35

Global Environmental Research ©2014 AIRIES 18/2014: 35-46 printed in Japan

Tsunami-induced Changes in a Shallow Brackish Lagoon Ecosystem

(Gamo Lagoon) in Sendai Bay, Japan

Gen KANAYA1*, Hideaki MAKI1, Takao SUZUKI2, Waka SATO-OKOSHI3 and Eisuke KIKUCHI4

*1National Institute for Environmental Studies,

16-2 Onogawa, Tsukuba, Ibaraki, 305-8506 Japan 2Graduate School of Life Science, Tohoku University,

Aramaki-Aoba, Aoba-ku, Sendai, 980-8578 Japan 3Graduate School of Agricultural Science, Tohoku University,

1-1 Amamiyamachi, Tsutsumidori, Aoba-ku, Sendai, 981-8555 Japan 4EEC, Miyagi University of Education,

149 Aramaki-Aoba, Aoba-ku, Sendai, 980-0845 Japan *email: gen@nies.go.jp

Abstract Impacts of the 2011 tsunami disaster were investigated in a shallow brackish lagoon (Gamo Lagoon) in

Sendai City, Japan. The tsunami caused drastic changes in lagoon topography, vegetation and sediment characteristics. Muddy sediment was flushed away and the sediment became much sandier. Changes in nutrients and chlorophyll a (chl. a) budgets implied that the lagoon became less eutrophic. Forty-seven macrozoobenthos species and two macroalgal species were nearly extirpated in 2011, while six species including opportunistic polychaetes and amphipods rapidly recovered their population within five months after the tsunami. In 2012 and 2013, some of the species that disappeared, including bivalves, amphipods, decapods and macroalgae, recruited and increased their population sizes. Shorebird populations in Gamo Lagoon and nine other sites in eastern Japan were less affected by the tsunami, since the population size and number of species did not change notably. Compositions of polycyclic aromatic hydrocarbons (PAHs) and normal alkanes (n-alkanes) in lagoon sediment implied petroleum contamination, which was discharged from an oil refinery in Sendai Port although the levels were quite low. A sharp increase in lagoon salinity caused an invasion of stenohaline marine species and limited the growth of the marsh reed Phragmites australis after the tsunami. Therefore lagoon salinity should be regulated to restore the reed-dominated brackish marsh and associated biota of this lagoon. The ongoing construction of huge sea walls is another potential threat to the lagoon ecosystem. Care should be taken to conserve coastal habitats such as tidal flats, salt marshes, sand dunes, as well as their backside landward areas during restoration.

Key words: biotic community, brackish lagoon, macrozoonthos, oil spill, tidal flat, tsunami

1. Introduction

1.1 Background

The Pacific coast of eastern Japan was struck by a huge tsunami on March 11, 2011, following the Great East Japan Earthquake (M 9.0). The tsunami caused extreme physical disturbances of tidal flats and salt marshes in the Tohoku District (e. g., Suzuki, 2011). As a result, biotic communities in coastal soft-bottom habitats were modified drastically (e. g., Kanaya et al., 2012; Seike et al., 2013; Urabe et al., 2013). At present, how-ever, few studies have reported tsunami-induced changes in environmental characteristics of tidal flats (but see Kanaya et al., 2012; Sera et al., 2012). Therefore it is

necessary to know how the tsunami modified the struc-ture and functioning of tidal-flat ecosystems to assess the ecological consequences of the tsunami disaster.

The coastal area of Sendai City, Miyagi Prefecture, was struck by a huge tsunami with a height of 7.2 m, which caused intensive physical disturbances in the area (Sendai City, 2013). In this paper, we report the tsunami-induced changes in a shallow brackish lagoon ecosystem, Gamo Lagoon, in Sendai City, based on a series of surveys conducted during 2011 to 2013. We discuss the direct and indirect (i.e., subsequential) effects of the tsunami disaster and potential threats for conser-vation and management of Gamo Lagoon.

36 G. KANAYA et al.

1.2 Gamo Lagoon Gamo Lagoon is a shallow hypertrophic lagoon

(mean water depth: 0.8 m; area: 0.13 km2) located at the north side of the Nanakita River estuary in Sendai City (Fig. 1). Detailed and long-term data sets on this lagoon before the tsunami disaster helped us assess the tsunami-induced changes in the ecosystem with high reliability (see Gamo Higata Shizen Saisei Kyogikai, 2008; Kanaya et al., 2012 and references therein). Salinity fluctuated tidally from near 0 to over 30 and the spring tidal range was about 50 cm before the tsunami (Gamo Higata Shizen Saisei Kyogikai, 2008). Sandy sediment was distributed near the levee, while organi-cally polluted mud predominated in the inner portion (Kanaya & Kikuchi, 2011). A stone levee with three water gates restricted water exchange. Adjacent fish farming ponds continuously discharged freshwater into the inner portion of the lagoon. The lagoon maintained high phytoplankton biomass in the water column (chl. a; > 100 μg L-1) due to water stagnation and sufficient sup-ply of regenerated nutrients from the highly reduced mud (Kikuchi et al., 1992).

