Comparison of age and growth of grass and silver carp above and below Changzhou Dam in the Pearl River, China

Abstract The life cycles of potamodromous fishes are generally interrupted by the presence of a dam, and populations often display different life history traits between upstream and downstream. This study compared the age and growth rates of grass carp and silver carp above and below Changzhou Dam in the Pearl River. Standard length (SL) composition and back-calculated SL at ages one and two years were greater for grass carp upstream than for downstream individuals; and the length–weight relationship differed between upstream and downstream populations, while the sex ratio, age structure and growth were not significantly different. For silver carp, only SL compositions differed between upstream and downstream populations; sex ratio, age structure, SL-at-age and growth rates were similar. Mean SL-at-age and growth rates of both species were lower at high latitudes than at middle and low latitudes, and populations in the Yangtze River had greater mean SL-at-ages and faster growth relative to silver carp at low latitudes. There were no differences in mean SL or growth among populations at middle and low latitudes. For grass carp, age and growth varied with small-scale ecological processes and regions or latitude, while for silver carp, such variations were proportionate to latitude and the large-scale geographical range. This study provides not only guidelines for fisheries management and conservation of exploited potamodromous fishes, but also insights into the relationship between life history trait variation and adaptive capacity. Highlights Size composition, the relationship between SL and body weight, size-at-age of grass carp were different, and silver carp only had difference in size composition between upstream and downstream of Changzhou Dam in Pearl River, China. Grass and silver carp showed similar age structure, sex ratio and growth rate between upstream and downstream of Changzhou Dam in Pearl River, China. Size at age and growth rate varied with latitudes to some extent and showed variations among populations at middle and low latitudes in China.

• Grass and silver carp showed similar age structure, sex ratio and growth rate between upstream and downstream of changzhou Dam in Pearl River, china.• size at age and growth rate varied with latitudes to some extent and showed variations among populations at middle and low latitudes in china.

Introduction
Most of the major rivers of the world are regulated by numerous dams (Koster et al. 2018).Dams disrupt river connectivity and alter both upstream fluvial habitats and downstream riverine ecosystems (Cheng et al. 2015;Smith et al. 2017).Upstream of a dam, physicochemical variables are altered; lotic rivers become lentic habitats, and suspended sediments settle in the impoundments.Downstream of a dam, the natural river regime is dramatically altered, with decreased water flow and flood frequency, weakened flood magnitude and shorter flood duration (Cheng et al. 2015;Harrison et al. 2016;Enders et al. 2017).Moreover, hypolimnetic discharge generates a lower water temperatures, and sediment retention above the dam results in clearer water and a lowered nutrient level (Song et al. 2018).As a result, fish assemblages will have different species structures above and below a dam, and population structure and biological characteristics will vary between upstream and downstream (Enders et al. 2017;Broaddus and Lamer 2022).Rheophilic and migratory fishes are particularly affected by dam construction and the resulting altered habitats and unnatural flow regime.Rheophilic fish lose preferential lotic habitats and can barely survive in lentic impoundments, thus facing the threat of population decline and extinction.Dams block migration routes, destroy spawning grounds, engender changes in spawning cues, and delay spawning time and recruitment of potamodromous fishes (Young et al. 2011;Cheng et al. 2015;Koster et al. 2018).Populations of potamodromous fish usually decrease rapidly soon after dam construction (Duan et al. 2009).However, some migratory fish species gradually adapt to the new habitats and altered environmental conditions over time and establish stable populations above and below the dam.One population remains in the large impoundment connected to the lotic upriver reaches, whereas another inhabits the lower reaches of the dam, and gene flow reduces due to the obstruction of the dam (Ellender et al. 2016).The decrease of gene flow and differences in habitats and environmental conditions can result in variation in life history traits of the populations above and below the dam (Whiterod et al. 2018).Information on age and growth is valuable in fisheries management and for developing conservation strategies for economically important and rare fish species.Analyzing and comparing age and growth parameters above and below the dam are fundamental for implementing reasonable fishery conservation and restoration strategies for potamodromous fishes (Haworth and Bestgen 2016).
