Blast injury on harbour porpoises ( Phocoena phocoena ) from the Baltic Sea after explosions of deposits of World War II ammunition

period from 28 to 31 August 2019 by a NATO unit in the German Exclusive Economic Zone within the marine protected area of Fehmarn Belt in the Baltic Sea, Germany. Between September and November 2019, 24 harbour porpoises were found dead in the period after those clearing events along the coastline of the federal state of Schleswig-Holstein and were investigated for direct and indirect effects of blast injury. Health evaluations were conducted including examinations of the brain, the air-filled (lungs and gastrointestinal tract) and acoustic organs (melon, acoustic fat in the lower jaw, ears and their surrounding tissues). The bone structure of the tympano-periotic complexes was examined using high-resolution peripheral quantitative computed tomography (HR-pQCT). In 8/24 harbour porpoises, microfractures of the malleus, dislocation of middle ear bones, bleeding, and haemorrhages in the melon, lower jaw and peribullar acoustic fat were detected, suggesting blast injury. In addition, one bycaught animal and another porpoise with signs of blunt force trauma also showed evidence of blast injury. The cause of death of the other 14 animals varied and remained unclear in two individuals. Due to the vulnerability and the conservation status of harbour porpoise populations in the Baltic Sea, noise mitigation measures must be improved to prevent any risk of injury. The data presented here highlight the importance of systematic investigations into the acute and chronic effects of blast and acoustic trauma in harbour porpoises, improving the understanding of underwater noise effects and herewith develop effective measures to protect the population level.


