ORIGINAL ARTICLE

Experimental study on the deterioration and natural remediationof the Ariake Sea tidal mud caused by the sea laver treatmentacid practice and the upward seepage of pore water liquid

Yan-Jun Du Æ Song-Yu Liu Æ Shigenori Hayashi

Received: 18 July 2007 / Accepted: 11 September 2007 / Published online: 3 October 2007

� Springer-Verlag 2007

Abstract This study presents a laboratory study of the

following two aspects: (1) the influence of sea laver

treatment acid on the geoenvironmental properties of Ari-

ake Sea tidal mud, and (2) the natural remediation effect on

the sea laver treatment acid contaminated Ariake Sea tidal

mud caused by the upward seepage of pore water liquid in

the mud. Firstly, the mechanisms of the transport of sea

laver treatment acid in the Ariake Sea tidal mud and the

generation mechanisms of the upward seepage flow in the

Ariake Sea tidal mud are discussed. Secondly, a series of

one-dimensional laboratory infiltration tests were carried

out to investigate the deterioration of the Ariake Sea tidal

mud caused by the sea laver acid treatment practice. Test

results reveal that the acid treatment practice caused con-

siderable change in the geochemical properties of the mud

in terms of increase in sulfide content and decrease in pH

value. After the treatment by the sea laver treatment acid,

the sulfide content of the mud even exceeded the safe limit

value for the benthos, which represents undesirable living

condition for benthos. Thirdly, series of laboratory fresh

seawater infiltration tests for the deteriorated Iida site mud

were conducted to illustrate this natural remediation effi-

ciency. It is found that with the infiltration of the fresh

seawater, the sulfide content of the Iida site mud was

considerably reduced and pH value increased to an

acceptable range for benthos living in the tidal flat mud.

With the increase in the infiltration time and the hydraulic

gradient, the remediation efficiency could be increased.

Keywords Ariake Sea � Benthos � Deterioration �Geoenvironment � Remediation � Sea laver treatment acid �Tidal mud

Introduction

The Ariake Sea, which is located in the Kyushu region of

Japan, is one of the well-known semi-closed shallow seas

in Japan. The Ariake Sea has vast tidal flat area, which

almost covers 40% of the total tidal flat in Japan. The tidal

flat area is well known for large amount of production of

sea lavers (Porphyra spp.) and benthos such as Agemaki

shell (Sinonovacula constricta), Tairagi shell (Atrina

pectinata), and Oyster (Crassostreagigas). According to

the unpublished data from the Ministry of Agriculture,

Forestry and Fisheries of Japan, recently the annual catch

of some benthos decreased. For example, the catch of the

Agemaki shell which lives in 0–700 mm zone below mud

surface dropped from 170 tons in 1976 to practically nil by

1992 (Fig. 1). So far, the possible reasons for this observed

decrease are: (1) the man-made changes such as the Isa-

haya land reclamation project started from 1988 to 1998

(Unoki 2002; Kotama et al. 2005); (2) the frequently

occurrence of red tides in the Ariake Sea after 1998

(Tsuzumi 2003), and (3) the deteriorated environment in

the mud of the Ariake Sea. The first reason is mostly due to

the change in the tidal height and tidal velocity in the

Ariake Sea, and change in the water quality inside the

reservoir of the Isahaya reclamation project. The Isahaya

land reclamation was initially constructed for the purpose

of increasing farmlands and against flood disaster. Unoki

Y.-J. Du (&) � S.-Y. Liu

Institute of Geotechnical Engineering, Southeast University,

Si Pai Lou #2, Nanjing 210096, Jiangsu Province, China

e-mail: duyanjun@hotmail.co.jp

S. Hayashi

Institute of Lowland Technology, Saga University,

Honjo 1, Saga 840-8502, Japan

123

Environ Geol (2008) 55:889–900

DOI 10.1007/s00254-007-1040-z

(2002) reported that due to the construction of the reservoir

of the Isahaya land reclamation project, both the tidal

velocity and tidal height near the reservoir and those away

from the reservoir decreased. Unoki (2002) indicated that

the average decrease in the tidal velocity in the whole

Ariake Sea was theoretically about 2–3% due to the con-

struction of the dike of the Isahaya land reclamation

project. The reduced tidal velocity in the Ariake Sea would

result in the water density stratification and the oxygen-

depleted water in the tidal flat areas of the Ariake Bay head

(Kotama et al. 2005), which causes impact on the living

conditions of benthos inhabiting in the tidal flat muds of the

Ariake Sea. The second reason is mostly attributed to the

fact that occurrence of red tide in the Ariake Sea has been

frequently observed since 1998 as well as its scale. Tsuz-

umi (2003) proposed an index namely Index of Occurrence

Scale of Red Tide to quantify the occurrence frequency of

the red tide in the Ariake Sea. He indicated that this index

gradually increased from 1997 and reached the maximum

in the year of 2000 during the period of year of 1981–2002.

Due to the toxicity, the red tide caused considerable

damage to the benthos that inhabit in the shallow depth of

the tidal mud. Although the aforementioned two reasons

can partly explain the observed phenomenon shown in

Fig. 1, none of them can reasonably explain why the catch

of Oyster shell and Tairagi shell decreased since the

middle of 1970s and the end of 1970s, respectively.