The four major Chinese carp, composed of black carp (Mylopharyngodon piceus), grass carp (Ctenopharyngodon idella), silver carp (Hypophthalmichthys molitrix) and bighead carp (H.nobilis), are economically important native fish species in China.The species are widely distributed throughout China, and there were large fishing catches in central and southern China before the twentieth century (Duan et al. 2009;Li et al. 2016).The four major Chinese carp are typical potamodromous fishes.They spawn in the rivers and tributaries given sufficient flow rates, and the semi-buoyant fertilized eggs drift downstream with currents into slow-flowing and lentic habitats, i.e. lakes, reservoirs, impoundments, lagoons and backwaters, where in the larvae and juveniles grow and develop.During the adult stages, the fish begin to undertake yearly spawning migration toward lotic habitats during spring and summer (Li et al. 2016;Song et al. 2018).Many dams are operating in the large rivers of China, and consequently the major Chinese carp have lost many historical spawning grounds and experienced delays in spawning.The fertilized eggs of the Chinese carp are semi-buoyant and settle to the bottom and suffocate in the slow-flowing waters, and larvae develop and grow in suboptimal conditions, resulting in high larval mortality and low juvenile survival (Ridgway and Bettoli 2017;Song et al. 2018).Fishery resources of the major Chinese carps have dramatically declined in the Yangtze River and Pearl River since the 1980s due to the construction of large-scale hydroelectric power projects and cascaded dams (Duan et al. 2009;Li et al. 2016).Before the twentieth century, the major Chinese carp were important commercial fish in the Pearl River, the largest river in southern China (Lu 1990).There were two important spawning grounds in the middle reaches of the Pearl River to which the carp migrated from the lower reaches for reproduction (Tan et al. 2010).Eleven cascaded dams have been erected in the mainstream of the Pearl River, and the Chinese carp populations have sharply declined.In a 2015 fishery investigation, the landing of grass and silver carp accounted for 2.74% and 2.01% of the total catch, respectively, and black grass carp only contributed 0.29% of the total catch (Shuai et al. 2017).At present, grass and silver carp are found in the middle and lower reaches of the Pearl River and are separated by the Changzhou Dam.The dam, which was completed in 2007, is situated in the end of the middle reaches of the river and the carps above and below the dam may develop into new populations (Tan et al. 2015).The population above the dam may spawn in the middle reaches of the free-flowing river and feed and grow in the impoundment.The populations below the dam may spawn in the lower reaches and drift into the Pearl River Delta for growth and feeding (Shuai et al. 2017;Zhang et al. 2020).Some studies reported that carps can pass upstream and downstream through the dam, but the Changzhou Dam may lessen gene exchange between the two populations, because a few carps make upstream and downstream passage through the dam and there is also potential for high mortality if the eggs/larvae drift through the dam (Yang et al. 2017;Whitty et al. 2022;Turney et al. 2022).Therefore, the two populations above and below the dam may show differences in life history traits.Age and growth of grass and silver carp in the middle and lower reaches of the Pearl River have been studied but have not been compared between upstream and downstream populations (He et al. 2021;Zhu et al. 2021).In the Mississippi River, silver carp were found to have different growth rates and maximum length between the upstream and downstream areas of Dam 19 (Broaddus and Lamer 2022).Understanding the differences in population age structure and growth rates for grass and silver carp upstream and downstream of Changzhou Dam can provide insight into the management and production of the fishery and the effects of the dam on life history traits of the carps.Thus, the objective of this study was to compare population structure, sex ratio, size-at-age and growth rate for grass and silver carp between upstream and downstream populations.Many studies have reported age and growth of grass and silver carp, here, we compared the Pearl River populations to other populations in China and analyze the spatial patterns with latitude and region.

