See discussions, stats, and author profiles for this publication at: [Link]

net/publication/317318851

Modeling whale-vessel encounters: the role of speed in mitigating collisions

with humpback whales (Megaptera novaeangliae)

Article in Journal of Cetacean Research and Management · May 2017

CITATIONS READS

9 273

3 authors:

Jens Currie Stephanie Stack

Pacific Whale Foundation Pacific Whale Foundation
32 PUBLICATIONS 157 CITATIONS 28 PUBLICATIONS 99 CITATIONS

SEE PROFILE SEE PROFILE

Greg Kaufman
Pacific Whale Foundation, Maui, HI USA
42 PUBLICATIONS 754 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Preventing Ocean Pollution View project

Assessing the Impacts of Human Activities on Whales and Dolphins View project

All content following this page was uploaded by Stephanie Stack on 02 February 2018.

The user has requested enhancement of the downloaded file.

 J. CETACEAN RES. MANAGE. 17: 57–64, 2017 57

Modelling whale-vessel encounters: the role of speed in mitigating

collisions with humpback whales (Megaptera novaeangliae)
J.J. CURRIE, S.H. STACK AND G.D. KAUFMAN
Pacific Whale Foundation, 300 Ma’alaea Road, Suite 211, Wailuku, Maui, HI 96793, USA
Contact e-mail: research@[Link]

ABSTRACT
Increasing whale populations and vessel traffic worldwide has led to an increase in reported whale-vessel collisions. This paper reports on factors
that affect the rate of whale-vessel collisions in the four-island region of Maui, Hawai’i. More specifically, it aims at quantifying the probability of
a whale-vessel collision with varying vessel speeds using encounter distances as a proxy. A change point model was used to identify a speed threshold
of 12.5kts (6.4m/s), which showed a significant change in the relationship between speed and mean sighting distance. A 3.4-fold decrease in close
encounters with humpback whales was observed when vessels travelled at speeds of 12.5kts (6.4m/s) or less. Furthermore, results indicate that lone
adult whales and calves are the most likely to be involved in a collision. A speed limit of 12.5kts (6.4m/s) is warranted in areas and/or during seasons
where a high density of whales occurs. This limit aligns with a reduction in lethal vessel strikes with speed from previous studies which found a
significant increase in the likelihood of mortality when vessel speeds exceeds 12kts.
KEYWORDS: MODELLING; SHIP STRIKES; HUMPBACK WHALE; PACIFIC OCEAN; SURVEY–VESSEL; CONSERVATION

