepest part of the sedimentary Bornholm Basin, characterized by strong stratification in the water column with near-bottom currents rarely exceeding sediment resuspension thresholds [8]. Depths up to 100 m fall below the halocline/pycnocline, and surface sediments are dominated by muds and muddy sands affected by almost permanent hypoxia or anoxia [9]. Vast areas of the Bornholm Basin are covered with a “fluffy layer” from a constant accumulation of settling organic flocs formed in the water column above the pycnocline. These muds (the fine fraction of sediments < 2 µm) act as a major sink of pollutants due to their large surface area, abundance of clay minerals and associated coatings of organic matter and iron/manganese oxides and oxyhydroxides [10]. Therefore, a major part of heavy metals and organic contaminants contained within the bottom sediments will be associated with this fraction in the upper 25 cm of the sediment, which at local sedimentation rate at 0.52 ± 0.02 up to 0.82 ± 0.10 mm yr−1 were building up throughout the whole Anthropocene.
The detonations took place approximately 20 km away from the designated CWA dumpsite. Within a 20 kms radius zone of the explosions of the Nord Stream 1 pipeline, 39 samples containing measurable CWA levels have been recorded [13]. No samples were collected closer than 20 kms to the explosion of Nord Stream 2 pipeline. As with other pollutants, the highest CWA concentration is in the top 5 cm of the sediment layer [14]. There are several different CWA residue compounds originating from either mustard gas or arsenic-based CWAs. In this assessment we combine the compounds into two classes with two mean values used in the assessment (Tab. 1). We assume conservatively that the sediment near the explosion sites contained the mean CWA residue concentration of these samples.
Table 1. Sediment concentration of CWA near the Explosions in the Bornholm Deep
Compound
|
CWA concentration (µg/kg)
|
n
|
Mean
|
Maximum
|
Std.Dev.
|
Mustard related CWAs
|
1,4-Dithiane
|
11.0
|
3.1
|
32.0
|
9.6
|
1-oxa-4,5-dithiepane
|
5.0
|
5.4
|
24.0
|
10.5
|
1,2,5-Trithiepane
|
19.0
|
0.7
|
2.8
|
0.6
|
Sum Mustard related
|
39.0
|
9.2
|
56.0
|
13.8
|
As based CWAs
|
Adamsite
|
1.0
|
27.0
|
27.0
|
0
|
5,10-Dihydrophenoarsazin-10-ol 10-oxide
|
6.0
|
1.0
|
6.1
|
2.5
|
Clark I
|
2.0
|
1641.3
|
3262.8
|
2293.2
|
Diphenylarsenic acid
|
6.0
|
12.3
|
68.0
|
27.4
|
Diphenylpropylthioarsine
|
6.0
|
8.0
|
27.0
|
11.9
|
Triphenyl arsine
|
8.0
|
44.0
|
172.3
|
76.6
|
Triphenyl arsine oxide
|
8.0
|
6.9
|
43.6
|
15.1
|
Phenylarsonic acid
|
7.0
|
1.7
|
6.0
|
2.9
|
Dipropyl phenylarsonodithioite
|
14.0
|
5.2
|
41.0
|
10.7
|
Sum Arsenic-based
|
39.0
|
129.2
|
3461.6
|
550.4
|
Pollutants in Baltic Sea sediments have been monitored for several decades by HELCOM Contracting Parties[4]. The following contaminants have been prioritized for the Bornholm Basin: Mercury (Hg), polybrominated diphenyl ethers (PBDEs), hexabromocyclododecanes (HBCDs), Cesium-137, anthracene, cadmium (Cd), lead (Pb), and tributyltin (TBT). Table 2 contains the mean sediment concentration of the prioritized problematic contaminants in the Bornholm Deep reported by HELCOM and European Environmental Quality Standards (EQS) for “prioritized substances and certain other pollutants” in “other surface waters” are used[5] (Tab. 2). In the absence of European EQS values, national Danish EQS values are applied[6]. The toxicity of arsenic-based CWAs is evaluated according to inorganic or organic As toxicity. For mustard gas and its degradation products the lowest measured toxicity threshold value is used (Chmielińska et al., 2018).
Table 2. Mean sediment concentrations of pollutants in the Bornholm Deep (HELCOM, BSEP, 157, 2016) with estimations of their resuspended tonnage, EQS and relative mixture toxicity contributions.
