Impacts of sea cucumber farming on biogeochemical ...cwr.bnu.edu.cn/userfiles/files/1-s2_0-S1474706516302613...2012). The rapid growth of Apostichopus japonicus farming pro-moted a

Page 1

lable at ScienceDirect

Physics and Chemistry of the Earth xxx (2017) 1e12

Contents lists avai

Physics and Chemistry of the Earth

journal homepage: www.elsevier .com/locate/pce

Impacts of sea cucumber farming on biogeochemical characteristics inthe Yellow River estuary, Northern China

Jing Fu a, Hisashi Yokoyama b, Baoshan Cui c, Jin Zhou d, Jiaguo Yan c, Xu Ma c,Shozo Shibata a, *

a Graduate School of Global Environmental Studies, Kyoto University, Kyoto 6068501, Japanb Field Science Education and Research Center, Kyoto University, Kyoto 6068502, Japanc State Key Laboratory of Water Environment Simulation, School of Environment, Beijing Normal University, Beijing 100875, PR Chinad East China Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Shanghai 200090, PR China

a r t i c l e i n f o

Article history:Received 20 October 2016Received in revised form8 December 2016Accepted 16 December 2016Available online xxx

Keywords:Sea cucumber farmingApostichopus japonicusEnvironmental impactAquaculture wasteStable isotopeYellow River estuary

* Corresponding author. Present address: Graduatmental Studies, Kyoto University, Yoshida-HonmachiJapan.

E-mail addresses: [email protected] (S. ShibFu).

http://dx.doi.org/10.1016/j.pce.2016.12.0061474-7065/© 2016 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Fu, J., et aNorthern China, Physics and Chemistry of th

a b s t r a c t

To investigate the potential environmental effects of pond farming for Apostichopus japonicas in YellowRiver estuary, we examined discrepancies of distance-based typical pollution indicators (TOC, TN, NO3

�,NH4

þ, NO2� and PO4

3�) and biochemical tracers (d13C and d15N) in water column and sediment, as well asdietary characteristics of dominant macrobenthos between farming and non-farming areas. The resultsrevealed that studied variables in water column showed no uniform spatial differences. Meanwhile,those in sediment displayed similar decrease tendencies from farming pond to the adjacent tidal flat,which was considered to represent the environmental effects of farming. Biochemical tracers (d13C andd15N) in both water column and sediment confirmed the origin of organic matters from the aquaculturewaste. The detectable dispersion distance of aquaculture waste was restricted to an area within 50 mdistance as determined by most variables in sediment (TOC, TN, NO3

� and NH4þ), particularly by C:N ratio

and d13C with which origins of the wastes were traced. Bayesian mixing models indicated that in thefarming area BMA had a larger contribution, while POM(marine) showed a smaller contribution to the dietsof Helice tridens and Macrophthalmus abbreviates compared to those in the non-farming area. The overallresults showed that pond farming for Apostichopus japonicus in the Yellow River estuary altered the localenvironment to a certain extent. For methodological consideration, sediment biogeochemical charac-teristics as a historical recorder much more effectively reflected aquaculture waste accumulation, andstable isotope approaches are efficient in tracing the origin and extent of various allogenous sources.

© 2016 Elsevier Ltd. All rights reserved.

1. Introduction

Coastal aquaculture has rapidly grown in recent years with anincreasing demand for seafood production (Naylor et al., 2000;Tanner and Fernandes, 2010). The expansion of coastal aquacul-ture, however, has raised widespread concerns about environ-mental costs and sustainable aquaculture techniques (Puscedduet al., 2007; Tanner and Fernandes, 2010; Troell et al., 2003).

Aquaculture farming releases large amounts of dissolved andparticulate nutrients originating from excretion process of reared

e School of Global Environ-, Sakyo-ku, Kyoto, 606-8501,

ata), [email protected] (J.

l., Impacts of sea cucumber fae Earth (2017), http://dx.doi

organisms and uneaten feed and fecal materials (Holby and Hall,1991; Lupatsch and Kissil, 1998). The redundant particulateorganic carbon and nitrogen loads result in organic enrichment andsediment biogeochemical characteristics alternation (Brooks andMahnken, 2003; Holby and Hall, 1991; Sar�a et al., 2004). Theexcess dissolved inorganic nutrients that easily dispersed in thewater and deposited in the sediment are the most concernedcontaminant contributed to coastal eutrophication (Paerl, 2006;Smith et al., 1999; Wu et al., 2014).

In recent years, sea cucumber farming has increased tremen-dously encouraged by the strong and growing market demand(Chen, 2003; FAO, 2008; Han et al., 2016) for its huge profit andhigh nutritional and pharmaceutical values (Bordbar et al., 2011).Consequently, much attention had been recently focused on itsculturing manner for better production and development (Donget al., 2010; Liu et al., 2010; Paltzat et al., 2008; Yokoyama, 2013).

rming on biogeochemical characteristics in the Yellow River estuary,.org/10.1016/j.pce.2016.12.006

Page 2

J. Fu et al. / Physics and Chemistry of the Earth xxx (2017) 1e122

Although a variety of studied related to sea cucumber industry,such as hatchery production, culturing manner, sociological issuesof stocking were conducted in previous studies (Han et al., 2016;Ren et al., 2012), there existed a relative paucity of details con-cerning the environmental effects of sea cucumber farming activ-ities (Li et al., 2014; Ren et al., 2010; Zheng et al., 2009). One of themost important reason probably lies in that most sea cucumbersare deposit feeders. They are always considered as the “environ-mental cleaners or scavengers” (Yang et al., 1999) and therefore noexogenous nutrients were introduced. Actually, the exogenousnutrients input mainly depend on culturing manners. In most areasof China, commercial feeds which mainly composed of fish mealand soybean meal (accounting for 15% and 30% of total weight,respectively) were commonly used in sea cucumber ponds. Theexcessive feeds always make a large amount of dissolved and par-ticulate waste which result in nutrients enrichment of water andsediment in the pond and adjacent environment.

Sea cucumber farming in China increased tremendously alongcoasts, and the yield reached 205,791 tons (dry weight) in 2015(China Fisheries Yearbook, 2016). For the present study area YellowRiver estuary, sea cucumber production exponentially increased inthe past ten years. The farming ponds area increased from 21.3 ha in2003 to 21,734 ha in 2014 with an annual production of 11,241 tons(dry weight) annually. Among the sea cucumber species cultured(nearly ten species) around the world at present, Apostichopusjaponicus was the largest species in production, which was mainlyartificially cultured in China (Han et al., 2016; Purcell et al., 2012).The annual hatchery production of juvenile of this species wasapproximately 6 billion in China, and the sum exceeded the totalproduction of all other reared sea cucumber species (Purcell et al.,2012). The rapid growth of Apostichopus japonicus farming pro-moted a number of studies focusing on genetics (Li et al., 2009),energetic (Liu et al., 2009), neurology (SEO and LEE, 2011) andfeeding ecology and physiology (Sun et al., 2013) etc. However, fewstudies reported the impacts of Apostichopus japonicus farming onpond sediment (Ren et al., 2012; Zheng et al., 2009) or coastalwaters (Kang and Xu, 2016), no comprehensive investigations ofwater, sediment and biological characteristics were conducted todate.

Environmental variables selected in this research were all uni-versally recognized by previous studies. Some of them werespecially emphasized as highly informative descriptors of hetero-trophic loadings from aquaculture, such as dissolved inorganicnutrients (DIN) like nitrate (NO3

�), nitrite (NO2�), ammonium (NH4

þ)and phosphate (PO4

3�) in water (Sar�a, 2007), available phosphorus(AP) and acid volatile sulphide (AVS) in sediment (Yokoyama,2003). Other studied variables like total organic carbon (TOC) andtotal nitrogen (TN) has been widely used as important indices torepresent organic matter contents in water (Trott et al., 2004) andsediment (Bai et al., 2016; Gao et al., 2012; Wang et al., 2016; Zhaoet al., 2016). Similarly, C:N ratio, d13C and d15N were commonlyutilized in recent years as effective tools to trace and verify theorigin, extent and fate of nutrient sources in the environment andecosystem (Andrews et al., 1998; Sakamaki et al., 2010). The com-bination of these biogeochemical indicators will sufficientlydescribe and interpret the local environment variations (Kollongeiand Lorentz, 2014) induced by aquaculture activities.