Study area and samples collection
The Pearl River has a length of 2218 km and originates in Eastern Yunnan Province.The river flows through Guangxi and Guangdong Provinces and enters the South China Sea (Figure 1(a)).The middle reaches are from Sanjiangkou to Wuzhou and was divided into Qianjiang and Xunjiang River; the lower reaches, which is called Xijiang River, are confluent with the Bei River and Dong River and form the Pearl River Delta.The Qianjiang is from Sanjiangkou to Guiping and has a length of 122 km; the Xunjiang has a length of 172 km; and the Xijiang River has a length 208 km.The flood season is from May to September during which the runoff account for 80% of the total annual runoff.The middle and lower reaches of the Pearl River had a yearly mean water temperature of 19.4-24.2°C (Lu 1990).
The Changzhou Dam is 350 km upstream from the mouth of the Pearl River and is divided by Chouzhou Island and Sihuazhou Island into three sections.It is a low-head dam with a height of 34.6 m.The Changzhou Dam is the longest dam in the Pearl River, with a length of 3.5 km.It has 43 spillway gates, two hydropower plants, two navigation locks and a fishway.Water depth of impoundment is 20.6 m during the dry season from November to April and 18.6 m during the flood season from May to October.The hydropower plants halt operation completely when water discharge exceeds 16,300 m 3 /s, and water pass through spillway gates; spillway gates are completely removed from the water when water discharge exceeds 21,000 m 3 /s, and the dam is at open-river condition and free-flowing (Zhao et al. 2023).In 2020, 147,000 cargo-vessels pass through the navigation locks of the Changzhou Dam (Ni 2021).It is found that carps made upstream and downstream passages through the spillway gates during open-river condition and through the lock chamber of locks and dams in the Upper Mississippi River (Finger et al. 2020;Turney et al. 2022;Whitty et al. 2022).Hereby, it is reasonably believed that silver and grass carp may pass through spillway gates and navigation locks of the Changzhou Dam.The fishway of the Changzhou Dam is 1.5 km long and 5 m wide.During the spawning periods of 2011-2014, very few silver carps and one grass carp were found to occur in the fishway of the Changzhou Dam (Tan et al. 2015).The Changzhou Dam reservoir has a backwater area of 164 km in length and two important spawning grounds of the Chinese carps is inundated (Tan et al. 2010).
Fish samples were collected during fishing seasons in July and October 2019, July and December 2020, and August 2022.Grass carp and silver carp were purchased from local commercial catches and were captured by gill nets (height 2.5 m × stretched length 100 m, 20, 40, 60 and 100 mm mesh size).Standard length (SL) to the nearest 0.1 cm and body weight (BW) to the nearest 0.1 g were recorded for each fish.The fish were dissected and sexed through visual observation of the gonads.Lapillar otoliths were removed from each fish, cleaned, stored in plastic vials, and allowed to dry for at least two weeks prior to further processing.Lapilli of grass and silver carp have been suggested to provide more reliable age estimates than other hard structures (Ridgway and Bettoli 2017).Therefore, we used lapilli for age determination.

Age determination
After drying was completed, one otolith from each individual was mounted on a glass slide using transparent enamel resin, with the distal surface facing upward.Both sides of the otoliths were ground to sequentially finer grades using carborundum paper (600#-1500#), and then polished with diamond sandpaper (3000#) to enhance clarity.Otolith thin-sections were viewed and photographed through an Oplenic PSC603-07S digital camera attached to an optical microscope (Olympus BX53, Tokyo, Japan) under transmitted light.Similar to Culter alburnus and Hemiculter leucisculus, the otolith cores of grass and silver carp were offset toward the anterior and ventral edges of the lapillus, and the counts and measurements were undertaken along the posterior axis from the core to the posterior margin (Huang et al. 2021(Huang et al. , 2022)).The lapillar core was composed of a black primordium and a wide opaque area that is thought to be formed at larval and juvenile stages (Figure 2, Huang et al. 2022).Outside the core, grass carp had an inconspicuous translucent band and a light opaque band in the otolith, whereas silver carp had a wide translucent band and a narrow opaque band in the otolith.Thereafter, translucent bands and opaque bands were alternately formed.The first annulus was defined as the outer margin of the first opaque band following the core.The number of annuli was determined by counting opaque bands in each otolith.Annulus counts were performed independently twice; when readings disagreed, the otolith was read a third time before a final age was assigned.