The third reason is from the questionnaires of the

Kyushu local fishermen. Most fishermen thought that the

mud in the tidal flat areas of the Ariake Sea has been

deteriorated in terms of the appearance that unpleasant

odor from the mud was found in the Ariake Sea tidal flats

and the color of the Ariake Sea tidal flats became darker,

recently. These fishermen thought that the observed

unpleasant odor was toxic and harmful to the clamps like

Agemaki shells and Tairagi shells. These fishermen thought

that the considerable unpleasant odor is due to the sea laver

acid treatment practice, which is used by the local sea laver

farmers as the disinfectant acid to treat the sea lavers as

well as an effective way to remove the attachments. The

color of attachments is different from the sea lavers culti-

vated in the Ariake Sea. By doing so, the farmers found

that the sea laver could keep its characteristic color for a

good sale. Sea laver farmers usually dilute the sea laver

treatment acid to 1% (by volume) by the seawater before

treating the sea lavers. In Saga Prefecture, the diluted

concentration of the sea laver treatment agent is usually 1%

(by the volume of seawater). This practice has been widely

undertaken all over Japan from 1978 to present. In 1993,

large scaled acid treatment practice was officially under-

taken in Saga Prefecture although before that acid

treatment practice had been undertaken without any official

record. In the Kyushu region before 2002, the residual sea

laver treatment agents were directly dumped into the Ari-

ake Sea without any pre-treatment. During the period of

1977–2001, annually about 2,800 tons of sea laver treat-

ment acid has been dumped into the Ariake Sea water

within the sea laver farming areas (The Oceanographic

Society of Japan 2005). Since the density of the treatment

1972 1976 1980 1984 1988 1992 1996 20000

200

400

600

800

1000

a)Offically recorded sea

laver treatment practice started in Saga Prefecture

Reduction of Pcontent in sea laver

treatment acidSea laver treatment

practice started

Year

1972 1976 1980 1984 1988 1992 1996 2000

1972 1976 1980 1984 1988 1992 1996 2000

Cat

ch (

ton)

Cat

ch (

ton)

Cat

ch (

ton)

0

5000

10000

15000

20000

b)Reduction of P

content in sea laver treatment acid

Offically recorded sea laver treatment practice

started in Saga Prefecture

Sea laver treatment practice started

Year

0

2000

4000

6000

8000

10000

12000

14000

16000

c) Reduction of P content in sea laver

treatment acid

Offically recorded sea laver treatment practice

started in Saga Prefecture

Sea laver treatment practice started

Year

Isahaya landreclamation

Isahaya landreclamation

Isahaya landreclamation

Fig. 1 Change of annual catch of some clams in the Ariake Sea and

some recorded human activities: a Agemaki shell; b Oyster shell; cTairagi shell

890 Environ Geol (2008) 55:889–900

123

acid (17 kN/m3) is usually higher than that of the sea water

(10 kN/m3), considerable amount of sea laver treatment

acid would settle down to the surface of the tidal mud in

the acid treatment practice. It is thought that during the sea

laver cultivation season, the sea laver treatment acid would

transport into the mud driven by the downward seepage.

Under appropriate condition (i.e., relatively high tempera-

ture such as in spring and summer seasons), phosphorous

(P) contained in the sea laver treatment acid would enhance

the activity of the sulfate-reducing bacteria and thereby

enhance the bio-chemical reactions that generally occur in

the marine sediments. However, so far the mechanisms of

transport of the sea laver treatment acid in the Ariake Sea

tidal mud and detailed studies on how the sea laver treat-

ment practice affects the geoenvironmental properties of

the tidal mud have not received sufficient attention.

On the other hand, the field test on the Iida tidal mud of

the Ariake Sea shows that the deteriorated tidal mud is

becoming ‘‘clean’’ now in terms of decrease in the sulfide

content at a depth of 100 mm, as shown in Fig. 2. One of

the reasons for this observation is due to the controlled sea

laver treatment practice in which P content in the treatment

acid was reduced from the initial 14–5, 4, and 3% (by

weight) in the year of 2002, 2003, and 2004 to present,

respectively. Such a countermeasure limits the activity of

the sulfate reduction bacteria and the consequent interac-

tion between the tidal mud and the sea laver treatment acid.

Another possible reason may be due to that an upward

seepage of pore water liquid was observed in the Iida tidal

mud (Fig. 3), which may ‘‘wash out’’ those chemical

compounds that lead to the unpleasant odor (e.g., hydrogen

sulfide, H2S) and dark color (e.g., iron sulfide, FeS) in the

mud. In another words, the upward seepage of the pore

water liquid in the tidal mud has a natural remediation

effect. However, the mechanism of generation of the

upward seepage in the tidal mud is not clear.

The purpose of this study is to: (1) propose mechanisms

for explaining the transport of the sea laver treatment acid

in the Ariake Sea tidal mud and the generation of the

upward seepage of pore water liquid in the mud; (2)

investigate the effects of the sea laver treatment practice on

the geoenvironmental properties of the Ariake Sea tidal

mud; and (3) investigate the factors controlling the natural

remediation effect caused by the upward seepage of pore

water liquid in the mud.

Proposed mechanisms of transport of sea laver

treatment acid in the Ariake Sea tidal mud

and natural remediation effect

The downward and upward seepage of pore water liquid in

the mud may be induced by three possible mechanisms: (1)

tidal flow induced seepage in the tidal mud, as shown in

Fig. 3, (2) coupled heat-pore water vapor-pore water liquid

flow; and (3) heat-free pore gas bubble interaction induced

pore water liquid flow in the tidal mud. The first aspect has

been proved by the field test in the Iida tidal area (see

Fig. 3) and by the laboratory test (Okuzono 2006). The

reason for the different corresponding time for arriving at

the peak total head is mainly due to the presence of the free

gas bubble (including oxygen, methane, hydrogen sulfide

and carbon oxide etc.) in the pore water of the Ariake Sea

tidal mud, which usually has a concentration of about 3–

5% (by volume). Due to the presence of the free gas bubble

in the pore water, the transfer of the flood tide loading to

the tidal mud is delayed. As a result, at the flood tide, the

time for arriving at peak pore water pressure at the shallow

Oct0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

Year

Month

200720062005200420032002

Apr OctApr OctApr OctApr OctApr

)g/gm( tnetnoc edifluS

Fig. 2 Annual change in sulfide content of the Iida tidal mud at the

depth of 100 mm

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5

flood tidedownward

ebb tideupward

A: 22:37 (time), 6.28m (total head)B: 23:04 (time), 6.21m (total head)C: 23:26 (time), 6.19m (total head)

