LI De-Liang， ZHANG Jian， PI Jie， GAO Zi-Han， XIAO Tiao-Yi and CHEN Yu-Shun

(1. Collaborative Innovation Center for Efficient and Health Production of Fisheries in Hunan Province， Hunan Engineering Technology Research Center of Featured Aquatic Resources Utilization， College of Animal Science and Technology， Hunan Agricultural University， Changsha 410128， China; 2. State Key Lab for Freshwater Ecology and Biotechnology， Institute of Hydrobiology， Chinese Academy of Sciences， Wuhan 430072， China; 3. University of Chinese Academy of Sciences，Beijing 100049，China)

Abstract: Samples of Corbicula fluminea (Müller， 1774) were collected monthly from March 2011 to February 2012 in the Datong Lake to study its population dynamics and secondary production in its native environment. A total of 4，108 C. fluminea were collected and measured， with shell length (SL) ranging from 3.41 to 29.58 mm， and wet weight (WW) ranged from 0.005 to 10.03 g. The calculated SL-WW equation was log WW = –3.52+3.08 log SL (R2=0.97)， which showed that the growth curve of C. fluminea followed a positive allometric pattern. The recruitment patterns peaked in one from March to April and the other one during the months of August to October. The mean annual abundance and biomass were (274±95) ind./m2 and(20.1±5.7) g AFDW/m2 (Ash Free Dry Weight， AFDW)， respectively. The annual secondary production (P)was estimated to be 23.90 g AFDW/m2·year， resulting in a ratio of 1.20/year and a turnover time of 298 days. The asymptotic maximum length (L∞)， curvature parameter (K) and seasonal oscillation in growth rate(C) derived from the von Bertalanffy Seasonal Growth Formula (VBSGF) were 31.91 mm， 0.45/year， and 0.84， respectively. The slowest growth period was in January， and the potential lifespan (tmax) was estimated to be 4.44 years. The total mortality (Z)， natural mortality (M) and fishing mortality rates (F) estimated were 1.68/year， 0.89/year and 0.79/year， respectively. Its exploitation rate (E=0.47) indicated a slightly high exploited stock that needs some management intervention to maintain the sustainability of this fishery resource.

Key words: Corbicula fluminea; Population dynamics; Secondary Production; Growth; Mortality;Datong Lake

Corbicula fluminea(Müller， 1774) is a freshwater bivalve mollusk， confined to its extant native distribution in Asia， the Middle East， Africa and Australia at the beginning of the 20th century[1]. In the past 80 years，C. flumineahas extended its range rapidly to American and European ecosystems by a combination of human and natural dispersion mechanisms[2—4]. It was first recorded outside its native range in 1924 in Vancouver Island， British Columbia[5].Since then， it has become established in most major rivers of the continents. Nowadays，C. flumineahas spread almost all over the world and been considered as one of the most important non-indigenous invasive species[6，7]. In many areas where introduced，C. flumineahas caused great ecological and economic impacts[4，7]. Ecological impacts ofC. flumineainvasion include altering nutrient dynamics and competing with other filter-feeding bivalves[6]. Economic impacts mostly come from macrofouling of water conduction systems in fossil-fueled or nuclear power station， and enhancement of sedimentation rates in irrigation channels[1]. Consequently， the invasion， dispersal trajectory and potential control methods ofC. flumineahave been the core of studies dealing with biological invasions in aquatic ecosystems[8].

In China，C. flumineais not only an economically important food resource but also explored for medicine with its pharmacological activity from ancient time[9]. Datong Lake is a typical native lake ofC. fluminain the central Yangtze River Basin， and the largest lake for aquaculture in Hunan Province，China， with its lake area of 82.67 km2[10].Corbicula flumineain the Datong Lake has increased gradually from 1990s onward when fertilization technology in aquaculture was used[11]. Now it reaches considerable mean abundance (502±67) ind./m2and biomass(447.2±48.7) g/m2[12]. In recent years， the harvest ofC. flumineain lakes in the Yangtze River basin such as the Datong Lake and Hongze Lake， has been enhanced due to its economic benefits[12，13]. It is essential to develop studies on its population dynamics and secondary production characteristics to effectively manage this important aquatic resource[9].

The present study was designed to investigate the population dynamics and secondary production ofC. flumineain its native lake (Datong) in the central Yangtze River Basin， China. These results will provide us valuable information to effective resource management ofC. flumineaand better understanding of the ecology of this invasive species in its native environment.