The lagoon was nominated a National Wildlife Pro-tection Area by the Environmental Agency (currently the Ministry of the Environment, MOE) in 1987. The inter-tidal flats were important habitat for migratory shorebirds and macrozoobenthos, including polychaetes, amphipods, bivalves and decapods (Gamo Higata Shizen Saisei Kyogikai, 2008). In muddy areas, by contrast, macro-

zoobenthos had low species richness and density due to bottom water hypoxia and sulfide accumulation in the sediment (Kanaya & Kikuchi, 2011). Reeds, Phragmites australis, vegetated densely along the lagoon edge, which was a resting place for migratory ducks and geese. Sand dune vegetation was well developed on sand bars. The red alga Gracilaria vermiculophylla and green alga Ulva intestinalis (formerly reported as Enteromorpha prolifera) grew densely in the inner subtidal zone, especially in warmer months (Aikins & Kikuchi, 2002).

2. Topography, Vegetation, and Sediment

Characteristics The tsunami caused distinctive changes in the to-

pography and vegetation of Gamo Lagoon. The sand bars almost disappeared after the tsunami, when sea water directly entered the lagoon (satellite image on 14 March 2011, Google TM Earth). Within three months, however, the sand bar had mostly recovered due to deposition of drift sand, and the lagoon recovered its semi-enclosed shape (Fig. 2). A GPS-based vegetation map (see Kanaya et al., 2012 for methods) showed that the pine forest decreased from 4.2 to 2.0 ha, sand dune vegetation from 8.6 to 0.1 ha, and reed marshes from 7.8 to 1.2 ha. In contrast, the area of bare tidal flat increased from 4.7 to 5.3 ha due to the dissapearance of reed marshes (Kanaya et al., 2012). Marsh and sand dune vegetation have not recovered yet as of 2013.

Fig. 1 Location of Gamo Lagoon on the northern side of the Nanakita River Estuary in Sendai City, Japan.

Tsunami-induced Changes in a Shallow Brackish Lagoon Ecosystem 37

Before the tsunami, macroalgae, including red algae, Gracilaria vermiculophylla, and green algae, Ulva intestinalis, grew densely in the inner portion of the lagoon (Fig. 3). The macroalgae were a habitat for algae-associated macrozoobenthos such as free-living amphipods (Aikins & Kikuchi, 2002). Almost all macroalgae, however, were washed away by the tsunami.

In 2011, only small patches of G. vermiculophylla were found in the inner portion (coverage < 30%) and U. intestinalis had completely disappeared. In 2012, their biomass gradually recovered in the inner portion. In 2013, their patches became more dense and expanded to the central lagoon. In 2013, the stenohaline marine species U. pertusa (Yamochi, 2013) grew densely, though it was

Fig. 2 Lagoon topography and vegetation. Areas of tidal flats, pine forest, sand dune vegetation

and reed marshes are shown. The survey was conducted in June 2011 using a handy GPS (see Kanaya et al., 2012 for methods). Data before the tsunami were obtained from Gamo Higata Shizen Saisei Kyogikai (2008).

38 G. KANAYA et al.

rarely found before the tsunami. The sediment characteristics also changed distinc-

tively. Before the tsunami, organically polluted mud was widely distributed in the inner lagoon (Fig. 4). In that area, the biodiversity and abundance of macrozoobenthos were very low due to the toxicity of sedimental sulfide (Kanaya & Kikuchi, 2011). However, the tsunami removed the mud from the inner lagoon, resulting in a sharp decline in the silt-clay content of the sediments (particles < 0.063-mm mesh sieve). We found that sedi-ment in Gamo Lagoon remained sandy for three years at most. This implies that the tsunami-induced changes in sediment granulometry significantly improved the habi-tat quality for macrozoobenthos in the inner lagoon.

3. Changes in Nutrient Budget and Trophic States Gamo Lagoon was hypertrophic with a maximum chl.

a concentration of >100 μg L-1 which was supported by a sufficient supply of regenerated nutrients from muddy reduced sediment in the inner lagoon (Kikuchi et al., 1992; Kanaya et al., 2007). Accordingly the lagoon acted as a source of nutrients and phytoplankton, since it dis-charged water containing high levels of NH4

+ , PO43-, and

chl. a into adjacent waters during ebb tides (Kikuchi et al., 1992).

In August 2011, we measured the tidal flux of dis-solved nitrogen, dissolved phosphorus, and particulate materials including chl. a to assess changes in lagoonal trophic states. Figure 5 shows the temporal changes in

Fig. 3 Changes in coverage of the red algae Gracilaria vermiculophylla (a), green algae Ulva intestinalis (formerly classified as Enteromorpha prolifera) (b), and stenohaline marine species Ulva pertusa following the tsunami (c). Coverage (%) was measured at 62, 48, and 45 stations in 2011, 2012 and 2013, respectively, and are indicated with different colors. Data before the tsunami were obtained from Gamo Higata Shizen Saisei Kyogikai (2008).

Tsunami-induced Changes in a Shallow Brackish Lagoon Ecosystem 39

quality of inflowing / outflowing waters at the lagoon mouth. The inflowing water was rich with NH4

+, NO3-,

dissolved organic nitrogen (DON), and dissolved organic phosphorus (DOP), while the outflowing water was rich with NO2

- and PO43-. Increasing NO2

- with decreasing NH4

+ during the tidal cycle implies further nitrification in the lagoon (Dähnke et al., 2012). Increasing PO4

3- with decreasing DOP during the tidal cycle may possibly have been due to phosphate regeneration from DOP and/or particulate organic phosphorus in the lagoon (e.g., Nowicki & Nixon, 1985; Fang, 2000). A declining con-centration of chl. a, suspended solids (SS), particulate organic carbon (POC), and particulate nitrogen (PN) during the tidal cycle indicates deposition of these particulate materials onto the lagoon sediment.