INTRODUCTION frequency of vessel strikes are likely to be underestimates,

Vessel collisions with cetacean species are a growing owing to under reporting, whether intentional or
concern worldwide (IWC, 2011; Douglas et al., 2008; Laist unintentional (Van Waerebeek et al., 2007; Neilson et al.,
et al., 2001; Van Waerebeek et al., 2007). Although a wide 2012). In addition, population level effects of collision
range of cetacean species are struck by vessels, collisions are mortality are also not well understood for most whale species
a key mortality factor for larger whale species, including (van der Hoop et al., 2013).
those found on the endangered species list (Laist et al., 2001; Hawai’i is an area where humpback whale habitat and
Redfern et al., 2013). Large whales, including humpback high human use overlap. Over 8.1 million people visited
whales (Megaptera novaeangliae) are more susceptible to Hawai’i in 2014 (DBEDT, 2015), with vessel-based
collisions in areas where their habitat overlaps with heavy activities being a major source of revenue for the tourism
vessel traffic. This risk is increased when whales are resting sector (Lammers et al., 2013) owing to the thousands of
or moving slowly at the surface (Constantine et al., 2015; humpback whales that migrate to Hawai’i each winter to
Laist et al., 2001; Vanderlaan and Taggart, 2007). breed and calve. More than half, 53.6%, of the North Pacific
The increased rate of whale-vessel collisions over the past humpback whale population migrates to Hawai’i each year
few decades constitutes an important conservation issue (Calambokidis et al., 2008) with the population growing by
(IWC, 2011; Douglas et al., 2008; Laist et al., 2001; Silber 5.5–7.0% annually (Calambokidis and Barlow, 2004;
et al., 2010), as they can often be lethal to the animal. Mobley et al., 2001). As humpback whale numbers continue
Collisions which seriously injure or kill large whales are an to grow, so too does the concern about potential increases in
important factor threatening the viability of certain whale-vessel collisions.
populations or sub-populations (Knowlton and Kraus, 2001; The majority of whale-vessel collisions reported in
Panigada et al., 2009). Hawai’ian waters by NOAA occur between the islands of
Increased vessel traffic globally, as well as an increase in Maui, Molokai, Kaho‘olawe, and Lana‘i, collectively
size and speed of vessels, has contributed to the rise in referred to as the four-island region of Maui (Laist et al.,
whale-vessel collisions (Dolman et al., 2009; Jensen and 2001; Lammers et al., 2013). From 2013 to 2015, 17 vessel
Silber, 2004; Vanderlaan et al., 2009). Vessels of all types collisions were reported to NOAA, of which 82% (n = 14)
and sizes are known to be involved in collisions with were recorded in the four-island region of Maui (Ed
cetaceans, but larger and faster vessels account for higher Lyman, NOAA/HIHWNMS, pers. comm., 2015). The high
instances of lethal collisions (Laist et al., 2001; Panigada et percentage of whale vessel strikes within the Hawai’ian
al., 2006; Silber et al., 2010; Vanderlaan and Taggart, 2007). Islands Humpback Whale National Marine Sanctuary
At a speed of 12kts (6.2m/s), 45–60% of collisions between (HIHWNMS) is not surprising, given that the greatest
a vessel with mass significantly exceeding that of the whale density of humpback whales occurs within this region
are lethal; at speeds ≥ 19kts (9.8m/s) 100% of collisions are (Mobley et al., 2001); which is in conjunction with over half
lethal (Vanderlaan and Taggart, 2007). of the Hawai’ian whalewatching operations (O’Connor et
While various models provide insight into the al., 2009) (Appendix Fig. 1) and a multitude of other
survivability of vessel strikes among large whales, our commercial and non-commercial vessels (Appendix Fig. 2).
understanding of the true frequency of strikes and the factors Increasing reports of whale-vessel collisions in Hawai’i
that lead to them is limited. Published figures for the are likely to be caused in part by an increasing number of
58 CURRIE et al.: MODELLING WHALE – VESSEL ENCOUNTERS

humpback whales and increased monitoring efforts over METHODS

the last 20 years (Lammers et al., 2013). However, the Study area
incidence of reported collisions is increasing more than The study region covered an area of 798.0km2 located within
would be expected for these reasons alone. Other potential the four-island region of Maui, Hawai’i, and was chosen to
factors include an increase in registered vessels between cover a large section of the HIHWNMS (Fig. 1). The area
7.9 and 19.8m, the size class which is responsible for experiences high levels of vessel traffic during the
more than two thirds of strikes in Hawai’i (Lammers et al., whalewatching season from December to April each year
2013). (DBEDT, 2015).
Although mostly limited to vessels that have the ability to
avoid collisions, to date very few studies have attempted to Data collection
quantify the risk of collisions by taking into account not only Surveys were conducted from an 8m Power Cat research
distance to whale (Gende et al., 2008) but also vessel speed vessel equipped with two 150 horsepower outboard engines.
at the time of initial sighting. A better understanding of Data were collected using systematic line transect
specific factors that influence the incidence of collisions, methodologies (Buckland et al., 2004) during the winter
particularly in the seconds prior to contact, is crucial to months from 2013 to 2016. Observations were undertaken
reduce this threat. This paper looks at data collected from a by two experienced observers and the boat operator using a
dedicated research platform that recorded distances to first continuous scanning methodology by naked-eye or reticle
sightings of humpback whales travelling at different speeds. binoculars (Bushnell 7x50), while a fourth person acted as a
These data were then used to assess the frequency and data recorder. Only whales sighted within 300m or less,
proximity of encounters between small vessels (< 10m) and forward and abeam of the vessel were recorded to represent
humpback whales in relation to vessel speed and to identify whales at risk of collision. Within this distance, encounters
a speed guideline for the Hawai’i regions or similar areas. were further classified into surprise encounters (SE) and near
To our knowledge, this is the first systematic study aimed at misses (NM) defined as sighting within 80–300m and 0–80m
better understanding how speed influences the encounter respectively, as outlined in Stack et al. (2013). In the context
distance between humpback whales and small vessels, of this paper and throughout the remaining text, SE and NM
utilising an in situ approach to developing a whale-vessel will be collectively referred to as ‘encounters’ and refers to
collision model for management purposes. whales sighted within 0–300m forward and abeam of the