The main purposes of this study were to (1) examine the im-pacts of sea cucumber pond discharge on local tidal flat by evalu-ating spatial differences (between farming and non-farming area oramong stations with a gradient distance to the aquaculture pond)of environmental variables in water and sediment; (2) detect dif-ferences in feeding strategies of dominant macrobenthos betweenfarming and non-farming areas. We hypothesized that significantspatial differences exist for contents of selected environmental

Please cite this article in press as: Fu, J., et al., Impacts of sea cucumber faNorthern China, Physics and Chemistry of the Earth (2017), http://dx.doi

variables by aquaculture activities with varying degrees. Besides,stable isotopes serve as an effective tool for revealing the sourceand fate of aquaculture induced organic matters in theenvironment.

2. Materials and methods

2.1. Study area

This study was conducted in the north part (37�500e38�100 N,118�200e119�000 E) of the Yellow River estuary (Fig. 1b) which islocated in the northeast of Shandong Province, the People's Re-public of China (Fig. 1a). The study site is characterized by a warm-temperate continental monsoonal climate, with mean annualtemperature and annual mean precipitation of 11.9 �C and 640 mm,respectively (Bai et al., 2015; Gao et al., 2014).

A total area of 66.67 km2 was constructed for sea cucumberponds along the north coast of Yellow River estuary. Each pond wasconstructed with 3 e 6 ha in size and 1.5 e 3 m in depth. Apos-tichopus japonicus are cultured in ponds with a density of 10 in-dividuals m�2. Commercial feed (fish meal 15%, soybean meal 20%,peanut flour 10% and seaweed meal 40%) is added to ponds 1e2times daily. Water exchange is managed uniformly twice a monthwhen spring tide occurs. Pond influent water is taken from onetidal creek using pumping systems. Pond effluents are released intounified drainage channels then discharged into adjacent wetlanddirectly through a sluice gate.

The tidal flat (C-tidal flat) adjacent to the sea cucumber pondarea is separated from sea cucumber ponds by a 7 m width road. Itis dominated by salt marsh plant Suaeda salsa and two macro-benthic species, Helice tridens and Macrophthalmus abbreviateswhich are widely distributed in this locality. In order to eliminatepotential effects of farming activities, we investigated the envi-ronments of a referential tidal flat (N-tidal flat) in Natural reservearea (non-farming area) which is 15 km departing from the C-tidalflat and was strictly considered with similar natural characteristics(such as sediment type, tidal characteristics and elevation).

2.2. Field sampling and laboratory procedures

Field samplingwas conducted in the sea cucumber pond area, C-tidal flat and N-tidal flat from May 1 to May 21, 2016. In the C-tidalflat, there were no other allochthonous nutrients loading exceptaquaculture effluent. In the sea cucumber pond area, two stations(Fig. 1c) were positioned in the sea cucumber pond and effluentchannel, respectively. Two gradient-based sampling were set up inC-tidal flat and N-tidal flat. Six stations (Fig. 1c) in the C-tidal flatcreek were set at 0 m, 50 m, 100 m, 200 m, 500 m and 800 mdistances from the sluice gate using an electronic distance meter(Nikon COOLSHOT 40i Laser Rangefinders) and a GPS receiver(Garmin GPS map64s). Four stations (Fig. 1d) in the N-tidal flatcreek were set at 0 m, 50 m, 100 m and 200 m distances from thereclaimed land. Other two stations (CM and NM) were set at themarine area of C-tidal flat and N-tidal flat (Fig. 1c and d) for wateranalysis (organic analysis only).

At each station (except for CM and NM stations), both water andsediment samples were collected for analyses of TOC, TN, C:N ratio,d13C and d15N, and dissolved inorganic nutrients (NO3

�, NO2�, NH4

þ

and PO43�/AP). The chemical items analyzed in this study were

shown with their abbreviations in Table 1. Dominant consumersand primary producers were sampled at 0 m, 50 m and 100 mstations in each tidal flat. All of these samplings were conducted inthe ebb tides.

At each station, three replicate water samples were collected bya bucket from the surface layer of the water for the analyses of

rming on biogeochemical characteristics in the Yellow River estuary,.org/10.1016/j.pce.2016.12.006

Page 3

Fig. 1. Map of the study area (a, b) showing sampling stations in sea cucumber farming area (c) and non-farming area (d). Farming area: sea cucumber pond (pond), effluent channel(EC), C-tidal flat (0 m to 800 m) and coastal area of C-tidal flat (CM); Non-farming area: N-tidal flat (0 m to 200 m) and coastal area of N-tidal flat (NM).

J. Fu et al. / Physics and Chemistry of the Earth xxx (2017) 1e12 3

particulate organic matter (POM) and dissolved inorganic nutri-ents. Marine POM (POM(marine)) in CM and NM stations werecollected at the sea surface of each adjacent offshore. POM sampleswere collected by filtering 0.4 e 3 L surface seawater through a125 mm mesh (to remove zooplankton) followed by glass-fiber fil-ters (Whatman GF/F, 447 mm). To remove carbonates, POM sam-ples were treated with 1.2 N HCl overnight and then oven dried at60 �C for stable isotope analysis. The remained filtered watersamples were treated immediately for nutrient analysis usingSANþþ continuous flow analyzer (SKALAR, Netherlands).

To detect the biogeochemical characteristics of deposited sedi-ment, three replicate surface layer (0 to 1 cm) sediments werecollected by using a 48mm (inner diameter) acrylic tube. An aliquotof the samples was used for the analysis of acid volatile sulphide

Table 1Chemical items analyzed in the present study and their abbreviations.

Analyzing items Water Sediment

Particulate organic matterIn the pond area POM(pond)

In the tidal flat area POM(tidal flat)

In the coastal area POM(marine)

Total organic carbon TOC(POM) TOC(S)Total nitrogen TN(POM) TN(S)

Carbon to nitrogen ratio C:N ratio(POM) C:N ratio(S)Stable carbon isotope ratio d13C(POM) d13C(S)Stable nitrogen isotope ratio d15N(POM) d15N(S)

Nitrate NO3�(water) NO3

�(S)

Nitrite NO2�(water)

Ammonium NH4þ(water) NH4

þ(S)

Phosphate PO43�

(water) AP(S)Acid volatile sulfides AVS

Please cite this article in press as: Fu, J., et al., Impacts of sea cucumber faNorthern China, Physics and Chemistry of the Earth (2017), http://dx.doi

(AVS) concentration to present sediment condition affected byaquaculture (Yokoyama, 2010). The remainder was divided into twogroups (group A and group B). Group Awas used for elemental (TOC& TN) and stable isotope analysis (d13C & d15N), group B was usedfor the analyses of dissolved inorganic nutrients in sediment.Sediment samples of group Awere weighed, dried at 60 �C for 72 h,then weighed again to determine the water content, and ground toa fine powder. Prior to analyses, samples for carbon isotope analysiswere soaked in 1.2 N HCl overnight to remove carbonates, filteredon a Nuclepore polycarbonate track-etch membrane filter (poresize ¼ 0.2 mm) and dried again. Aliquots of samples not soaked in1.2 N HCl were used for nitrogen isotopic analysis. Group B samplesused for dissolved inorganic nutrients analysis (Jia et al., 2017) wereair-dried at room temperature and passed through a 2 mm meshsieve. Samples for NO3

� and NH4þ analysis (5.0 g) were extracted

with 2 mol l�1 KCL (25 ml), and samples for AP (1.2 g) wereextracted with 0.5 mol l�1 NaHCO3 (25 ml) for determining theexchangeable nitrogen and phosphorous. Aftershock extraction for30 min, the samples were filtered and analyzed by SANþþ contin-uous flow analyzer to determine the concentrations of NO3

�, NH4þ

and AP.Benthic microalgae (BMA) samples (n ¼ 3) were extracted from

the surface sediment in each tidal flat (Couch, 1989; Riera andRichard, 1996). The sediment was brought back to the laboratorywithin 30 min and then spread on a flat tray (1 cm depth) andcovered by a nylon screen (63 mm mesh). A 4 e 5 mm layer ofcombusted sand (acid-washed and burned, 125 e 500 mm) wasspread over the nylon screen. The sand was kept wet using filteredseawater and then cultured under fluorescent light for 12e15 h. Thetop 2 mm of sand was then removed and filtered through a 63 mmfiber sieve by washing with distilled water. Water containing the

rming on biogeochemical characteristics in the Yellow River estuary,.org/10.1016/j.pce.2016.12.006

Page 4

J. Fu et al. / Physics and Chemistry of the Earth xxx (2017) 1e124

microalgae was filtered onto GF/F (425 mm) filters, and oven driedat 60 �C for stable isotope analysis.