The diurnal budget of each component was calculated from the concentration and flow rate data (Table 1, see Kikuchi et al., 1992 for methods). We compared them with those measured in 1985 by Kikuchi et al. (1992). In 1985, the lagoon was a significant source of NH4

+ (net export; 21.7 kg N day-1), PO4

3- (8.7 kg P day-1), and chl. a

(0.6 kg day-1) to adjacent waters. In 2011, by contrast, the lagoon became a sink for NH4

+ (net import: 0.7 kg N day-1) and chl. a (net import: 0.7 kg day-1). Though PO4

3- was still exported to adjacent waters, the amount (net export; 1.5 kg P day-1) was a magnitude lower than that in 1985. Such changes would mainly be due to the disap-pearance of muddy sediment in the inner lagoon.

Before the tsunami, chl. a concentration in inner lagoon water (> 100 μg L-1, Kikuchi et al., 1992; Kanaya et al., 2007) was comparable with or much higher than those reported from other eutrophic estuaries of the world (1.6 to 100 μg L-1; Heip et al., 1995). It became much lower, however, after the tsunami (11.5 to 25.9 μg L-1, measurements in July, August, and October 2011). This implies decreasing phytoplanktonic productivity in the water column. We measured the chl. a sedimentation rate in August and September 2011 using sediment traps (see Kanaya et al., 2008 for methods) and found that in 2011 it was less than half that of 2005 (Fig. 6). This indicates reduced organic loading from the water column to the sediment surface in Gamo Lagoon after the tsunami.

This suggests that the essential ecosystem functions of the lagoon, e.g., decomposition of organic matter, nutrient cycling and primary productivity (see the review by Levin et al., 2001) had largely been modified by the tsunami. Reduction in phytoplanktonic productivity

Fig. 4 Silt-clay content (particles < 0.063-mm mesh; %)

at a 1 to 3 cm deep layer of sediment. n = 45, 62, 48, and 45 for 1997, 2011, 2012, and 2013, respectively. Methods were the same as in Kanaya and Kikuchi (2008).

Fig. 5 Temporal changes in water quality at the lagoon mouth. Measurements were conducted on August 4 to 5, 2011, for one tidal cycle. Closed circles, inflowing water; open circles, outflowing water. Abbreviations: DON, total dissolved organic nitrogen; DOP, total dissolved organic phosphorus; SS, suspended solid; POC, particulate organic carbon; PN, particulate nitrogen. Nutrients, DON, and DOP were quantified by Traacs-800 (Bran + Luebbe GmbH). All particulate materials were analyzed the same way as in Kanaya et al. (2008).

40 G. KANAYA et al.

would prevent the sediment from being organically enriched (i.e., increased organic pollution), since the excess deposition of phytoplanktonic detritus was the major cause of sediment deterioration in this lagoon (Kanaya & Kikuchi, 2011).

4. Macrozoobenthos and Shorebirds

In a 2011 survey, 47 of 79 macrozoobenthic species

identified before the tsunami were found to have nearly disappeared (Kanaya et al., 2012). For example, algae- associated amphipods (Eogammarus possjeticus and Melita sp.), sessile fauna (e.g., Fistulobalanus albicostatus and Crassostrea gigas), marsh-associated fauna (Chiromantes dehaani, Assiminea hiradoensis, and Cerithidea rhizophorarum), motile epifauna (Pagurus minutus and Reticunassa festiva), and infaunal decapods and bivalves (Macrophthalmus japonicus, Upogebia

yokoyai, Ruditapes philippinarum, Macoma contabulata, Laternula marilina, and Mya arenaria) were not found shortly after the tsunami (Fig. 7).

In April to June 2011, marine species were found in the lagoon possibly due to stranding from the adjacent sea. These species rarely occurred there before the tsunami since the salinity in Gamo Lagoon was too low for stenohaline marine species to survive (often less than 20; Kurihara, 1990; Kurihara et al., 2000). Most of them, however, were absent until August 2011 as the sand bar recovered and lagoonal salinity gradually declined (Kanaya et al., 2012).

In Gamo Lagoon, seven species, including oppor-tunistic polychaetes and amphipods, rapidly recovered their density within five months after the tsunami. In general, opportunistic species can rapidly colonize empty habitats created by pulsed disturbances (e.g., Grassle & Grassle, 1974). In Gamo Lagoon, therefore, newly created sandy bottoms would provide the opportunists with a preferable “empty habitat” to colonize.

Several species recruited and recovered their popula-tion during 2011 to 2013 (see Fig. 7). In the fall of 2011, juveniles of A. hiradoensis were found in dense con-centrations around the reed marshes in the inner lagoon. In 2012, the amphipods E. possjeticus and Melita sp. occurred with high densities in macroalgal patches. In 2012, small individuals of F. albicostatus and C. gigas were found on hard substrates in the intertidal zone. Recruitments of ghost shrimp N. japonica and the bivalves R. philippinarum, M. contabulata, and L. marilina were also found in 2012 and/or 2013. Although several species such as U. yokoyai, P. minutus, C. rhizophorarum, and R. festiva are still absent or rarely found, the macrozoobenthic community gradually re-covered after the tsunami.

