ArticlePDF Available

Assessing the Impact of Underwater Clearance of Unexploded Ordnance on Harbour Porpoises (Phocoena phocoena) in the Southern North Sea

Authors:

Abstract and Figures

Large amounts of legacy unexploded ordnance (UXO) are still present in the North Sea. UXO are frequently accidentally encountered by fishermen and dredging vessels. Out of concern for human safety and to avoid damage to equipment and infrastructure from uncontrolled explosions, most reported UXO found in the Dutch Continental Shelf (DCS) are detonated in a controlled way. These underwater detonations produce high amplitude shock waves that may adversely affect marine mammals. The most abundant marine mammal in the DCS is the harbour porpoise (Phocoena phocoena), a species demonstrated to be highly sensitive to sound. Therefore, an assessment of potential impacts of underwater explosions on harbour porpoises was undertaken. Information regarding UXO cleared in the DCS provided by the Netherlands Ministry of Defence was used in a propagation model to produce sound exposure maps. These were combined with estimates of exposure levels predicted to cause hearing loss in harbour porpoises and survey-based models of harbour porpoise seasonal distribution on the DCS. It was estimated that in a 1-y period, the 88 explosions that occurred in the DCS very likely caused 1,280, and possibly up to 5,450, permanent hearing loss events (i.e., instances of a harbour porpoise predicted to have received sufficient sound exposure to cause permanent hearing loss). This study is the first to address the impacts of underwater explosions on the population scale of a marine mammal species. The methodology is applicable to other studies on the effects of underwater explosions on the marine environment.
Content may be subject to copyright.
Aquatic Mammals 2015, 41(4), 503-523, DOI 10.1578/AM.41.4.2015.503
Assessing the Impact of Underwater Clearance of Unexploded
Ordnance on Harbour Porpoises (Phocoena phocoena)
in the Southern North Sea
Alexander M. von Benda-Beckmann,
1
Geert Aarts,
2, 3
H. Özkan Sertlek,
4, 5
Klaus Lucke,
2, 6
Wim C. Verboom,
7
Ronald A. Kastelein,
8
Darlene R. Ketten,
9
Rob van Bemmelen,
2
Frans-Peter A. Lam,
1
Roger J. Kirkwood,
2
and Michael A. Ainslie
1, 10
1
TNO Acoustics and Sonar, Oude Waalsdorperweg 63, 2597 AK, The Hague, The Netherlands
E-mail: sander.vonbenda@tno.nl
2
IMARES, Wageningen UR, Zuiderhaaks 5, 1797 SH ’t, Horntje, The Netherlands
3
Department of Aquatic Ecology & Water Quality Management (AEW), Wageningen University,
Droevendaalsesteeg 3a, Building 100, 6708 PB, Wageningen, The Netherlands
4
Institute of Biology Leiden, Leiden University, Sylviusweg 72, 2333 BE, Leiden, The Netherlands
5
Electronics Engineering Department, Gebze Institute of Technology, PO Box 141, 41400, Gebze, Turkey
6
Centre for Marine Science & Technology, Curtin University, GPO Box U1987, Perth, WA 6845, Australia
7
JunoBioacoustics, Dorpsstraat 1-a, 1731 RA, Winkel, The Netherlands
8
SEAMARCO, Julianalaan 46, 3843 CC, Harderwijk, The Netherlands
9
Hanse Wissenschaftskollegg Institute for Advanced Studies, Neurosciences
and Marine Sciences, Delmenhorst, 27753 Germany
10
Institute of Sound and Vibration Research, University of Southampton, Highfield, Southampton SO17 1BJ, UK
Abstract
Large amounts of legacy unexploded ordnance
(UXO) are still present in the North Sea. UXO are
frequently accidentally encountered by fishermen
and dredging vessels. Out of concern for human
safety and to avoid damage to equipment and
infrastructure from uncontrolled explosions, most
reported UXO found in the Dutch Continental
Shelf (DCS) are detonated in a controlled way.
These underwater detonations produce high
amplitude shock waves that may adversely affect
marine mammals. The most abundant marine
mammal in the DCS is the harbour porpoise
(Phocoena phocoena), a species demonstrated
to be highly sensitive to sound. Therefore, an
assessment of potential impacts of underwater
explosions on harbour porpoises was undertaken.
Information regarding UXO cleared in the DCS
provided by the Netherlands Ministry of Defence
was used in a propagation model to produce sound
exposure maps. These were combined with esti-
mates of exposure levels predicted to cause hear-
ing loss in harbour porpoises and survey-based
models of harbour porpoise seasonal distribution
on the DCS. It was estimated that in a 1-y period,
the 88 explosions that occurred in the DCS very
likely caused 1,280, and possibly up to 5,450,
permanent hearing loss events (i.e., instances of a
harbour porpoise predicted to have received suffi-
cient sound exposure to cause permanent hearing
loss). This study is the first to address the impacts
of underwater explosions on the population scale
of a marine mammal species. The methodology is
applicable to other studies on the effects of under-
water explosions on the marine environment.
Key Words: underwater explosions, harbour
porpoise, Phocoena phocoena, hearing loss,
marine acoustics, anthropogenic noise, Dutch
Continental Shelf
Introduction
Large numbers of unexploded ordnance (UXO),
such as aerial bombs, ammunition, mines, and
torpedoes, mostly legacies of World War II,
remain in the North Sea. UXO are frequently
accidentally encountered by fishermen and dredg-
ing vessels (OSPAR Commission, 2010). There is
a societal need for clearance of UXO as they pose
a risk to offshore activities such as fishing, dredg-
ing, and pipe-laying. For example, in 2005, three
Dutch fishermen were killed by a bomb that was
caught in their net and exploded on deck (OSPAR
Commission, 2010). The Royal Netherlands Navy
(RNLN) is tasked with clearing UXO from the
Dutch Continental Shelf (DCS) and executes on
504 von Benda-Beckmann et al.
average approximately 120 underwater detona-
tions per year.
Underwater detonations of explosives pro-
duce some of the highest peak sound pressures
of all underwater anthropogenic sound sources.
The number of underwater detonations in the
North Sea, as well as their high intensity shock
waves and sound levels, has raised concerns about
their impact on marine life (Nützel, 2008; Ainslie
et al., 2009; Camphuysen & Siemensma, 2011;
Koschinski, 2012). The high amplitude shock
waves and the attendant sound wave produced by
underwater detonations of UXO have the potential
to cause injury or death to marine vertebrates and
invertebrates (e.g., Ketten et al., 1993; Richardson
et al., 1995; Lewis, 1996; Ketten, 2004, 2012;
Danil & St. Leger, 2011).
The harbour porpoise (Phocoena phocoena)
is the most abundant marine mammal species
in the DCS and adjacent waters. In recent years,
the harbour porpoise population has undergone a
redistribution across its North Sea range result-
ing in an increase in abundance in Dutch waters
(Camphuysen, 2011), with some 10,000s of ani-
mals now present (Geelhoed et al., 2012).
Harbour porpoises have sensitive hearing,
making them potentially exceptionally vulner-
able to noise-induced effects from anthropogenic
sound-producing activities at sea (Culik et al.,
2001; Kastelein et al., 2002, 2010, 2012a, 2012b,
2013, 2014a, 2014b, 2015a, 2015b; Ketten, 2004;
Lucke et al., 2009; Tougaard et al., 2009). The
main potential effects of concern of underwater
explosions on an individual animal are (1) trauma
(from direct or indirect blast wave effect injury)
such as crushing, fracturing, hemorrhages, and
rupture of body tissues caused by the blast wave,
resulting in immediate or eventual mortality;
(2) auditory impairment (from exposure to the
acoustic wave), resulting in a temporary or per-
manent hearing loss such as temporary threshold
shift (TTS) and permanent threshold shift (PTS);
or (3) behavioural change, such as disturbance to
feeding, mating, breeding, and resting. Studies
of blast effects on cetaceans indicate that smaller
species are at greatest risk for shock wave or blast
injuries (Ketten, 2004).
Underwater sound plays an important role in
the lives of harbour porpoises. They use sound
actively and passively for social interaction, com-
munication, navigation, predator avoidance, and
foraging (Tyack, 2008; Clausen et al., 2010). Inter-
ference or impairment of their hearing ability as
a result of underwater explosions, therefore, may
directly or indirectly affect their reproduction and
longevity. For instance, loss of important social
bonds due to impaired communication (e.g.,
contact between mother and calf) or loss of prey
could produce significant population-level conse-
quences (National Research Council, 2005).
The objective of this study was to assess the
potential impact of underwater explosions due to
clearance activities of UXO on North Sea harbour
porpoises. Sound levels were estimated for explo-
sions in shallow water, which were then compared
to risk thresholds and distributions of harbour por-
poises. It was decided to focus on explosion levels
high enough to cause hearing loss in harbour por-
poises, which are likely to occur at lower levels
than possible direct traumatic injuries. Potential
behavioural responses of harbour porpoises to
single explosion events were not considered in
this study. A paucity of observations of wild har-
bour porpoise responses to explosions (single
sound transients) made it impossible to address the
severity of responses to single explosion events.
Since there is a lack of acoustic measurements
for large explosions (charge mass much larger than
1 kg TNT equivalent) in shallow waters (bottom
depths less than 50 m) (Weston, 1960; Chapman,
1985; Soloway & Dahl, 2014), measurements
of sound produced by several large underwater
explosives in shallow water environments were
carried out. These measurements were used to
correct deep water source models for explo-
sions and were used in combination with a shal-
low water propagation model to generate maps
of received sound exposure level (SEL) for the
reported explosions in 2010 and 2011. Reported
levels of impulsive sound that cause blast injury to
harbour porpoises were reviewed (Ketten, 2004)
to determine a range of onset thresholds at which
explosions are expected to cause permanent hear-
ing loss. Herein, permanent hearing loss refers to
a permanent reduction (at least partial) in hearing
capability resulting from the blast wave or acous-
tic components of the explosion.
The sound propagation maps, available risk
thresholds, and aerial survey derived harbour por-
poise density estimates (Geelhoed et al., 2012;
Aarts et al., in prep.) were combined to obtain an
estimate of the annual number of possible impact
events (i.e., instances of a harbour porpoise re-
ceiving sufficient sound exposure to cause hearing
loss). The estimated number of impact events were
used to discuss the implications of UXO clearance
for the North Sea harbour porpoise population.
Methods
Overview of Clearance Activities by the RNLN in
the Southern North Sea
Since 2005, the RNLN has recorded the approxi-
mate location, type of ordnance, and date of deto-
nation for destroyed underwater ordnance but not
Impact of Explosions on the North Sea Harbour Porpoise 505
always the charge mass. Because the detonations
can interfere with geoseismic monitoring, all
detonations exceeding 25 kg (TNT eq.) were also
reported to the Royal Netherlands Meteorological
Institute (KNMI). These KNMI reports provided
precise locations, times, and charge masses of the
explosions. For this study, activities over a 2-y
period (2010 and 2011) were investigated using
the combined RNLN and KNMI datasets. These
lists were merged into a final list by removing the
same detonations occurring in both lists; in such
cases, the information in the KNMI report was
retained. In some cases, small counter-charges
used to detonate the main charge were listed sepa-
rately. In such cases, only the main charge mass
(usually much larger) was considered.
The final list comprised 210 explosions of
which 181 were within the DCS and 29 outside
the DCS (a detailed list of explosions can be found
in von Benda-Beckmann et al., 2015). Of the 210
explosions, 62 had an unknown charge mass.
These were most likely smaller explosions (e.g.,
of grenades and other small ammunition) that
were below the 25 kg (TNT eq.) that is required to
be reported to the KNMI. An expert on explosives
estimated the charge mass of these smaller explo-
sives to be in the range of 5 to 20 kg (TNT eq.)