The dominant salt marsh plant Suaeda salsa was collected byhand. Leaves from 10 plants were taken as a single sample. Plantdebris was collected by hand nets. The dominant macrobenthicanimals (two crabs) were caught by a long-handled scoop aroundthe sampling station within 5 m distance. Each sample was thenpreserved on ice and returned to the laboratory for sorting andsampling. Muscle tissue of crab was taken for animal isotopesample. All animal and plant samples were first treated with 1.2 NHCl to remove carbonates and then oven dried at 60 �C and groundto fine powder for isotope analysis.

The elemental and isotope analysis of samples were analyzedusing a Delta XP plus isotope ratio mass spectrometer linked to aConflo IIII interface (Thermo Fisher Scientific, USA) with a FlashEA1112 elemental analyzer.

Stable isotope ratios were expressed in the typical notation (d13Cor d15N), defined by the following equation:

d13C or d15N ¼ (Rsample/Rstandard � 1) � 103

and R ¼ 13C/12C or 15N/14N

The standard reference materials were Pee Dee Belemnite (PDB)and atmospheric N2 for carbon and nitrogen isotopes, respectively.

2.3. Data analysis

Principle components analysis (PCA; Manly, 1986) was con-ducted in R V3.3.1 to explore the relationships among stations thatdriven by sediment and water variables using the function prcompand princomp of R package stats. To evaluate spatial differences inN-tidal flat, C-tidal flat, effluent channel and sea cucumber pond, amultivariate techniquewas applied using the PRIMER 6.0 ecologicalsoftware package developed by the Plymouth Marine Laboratory.Ordination was computed via hierarchical cluster analysis (CLUS-TER) based on the Euclidean distance of studied environmentalvariables and isotopic compositions. Seriation test and SIMPER(Similarity percentages) analysis were performed to analyze thecommunity structure. One-way ANOSIM was performed to test thedifferences in the studied areas.

Multiple comparisons of each environmental variable betweenstations were conducted to detect spatial differences betweenfarming and non-farming areas or among stations with a gradientdistance to the aquaculture pond. One-way ANOVAwas used whendata is normally distributed (Shapiro and Wilk, 1965) and homo-scedastic (Bartlett, 1937) followed by Bonferroni post hoc test (Rice,1989). Kruskal-Wallis test (Kruskal and Wallis, 1952) was per-formed when the data is not normally distributed or homosce-dastic, followed by Gao post hoc test (Gao et al., 2008; Konietschkeet al., 2015) to compare differences among stations for the selectedenvironmental variable. Statistical significance level was consid-ered at 5 %. Statistical analyses were performed using the functionWilk-Shapiro test, Bartlett's test, Bonferroni test and Kruskal test ofthe R package stats and the function Gao of the R package npar-comp (Ihaka and Gentleman, 1996; Shirley, 1977) and oneway_testfunction of the R package coin (Ihaka and Gentleman, 1996) in RV3.3.1. As no significant differences (ANOVA with Bonferroni’s posthoc tests, p > 0.05) were observed in the N-tidal flat along thegradient (0 m, 50 m, 100 m, 200 m), the data from N-tidal flatsampling stations were calculated together to represent a referencelocation expressed as N-tidal flat.

Carbon and nitrogen stable isotopes of primary producers (foodsources) and consumers were initially investigated using biplotdiagrams. Stable isotope values of consumers and primary

Please cite this article in press as: Fu, J., et al., Impacts of sea cucumber faNorthern China, Physics and Chemistry of the Earth (2017), http://dx.doi

producers were calculated together in each tidal flat due to nosignificant differences (Bonferroni’s post hoc tests, p > 0.05) ofthem were found along the gradient in each tidal flat. A t-test wasperformed to test significant differences of stable isotope values ofconsumers and primary producers between different tidal flats. TheBayesian stable isotope mixing model, SIAR v 4.0 (Stable IsotopeAnalysis in R) (Parnell et al., 2010) was used to determine propor-tional contributions of potential food sources to dominant con-sumers. Contributions of dietary sources were represented as meanand 95% credibility interval (Ci). The fractionations d13C and d15Nfor crustacean (muscle, acid treated) calculated from raw datapresented in Yokoyama et al. (2005) were d13C ¼ 1.93 ± 0.40‰,d15N ¼ 3.97 ± 0.49‰, respectively.

3. Results

3.1. Environmental variables

Each sampling location (sea cucumber pond, effluent channel, C-tidal flat and N-tidal flat) were clearly separated from each other byPCA analysis (Fig. 2a) with seventeen water and sediment variables(Table 1). The first three principal components (PC1, PC2 and PC3)explained 73% of the total variance. The main contributions to PC1included sediment variables TOC(S), TN(S), C:N ratio(S), d13C(S), NO3

�(S),

NH4þ

(S), AP(S) (with large positive coefficients), d15N(S) and d13C(POM)

(with negative coefficient). Vector plots of sediment variables andd13C(POM) showed that PC1 represented axes of increasing organicaccumulation and nutrients concentrations in sediment anddecreasing d15N(S) and d13C(POM) from tidal flat to effluent channeland then followed by sea cucumber pond. As for PC2 (Fig. 2b),POM variables (except for d13C(POM)) and PO4

3�(water) in water were

closely related to PC2 as indicated by large positive coefficients(TOC(POM), TN(POM), C:N ratio(POM), d15N(POM)) and negative coeffi-cient (PO4

3�(water)), respectively. PC2 represented axes of increasing

organic accumulation in POM and decreasing PO43�

(water) concen-tration from sea cucumber pond area to N-tidal flat. The maincontribution to PC3 (Fig. 2c) was fromDIN(water) that showed highlypositive coefficients (NO3

�(water), NO2

�(water), NH4

þ(water)). As thevector

plots of DIN(water) that PC3 represented, however, no obviousgradient variations were found across stations.

The cluster analysis (Fig. 3) showed that studied localities couldbe distinguished into three groups (sea cucumber pond, the com-bination of C-tidal flat and effluent channel, N-tidal flat) with thethreshold Euclidean distance more than 7.0 (Fig. 3). The results ofthe pairwise test in ANOSIM showed that significant differenceswere present between N-tidal flat and sea cucumber pond (R ¼ 1,p < 0.01), N-tidal flat and the combination of C-tidal flat andeffluent channel (R ¼ 0.529, p < 0.01), sea cucumber pond and thecombination of C-tidal flat and effluent channel (R ¼ 0.991,p < 0.01). With the subsequent SIMPER analysis, the primarycontributors to three groups were TOC(S) (89.60% contribution),TOC(POM) (63.16% contribution) and TOC(S) (97.82% contribution).

3.2. Water and sediment biogeochemical characteristics

3.2.1. Water biogeochemical characteristicsTOC(POM) and TN(POM) concentrations (Fig. 4a and b) detected in

farming area (mean ± SD ¼ 0.4 ± 0.2 and 0.07 mg l�1, respectively)were significantly lower than those in N-tidal flat (1.4 ± 0.6 and0.2 ± 0.1 mg l�1, respectively). C:N ratios of POM (Fig. 4c) rangedfrom 4.7 to 9.1 and showed significant differences between seacucumber pond and N-tidal flat (ANOVAwith Bonferroni’s post hoctests, p < 0.001). d13C(POM) in sea cucumber pond (�20.6 ± 1.0 ‰),effluent channel (�19.9 ± 1.2‰) and C-marine stations(�21.1 ± 0.1‰) showed significantly lower values than those in C-

rming on biogeochemical characteristics in the Yellow River estuary,.org/10.1016/j.pce.2016.12.006

Page 5

Fig. 2. Principal component analysis (PCA) based on the seventeen sediment and water parameters. (a) 3D scores plot (PC1, PC2 and PC 3) of the environmental variables to thesamples. (b) PC1 and PC2. (c) PC1 and PC3.