1 Materials and Methods

1.1 Study area

The Datong Lake (29°11′N(xiāo)—29°15′N(xiāo)， 112°26′E—112°33′E) is located in the Central Yangtze River Basin， and was separated from the Dongting Lake since the land reclamation in 1951 (Fig. 1)[10]. The lake water is slightly alkaline (pH=7.7—9.6) with higher reducibility [(–88.40±9.10) mv]， and dissolved oxygen values are always around saturation levels (e.g.， 7.3—13.5 mg/L). Annual mean water temperature of the lake is around 20.3℃， usually ranging from 9.4 to 33.9℃ over a year. Water depth of the lake is varied from 138 to 250 cm. Bottom substrates in the lake were primarily dominated by silt with minor amounts of sand and rarely fine gravel.Over the past 60 years， the Datong Lake has transited the fishery forms from natural proliferation to artificial culture， and intensive aquaculture was enhanced gradually since 1990s when fertilization technology was used[10]. Now， the Datong Lake has transformed from a submerged grass-based to an algae-based lake.The annual mean cell density of the phytoplankton in the lake was 1.84×106cells/L， being the highest in summer (16.4×106cells/L) and ranged from 1.71×106cells/L to 1.98×106cells/L in other three seasons[11].

1.2 Sampling and laboratory procedures

Fig. 1 General geographic setting and sampling sites of the Datong Lake

Monthly benthic samples were collected late each month from ten sites in the Datong Lake from March 2011 to February 2012 (Fig. 1). In order to reduce the potential effects of aquaculture activity on distribution ofC. fluminea， the sampling sites were selected from the open water areas where similar aquaculture species and density exist. At each site，sediments were collected twice using a Peterson grab sampler with a sampling area of 0.0625 m2. The collected benthic materials were mixed and sieved through a 450-μm-mesh screen. Shell length (SL) of allC. flumineaindividuals were measured with a digital Vernier caliper to the nearest 0.01 mm， and wet weight with shell (WW) were measured to the nearest 0.1 g. Then， theC. flumineaspecimens sampled were separated and divided into different size classes for analysis.

Corbicula. flumineabiomass was expressed as grams of ash free dry weight (AFDW) converted based on an allometric equation betweenWWandAFDW. The relationship betweenWWandAFDWwas calculated and was given by the following equation:log (AFDW)=0.9833×logW(g)–1.2924 (n=192，R2=0.96， range ofWW=0.0209–8.8334 g). This equation was estimated from a set of 192 individuals，which were dried at 60℃ until stable weight reached for dry weight estimations and followed by combustion at 550℃ for 4h forAFDWestimations[14].

1.3 Secondary Production

Annual secondary production estimates were based upon the size-frequency method， thus individuals were grouped into 2 mm size classes for the initial production calculation. This method uses the average size-frequency distribution to sum total losses between size groups based on samples taken throughout the year and followed the procedure[15]:

wherePis the annual production;nis the number of size classes;Njis the mean number of individuals in size classj;Wjis the mean weight of individuals in size classj;CPIis cohort production interval in days from hatching to the attainment of the largest size class. Because shelled veligers are shed directly and life span (tmax) was calculated to be 4.44 years (see below)， a cohort production interval (CPI) correction was made based onCPI=1620 days.

1.4 Growth, lifespan and mortality

In order to estimate the growth parameters ofC.flumineain the Datong Lake， monthly shell length distributions were grouped into 1 mm size classes and analyzed with the aid of the ELEFAN I’ (Electronic Length Frequency Analysis) routine from the FISAT package (FAO-ICLARM Stock Assessment Tools).The growth formula used for estimating growth parameters is the von Bertalanffy model with seasonal oscillation， the von Bertalanffy Seasonal Growth Formula (VBSGF)[16]is:

whereLtis the length (mm) at a given timet;L∞is the asymptotic maximum length (mm);K(/year) is the curvature parameter;Cis a constant defining the degree of seasonal oscillation [ranging from 0 indicating continuous non-oscillating growth， to 1 when growth comes to a complete halt at the ‘winter point’(WP)];t0is the theoretical age at zero length (year)andtsis the initial point of oscillation in relation tot=0 and theWP. The different subroutines of ELEFAN I (K-scanning) were used to identify the VBSGF that best fitted the monthly shell length data， using theRnvalue as a criterion of fit[17].

The theoretical lifespan (tmax) was estimated by an inverse of the VBSGF， considering maximum shell length as 95% of the asymptotic length[18]:

whereL95%represent 95% of the maximum shell length recorded during field sampling.

Total mortality (Z) was calculated from lengthconverted catch curve[19]yielded by the ELEFAN I routine of the FISAT II program[20].Zwas estimated by:

whereNis the number of individuals;gis the regression intercept，Z(/year) is the unbiased mortality estimate; andtis the estimated age (year) for each cohort[19].

The natural instantaneous mortality rate (M) was estimated using the empirical relationship defined by Pauly[21]:

whereTis the mean annual lake water temperature andL∞is the maximum asymptotic length (mm) thatC. flumineacan reach. The mean annual lake water temperature was estimated to be 20.1℃[11].