(R. van Wees, pers. comm.); therefore, for the pur-
pose of this study, unknown charge masses were
assumed to be 10 kg (TNT eq.).
Pressure Measurements of Large Explosions
in Shallow Water
The acoustical properties of explosives are sum-
marised by Weston (1960, 1962). The main fea-
ture is a short shock wave comprising a sharp—
almost instantaneous—rise in pressure followed
by an exponential decay with a time constant of a
few hundred microseconds. The expanding shock
wave creates a large pulsating bubble with succes-
sive expansions and contractions that give rise to a
series of weaker, more symmetrical bubble pulses
as happens also with an airgun pulse (Weston,
1960; Cole, 1965). In shallow water, the multi-
ple interactions of the shock and acoustic waves
create a more complex pattern. Because of a lack
of published measurements for explosions in shal-
low water (see Soloway & Dahl, 2014), especially
for a charge mass much larger than 1 kg, a field
experiment was carried out in 2010 to measure
multiple detonations of aerial bombs at varying
ranges up to approximately 2 km from the detona-
tion site.
Measurements of underwater pulse dura-
tion (in s), sound exposure level (SEL in dB re
1 μPa
2
s), and peak overpressure (in kPa) were
obtained in September 2010 for the underwater
explosions of seven aerial bombs (six 1,000 lb and
one 500 lb with TNT eq. charge masses of 263 and
121 kg, respectively). These bombs were found on
land and detonated on the seafloor in an area with
a water depth of 26 to 28 m and a sandy bottom.
Measurements were obtained using a hydrophone
(B&K 8105) and ICP tourmaline pressure gages
(type W138A01/M038CY060), measuring at a
sampling rate of 50 kHz, set at depths of 4, 5, 13,
15, 23, and 25 m, and distances between 100 and
2,000 m from the source.
Underwater explosions caused a series of typi-
cal pressure changes (see Figure 1). Following
Madsen (2005), the pulse duration, t
90, was mea-
sured as the time containing 90% (5 to 95%) of
the energy of the signal. The SEL within one tenth
of a decade (decidecade) bands as well as the
broadband level were measured from each explo-
sion within the pulse duration in a frequency band
of 0 to 20 kHz. The peak overpressure (or peak
compressional pressure in kPa) of the shock wave,
defined as the maximum value of the positive
overpressure (pressure minus ambient pressure,
when this quantity is positive), was measured
within a set time interval (herein taken to be 0.5 s
to cover the full explosion event).
Risk Thresholds for Hearing Loss for
Harbour Porpoises Exposed to Explosions
This study focused mainly on the risk of explo-
sions leading to permanent hearing loss as a result
of direct blast injury caused by the shock wave
close to the source or PTS from the acoustic wave
of the blast incident farther away from the source.
Permanent hearing loss estimated herein may be
partial (such as acoustically derived damage to
some portion of the inner ear), occurring within a
specific frequency range, or total (such as severe
trauma to the auditory bulla or middle or inner
ear tissues), depending on which components of
the auditory system are affected and how they
are affected. More severe trauma was expected to
occur at higher levels than those required to cause
permanent hearing loss and, as such, this was con-
sidered as a lower limit.
Permanent and Temporary Threshold Shifts
Marine mammals exposed to intense underwater
sounds may suffer hearing loss, resulting in hear-
ing threshold shifts that may be temporary (TTS)
or permanent (PTS). The extent of the hearing
threshold shift and the frequency range affected
may depend on the sound level, spectral content,
temporal pattern, and exposure duration. TTS
is defined as a threshold shift that recovers to
normal sensitivity. The course and time of recov-
ery depend on several factors: the individual’s
recovery ability, the sound to which an animal
was exposed, and the amount of shift that incurred
(Kastelein et al., 2012b, 2014b, 2015a; Finneran
506 von Benda-Beckmann et al.
& Schlundt, 2013). TTS onset defined herein is
consistent with the criteria used in most studies
on marine mammals—that is, as a TTS of 6 dB
or greater measured shortly (1 to 4 min) after ces-
sation of the exposure, following Southall et al.
(2007).
Different TTS onset SEL thresholds have been
measured in harbour porpoises for a variety of
pulse types (Lucke et al., 2009; Kastelein et al.,
2012b, 2014b, 2015a, 2015b). Lucke et al. (2009)
determined a masked TTS-onset SEL level for a
harbour porpoise subjected to single airgun tran-
sients, measuring auditory evoked potentials. As the
sound stimuli of Lucke et al. resembled explosion
sound better than other TTS studies currently avail-
able (Kastelein et al., 2012b, 2014b, 2015a, 2015b;
Tougaard et al., 2015), our estimates for SEL thresh-
olds for TTS and PTS were based on that study.
There is currently no experimental data available
for predicting PTS-onset levels in marine mam-
mals. Observations for humans show that a TTS
of 40 to 50 dB results in a significant risk of PTS
(Kryter et al., 1966; Kryter, 1994). Southall et al.
(2007) used reported TTS onset measurements for
marine mammals to estimate the SEL required for
a 40 dB TTS based on the TTS growth curve for
land mammals, which has a 2.3 dB increase in TTS
per 1 dB increase in SEL (i.e., the TTS growth rate
is 2.3 dB/dB). Based on this, PTS was predicted
to occur for SEL values 15 dB above SEL, caus-
ing TTS onset (Southall et al., 2007). Recent mea-
surements for continuous noise exposure and sonar
sound suggest that the growth rates may well exceed
2.3 dB/dB in some cases (Kastelein et al., 2013,
2014b). Given the lack of data on growth rates in
harbour porpoises for impulsive sounds, let alone
sound from explosions, the approach for extrapolat-
ing the TTS growth that is outlined by Southall et al.
(2007) was used.
To account for the frequency selective sensi-
tivity of the mammalian hearing system in sound
exposure calculations, frequency-selective weight-
ing functions are often used (Fletcher & Munson,
1933; Suzuki & Takeshima, 2004; Southall et al.,
2007). Several weighting curves for harbour
porpoises have been proposed in the literature
(Verboom & Kastelein, 2005; Southall et al., 2007;
Finneran & Jenkins, 2012; Wensveen et al., 2014;
Tougaard et al., 2015). No frequency weighting
study has been conducted for an explosive sound
source. Because air guns are an impulsive and low
frequency source, and are fairly representative of
Figure 1. Examples of recorded sound pressure time-series of a detonation of a 263-kg TNT equivalent aerial bomb
(measured at a sampling frequency of 50 kHz), recorded at a depth of 15 m at a distance of 982 m (upper) and at 13-m depth
at a distance of 1,978 m (lower) from the explosion site. An early arrival of the ground shock wave can be observed, followed
by the water-borne shock wave (at time = 0 s), and reflections and bubble oscillations.
Impact of Explosions on the North Sea Harbour Porpoise 507
an explosion sound at larger distances in shallow
water, the data from an airgun study (Lucke et al.,
2009) were used to define TTS onset. Therefore, no
frequency weighting was applied to the SEL when
estimating the risk of TTS and PTS for explosions.
Lucke et al. (2009) concluded that the pre-
defined TTS criterion (measured at 4 kHz) was
exceeded at received SELs greater than 164 dB re
1 μPa
2
s. Based on this, it, therefore, was assumed
that TTS onset was very likely to occur for har-
bour porpoises exposed to explosion sounds
at an (unweighted) SEL of 164 dB re 1 μPa
2
s.
Following Southall et al. (2007), PTS onset was
then estimated by adding 15 dB to the TTS onset
SEL. Given the uncertainties involved in this
extrapolation, this threshold should be considered
as a lower limit below which PTS is considered
unlikely. The resulting lower limit for onset of
PTS occurred at an unweighted SEL greater than
179 dB re 1 μPa
2
s. At higher SELs, the risk of PTS
was considered to become increasingly likely. To
denote a threshold where the onset of PTS was
very likely to occur, the following section consid-
ers the observations of ear trauma observed in har-
bour porpoises exposed to underwater explosions.
Primary Blast Injury—A study on primary blast
injury in marine mammals caused by shock waves
was reported by Ketten (2004). In that study,
fresh odontocete cadavers (including harbour por-
poises) were exposed to explosions from varying
charge masses in a controlled environment. In
some cases, the animals were tested within 24 h
postmortem after chilling. In others, they were
frozen within 4 h postmortem and prepped by
thawing under controlled conditions (chiller and
water bath). In all test specimens, the cadaver was
scanned on receipt when fresh and rescanned just
prior to testing to ensure that the lungs, brain, air-
ways, and other major organs had tissue appear-
ances consistent with normal anatomy and that
there were no significant changes from their fresh
condition. The specimens were then implanted
with pressure gages for recording internal received
PSI and again scanned to document the position
of the gages. The necropsy and measurement of
the blast effects were performed by a team con-
sisting of experts on marine mammal physiology
as well as experts on explosion-induced trauma.
Trauma found included mild to severe and likely
lethal injuries, particularly haemorrhages, as well
as lung, esophageal, liver, brain, and ear injuries.
At peak blast, overpressures exceeding 172 kPa
(25 psi) ear trauma were always observed, which
was consistent with the ear being the most sen-
sitive organ for blast-related trauma. No effects
were observed for peak overpressures below 69 to
83 kPa (10 to 12 psi).
Blast trauma to the harbour porpoise’s middle
nd inner ear would likely result in a permanent,
cute hearing loss, which could be broad spectrum
ch as in the case of middle ear ossicular chain
isarticulation (a probable 36 dB loss overall
ue to the dysfunction of the middle ear bones;
chuknecht, 1993) or elevated thresholds in only
me frequencies, depending upon the received
coustics impacting the inner ear. Observations of
ar trauma and related peak overpressures, there-
re, could serve as a proxy for moderate to severe
ermanent hearing loss.
Dose-Response Relationship for Hearing
oss—To delimit the range of SEL values at
hich onset of permanent hearing loss—either
oise-induced PTS or ear trauma caused by the
last wave—could occur, levels at which ear
auma resulting from primary blast injury was
bserved were considered. Peak overpressure was
sed by Ketten (2004) to predict the occurrence
f primary blast injury. Here, instead, the peak
verpressure was empirically related to SEL for
easured explosion sounds in shallow water due
the difficulties of predicting peak overpressure
shallow water. From those measured data, an
ffective SEL that corresponded to peak over-
ressures resulting in ear trauma was estimated.
corresponding upper limit for the onset of PTS
as estimated using the observations by Ketten
004) in combination with measured SEL and
eak overpressure for shallow water explosions
ee Appendix A for full details). The expectation
as that levels just below those required to cause
inimal detectable ear trauma were very likely
be sufficient to cause PTS (i.e., the border-
ne between detectable blast injury and acoustic
ffects). Table 1 summarizes the SEL thresholds
dopted in this study for the risk of temporary
nd permanent hearing loss for harbour porpoises,
hich were considered appropriate for harbour
orpoises exposed to underwater explosions in
allow waters (less than 50-m depth).
odelling of Sound Exposure Levels of
xplosions at Large Distances in Shallow Water
lthough the measurements of the charges pre-
nted in the previous section are representative
f typical UXO found in the North Sea, the inven-
ries of the 2010 and 2011 detonations showed a
ide range of explosive charge masses found in
range of water depths. The water depth affects
e sound propagation characteristics, particu-
rly of low frequency sound. A source model
as required, therefore, to estimate the effect of
nderwater explosions for which there were no
mpirical data. The model used in this study for
redicting the impact area for harbour porpoises
a
a
su
d
d
S
so
a
e
fo
p
L
w
n
b
tr
o
u
o
o
m
to
in
e
p
A
w
(2
p
(s
w
m
to
li
e
a
a
w
p
sh
M
E
A
se
o
to
w
a
th
la
w
u
e
p
consisted of two parts: (1) a source model for
508 von Benda-Beckmann et al.
underwater explosions and (2) a sound propaga-
tion model for shallow water environments.