Each fish was aged based on the number of annuli, the capture date and the assumed hatching date.Grass and silver carp spawn from April to June; therefore, the assumed hatching date approximately corresponds to the middle of the spawning period (1 May; Lu 1990).All fish were captured after the spawning season and had a translucent otolith edge, so age was equal to the number of opaque bands.Twenty grass carp and nine silver carp were not aged owing to broken and unreadable otoliths, and their ages were estimated using age-length keys (Huang et al. 2022), and they were removed from the growth analysis.

Data analysis
Population structure was examined with SL and age frequency distributions and compared between sampling sites.A two-sample Kolmogorov-Smirnov test was used to determine whether the SL frequency distributions differed between upstream and downstream populations.One hundred and eighty-nine grass carp and 132 silver carp were sexed.The differences in sex ratio (female/male) between upstream and downstream were tested using Chi-squared tests.The relationship between SL and BW was fitted to the power function: BW = aSL b , where a is a scaling constant and b is the allometric growth parameter.Linear regressions of SL-BW data were compared between upstream and downstream using an analysis of covariance (ANCOVA) after log10 transformation of both length and weight.
A decimal age assigned to each individual used for otolith analysis in order to construct a growth model (Zhao et al. 2019).Age was calculated as follows: where D c is the number of days from the date of the average birth (May 1) to the capture, and annuli is the number of opaque bands.Then, SL-at-age data were fitted to the von Bertalanffy growth function (VBGF) using the maximum-likelihood method.The formula was as follows: where L t is the SL at time t, L ∞ is the asymptotic length, k is the growth coefficient and t 0 is the hypothetical age at which SL is equal to zero.An analysis of residual sum of squares (ARSS) was used to compare growth rates between upstream and downstream groups.
The relationship between SL and otolith radius was fitted to a linear regression.R 2 values of the linear regressions were 0.78 and 0.80 for grass carp and 0.77 and 0.71 for silver carp upstream and downstream, respectively, substantiating our assumptions that somatic and otolith growth (R) were proportional.Lapilli are formed before hatching, and SL-at-age was back-calculated using Lee's formula: SL/R = SL n /R n , where SL n and R n are the length and otolith radius at the nth year.SL n was calculated by the modified Lee formula as follows: Back-calculated SL-at-age was compared at each age between upstream and downstream populations using independent-sample t-tests.Back-calculated SL data were tested for normality using Kolmogorov-Smirnov tests prior to the t-tests.
To observe growth rate variation across latitudes, the data of back-calculated SL-at-age of native populations at different latitudes and regions in the Heilongjiang River, Liaohe River, Yellow River and Yangtze River (Figure 1(b)) were obtained from previous studies (Si et al. 2002;Yan et al. 2007;Xiong et al. 2014;Zhang et al. 2015;Chen et al. 2016;Pan et al. 2019;Wang et al. 2020;Peng 2022).Differences in growth rate among populations were compared using an ANCOVA.In this model, the back-calculated SL was the dependent variable; log 10 (age) was an independent variable, and population was a block variable.Differences in growth rates were assessed by examining differences in the population × log 10 (age) interaction term.If growth rates were significantly different among populations, further ANCOVAs were conducted for pairwise comparisons between the regressions.The significance level for these pairwise comparisons was adjusted to p = .05/nbased on the Bonferroni correction, where n is the number of comparisons made.
The data were expressed as mean ± standard deviation (SD).All significance levels were set at p < .05.Statistical analyses were performed using SPSS (version 16.0; SPSS Inc., Chicago, IL), and plots were erected by STATISTICA 10.0 (TIBCO Software, Inc., Palo Alto, CA).