Introduction
The Baltic Sea still contains large deposits of ammunition submerged before and during World War II, which may be detonated in a controlled manner for safety reasons due to shipping routes or offshore constructions (e.g.bridges, tunnels, wind farms, gas pipelines).As these detonations generate very high sound energy levels in the low frequency range, which can propagate over long distances and have far-reaching, adverse effects on marine mammals (Aarts et al., 2016;von Benda-Beckmann et al., 2015;Merchant et al., 2020;Sertlek et al., 2019;Soloway, 2018), there is a need to increase our understanding on the effects of explosions on marine fauna.
The harbour porpoise is the only native cetacean species in the German waters of the Baltic Sea (Hammond et al., 2013(Hammond et al., , 2017;;Scheidat et al., 2008;Viquerat et al., 2014).Two harbour porpoise subpopulations are found in Germany's coastal waters, of which the "Belt Sea subpopulation" with an estimated 42,324 harbour porpoises (95% confidence interval: 23,368-76,658, Hammond et al., 2017) is a substantially larger subpopulation.The "Baltic Proper subpopulation" is estimated at 497 individuals (95% confidence interval: 80-1091, Amundin (2016) and is therefore considered to be "threatened with extinction" (Carlén et al., 2018;Galatius et al., 2012;Viquerat et al., 2014;Wiemann et al., 2010).Under European law, harbour porpoises are protected by various agreements, such as ASCOBANS (Agreement on the Conservation of Small Cetaceans of the Baltic, North-East Atlantic, Irish and North Seas), HELCOM (Helsinki Commission), the EU Habitats Directive and MSFD (Marine Strategy Framework Directive), which are designed to ensure the management and conservation of harbour porpoise populations.Studies on reproductive capacity and age structure have shown that the mean age at death of female harbour porpoises in the Baltic Sea is only 3.67 (±0.3) years, although harbour porpoises were reported up to 20-25 years old (Kesselring et al., 2017).In addition, female harbour porpoises do not reach sexual maturity until 4.95 (±0.6) years, highlighting the negative potential of such early mortality concerning population sustainability and conservation (Kesselring et al., 2017).
Depending on the distance of the animals from the site of detonation, the size of the explosive charge, the frequency of detonations, the substrate and the water depth, the effects, especially of animals in close proximity, vary in severity (Danil et al., 2021;von Benda-Beckmann et al., 2015;Soloway, 2018).Within a distance of up to several kilometres around the detonation site, animals can be directly injured.The shock wave may cause tissue ruptures or even fractures by blunt trauma.Gross lesions of such traumata represent pathologic entities rather than a clinical expression of diseases, such as haemorrhages in the ear, ossicle fractures or ruptures/lacerations of fatty tissue, brain, lungs, alveolar walls, lungs, or gastrointestinal tract (Clemedson, 1956).Studies on humans and terrestrial mammals exposed to blasts from explosions showed dislocation of the middle ear ossicles (Mayo and Kluger, 2006;Patterson and Hamernik, 1997) and damage to the sensory cells in the cochlea (Patterson and Hamernik, 1997), as well as decreased spiral ganglion neurons and afferent nerve synapses (Cho et al., 2013).These types of lesions are named blast injuries and are caused by the pressure component of an explosion (see review by Hirsch, 1968).However, little is known about the effects of blast injuries in cetaceans.Ketten et al. (1993) found round window rupture, ossicular chain disruption, serosanguinous effusion of peribullar spaces, dissection of the middle ear mucosa with pooled sera and bilateral periotic fractures on humpback whales that had stranded and died following an explosion.In addition, experiments performed post-mortem with exposure to different predetermined blast pressures revealed that lesions were consistently manifest post-mortem, after the carcases had been frozen and thawed at high psi (100-300 psi received).Injuries included liver ruptures and haemorrhages, classic blast lungs, laryngeal haemorrhages, segmental gut haemorrhages, cerebral ventricular inflations, intraorbital haemorrhages, middle ear ossicular fractures and inner and middle ear haemorrhages (Ketten, 2006).
At greater distances (depending on the explosive charge) from the blast site with lower pressure waves that do not cause direct injuries, shifts in hearing thresholds can occur, which may either be reversible (temporary threshold shifts or TTS, Lucke et al. (2009), Schaffeld et al. (2019)) or, in the case of particularly high sound intensities and/or repetitive exposure, irreversible (permanent threshold shift or PTS, Reichmuth et al. (2019)).Ketten (1995) estimated the theoretical zones for blast trauma, PTS and TTS for marine mammals exposed to underwater detonations of Class A explosives (TNT derivatives with fast rise-time waveforms) of different charge.For a 1200 lb charge, the lethal zone was estimated up to 300 m from the source and up to 800 m for a 10000 lb charge.The zones where PTS would take place were estimated as 100-500 m (for 1200 lb charge) and 150-900 m (for 10000 lb charge).The habitat of the harbour porpoise in the Baltic Sea is intensively influenced by different human activities and factors which may have a negative impact on harbour porpoise subpopulations (Siebert et al., 2012).These include shipping, seismic surveys, military activities, fisheries, offshore constructions, blasting of munition, chemical and pharmaceutical contamination, and marine litter (ICES, 2019;Panti et al., 2019;Siebert et al., 2012Siebert et al., , 2020;;Unger et al., 2017).Due to high pressure from anthropogenic factors, animals from the Baltic and North Seas are in a poorer health status than harbour porpoises from Arctic waters, which are currently less exposed to anthropogenic factors, showing a better health status in comparison (Beineke et al., 2005;Das et al., 2006;Jepson et al., 2005Jepson et al., , 2016;;Siebert et al., 2001Siebert et al., , 2006Siebert et al., , 2009Siebert et al., , 2020;;Wünschmann et al., 2001).
Unintentional bycatch of harbour porpoises thus poses a significant threat to the subpopulations, especially in the Baltic Sea (Agreement on the Conservation of Small Cetaceans of the Baltic, North-East Atlantic, Irish and North Seas (ASCOBANS) (Carlén et al. 2018).Although it is assumed that bottom-set gillnets can be detected by harbour porpoises by means of echolocation, the weak echoes only allow detection at very short distances of between three and 26 m (Kastelein et al., 2000;Villadsgaard et al., 2007).The evoking of behavioural responses or any hearing impairment can not only affect the survival of individuals but can also have consequences at population level when anthropogenic interventions negatively affect the fitness of many individuals (King et al., 2015).
Between 28 and 31 August 2019, with involvement of the German Navy, forty-two British ground mines of the type MK 1-7 from World War II were cleared by means of blasting by a NATO unit in the Exclusive Economic Zone within the marine-protected area (MPA) Fehmarn Belt, Baltic Sea.The area represents an important foraging habitat for harbour porpoises and has been designated for the conservation of harbour porpoises and other protected marine species.The explosions took place during a sensitive time of the year, the birth and mating period of harbour porpoises (Hasselmeier et al. 2004, Kesselring et al. 2019).In the present multi-scale analysis, we evaluate the characteristics of dead harbour porpoises recovered in close proximity after mine-blasting was undertaken to assess potential effects of blast injury among small cetaceans.
To answer the question whether distinct lesion patterns occur in cetaceans, we conducted full post-mortem examinations in harbour porpoises collected between 7 September and 24 November 2019 in the waters and on the coastline of Schleswig-Holstein to determine if the dead animals showed evidence of blast injury.The aim was to examine not only for immediate lethal damage but also for sublethal changes, which can lead to poor feeding and disorientation of harbour porpoises, increasing the risk of the animals for developing sickness, vessel strike or entanglements, and being exposed to collisions or bycatch.A range of ancillary diagnostic and research studies were undertaken to rule out other disease processes and render a diagnosis of blast injury.To the best of our knowledge, this is the first multi-scale analysis of the environmental impacts of ammunition blasts on marine mammals in the Baltic Sea.
U. Siebert et al.