Fig. 1. Transect lines depicting survey area in the four-island region of Maui, Hawai‘i, between 2 February
2013 and 31 March 2016 including the Hawai’ian Islands Humpback Whale National Marine Sanctuary
(HIHWNMS) boundary.
 J. CETACEAN RES. MANAGE. 17: 57–64, 2017 59

vessel. The division of encounters into SE and NM allowed determine the density of encounters, the study area was
for subsequent analysis to determine if specific age classes divided into 184 grid cells measuring 1.5x1.5km, each with
were more susceptible to NM and/or SE. In addition, the an area of 2.25km2. Each grid cell was summarised by the
following data were also recorded: time and location count of encounters occurring in that cell and the total on
(latitude and longitude) of sighting, vessel speed, age-class effort distance travelled in that cell. Density of encounters
of whale, number of whales in the pod, angle to pod was calculated by dividing the total number of encounters
(measured in magnetic degrees), and direction of travel by by the on effort distance per grid cell. Only grid cells that
the whale. Additional environmental variables including had a total on effort distance of ≥ 5km were included in final
Beaufort Sea state (BSS) as a measure of wind speed and density estimates. Maps and grid were created using ArcGIS
Douglas Sea state (DSS) as a measure of wave height, were 10.1 (ESRI, 2011).
recorded at the start of each transect line, and updated as they
changed throughout the survey. Probability of encounters with varying vessel speed and
To quantify rate encounters with varying vessel speed, a month
total of seven different speeds were randomly selected for the A General Linear Model (GLM) with a binomial error
start of each transect, and speed was increased at 5kt (1.3m/s) distribution and logit link function was used to model the
increments every 15 minutes until the transect was completed. relationship between encounters and vessel speed:
Depending on the length of the transect line, between two and PSE = e β0 + βsp + ε
three speed intervals were completed for each transect.
Speeds used were 5.0, 7.5, 10, 12.5, 15, 17.5 and 20.0kts where PSE is the probability of encounter, β0 is the intercept,
(2.6–10.3m/s) and this range was chosen to best represent the βsp is the speed ranging from 5.0 to 20.0kts (2.6 to 10.3 m/s),
most frequently travelled vessel speeds in the study area. and is the binomial error.
To account for the variation in number of humpback
Analysis whales from December to April resulting from a progressive
Analyses performed were: (1) assess the composition of SE influx in numbers leading to peak season (Baker and
and NM with varying age class and group composition; (2) Herman, 1981), analysis was divided into five months to
change-point modelling to determine a threshold speed at represent the primary mating/birthing season in Hawai’i
which a change in mean distance at initial sighting of the waters: December, January, February, March and April.
whale occurred for all encounters; (3) quantification and
distribution of encounters above and below identified Model fit
threshold speed; (4) probabilities of encounters with varying All computations were completed using the ‘stats’ package
vessel speed. in R (Wood, 2011). Final model selection was based on
minimizing the AIC values (Akaike, 1973). To ensure proper
Change-point modelling model fit and adherence to assumptions, model residuals
To determine if there was a threshold speed which caused a were graphed and visually checked for violations.
change in the mean encounter distance a change-point
analysis was completed (Gende et al., 2011) using the RESULTS
‘changepoint’ package in R (Killick and Eckley, 2014). Survey effort
Encounter data from 2013–2016 were binned into 2.5 knot Between 2 February 2013 and 31 March 2016, 143 survey
speed increments, which were summarised by the mean days allowed for sampling of 608 transect lines in the four-
sighting distance derived from a minimum of 30 observations. island region of Maui. Each transect line was surveyed a
Encounter data for each set of changes were then checked for minimum of 23 and maximum of 29 times throughout the
normality and independence to ensure adherence to change- study period. This corresponded to a total of 4,477.6 nautical
point distribution assumptions. As the goal of the analysis was miles ([Link]) on effort and 5,009.4 [Link] off effort survey
to identify a speed threshold and assess the frequency of distances.
encounters above and below this threshold, the At-Most-One-
Change method was considered most appropriate for the Composition of SE and NM
change-point model fit with a normal distribution: A total of 529 SE and 25 NM were recorded during the study
period. Calves were present in 23.1% (n = 122) of SE and
Eni ~ (βj , σ 2)
48.0% (n = 12) of NM. Of all SE and NM involving calves,
Where Eni are the encounters (i) including a speed (kts) and 54.5% (n = 73) were mother-calf pairs, 26.1% (n = 35) were
distance (m) with mean (βj ) and variance (σ 2), and j is the mother-calf-escort pods, and 19.4% (n = 26) were lone
mean distance of sighting above and below the identified calves (i.e. mother did not surface). Lone adults accounted
change point. for 48.3% (n = 255) of SE and 32.0% (n = 8) of NM, while
pods consisting of ≥ 2 adults, accounted for 22.3% (n = 118)
Distribution of encounters above and below the threshold of SE and 44.0% (n = 11) of NM.
speed
To determine the location and frequency of encounters, all Change-point modelling
on effort sighting and GPS track data collected from The change point model identified a change in the
February 2013 to April 2016 were combined. Data were then relationship between speed and mean sighting distance at
subdivided into two groups: encounters above and below the 12.5kts (6.4m/s) (Fig. 2). The mean sighting distance before
threshold speed identified using the change-point model. To and after the change point was 211.2m and 189.4m,
60 CURRIE et al.: MODELLING WHALE – VESSEL ENCOUNTERS