Tidal flats along the Japanese coast are important habitat for migratory shorebirds or waders (Ministry of the Environment, 2013). To assess the effects of the tsunami disaster on the shorebird community, we com-piled data from a five-year survey at Gamo Lagoon and nine monitoring sites along the Pacific coast of Japan (data were provided by the Monitoring Sites 1000 Project, MOE). Figure 8 shows the individual number and species diversity of shorebirds in 2008 to 2012 at Takase River (Aomori Pref.), Gamo and Torinoumi Lagoons (Miyagi Pref.), Matsukawaura Lagoon and Natsui River (Fukushima Pref.), Kashimanada Shore (Ibaraki Pref.), and Ichinomiya River, Banzu Tidal flat, Yatsu Lagoon, and Sanbanze Tidal flat (Chiba Pref.). The results showed that the effects of the tsunami disaster were inconspicuous at these sites. At the sites, habitat sub-strates (i.e., tidal flats and sandy shores) were retained after the tsunami (judging from satellite images on March 18 and 31 and April 6, 2011; Google TM Earth). Rapid recovery of opportunistic macrozoobenthos, as found in Gamo Lagoon, may also assure a supply of potential prey for shorebirds. This suggests that the population of migratory shorebirds was less affected by the tsunami disaster.

Table 1 Diurnal budgets (kg day −1) of dissolved and particulate materials in Gamo Lagoon. See Kikuchi et al. (1992) for methods.

Flux (kg day −1) a Budget (Δ) 1985 2011 1985 2011 In Out In OutNO3

−+NO2−−N 23.6 18.5 2.3 7.4 −5.1 +5.1

NO2−−N - - 1.0 7.3 - +6.2

NO3−−N - - 1.3 0.1 - −1.2

NH4+−N 22.0 43.7 48.3 47.6 +21.7 −0.7

PO43−−P 7.7 16.5 2.3 3.8 +8.7 +1.5

DON - - 36 25 - −11DOP - - 1.9 0.8 - −1.0SS - - 1844 1297 - −547Chlorophyll a 0.5 1.1 1.0 0.3 +0.6 −0.7POC - - 168 58 - −110PN - - 31 11 - −20

a Data were obtained in July to October 1985 (4 days) and in August in 2011 (1 day). Data in 1985 were from Kikuchi et al. (1992). Abbreviations: -, no data; DON, total dissolved organic nitrogen; DOP, total dissolved organic phosphorus; SS, suspended solids; POC, particulate organic carbon; PN, particulate nitrogen.

Fig. 6 Chlorophyll a sedimentation rates (mg m-2 d-1) measured

by sediment traps (3.3 cm Ø, 7.8 cm high, n = 3) set in field for two tidal cycles. This was conducted on July 21 and August 21, 2005, and August 4 and September 1, 2011. The methods were the same as in Kanaya et al. (2008). Bars indicate SD.

Tsunami-induced Changes in a Shallow Brackish Lagoon Ecosystem 41

Fig. 7 Macrozoobenthos in Gamo Lagoon. Representative taxa are shown for three groups, i.e., species that have

disappeared or nearly disappeared (47 taxa) (a), temporal immigrants from the sea, April to July 2011 (12 taxa) (b), and rapidly recovered species (7 taxa) (c). For methods and original data, see Kanaya et al. (2012).

42 G. KANAYA et al.

5. An Oil Spill An oil refinery near Sendai Port was destroyed on

March 11, 2011 and burned for several days afterwards (Sendai City, 2013). As a result, crude oil and its pyrolytic products spilled into the sea. In the summer to fall of 2011, we collected sediment samples for analysis of normal alkanes (n-alkanes) and polycyclic aromatic hydrocarbons (PAHs) from several locations in Gamo Lagoon. These components were analyzed by gas chromatograph / mass spectrometer (GC-MS) according to the methods suggested by the Ministry of the En-vironment (2003) with some modifications. Each sedi-ment sample was subjected to subsequential batch extraction with acetone and dichloromethane, and the

extracts were cleaned up and fractionated by silica gel column chromatography. Fractions for aliphatic or aromatic hydrocarbons with internal standards were separately injected into GC-MS with a fused silica capil-lary column in the splitless mode. Each peak of n-alkanes and PAHs was qualified and quantified in the selected ions monitoring mode.

Typical chromatograms and contents (μg kg-1) of n-alkanes, pristane, and phytane are shown in Fig. 9. The composition differed among sediments and oils meas-ured in this study (Table 2). The carbon preference index (CPI C24-34), i.e., ratio of n-alkanes with even carbon numbers (C24 to C34) to those with odd carbon numbers (C25 to C33), is used as a source indicator of n-alkanes in marine sediments (e.g., Bray &Evans, 1961). This is due

Fig. 8 Individual numbers and species richness of shorebirds at ten sites along the Pacific Coast of eastern Japan. (a) Takase Riv., (b) Gamo, (c) Torinoumi (d) Matsukawaura, (e) Natsui Riv., (f) Kashimanada, (g) Ichinomiya Riv., (h) Banzu, (i) Yatsu, (j) Sanbanze. Data are from the MOE Monitoring Sites 1000 Project (Shorebirds Datapackage 2012.zip, http://www.biodic.go.jp/ moni1000/findings/data/index.html). Maximum densities in three seasons, spring (April–May), fall (August–September), and winter (December–February), were summed for each year. For 2012, the data were multiplied by 1.5 since they were collected only in spring and fall.