Location and date of finding
After 42 British base mines (MK 1-7 types) from World War II were detonated between 28 and 31 August 2019, 24 harbour porpoises were found dead on the Baltic Sea coasts of the German federal state Schleswig-Holstein between 7 September and 24 November 2019 (Fig. 1, Table 1).Investigations into the health status and cause of death of recovered porpoises as well as primary and secondary effects of explosions were carried out.Twenty-three animals were found on the beaches, and one bycaught was presented by a fisherman.The location/ coordinates and date of finding the porpoises were recorded.All animals were transported fresh, six animals were directly examined and 18 porpoises were temporarily stored in freezers and thawed prior to necropsy.

Necropsy
The harbour porpoises were measured and weighed according to Siebert et al. (2001).Six teeth from the lower jaw were removed for age determination by counting the annual growth layers (Lockyer 1995).The age class of the animals was estimated based on the habitus, weight, length, sex, date of discovery and gonadal activity (Siebert et al. 2001).The sex of the animals was routinely determined on the basis of primary and secondary sexual characteristics (Kesselring et al. 2017).The classification of the post-mortem state (1 [fresh, not frozen] − 5 [mummified]) followed an international protocol (IJsseldijk et al., 2019;Siebert et al., 2001Siebert et al., , 2020)).The nutritional status of the animals was judged based on their body weight, length, blubber thickness of the seasonally physiologically varying blubber layer, muscle mass and estimated age (Siebert et al., 2001).Due to the high degree of variability of such parameters in neonates, the nutritional status of this age class could not be readily assessed.
The necropsies were carried out in accordance with international guidelines for small cetaceans (IJsseldijk et al., 2019;Siebert et al., 2001Siebert et al., , 2020)).Each organ was removed according to its state of preservation, further surface appearance, weight and size were assessed, and after opening and making an incision of the organs, the contents and tissues were evaluated macroscopically.The air-filled organs, such as the lungs and the gastrointestinal tract and the acoustic organs, which include the melon and acoustic fat in the lower jaw, as well as the ear and its surrounding tissues were examined.Other signs of trauma such as haematomas in the musculature, fat and subcutaneous tissue, pericardium, chest cavity and lungs were also assessed.
In better preserved animals, classified code 2-3, a variety of tissue samples and swabs from various organs and parasites, if present, were collected for further investigations.These included histopathological, virological, serological, microbiological, and parasitological examinations.The ears and cranial sinuses (pterygoid and peribullar) were also examined for parasites and swabs were collected for microbiological and virological examination.Detailed lists of the location, signalment and stranding date of the recovered porpoises are given in Table 1.

Imaging
Prior to necropsy, 14 harbour porpoises were scanned using computed tomography (CT) at the Röntgenpraxis Heide (Heide, Germany).This imaging procedure was used to obtain an overview of possible traumatic injuries to the animal prior to necropsy, so that it could then be prepared in a targeted manner.The Röntgenpraxis Heide has permission by the health department of Schleswig-Holstein, Germany to examine living and dead animals.
Tympano-periotic complexes (Fig. 2a-c) of all collected harbour porpoises were examined at the Department of Osteology and Biomechanics (University Medical Centre Hamburg-Eppendorf, Hamburg, Germany) by first and second generation high-resolution peripheral quantitative computer tomography (HR-pQCT, XtremeCT®, Scanco Medical, Brüttisellen, Switzerland, voxel size 61 µm and XtremeCT II®, Scanco Medical, Brüttisellen, Switzerland, voxel size 42 µm).This technique allows detailed imaging and quantification of bone microstructure.In case of suspected intraosseous lesions or microfractures, additional scans of the affected areas were performed with a voxel size of 31 µm.By means of this imaging technique, high spatial resolution was achieved for the entire tympano-periotic complex as a region of interest.