Fig. 2. Mean sighting distance of humpback whales with increasing speed

(points) and the identified change point (solid line) recorded within the
four-island region of Maui, Hawai’i between 2 February 2013 and 31
March 2016.

respectively. In the field, encounters were reduced 3.4 fold Fig. 4. Probabilities of encounters with humpback whales at varying vessel
when the vessel travelled at speeds of 12.5kts (6.4m/s) or speeds, where lines represent monthly predictions based on binomial
regression and the shaded area represents the 95% confidence interval.
less. As such, encounters occurred for every 37.0 on effort
nautical miles when travelling 12.5kts or less and every 10.9
on effort nautical miles when travelling faster than 12.5kts. encounters and speed: December (p-value: 0.03, [Link] =
76), February (p-value: 0.006, [Link] = 213), and March
Distribution of encounters above and below the threshold (p-value: 0.003, [Link] = 275) (Fig. 4).
speed
There was no clear trend in distribution of encounters DISCUSSION
when travelling at speeds below 12.5kts (6.4m/s) (Fig. 3). Whale-vessel collisions are a matter of concern globally. To
However, when travelling at speeds greater than 12.5kts date, very few studies have attempted to quantify the risk of
(6.4m/s), a higher frequency of encounters was observed in a whale being struck by a vessel by taking into account the
the Au’Au Channel, which is covered by transect lines 1–9. frequency of close encounters at varying vessel speeds
(Richardson et al., 2011). Previous studies have assessed the
Probability of encounters with varying vessel speed by risk of whale-vessel collisions by establishing co-occurrence
month of whales within major shipping routes (Redfern et al.,
A significant positive relationship between speed and 2013). The implications of speed on mortality rate
probability of encounter was identified (p = value: < 0.001, (Vanderlaan et al., 2009) and encounter distance (Gende, et
[Link] = 798). When data were further divided by month, al., 2011) has also been investigated. This study aimed at
three months were found to significantly vary from intercept assessing the rate of close encounters (< 300m) with
only models showing a postive relationship between humpback whales at varying vessel speeds.

Fig. 3. Encounters per km travelled at speeds (A) above and (B) below the identified 12.5kts (6.4m/s) threshold within the four-island region of Maui, Hawai’i
between 2 February 2013 and 31 March 2016.
 J. CETACEAN RES. MANAGE. 17: 57–64, 2017 61