Tsunami-induced Changes in a Shallow Brackish Lagoon Ecosystem 43

to the fact that natural uncontaminated sediments with terrigenous sources are dominated by heavy n-alkanes with odd carbon numbers (e.g., Bray & Evans, 1961; Gao & Chen, 2008). Accordingly, a CPI close to 1 implies that the n-alkanes were derived mainly from petroleum (1.0 to 1.1, see Table 2), while 3 to 6 indicate the dominance of biogenic sources. In Gamo Lagoon, the CPI of the sediment was 1.6 ± 1.0 (mean ± SD, n = 7), which is intermediate between those of biogenic sources and petrol, suggesting a mixture of these sources. The pristane / phytane ratio is generally higher than 1 (3 to 5) in natural uncontaminated sediments, while it is close to 1.0 under petroleum contamination (e.g., Gao & Chen, 2008). In Gamo Lagoon, the ratio was 0.55 ± 0.30 and was closer to that of the stranded crude oil (0.60). This indicates that the n-alkanes in the sediment were derived mainly from stranded crude oil.

The composition of PAHs also differs among sources,

e.g., petroleum, sewage, biomass combustion, and petro-leum combustion (e.g., Gao & Chen, 2008). In Gamo Lagoon, the high fractional contribution of Σ alkylated PAHs to total PAHs implies the contribution of petrogenic sources (Table 2). We also calculated the ratios of phenanthrene / anthracene (PH / AN) and fluoranthene / pyrene (FLU / PY) as source indicators. In general, sediment polluted by petrol exhibits high PH / AN (>25), while pyrolytic products show high FLU / PY (>1) (Baumard et al., 1998). In Gamo Lagoon, the ratios were 76 ± 83 for PH / AN and 2.2± 3.3 for FLU / PY, suggesting a mixture of petrogenic input and its pyrolytic products in the PAH pool.

Our data indicated that the sediment in Gamo Lagoon was polluted with petrol and its pyrolytic products. The total n-alkane content (875 ± 476 μg kg-1) and total PAHs (102 ± 86 μg kg-1, sum of all PAHs in Table 2) , however, were not considerably higher than those in other estua-rine coastal sediments (see Colombo et al., 1989;

Table 2 Normal alkane (n-alkane) and polycyclic aromatic hydrocarbon (PAH) content in Gamo lagoon sediment in 2011 (Gamo), stranded crude oil at Sendai Port in 2011 (Strand.), heavy oil from the Fukushima thermal power station (Heavy), and thin fuel oil A (A-fuel). The carbon preference index (CPI C24-34) was calculated according to Bray and Evans (1961). Data are the mean ± SD (n = 7).

Sediment (μg kg-1) Oil (μg g-1) Gamo (SD) Strand. Heavy A-fuelAlkanes Total n-alkanes (C12-38) 875 (476) 29641 22194 101589CPI C24-34 1 1.6 (1.0) 1.0 1.1 1.0 Pristane 13 (9.3) 576 448 2715 Phytane 30 (26) 963 285 2750 Pristane / Phytane 0.55 (0.30) 0.60 1.6 0.99 PAHs

Acenaphthylene 0.02 (0.05) 1.8 0.29 174 Acenaphthene 0.14 (0.11) 14 12 943 Naphthalene 0.26 (0.21) 4.4 n.d. 2083 ΣC1-4 Naphthalenes 8.5 (6.32) 910 434 102458Dibenzothiophene (DBT) 0.65 (0.43) 33 11 1180 ΣC1-3 DBTs 15 (13.6) 328 131 11647Anthracene (AN) 0.23 (0.33) 3.6 3.9 228 Phenanthrene (PH) 5.4 (2.9) 362 161 2291 ΣC1-3 Phenanthrenes 27 (30) 1764 1281 10955Fluorene 0.27 (0.22) 68 27 832 ΣC1-2 Fluorenes 6.8 (5.1) 361 397 7839 Fluoranthene (FLU) 3.8 (3.2) 25 32 73 Pyrene (PY) 5.9 (4.9) 3.5 2.9 451 Benzanthracene 2.6 (2.5) 36 12 9.5 Chrysene 6.2 (6.0) 231 262 18 Benzo[b,k]fluoranthene 1.9 (2.1) 29 20 0.41 Benzo[a]pyrene 3.2 (4.2) 0.51 0.69 1.5 Benzo[e]pyrene 8.7 (9.1) 2.5 3.4 7.2 Indeno[1,2,3-cd]pyrene 0.66 (0.77) 0.38 0.52 1.1 Dibenz[a,h]anthracene 0.16 (0.22) 0.8 1.0 2.2 Benzo[ghi]perylene 4.0 (4.4) n.d. n.d. n.d. Total PAHs 102 (86) 4178 2791 141192Σ alkylated PAHs 58 (53) 3363 2243 132898PH/AN 76 (83) 100 42 10 FLU/PY 2.2 (3.3) 7.1 11 0.16

1 CPI = 1/2[(C25+C27+C29+C31+C33)/(C24+C26+C28+C30+C32)+(C25+C27+ C29+C31+C33)/(C26+C28+C30+C32+C34)]. Sediment samples in Gamo Lagoon were collected in June (n = 1), July (n = 4), and October of 2011 (n = 2).