Histopathology
The tissue samples for histological examination were fixed in 10% neutral buffered formalin and embedded in paraplast (Leica Paraplast® Standard, Leica biosystems, IL, USA) according to a standard laboratory protocol.The 5 µm thick sections were stained with haematoxylin and eosin.In selected cases, additional special stainings included Elasticavan-Gieson and Ziehl-Neelsen as well as Grocott ś methenamine silver impregnation (Siebert et al., 2001;Wohlsein et al., 2019).
The ears were stored at room temperature and decalcified for approximately eight to 10 days using the commercial rapid decalcifier RDO® (Apex Engineering Products Corporation, Aurora, IL, USA), following a previously optimised protocol (Morell et al., 2009) (first day in 50% RDO®, and remaining seven to nine days in 25% RDO®).The ears were then sectioned transversely on four levels (Fig. 2a).The first incision plane was immediately caudal to the sigmoid process.The second plane was sectioned 2 mm rostrally from the first incision, and the other two were each sectioned 2 mm further rostrally from the first plane (Fig. 2a).The tissue sections were placed in cassettes, rinsed in 50% ethanol and routinely embedded in paraffin (Fig. 2b).Three to five µm thick sections were prepared from the paraffin blocks and stained with haematoxylin and eosin.

Microbiology
For bacteriological and mycological examination, samples of lesions and a representative suite of tissues, including the lung, liver, kidney, intestine, mesenteric lymph nodes and spleen were collected and stored at − 25 • C. The investigations were carried out at the Institute for Hygiene and Infectious Diseases of Animals at the Justus Liebig University of Gießen in accordance with the conventional methods of Prenger-Berninghoff et al. (2008) and Siebert et al. (2009Siebert et al. ( , 2017)).

Virology
All organs were examined macroscopically and histologically for evidence of viral infections.Individual organ samples of 21 harbour porpoises were also analysed by polymerase chain reaction (PCR) techniques for the presence of influenza A virus, morbillivirus and herpesvirus (Shin et al., 2019;Van Devanter et al., 1996;Verna et al., 2017).Organ samples were homogenized and viral nucleic acid was extracted using commercial kits, either manually or automatically using the KingFisher Duo Prime (Thermo Fisher Scientific, MA, USA) purification system.Real-time PCR was used to test for influenza A viruses, while morbilli and herpes viruses were detected using gel-based PCR methods.Negative and positive controls were included to check nucleic acid extraction and PCR reactions.

Parasitology
Macroscopically visible parasites were preserved for parasitological determination in 70% ethanol and speciated according to Lehnert et al. (2005).

General information
The porpoises were recovered between 7 September and 24 November 2019 in the area extending from Eckernförder Bay to Lübecker Bay (Table 1, Fig. 1).Twenty-three individuals were found on the beach and one animal was submitted as bycatch.There were 14 females and 10 males, of which three were categorised as neonates, 15 as juveniles and six as adults (Table 1).

Pathological findings at necropsy and histopathology
The respiratory tract was the organ system with the highest number of morphological lesions, including parasites in the bronchial tree and pulmonary blood vessels, bronchopneumonia, and bronchitis as well as foreign bodies, alveolar histiocytosis and calcification (Table 2).The skin and skeletal systems also had a number of lesions, including bleeding and haematoma in the skin, muscles and subcutis, fracture of the skull, netmarks, panniculitis and atrophy of the skeletal muscles.Nematodes in the peribullar and pterygoid sinuses were found in 15 porpoises, and 10 animals showed protein-rich liquid in the scalae of the cochlea.Additional findings were haemorrhage in the melon (Fig. 3), inner ear, acoustic fat surrounding the ear (Fig. 4) and in the lower jaw (Fig. 5), otitis media, and bleeding as well as microgliosis in the brain.All pathological findings are summarised in Table 2.