Change-point modelling ACCOBAMS workshop on ship strikes noted that there

Despite the relatively small change in mean sighting distance could be many interpretations of a near miss and a clear
identified using the change point model, the speed threshold definition is required (IWC, 2011). The terms as outlined in
of 12.5kts (6.4m/s) showed a significant decrease in the this study were designed to quantify the risk of vessel strikes
frequency of encounters when traveling below this threshold. by using close encounters (< 300m) as proxies for whale-
The small change in mean sighting distance is likely a result vessel collisions. Results from this study relating speed to
of analysing only whales that were sighted within 300m of probability of encounters, in conjunction with other studies
the vessel. The speed threshold of 12.5kts (6.4m/s) is similar relating speeds to encounter distance and lethality (Gende et
to results presented in Gende et al., (2011), which identified al., 2011; Vanderlaan and Taggart, 2007), all point to similar
a threshold speed of 11.8kts (6.1m/s), despite utilising large speed thresholds of 11–13kts (5.7–6.7m/s).
cruise ships and including encounter distances up to 1000m.
These results suggest that speeds in excess of ~ 12kts Age-class and susceptibility to whale-vessel collisions
(6.2m/s), regardless of vessel size, will significantly increase SE occurred across all age-classes. However, lone adults
the likelihood of whale-vessel collisions. were more likely to be involved in a SE than other
compositions recorded. This differs from other findings
Distribution of encounters above and below the threshold which show a significantly greater proportion of calves and
speed sub-adults involved in SE than the general population (e.g.
A reduction in speed may be favorable for preventing whale- Richardson et al., 2011),. The number of lone adult SE
vessel collisions over other options, such as reduced or increased from 2014 to 2015, suggesting that there are yearly
closed traffic areas, as we observed a threefold reduction in variations in the population, as shown by Tonachella et al.
encounters when vessel speeds were reduced to 12.5kts (2012). If some years are peak years for calving, there will
(6.4m/s) or less, and noted no clear trends in distribution of be more young whales present and therefore an increased
encounters for the four-island region. Furthermore, the susceptibility of that age class to a collision. If, however,
implementation of a speed limit is much easier and is more there are lulls in the calving rate, the opposite will be true
likely to become adopted rather than trying to minimise and more SE with adult whales would be expected.
traffic within an area. The age-class composition of NM revealed that 48.0% of
all NM involved a calf, and yet calves comprise only 7.0–
Probability of encounters with varying vessel speed by 9.0% of the Hawai’ian population of humpback whales
month (Mobley et al., 2001). This supports earlier research findings
The contrast in the monthly rate of encounters suggests that indicating that calves and juveniles are highly vulnerable to
the risk of a whale-vessel collision varies with month and vessel strikes (Laist et al., 2001; Lammers et al., 2013). This
whale abundance. During February, the peak humpback is likely due to a combination of calf related traits such as:
whale season in Hawai’i, the probability of encounter more time spent at the surface to breathe than adults, surfacing
increases from ~35% to 50% when vessel speed is increased often without the mother if the pod is stationary, being less
from below 12.5kts (6.4m/s) to above. Similarly, a visible than adults, and being relatively naive to interactions
probability analysis modelling the lethality of vessel strikes with vessels (Laist et al., 2001; Lammers et al., 2013). In
with speed found a significant increase in the likelihood of Hawai’ian waters, 63.5% of the 52 collisions with humpback
lethality when vessel speeds exceeds 12kts (6.2m/s) whales between 1975–2011, in which age-class was specified,
(Vanderlaan and Taggart, 2007). As such, vessel speed involved either a calf or juvenile (Lammers et al., 2013).
restrictions are being used as mitigation measures in various
locations (e.g. USA: Gulf of Maine and Glacier Bay, Alaska; Recommendations
New Zealand: Hauraki Gulf, Auckland) to reduce the Although data was collected within the four-island region,
occurrence and/or severity of whale-vessel collisions with results from previous literature (Constantine et al., 2015;
large whale species (Constantine et al., 2015; Gende et al., Currie et al., 2014; Guzman et al., 2013; Laist et al. 2014;
2011; Vanderlaan et al., 2009). Lammers et al., 2013; Richardson et al., 2011; Stack et al.,
Combining information on the rate of near collisions with 2013; van der Hoop et al., 2014; Vanderlaan and Taggart,
the severity (Vanderlaan et al., 2009) of whale vessel 2007) suggest implementation of a 12–13kts (6.2–6.7m/s)
collisions based on differing vessel speeds provides insight speed limit in areas and/or in seasons with high densities
into the efficacy of speed restrictions as a management tool. of humpback whales is warranted. Furthermore, speed
Reduced speed will not only allow whales more time to restrictions have been proven a successful mitigation
manoeuvre, but also increases reaction time for a vessel to measure (Gende et al., 2011; Vanderlaan et al., 2009).
stop or change course if they are able (Stack et al., 2013). Instances of whale-vessel collisions still occur at speeds
Consequently, this could reduce the incidence of collisions. below this threshold (Laist et al., 2001; Vanderlaan and
The average speed of whale-vessel collisions reported from Taggart, 2007) and adoption of programs such as the ‘Be
1979–2011 in Maui was 14.7kts (7.6m/s) and, of these Whale Aware’ by Pacific Whale Foundation (PWF, 2015)
collisions, 52.9% were at speeds ≥ 15kts (7.7m/s) (Lammers and ‘Ocean Etiquette and guidelines’ by NOAA (NOAA,
et al. 2013). Current results suggest implementation of a 2015b) should continue to be implemented to help further
speed guideline in the four-island region of Maui would be mitigate whale-vessel collisions. As both whale and human
most effective during peak whale season (February–March). populations increase, with a concurrent increase in
The defining of SE and NM at distances of 300m and anthropogenic activities in the marine environment, more
80m respectively (Stack et al., 2013) differs from the term scientific research leading to sound management strategies
near miss defined in IWC (2011) as 100m. The IWC- will ensure that both humans and animals can safely co-exist.
62 CURRIE et al.: MODELLING WHALE – VESSEL ENCOUNTERS