The source model was based on the empiri-
cal deep-water formula for explosion energy as
a function of distance by Weston (1960). That
model contained the energy contribution of both
the shock wave as well as the first two bubble
oscillations. From the Weston empirical relation, a
linear source model was estimated by considering
the energy spectrum at a distance of 5,000 charge
radii beyond which the sound wave is assumed to
become linear (Ainslie, 2010) and back-comput-
ing this spectrum to a source energy level using
spherical spreading. Thus, a spectral energy level
was obtained for each decidecade frequency bin in
a range from 10 Hz to 20 kHz.
The propagation model SOPRANO, proposed
by Sertlek & Ainslie (2014) and Sertlek et al.
(in prep.), was used for calculating propagation
loss in shallow water environments. SOPRANO
is a hybrid model based on normal mode and
flux theories. It is a fast analytical model which
can estimate incoherent propagation loss with
an accuracy similar to that of other propagation
models (Sertlek et al., in prep.). The SEL was then
obtained by subtracting the propagation loss from
the source energy level.
This model enabled the generation of sound
maps (i.e., geographical distributions of SEL for
each decidecade band) with high resolution and
broad range frequencies, including range-depen-
dent bathymetry (with adiabatic approximation)
and sediment type effects. For bathymetric data,
the EMODNET database was used, which has a
grid size of 0.125 × 0.125 min (www.emodnet-
hydrography.eu). Information for describing the
geoacoustical properties of the seabed was obtained
from the DINO database (https://www.dinoloket.
nl). Propagation loss was calculated with 25-m
resolution, and the computed SEL was then
mapped onto a sound map with resolution of 1 km
× 1 km (Figure 2). Sea surface was assumed to
be flat, and effects of non-constant sound speed
profiles were neglected.
To relate effects of explosions to the risk of hear-
ing loss, an estimate for the peak overpressure was
Table 1. Thresholds related to temporary and permanent hearing loss caused by a single underwater explosion in shallow
water (< 50-m depth). Permanent hearing loss can be either noise-induced permanent threshold shift (PTS) or be due to ear
trauma caused by the blast wave. The right column indicates the best estimate for unweighted broadband (measured in 0
to 20 kHz) SEL risk thresholds for “permanent hearing loss” induced by explosions. To indicate the chance of an effect to
occur, the following terminology was adopted: “very likely” indicates a probability exceeding 95%, and “unlikely” indicates
a probability of less than 5%. “Increasingly likely” is then anything between 5 and 95% probability. Arrows = ranges of SEL
thresholds for onset of permanent hearing loss that are estimated to occur (see text and Appendix A for more details).
SEL Noise- Blast wave-
(unweighted) Noise- induced induced ear Permanent
(dB re 1 µPa
2
s) induced TTS PTS trauma hearing loss
> 203 Very likely
Very likely* Very likely
Increasingly
190-203
likely
Very likely
<
Increasingly Increasingly
179-190
likely likely
<
Unlikely
164-179
Unlikely Unlikely
< 164 Unlikely
*Based on expert judgement
Impact of Explosions on the North Sea Harbour Porpoise 509
required. For deep water explosions, peak over-
pressure can be estimated using empirical models
(Weston, 1960; Chapman, 1985). However, in
shallow water, estimation of peak sound pressure
is challenging because the estimate is very sensi-
tive to bandwidth and geometry (which is poorly
known). Uncertainties in geometry lead to small
phase uncertainties in arrivals of different sound
paths, which can have a large effect on the pre-
dicted peak pressure. The estimate of peak sound
pressure, therefore, would require a sensitivity
study on top of the existing computation load for
a single time series. As this was not possible here,
peak overpressures for which trauma from explo-
sions were observed in harbour porpoises were
related instead to effective shallow water SEL
thresholds. These were more easily calculated and
robust to environmental uncertainties. The empiri-
cal relationship between the measured peak over-
pressure and SEL is described in Appendix A.
Harbour Porpoise Density Estimates
Line transect surveys were carried out in the
Dutch section of the North Sea during four sea-
sons in 2010 and 2011 (i.e., in March, July, and
October/November 2010, and in March 2011;
Geelhoed et al., 2012; Scheidat et al., 2012), pro-
viding a large temporal overlap with the occur-
rence of explosions. In order to estimate absolute
abundance, one needs to correct for nondetect-
ability (e.g., Hiby & Lovell, 1998). To do so, the
one-sided effective strip width (taking the detec-
tion probability or g[0] values into account) was
calculated and defined as 76.5 m for good sight-
ing conditions and 27 m for moderate sighting
conditions (Scheidat et al., 2005, 2008; Gilles
et al., 2009). Next, for each grid cell (10 km ×
10 km), the number of harbour porpoise sight-
ings and the effective surveyed area were calcu-
lated. A Bayesian spatial model was fitted to the
data wherein the number of sightings in each grid
cell was treated as a Negative Binomial distrib-
uted count, and the log of the area surveyed was
treated as an offset. Spatial variability in sighting
rate was modelled as a latent Gaussian random
field using a two-dimensional autoregressive cor-
relation function of order-1 (Rue et al., 2009). The
final seasonal absolute harbour porpoise density
estimates (illustrated in Figure 3) were overlaid
with the generated soundmaps. The number of
affected animals receiving an SEL as defined by
the risk thresholds was then estimated by sum-
ming the number of predicted animals for each
grid cell within a contour on the sound exposure
map (for more detail, see Aarts et al., in prep.).
This resulted in an estimate of the number of har-
bour porpoises affected by each explosion.
Figure 2. Examples of unweighted broadband SEL maps for a single explosion (263-kg TNT eq. charge mass) calculated
for different receiver depths (at 1 m above the sea floor, left, and 1 m from the sea surface, right) computed with the model
described in Sertlek & Ainslie (2014) and Sertlek et al. (in prep.). The water depth at the detonation site is approximately 28 m.
510 von Benda-Beckmann et al.
Results
Overview of Explosion Activities by the RNLN in
the Southern North Sea
In 2010 and 2011, detonations were located pre-
dominantly in the southern DCS (Figure 4).
Reported charge masses range from 10 to 1,000 kg,
with most at 125 to 250 kg (Figure 4). Most detona-
tions occurred in water depths between 20 and 30 m.
There was a distinct seasonal pattern to the
explosions, with a peak in March and smaller
peaks in August and November (Figure 5). The
March peak coincided with a peak of fishing
activity and, thus, an increase in encounter rate of
UXO (OSPAR Commission, 2010).
Measured SEL and Peak Overpressure of
Explosions in Shallow Water
In the measured explosions, large differences in
received levels were noticeable. They were attrib-
uted to super-positioning of direct and surface
reflected paths (Figure 6). SELs were on average
lower near the surface than near the bottom or in
the middle of the water column. This suggested
that the depth where the harbour porpoise was
most likely to be located needed to be considered
when discussing the impact of the sound.
Figure 3. Example of the model-based estimates of harbour porpoise density (grey scale: number of harbour porpoises/ km
2
)
within the Dutch Continental Shelf (DCS; polygon) for March 2011 based on the aerial survey performed in that month.
Impact of Explosions on the North Sea Harbour Porpoise 511
Comparison of Measured and Modelled Sound
Levels of Explosions in Shallow Water
Measured explosions were run through the propa-
gation model to test its efficacy (Figure 7). The
model was in agreement with the data at frequen-
cies containing the peak energy in the spectrum
between the cut-off frequency of ~20 and 100 Hz
but systematically overestimated the measured
SEL at frequencies above 100 Hz (Figure 7, right
panel).
The systematic overestimate of SEL by the
model (likely causes are interpreted in the
“Discussion” section) resulted in an overestima-
tion of 8.1 dB (Standard Deviation [SD] = 3.6 dB)
of the broadband SEL. Such an overestimate
would lead to a systematic overestimation of the
impact ranges and, therefore, required a correction
factor. For this, the power mean over each decide-
cade band was subtracted from the modelled SEL.
After application of the correction factor, the
modelled broadband SEL agreed within 0.1 dB
(SD = 3.7 dB) from the measured broadband SEL
(see Figure 6) and was used to estimate impact
distances and number of affected animals.
Effect Distances for Shallow Water Detonations
Based on Field Measurements
The largest distance at which the peak overpres-
sure corresponded to risk of observed ear trauma
was at approximately 500 m (Figure 8). Between
Figure 4. Geographical distribution of detonation events for the years 2010 and 2011 (left). Distribution of charge mass (TNT
equivalent) (top right) and depth (bottom right) at the charge detonation site for the years 2010 and 2011. The 62 explosions
for which charge masses were not available are indicated by the grey bar in the left plot. A charge mass of 10-kg TNT-eq. was
assumed for these for the purpose of this study (see text for details).
Figure 5. Number of explosion events per month in the North
Sea for the years 2010 and 2011 based on the information
provided by the Netherlands Ministry of Defence
512 von Benda-Beckmann et al.
1,200 and 1,800 m, the peak overpressures fell
below the limit at which no ear trauma occurred.
The minimum SEL measured within 2 km was
191 dB re 1 µPa
2
s, which exceeded by 1 dB the
SEL-based risk threshold above which PTS was
considered very likely (190 dB re 1 µPa
2
s), and
exceeded by 12 dB, the lower limit of PTS onset
(179 dB re 1 µPa
2
s).
Modelled Effect Distances for Harbour Porpoises
Exposed to Explosions in 2010 and 2011
Model predictions of effect distances as a func-
tion of SEL threshold were made for all explosion
events occurring in the DCS in the years 2010 and
2011 (Figure 9). Effect distances for the lower
limit of PTS (179 dB re 1 µPa
2
s ) varied between
hundreds of meters and 15 km, which were higher
than the empirical relationship by Weston (1960)
for deep water environments. There was a trend of
increasing effect distances with increased charge
mass, with substantial scatter due to variations in
water depth in which explosives were detonated.
This indicated that the water depth in which the
explosion occurs has a significant influence on the
effect range for a given charge mass (Figure 10).
The density of harbour porpoises in the DCS in
2010 was dependent on season, varying between
1.34 animals/km
2
in early spring to 0.60 animals/
km
2
in summer (these model-based values are
similar to the values reported by Geelhoed et al.,
2012, and Scheidat et al., 2012, based on tradi-
tional distance analysis). Thus, a fixed impact
Figure 6. The 263-kg charge mass detonations measured in 2010: shown are SEL (dB re 1 μPa
2
s) (left), peak overpressure
(kPa) (right) measured in a frequency band of 1 Hz to 20 kHz, at various distances from the detonation site, with recorders at
varying depths (near surface = circles, midwater = squares, near bottom = triangles). SEL values for measurements close to
the surface are lower due to pressure release of sound near the surface.
Figure 7. Left: Comparison of decidecade band SEL levels as a function of frequency for a midwater measurement at ~2-km
distance (black) from an explosion, with model predictions (uncorrected: grey; corrected: dashed grey). Right: Comparison
of all decidecade band SEL levels difference between model and measurements (grey lines), showing that the model tends
to systematically overpredict the measured SEL, particularly at higher frequencies. The black line indicates the power mean
used as the empirical correction factor to the source level used to correct for the overestimation of the model.
Impact of Explosions on the North Sea Harbour Porpoise 513
distance would affect a different number of
animals depending on the season of the event.
Herein, explosions occurring in the DCS between
March 2010 and March 2011 were considered (88
in total), providing a maximal temporal overlap
with the aerial line-transect surveys in the DCS.