C B A

2003/11/25 09:48

2003/11/25 19:24

2003/11/26 05:00

2003/11/25 05:00

)m( daeh lato

T

Elapsed time (Year/month/date, hr:min)

0 depth 1.5m depth 3.0m depth

seepage

flood tidedownward

seepage seepage

ebb tideupwardseepage

Fig. 3 Calculated total head in the Iida tidal mud based on the

measurements of pore water pressure data

Environ Geol (2008) 55:889–900 891

123

depth is shorter than that at the deep depth, which results in

a downward seepage flow of pore water liquid. At the ebb

tide, the time for arriving at peak pore water pressure at the

shallow depth is longer than that at the deep depth, which

results in a upward seepage flow of pore water liquid. The

second and third aspects are related to the temperature

change in the tidal mud. Moqsud et al. (2006) showed that

during spring and summer, the temperature at the shallow

depth of the Iida tidal mud of the Ariake Sea is higher than

that of the deep depth, whereas opposite phenomenon was

found during autumn and winter (see Fig. 4). The tem-

perature gradient in the mud causes pore water to move in

the vapor phase from a higher temperature site to a lower

temperature site. The vapor condensates at the lower

temperature site, which increases the total head and drives

the water liquid phase from lower temperature site to the

higher temperature site (Nassar et al. 2000). Aforemen-

tioned process is titled coupled heat-pore water vapor-pore

water liquid flow, as shown in Fig. 5. Aspect (3) is due to

the temperature change in the tidal mud and the presence of

the free gas bubble in the pore water of the Iida tidal mud.

The internal pressure p of the free gas bubble is expressed

as (Terzagi 1943):

p ¼ uw þ pa þ2Ts

r0

ð1Þ

in which uw = the pore water pressure, pa = the

atmospheric pressure, Ts = the surface tension pressure

acting on the gas bubble, and r0 = the radius of the gas

bubble. Since the temperature changes in the tidal mud at

different seasons, Ts changes. At the same time, based on

the ideal gas low, the volume of the free gas bubble intends

to change when the temperature in the tidal mud changes,

as expressed by:

pV ¼ nRT ð2Þ

in which V = the volume, n = the amount of substance of

dissolved gas, R = the gas constant, and T = the temper-

ature in Kelvin. However, due to the less or more confined

pore space (i.e., confining pressure u, pa, and Ts acting on

the free gas bubble) in the tidal mud, the total pore space

volume which includes pore water liquid volume and free

gas bubble volume hardly changes instantaneously (Xu

and Ruppel 1999; Xu and Germanovich 2006). To satisfy

Eq. (2), p should increase when temperature increases

during spring and summer seasons while p decreases

during autumn and winter seasons. Since Ts changes

opposite to the change in temperature and pa is assumed

almost constant, uw changes resulting in the generation of

excess pore water pressure. The development of the excess

pore water pressure is more significant in the deep tidal

mud than that in the shallow depth where confining

pressure is less and gas emission to the atmosphere fre-

quently occurs. As a result, a difference between the

excess pore water pressure at the shallow depth and deep

depth exists, which generates seepage flow of pore water

liquid. During the spring and summer seasons, positive

excess pore water pressure at the deep depth is higher than

that at the shallow depth due to the relatively high tem-

perature at the shallow depth. Consequently, an upward

fluid seepage of pore water fluid develops. During the

autumn and winter seasons, an opposite seepage flow

(downward seepage flow) develops due to the relatively

high negative excess pore water pressure in the shallow

depth of tidal mud (Fig. 6). On the other hand, due to the

emission of the free gas bubble to the atmosphere at the

surface of the tidal mud, the amount of free gas bubble

will be less than that at the deep depth. In another words, a

concentration gradient of free gas bubble exists in the tidal

mud. Therefore, the free gas bubble will transport upward

mainly driven by diffusion.

For simplicity, in this study, it is simply assumed that

the net pore water liquid seepage in the tidal mud is the

sum of the aforementioned three mechanisms. As a result,

during the spring and summer seasons, upward seepage

will be generated at the ebb tide. At the flood tide, aspect

(1) will induce downward seepage while aspects (2) and (3)

will induce upward seepage. The net seepage flow will

depend on the sum of the downward seepage and upward

seepage. During the autumn and winter seasons, downward

seepage will be generated at the flood tide. At the ebb tide,

aspect (1) will induce upward seepage while aspects (2)

and (3) will induce downward seepage. The net seepage

flow will depend on the sum of the downward seepage and

the upward seepage. During the sea laver treatment prac-

tice at the later December to the next February, the sea

laver treatment acid, which has settled down to the surface

Temperature

htpeD

Summer Winter

Spring Autumn

Fig. 4 Measured temperature distribution in the Iida tidal mud during

the four seasons (after Moqsud et al. 2006)

892 Environ Geol (2008) 55:889–900

123

of the tidal mud due to the higher density than that of the

seawater, would transport down in the tidal mud driven by

the downward seepage of pore water liquid at the flood

tide. The interaction between the migrated acid and the

mud may not be significant because that the low temper-

ature limits the sulfate reduction bacteria (Cook and Kelly

1992) and thereby the generation of the H2S and FeS

(Hayashi et al. 2003). In the spring and summer seasons,

when the temperature increases, considerable amount of

chemical compounds like H2S and FeS are generated due

to the enhanced activity of the sulfate reduction bacteria in

the tidal mud (Richard and Morse 2005; Hayashi and Du

2005). At the ebb tide, these chemical compounds would

be washed out of the tidal mud due to the upward pore

water liquid seepage and consequently would induce a

natural remediation effect on the mud. As a result, the

measured content of H2S decreases recently (see Fig. 2). In

this study, only the effect of the aforementioned aspect (1)

related pore water liquid seepage on the transport of the sea

laver treatment acid in the mud and the natural remediation

effect will be presented in the later part. A series of current

laboratory tests are undertaken to investigate the aspects

(2) and (3) related pore water liquid seepage.