Imaging characteristics in the tympano-periotic complex
Seven out of 24 scanned animals revealed osseous irregularities, comprising six individuals with dislocations of the middle ear ossicles and one with a micro-fracture in the malleus and fractures in the tympano-periotic complex at the level of the processus sigmoideus and/or region around the mallear ridge (Fig. 2 A-C, 6 A-E).There was no evidence of extraction-related damage in these specimens.

Microbiological findings
Microbiology of various tissues (Table 3) and swabs of 22 harbour porpoises recovered only a few potentially pathogenic bacteria such as Pasteurella multocida, which was causing a septicaemia in one animal.Besides Vibrio spp., no other zoonotic bacteria were isolated.In the majority of cases, bacteriology recovered an unspecified mixed flora, partly in high quantities.A total of 38 different bacteria and fungi genus were found in the 22 harbour porpoises (Table 3).

Virological findings
The virological investigations did not detect morbillivirus, influenza and herpesvirus in any of the 21 investigated cases.

Parasitological findings
Metazoan parasites were in 15/24 harbour porpoises, found in the respiratory (14 cases) and gastrointestinal tracts (stomach compartments: three cases, intestine: two cases), liver (three cases), pancreas (one case), heart (three cases) and ear (15 cases) (Table 2).Parasites in the respiratory tract were identified as Pseudalius inflexus, Torynurus convolutus and/or Halocercus invaginatus, in the ears as Stenurus minor, in the liver and pancreas as Campula oblonga and in the stomach as Pholeter gastrophilus.The parasites in the heart originated from the respiratory tract (Pseudalius inflexus, Torynurus convolutes) and those in the intestine from the ears (Stenurus minor).The analysis showed that the spectrum of parasite species was similar to other surveys (Lehnert et al, 2005;Siebert et al., 2020) and no new species were detected.

Causes of death / main cause of disease
Based on the pathologic dislocation and fracture of the middle ear bones, bleeding in the acoustic fat of the lower jaw, peribullar sinus and melon (Table 4,, the cause of death of eight harbour porpoises was judged to be in association with a blast injury.The geographic location where the animals were found is shown in Fig. 7. Another individual showed lesions consistent with blast injury and pathological findings of a blunt force trauma, which included severe bleeding and haematoma in the muscles and blubber of the left side of the body.The full stomach of this animal was indicative of a sudden death.One juvenile harbour porpoise in good nutritional status and with milk in the stomach had pathologic findings consistent with blast injury, but net marks were also identified which led to the determination that the animal may have succumbed due to bycatch (Table 4).
Two harbour porpoises displayed lesions of blunt force trauma (haemopericardium, massive bleeding and hematoma in the blubber and muscles and fractures of the skull).One healthy animal was handed over as bycatch by a fisherman, with typical net marks, severe pulmonary oedema, and congestion.For two juvenile porpoises in good nutritional status, a cause of death could not be determined (Table 4).