Compound
|
Mean concentration in Bornholm Deep sediments [µg/kg]
|
Resuspended tonnage
[Tons]
|
EQS
[µg/L]
|
Toxic contributions
[%]
|
Organics
|
PBDE
|
0.31
|
< 0.1
|
0.0049
|
0.04%
|
HBCD
|
0.92
|
0.2
|
0.0008
|
0.67%
|
Anthracene
|
6.3
|
1.6
|
0.1
|
0.04%
|
Metals
|
Hg
|
60
|
15
|
0.05
|
0.7%
|
MeHg
|
0.375
|
< 0.1
|
0.0065
|
0.03%
|
Cd
|
1140
|
285
|
0.2
|
3.3%
|
Pb
|
57700
|
14425
|
1.3
|
25.9%
|
Cs-137
|
77 Bq/kg
|
-
|
0,015 Bq/L[7]
|
3.0%
|
Cr
|
14700
|
3675
|
3.4
|
2.5%
|
Cu
|
15500
|
3875
|
1
|
9.0%
|
Zn
|
14600
|
3650
|
7.8
|
1.1%
|
Ni
|
4000
|
1000
|
8.6
|
0.3%
|
As
|
1800
|
450
|
0.6
|
1.8%
|
TBT
|
17.7
|
4,4
|
0.0002
|
51.6%
|
Chemical Warfare Agents
|
Sum As based
|
129.2
|
32,3
|
0.6
|
0.1%
|
Sum mustard gas related
|
9.2
|
2,3
|
830
|
0.00001%
|
Marine mammals impact assessment:
The range at which there was a risk of blast injury was estimated by equations provided by Yelverton et al (1973)[16] in the SI Figure 1. These equations provide estimates of the acoustic energy (acoustic impulse) as function of the size of the charge and the distance from the explosion. Impact ranges were found when the distance at which the acoustic impulse dropped to 30 µPa·s. This threshold was provided by Lance et al (2015)[17] and corresponds to less than 10% probability of injury to lungs or intestines in human divers. The range within which there was a risk of damage to the inner ear (acoustic trauma) was estimated based on measurements of the sound exposure level from explosions of different charge sizes up to 295 kg at different ranges for detonations of unexploded ordnance (UXO) in the North Sea [18] and extrapolated these out to distances where the levels exceeded the thresholds for onset of noise induced permanent threshold shift (PTS) [5]. The maximum range at which harbour porpoises were susceptible to PTS was extrapolated by linear regression from the abovementioned results in the range 1 kg to 200 kg. Seals, however, are considered less susceptible to PTS at low frequencies than baleen whales [5]. Maximum impact ranges for seals were therefore estimated by correcting the results for baleen whales for both the lower frequency weighting levels and lower sensitivity of seals to PTS compared to baleen whales [18]. The ranges within which temporary threshold shift (TTS) could have occurred was estimated from the propagation loss curves [18], by identifying the distance from the explosion where the received sound exposure level was 15 dB lower than the level at the maximum range for PTS [5]. For further information see the SI section 3.
Sediment resuspension:
During an underwater explosion sediment containing residues pollutants is resuspended into the water column dependent upon the location of the charge (explosion on the surface, below or above the ground), as well as the density and type of seabed. The volume of ejected sediment from a 500 kg load of TNT, assumed to have been used here, placed on the surface of the bottom sediment is approximately 13 m3. In addition, the created gas bubbles, moving upwards, lifts sediment into the water column at volume equal to about half of the volume of the bubble, which in the case of detonation of 500 kg of TNT is in the order of 1000 m3 [19].
The third factor that influences the resuspension of sediments is the gas leak from the damaged pipelines. Due to its buoyancy, the gas will be transported rapidly to the sea surface. The pressure in the pipeline before explosion was about 115 bar [1]. The rapid pressure drops in the pipeline due to the explosion generated a jet propagating towards the axis of the gas line.
To assess the of development of the resuspended sediment a 3D hydrodynamic model, coupled to a sediment transport model was implemented, based on the MIKE powered by DHI tools[8] [20, 21]. The model is based on a flexible, triangular mesh model, enabling high spatial resolution in the area of the gas pipeline explosion. The model includes various dynamically coupled modules [22], including the mud transport MIKE 3 MT [23] for simulation of the transport, sinking and deposition of sediment in the marine environment. The hydrodynamic part of the model has been obtained with the MIKE 3 FM HD module [24]. The influence of wind waves was also considered by using the Spectral Wave MIKE 21 FM SW module [25]. The modules were applied using operational meteorological data that were delivered to the system using an external Weather Research and Forecasting Model covering the entire Baltic Sea area [26, 27]. For further information see the SI4 section.
Marine toxic risk:
To assess if the marine environment is at risk as a consequence of the release of pollutants associated with suspended sediment particles outlined above, the total toxic mixture risk characterization ratio (RCR) based on the data from Table 2 is calculated as:
Concentration additivity of the individual pollutants in Table 2 is assumed. As a conservative approach it is moreover assumed that the sediment particles taken up by biota release the entire amount of associated pollutants, which are subsequently taken up by the organism. We conservatively, furthermore disregards sorption and de-sorption of pollutants during transport in the water. Consequently, resuspended sediment particles have concentrations of pollutants corresponding to the mean values in Tables 1 and 2, at any given time and place. The predicted environmental concentration (PEC) for a pollutant is thus:
The resulting critical sediment concentration of resuspended sediment particles in the water is thus obtained when Eq.1 equals 1. Inserting the values from Table 1 and 2 in Eq. 1 and Eq. 2, gives:
Hence, the toxic threshold of suspended sediments concentration (SSC) in the water column is 5.8 mg/L, above which there is a predicted marine environmental risk.
[5] Directive 2013/39/EU Regarding priority substances in the field of water policy. Official journal to the European Union. Available at. https://eur-lex.europa.eu/legalcontent/DA/ALL/?uri=CELEX%3A32013L0039
[6] BEK nr. 1625 af 19/12/2017 Bekendtgorelse om fastlaggelse af miljomal for vandlob, soer, overgangsvande, kystvande og grundvand. Miljo- og Fodevareministeriet. Available at. https://www.retsinformation.dk/Forms/R0710.aspx?id=196701.
[7] HELCOM 2017: https://www.Radioactive-substances-HELCOM-core-indicator-2018.pdf&usg=AOvVaw12X4V6wPWqlZ9HUh1oeHBa
[8] https://www.mikepoweredbydhi.com/