Materials and test method

The soils used for the laboratory tests were sampled from

the Higashiyoka site tidal flat area, Ariake Sea (See Fig. 7).

The basic physico-chemical properties of the sampled soils

are tabulated in Table 1. Two types of laboratory tests were

performed: (a) Type-1, infiltration of sea laver treatment

acid together with seawater; and (b) Type-2, infiltration of

fresh seawater. Type-1 test represents the field scenario that

sea laver treatment acid transports in the tidal mud driven

by the downward seepage occurring at the flood tide during

the sea laver cultivation season. It is noted that although the

real tidal mud temperature in the field during the sea laver

cultivation (7–16�C) is lower than that of the laboratory

test (25�C), the laboratory test represents an acceleration

condition in which the tidal temperature is elevated to

enhance the activity of the sulfate reduction bacteria and

thereby enhance the deterioration of the mud. With the

elevated temperature, it was both time-effective and cost-

effective. Type-2 test represents the field scenario that the

upward seepage of pore water liquid occurs in the tidal

mud at the ebb tide during the cultivation season. For

Type-1 test, about 6 kg of mud under the field water

temperature

High temperature

Low total headEvaporation

CondensationLow

Low temperature

Hightemperature

Spring ~ Summer16°C - 29°C

Autumn ~ Winter16°C - 7°C

Hot water Cold water

High total head

High totalhead

Condensation

EvaporationLow total

head

Temperature

Summer Winter

Spring Autumn

Sun Sun

Mud

dep

th

Mud

dep

th

Pore

wat

er v

apor

Pore

wat

er li

qurd

Hea

t

Pore

wat

er v

apor

Pore

wat

er li

qurd

Hea

t

Fig. 5 Proposed concept of

coupled heat-pore water vapor

and coupled heat-pore water

liquid flow in the tidal mud

free

gas

free

gas

Low positive excess pore

water pressure

High volumetricswell tendency,

Low content

Low volumetricswell tendency,High content

Lowtemperature

High temperature

Hightemperature

Low temperature

Spring ~ Summer16°C -29°C

Autumn ~ Winter16°C -7°C

Hot water Cold water

High positive excess pore

water pressure

Low negative excess pore

water pressure

High volumetricshrink tendency,

Low content

Low volumetricshrink tendency,

High content

High negative excess pore

water pressure

Sun Sun

Temperature

Summer Winter

Spring Autumn

Mud

dep

th

Mud

dep

th

Pore

wat

er li

qurd

Pore

wat

er li

qurd

Hea

t

Hea

t

Fig. 6 Proposed concept of

heat-free pore gas interaction

induced pore water liquid flow

in the tidal mud

Environ Geol (2008) 55:889–900 893

123

content condition was thoroughly mixed with the seawater

taken near the estuary of the Rokkaku River until the water

content reached 2 times the liquid limit. The initial pH and

sodium (Na+) concentration of the sampled seawater were

7.6 and 25 g/L, respectively (Table 2). Choosing the Hig-

ashiyoka tidal mud is mainly because that the Higashiyoka

tidal mud was thought to be less affected by the sea laver

treatment practice, which is suitable for the tests. The test

was performed in a test apparatus shown in Fig. 8. The

slurry was poured into test apparatus together with the

sampled seawater. The height of the soil-water suspension

was 900 mm. After that the soil-water suspension was

allowed for settling undisturbed until self-weight consoli-

dation finished (about 5 days). Four parallel tests were

conducted over a period of 30 days. In one test (labeled

B0), the seawater was continuously drained from an acrylic

container (see Fig. 8) and every day the sea water level was

refilled until the water level reached 900 mm (from the

base of the soil specimen). This process was repeated for

30 days. At the completion of the test, the water above the

mud was drained away and the sample was extruded and

sliced for determining salt concentration, pH and sulfide

content. pH value was measured using a portable pH meter

HORIBA D-52. Sulfide content was measured using an

apparatus titled GASTEC 201 L/H (Wu et al. 2003, here-

after called GASTEC method). The reason for choosing the

GASTEC 201 L/H for measuring the soil sulfide content is

mainly because this method is both time-effective and cost-

effective in the case of dealing with large amount of marine

sediment samplings as compared with the traditional ana-

lytical method (Montani 2002). Moreover, Montani (2002)

indicated that in most cases, the measured sulfide contents

of marine sediments using the GASTEC method are well

consistent with the values measured using the analytical

method suggested by Berner (1964). Sulfide content mea-

sured by this method consists of contents of hydrogen

sulfide (H2S), hydrosulfide ion (HS–), ferrous sulfide (FeS),

and sulfur ion (S2–) contained in the mud. For these tests,

B0 means the concentration of the sea laver treatment acid

was 0%. Another test (labeled B0.1), was carried out with

refilling the seawater mixed with the sea laver treatment

acid with the concentration of 0.1% (by volume of sea-

water) until the water level reached 900 mm (from the base

of the soil specimen). The sea laver treatment acid used in

the laboratory test has the same composition with that used

in the field of the Saga Prefecture (before diluted up to

1%). It has a pH value of about 2.0, density of 1.7 g/cm3

and phosphate ion (PO43–) concentration of 18 g/L. In

addition to 0.1 % sea laver treatment acid, similar tests

were conducted at 0.01 and 0.03% (by volume of seawa-

ter). All of the tests aforementioned were performed at the

temperature of 25�C.

