106
(depth 551 m, no. 12 in Table 2 ), we discovered
high abundances of invertebrates and fi sh (such
as Ophiura sarsii , sea stars, Anguilliformes sp.,
and Sebastolobus macrochir (Günther 1877)) in a
submarine canyon area where much debris had
accumulated. An extremely large number of the
anthomedusan jellyfi sh Euphysa japonica were
observed near the seafl oor at some stations, but
the identity and substrate habitat preferences of
its benthic polyp stages have not yet been determined (Lindsay et al. 2008 ), so an analysis of
correlations with debris distributions was not
possible at the present time.
During the ROV survey in 2013, high densities of brittle stars were observed at stations 1 and
2 (Table 1 and 2 ) but not as many occurred at station 4. At station 2, the body sizes of brittle stars
were signifi cantly different from those occurring
at the other stations. At station 4, we found drag
marks made by trawl nets and a lower diversity
and population size of macrofauna.
Although the area surveyed by the ROV
Crambon was limited, detailed information on
debris could be collected (Fig. 4 ). The number of
debris items was obviously higher within the bay
area (no. 5; the Yamada Bay) on the July of 4th.
Eight debris items were observed in an approximately 50 × 50 m area.
Surveys by side-scan sonar were able to investigate broader areas of the deep-sea fl oor,
although the resolution of acoustic images was
lower than that of the video images obtained during the ROV survey (nos. 21 to 23 in Table 2 ).
The acoustic survey off Miyagi found a large
debris item of approximately 10–15 m in length
(no. 23). In coastal areas, numerous small debris
items, mostly less than 1 m in length, were able to
be distinguished in the acoustic images (Fig. 5 ;
nos. 24–26 in Table 2 ).
As shown in Tables 1 and 2 , the amount of
debris decreased substantially with increasing
water depth and/or distance from the coastline.
The average number of debris items at the ten
main, video-surveyed study sites (1, 2, 4, 12,
14–19) at depths from 289 to 890 m and at distances from 11 to 57 km from the coastline, was
5,195 (standard deviation [SD] = 5,133) per km
2
(Tables 1 and 2 ). It was 1,116 per km
2 (SD = 961)
at the fi ve sites (7, 9–11, 13) in deeper areas near
the trench (>1,000 m depth), which were over
100 km from the coastline. In canyon areas (2, 4,
12, 18, 19), the average number of debris items
per km
2 was 6,761 (SD = 3,528), and this was
higher than at other areas of similar depth that we
studied (mean = 3,629, SD = 6,381). Of the
observed debris categories, “plastics” were most
numerous (113), followed by “wood and paper”
(85). A large subset of the debris (over 237 items)
were unclassifi able.
4
Discussion
Our preliminary results suggest that the direct
impact of tsunami-derived debris in deep-sea
areas off the Sanriku region (ca. 200–800 m
depth) is not as severe as in coastal environments
(Seike et al. 2013 ; Urabe et al. 2013 ).
Accumulations of tsunami-derived debris were
found mainly within submarine canyons in deepsea areas, while in coastal areas the debris was
scattered across the seafl oor. The transportation
power of tsunamis, turbidity currents, and the
geomorphological features of the seafl oor appear
to greatly affect the distribution pattern of marine
debris (Arai et al. 2013 ).
Our observations also showed that different
kinds of debris, such as wood, plastics, and construction materials, still occur in mixed aggregations. These debris items accumulate sessile
species that need hard substrates on which to
attach. This is assumed to be because of the lack
of such surfaces on the largely muddy deep-sea
fl oor, as was observed in the canyon area.