Population size structure
Totals of 92 and 129 grass carp were captured from upstream and downstream sites of Changzhou Dam, respectively.Upstream of the dam, the size range of grass carp was 10.3-53.4cm, with a mean ± SD of 32.96 ± 8.97 cm; downstream of the dam, SL ranged from 12.8 to 61.9 cm, with a mean ± SD of 30.12 ± 8.75 cm (Figure 3).Grass carp was dominated by fish of 30-35-cm SL and of 25-30-cm SL for upstream and downstream sites, respectively.Length-frequency distributions were significantly different (Z = 1.688, p = .007)between upstream and downstream groups.The sex ratios of fish collected above and below the dam were 0.68 and 0.71 and not significantly different between sampling sites (p = .886).The relationship between SL and BW was described by a power function (Figure 4): BW = 0.022SL 2.95 (upstream, r 2 = 0.99, F (2, 127) = 10156.03)and BW = 0.0167SL 3.04 (downstream, r 2 = 0.96, F (2, 87) = 3267.10).The intercept (p = .009,F = 6.88) and slope (p = .026,F = 4.44) of the SL-BW relationship differed significantly between upstream and downstream groups.
Totals of 65 and 112 silver carp were collected from upstream and downstream sites of Changzhou Dam, respectively.Size ranged from 17.62 to 54.00 cm SL, with a mean ± SD of 34.92 ± 6.82 cm for fish above the dam and 7.39-53.10cm SL, with a mean ± SD of 35.81 ± 11.71 cm for individuals below the dam (Figure 3).Length frequency distributions had a mode in the 30-35-cm SL class upstream and in the 45-50-cm SL class downstream and were significantly different (Z = 1.685, p = .007)between upstream and downstream groups.The sex ratios above and below the dam were 0.65 and 0.71 and were not significantly different between sampling sites (p = .779).The relationship between SL and BW was expressed by the following function (Figure 4): BW = 0.011SL 3.13 (upstream, r 2 = 0.97, F (2, 110) = 4380.95)and BW = 0.04SL 2.74 (downstream, r 2 = 0.95, F (2, 63) = 2441.76).The SL-BW relationships were not significantly different with respect to the intercept and slope (p > .05) between upstream and downstream groups.

Age and growth
Upstream of Changzhou Dam, grass carp age estimates ranged from 0 to 4 years, with age 2 being the most abundant (57.1%; Figure 5); downstream of the dam, ages ranged from 1 to 5 years, with the most prominent age class being 2 years (41.9%).There was no difference in age frequency distributions of grass carp between upstream and downstream populations (Z = 0.724, p = .67).
Upstream of Changzhou Dam, the ages of silver carp ranged between 1 and 4 years, with age 2 being the most abundant (52.3%; Figure 5); downstream of the dam, fish were aged between 0 and 5 years, with the dominant age classes of 3 (30.4%)and 2 years (22.3%).Age structure did not differ significantly between upstream and downstream (Z = 1.125, p = .159).
Back-calculated SL-at-age for each age satisfied the normality assumption (p > .05;Table 1).SLs at ages 1 and 2 were significantly greater for grass carp above the dam relative to those below the dam (p < .05).SLs at age 3 did not significantly differ between  upstream and downstream groups (p > .05).Silver carp showed similar SL at ages 1, 2 and 3 between upstream and from downstream populations.