The soils used for Type-2 test were sampled at Iida tidal

flat areas of the Ariake Sea. The basic properties are tab-

ulated in Table 1. The sampled soils were mixed with the

sea laver treatment acid. The concentration of the sea laver

acid was controlled at 0.2% (by volume of the soil pore

water). The sea laver treatment acid used for Type-2 test is

the same with that used for Type-1 test, and its composi-

tions are tabulated in Table 2. The soils were placed into

the cylinders, which have the same shape and size with that

used for Type-1 tests except that there were some circular

openings with a diameter of 7 mm and interval of 2 mm on

the surface of the cylinder. These openings were used for

periodical mud sampling during the infiltration test. The

soils were allowed for curing undisturbed under the con-

ditions of 25�C and light shielding by covering black sheets

on the cylinders. The curing took 9 days until the sulfide

Fig. 7 A map showing the soil sampling locations

Table 1 Physico-chemical properties of the soils used in this study

Parameter Iida mud Higashiyoka mud

Specific gravity, Gs 2.69 2.71

Water content (%) 235 168

Liquid limit, wL (%) 158 140

Plasticity index (%) 101 90

Clay particle (%) 60 45

pH 6.9 8.0

Salt content (mg/L) 20 15

Sulfide content (mg/g) 0.81 0.09

894 Environ Geol (2008) 55:889–900

123

content reached practically constant in the soils. It was

observed that gravity consolidation of the soil finished at

4 days after pouring into the cylinder. The sea laver

treatment acid used for the test is the same with that used

for Type-1 test. The Ariake Sea water taken from the

estuary of the Rokkaku River was then introduced to the

cylinders in a careful manner until the downward hydraulic

gradients reached 1 and 5, respectively. The basic chemical

properties of the seawater are summarized in Table 2.

Throughout the test, the volume of the infiltrated seawater

from the soil base was measured periodically and the same

amount of the fresh seawater was introduced above the

soils to maintain the constant hydraulic gradient. During

the test, a syringe was inserted into the openings of the

cylinder to take soil samples from different depth for the

regular measurement of pH and sulfide content. The infil-

tration test was performed at the temperature of 25�C.

Results and discussion

Type-1 test

The measured sulfide content along the soil depth is plotted

in Fig. 9. It can be seen that the sulfide content in the case

of the Higashiyoka mud experienced acid treatment (B0.1

test) was significantly higher than the original low values

(B0 test), especially at the depth of 0–40 mm. The sulfide

content at this depth range even exceeded the safe limit

value of 0.2 mg/g, as required by the Japan Fishery Water

Quality Standard (Japan Fisheries Resource Conservation

Association, 2000). For B0 test, which represents an field

scenario that mud was not contaminated by the sea laver

acid treatment practice, the sulfide content is much lower at

the depth of 0–40 mm, as compared with B0.1 test. At the

depth of 40–120 mm, the sulfide content did not change

considerably in the case of B0.1 test, i.e., practically

constant.

Table 2 Chemical properties of seawater and sea laver treatment

acid used for both Type-1 and Type-2 tests

Parameter Seawater Sea laver treatment acid

Density (g/cm3) 1.0 1.7

pH value 7.6 2.0

Na+ 25.1 0.3

K+ 0.9 0.03

Ca2+ 1.1 ND

Mg2+ 2.5 ND

Cl– 35.8 ND

SO42– 3.3 171.6

PO43– ND 124.4

SO42– 3.3 171.6

PO43– ND 124.4

drain

002

007

200Unit: mm

Mud

Seawater 4-tie rods, @90°

acrylic cylinder

porous plate

acrylic base container

flange

4-tie rods, @90°

flange

Fig. 8 Schematic of laboratory test apparatus used for this study

120

100

80

60

40

20

00.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2

Safe limit

Sulfide content (mg/g)

B0 test

B0.1

test

Dep

th (

mm

)

Fig. 9 Change of sulfide content versus soil depth for Type-1 test

Environ Geol (2008) 55:889–900 895

123

Change in pH along soil depth was plotted in Fig. 10.

After treated by the sea laver treatment acid, soil pH

dropped from averagely 7.8 (B0 test) to a range of 6.6–7.3

(B0.1 test). Based on the Japan Fisheries Resource Con-

servation Association (2000), a pH value in the range of

7.8–8.4 is suitable for benthos living in the tidal flat mud.

Therefore, after treatment of sea laver treatment acid (B0.1

test), mud had an undesirable condition for inhabitation of

benthos.

Figure 11 shows the change in salt content along soil

depth. Salt content of the soil experienced B0 test is in the

range of 14–23 g/L while salt content of soil experienced

B0.1 test is lower with a range of 11–13 g/L. Based on the

Japan Fisheries Resource Conservation Association (2000),

a salt content value in the range of 20–25 g/L is suitable for

benthos like clamp shells living in the tidal flat mud. For

the salt content of the mud less than 20–25 g/L, lower salt

content represents worse inhabitation condition for ben-

thos. Therefore, after treatment of sea laver treatment acid

(B0.1 test), mud had an undesirable condition for benthos’

inhabitation.

Based on the laboratory infiltration tests, the increase in

the sulfide content with the sea laver treatment acid con-

centration is depicted in Fig. 12. It can be seen that with

the increase in the concentration of the sea laver treatment

acid, the sulfide content of the mud at almost depth less

than 40 mm increased considerably while the sulfide con-

tent at the depth of 40 mm increased marginally. When the

sea laver treatment acid concentration was higher than

0.025% (for the case of 1 mm depth) or higher than

0.012% (for the case of 40 mm), sulfide content was higher

than the safe limit value (0.2 mg/g), which indicates a

serous undesirable condition.