Discussion
Harbour porpoises in German Baltic waters of Schleswig-Holstein are the only native cetacean species and have been studied only since 1990 (Benke et al. 1998).Their habitat is exposed to a large variety of anthropogenic effects (Siebert et al. 2012).Based on genetic investigations, it is evident that harbour porpoises in German Baltic waters belong to two subpopulations: the Belt Sea population and seasonally, to the Baltic Proper (Lah et al. 2016).Ewingella americana The majority of harbour porpoises in the present study were neonates or juveniles (18/24, 75%) with only six adults (6/24, 25%).Previous studies of Baltic harbour porpoises were also dominated by immature animals and only 17.7 to 33.0% of the animals were older than three years (Siebert et al. 2001, 2006, 2020, Wünschmann et al. 2001).Harbour porpoises can be older than 20 years and currently the oldest known female lived to the age of 25 years (Lockyer and Kinze, 2003).In mammals, a high mortality rate during the first months after birth can be expected, but naturally an increase in mortality is only expected towards the end of their life span.In the case of the Baltic Sea harbour porpoises, the life span of free-living harbour porpoises has been truncated over the last decades (Kesselring et al. 2017).
Little is known about the effects of blast injuries in cetaceans (Ketten, 1995(Ketten, , 2006)), in particular on specific tissues at different distances to the source or depths where the individuals are located.Ketten (2006) described a cascade of tissue damage depending on the blast pressure to which individual animals were exposed post-mortem.However, the results from our study show different associations of lesions.The difference in both studies could be due to the fact that porpoises in the Baltic were located at different depths during the exposure, affecting their tissues differently than if they had been located at or near the surface, or it is possible that part of the lesions were sublethal and the individuals could partially recover.Nonetheless, the combination of alterations within the same individual is important to understand and establish the progression and severity of damage that an animal may incur when exposed to explosions.This is a seminal study describing the pathological findings associated with blast injury and lesions that may potentially be detected months after the explosions occurred.Although 10 of the 24 porpoises displayed muscle atrophy, especially the neonates and juvenile animals often showed no or mild lesions and had stomachs filled with either milk or fish remains, indicating an acute death.Sudden death can be caused by several processes, such as 1) bycatch, 2) blunt force trauma and 3) blast injury and/or acoustic trauma.Indications of bycatch are usually net marks, severe pulmonary oedema, congestion of the lungs, bleeding in the ocular fluid and in the area of the net marks (Kuiken et al., 1994;Siebert et al., 2020;IJsseldijk et al., 2021).Evidence of blunt force trauma includes haemorrhages in the muscles, internal organs, dislocations and fractures (Grosse-Brockhoff, 1954).Haemorrhages in the acoustic fat around the lower jaws, melon, cranial sinuses, as well as a change in the ossicles and surrounding bones, indicate blast injury.In particular, dislocation of the ossicles has been described as consequences of blast injury in humans (Danil et al., 2021;Chandler and Edmond, 1997;Mayo and Kluger, 2006).
In the present study, 10 harbour porpoises showed bleeding and haematoma in the acoustic fat of the lower jaw, peribullar area and the melon.There were seven cases of microfractures or dislocation of the middle ear ossicles and tympano-periotic complex are indicative of blast injury.Similar lesions are also described in terrestrial mammals suffering from blast injury (Clemedson, 1956).One of the 10 animals was bycaught showing net marks, and another animal displayed large areas of subcutaneous haemorrhages on one body side.However, both animals may have suffered from the blast injury and consequently not have been fully aware or were unable to use their echolocation system for orientation and avoid fishing nets or collision with e.g. a ship.
Depending on the frequencies and intensities of the blast, indirect damage can occur during explosions secondary to the direct injury to e.g hearing/orientation.These include the displacement of the animals and/or changes in behaviour, for example the interruption of food intake, chronic stress and a reduced ability of the immune system to defend itself (Richardson et al., 1995).Animals can also be permanently or temporarily impaired in their ability to orient and detect obstacles,  and therefore the risk of being bycaught increases.Another possible effect is the separation of mother-calf pairs, which usually results in the death of the calves and cannot be ruled out for the neonates in this study.
One of the major aims of this study was to assess the extent of such detonations within the middle ears of echolocating toothed whales.Herein, we revealed osseous damage potentially associated with blast trauma.In this regard, it is interesting to note that bycaught harbour porpoises from the Baltic Sea have previously shown indications of acoustic trauma, indicating that civil and military explosions in the Baltic have been a threat to these animals in the past (Wohlsein et al. 2019).Previously, one report proposed that such injuries might be more lethal for echolocating toothed whales than to baleen whales, as toothed whales rely on hearing for navigation with reduced perception of gillnets and an increased risk of bycatch (Wohlsein et al. 2019;Siebert et al., 2020).Supporting this hypothesis, toothed whales exhibited fewer remodelled fractures compared to baleen whales, these probably causing a shorter period between injury and death.However, as the precise timepoint of the lesion remains unknown in the animals presented here, the question remains whether the observed alterations were directly associated with the specific mine blast or resulted from unrelated accidents.
In this context, the number of individuals with evidence of the middle ear being affected by blast injury was high compared to a previous survey stating an incidence of cetacean ear bone fractures of 0.1%, pointing towards a relevant extent of external pressure on the toothed whales leading to life-ending results.However, the retrospective study design has methodological limitations, and the reason for the observed injuries cannot be established but may be elucidated by future examinations.
The impulsive blasting noise can affect harbour porpoise hearing in several ways, ranging from irreversible changes in the hearing apparatus (Richardson et al., 1995) to temporarily impaired hearing abilities (Lucke et al., 2009), which is considered as an injury in Germany (BMU 2014).In order to prevent a TTS during the construction of offshore windfarms, a strict noise mitigation concept must be followed in Germany (BMU 2014), consisting of previous deterrence, a soft-start with reduced energy and the compliance with a maximum threshold of 160 dB re 1 µPa 2 s at a distance of 750 m.Although these regulations are the strictest worldwide, further measures have been suggested regarding effects of multiple exposure (Schaffeld et al. 2020) and adjustments for the acoustic deterrence prior to pile-driving (Schaffeld et al. 2019).