Fig. 3. Dendrogram showing cluster relationship between samples. Different sampling locations were shown in different symbols (filled circle: N-tidal flat, open circle: C-tidal flat,triangle: effluent channel, square: sea cucumber pond). See Fig. 1 for the sample numbers.

J. Fu et al. / Physics and Chemistry of the Earth xxx (2017) 1e12 5

tidal flat (�16.7 ± 1.6‰), N-tidal flat (�16.9 ± 1.2‰) and N-marine(�18.4 ± 0.4‰) stations (Fig. 4d). d15N(POM) values in the sea cu-cumber pond (�2.0 ± 0.2‰) was significantly lower than those inother stations (Bonferroni’s post hoc tests, p < 0.001). No remark-able differences were found within C-tidal flat stations(0.9 ± 0.6‰).

NO3� and NH4

þ showed no significant differences (ANOVA, p >0.05) among stations (Fig. 4f and h), whereas significant differenceswere present in NO2

� and PO43� (Fig. 4g, i; Kruskal-Wallis test, p <

0.01 and p < 0.001, respectively). The NO2� concentration in sea

Please cite this article in press as: Fu, J., et al., Impacts of sea cucumber faNorthern China, Physics and Chemistry of the Earth (2017), http://dx.doi

cucumber pond (53 ± 2 mg l�1) was clearly higher than that in N-tidal flat (40 ± 9 mg l�1), but no distinct differences were foundamong stations in the farming area or between C-tidal flat and N-tidal flat. The PO4

3� concentrations in sea cucumber pond(32 ± 1 mg l�1), effluent channel (33 ± 1 mg l�1) and C-tidal flat(29 ± 3 mg l�1) were clearly higher than that in N-tidal flat(23 ± 2 mg l�1), but no distinct differences were found within sta-tions in the farming area.

rming on biogeochemical characteristics in the Yellow River estuary,.org/10.1016/j.pce.2016.12.006

Page 6

Fig. 4. Comparisons of water parameters including TOC (a), TN (b), C:N ratio (c), d13C (d), d15N (e), NO3� (f), NO2

� (g), NH4þ (h) and PO4

3� (i). Different letters above bars indicatesignificant differences (p < 0.05) by Bonferroni post hoc test (TOC, TN, C:N ratio, d15N, NO3

�, NH4þ) and Gao post hoc test (d13C, NO2

�, PO43�). Boxes represent 25% e 75% percentiles,

range bars represent the 5% and 95% percentiles, solid dots in the boxes represent mean values, hollow dots represent values outside of 95% confidences interval.

J. Fu et al. / Physics and Chemistry of the Earth xxx (2017) 1e126

3.2.2. Sediment biogeochemical characteristicsAVS concentration detected in sea cucumber pond was

0.1 ± 0.5 mg g�1 (dry sediment) higher than those in the effluentchannel, C-tidal flat and N-tidal flat (all less than 0.01 mg g�1). ForTOC(S) and TN(S), significant differences (Kruskal-Wallis test, p <0.01 and p < 0.001, respectively) were observed among stations(Fig. 5a and b). TOC(S) and TN(S) in the sea cucumber pond(5.9 ± 1 mg g�1 and 0.8 ± 0.3 mg g�1, respectively) and effluentchannel (2.95 ± 0.43 mg g�1 and 0.51 ± 0.08 mg g�1, respectively)were significantly higher than those in the C-tidal flat(1.2 ± 0.3 mg g�1 and 0.3 ± 0.1 mg g�1, respectively) and N-tidal flat

Please cite this article in press as: Fu, J., et al., Impacts of sea cucumber faNorthern China, Physics and Chemistry of the Earth (2017), http://dx.doi

(1.8 ± 0.2 mg g�1 and 0.3 mg g�1, respectively). Both TOC(S) andTN(S) contents showed decreasing trends from sea cucumber pondto 0 m station, and no gradient variations along the dispersal dis-tance in C-tidal flat.

C:N ratios (Fig. 5c) in the pond, effluent channel, 0 m station andN-tidal flat had significant differences from other stations in C-tidalflat (50 m, 100 m, 200 m and 800 m). A decreasing trend was foundfrom pond to 0 m. d13C(S) and d15N(S) values of the sea cucumberpond were remarkably different from the effluent channel, C-tidalflat and N-tidal flat (Fig. 5d and e). A decreasing trend of d13C(S) wasfound from pond to 0 m station. d13C(S) of 0 m station showed a

rming on biogeochemical characteristics in the Yellow River estuary,.org/10.1016/j.pce.2016.12.006

Page 7

Fig. 5. Comparisons of sediment parameters including TOC (a), TN (b), C:N ratio (c), d13C (d), d15N (e), NO3� (f), NH4

þ (g) and AP (h). Different letters above bars indicate significantdifferences (p < 0.05) by Bonferroni post hoc test (d15N, AP) and Gao post hoc test (TOC, TN, C: N ratio, d13C, NO3

�, NH4þ). Boxes represent 25%e75% percentiles; range bars represent

the 5% and 95% percentiles, solid dots in the boxes represent mean values, hollow dots represent values outside of 95% confidences interval.

J. Fu et al. / Physics and Chemistry of the Earth xxx (2017) 1e12 7

large range of standard deviation and no significant differenceswith the effluent channel and other stations in C-tidal flat and N-tidal flat.

No significant differences (Fig. 5f) among stations were foundfor NO3

�(S) (Kruskal-Wallis test, p > 0.05). Both NO3

�(S) and NH4

þ(S)

were found decreasing trends from pond to 0 m station. The con-centration of NH4

þ(S) ranged from 1.8 to 18.1 mg kg�1 (Fig. 5g) and

varied across stations (Kruskal-Wallis test, p < 0.001). The highestNH4

þ(S) concentration found in sea cucumber pond

(15.8 ± 2.0 mg kg�1) was significantly higher than those fromeffluent channel (8.6 ± 2.9 mg kg�1), C-tidal flat (5.7 ± 2.6 mg kg�1)and N-tidal flat stations (4.3 ± 0.5 mg kg�1). NH4

þ(S) content in

effluent channel showed no significant differences with C-tidal flatstations (except for 500 m), but was obviously higher than that inN-tidal flat. AP contents (Fig. 5h) in pond (8.2 ± 0.9 mg kg�1) andeffluent channel (10.4 ± 3.9 mg kg�1) were significantly higher(Bonferroni’s post hoc tests, p < 0.001) than those in both C-tidalflat (3.6 ± 0.6 mg kg�1) and N-tidal flat (3.4 ± 2.1 mg kg�1).

Please cite this article in press as: Fu, J., et al., Impacts of sea cucumber faNorthern China, Physics and Chemistry of the Earth (2017), http://dx.doi

3.3. Feeding strategies of dominant macrobenthos

d13C value of H. tridens in the C-tidal flat (�13.0 ± 0.6‰) wassignificantly higher (t-test, p < 0.01) than that in the N-tidal flat(�15.5 ± 3.3‰; Table 2). d15N value of H. tridens in C-tidal flat(4.3 ± 1.1‰) was significantly lower than that in N-tidal flat(5.8 ± 0.4‰). As for the crabM. abbreviates, d13C values were similar(t-test, p > 0.05) between C-tidal flat (�12.1 ± 0.6 ‰) and N-tidalflat (�12.3 ± 0.4‰), whereas d15N value in C-tidal flat (2.1 ± 0.7‰)was significantly lower (t-test, p < 0.01) compared to that in N-tidalflat (3.8 ± 0.9‰).

In according with the presumptive trophic shift of crustaceans(1.9 ± 0.4‰ in d13C and 4.0 ± 0.5‰ in d15N) presented by Yokoyamaet al. (2005), the expected d13C and d15N values of the H. tridens dietwould be �17.4‰ and 1.8‰, respectively in N-tidal flat,and �14.9‰ and 0.3‰, respectively in C-tidal flat. The expectedd13C and d15N values of M. abbreviates diet would be �14.2‰ and�0.2‰, respectively in N-tidal flat, and �14.0‰ and �1.9‰,respectively in C-tidal flat.