The increased value of sulfide content with the increase

in the concentration of the sea laver treatment agent may be

explained by the geochemical reaction occurring in the

marine sediments. Field monitoring test results indicate

that both Iida mud and Higashiyoka tidal mud are under-

gone reduction condition (Hayashi et al. 2003). Under the

120

100

80

60

40

20

06.0

7.8 - 8.4

pH

B0 test

B0.1

test

Dep

th (

mm

)

6.5 7.0 7.5 8.0 8.5 9.0

Fig. 10 Change of pH versus soil depth for Type-1 tests

120

100

80

60

40

20

010 12 14 16 20 22 24 26

Salt concentration (g/L)

B0 test

B0.1

test

Dep

th (

mm

)Fig. 11 Change of salt content versus soil depth for Type-1 tests

0.00 0.02 0.04 0.06 0.08 0.100.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

2.2

Depth of 1 mm

Depth of 5 mm

Depth of 15 mm

Depth of 40 mm

Sulf

ide

cont

ent (

mg/

g)

Concentration of sea laver treatment acid (%)

Safe limit

Fig. 12 Change of sulfide content with the concentration of sea laver

treatment acid for Type-1 tests

896 Environ Geol (2008) 55:889–900

123

reduction condition, the sulfate (SO42–) in the pore water of

the mud is reduced to hydrogen sulfide (H2S) by the

organic matter deposition with the aid of sulfate-reducing

bacteria (Jorgenson 1991; Mitchell 1993). Simultaneously

the H2S reacts with soluble ferrous iron (Fe2+), which is

usually observed in the Ariake marine muds (Ohtsubo et al.

1995) and forms black amorphous FeS. The reduction

process of SO42– to H2S and formation of FeS can be

expressed by following equations.

SO2�4 þ CH3COO� þ Hþ ! H2Sþ CO2 þ H2O ð3Þ

H2Sþ Fe2þ ! FeS ð4Þ

in which CH3COO– symbolizes the organic matter. Usually

oxidation condition was observed in the shallow depth up

to 5 cm of the Ariake mud (Hayashi et al. 2003). Under this

condition, with the aid of the sulfur bacteria, H2S is

oxidized to SO42–, as expressed in Eq. (5):

H2Sþ H2O! SO2�4 þ Hþ ð5Þ

Jorgenson (1991) and Mitchell (1993) indicated that

aforementioned geochemistry reaction is typically occurred

in marine sediments showing the cycle of sulfur (S) in the

marine sediments. However, during the sea laver treatment

practice undertaken in the Ariake Sea, the reaction equi-

librium of Eqs. (3) and (4) might have been broken which

lead to significant occurrence of H2S in the tidal mud of

Ariake Sea. Since the phosphorous (P) contained in the sea

laver treatment agent (in the form of inorganic chemical

compound, NaH2PO4) provides source of feed to the sul-

fate-reducing bacteria, the activity of the sulfate-reducing

bacteria would become higher under relatively high tem-

perature ([18�C) and reduction condition (McGhee 1991).

Therefore, the decomposition of organic matter initially

contained in the mud and the additional organic matter like

DL-Malic acid from the sea laver treatment agent would be

enhanced. Consequently, according to Eqs. (1) and (2),

considerable amount of H2S and FeS would form. As a

result, the sulfide content of the mud increased. The

occurrence of black FeS was observed in the B0.1 test that a

blackish matter (identified as FeS) appeared in the water

above the mud after 10 days of test. At the end of test, a

thin layer of blackish fine mass (identified as FeS) was

observed on the surface of the mud.

The decrease in pH and increase in sulfide content as a

result of the sea laver acid treatment practice seems to have

also occurred in the Iida tidal flat areas where the non-

officially recorded local acid treatment practice before

1993 and the large scaled acid treatment practice after 1993

have been encountered. Such a result is reasonable for

explaining the phenomenon that the catch of the clamp

shells living in the shallow depths of the mud decreased

during the period of the sea laver treatment practice, as

depicted in Fig. 1.

Type-2 test

Figure 13 shows the change of sulfide content along the

soil depth before and after the fresh seawater infiltration

test. Compared with the original value, after infiltration of

fresh seawater, sulfide content decreased, indicating that

the infiltration of fresh seawater had a remediation effect

on the sea laver treatment acid contaminated Iida site tidal

mud. The decrease in sulfide content became significant

with the increase in testing time. This is mainly because

that with the increase in test duration, larger amount of

oxygen (O2) dissolved in the seawater might have been

consumed by the mud, which lead to the oxidation of H2S

as expressed by Eq. (5). As a result, sulfide content

decreased with the increase in time duration. Compared

with the case of the hydraulic gradient i = 1 (see Fig. 13a),

the case of the hydraulic gradient i = 5 (see Fig. 13b)

resulted in a better remediation efficiency in terms of

decrease in the sulfide content in the soil at the same testing

duration. The reason that the sulfide content is still higher

than the safe limit, 0.2 mg/g is mainly due to the limited

short test duration.

The change in pH value along the soil depth before and

after the test was shown in Fig. 14. Compared with the

original value, pH of soil increased after test. It increased

with the increase in testing duration. After 30 days, pH

value exceeds 7.8, which is in the acceptable range for

benthos living in the tidal mud (pH = 7.8–8.4). However, it

was found that the effect of hydraulic gradient on change in

pH was not considerable.

To evaluate the remediation efficiency, the percentage

of sulfide content reduction, A%, is used in this study as

expressed by following:

A% ¼ SC0 � SCðtÞSC0

� 100 ð6Þ

in which SC0 = the original sulfide content and SC(t) = the

sulfide content at specified time. A high value of A%

represents that sulfide content at specified time is low,

indicating that the remediation efficiency is high. A nega-

tive value of A% means that after the infiltration of sea

water, sulfide content became higher than the original

value.