Fig. 9 Typical mass chromatograms and content of n-alkanes

(C12-38), pristane (Pri), and phytane (Phy) in lagoon sediment (a) and in stranded crude oil at Sendai Port (b) in 2011. Bars in (a) indicate SD (n = 7). The peak for each n-alkane is indicated with the carbon number.

44 G. KANAYA et al.

Baumard et al., 1998; Chen & Chen, 2011). Page et al. (2002) reported that the lower toxicity threshold of total PAHs for the amphipod Rhepoxynius abronius was 2600 μg kg-1. In Gamo Lagoon, the level of total PAHs was a magnitude lower than that of the lower toxicity threshold. Rapid recovery of the amphipod and polychaete popula-tions, as discussed above, also indicated that the petrol pollution hardly affected the macrozoobenthic commu-nity in Gamo lagoon.

6. Threats to the Conservation of the Gamo

Lagoon Ecosystem Several problems arose after the tsunami disaster,

which could potentially modify the structure, functioning and biodiversity of the lagoon ecosystem. The first is subsequent changes in the saline regime after the tsunami. Figure 10 shows the changes in diurnal maximum and minimum salinity in 1989-1990, 1999-2000, and 2012-2013 measured by a data logger set in field (data from Kurihara, 1990; Kurihara et al., 2000). Diurnal minimum salinity was less than 15 in 1989-1990 and less than 20 in 1999-2000 on most days. By contrast, it was over 20 in 2012-2013. Diurnal maximum salinity was also constantly high after the tsunami. These suggest that the freshwater input significantly declined after the

tsunami, possibly due to topographical changes near the lagoon mouth and cessation of the inflow of freshwater from the fish farming ponds. High salinity should have negative effects on the growth and survival of marsh reeds (e.g., Lissner & Schierup, 1997). Furthermore, increasing salinity would result in structural changes in the biotic community in Gamo Lagoon. In fact, stenohaline marine species such as the crab Portunus trituberculatus and the green algae U. pertusa were found in the summer of 2013.

The second is anthropogenic disturbances arising through the reconstruction of a huge sea wall. The admin-istration plans to construct huge sea walls with a height of 7.2 m T.P. along the coastline of Sendai City (e.g., http://www.thr.mlit.go.jp/Sendai/kasen_kaigan/fukkou/kouzou.html). Since salt-marsh plants, sand-dune vege-tation and macrozoobenthos often show patchy distribu-tion (e.g., Gamo Higata Shizen Saisei Kyogikai, 2008; Japanese Association of Benthology, 2012), construction of sea walls can easily destroy habitats at a local scale. Huge sea walls can also cause habitat quality to deterio-rate by breaking connectivity among habitats. For exam-ple, semi-terrestrial crabs Chiromantes dehaani and C. haematocheir use sand dunes, grasslands, and forests as their main habitats, while they must migrate to the inter tidal zone to breed (Kobayashi, 2000). Therefore, it is

Fig. 10 Diurnal maximum (red solid line) and minimum (blue broken line) salinity in Gamo Lagoon in 1989-1990 (a), 1999-2000 (b), and 2012-2013 (c). Data were obtained by a logger set at 115 m (before the tsunami) and 525 m (after the tsunami) inward from the levee. The data before the tsunami were from Kurihara (1990) and Kurihara et al. (2000). The grey areas indicate no data.

Tsunami-induced Changes in a Shallow Brackish Lagoon Ecosystem 45

necessary to maintain connectivity among habitats for maintaining populations of such cross-habitat taxa.

During reconstruction of sea walls, destruction of reed marshes and tidal flats should be carefully mini-mized. Creation of passages across the sea wall (i.e., like fishways in dams) is one possible solution for keeping the connectivity between the lagoon and back-land habi-tats for migrating species. Setting the sea wall back from the lagoon area would, at present, be one of the most effective and ideal resolutions to conserve the lagoon ecosystem. Proper assessment and countermeasures are strongly needed before sea wall restoration to avoid degradation of habitat quality in estuarine wetlands, i.e., endangered ecosystems along the Japanese Coast.

Acknowledgements

We would like to thank Y. Nakamura, Y. Takagi, T.

Kondo, and the Alumni of Kikuchi Laboratory for their help in field work. K. Kinoshita, A. Oishi, J. Nakano, Y. Okada and the staff of the Marine Biological Research Institute of Japan Co., Ltd., are acknowledged for their help in laboratory work. The editors and two anonymous referees are acknowledged for their reviews and critical comments. This research was partly supported by TEAMS (Tohoku Ecosystem-Associated Marine Sci-ences) from MEXT. Data sets on shorebirds were provided by the Monitoring Sites 1000 Project, MOE.

References Aikins, S. and E. Kikuchi (2002) Grazing pressure by amphipods

on microalgae in Gamo Lagoon, Japan. Marine Ecology Pro-gress Series, 245: 171-179.