Among the primary producers, salt marsh plant Suaeda salsa

rming on biogeochemical characteristics in the Yellow River estuary,.org/10.1016/j.pce.2016.12.006

Page 8

Table 2Carbon and nitrogen isotopic compositions of consumers and their potential food sources in the tidal flat of Natural Reserve Area and sea cucumber pond area.

Consumers/producers N-tidal flat C-tidal flat

d13C (‰) d15N (‰) n d13C (‰) d15N (‰) n

Helice tridens �15.5 ± 3.3a 5.8 ± 0.4a 12 �13.0 ± 0.6b 4.3 ± 1.1b 21Macrophthalmus abbreviates �12.3 ± 0.4a 3.8 ± 0.9a 10 �12.1 ± 0.6a 2.1 ± 0.7b 7POM (marine) �18.4 ± 0.3a 1.5 ± 0.3a 3 �21.1 ± 0.1b 2.4 ± 0.3b 3POM (tidal flat) �16.9 ± 1.2a 2.7 ± 0.4a 12 �16.7 ± 1.6a 0.9 ± 0.6b 18POM (pond) �20.6 ± 0.9 �2.0 ± 0.2 3BMA �8.1 ± 0.4a 2.2 ± 0.4a 3 �10.7 ± 1.5a 3.3 ± 0.4b 3Salt marsh plant �30.4 ± 0.5a 6.3 ± 1.5a 9 �29.7 ± 1.3a 6.1 ± 0.9a 9Plant debris �25.8 ± 0.9a 0.7 ± 0.4a 3 �17.7 ± 1.6b �0.1 ± 0.4a 3

The mean ± SD values are given for food items with the number of samples (n)� 3. BMA: benthic microalgae. Different letters (a, b) on stable isotope values indicate significantdifference (p < 0.05) by t-test.

J. Fu et al. / Physics and Chemistry of the Earth xxx (2017) 1e128

had the lowest d13C (around �29.7 ± 1.3 to �30.4 ± 0.5‰) andhighest d15N (around 6.1 ± 0.9 to 6.3 ± 1.5‰) values. There werelarge differences in isotope compositions between Suaeda salsa andhypothetical diet of the two crabs. The d13C values of hypotheticaldiet for two crabs were between POM (POM(marine), POM(tidal flat)and POM(pond)) and BMA (Fig. 6).

SIAR analysis result (Fig. 7a and b) revealed that in N-tidal flat,the contributions of POM(marine) to the diet of H. tridens (25%, Ci0 e 49%) was similar to those of BMA (27%, Ci 8 e 45%) and plantdebris (25%, Ci 8 e42%; Fig. 7a), while POM(tidal flat) (19%, Ci 0e37%)made small contribution and salt marsh plant Suaeda salsa (4%, Ci0e 9%) mademinor contribution. In C-tidal flat, POM(marine) made asmaller contribution to the diet of H. tridens (3%, Ci 0 e 7%), whileBMA made a greater contribution (45%, Ci, 36 e 54%) (Fig. 7b).Besides, the contributions of POM(tidal flat) (15%, Ci 0 e 32%), plantdebris (18%, Ci 0 e 36%) and Suaeda salsa (1%, Ci 0 e 2%) wererelatively low. In addition, POM(pond) contributed 19% (Ci 2 e 34%)of the diet of H. tridens in C-tidal flat.

As for crab M. abbreviates (Fig. 7c and d), BMA overwhelminglydominated the diet in N-tidal flat (51%, Ci 39 e 63%) and C-tidal falt(54%, Ci 37e69%). But samewithH. tridens, POM(marine) showed themuch lower contribution to the diet of M. abbreviates in C-tidal flat(6%, Ci 0 e 16%) than that in N-tidal flat (22%, Ci 0 e 44%). Likewise,POM(tidal flat) and plant debris made small contributions, and Suaedasalsa made a minimal contribution to the diet of M. abbreviates inboth N-tidal flat and C-tidal falt. But, POM(pond) contributed arelatively lower amount (9%, Ci 0e 22%) to the diet ofM. abbreviatesthan it contributed to H. tridens in C-tidal flat.

Fig. 6. Stable isotope biplots of d13C and d15N for dominant consumers and their po-tential food sources in N-tidal flat (a) and C-tidal flat (b). Grey open circle: expectedisotope composition of the diet of each crab. Error bars represent SD.

4. Discussion

4.1. Differences of distance-based environmental variables and theirimplications

Overall, this study intended to evaluate the environmental im-pacts of sea cucumber farming based on two kinds of analyses. Thefirst one was the detection of discrepancies among distance-basedpollution indicators in the water column and sediment. The secondone was the utilization of biochemical tracers (d13C and d15N) totrace the origin of organic matters. Both methods had been widelyconfirmed as reliable tools to evaluate environmental effects ofaquaculture (Gowen and Bradbury, 1987; Sar�a, 2007; Sar�a et al.,2004; Wai et al., 2011). A total of nine typical environmental vari-ables (TOC, TN, C:N ratio, AVS, NO3

�, NH4þ, NO2

�, PO43� and AP) were

studied in this paper. For the studied parameters in the water col-umn, they showed no uniform variation tendencies. For example,NH4

þ(water) and NO3

�(water) contents were similar between farming

and non-farming areas, or among stations within the farming area(such as pond and its adjacent tidal flat). TOC(POM), TN(POM),

Please cite this article in press as: Fu, J., et al., Impacts of sea cucumber faNorthern China, Physics and Chemistry of the Earth (2017), http://dx.doi

NO2�(water) and PO4

3�(water) contents in the farming area were

different from those in N-tidal flat, while they did not differ amongfarming areas (Fig. 4). In our study, distribution patterns of studiedparameters may have many potential reasons. One possible reasonlies in that eutrophication, related to the rapid development ofmariculture (Bouwman et al., 2013) or other anthropogenic activ-ities, is widely present in the Yellow River estuary and its adjacentwaters (Chen et al., 2013; Kang and Xu, 2016; Zhang et al., 2008).

rming on biogeochemical characteristics in the Yellow River estuary,.org/10.1016/j.pce.2016.12.006

Page 9

Fig. 7. Proportionate contribution of potential food sources to the diet of Helice tridents and Macrophthalmus abbreviates by SIAR boxplots. Plant: salt marsh plant. Debris: plantdebris. Dark gray, gray and light gray boxes show 50, 75, 95 credibility interval respectively.

J. Fu et al. / Physics and Chemistry of the Earth xxx (2017) 1e12 9

High background values hinder the statistical differences.In previous studies, NO3

�(water) was recognized as the typical

dominant part of DIN in the studied coastal waters (Gong, 2012;Zhang et al., 2008). But our study found the same result as Kangand Xu, 2016, that NH4

þ(water) represented the relative major pro-

portion of DIN than NO3�(water) in the mariculture zone. This result

corresponded well to the meta-analysis of Sar�a (2007) thatNH4

þ(water) appears to be the most affected compound by aquacul-

ture loadings than NO2�(water), NO3

�(water) and followed by PO4

3�(-

water). However, discrepancies among distribution patterns ofstudied parameters prevented us from drawing much more clearconclusions.

Despite the distinct distribution patterns of nutrients in thewater column, TOC(S), TN(S), NO3

�(S), NH4

þ(S), AP(S) and AVS contents

in sediment showed a uniform distance-based differences. Theyhad significantly higher values in farming area than those in non-farming areas. Within the farming area, a decrease transect frompond to 0 m station was detected. These findings clearly demon-strated the presence of aquaculture waste from the sea cucumberfarming, and additionally implied a restricted dispersion distance(approximately 50 m in C-tidal flat) of aquaculture waste thataccumulated in sediment. It was also worth noting that althoughAVS content in sea cucumber pond in this study was below thethreshold (0.2 mg/g) in some rules ensuring a sustainable aqua-culture (Uede, 2008; Yokoyama, 2003; Yokoyama et al., 2007), the

Please cite this article in press as: Fu, J., et al., Impacts of sea cucumber faNorthern China, Physics and Chemistry of the Earth (2017), http://dx.doi

content was much higher than those reported from other naturalsediments. Similar cases were also found in TOC(S) and TN(S) con-centrations in sea cucumber pond. This phenomenon was alsoconsidered as a result of sea cucumber aquaculture.