By comparing modeled effect areas to measured
harbour porpoise densities, the mean number of
affected animals and the total number of impact
events as a function of SEL threshold per explo-
sion were estimated (Figure 11; Table 2).
Harbour porpoises were likely to receive lower
SEL when near the surface than when at depth
(see Figure 6). To account for this in the expo-
sure determinations, it was assumed that harbour
porpoises spent 50% of their time near the surface
and 50% near the bottom (Table 2). This assump-
tion was based on published data for normal har-
bour porpoise behaviour (Westgate et al., 1995;
Akamatsu et al., 2007). The average number of
animals affected by each underwater explosion
was then estimated by taking the mean of the
number of animals exposed at 1 m below the sur-
face and 1 m above the bottom.
A typical underwater explosion was esti-
mated to lead to PTS in multiple individuals,
with mean overall explosions ranging from 8 to
51 individuals, depending on the chosen SEL
threshold level, and TTS in several hundreds of
individuals (Figure 11). Although the impact area
and corresponding number of harbour porpoises
increased as a function of charge weight, the rela-
tionship was weakened by spatial and seasonal
variability in harbour porpoise density and loca-
tion of the explosion relative to topography (i.e.,
shallower regions more strongly attenuate sound).
Between March 2010 and March 2011 (the
period with temporal overlap between harbour
porpoise surveys and explosions), a total of 88
UXO were detonated by the RNLN in the DCS.
These would have resulted in a total of approxi-
mately 5,450 impact events (i.e., instances in
which the SEL at a harbour porpoise exceeded
the estimated risk threshold) for the lower limit
for permanent hearing loss and more than 28,000
TTS-onset impact events (Table 2). This was pos-
sibly an overestimation of the number of individ-
ual animals affected since some may have been
affected on several occasions. The frequencies of
multiple exposures depended on how individual
animals move, but the probability of multiple
exposures for high SEL thresholds that lead to
PTS was small (Aarts et al., in prep.).
Figure 8. Effect of distances based on measured peak overpressure for a charge mass of 263 kg in water depth of 26 m.
Superimposed in grey are risk thresholds for blast injury from Ketten (2004).
514 von Benda-Beckmann et al.
Based on the limited number of studies of actual
explosion-induced ear trauma in fresh cadavers
(Ketten, 2004), it was judged that a SEL >190 dB
re 1 µPa
2
s would very likely result in permanent
hearing loss. The model estimated an annual total
of approximately 1,280 impact events in which
animals were very likely to sustain permanent
hearing loss (Table 2).
Figure 9. Modelled effect distances for a harbour porpoise at 1 m above the bottom (top) and 1 m below the surface (bottom)
as a function of SEL threshold. Each black curve indicates a single explosion in the years 2010 and 2011. Vertical lines
bordering the pink shaded areas represent the TTS, PTS, and ear trauma onset threshold values (see Table 1).
Impact of Explosions on the North Sea Harbour Porpoise 515
Figure 10. Modelled effect distances for a harbour porpoise at 1 m from the bottom as a function of charge mass for the lower
limit for PTS (179 dB re 1 μPa
2
s). Symbols indicate different water depths at the location of explosion, showing lower effect
distances for shallow water explosions. For comparison, the dashed line shows the deep-water prediction using the Weston
(1960) model.
Figure 11. Estimated number of harbour porpoises affected by each explosion (black lines) as a function of received
(unweighted) SEL for 88 explosions occurring in the DCS in the period between March 2010 and March 2011, assuming
animals are 50% at the surface and 50% near the bottom. Model SEL estimates were empirically corrected to match field
acoustic measurements (see “Methods” section for details). Vertical lines bordering the pink shaded areas represent the TTS,
PTS, and ear trauma onset threshold values (see Table 1).
516 von Benda-Beckmann et al.
Discussion
Implication of Underwater Explosions for the
North Sea Harbour Porpoise Population
Risk of Permanent Hearing Loss—This study
aimed to quantify the potential impact on harbour
porpoises of underwater ordnance explosions
on the Dutch Continental Shelf conducted by
the RNLN. It was estimated that in a 1-y period
(March 2010 to March 2011), the 88 explosions
that occurred in the DCS very likely caused 1,280,
and possibly up to 5,450, permanent hearing loss
events, with some animals potentially exposed
multiple times.
The implication of a permanent hearing loss
for a harbour porpoise depends on the severity of
the hearing loss and the frequency range that is
affected. Because of the configuration and map-
ping of frequencies in the inner ear, the impulse
from a blast results in a broad movement of the
inner ear membranes that is not frequency spe-
cific. Since high frequencies are encoded shortly
after the entry to the cochlea, they are subject to
impulse-related effects even though the peak in
the spectrum of the signal is lower. Such effects
were also observed in Kastelein et al. (2014a),
who reported TTS to occur at increasingly higher
frequencies with increasing noise exposure. When
received levels of an explosion impulse exceed
the threshold for causing a permanent hearing loss
to part of the harbour porpoise’s hearing range, a
wider range of hearing is also likely to be nega-
tively affected. Although the frequency range
affected remains unknown, when explosion levels
are high enough to cause primary blast injury
resulting in ear trauma, exposures near those
levels will very likely result in a large reduction
in hearing ability over a broad frequency range.
While the dominant parts of the sound energy
from the explosion shock wave is contained in
low frequencies (< 1 kHz), harbour porpoises
are known to echolocate at frequencies above
100 kHz (Møhl & Andersen, 1973). They do
produce low-level sounds at frequencies below
1 kHz that have been attributed to communica-
tion (Verboom & Kastelein, 1997), while others
argue that they communicate solely at frequencies
> 100 kHz (Hansen et al., 2008; Clausen et al.,
Table 2. The estimated average number of harbour porpoises impacted with a temporary or permanent hearing loss by a
single explosive clearance and the total impact events (i.e., instances where a harbour porpoise was exposed to levels high
enough to cause a specified effect) from the total of 88 explosions in the Dutch Continental Shelf (DCS) waters between
March 2010 and March 2011. The effects considered were temporary and permanent hearing loss by explosions. Estimates
are made for near the sea surface (i.e., at 1 m depth), near the bottom (i.e., 1 m above the bottom), and for the mean of these
two, which assumed 50% of the harbour porpoises were near the surface and 50% were near the bottom.
# of harbour porpoises
(single explosion – average)
# of impact events
(88 explosions)
Type
Threshold
unweighted
SEL
(dB re 1 μPa
2
s)
If all
near
surface
If all
near
bottom
If 50% near
surface and
50% near
bottom
If all
near
surface
If all
near
bottom
If 50% near
surface and
50% near
bottom
Blast wave
ear trauma 203 0 1 1
1 119 60
Very likely
Permanent
hearing loss
Very likely
190 2 27 15
205 2,362 1,283
Permanent
hearing loss
Increasingly
likely
179 18 106 62
1,584 9,318 5,451
TTS
Very likely
164 190 448 319
16,721 39,413 28,067
Impact of Explosions on the North Sea Harbour Porpoise 517
2010). Harbour porpoises rely on their echoloca-
tion ability on a daily basis to catch prey and navi-
gate (Kastelein, 1997). Any impairment of their
echolocation hearing abilities would likely lead to
significant negative effects on their fitness. The
implications of TTS are less clear. The estimated
number of impact events for TTS by the model
suggested that a large number of animals could
have been exposed to one instance of TTS, and
some could have experienced multiple exposures
each year (see Aarts et al., in prep.). In principle,
TTS is expected to recover quickly (within hours
to days). Minor TTS in a low-frequency range out-
side the frequency regimes used for communica-
tion and echolocation are unlikely to significantly
affect the animal’s fitness. Repeated or severe
TTS might result in noticeable, even if transitory,
damage to the inner ear at the submicroscopic
level (Kujiwa & Liberman, 2009), which could
have significant consequences for fitness and sur-
vival. It is unknown whether harbour porpoises
would have responses similar to the species tested
in Kujiwa & Liberman.
The detonation activities in the DCS strongly
peaked in March when the harbour porpoise abun-
dance also peaks (Geelhoed et al., 2012). The per-
centage of animals estimated to have suffered from
permanent hearing loss per year due to exposure to
underwater explosions in the DCS could be at least
1.3% and potentially up to 8.7% of the harbour
porpoise population present in the DCS (based on
the average population size in the DCS of the four
survey estimates in Geelhoed et al., 2012).
Uncertainty in Estimated Impacts
The predicted impacts contained uncertainties,
which were attributed to the following factors:
(1) difficulty of predicting the generation and
propagation of underwater shock waves in shal-
low water at large distances, (2) lack of data
on sensitivity of harbour porpoises exposed to
explosion shock waves and broadband impulsive
sound, and (3) a lack of knowledge on habitat
use and movement patterns of harbour porpoises
in the North Sea. The main uncertainties in the
model predictions are discussed in the following
sections. Uncertainties in habitat use are discussed
in Aarts et al. (in prep.).
Sound Exposure Levels of Explosions in
Shallow Water
—The impact assessment relied on
the validity of the propagation model used to esti-
mate SEL for different explosive charge mass and
water depths. With the adopted deep water model
for the explosion source energy level from Weston
(1960), prior to correction, the model systemati-
cally overestimated the observed broadband SEL
(by approximately 8 dB), particularly at frequen-
cies higher than 200 Hz.
Different mechanisms could be responsible for
the overestimation of the (uncorrected) model.
For example, interaction with the sea floor for an
explosive detonated near or on the bottom, energy
losses due to a blow-out as the shock wave first
reaches the surface, energy loss by cavitation close
to the surface, a lower effective charge mass due
to degradation of the explosive charge, or propa-
gation losses due to multiple interactions with
wind-generated bubbles at larger distances could
all lead to reductions in the estimated resulting
SEL (Cole, 1965; Ainslie, 2010; Pfeiffer, 2014).
The explosion source model assumed that no
blow-out at the surface occurred. Surface blow-out
may lead to pressure release in the bubble, leading
to lowered (horizontal) radiation efficiency. This
may occur for large explosives in very shallow
areas; however, the effect on the radiated energy
in the shock wave at large horizontal distances is
not well understood (Weston, 1960). No surface
blow-out occured during the measured explosions,
which were both typical in charge mass and water
depth (see Figure 4). The UXO were also assumed
to detonate at full power, and possible degrada-
tion that may have occurred due to long contact
with sea water (Pfeiffer, 2014) was not taken into
consideration. Degradation occurs more strongly
for some types of UXO (especially thin-walled
mines or aluminium torpedoes), but other types
of ammunition show less signs of strong deterio-
ration (Pfeiffer, 2014). Since most of the larger
UXO found at sea by the RNLN are aerial bombs,
the assumption of full power detonation is likely
a reasonable assumption for the majority of the
UXO considered herein (von Benda-Beckmann
et al., 2015).
To adjust for the overprediction in SEL, the
acoustic source model was corrected to match
measured SEL levels at various depths. Because
these experiments were representative of typical
charges and water depths at which these detona-
tions occur, the corrected model predictions were
expected to give a realistic estimate of the impact
distances for most of the explosions modelled.
Also, for very small charges (~1-kg TNT eq.
charge mass), the uncorrected source model sys-
tematically overestimated broadband SEL mea-
surements by Soloway & Dahl (2014) by ~7.6 dB
(SD = 2.1 dB) at distances up to 1 km. This over-
estimate also occurred for explosions in midwa-
ter, and these explosives were calibrated charges.
As a result, uncertainty in charge mass and effect
of bottom interactions of a bottom explosion
were unlikely causes for the model overestimat-
ing the SEL of an explosion. After applying the
empirical correction factor proposed in this study,
a better match was obtained for the Soloway &
Dahl (2014) measurements. The corrected model
518 von Benda-Beckmann et al.
slightly underestimated the Soloway & Dahl
(2014) measurements by 1.5 dB (SD = 2.3 dB)
(1-sigma STD).