The fishing mortality rate (F) was calculated as:

The exploitation rate (E) was calculated as:

2 Results

2.1 Population size and recruitment

A total of 4，108C. flumineawere collected and measured， withSLranging from 3.41 to 29.58 mm，andWWranged from 0.005 to 10.03 g. The calculatedSL-WWequation was logWW= –3.52+3.08 logSL(R2=0.97). The linear regression showed a significant relationship betweenSLandWW(P＜0.05). The morphometric relationship betweenWW/SL(b=3.08)indicated positive allometric growth. There were two major recruitment peaks observed during the study period: one from March to April and the other one occurring during the months of August-October when most of the young clams (i.e.， length＜6 mm) were observed (Fig. 2).

2.2 Abundance, biomass and secondary production

The monthly mean abundance (±SE) ofC. fluminearanged from (150±36) ind./m2(May 2011) to(367±151) ind./m2(March 2011) with an annual mean of (274±95) ind./m2(Fig. 3). Initial abundance in March was high， decreased gradually until May 2011，and then increased rapidly until July 2011. Monthly mean biomass ofC. fluminearanged from (11.2±2.0) gAFDW/m2(May 2011) to (27.4±7.9) gAFDW/m2(February 2012) with an annual mean of (20.1±5.7) gAFDW/m2(Fig. 3).

Fig. 2 Monthly shell length distributions of C. fluminea in the Datong Lake from March 2011 to February 2012 (N=number of specimens)

The annual secondary production ofC. flumineaestimated by size-frequency method was 23.90 gAFDW/m2·year and the mean annual biomass was estimated to be 19.52 gAFDW/m2， which resulted in aratio of 1.20/year and a turnover time of 298 days(Tab. 1).

2.3 Growth, lifespan and mortality

Seasonal growth curve estimated by the VBSGF analysis forC. flumineahad an asymptotic maximum length (L∞) of 31.91 mm and a growth rate (K) of 0.45/year (Fig. 4). The relative amplitude of seasonal oscillation in growth rate wasC=0.84， and the winter point wasWP=0.09. Thetmaxwas estimated to be 4.44 years.Zwas estimated to be 1.68/year， whereasMwas 0.89/year andFwas 0.79/year.Ewas estimated to be 0.47.

Fig. 3 Monthly variations of C. fluminea mean abundance (±SE)(ind./m2) and biomass (±SE) (g AFDW/m2) from March 2011 to February 2012 in the Datong Lake

3 Discussion

3.1 Population size and recruitment

The allometric coefficientb(3.08) confirmed that theC. flumineapopulation in the Datong Lake has a positive allometric pattern. Similar exponential values were also reported in the Taihu Lake， China(b=3.183)[22]. However， an isometric pattern for individuals withSLlarger than 6 mm was observed in the Datong Lake in previous study (b=3.00)[12]. The discrepancies in the value ofbinSL–WWrelationships could be attributed to the differentSLranges of samples analyzed. In addition， the value ofbis also related to the environmental conditions， food availability， size range， as well as growth process[23]. For the growth process ofC. flumineafrom juveniles to adults， their growth pattern can be summarized as: allometric growth at the beginning， then an isometric growth， and then allometric growth again[22].

Tab. 1 Estimation of secondary production of C. fluminea in the Datong Lake by the size-frequency method. Biomass is expressed as ash free dry weight

The results reported in the literature suggest that the number of reproductive events ofC. flumineavaries each year (Tab. 3). Most of theC. flumineapopulations of Asia and North America have two reproduction periods during the year， one in spring and the other in autumn， and have large reductions or total in-terruption of juvenile release in the summer and winter[24—27]. In contrast， some studies found thatC.flumineahad only one reproductive period， while three were found in others， with differences among years even at the same site[28，29]. The presence of two main recruitment peaks for the population in the Datong Lake was in agreement with reported existence of the two reproductive periods. However， distinct cohorts were difficult to be discriminated in the length–frequency distributions in the present study(Fig. 2).