Based on underwater noise recordings from 88 controlled underwater explosions related to the clearance of unexploded ordnance on the Dutch continental shelf and corresponding harbour porpoise density in the area, TTS and even PTS events were considered as probable (Aarts et al. 2016).It was found that received levels close to the sea-bed were much higher, resulting in a maximum distance of 2 km, where blast wave ear trauma were estimated to occur very likely (von Benda-Beckmann 2015).
Furthermore, TTS impact zones from underwater clearance were estimated at maximum distances exceeding 11 km (von Benda-Beckmann et al. 2015).This estimated effect distance of blast wave trauma potential cannot be directly transferred to the results of this study.The sound propagation probably differs due to differences in water depth, salinity, temperature, and sediment between the study areas.Also, the source level of underwater explosions is exponentially correlated to the TNT charge mass, which is unknown for the detonations in this study.The findings of this study clearly demonstrate that the deterrence of animals was insufficient, since blast trauma were found in eight animals, which might have been too close to detonation sites.Accordingly, the number of animals which might have been affected by TTS has very likely been been much higher.This shows that noise mitigation and animal deterrence must be improved for future clearance to avoid hearing impairments in noise-sensitive species.Although evidence for hearing impairment exists and effects of blast are more severe than for pile-driving, no legal obligatory noise mitigation concept exists to date in Germany.
Besides direct effects by explosions, triggering of behavioural responses or any impairment of hearing, this impulsive noise can affect the survival rate of individuals and can also have consequences at population level when anthropogenic interventions negatively affect the fitness of many individuals (King et al., 2015).Based on the high mortality of young individuals before and just after reaching sexual maturity (Siebert et al. 2006, 2020, Kesselring et al. 2017) and the critical health status of older individuals in comparison to animals from Arctic areas (Siebert et al. 2001(Siebert et al. , 2006(Siebert et al. , 2009(Siebert et al. , 2020)), it can be confirmed that the status of the harbour porpoise subpopulations in the Baltic is in a critical state (Carlén et al. 2018).As the Baltic Sea still contains large deposits of World War II munition, which continues to degrade and/or conflicts with the increasing number of constructions, removal thereof -including controlled explosions -will continue for many years.If removing the device, representing the preferred method, proves impossible, underwater explosions need to be conducted.These detonations generate very high levels of sound energy in the low frequency range, which can propagate over long distances and these have far-reaching, negative effects on marine mammals (Aarts et al., 2016;von Benda-Beckmann et al., 2015;Merchant et al., 2020;Sertlek et al., 2019;Soloway, 2018).At a distance of up to several kilometres around the site of the detonation, it must be expected that the animals are directly harmed.Therefore, a concept for the Baltic Sea is needed, including a maximum budget of cumulative explosion impact which should be allowed for a certain area and respite to give the animals time to recover from previous impacts.In addition, other impacts such as disturbance from shipping, fisheries, chemical pollution and offshore constructions also need to be considered in the management and protection of harbour porpoises.Some attempts have been made by averting the impact on harbour porpoises before detonating explosive devices, including the use of bubble curtains.For this purpose, also acoustic warning devices (seal scarers) are used for which a deterrent effect has been demonstrated (Brandt et al., 2013a;2013b).However, the use of these warning devices requires further regulations, as it has recently been shown that seal scarers alone also have the potential to cause TTS (Schaffeld et al., 2019).Systematic pathological, microbiological, virological, and parasitological investigations of the 24 harbour porpoises showed that the respiratory tract displayed a large number of lesions including parasitic infections, bronchopneumonia, alveolar histocytosis, calcification and fibrosis.Less frequent were parasitic infections and associated lesions in the alimentary tract, liver, or pancreas.These findings are consistent with studies on harbour porpoises from the same area (Siebert et al. 2001, 2020, Wünschmann et al. 2001), as well as the North Sea (Jepson et al. 2000, Jauniaux et al. 2002, Siebert et al. 2001, IJsseldijk et al. 2020).The bacteria were mainly considered as unspecific.Parasites in the respiratory tract, alimentary system, liver, pancreas, and ears were similar to the previously described spectrum of species (Lehnert et al. 2005, Siebert et al. 2020) and no new parasites were identified.
The systematic microbiological investigations revealed no evidence of septicaemia or severe bacterial infections as found in previous studies (Prenger-Berninghoff, Siebert et al. 2009Siebert et al. , 2020)).No signs of a viral epidemic due to morbillivirus, influenza and herpesvirus were detected at histopathology or viral investigations by PCR so that an epidemic due to these viral diseases among the 24 harbour porpoises could be excluded.So far, larger die-offs due to morbillivirus and influenza virus have only occurred in harbour seals in those waters (Härkönen et al. 2006, Bodewes et al. 2015).
In conclusion, this study presented pathological investigations on protected harbour porpoises in the Baltic Sea stranded after exposure to underwater explosions.It underlines the high potential of harming cetaceans and especially harbour porpoises by acute and chronic, morphological, and behavioural effects by explosions.Moreover, it appears to be of pivotal importance to conduct routine examinations on the ear of all bycaught and stranded harbour porpoises, including highresolution imaging of the inner ear to extend scientific knowledge concerning the incidence and relevance of osseous abnormalities within the middle ear.Likewise, complete post-mortem examinations are required to evaluate the proportion of irregularities detectable within different organs, such as bleeding or haematoma.It appears of particular interest to conduct prospective long-term monitoring of species exposed to explosions, combined with health assessments of necropsies of stranded cetaceans to assess the effects of explosions on cetaceans and evaluate mitigation measures for improved decision-making.This is primarily important as there is a strong need for further investigations to understand and predict the effects from detonation of ordnances on single animals and especially at population level.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Fig. 1 .
Fig. 1.Overview map (top right) showing the location of the study area (red square) in the German Baltic Sea.The large map presents the locations of retrieved dead harbour porpoises in the study area and the MPA Fehmarn Belt.Coloured points illustrate different age classes and the point size reflects the number of porpoises found at this site (small = 1 individual, intermediate = 2 individuals, large = 8 individuals).Points in black boxes show number of porpoises from different age classes found at the same or adjacent sites.(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2 .
Fig.2.a) Tympano-periotic complex of a harbour porpoise, indicating the transverse sectional planes used for histopathology.b) Decalcified tympano-periotic complex of a porpoise in cross-section.c) 3D reconstruction of a tympano-periotic complex of a harbour porpoise by HR-pQCT scan.