A plausible cause for the overestimation using
the uncorrected model was the effect of the cavi-
tation zone on the propagation of the generated
shock wave. The cavitation zone created by the
shock wave was likely to lead to energy loss and
also could act as a bubble screen for the bottom-
reflected paths, including the shock wave. In deep
water, shielding by the cavitation zone is unlikely
to have such a noticeable effect on the measured
SEL because a large fraction of the bottom reflec-
tions would have travelled beyond the cavitation
zone or would arrive much later than the direct
blast. The size of the cavitation zone would be
a function of the charge mass and depth of the
explosion. The slight overestimation observed
when comparing the model to the Soloway & Dahl
(2014) datasets agreed with this interpretation as
a smaller charge mass would result in a smaller
cavitation zone and, hence, less absorption.
Another effect that becomes important at large
distances is the attenuation by wind-generated
bubbles. This is most noticeable at higher fre-
quencies and longer distances. Attenuation by
wind-generated bubbles could in part explain
the large discrepancy between the modeled esti-
mate and the measurements at frequencies above
10 kHz. Alternatively, this discrepancy could be
caused by the low-pass filter on the measurement
devices. However, since the energy above 10 kHz
did not contribute significantly to the broad-band
energy at a distance of 2 km, its impact on harbour
porpoises could be ignored.
It is concluded, therefore, that the corrected
model provided a reasonable estimate of the SEL
within the first 2 km around the explosion. This
corresponded to the lower limit for effect distances
for smaller charges (< 25 kg). For larger explo-
sions (> 25 kg), for which the effect distances for
the lower limit of PTS onset (SEL > 179 dB re
1 μPa
2
s) approached the 10 to 20 km, the model
requires further validation.
Uncertainties in Dose-Response Relationships
for Hearing Loss—As there are no measurements
of onset thresholds of permanent hearing loss
in any marine mammal species, we considered
a range of SEL thresholds at which permanent
hearing loss could occur. A lower limit threshold
above which permanent hearing loss was consid-
ered increasingly likely to occur was obtained
by following the approach proposed by Southall
et al. (2007). That study assumed that PTS occurs
at a level of 15 dB above the TTS onset. Data from
Kastelein et al. (2012b) suggest there could be a
higher TTS/PTS exposure difference. Kastelein
et al. (2014a) demonstrated a positive correlation
between frequency and rate of TTS increase. This
would result in a wider offset between TTS and
PTS than assumed in the present study and, con-
sequently, a higher PTS threshold, especially for
the low-frequency range containing most of the
propagated energy from an explosion. Therefore
this SEL threshold of 179 dB re 1 μPa
2
s might be
conservative and may lead to an overestimation of
the number of animals that could be affected by
permanent hearing loss.
In this study, the effect of frequency weighting
of the SEL was neglected when estimating the risk
of TTS and PTS on the basis that our risk criterion
was based on observation of TTS onset due to a
single airgun exposure (Lucke et al., 2009), which
was more representative of the explosion sound
than other stimuli for which TTS measurements
have been obtained in harbour porpoises.
Weighting of the received SEL would need to
be compared to the corresponding weighted expo-
sure levels at which TTS was measured by Lucke
et al. (2009). Since the very low frequencies prop-
agate poorly in shallow waters, the frequency con-
tent was similar to that of an airgun, and, hence,
the frequency weighting was unlikely to affect the
results as the risk threshold for TTS will be low-
ered by the same order as the predicted weighted
SEL. Therefore, it was expected that the lack of
frequency weighting does not affect the estimated
impact of temporary and permanent hearing loss.
Values of peak overpressure levels at which
ear trauma was observed in eight cadaveric har-
bour porpoises exposed to underwater explosions
(Ketten, 2004) were used to put an upper limit to
the PTS SEL threshold. Ketten (2004) noted that
other characteristics of the shock wave are likely
to be important in determining the risk of injury
and need to be considered such as near field vs
far field loading effects; exponential vs sinusoidal
bursts; and synergistic effects of rate of pressure
increase, peak sound pressure, waveform, dura-
tion, and rise time, coupled with body mass (spe-
cies with smaller average body mass have greater
liability for shock-induced trauma, particularly in
air-filled tissues and from bubble formation and
oscillation, lesser body shadow protection, etc.).
It is not obvious how to translate the experimental
conditions employed by Ketten to large explosions
at sea and with greater distances from the source-
except in terms of received peak sound pressures.
That is equivocal as well, however, because of
the complex physics of propagating shock waves.
Further, it is important to note that tissue and
whole system responses are different in live vs
postmortem specimens. Even in the freshest mate-
rial, the mechanical properties for some or all tis-
sues may be altered due to death. In some cases,
postmortem material may be more responsive
Impact of Explosions on the North Sea Harbour Porpoise 519
than living tissue; in others, less. Thus, cadaveric
results are not definitive for live cases. They do
provide insights into purely mechanical responses
that may occur in marine mammals due to shock
waves, however. For this reason, the Ketten study
focused on the tissue changes that are essentially
mechanical and should be robust even in cadavers
such as ossicular fracture; middle, fenestral, and
inner ear membrane responses; and the minimal
end and massive, distinct trauma in other tissues
such as the liver and brain.
Conclusion
In this study, the impact of explosive clearance
activities of historical UXO on harbour porpoises
in the southern North Sea was assessed. This is the
first study to address on a broad scale the potential
impacts of underwater explosions on any marine
mammal species. This assessment was based on
actual explosion events carried out by the RNLN
in 2010 and 2011, using information on explosive
type, location, and timing of the detonations as
made available by the Netherlands Ministry of
Defense (NLMoD). For these explosion events,
impact areas were modelled and compared to the
aerial survey-based estimates of harbour porpoise
distribution during the same period. While the
data used in the analysis were site specific, the
methodology used in this study is broadly appli-
cable to other areas in which underwater explo-
sions occur.
Within the period of 2010 and 2011, a total
of 210 explosions occurred, with charge masses
ranging from several to one thousand kilograms
TNT eq. The model estimated that the 88 explo-
sions that occured in the DCS between March
2010 and March 2011 (the period of overlap
between aerial surveys and reported explosion
events) were very likely to have caused 1,280,
and possibly up to 5,450, permanent hearing loss
events (i.e., instances of a harbour porpoise pre-
dicted to have received sufficient sound exposure
to cause permanent hearing loss). Distances to
which there was a risk of permanent hearing loss
were on the order of one to several kilometres, and
possibly further for larger explosions.
The predicted impacts in this study contain
uncertainties, which were attributed to the follow-
ing factors:
Lack of data on sensitivity of harbour porpoises
to explosion shock waves and broadband im-
pulsive sound
Difficulty in predicting the generation and
propagation of underwater shock waves in
shallow water at large distance
Lack of knowledge on movement patterns and
habitat use of harbour porpoises
This study confirms earlier concerns (Nützel,
2008; Ainslie et al., 2009; Camphuysen &
Siemensma, 2011: Koschinski, 2012) that explo-
sions in the North Sea pose a risk to harbour por-
poises and has prompted the NLMoD to investi-
gate and implement mitigation measures to reduce
the adverse effects of underwater explosions.
Even the most optimistic estimates presented in
this study suggested that there was a significant
risk that hundreds of individual harbour por-
poises were negatively affected by the explosions,
although the population-level consequences could
not be judged. Models are being developed to
assess population-level consequences of under-
water sound on harbour porpoises (Lusseau et al.,
2012; Harwood et al., 2013; Nabe-Nielsen et al.,
2014). The detonation activities by all nations
active in the North Sea, as well as information on
distribution of harbour porpoises, should be con-
sidered in order to judge the cumulative effect of
all explosions on the harbour porpoise population
in the North Sea. To enable such assessment, it
is recommended that reportings of detonations
contain the necessary information (e.g., location,
depth, time, and estimated charge mass) to enable
an impact assessment.
Acknowledgments
We acknowledge the valuable discussions with
Lieutenant Commander Jurriën van Kasteren and
Lieutenant Commander Menno van der Eerden on
the practices and procedures of underwater explo-
sive clearance activities by the RNLN and Defence
Explosive Ordnance Disposal Service (DEODS).
We further thank the Netherlands Ministry of
Defence and the KNMI for making available the
reported explosion events used in this study, and
Lieutenant Commander René Dekeling and Christ
de Jong for a valuable discussion on the risk
assessment. This work greatly benefited from dis-
cussions with Rolf van Wees, Hans van Aanhold,
and Mark Tyler Street on the physics and the
measurement and modelling of underwater explo-
sions in shallow water environments. Further,
we thank Reinoud van de Kasteele and Stefania
Giodini for helping in the analysis of the 2010
explosion measurements. Finally, we thank the
two anonymous reviewers for their useful com-
ments and suggestions, which helped to improve
520 von Benda-Beckmann et al.
the quality of the manuscript. Özkan Sertlek and
Geert Aarts have been funded by NWO-ZKO
grant “Effects of Underwater Noise on Fish and
Marine Mammals in the North Sea.” D. R. Ketten
was supported by the Living Marine Resources
and ONR programmes of the U.S. Navy to study
underwater blast trauma and noise impacts, and
by the Hanse-Wissenschaftskolleg.
Literature Cited
Aarts, G., von Benda-Beckmann, A. M., Lucke, K.,
Sertlek, H. O., van Bemmelen, R., Geelhoed, S. C. V.,
. . . Kirkwood, R. (In prep.). Movement behaviour influ-
ences population level exposure to underwater sound:
Impact of explosions on harbour porpoises.
Ainslie, M. A. (2010). Principles of sonar performance
modeling. Chichester, UK: Springer-Praxis. http://dx.
doi.org/10.1007/978-3-540-87662-5
Ainslie, M. A., de Jong, C. A. F., Dol, H. S., Blacquière,
G., & Marasini, C. (2009). Assessment of natural and
anthropogenic sound sources and acoustic propagation
in the North Sea (TNO Report TNO-DV 2009 C085).
The Hague: TNO.
Akamatsu, T., Matsuda, A., Suzuki, S., Wang, D., Wang,
K., Suzuki, M., . . . Oota, K. (2005). New stereo acous-
tic data logger for free-ranging dolphins and porpoises.
Marine Technology Society Journal, 38, 6-12. http://dx.
doi.org/10.4031/002533205787443980
Camphuysen, C. J. (2011). Recent trends and spatial pat-
terns in near-shore sightings of harbour porpoises
(Phocoena phocoena) in the Netherlands (Southern
Bight, North Sea), 1990-2010. Lutra, 54, 39-47.
Camphuysen, C. J., & Siemensma, M. L. (2011).
Conservation plan for the harbour porpoise Phocoena
phocoena in The Netherlands: Towards a favourable
conservation status (NIOZ Report 2011-07). Texel:
Royal Netherlands Institute for Sea Research.
Chapman, N. R. (1985). Measurement of the waveform
parameters of shallow explosive charges. The Journal of
the Acoustical Society of America, 78, 672-681. http://
dx.doi.org/10.1121/1.392436
Clausen, K. T., Wahlberg, M., Beedholm, K., DeRuiter, S.,
& Madsen, P. T. (2010). Click communication in har-
bour porpoises Phocoena phocoena. The International
Journal of Animal Sound and Its Recording, 20, 1-28.
http://dx.doi.org/10.1080/09524622.2011.9753630
Cole, R. H. (1965). Underwater explosions. [Originally
published in 1948 by Princeton University Press, Dover,
NY].
Culik, B., Koschinski, S., Tregenza, N., & Ellis, G. M.
(2001). Reactions of harbor porpoises Phocoena pho-
coena and herring Clupea harengus to acoustic alarms.