ACKNOWLEDGEMENTS Knowlton, A.R. and Kraus, S.D. 2001. Mortality and serious injury of
northern right whales (Eubalaena glacialis) in the western North Atlantic
The authors wish to thank Emmanuelle Martinez, Betsy Ocean. J. Cetacean Res. Manage. 2:193–208.
Davidson and Shannon Easterly for their help in field Laist, D.W., Knowlton, A.R., Mead, J.G., Collet, A.S. and Podesta, M. 2001.
Collisions between ships and whales. Mar. Mamm. Sci. 17(1):35–75.
work and data collection. Additionally, we thank Pacific Laist, D.W., Knowlton, A.R. and Pendelton, D. 2014. Effectiveness of
Whale Foundation’s members and supporters for providing mandatory vessel speed limits for protecting North Atlantic right whales.
the funding necessary to conduct this study. Finally, Endanger. Species Res. 23:133–147.
Lammers, M.O., Pack, A.A., Lyman, E.G. and Espiritu, L. 2013. Trends in
our gratitude goes to all the research interns between collisions between vessels and North Pacific humpback whales
2013 and 2016, who contributed significantly to research (Megaptera novaeangliae) in Hawaiian waters (1975–2011). J. Cetacean
efforts. This research was conducted under NMFS permit Res. Manage. 13(1):73–80.
Mobley, J.M., Spitz, S., Grotefendt, R., Forestell, P., Frankel, A. and Bauer,
16479. G. 2001. Abundance of humpback whales in Hawaiian waters: Results
of 1993–2000 aerial surveys. Report to the Hawaiian Islands Humpback
REFERENCES Whale National Marine Sanctuary. 9pp.
Neilson, J.L., Gabriele, C.M., Jensen, A.S., Jackson, K. and Straley, J.M.
Akaike, H. 1973. Information theory and an extension of the maximum 2012. Summary of reported whale-vessel collisions in Alaskan waters. J.
likelihood principle. pp.267–81. In: B.N. Petrov and F. Csáki (eds.) Mar. Biol. 2012:1–18.
Proceedings of the 2nd International Symposium on Information Theory, NOAA. 2015. Ocean Etiquette. [Available at: [Link]
Tsahkadsor, Armenia, USSR, 2–8 September 1971, Budapest: Akadémiai facts/[Link] Accessed on 15 April 2015.]
Kiadó. O’Connor, S., Campbell, R., Cortez, H. and Knowles, T. 2009. Whale Watching
Baker, C.S. and Herman, L.M. 1981. Migration and local movement of Worldwide: tourism numbers, expenditures and expanding economic
humpback whales (Megaptera novaeangliae) through Hawaiian waters. benefits. Special report from the International Fund for Animal Welfare.
Can. J. Zool. 59(3): 460–9. Yarmouth, International Fund for Animal Welfare, Yarmouth, MA, USA.
Buckland, S.T., Anderson, D.R., Burnham, K.P., Laake, J.L., Borchers, D.L. Pacific Whale Foundation. Be Whale Aware. Accessed on 15 October 2015.
and Thomas, L. 2004. Advanced Distance Sampling. Oxford University 295pp. [Available at: [Link]
Press, Oxford. 434pp. accessed 15 October 2015]
Calambokidis, J., Falcone, E.A., Quinn, T.J., Burdin, A.M., Clapham, P.J., Panigada, S., Pesante, G., Zanardelli, M., Capoulade, F., Gannier, A. and
Ford, J.K.B., Gabriele, C.M., LeDuc, R., Mattila, D., Rojas-Bracho, L., Weinrich, M.T. 2006. Mediterranean fin whales at risk from fatal ship
Straley, J.M., Taylor, B.L., Urbán, J.R., Weller, D., Witteveen, B.H., strikes. Mar. Poll. Bull. 52(10): 1,287–98.
Yamaguchi, M., Bendlin, A., Camacho, D., Flynn, K., Havron, A., Panigada, S., Leaper, R. and Arcerdillo, A. 2009. Ship strikes in the