4.2. Detection of origins of organic matters and their implications

In the water column, the result of C:N ratios(water) and isotopiccomposition analysis revealed that sea cucumber pond haddifferent organic origins from adjacent tidal flat and non-farmingarea. For example, C:N ratios(POM) detected in pond, 0 m, 50 m,100 m and 500 m stations were close to the Redfield ratio(C:N ¼ 106:16) for marine plankton (Andrews et al., 1998;Burkhardt et al., 1999; Redfield, 1963), which might indicate thatthe main contributor of POM in these stations was marine phyto-plankton. The relatively higher values of C:N ratios(POM) in theeffluent channel, 200 m, 800m and N-tidal flat suggested that POMresulted from amixture of marine phytoplankton and other organicmatter sources. Low C:N ratios(POM) detected in N-marine and C-marine areas might be attributed to the contribution of marinezooplankton with low C:N ratio (Båmstedt, 1986; Walve andLarsson, 1999). The lowest d13C(POM) and d15N(POM) values found insea cucumber pond maybe due to the allogenous sources. Terres-trial plant (soybean and peanut) and seaweed flour (seaweedcultured in offshore) introduced by aquaculture (commercial feeds)

rming on biogeochemical characteristics in the Yellow River estuary,.org/10.1016/j.pce.2016.12.006

Page 10

J. Fu et al. / Physics and Chemistry of the Earth xxx (2017) 1e1210

had low d13C and d15N values, and the presence of allogenoussources resulted in low isotopic values. In this case, the distributionpattern of d13C(POM) and d15N(POM) values indicated the environ-mental impact of aquaculture.

In the sediments, high d13C(S) and low d15N(S) values weredetected in sea cucumber pond. Similar to the reason of lowd15N(POM) compositions in the water column, low d15N(S) valuesshould be due to the allogenous organic matter input. d13C(S) inpond sediment was approximately equal to d13C(POM) of POM,which suggested that they had homogeneous resources. In theadjacent and natural tidal flat, deposition of organic matter fromsalt marsh plant (d13C ¼ 29.74 ± 1.31‰, d15N ¼ 6.1 ± 0.91‰)contributed to the sediment with much lower d13C and high d15Nvalues. Therefore, high d13C and low d15N values in the pond were areflection of the disturbance of aquaculture waste.

4.3. Effects of pond farming waste on food sources for dominantmacrobenthos

SIAR analysis results revealed that BMAwas the most importantfood sources for H. tridens and M. abbreviates in both farming andnon-farming area. However, for both species, contributions of BMAto the diet in the C-tidal flat were higher than those in N-tidal flat.This founding might imply that nutrient enrichment in C-tidal flatmay enhance the primary production of BMA which furthermoreenabled a higher contribution to the consumer's diet. Meanwhile,the contribution of POM(marine) in C-tidal flats was lower than thatin N-tidal flats and the POM(pond) showed opposite tendencies.These differences indicated that sea cucumber farming activitiesmight modify the dietary characteristics of benthic animals to acertain extent. The similar modifications of food composition ofdominant benthic animals by aquaculture were confirmed in pre-vious studies of other aquaculture types, such as fish farming(Mazzola and Sar�a, 2001; Yokoyama and Ishihi, 2007) and shrimppond farming (Kon et al., 2009).

4.4. Implications for operation and management of sea cucumberfarming

The overall conclusion of the present study was that sea cu-cumber farming slightly altered the local environment, however,this modification was restricted to a limited area (within 50 m). Upto present, there existed a relative paucity of details concerning theenvironmental effects of sea cucumber farming (Li et al., 2014; Renet al., 2010; Zheng et al., 2009). Therefore, it's hard to make acomparison of our results with the previous studies. In this study,the short dispersal distance might be attributed to the ideal oper-ation mode and management of sea cucumber pond farming in thestudied region. Firstly, the main feed for Apostichopus japonicus iscomposed of much seaweed flour (40%) and less protein feeds(fishmeal ¼ 15%) in comparison with the shrimp feed (around 45%)(Dierberg and Kiattisimkul, 1996; Paez-Osuna, 2001; Trott et al.,2004), that reduced the accumulation of organic matter from thewaste feed. Besides, the intense of sea cucumber farming activitieswas moderate, with low densities (10 individuals m�2) that farbelow the typical densities (20 individuals m�2) practiced by Chi-nese farmers (Dong et al., 2010; Kang and Xu, 2016). In addition,waste waters from sea cucumber ponds, connected by internalpipes, were gathered in large channels and drained intermittentlyby valves, which may result in a higher deposition before thedischarge. The modification in sediment was restricted to 50 mdistance, which indicated that farming pond and effluent channelwere heavily disturbed. Consequently, high natural feeds content,low stocking densities, and line effluent channels reduced thedeleterious effects of sea cucumber farming on the adjacent

Please cite this article in press as: Fu, J., et al., Impacts of sea cucumber faNorthern China, Physics and Chemistry of the Earth (2017), http://dx.doi

environment. Besides, deposited sludge should be more efficientlyremoved by physical efforts or by filters like salt marsh plant, bi-valves and macroalgae.

For the management of aquaculture activities, one of the chal-lenges is to develop practical, cost-effective monitoring techniquesthat reflect the impact and processes (Burford et al., 2003). Thepresent study provides meaningful information about environ-mental impacts of Apostichopus japonicus farming. The uniquecharacteristics of Apostichopus japonicus farming discharges indi-cated that the predominant effects of sea cucumber farmingdischarge were in the sediment processes, rather than water col-umn processes. In this case, environmental variables should beunequally treated.

In conclusion, the present study considered that pond farmingfor Apostichopus japonicus in the Yellow River estuary altered localenvironment to a certain extent, as determined by the discrep-ancies among distance-based environmental parameters in sedi-ment (TOC(S), TN(S), NO3

�(S), NH4

þ(S), AP(S) and AVS) as well as isotopic

signatures (d13C and d15N). Such kind of modification coveredfarming pond, effluent channel, and adjacent wetland within 50 mrange. Apostichopus japonicus farming also modified dietary char-acteristics of some dominant macroinvertebrates. For methodo-logical consideration, sediment biogeochemical characteristics as ahistorical recorder much more effectively reflected both organicand dissolved inorganic nutrients induced by aquaculture.

Acknowledgements

This study was supported financially by National Key BasicResearch Programme of China (2013CB430406). The present studywas conducted using Cooperative Research Facilities (Isotope RatioMass Spectrometer) of Center for Ecological Research, Kyoto Uni-versity. We are grateful to the support from the lab of AquaticEnvironmental Biology, Kyoto University. We thank Dr. EdouardLavergne for the idea of statistical analyses in environment vari-ables. And thank Prof. Yoh Yamashita, Dr. Bunei Nishimura and Dr.Zou Yuxuan for their assistance in different aspects of this study.

References

Andrews, J., Greenaway, A., Dennis, P., 1998. Combined carbon isotope and C/Nratios as indicators of source and fate of organic matter in a poorly flushed,tropical estuary: Hunts Bay, Kingston Harbour, Jamaica. Estuar. Coast. Shelf Sci.46, 743e756.

Bai, J., Zhao, Q., Lu, Q., Wang, J., Reddy, K.R., 2015. Effects of freshwater input ontrace element pollution in salt marsh soils of a typical coastal estuary. China. J.Hydrol. 520, 186e192.

Bai, J., Zhang, G., Zhao, Q., Lu, Q., Jia, J., Cui, B., Liu, X., 2016. Depth-distributionPatterns and Control of Soil Organic Carbon in Coastal Salt Marshes withDifferent Plant Covers. Scientific reports 6.

Båmstedt, U., 1986. Chemical Composition and Energy Content. The BiologicalChemistry of Marine Copepods. Clarendon Press, Oxford, pp. 1e58.

Bartlett, M.S., 1937. Properties of sufficiency and statistical tests. In: Proceedings ofthe Royal Society of London. Series a, Mathematical and Physical Sciences,pp. 268e282.

Bordbar, S., Anwar, F., Saari, N., 2011. High-value components and bioactives fromsea cucumbers for functional foodsda review. Mar. Drugs 9, 1761e1805.