Marine Ecology Progress Series, 211, 255-260. http://
dx.doi.org/10.3354/meps211255
Danil, K., & St. Leger, J. A. (2011). Seabird and dol-
phin mortality associated with underwater detonation
exercises. Marine Technology Society Journal, 6, 89-95.
http://dx.doi.org/10.4031/MTSJ.45.6.5
Finneran, J. J., & Jenkins, A. K. (2012). Criteria and
thresholds for U.S. Navy acoustic and 51 explosive
effects analysis. San Diego: SPAWAR Systems Center
Pacific. 64 pp.
Finneran, J. J., & Schlundt, C. E. (2013). Effects of fatigu-
ing tone frequency on temporary threshold shift in bot-
tlenose dolphins (Tursiops truncatus). The Journal of
the Acoustical Society of America, 133(3), 1819-1826.
http://dx.doi.org/10.1121/1.3641449
Fletcher, H., & Munson, W. A. (1933). Loudness, its defini-
tion, measurement and calculation. The Journal of the
Acoustical Society of America, 5, 82-108. http://dx.doi.
org/10.1121/1.1915637
Geelhoed, S., Scheidat, M., Aarts, G., van Bemmelen, R.,
Janinhoff, N., Verdaat, H., & Witte, R. R. H. (2012).
Shortlist masterplan wind aerial surveys of harbour
porpoises on the Dutch Continental Shelf (IMARES
Report Number C103/11). Texel: Den Burg.
Gilles, A., Scheidat, M., & Siebert, U. (2009). Seasonal
distribution of harbour porpoises and possible interfer-
ence of offshore wind farms in the German North Sea.
Marine Ecology Progress Series, 383, 295-307. http://
dx.doi.org/10.3354/meps08020
Hansen, M., Wahlberg, M., & Madsen, P. T. (2008).
Low-frequency components in harbour porpoise
(Phocoena phocoena) clicks: Communication signal,
by-product, or artifacts? The Journal of the Acoustical
Society of America, 124, 4059-4068. http://dx.doi.org/
10.1121/1.2945154
Harwood, J., King, S., Schick, R., Donovan C., & Booth, C.
(2013). A protocol for implementing the interim popu-
lation consequences of disturbance (PCOD) approach:
Quantifying and assessing the effects of UK offshore
renewable energy developments on marine mammal
populations (Report SMRUL-TCE-2013-014; Scottish
Marine and Freshwater Science, 5[2]).
Hiby, L., & Lovell, P. (1998). Using aircraft in tandem
formation to estimate abundance of harbour por-
poise. Biometrics, 54, 1280-1289. http://dx.doi.org/10.
2307/2533658
Kastelein, R. A. (1997). Food consumption and growth of
marine mammals (Ph.D. thesis). Landbouwuniversiteit
Wageningen, The Netherlands. 647 pp.
Kastelein, R. A., Gransier, R., Hoek, L., & de Jong, C. A. F.
(2012a). The hearing threshold of a harbor porpoise
(Phocoena phocoena) for impulsive sounds (L). The
Journal of the Acoustical Society of America, 132(2),
607-610. http://dx.doi.org/10.1121/1.4733552
Kastelein, R. A., Gransier, R., Hoek, L., & Olthuis,
J. (2012b). Temporary threshold shifts and recov-
ery in a harbor porpoise (Phocoena phocoena) after
octave-band noise at 4 kHz. The Journal of the
Acoustical Society of America, 132, 3525. http://dx.doi.
org/10.1121/1.4757641
Kastelein, R. A., Gransier, R., Hoek, L., & Rambags, M.
(2013). Hearing frequency thresholds of a harbour
Impact of Explosions on the North Sea Harbour Porpoise 521
porpoise (Phocoena phocoena) temporarily affected by
a continuous 1.5 kHz tone. The Journal of the Acoustical
Society of America, 134, 2286-2292. http://dx.doi.org/
10.1121/1.4816405
Kastelein, R. A., Gransier, R., Marijt, M. A. T., & Hoek,
L. (2015a). Hearing frequencies of a harbor porpoise
(Phocoena phocoena) temporarily affected by played
back offshore pile-driving sounds. The Journal of the
Acoustical Society of America, 137, 556. http://dx.doi.
org/10.1121/1.4906261
Kastelein, R. A., Gransier, R., Schop, J., & Hoek, L.
(2015b). Effects of exposure to intermittent and continu-
ous 6-7 kHz sonar sweeps on harbor porpoise (Phocoena
phocoena) hearing. The Journal of the Acoustical
Society of America, 137(4), 1623-1633. http://dx.doi.
org/10.1121/1.4916590
Kastelein, R. A., Gransier, R., Schop, J., & Hoek, L.
(2014a). Frequency of greatest temporary hearing
threshold shift in harbor porpoises (Phocoena pho-
coena) depends on the noise level. The Journal of the
Acoustical Society of America, 136, 1410-1418. http://
dx.doi.org/10.1121/1.4892794
Kastelein, R. A., Bunskoek, P., Hagedoorn, M., Au,
W. W. L., & de Haan, D. (2002). Audiogram of a harbor
porpoise (Phocoena phocoena) measured with narrow-
band frequency-modulated signals. The Journal of the
Acoustical Society of America, 112, 334-344. http://
dx.doi.org/10.1121/1.1480835
Kastelein, R. A., Hoek, L., Gransier, R., Rambags, M., &
Claeys, N. (2014b). Effect of level, duration, and inter-
pulse interval of 1-2 kHz sonar signal exposures on
harbour porpoise hearing. The Journal of the Acoustical
Society of America, 136, 412-422. http://dx.doi.org/10.
1121/1.4883596
Ketten, D. R. (2004). Experimental measures of blast and
acoustic trauma in marine mammals (ONR Final Report
N000149711030).
Ketten, D. R. (2012). Marine mammal auditory system
noise impacts: Evidence and incidence. In A. N. Popper
& A. Hawkins (Eds.), The effects of noise on aquatic
life (pp. 207-212). New York: Springer. http://dx.doi.
org/10.1007/978-1-4419-7311-5_46
Ketten, D. R., Lien, J., & Todd, S. (1993). Blast injury in
humpback whale ears: Evidence and implications. The
Journal of the Acoustical Society of America, 94(3/2),
1849-1850. http://dx.doi.org/10.1121/1.407688
Koschinski, S. (2012). Underwater noise pollution from
munitions clearance and disposal, possible effects
on marine vertebrates, and its mitigation. Marine
Technology Society Journal, 45, 80-88. http://dx.doi.
org/10.4031/MTSJ.45.6.2
Kryter, K. D. (1994). The handbook of hearing and the
effects of noise. New York: Academic Press. 673 pp.
Kryter, K. D., Ward, W. D., Miller, J. D., & Eldredge,
D. H. (1966). Hazardous exposure to intermittent
and steady state noise. The Journal of the Acoustical
Society of America, 39, 451-464. http://dx.doi.org/10.
1121/1.1909912
Kujiwa, S. G., & Liberman, M. C. (2009). Adding insult to
injury: Cochlear nerve degeneration after “temporary”
noise-induced hearing loss. Journal of Neuroscience, 20,
14077-14085. http://dx.doi.org/10.1523/JNEUROSCI.
2845-09.2009
Lewis, J. A. (1996). Effects of underwater explosions on life
in the sea. Melbourne, Victoria: Department of Defence,
Defence Science and Technology Organisation. 38 pp.
Lucke, K., Siebert, U., Lepper, P. A., & Blanchet, M-A.
(2009). Temporary shift in masked hearing thresholds in
a harbor porpoise (Phocoena phocoena) after exposure
to seismic airgun stimuli. The Journal of the Acoustical
Society of America, 125(6), 4060-4070. http://dx.doi.
org/10.1121/1.3117443
Lusseau, D., Christiansen, F., Harwood, J., Mendes, S.,
Thompson, P. M., Smith, K., & Hastie, G. D. (2012).
Assessing the risks to marine mammal populations from
renewable energy devices – An interim approach (Final
Workshop Report). 29 pp.
Madsen, P. A. (2005). Marine mammals and noise: Problems
with root mean square sound pressure levels for tran-
sients. The Journal of the Acoustical Society of America,
117(6), 3952-3957. http://dx.doi.org/10.1121/1.1921508
hl, B., & Andersen, S. (1973). Echolocation: High-
frequency component in click of harbor porpoise
(Phocoena ph. L.). The Journal of the Acoustical
Society of America, 54, 1368-1372. http://dx.doi.org/10.
1121/1.1914435
Nabe-Nielsen, J., Sibly, R. M., Tougaard, J., Teilmann, J.,
& Sveegaard, S. (2014). Effects of noise and by-catch
on a Danish harbour porpoise population. Ecological
Modelling, 272, 242-251. http://dx.doi.org/10.1016/j.
ecolmodel.2013.09.025
National Research Council. (2003). Ocean noise and
marine mammals. Washington, DC: The National
Academies Press.
Nützel, B. (2008). Untersuchungen zum schutz von
schweinswalen vor schockwellen [Investigations for
the protection of harbour porpoises to blast waves]
(Forschungsanstalt der Bundeswehr für Wasserschall
und Geophysik [FWG], Technischer Bericht TB 2008-
7).
OSPAR Commission. (2010). Quality status report:
Assessment of the impact of dumped conventional and
chemical munitions. London: OSPAR Commission.
Pfeiffer, F. (2014). Changes in properties of explosives due
to prolonged seawater exposure. Marine Technology
Society Journal, 102-110. http://dx.doi.org/10.4031/MT
SJ.46.1.5
Richardson, W. J., Greene, C. R., Jr., Malme, C. I., &
Thomson, D. H. (1995). Marine mammals and noise.
San Diego: Academic Press. 576 pp.
Rue, H., Martino, S., & Chopin, N. (2009). Approximate
Bayesian inference for latent Gaussian models by using
integrated nested Laplace approximations. Journal
of the Royal Statistical Society: Series B (Statistical
Methodology), 71, 319-392. http://dx.doi.org/10.1111/
j.1467-9868.2008.00700.x
522 von Benda-Beckmann et al.
Scheidat, M., Gilles, A., & Siebert, U. (2005). Applying
the circle-back method to estimate g(0)—Experiences
and results from aerial surveys in German waters.
Proceedings of the “Workshop on Estimation of g(0)
in Line Transect Surveys of Cetaceans” held at the
European Cetacean Society’s 18th Annual Conference,
Kolmården Djur Park, Kolmården, Sweden.
Scheidat, M., Verdaat, H., & Aarts, G. ( 2012). Using aerial
surveys to estimate density and distribution of harbour
porpoises in Dutch waters. Journal of Sea Research, 69,
1-7. http://dx.doi.org/10.1016/j.seares.2011.12.004
Scheidat, M., Gilles, A., Kock, K. H., & Siebert, U. (2008).
Harbour porpoise Phocoena phocoena abundance in the
southwestern Baltic Sea. Endangered Species Research,
5, 215-223. http://dx.doi.org/10.3354/esr00161
Schuknecht, H. F. (1993). Pathology of the ear (2nd ed.).
Philadelphia: Lea-Felbiger. http://dx.doi.org/10.1002/
hed.2880160317
Sertlek, H. Ö., & Ainslie, M. A. (2014). A depth-dependent
formula for shallow water propagation. The Journal
of the Acoustical Society of America, 136(2), 573-582.
http://dx.doi.org/10.1121/1.4884762
Sertlek, H. Ö., Ainslie, M. A., & Heaney, K. D. (In prep.).
Range-dependent propagation equations and results
for Pekeris waveguide. The Journal of the Acoustical
Society of America.