3.2 Abundance, biomass and secondary production

The peak abundance observed in March 2011[(367±151) ind./m2] was directly associated with the recruitment period of the generation born. The lowest abundance was in May 2011 [(150±36) ind./m2]， accompanied by the sufficient death assemblages (measured by the available empty shells)， for which may be the part reason that the present results were lower than that in 2010 and 2011[12]. Overall， annual mean abundance was relatively high contrasting with manyC. flumineapopulations in the middle and lower reaches of the Yangtze River Basin， such as the Poyang Lake[28]， Chaohu Lake[30]， and Hongze Lake[28](Tab. 2). Cai，et al.[22]have found that the abundance ofC. flumineawas similar to ours in the Tai Lake， averaging about 266 ind./m2over about 2427.8 km2. Biomass was almost constant all year around， varying from (11.2±2.0) to (27.4±7.9) gAFDW/m2， with low values in spring months associated with the high mortality mentioned above. Nonetheless， secondary production ofC. flumineafrom the Datong Lake (23.9 gAFDW/m2) was much higher than those reported in a few other systems (Tab. 2)， but lower than that in the River Minho estuary， Portugal[6](Tab. 3). The fertilization technology used in the Datong Lake which has provided enough available food for theC. flumineamay be one of the most important factors responsible for the high annual mean abundance and secondary production. Meanwhile， high production rate partially implies an important role ofC. flumineain organic matter and nutrient cycling of the lake ecosystem.Previous study showed that smaller， younger individuals are more likely to migrate than larger， older clams[31]， presumably to more acceptable microhabitats after having been redistributed during hydrologic events. Thus， the slight low annualvalues estimated in the present study might be most likely influenced by a relatively low fraction of juvenile clams，i.e.， larger， older individuals may have been oversampled after they were collected and remained at specific sites.

3.3 Growth, lifespan and mortality

Fig. 4 Seasonal growth curve estimated by the VBSGF analysis from monthly shell length data throughout the sampling period in the Datong Lake

Tab. 2 Standing crops of C. fluminea in lakes in the middle and lower reaches of the Yangtze River， China

There were wide variations in the population parameters estimated， which was attributed to different environmental conditions. In the present study，L∞(31.91 mm) is similar to that of previous studies(Tab. 3). The highest reportedL∞(49.8 mmSL) was observed in the River Minho estuary， Iberian Peninsula[6].Corbicula flumineaexhibited a slower growth rate (K=0.45/year) in the current study compared withK=0.72 year in the Hongze Lake， China[13]，K=0.68/year in the River Minho estuary， Iberian Peninsula[6]andK=0.65/year in the Paraná River Delta， Argentina[32].The growth dynamics ofC. flumineaare mainly determined by the amount of available food， pH/calcium value and ambient water temperature[33].Corbicula flumineaseems to grow well only in a fairly narrow range of temperatures (18—25℃)[34]， and some studies ofC. flumineahave shown that maximum growth rate usually occurs between 20 and 25℃， and growth rate declines rapidly as water temperature falls below 15℃[33]. In the Datong Lake， water temperatures fall below 15℃ for roughly 4 months (from November to February)， and they are consistently above 26℃ for another 3 months (from June to August)[35]. Therefore， the low growth rates ofC. flumineain the Datong Lake during the midsummer and the lack of growth in the midwinter were probably due to water temperature. As the bottom substrates were predominantly silt with minor amounts of sand and rarely fine gravel in the Datong Lake， oxygen content in sediment may be limited as well. Thus， it seems that water temperatures and oxygen content in sediment were the most likely factors involved in creating a stressful environment forC. flumineain the current study. Food quality and quantity may not play a similarly important role as water temperature does in limitingC. flumineain the Datong Lake， because primary production in the lake were kept at high levels throughout the year[11].Corbicula flumineaexhibited seasonal growth (C=0.84)， with very significant summer-to-winter differences in growth rates.WPwas 0.09-year units， which implied that the lowest growth rate occurred in January (0.09×12=1.08 months).

The most commonly used methods for age analysis of bivalve species are based on analysis of external surface rings， internal growth lines and microgrowth bands in shells[36]. In addition， the approximate lifespan of bivalve species can be estimated on the basis of VBSGF parameters[18]. In the Datong Lake，C. flumineagrowth was continuous throughout its life cycle and its lifespan was calculated as 4.44 years， which is similar to some reported populations[32]，whereas some other populations have shorter[23，25]or longer[2，3]lifespans than that in the current study.These differences are probably related to latitude，temperature， available food resources， and other factors[6]. It should be noted that the estimation of lifespan based on the inverse VBSGF and the maximum length of the largest individual could be overestimated， which implies very likely selecting the fastest growing individual， but not necessarily the oldest one[37].

Tab. 3 Comparisons of population parameters for C. fluminea in the Datong Lake and other regions

The fishing mortality (F) was less (0.79/year)compared to natural mortality (M， 0.89/ year)， indicating a balanced stock ofC. flumineain the Datong Lake. Estimating of the currentEwas one of the main approaches used to evaluate stock status[42]. Relative to the limit management reference point (E=0.4)，which was recommended to be consistent with high long-term yields[43]，C. flumineastock in the Datong Lake may be considered as being exploited slightly higher (E=0.47). Hence， there is an urgent need for regulatory measures aiming at optimizing yield to avert the potential decline of this fishery resource in the area.

4 Conclusions

This study investigated the population dynamics and secondary production ofC. flumineain a typical aquacultural lake (Datong) of the Central Yangtze River Basin， China. We found that the stock ofC. flumineaanalyzed is currently at a slightly higher exploitation level than it should be. Results from this study also provided basic information that may facilitate conservation and stock management policies forC. flumineain the area.