Baumard, P., H. Budzinski and P. Garrigues (1998) Polycyclic aromatic hydrocarbons (PAHs) in sediments and mussels of the western Mediterranean Sea. Environmental Toxicology and Chemistry, 17: 765-776.

Bray, E.E. and E.D. Evans (1961) Distribution of n-paraffins as a clue to recognition of source beds. Geochimica et Cosmochimica Acta, 22: 2-15.

Chen, C.-W. and C.-F. Chen (2011) Distribution, origin, and potential toxicological significance of polycyclic aromatic hydrocarbons (PAHs) in sediments of Kaohsiung Harbor, Taiwan. Marine Pollution Bulletin, 63: 417-423.

Colombo, J.C., E. Pelletier, C. Brochu and M. Khalllt (1989) Determination of hydrocarbon sources using n-alkane and polyaromatic hydrocarbon distribution indexes. Case study: Rio de La Plata Estuary, Argentina. Environmental Science and Technology, 23: 888-894.

Dähnke, K., A. Moneta, B. Veuger, K. Soetaert and J.J. Middelburg (2012) Balance of assimilative and dissimilative nitrogen processes in a diatom-rich tidal flat sediment. Biogeosciences, 9: 4059-4070.

Fang, T.H. (2000) Partitioning and behaviour of different forms of phosphorus in the Tanshui estuary and one of its tributaries, northern Taiwan. Estuarine, Coastal and Shelf Science, 50: 689-701.

Gamo Higata Shizen Saisei Kyogikai (2008). A Total Plan for Nature Restoration in Gamo Tidal Flat. Miyagi Prefecture, Sendai, Japan. (in Japanese)

Gao, X. and S. Chen (2008) Petroleum pollution in surface sedi-ments of Daya Bay, South China, revealed by chemical finger-

printing of aliphatic and alicyclic hydrocarbons. Estuarine, Coastal and Shelf Science 80: 95-102.

Grassle, J.F and J.P. Grassle (1974) Opportunistic life histories and genetic systems in marine benthic polychaetes. Journal of Marine Research, 32: 253-284.

Heip, C.H.R., N.K. Goosen, P.M.J. Herman, J. Kromkamp, J.J. Middelburg and K. Soetaert (1995) Production and consump- tion of biological particles in temperate tidal estuaries. Oceanography and Marine Biology: An Annual Review, 33: 1-149.

Japanese Association of Benthology (2012) Threatened Animals of Japanese Tidal Flats: Red Data Book of Seashore Benthos. Tokai University Press, Tokyo, Japan. (in Japanese)

Kanaya, G. and E. Kikuchi (2011) Influences of anthropogenic eutrophication on estuarine benthic ecosystems. Chikyu Kankyo, 16: 33-44. (in Japanese)

Kanaya, G., S. Takagi, E. Nobata and E. Kikuchi (2007) Spatial dietary shift of macrozoobenthos in a brackish lagoon revealed by carbon and nitrogen stable isotope ratios. Marine Ecology Progress Series, 345: 117-127.

Kanaya, G., S. Takagi and E. Kikuchi (2008) Dietary contribution of the microphytobenthos to infaunal deposit feeders in an estua-rine mudflat in Japan. Marine Biology, 155: 543-553.

Kanaya, G., T. Suzuki, H. Maki, Y. Nakamura, Y. Miyajima and E. Kikuchi (2012) Effects of the 2011 tsunami on the topography, vegetation, and macrobenthic fauna in Gamo Lagoon, Japan. Japanese Journal of Benthology, 67: 20-32. (in Japanese with English abstract)

Kikuchi, E., G. Urano and Y. Kurihara (1992) The use of dialysis culture in a study of chlorophyll budget in a brackish lagoon, Japan. Hydrobiologia, 245: 1-13.

Kobayashi, S. (2000) Distribution pattern and ecology of brachy-uran crabs in the riverine environment: their significance in the ecosystem and present condition. Ecology and Civil Engineer-ing, 3: 113-130. (in Japanese with English abstract)

Kurihara, Y. (1990) Report of Fundamental Investigation of Environmental Conservation Measure for Gamo Tidal Flat in 1989. Miyagi Prefecture, Sendai, Japan. (in Japanese)

Kurihara, Y., E. Kikuchi, T. Uehara and T. Suzuki (2000) Report of Fundamental Investigation of Environmental Conservation Measure for Gamo Tidal Flat in 1999. Miyagi Prefecture, Sendai, Japan. (in Japanese)

Levin, L.A., D.F. Boesch, A. Covich, C. Dahm, C. Erséus, K.C. Ewel, R.T. Kneib, A. Moldenke, M.A. Palmer, P. Snelgrove, D. Strayer and J.M. Weslawski (2001) The function of marine critical transition zones and the importance of sediment biodiversity. Ecosystems, 4: 430-451.

Lissner, J. and H.-H. Schierup (1997) Effects of salinity on the growth of Phragmites australis. Aquatic Botany, 55: 247-260.

Ministry of the Environment (2003) Manual for Analytical Methods on the Chemicals to be Monitored. Ministry of the Environment, Tokyo, Japan. (in Japanese)

Ministry of the Environment (2013) The Report of the Shorebirds Census in Japan (Spring 2013). Monitoring Site 1000, Biodi-versity Center, Ministry of the Environment. (in Japanese)

Nowicki, B.L. and S.W. Nixon (1985) Benthic nutrient reminerali-zation in a coastal lagoon ecosystem. Estuaries, 8: 182-190.