The variations of A% along the mud depth for both the

hydraulic gradient i = 1 and the hydraulic gradient i = 5

are shown in Fig. 15. For both cases (i = 1 and i = 5), at

the shallow depth, soils have higher A% value than the

deep depth, indicating that the remediation effect at the

Environ Geol (2008) 55:889–900 897

123

shallow depth is better than that of the deep depth. At some

depths, the value of A% even became negative. The neg-

ative value at the deep depth may be due to that with the

downward seepage of seawater, the FeS contained in the

deteriorated mud might have moved from shallow depth to

the deep depth mainly driven by the downward seepage. As

a result, the measured sulfide content at some deep depth

became higher than the original value (see Fig. 13) and A%

value at some deep depth became negative (see Fig. 15).

A% value at the shallow depth is higher than that at the

great depth, i.e., a gradient exists along the soil depth. This

may be because that with the downward seepage of sea-

water, FeS contained in the deteriorated mud at the shallow

depth was washed down from the shallow depth to the

great depth driven by the seepage pressure. As a result, the

sulfide content at the shallow depth was less than that at the

great depth.

With the increase in hydraulic gradient, A% value

increased. For the case of the hydraulic gradient i = 1, A%

varied in a range of 10–50%. For the case of i = 5, A%

varied in a range of 5–90%, indicating a higher remediation

effect. This may be due to that with the downward seepage

of seawater, the FeS contained in the deteriorated mud at

the shallow depth was washed down driven by the seepage

160

140

120

100

80

60

40

20

00.0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4

0.0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4

a)

i = 1

Sulfide content (mg/g)

Sulfide content (mg/g)

)m

m( htpeD

Original after 5 days after 10 days after 15 days after 20 days after 30 days

160

140

120

100

80

60

40

20

0b)

i = 5

Original after 5 days after 10 days after 15 days after 20 days after 30 days

)mc( htpe

D

Fig. 13 a Change of sulfide content versus soil depth for the case of

i = 1 in Type-2 test. b Change of sulfide content versus soil depth for

the case of i = 5 in Type-2 test

160

140

120

100

80

60

40

20

06.5

a)

i = 1

pH value

)m

m( htpeD

)m

m( htpeD

Original after 5 days after 10 days after 15 days after 20 days after 30 days

160

140

120

100

80

60

40

20

0b)

i = 5

pH

Original after 5 days after 10 days after 15 days after 20 days after 30 days

7.0 7.5 8.0 8.5

6.5 7.0 7.5 8.0 8.5

Fig. 14 a Change of pH versus soil depth for the case of i = 1 in

Type-2 test. b Change of pH versus soil depth for the case of i = 5 in

Type-2 test

898 Environ Geol (2008) 55:889–900

123

pressure, icwH, in which cw = the density of seawater and

H = the thickness of soil. Since H was almost constant

(approximately 180 mm in this study), the seepage pres-

sure for the case of i = 5 is five times higher than i = 1.

Thereby, higher amount of FeS was washed out from the

soil. As a result, value of A% for the case of i = 5 is higher

than that for the case of i = 1.

The relationship between pH and sulfide content for

both cases i = 1 and i = 5 is plotted in Fig. 16a and b,

respectively. Generally, the locations of the data measured

after treatment by the sea laver treatment acid are away

from the original locations. With the increase in pH, soil

sulfide content decreased. The co-relationship coefficients,

R, were 0.38 and 0.66 for the cases of i = 1 and i = 5,

respectively, indicating that pH was one of the factors

controlling the soil sulfide content. It was observed that five

data were far below others in the case of i = 5 (see

Fig. 16b), while such a phenomenon was not observed in

the case of i = 1 (see Fig. 16a). This is mainly because that

these five data are mainly from the shallow depth of the soil

(10 cm depth). As discussed in the earlier part, under the

condition of i = 1, at the shallow depth, larger amount of

140

160

120

100

80

60

40

20

0

a)

i =1

Variation in sulfide content, A% )

mm( htpe

D

160

140

120

100

80

60

40

20

0-100 -80 -60 -40 -20 -0 20 40 60 80 100

-100 -80 -60 -40 -20 -0 20 40 60 80 100

b)

i = 5

Variation in sulfide content, A%

)m

m( htpeD

after 5 days after 10 days after 15 days after 20 days after 30 days

after 5 days after 10 days after 15 days after 20 days after 30 days

Fig. 15 a Change of A% versus soil depth for the case of i = 1 in

Type-2 test. b Change of A% versus soil depth for the case of i = 5 in

Type-2 test

0.0

0.3

0.6

0.9

1.2

1.5

1.8

2.1

2.4

y = -0.37x+4.25R = 0.38

)g/gm( tnetnoc edifluS

)g /gm( tnetnoc edifluS

pH

Original after 5 days after 10 days after 15 days after 20 days after 30 days

0.0

0.3

0.6

0.9

1.2

1.5

1.8

2.1

2.4

y = -0.51x+5.12R = 0.66

pH

Original after 5 days after 10 days after 15 days after 20 days after 30 days

6.5 7.0 7.5 8.0 8.5

6.5 7.0 7.5 8.0 8.5

a)

b)

Fig. 16 a Relationship between pH and sulfide content for the case of

i = 1. b Relationship between pH and sulfide content for the case of

i = 5

Environ Geol (2008) 55:889–900 899

123

FeS initially contained in the soil might had been washed

down to the great depth. However, change of pH at the

shallow depth was not sensitive to the hydraulic gradient

(see Fig. 14). Therefore, the measured sulfide contents at

this depth are much lower than others (see Fig. 13b). As a

result, five data measured from the shallow soil depth were

below others and the trend line in Fig. 16b.