Bouwman, L., Beusen, A., Glibert, P.M., Overbeek, C., Pawlowski, M., Herrera, J.,Mulsow, S., Yu, R., Zhou, M., 2013. Mariculture: significant and expanding causeof coastal nutrient enrichment. Environ. Res. Lett. 8, 044026.

Brooks, K.M., Mahnken, C.V., 2003. Interactions of Atlantic salmon in the Pacificnorthwest environment: II. Organic wastes. Fish. Res. 62, 255e293.

Burford, M.A., Costanzo, S.D., Dennison, W.C., Jackson, C.J., Jones, A.B.,McKinnon, A.D., Preston, N.P., Trott, L.A., 2003. A synthesis of dominantecological processes in intensive shrimp ponds and adjacent coastal environ-ments in NE Australia. Mar. Pollut. Bull. 46, 1456e1469.

Burkhardt, S., Zondervan, I., Riebesell, U., 1999. Effect of CO2 concentration on C: N:P ratio in marine phytoplankton: a species comparison. Limnol. Oceanogr. 44,683e690.

Chen, J., 2003. Overview of sea cucumber farming and sea ranching practices inChina. SPC beche-de-mer Inf. Bull. 18, 18e23.

Chen, P., Liu, S., Zhang, G., Li, L., Gu, P., Cao, X., 2013. Monthly variation of nutrient

rming on biogeochemical characteristics in the Yellow River estuary,.org/10.1016/j.pce.2016.12.006

Page 11

J. Fu et al. / Physics and Chemistry of the Earth xxx (2017) 1e12 11

concentrations and fluxes in the lower Huanghe River: under the influence ofartificial floods. Acta Oceanol. Sin. 35, 59e71.

Couch, C.A., 1989. Carbon and nitrogen stable isotopes of meiobenthos and theirfood resources. Estuar. Coast. Shelf Sci. 28, 433e441.

Dierberg, F.E., Kiattisimkul, W., 1996. Issues, impacts, and implications of shrimpaquaculture in Thailand. Environ. Manag. 20, 649e666.

Dong, S., Liang, M., Gao, Q., Wang, F., Dong, Y., Tian, X., 2010. Intra-specific effects ofsea cucumber (Apostichopus japonicus) with reference to stocking density andbody size. Aquac. Res. 41, 1170e1178.

FAO, 2008. Sea Cucumbers, a Global Review of Fisheries and Trade. Food andAgriculture Organization, Rome, Italy.

Gao, H., Bai, J., He, X., Zhao, Q., Lu, Q., Wang, J., 2014. High temperature and salinityenhance soil nitrogen mineralization in a tidal freshwater marsh. PloS One 9,e95011.

Gao, X., Alvo, M., Chen, J., Li, G., 2008. Nonparametric multiple comparison pro-cedures for unbalanced one-way factorial designs. J. Stat. Plan. Inference 138,2574e2591.

Gao, X., Yang, Y., Wang, C., 2012. Geochemistry of organic carbon and nitrogen insurface sediments of coastal Bohai Bay inferred from their ratios and stableisotopic signatures. Mar. Pollut. Bull. 64, 1148e1155.

Gong, Y., 2012. Influencing Factor of Nutrient Transport in the Lower Reach ofYellow River. Doctoral Dissertation. Ocean University of China, Qingdao,pp. 1e110.

Gowen, R.J., Bradbury, N.B., 1987. The ecological impact of salmonid farming incoastal waters - a review. Oceanogr. Mar. Biol. 25, 563e575.

Han, Q., Keesing, J.K., Liu, D., 2016. A review of sea cucumber aquaculture, ranching,and stock enhancement in China. Rev. Fish. Sci. Aquac. 24, 326e341.

Holby, O., Hall, P., 1991. Chemical fluxes and mass balances in a marine fish cagefarm. II. Phosphorus. Mar. Ecol. Progr. Ser. 70, 263e272.

Ihaka, R., Gentleman, R., 1996. R: a language for data analysis and graphics.J. Comput. Graph. Stat. 5, 299e314.

Jia, J., Bai, J., Gao, H., Wen, X., Zhang, G., Cui, B., Liu, X., 2017. In situ soil net nitrogenmineralization in coastal salt marshes (Suaeda salsa) with different floodingperiods in a Chinese estuary. Ecol. Indic. 73, 559e565.

Kang, P., Xu, S., 2016. The impact of mariculture on nutrient dynamics and identi-fication of the nitrate sources in coastal waters. Environ. Sci. Pollut. Res. Int. 23,1300e1311.

Kollongei, K.J., Lorentz, S.A., 2014. Connectivity influences on nutrient and sedimentmigration in the Wartburg catchment, KwaZulu-Natal Province, South Africa.Phys. Chem. Earth, Parts A/B/C 67, 12e22.

Kon, K., Kawakubo, N., Aoki, J.I., Tongnunui, P., Hayashizaki, K.I., Kurokura, H., 2009.Effect of shrimp farming organic waste on food availability for deposit feedercrabs in a mangrove estuary, based on stable isotope analysis. Fish. Sci. 75,715e722.

Konietschke, F., Placzek, M., Schaarschmidt, F., Hothorn, L.A., 2015. nparcomp: an RSoftware Package for Nonparametric Multiple Comparisons and SimultaneousConfidence Intervals, 2015 64, 17.

Kruskal, W.H., Wallis, W.A., 1952. Use of ranks in one-criterion variance analysis.J. Am. Stat. Assoc. 47, 583e621.

Li, J., Dong, S., Gao, Q., Wang, F., Tian, X., Zhang, S., 2014. Total organic carbon budgetof integrated aquaculture system of sea cucumber Apostichopus japonicus, jel-lyfish Rhopilema esculenta and shrimp Fenneropenaeus chinensis. Aquac. Res.45, 1825e1831.

Li, Q., Chen, L., Kong, L., 2009. A genetic linkage map of the sea cucumber, Apos-tichopus japonicus (Selenka), based on AFLP and microsatellite markers. Anim.Genet. 40, 678e685.

Liu, G., Yang, H., Liu, S., 2010. Effects of rearing temperature and density on growth,survival and development of sea cucumber larvae, Apostichopus japonicus(Selenka). Chin. J. Oceanol. Limnol. 28, 842e848.

Liu, Y., Dong, S., Tian, X., Wang, F., Gao, Q., 2009. Effects of dietary sea mud andyellow soil on growth and energy budget of the sea cucumber Apostichopusjaponicus (Selenka). Aquaculture 286, 266e270.

Lupatsch, I., Kissil, G.W., 1998. Predicting aquaculture waste from gilthead seabream(Sparus aurata) culture using a nutritional approach. Aquat. Living Resour. 11,265e268.

Manly, B.F., 1986. Randomization and regression methods for testing for associa-tions with geographical, environmental and biological distances betweenpopulations. Res. Popul. Ecol. 28, 201e218.

Mazzola, A., Sar�a, G., 2001. The effect of fish farming organic waste on food avail-ability for bivalve molluscs (Gaeta Gulf, Central Tyrrhenian, MED): stable car-bon isotopic analysis. Aquaculture 192, 361e379.

Naylor, R.L., Goldburg, R.J., Primavera, J.H., Kautsky, N., Beveridge, M.C.M., Clay, J.,Folke, C., Lubchenco, J., Mooney, H., Troell, M., 2000. Effect of aquaculture onworld fish supplies. Nature 405, 1017e1024.

Paerl, H.W., 2006. Assessing and managing nutrient-enhanced eutrophication inestuarine and coastal waters: interactive effects of human and climatic per-turbations. Ecol. Eng. 26, 40e54.

Paez-Osuna, F., 2001. The environmental impact of shrimp aquaculture: causes,effects, and mitigating alternatives. Environ. Manag. 28, 131e140.

Paltzat, D.L., Pearce, C.M., Barnes, P.A., McKinley, R.S., 2008. Growth and productionof California sea cucumbers (Parastichopus californicus Stimpson) co-culturedwith suspended Pacific oysters (Crassostrea gigas Thunberg). Aquaculture275, 124e137.

Parnell, A.C., Inger, R., Bearhop, S., Jackson, A.L., 2010. Source partitioning usingstable isotopes: coping with too much variation. PloS One 5, e9672.