Different materials are expected to move and
decompose in different ways, and the composition of the debris aggregations is therefore
expected to change over time due to transportation and decomposition. For example, plastics
can be harmful when accidentally ingested
(Ramirez-Llodra et al. 2011 ). Flustering of plastics is also suspected to contribute to the accumulation of harmful substances in organisms, where
micro-plastics are known to be a problem
(Andrady 2011 ). In the case of wood or mixtures
of wood and fabric materials, such debris attracts
T. Yamakita et al.
(depth 551 m, no. 12 in Table 2 ), we discovered
high abundances of invertebrates and fi sh (such
as Ophiura sarsii , sea stars, Anguilliformes sp.,
and Sebastolobus macrochir (Günther 1877)) in a
submarine canyon area where much debris had
accumulated. An extremely large number of the
anthomedusan jellyfi sh Euphysa japonica were
observed near the seafl oor at some stations, but
the identity and substrate habitat preferences of
its benthic polyp stages have not yet been determined (Lindsay et al. 2008 ), so an analysis of
correlations with debris distributions was not
possible at the present time.
During the ROV survey in 2013, high densities of brittle stars were observed at stations 1 and
2 (Table 1 and 2 ) but not as many occurred at station 4. At station 2, the body sizes of brittle stars
were signifi cantly different from those occurring
at the other stations. At station 4, we found drag
marks made by trawl nets and a lower diversity
and population size of macrofauna.
Although the area surveyed by the ROV
Crambon was limited, detailed information on
debris could be collected (Fig. 4 ). The number of
debris items was obviously higher within the bay
area (no. 5; the Yamada Bay) on the July of 4th.
Eight debris items were observed in an approximately 50 × 50 m area.
Surveys by side-scan sonar were able to investigate broader areas of the deep-sea fl oor,
although the resolution of acoustic images was
lower than that of the video images obtained during the ROV survey (nos. 21 to 23 in Table 2 ).
The acoustic survey off Miyagi found a large
debris item of approximately 10–15 m in length
(no. 23). In coastal areas, numerous small debris
items, mostly less than 1 m in length, were able to
be distinguished in the acoustic images (Fig. 5 ;
nos. 24–26 in Table 2 ).
As shown in Tables 1 and 2 , the amount of
debris decreased substantially with increasing
water depth and/or distance from the coastline.
The average number of debris items at the ten
main, video-surveyed study sites (1, 2, 4, 12,
14–19) at depths from 289 to 890 m and at distances from 11 to 57 km from the coastline, was
5,195 (standard deviation [SD] = 5,133) per km
2
(Tables 1 and 2 ). It was 1,116 per km
2 (SD = 961)
at the fi ve sites (7, 9–11, 13) in deeper areas near
the trench (>1,000 m depth), which were over
100 km from the coastline. In canyon areas (2, 4,
12, 18, 19), the average number of debris items
per km
2 was 6,761 (SD = 3,528), and this was
higher than at other areas of similar depth that we
studied (mean = 3,629, SD = 6,381). Of the
observed debris categories, “plastics” were most
numerous (113), followed by “wood and paper”
(85). A large subset of the debris (over 237 items)
were unclassifi able.
4
Discussion
Our preliminary results suggest that the direct
impact of tsunami-derived debris in deep-sea
areas off the Sanriku region (ca. 200–800 m
depth) is not as severe as in coastal environments
(Seike et al. 2013 ; Urabe et al. 2013 ).
Accumulations of tsunami-derived debris were
found mainly within submarine canyons in deepsea areas, while in coastal areas the debris was
scattered across the seafl oor. The transportation
power of tsunamis, turbidity currents, and the
geomorphological features of the seafl oor appear
to greatly affect the distribution pattern of marine
debris (Arai et al. 2013 ).
Our observations also showed that different
kinds of debris, such as wood, plastics, and construction materials, still occur in mixed aggregations. These debris items accumulate sessile
species that need hard substrates on which to
attach. This is assumed to be because of the lack
of such surfaces on the largely muddy deep-sea
fl oor, as was observed in the canyon area.
Different materials are expected to move and
decompose in different ways, and the composition of the debris aggregations is therefore
expected to change over time due to transportation and decomposition. For example, plastics
can be harmful when accidentally ingested
(Ramirez-Llodra et al. 2011 ). Flustering of plastics is also suspected to contribute to the accumulation of harmful substances in organisms, where
micro-plastics are known to be a problem
(Andrady 2011 ). In the case of wood or mixtures
of wood and fabric materials, such debris attracts
T. Yamakita et al.