Soloway, A. G., & Dahl, P. H. (2014). Peak sound pressure
and sound exposure level from underwater explosions
in shallow water. The Journal of the Acoustical Society
of America, 136(3), EL219-EL223. http://dx.doi.org/
10.1121/1.4892668
Southall, B. L., Bowles, A. E., Ellison, W. T., Finneran,
J. J., Gentry, R. L., Greene, C. R., Jr., . . . Tyack, P. L.
(2007). Marine mammal noise exposure criteria: Initial
scientific recommendations. Aquatic Mammals, 33(4),
411-521. http://dx.doi.org/10.1578/AM.33.4.2007.411
Suzuki, Y., & Takeshima, H. (2004). Equal loudness-level
contours for pure tones. The Journal of the Acoustical
Society of America, 116, 918-933. http://dx.doi.org/
10.1121/1.1763601
Tougaard, J., Wright, A. J., & Madsen, P. T. (2015).
Cetacean noise criteria revisited in the light of proposed
exposure limits for harbour porpoises. Marine Pollution
Bulletin, 90(1-2), 196-208. http://dx.doi.org/10.1016/j.
marpolbul.2014.10.051
Tougaard, J., Carstensen, J., Teilmann, J., Skov, H., &
Rasmussen, P. (2009). Pile-driving zone of respon-
siveness extends beyond 20 km for harbour por-
poises (Phocoena phocoena [L.]). The Journal of the
Acoustical Society of America, 126(1), 11-14. http://dx.
doi.org/10.1121/1.3132523
Tyack, P. L. (2008). Implications for marine mammals of
large-scale changes in the marine acoustic environment.
Journal of Mammalogy, 89(3), 549-558. http://dx.doi.
org/10.1644/07-MAMM-S-307R.1
Verboom, W. C., & Kastelein, R. A. (1997). Structure of
harbour porpoise (Phocoena phocoena) click train
signals. In A. J. Read, P. R. Wiepkema, & P. E. Nachtigall
(Eds.), The biology of the harbour porpoise (pp. 343-
362). Woerden, The Netherlands: De Spil Publishers.
Verboom, W. C., & Kastelein, R. A. (2005). Some examples
of marine mammal “discomfort thresholds” in relation
to man-made noise. Proceedings of UDT Europe 2005,
Amsterdam.
von Benda-Beckmann, A. M., Aarts, G., Lucke, K.,
Verboom, W. C., Kastelein, R. A., Sertlek, O., . . .
Ainslie, M. A. (2015). Assessment of impact of under-
water clearance of historical explosives by the Royal
Netherlands Navy on harbour porpoises in the North
Sea (TNO Technical Report TNO 2014 R10916).
Wensveen, P. J., Huijser, L. E. A., Hoek, L., & Kastelein,
R. A. (2014). Equal latency contours and auditory
weighting functions for the harbour porpoise (Phocoena
phocoena). The Journal of Experimental Biology, 217,
359-369. http://dx.doi.org/10.1242/jeb.091983
Westgate, A. J., Read, A. J., Berggren, P., Koopman, H. N.,
& Gaskin, D. E. (1995). Diving behaviour of harbour
porpoises, Phocoena phocoena. Canadian Journal of
Fisheries and Aquatic Science, 52, 1065-1073. http://
dx.doi.org/10.1139/f95-104
Weston, D. E. (1960). Underwater explosions as acoustic
sources. Proceedings of the Physical Society, 76, 233-
249. http://dx.doi.org/10.1088/0370-1328/76/2/307
Weston, D. E. (1962). Explosive sources in underwater
acoustics. New York: Plenum Press.
Impact of Explosions on the North Sea Harbour Porpoise 523
Appendix A
To delimit the effect range at which permanent hearing loss becomes “very likely,” levels at which ear
trauma occurred in fresh animal cadavers resulting from primary blast injury were considered. Peak over-
pressure was used in Ketten (2004) to predict the occurrence of primary blast injury. Here, instead, the
peak overpressure was empirically related to SEL for measured explosion sound in shallow water due to
the difficulties of modelling peak overpressure in shallow water. Then, an effective SEL that corresponded
to peak overpressures resulting in ear trauma was estimated.
To indicate at what distances a risk of blast injury occurred, the measured peak overpressure was
related to measured SEL using the data from the monitored explosions (Figure A1). Measured peak over-
pressure and broadband SEL showed a roughly log-linear relationship with a greater scatter for measure-
ments close to the surface (depths smaller than 6 m).
Observations by Ketten (2004) showed that ear trauma occurred at peak overpressures exceeding
172 kPa (25 psi), indicated by the grey dashed line in Figure A1, which corresponded to a range in SEL
of 195 to 203 dB re 1 μPa
2
s, which suggests that ear trauma is very likely to occur at SEL > 203 dB re
1 μPa
2
s. No ear trauma could be observed for peak overpressures of 69 kPa (10 psi) as indicated by the grey
solid line, suggesting that ear trauma due to primary blast injury is unlikely to occur at SEL < 190 dB re
1 μPa
2
s. The probability of ear trauma to occur between SEL = 190 and 203 dB re 1 μPa
2
s was considered
to be increasingly likely.
It is expected that the acoustics of a blast event at a received level of SEL > 190 dB re 1 μPa
2
s is very
likely (> 95% probability) to cause PTS. The delimitation between the terms “unlikely” (less than 5%
probability), “increasingly likely” (probability between 5 and 95%), and “very likely” (greater than 95%)
adopted in this study are not based on a statistical evaluation of the data; rather, they should be considered
as indicative of the consensus opinion of the authors of the likelihood at which these effects occur.
Figure A1. Empirical relation between unweighted broadband SEL and peak overpressure for shallow water explosions of
263-kg TNT eq. charge mass, both measured at different distances (see “Results” for more details). Superimposed are risk
thresholds for a lower limit for PTS onset (black line, based on the Lucke et al. [2009] TTS measurement and adding 15 dB
following Southall et al. [2007]) and observed pathological effects (grey lines, observed by Ketten [2004]).
... The relationship between the charge weight and the marine environment, through the wave propagation, is usually studied for the case of relatively small charges (generally, smaller than a few-kilograms TNT-equivalent weight) and/or relatively deep seabeds (e.g., [8]- [10]). To the best of our knowledge, very few works are concerned with charges of a few-hundred-kilograms TNT-equivalent weight [11], [12] and located in coastal waters with a depth less than 50 m [13]- [15]. However, the UXO of WWII usually satisfies these conditions, in particular along the French coasts of Brittany and the Mediterranean Sea. ...
... Also, note that the characteristics of the explosion-induced signal may be slightly different according to the UXO location compared to the sea bottom [25]. Despite this abundant literature, works that are focused on the characteristics of charges of hundred kilograms TNT-equivalent weights and located in shallow water environments, except those reported in [13]- [15] and [26], are still lacking. This can be explained by the fact that the recording systems such as hydrophones or OBS, deployed in the near-field of the explosion, can be damaged. ...
... The work presented in a two-companion article only focuses on risks of damage to buildings and infrastructures on the adjacent shore. It does not address the risks of harming or disturbing marine life (e.g., [15]). These risks will be also under consideration in our future works. ...
... At shorter ranges acoustic waveforms from explosive sources are highly non-linear as a function of range (outlined in Section 5.2). To obtain useful far-field linear Source Level estimations requires the use of measured data from the explosion recorded at ranges where the acoustic waves are assumed to propagate linearly [von Benda-Beckmann et al. 2015;Ainslie 2010;Jones & Clarke 2005;Jones et al. 2006]. The predicted Source Level is then calculated by accounting for the propagation loss (equivalent to propagating back to 1 m from the source) to generate a monopole Source Level for use in propagation modelling of the type outlined in Section 5.3. ...
... Another approach is to use the energy spectrum of the source and incoherent propagation models [Weston 1960;von Benda-Beckmann et al. 2015] to calculate SEL. It is not possible to find peak sound pressure directly through this approach. ...
... It is not possible to find peak sound pressure directly through this approach. However, it is possible to derive a scale factor to relate the peak pressure with SEL [von Benda-Beckmann et al. 2015]. ...
... Since air guns are an impulsive and low-frequency source, they are fairly representative of an explosive sound at large distances in shallow water, as very low frequencies propagate poorly in shallow waters [62]. As such, the more fully defined thresholds for fish for seismic airguns have been adopted by EDGAR (Table 4), rather than the less conservative explosives guidelines [13]. ...
... Since air guns are an impulsive and low frequency source, they are fairly representative of an explosive sound at large distances in shallow water [62]. As such, the more fully defined thresholds for fish for seismic airguns have been adopted by EDGAR, rather than the less conservative explosives guidelines [13]. ...
Article
Full-text available
A simple underwater noise model suitable for use with explosive severance of well conductors and piles during the decommissioning of oil and gas subsea structures is introduced and evaluated against data from five projects in the US. This study focuses on a novel model for the determination of sound exposure levels. The model has been developed to enable determination of impact areas for marine mammals and fish. Simulated received underwater sound exposure levels were significantly correlated with measurements for all scenarios. The maximum total error achieved between simulations and measurements was 2.6%, suggesting that predictions are accurate to within 3% of the average measurement. A low relative bias was observed in the simulations when compared to measured values, suggesting only a small systematic underestimate (≤0.5% of average measurement) for most severance operations and a small overestimate (0.14%) for open water blasts.
... Since air guns are an impulsive and low frequency source, they are fairly representative of an explosive sound at large distances in shallow water as very low frequencies propagate poorly in shallow waters [39]. As such, the more fully defined thresholds for fish for seismic airguns ( Table 1) have been adopted by EDGAR, rather than the less conservative explosives guidelines [8]. ...
... Since air guns are an impulsive and low frequency source, they are fairly representative of an explosive sound at large distances in shallow water [39]. As such, the more fully defined thresholds for fish for seismic airguns have been adopted by EDGAR, rather than the less conservative explosives guidelines [8]. ...
Preprint
A simple underwater noise model suitable for use with explosive severance of well conductors and piles during the decommissioning of oil and gas subsea structures is introduced and evaluated against data from five projects in the US. This study focuses on a novel model for the determination of sound exposure levels. The model has been developed to enable determination of impact areas for marine mammals and fish. Simulated received underwater sound exposure levels were significantly correlated with measurements for all scenarios. The maximum total error achieved between simulations and measurements was 2.6%, suggesting that predictions are accurate to within 3% of the average measurement. A low relative bias was observed in the simulations when compared to measured values, suggesting only a small systematic underestimate (≤ 0.5% of average measurement) for most severance operations and a small overestimate (0.14%) for open water blasts.
... Numerous anthropogenic activities generate underwater noise, including industry, e.g. during the construction and operation of offshore wind farms (Dähne et al., 2013;Brandt et al., 2018;Graham et al., 2019), shipping (e.g. reviewed in Erbe et al., 2019) or the clearance of unexploded ordnance (von Benda-Beckmann et al., 2015;Siebert et al., 2022). Intense sound is also emitted during seismic explorations of the seafloor (e.g. ...
Article
Full-text available
Large rivers like the Elbe or the Weser are periodically entered by harbor porpoises of the North Sea. They may even move 97 km upstream to the port of Hamburg, where their presence is highest in spring. This migration is believed to be related to important anadromous prey species travelling upstream for spawning. An acoustic flowmeter in the port of Hamburg emits signals in the hearing range of harbor porpoises. The pulses have a duration of 0.2 ms, a peak frequency of 28 kHz, a source level of 210 dB re 1 µPa and an inter-pulse interval of 4.2 s. The signals are continuously emitted from both sides of the river at a location where the river is 400 m wide. We evaluated the potential of these signals to induce temporary threshold shift (TTS) in harbor porpoise hearing. Hearing tests with a harbor porpoise in human care were conducted to determine TTS onset. We modelled the acoustic field based on underwater noise measurements. The acoustic flowmeters emit pulses in a highly directional beam with a source level high enough for inducing TTS in harbor porpoises by a single exposure up to a distance of approximately 72 m. The received cumulative sound exposure levels for harbor porpoises travelling along the flowmeters are mainly dependent on the timing and distance to the sound source. Accordingly, a close approach to the flowmeter at the time of transmission should be prevented. This could be the case, if vessels force harbor porpoises to displace closer to the flowmeters. We therefore suggest to decrease acoustic flowmeter source levels. This case study emphasizes the need for a mandatory authorization process prior to the use of underwater sound for any purpose with potential effects on aquatic life. Such an authorization process should carefully consider potential effects for target and non-target species.