Conclusions

This study presents three proposed mechanisms for

explaining the transient seepage of pore water liquid of the

tidal mud, which contributes to the transport of sea laver

treatment acid in the Ariake Sea tidal mud and natural

remediation of the sea laver treatment acid contaminated

tidal mud. The mechanisms of the generation of the pore

water liquid seepage in the tidal mud are discussed.

The Type-1 test results show that for the sea laver

treatment acid treated Higashiyoka mud, with the increase

in the concentration of the sea laver treatment acid, the

sulfide content increased whereas the pH value decreased.

The mechanism of the increase in the sulfide content

caused by the sea laver treatment acid is explained based

on the enhanced geochemical reactions occurring in the

Ariake Sea tidal mud.

The Type-2 test results show that with the infiltration of

the fresh seawater in the sea laver acid contaminated mud,

the sulfide content decreased and pH values decreased,

indicating that the upward seepage of pore water liquid in

the mud has natural remediation effect.

The Type-2 test results show that with the increase in

hydraulic gradient, A% increased indicating that the reme-

diation efficiency increased. It is found that A% at the

shallow depth of the soil is higher than that at the great depth.

Acknowledgments This study is part of a big research grant titled

‘‘Proof Test of Sediment Improvement by Sand Filling and Mixing in

Enclosed Dikes and Evaluation’’ commissioned by the Japanese

Ministry of Education, Culture, Sports and Society and Technology

Agency (JST). The partial financial support from the International

Cooperation project titled ‘‘Recovery of Benthos in Ariake Sea Tidal

Mud’’ (No. 8621002036) between Southeast University, China and

Saga University, Japan, is appreciated. The former graduate students,

Mr. Yuji Ushihara, Mr. Kazuya Nakatake, Mr. Kengo Okuzono,

and Mr. Sei Tanaka, are thanked for their help with performing the

laboratory tests.

References

Berner RA (1964) Distribution and diagenesis of sulfur in some

sediments from the Gulf of California. Mar Geol 1(2):117–140

Cook RB, Kelly CA (1992) Sulphur cycling and flues in temperature

dimictic lakes. Sulphur cycling on the continents. Wiley, New

York

Hayashi S, Du YJ (2005) Effect of acid treatment agent of sea laver

on geoenvironmental properties of tidal flat muds in the Ariake

Sea. J ASTM Int 3(7):52–59

Hayashi S, Du YJ, Ushihara Y (2003) Investigation on effect of acid

treatment agent on the change in geoenvironmental properties of

sediment in the Ariake Sea. In: Proc 1st China–Japan Joint

Symposium on Geotechnical Engineering, Beijing, pp 264–271

Japan Fisheries Resource Conservation Association (2000) Fishery

water quality standard (in Japanese). Kouseisha-kouseikaku

Press Co., Tokyo

Jorgensen BB (1991) A thiosulfate shunt in the sulfur cycle of marine

sediments. Science 249(1):152–154

Kotama M, Tanaka K, Fujita T, Kimoto K, Okamura K, Mori Y

(2005) Variation of oxygen-depleted water and environmental

properties of bottom sediments in the western region of Ariake

Bay head (in Japanese). In: Proceedings of annual spring

conference of the oceanographic society of Japan, p 191

McGhee TJ (1991) Water supply and sewerage, 6th edn McGraw-Hill

Book Co., Singapore

Mitchell JK (1993) Fundamentals of soil behavior, 2nd edn. Wiley,

New York

Montani T (2002) Manual of measurement of sulfide content using

simple method (in Japanese). Japan Brine Water Fishing

Association, Tokyo

Moqsud MA, Hayashi S, Du YJ, Suetsugu D (2006) Evaluation of

Temperature trend in contaminated tidal flat in the Ariake Sea,

Japan. Am J Environ Sci 2(3):104–108

Nassar IN, Horton R, Flerchinger GN (2000) Simultaneous heat and

mass transfer in soil columns exposed to freezing/thawing

conditions. Soil Sci 165(3):208–216

Ohtsubo M, Egashira K, Kashima K (1995) Depositional and post-

depositional geochemistry, and its correlation with geotechnical

properties of the marine clays in Ariake Bay, Japan. Geotech-

nique 45(3):263–282

Okuzono K (2006) A laboratory test study of pore water seepage flow

in the Ariake Sea tidal mud. Master Thesis, Saga University,

Saga, Japan

Terzagi K (1943) Theoretical soil mechanics. Wiley, New York

The Oceanographic Society of Japan (2005) Recovery of the Ariake

Sea eco-system (in Japanese). Kouseisha-kouseikaku Press Co.,

Tokyo

Tsuzumi H (2003) The mechanisms of occurrence of large-scaled red

tide in the Ariake Sea where eutrophication process is not

significantly processed (in Japanese). Report to the General

Assessment Committee of Ariake Sea and Yachiyo Sea,

Japanese Ministry of the Environment

Unoki S (2002) Change in the tide amplitude and tidal velocity of

Ariake Sea caused by the Isahaya reclamation project (in

Japanese). Umi-to-Sora 78(1):19–30

Wu SS, Tsunami H, Tsukamoto K, Kogure K, Ohwada K, Wada M

(2003) Visualization of the respiring bacteria in sediments

inhabited by Capitella sp. 1. Fisheries Res 69(1):170–175

Xu W, Ruppel C (1999) Predicting the occurrence, distribution and

evolution of methane gas hydrates in porous sediments. J

Geophys Res 104:5081–5096

Xu W, Germanovich LN (2006) Excess pore pressure resulting

from methane hydrate dissociation in marine sediments: A

theoretical approach. J Geophys Res 111 (B01104). doi:10.1029/

2004JB003600

900 Environ Geol (2008) 55:889–900

123