Please cite this article in press as: Fu, J., et al., Impacts of sea cucumber faNorthern China, Physics and Chemistry of the Earth (2017), http://dx.doi

Purcell, S.W., Hair, C.A., Mills, D.J., 2012. Sea cucumber culture, farming and searanching in the tropics: progress, problems and opportunities. Aquaculture 368,68e81.

Pusceddu, A., Fraschetti, S., Mirto, S., Holmer, M., Danovaro, R., 2007. Effects ofintensive mariculture on sediment biochemistry. Ecol. Appl. 17, 1366e1378.

Redfield, A.C., 1963. The influence of organisms on the composition of sea-water.The Sea 26e77.

Ren, Y., Dong, S., Qin, C., Wang, F., Tian, X., Gao, Q., 2012. Ecological effects of co-culturing sea cucumber Apostichopus japonicus (Selenka) with scallopChlamys farreri in earthen ponds. Chin. J. Oceanol. Limnol. 30, 71e79.

Ren, Y., Dong, S., Wang, F., Gao, Q., Tian, X., Liu, F., 2010. Sedimentation and sedi-ment characteristics in sea cucumber Apostichopus japonicus (Selenka) cultureponds. Aquac. Res. 42, 14e21.

Rice, W.R., 1989. Analyzing tables of statistical tests. Evolution 43, 223e225.Riera, P., Richard, P., 1996. Isotopic determination of food sources of Crassostrea

gigas along a trophic gradient in the estuarine bay of Marennes-Oleron. Estuar.Coast. Shelf Sci. 42, 347e360.

Sakamaki, T., Shum, J.Y.T., Richardson, J.S., 2010. Watershed effects on chemicalproperties of sediment and primary consumption in estuarine tidal flats:importance of watershed size and food selectivity by macrobenthos. Ecosys-tems 13, 328e337.

Sar�a, G., 2007. A meta-analysis on the ecological effects of aquaculture on the watercolumn: dissolved nutrients. Mar. Environ. Res. 63, 390e408.

Sar�a, G., Scilipoti, D., Mazzola, A., Modica, A., 2004. Effects of fish farming waste tosedimentary and particulate organic matter in a southern Mediterranean area(Gulf of Castellammare, Sicily): a multiple stable isotope study (d13C and d15N).Aquaculture 234, 199e213.

SEO, J.Y., LEE, S.M., 2011. Optimum dietary protein and lipid levels for growth ofjuvenile sea cucumber Apostichopus japonicus. Aquac. Nutr. 17, e56ee61.

Shapiro, S.S., Wilk, M.B., 1965. An analysis of variance test for normality (completesamples). Biometrika 52, 591e611.

Shirley, E., 1977. A non-parametric equivalent of Williams' test for contrastingincreasing dose levels of a treatment. Biometrics 386e389.

Smith, V.H., Tilman, G.D., Nekola, J.C., 1999. Eutrophication: impacts of excessnutrient inputs on freshwater, marine, and terrestrial ecosystems. Environ.Pollut. 100, 179e196.

Sun, Z., Gao, Q., Dong, S., Shin, P.K.S., Wang, F., 2013. Seasonal changes in fooduptake by the sea cucumber Apostichopus japonicus in a farm pond: evidencefrom C and N stable isotopes. J. Ocean Univ. China 12, 160e168.

Tanner, J.E., Fernandes, M., 2010. Environmental effects of yellowtail kingfishaquaculture in South Australia. Aquac. Environ. Interact. 1, 155e165.

Troell, M., Halling, C., Neori, A., Chopin, T., Buschmann, A., Kautsky, N., Yarish, C.,2003. Integrated mariculture: asking the right questions. Aquaculture 226,69e90.

Trott, L.A., McKinnon, A.D., Alongi, D.M., Davidson, A., Burford, M.A., 2004. Carbonand nitrogen processes in a mangrove creek receiving shrimp farm effluent.Estuar. Coast. Shelf Sci. 59, 197e207.

Uede, T., 2008. Validity of acid volatile sulfide as environmental index and exper-iment for fixing its standard value in aquaculture farms along the coast ofWakayama Prefecture, Japan. Nippon. Suisan Gakkaishi 74, 402e411.

Wang, J., Bai, J., Zhao, Q., Lu, Q., Xia, Z., 2016. Five-year Changes in Soil OrganicCarbon and Total Nitrogen in Coastal Wetlands Affected by Flow-sedimentRegulation in a Chinese Delta. Scientific reports 6.

Wai, T.C., Leung, K.M.Y., Wu, R.S.S., Shin, P.K.S., Cheung, S.G., Li, X.Y., Lee, J.H.W., 2011.Stable isotopes as a useful tool for revealing the environmental fate and trophiceffect of open-sea-cage fish farm wastes on marine benthic organisms withdifferent feeding guilds. Mar. Pollut. Bull. 63, 77e85.

Walve, J., Larsson, U., 1999. Carbon, nitrogen and phosphorus stoichiometry ofcrustacean zooplankton in the Baltic Sea: implications for nutrient recycling.J. Plankton Res. 21, 2309e2321.

Wu, H., Peng, R., Yang, Y., He, L., Wang, W., Zheng, T., Lin, G., 2014. Mariculture pondinfluence on mangrove areas in south China: significantly larger nitrogen andphosphorus loadings from sediment wash-out than from tidal water exchange.Aquaculture 426e427, 204e212.

Yang, H.-s., Zhou, Y., Wang, J., Zhang, T., Wang, P., He, Y.-c., Zhang, F.-s., 1999.A modelling estimation of carrying capacities for Chlamys farreri, Laminariajaponica and Apostichopus japonicus in Sishilivan Bay, Yarntai, China. J. Fish. Sci.China/Zhongguo shui chan ke xue yan jiu yuan bian ji chu ban 7, 27e31.

Yearbook, C.F., 2016. China Bureau of Fisheries Beijing. China Agriculture Press,China.

Yokoyama, H., 2003. Environmental quality criteria for fish farms in Japan. Aqua-culture 226, 45e56.

Yokoyama, H., 2010. Monitoring, assessment and management of fish farm envi-ronments in Japan. Rev. Aquac. 2, 154e165.

Yokoyama, H., 2013. Growth and food source of the sea cucumber Apostichopusjaponicus cultured below fish cages d potential for integrated multi-trophicaquaculture. Aquaculture 372e375, 28e38.

Yokoyama, H., Ishihi, Y., 2007. Variation in food sources of the macrobenthos along alandesea transect: a stable isotope study. Mar. Ecol. Prog. Ser. 346, 127e141.

Yokoyama, H., Nishimura, A., Inoue, M., 2007. Macrobenthos as Biological Indicatorsto Assess the Influence of Aquaculture on Japanese Coastal Environments,Ecological and Genetic Implications of Aquaculture Activities. Springer,pp. 407e423.

Yokoyama, H., Tamaki, A., Harada, K., Shimoda, K., Koyama, K., Ishihi, Y., 2005.Variability of diet-tissue isotopic fractionation in estuarine macrobenthos. Mar.

rming on biogeochemical characteristics in the Yellow River estuary,.org/10.1016/j.pce.2016.12.006

Page 12

J. Fu et al. / Physics and Chemistry of the Earth xxx (2017) 1e1212

Ecol. Prog. Ser. 296, 115e128.Zhang, J., Liu, S., Zhang, Q., Liu, Y., 2008. Nutrient distribution and eutrophication

assessment for the adjacent waters of the Yellow River estuary. Mar. Sci. Bull.27, 66e71.

Zhao, Q., Bai, J., Liu, Q., Lu, Q., Gao, Z., Wang, J., 2016. Spatial and seasonal variations

Please cite this article in press as: Fu, J., et al., Impacts of sea cucumber faNorthern China, Physics and Chemistry of the Earth (2017), http://dx.doi

of soil carbon and nitrogen content and stock in a tidal salt marsh with Tamarixchinensis. China. Wetlands 36, 145e152.

Zheng, Z., Dong, S., Tian, X., Wang, F., Gao, Q., Bai, P., 2009. Sediment-water fluxes ofnutrients and dissolved organic carbon in extensive sea cucumber cultureponds. CLEANeSoil, Air, Water 37, 218e224.

rming on biogeochemical characteristics in the Yellow River estuary,.org/10.1016/j.pce.2016.12.006