Huggings, J., Maloney, N., Barlow, J. and Wade, P.R. 2008. SPLASH: Mediterranean Sea: Assessment and identification of mitigation measures.
Structure of populations, levels of abundance and status of humpback Paper SC/61/BC2 presented to the IWC Scientific Committee, 22–25
whales in the North Pacific. Report submitted by Cascadia Research June 2009, Madeira, Portugal (unpublished). 5pp. [Available from the
Collective to USDOC, Seattle, WA, USA. Contract AB133F-03-RP-0078 Office of this Journal].
(unpublished). [Available from the author – JC]. Redfern, J.V., McKenna, M.F., Moore, T.J., Calambokidis, J., Deangelis,
Calambokidis, J. and Barlow, J. 2004. Abundance of blue and humpback M.L., Becker, E. A., Barlow, J., Forney, K.A., Fiedler, P.C. and Chivers,
whales in the eastern North Pacific estimated by capture-recapture and S.J. 2013. Assessing the risk of ships striking large whales in marine
line-transect methods. Mar. Mamm. Sci. 20(1):63–85. spatial planning. Conserv. Biol. 27(2):292–302.
Constantine, R., Johnson, M., Riekkola, L., Jervis, S., Kozmian-Ledward, Richardson, D.T., Silva, I.F., Macie, A., Rankin, R.W., Maldini, D. and
L., Dennis, T., Torres, L.G. and Aguilar de Soto, N. 2015. Mitigation of Kaufman, G.D. 2011. Whale surprise encounters and near misses: proxies
vessel-strike mortality of endangered Bryde’s whales in the Hauraki Gulf, of vessel strikes in Maui County waters. Document SC/63/BC2 presented
New Zealand. Biol. Conserv. 186:149–57. to the IWC Scientific Committee, Tromso, Norway (unpublished). 19pp.
Currie, J.J., Stack, S.H., Davidson, E., Kaufman, G.D., Martinez, E. 2014. [Available from the Office of this Journal]
Results from two years of line transect surveys utilizing surprise Silber, G.K., Slutsky, J. and Bettridge, S. 2010. Hydrodynamics of a
encounters and near-misses as proxies of vessels collisions with ship/whale collision. J. Exp. Mar. Biol. Ecol. 391:10–19.
humpback whales (Megaptera novaeangliae) in the four-island region of Smultea, M.A. 1994. Segregation by humpback whale (Megaptera
Maui, Hawai’i, USA. Paper SC/65b/HIM01 presented to the IWC novaeangliae) cows with a calf in coastal habitat near the island of
Scientific Committee, 12–24 May 2014, Bled, Slovenia, (unpublished). Hawaii. Can. J. Zool. 72(5):805–11.
21pp. [Available from the Office of this Journal]. Stack, S.H., Currie, J.J., Davidson, E.H., Frey, D., Maldini, D., Martinez,
Department of Business, Economic Development and Tourism (DBEDT). E. and Kaufman, G.D. 2013. Preliminary results from line transect
2015. Research Economics and Analysis. Accessed May 2015. [Available surveys utilizing surprise encounters and near-misses as proxies of vessels
at: [Link] collisions with humpback whales (Megaptera novaeangliae) in Maui
Dolman, S.J., Weir, C.R. and Jasny, M. 2009. Comparative review of marine County waters, Hawaii, USA. Paper SC/65a/WW04 presented to the IWC
mammal guidance implemented during naval exercises. Mar. Poll. Bull. Scientific Committee, 3–15 June 2013, Jeju, Korea (unpublished). 20pp.
58:465–77. [Available from the Office of this Journal]
Douglas, A.B., Calambokidis, J., Raverty, S., Jeffries, S.J., Lambourn, D.M. Tonachella, N., Nastasi, A., Kaufman, G., Maldini, D. and Rankin, R. W.
and Norman, S.A. 2008. Incidence of ship strikes of large whales in 2012. Predicting trends in humpback whale (Megaptera novaeangliae)