Page, D.S., P.D. Boehm, W.A. Stubblefield, K.R. Parker, E.S. Gilfillan, J.M. Neff and A.W. Maki (2002). Hydrocarbon composition and toxicity of sediments following the Exxon Valdez oil spill in Prince William Sound, Alaska, USA. Environmental Toxicology and Chemistry, 21: 1438-1450.

Sendai City (2013) The Great Eastern Japan Earthquake Disaster in Sendai City – Report of the Activities in One Year After the Earthquake. Sendai City, Japan. (in Japanese)

Seike, K., K. Shirai and Y. Kogure (2013) Disturbance of shallow marine soft-bottom environments and megabenthos assem-blages by a huge tsunami induced by the 2011 M9.0 Tohoku-Oki earthquake. PLoS One, 8: e65417.

46 G. KANAYA et al.

Sera, K., F. Baba, S. Goto, C. Takahashi, Y. Saitoh and M. Matsumasa (2012) Analysis of plants and sediments from the tidelands of the coastal regions of the Tohoku District following the 2011 Tsunami. International Journal of PIXE, 22: 139-147.

Suzuki, T. (2011) Environmental changes in tidal flats and influ-ences on benthic invertebrates caused by the earthquake and tsunami at eastern Japan. Journal of Japan Society on Water Environment, 34: 395-399. (in Japanese)

Urabe, J., T. Suzuki, T. Nishita and W. Makino (2013) Immediate ecological impacts of the 2011 Tohoku earthquake tsunami on intertidal flat communities. PLoS ONE, 8: e62779.

Yamochi, S. (2013) Effects of desiccation and salinity on the outbreak of a green tide of Ulva pertusa in a created salt marsh along the coast of Osaka Bay, Japan. Estuarine, Coastal and Shelf Science, 116, 21-28.

Gen KANAYA Gen KANAYA is a researcher at the National Institute for Environmental Studies (NIES). He received his Doctor of Science degree in 2005 from the Biological Institute, Graduate School of Science, Tohoku University. His main research field is tidal flat ecology, focusing on the eco-logical functioning of macrozoobenthos in soft-

bottom habitats. He currently aims to reveal ecosystem succession patterns in this lagoon after the huge disturbance event.

Hideaki MAKI Hideaki MAKI is a senior researcher at the National Institute for Environmental Studies (NIES). He got his Ph.D. in engineering from Osaka University and spent post-doc time at the Marine Biotechnology Institute. He has studied microbial degradation of crude oil in marine environments and conducted some field experi-

ments on crude oil bioremediation. Recently, he has been conducting monitoring work on water and sediment environments relevant to hypoxia in Tokyo Bay. Since the time of the Great East Japan Earthquake, he has been involved in monitoring work on PAH contamination in sediments in the Tohoku coastal sea.

Takao SUZUKI Takao SUZUKI is an assistant professor at the Graduate School of Life Sciences, Tohoku Uni-versity. He studies community ecology of benthic invertebrate animals inhabiting tidal flats, salt marshes and seagrass beds. After the Great East Japan Earthquake and massive tsunami in 2011, he has continued to monitor the recovery pro-

cesses of benthic communities along the coastal area of the Tohoku district of Japan. Now he is promoting citizen participation research methods for studying macro-benthic animals in tidal flats, which would be useful for long-term monitoring by the public.

Waka SATO-OKOSHI Waka SATO-OKOSHI is an associate professor at the Graduate School of Agricultural Science, Tohoku University. After she got her Ph.D. from Tohoku University, she worked at the Natural History Museum of Chiba as chief curator of invertebrate zoology before returning to the university as an academic staff member. Since

then, she has focused on biological and ecological studies of not only polychaetes but also various other marine benthic animals which are strongly correlated with human activities. She has had opportunities to visit international institutes such as the University of Alberta (Canada), Murdock University (Australia), Chile JICA (Chile), Universidad la República (Uruguay), and Kasetsart University (Thailand) and study as a visiting scientist. She also took part in the Japanese Antarctic Research Expedition. Her interests focus mainly on biology and ecology of polychaetes but also focus on the effect of global warming on marine ecosystems. She is currently involved in the “Environmental and faunal changes before and after the Great East Japan Earthquake” project, which started in 2011 just after the tsunami-related damage.

Eisuke KIKUCHI Eisuke KIKUCHI is a professor emeritus at Tohoku University. He has studied wetland ecosystems such as rice fields, estuarine tidal flats and small shallow lakes, focusing on nutrient cycling and food chain dynamics of these ecosystems, especially with regard to the functional roles of benthic invertebrates. His main research areas

have included Gamo Lagoon, the strongly acidic Lake Katanuma and the Ramsar site Lake Izunuma.. After his retirement from Tohoku University in 2009, he has mainly studied the successional changes in aquatic community structure in a rice field ecosystem as a cooperative researcher of the EEC, Miyagi University of Education, and has spent considerable time on the management of the Miyagi Prefectural Izunuma-Utinuma Environmental Foundation as Director General.

(Received 17 January 2014, Accepted 26 May 2014)