... It has been estimated that a recent detonation of 88 naval mines on the Netherlands' continental shelf in a single year "very likely caused 1280, and possibly up to 5450, permanent hearing loss events (i.e., instances of a harbour porpoise predicted to have received sufficient sound exposure to cause permanent hearing loss)." [45] Harbour porpoise is particularly vulnerable to underwater noise. Based on their detailed measurements and model results, Salomons et al. concluded that "harbour porpoises are at risk of permanent hearing loss at distances of several kilometers from large explosives." ...
Article
Full-text available
The innovative development of the legal regime of straits has prevented the erection of ‘sovereignty-barriers’ to the movement of humans in and above straits. However, it overlooks to a great extent the significance of straits for marine organisms and birds. This article examines if it is necessary to supplement the legal regime of straits with rules that would allow circumnavigating ‘sovereignty-barriers’ also for wildlife movement considering the obligation to protect and preserve the marine environment. The main users of straits are not humans, but rather marine species. They rely on straits for moving from one ecosystem to another. That perspective to straits shifts the emphasis away from anthropocentric connectivity. It raises a question about the need to update the current navigation-oriented legal regime of straits with new wildlife-centric rules. This implies an additional scrutiny on human activities that have a significant negative effect on marine organisms and the fragile marine environment of straits. A wildlife-centred approach enables to reconsider the appropriateness of some human uses of the seas that are environmentally hazardous, but still relatively commonplace in straits. It is possible to facilitate the unimpeded movement of marine species through straits by the prohibition of some detrimental maritime practices that have a reasonable alternative. Such practices include, e.g., the detonation of naval mines in clearance operations, the construction of such causeways that are impassable for marine species, and the use of overhead power lines in straits. In addition, limits could be set to the use of sonars and to the speed of ships in straits.
... The underwater detonation of munition has a catastrophic impact on the organisms living in the surrounding area and their habitat. The shock wave destroys any organisms and can still harm the sensory organs of porpoises many kilometers away [19,20]. Incomplete detonations leave substantial quantities of the explosive material in the environment, which is absorbed by organisms (e.g., mussels) and enters the food chain [17]. ...
Article
Full-text available
Military munitions from World War I and II dumped at the seafloor are a threat to the marine environment and its users. Decades of saltwater exposure make the explosives fragile and difficult to dispose of. If required, the munition is blast-in-place. In August 2019, 42 ground mines were detonated in a controlled manner underwater during a NATO maneuver in the German Natura2000 Special Area of Conservation Fehmarn Belt, the Baltic Sea. In June 2020, four detonation craters were investigated with a multibeam echosounder for the first time. This dataset is represented here as maps of bathymetry, slope angle, and height difference to the surrounding. The circular craters were still clearly visible a year after the detonation. The diameter and depth of the structures were between 7.5–12.6 m and 0.7–2.2 m, respectively. In total, about 321 m2 of the seafloor was destroyed along the track line.
... The current levels in the Baltic biota indicate that PCB contamination remains a serious impediment to the health and reproductive status of the Baltic Proper harbour porpoise population, but lack of samples prevents direct studies. The lack of samples is due to a combination of the small population size and a low willingness to report and land bycaught harbour porpoises.Impulsive underwater noise sources occurring in the BalticProper can cause behavioral disturbance, hearing loss, and other physical injury to harbour porpoises(Kastelein et al., , 2017Ketten, 2004;Lucke et al., 2009;Pirotta et al., 2014;Sarnocińska et al., 2020;Thompson et al., 2013;von Benda-Beckmann et al., 2015). Data on loud sources of impulsive noise in the Baltic Sea are collated nationally and reported to an ICES registry in support of HELCOM(HELCOM, 2021; ICES, 2020b). ...
... Proper can cause behavioral disturbance, hearing loss, and other physical injury to harbour porpoises (Kastelein et al., , 2017Ketten, 2004;Lucke et al., 2009;Pirotta et al., 2014;Sarnocińska et al., 2020;Thompson et al., 2013;von Benda-Beckmann et al., 2015). (Forney et al., 2014;Moore et al., 2009). ...
Article
Full-text available
Abstract Knowing the abundance of a population is a crucial component to assess its conservation status and develop effective conservation plans. For most cetaceans, abundance estimation is difficult given their cryptic and mobile nature, especially when the population is small and has a transnational distribution. In the Baltic Sea, the number of harbour porpoises (Phocoena phocoena) has collapsed since the mid‐20th century and the Baltic Proper harbour porpoise is listed as Critically Endangered by the IUCN and HELCOM; however, its abundance remains unknown. Here, one of the largest ever passive acoustic monitoring studies was carried out by eight Baltic Sea nations to estimate the abundance of the Baltic Proper harbour porpoise for the first time. By logging porpoise echolocation signals at 298 stations during May 2011–April 2013, calibrating the loggers’ spatial detection performance at sea, and measuring the click rate of tagged individuals, we estimated an abundance of 71–1105 individuals (95% CI, point estimate 491) during May–October within the population's proposed management border. The small abundance estimate strongly supports that the Baltic Proper harbour porpoise is facing an extremely high risk of extinction, and highlights the need for immediate and efficient conservation actions through international cooperation. It also provides a starting point in monitoring the trend of the population abundance to evaluate the effectiveness of management measures and determine its interactions with the larger neighboring Belt Sea population. Further, we offer evidence that design‐based passive acoustic monitoring can generate reliable estimates of the abundance of rare and cryptic animal populations across large spatial scales.
Article
Full-text available
Harbour porpoises are under pressure from increasing human activities. This includes the detonation of ammunition that was dumped in large amounts into the sea during and after World War II. In this context, forty-two British ground mines from World War II were cleared by means of blasting in the 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.
Conference Paper
Full-text available
Worldwide a concern exists about the influence of man-made noise on marine life and particularly on marine mammals and fish. One of the acoustic polluters of the world's oceans is high-power active sonar, but also pile driving and seismic activities at sea are of concern with respect to animal welfare. At TNO, acoustic criteria are being developed to protect marine animals from severe disturbance (or worse) due to man-made noise. One of the 'stages' in 'dose-response relationships' is the 'discomfort threshold', the received noise level at which a marine animal turns when approaching a noise source. In The Netherlands discomfort thresholds for a number of sound types have been determined for harbour porpoises, harbour seals and some North Sea fish species. This paper shows how those measurements were carried out and compares some results with proposed TNO dose-response relationships for marine mammals.
Chapter
Full-text available
Broadband recordings of signals produced by 3 male harbour por¬poises in 28 low-repetition frequency click trains (repetition frequencies between 14 and 35 Hz) were subjected to Fast Fourier Transform analysis (freque-ncy range 1 kHz - 160 kHz). Individual signals consisted of low-fre¬quency (80 Hz - 10 kHz), mid-frequency (10 kHz - 100 kHz) and high-frequen¬cy (100 kHz - 160 kHz) components, one or more of which were represented in every signal. Low-fre¬quency com¬ponents were highly variable in fre¬quen¬cy, frequency range, duration and waveform, and could be divided into 3 types: 1) LF1, dominant damped sine waves (2 kHz) of short duration (2.5 - 4.5 ms), 2) LF2, low-amplitude, noise-like com¬ponents of longer dura¬tion (25 ms), and 3) LF3, con¬tinuous sine waves (fr¬equen¬cy 40 - 600 Hz). Mid-fre¬quency (MF1) compo¬nents consis¬ted of har¬monic pat¬terns based on a fun¬damen¬tal fre¬quency of 7 - 20 kHz. High-frequency compo¬nents, which may make up part of this har-monic pattern, were in¬variable in waveform, duration (150 - 300 ms) and frequen¬cy. Energy peaks were at around 120 kHz (HF1, 115 - 126 kHz) and at 140 kHz (HF2, 135 - 146 kHz); sometimes more than 2 peaks were present (HF3). The signals are shown to be more variable than previous¬ly believed, and the conclusions of Verboom and Kastelein (1995) regarding MF com¬ponents are revised.
Article
Full-text available
Safety criteria for mid-frequency naval sonar sounds are needed to protect harbor porpoise hearing. A porpoise was exposed to sequences of one-second 6–7 kHz sonar down-sweeps, with 10–200 sweeps in a sequence, at an average received sound pressure level (SPL av.re.) of 166 dB re 1 μPa, with duty cycles of 10% (intermittent sounds) and 100% (continuous). Behavioral hearing thresholds at 9.2 kHz were determined before and after exposure to the fatiguing noise, to quantify temporary hearing threshold shifts (TTS 1–4 min) and recovery. Significant TTS 1–4 min occurred after 10–25 sweeps when the duty cycle was 10% (cumulative sound exposure level, SELcum: ∼178 dB re 1 μPa2s). For the same SELcum, the TTS 1–4 min was greater for exposures with 100% duty cycle. The difference in TTS between the two duty cycle exposures increased as the number of sweeps in the exposure sequences increased. Therefore, to predict TTS and permanent threshold shift, not only SELcum needs to be known, but also the duty cycle or equivalent sound pressure level (Leq). It appears that the injury criterion for non-pulses proposed by Southall, Bowles, Ellison, Finneran, Gentry, Greene, Kastak, Ketten, Miller, Nachtigall, Richardson, Thomas, and Tyack [(2007). Aquat. Mamm. 33, 411–521] for cetaceans echolocating at high frequency (SEL 215 dB re 1 μPa2s) is too high for the harbor porpoise.
Article
Dr Ainslie’s book provides a long-awaited complete and modern treatment of sonar performance modelling (SPM). In this context, the word "sonar" is used in a broad sense, to mean any deliberate use of underwater sound, including by marine mammals. The acronym "SONAR" stands for "sound navigation and ranging", but this book demonstrates how sonar systems and methodology are used for a variety of sensing, communications and deterrence systems, and by a number of industries and end-users (military, offshore, fisheries, surveyors and oceanography). The first three chapters provide background information and introduce the sonar equations. The author then lays the main foundations with separate chapters on acoustical oceanography, underwater acoustics, signal processing and statistical detection theory. These disparate disciplines are integrated expertly and authoritatively into a coherent whole, with as much detail as necessary added for more advanced applications of SPM. The book is illustrated with numerous worked examples, at both introductory and advanced levels, created using a variety of modern SPM tools.
Article
Scitation is the online home of leading journals and conference proceedings from AIP Publishing and AIP Member Societies
Article
Mean dive depths and duration ranged from 14±16 to 41±32 m, and from 44±37 to 103±67 s, respectively. The maximum recorded dive depth and duration was 226 m and 321 s. This performance may not represent the maximum capacity of harbour porpoises but rather the maximum depth of the study area. Individual dives had similar rates of descent and ascent. Two porpoises with monitoring periods >2days demonstrated a diel pattern in their diving, making fewer, but deeper dives at nights. In the Bay of Fundy, porpoises made between 22 and 70% of dives to depths (range 20-130 m) where the majority of entanglements in ground fish gill nets were reported. -from Authors
Article
The present paper gives a survey of problems encountered with changes in properties of energetic materials in sea-dumped military munitions.