Washington State. J. Mar. Biol. Ass. U.K. 88(06): 1,121–32. abundance using citizen science. Pac. Conserv. Biol. 18(4):297–309.
ESRI. 2011. ArcGIS Desktop: Release 10.: Environmental Systems Van der Hoop, J.M., Moore, M.J. and Barco, S.G., Colde, T.V.N., Daoust,
Research Institute, Redlands, CA. [Available from: [Link] P-Y., Henry, A., McAlpine, D.F., McLellan, W.A., Winner, T., and Solow,
Gende, S.M., Hendrix, A.N., Harris, K.R., Eichenlaub, B., Nielson, J. and A.R. 2013. Assessment of management to mitigate anthropogenic effects
Pyare S. 2011. A Bayesian approach for understanding the role of ship on large whales. Conserv. Biol. 27:121–133.
speed in whale-ship encounters. Ecol. Appl. 21:2,232–40. Van Waerebeek, K., Baker, A.N., Félix, F., Gedamke, J., Iñiguez, M.,
Guzman, H.M., Gomez, C.G., Guevara, C.A. and Kleivane, L. 2013. Sanino, G.P., Secchi, E., Sutaria, D., Van Helden, A., and Wang, Y. 2007.
Potential vessel collisions with southern hemisphere humpback whales Vessel collisions with small cetaceans worldwide and with large whales
wintering off Pacific Panama. Mar. Mamm. Sci. 29(4):629–42. [Available in the Southern Hemisphere, an initial assessment. Latin Am. J. Aquat.
at: [Link] Mamm. 6:43–69.
International Whaling Commission. 2011. Report of the Joint IWC- Vanderlaan, A.S.M., and Taggart, C.T. 2007. Vessel collisions with whales:
ACCOBAMS Workshop on Reducing Risk of Collisions between Vessels the probability of lethal injury based on vessel speed. Mar. Mamm. Sci.
and Cetaceans, 21–24 September 2010, Beaulieu-Sur-Mer, France. Paper 23(1):144–156.
IWC/63/CC8 presented to the IWC Conservation Committee, July 2011, Vanderlaan, A.S., Corbett, J.J., Green, S.L., Callahan, J.A., Wang, C.,
Jersey, Channel Islands, UK. 41pp. [Paper available from the Office of Kenney, R.D., Taggart, C.T. and Firestone, J. 2009. Probability and
this Journal] mitigation of vessel encounters with North Atlantic right whales.
Jensen, A.S. and Silber, G.K. 2004. Large whale ship strike database. US Endanger. Species Res. 6(3):273–285.
Department of Commerce, National Oceanic and Atmospheric Weinrich, M., Pekarcik, C., and Tackaberry, J. 2010. The effectiveness of
Administration (NOAA) Technical Memorandum NMFS-OPR-25. 37pp. dedicated observers in reducing risks of marine mammal collisions with
[Available from: http//[Link]]. ferries: A test of the technique. Mar. Mamm. Sci. 26(2):460–70.
Killick, R. and Eckley, I.A. 2014. changepoint: An R Package for Wood, S.N. 2006. Generalized Additive Models: An Introduction with R.
Changepoint Analysis. J. Stat. Softw. 58(3):1–19. Taylor and Francis, UK. 384pp.
 J. CETACEAN RES. MANAGE. 17: 57–64, 2017 63

APPENDIX 1

Appendix Fig. 1. Map depicting ship traffic densities of vessels equipped with AIS transceivers in the four-island region of Maui,
Hawai’i over a one year period. Source: Data for map provided by PacIOOS ([Link] which is a part of the
US Integrated Ocean Observing System (IOOS), funded in part by National Oceanic and Atmospheric Administration (NOAA)
Awards, NA11NOS0120039 and NA16NOS0120024.

Appendix Fig. 2. Map depicting tourism vessel traffic densities of eight vessels in the four-island region of Maui, Hawai’i over
a one year period.
View publication stats