Fig. 5 .
Fig. 5. Bleeding and haemorrhages (arrows) in the acoustic fat of the lower jaw of a harbour porpoise.

Fig. 6 .
Fig. 6.Three-dimensional HR-pQCT scans indicate potential blast traumaassociated osseous damage.(A) A representative reconstructed image of an intact tympano-periotic complex (TPC).(B) The high-resolution scan reveals a fractured tympanic (red arrows) at the level of the sigmoid process.(C) Virtual cut-section showing the intact middleear ossicles (i.e.stapes, incus and malleus highlighted by false-colours), marked by a seamless osseous connection.(D) Segmentation of intact ossicles without any signs of dislocation and continuous bony connection.Falsecolour (red) indicates the cut-section.(E)In the specimen with the broken tympanic, in-depth analysis of the HR-pQCT scan revealed dislocations of the ossicles (red arrows).False-colour (red) indicates the cut-section.HR-pQCThigh-resolution peripheral quantitative computed tomography; IMCincudomalleolar complex; Mmalleus; Iincus; Sstapes.(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 7 .
Fig. 7. Map of the German Baltic Sea illustrating locations of the eight retrieved dead harbour porpoises showing indication of blast trauma.

Table 1
Location and date when the 24 harbour porpoises were found, age group, age estimated (in years) and sex.The neonates without teeth are labelled as "-".

Table 2
Pathological findings of the 24 harbour porpoises (necropsy and histopathological results).

Table 3
Microbiological findings of 22 investigated harbour porpoises.

Table 4
Causes of death of 24 harbour porpoises.