Comparison of mean SL-at-age and growth
Mean SL-at-age for grass carp varied with latitude among four populations (Figure 1(b)).The mean SL at ages 1-5 years in the population from Bositeng Lake at the high latitude was lesser than those of the other three populations from the Yangtze River and Pearl River.Then, mean SL at ages 1-4 years in the Pearl River was lower than those of the upper and lower Yangtze River.Growth rates were not significantly different among populations from Bositeng Lake, the Yangtze River and the Pearl River (F = 0.246, r 2 = 0.964, p = .86;Figure 7).The mean SL-at-age for silver carp showed differences among the six populations (Figure 1(b)).Mean SL-at-age for 1-6 years was greater in the Yangtze River (middle latitude) than in the Heilong River, Songhua River, Biliu Reservoir, Pear River and Dianchi Lake.Populations from Heilong River and Songhua River (high latitudes) had the lowest mean SL-at-ages one and two years.The mean SL at ages 2-4 years was greater in the Pearl River (low latitude) compared to in Biliu Reservoir (relatively high latitude) and Dianchi Lake (low latitude).Growth rates were significantly different among the six populations (F = 6.722, r 2 = 0.983, p = .001;Figure 7).Furthermore, pairwise comparisons revealed that growth rates in the Heilong River and Songhua River were significantly lower than in the other four rivers (p < .5/15).The population from the Yangtze River had a higher growth rate compared to individuals from Dianchi Lake (p < .5/15).Growth rates were not significantly different between populations from the Yangtze River and each of the other populations from Biliu Reservoir, Dianchi Lake and the Pearl River (p > .5/15).

Discussion
The current study found that grass carp showed differences in size structure, the SL-BW relationship and SLs at ages 1 and 2 between upstream and downstream areas of Changzhou Dam, whereas silver carp only exhibited variation in size structure between upstream and downstream areas.Sex ratio, age structure and growth rate for both species were not significantly different between upstream and downstream populations.The silver carp above and below the dam had similar SL-BW relationships and SLs at all ages.These results suggest that grass carp and silver carp displayed different strategies to adapt to different habitats above and below Changzhou Dam.Furthermore, silver carp and grass carp showed mean latitudinal and regional variation in SL-at-age and growth rate for populations in China.
Dam construction generally leads to a decline of potamodromous fishes, but some fishes gradually adapt to new habitats and altered environmental conditions and form stable population (Ellender et al. 2016;Zhao et al. 2019).Since the construction of Changzhou Dam, grass carp and silver carp have experienced the new habitats and environmental conditions above and below the dam for over 10 years.Compared to historical data, longevity measures of grass carp and silver carp in this study were consistent with those of populations in the low reaches of the Pearl River in the 1980s.The age structure of fish in this study was also similar to that of grass carp and silver carp in the 1980s (Lu 1990).This indicates that grass carp and silver carp tend to form stable populations and have adapted well to habitats and environmental conditions above and below the dam.
Longevity and age structure of fish usually vary with latitude and are influenced by fishing pressure (Sharpe and Hendry 2009;Huang et al. 2022).The similar longevity and age structure of grass carp and silver carp above and below the dam are reasonable in light of the same fishing season and the similar latitude.The sex ratio of most fish increases with size, and females tend to be larger than males at the same age (Alves et al. 2019).The low sex ratio for both species upstream and downstream of the dam may be attributed to the high proportion of smaller individuals in these populations.
Although living in different habitats upstream and downstream, the two populations of silver carp had similar SL-at-age and growth rates.Water temperature and prey abundance are the main factors influencing growth rate (Broaddus and Lamer 2022; Huang et al. 2022).During the sampling periods, chemical factors were not different between upstream and downstream areas of Changzhou Dam.Phytoplankton and zooplankton, the main prey of silver carp, are abundant below Changzhou Dam (Xia et al. 2022).Upstream of Changzhou Dam, abundant nutrients in the impoundments are favorable for productivity of plankton, and thus food resources of silver carp should be sufficient (Song et al. 2018;Zhao et al. 2019).Therefore, similar environmental factors and abundant food resources explain the invariable SL-at-age and growth rate of silver carp above and below Changzhou Dam.However, other studies have found spatial variation in life history traits of silver carp.The invasive populations of silver carp in the three tributaries of the Missouri River had the highest growth rates at the confluence sites compared to sites in the upper and middle reaches of each river (Hayer et al. 2014).In the Mississippi River, length-frequency distributions and mean length-at-age after age five were different between upstream and downstream areas of the pool and Dam 19 due to lower fish density above the dam (Broaddus and Lamer 2022).Herein, although silver carp had a lower density upstream than downstream (Shuai et al. 2017), the low density seemed to have little influence on the variations in the age structure or growth between upstream and downstream populations.
Grass carp had similar growth rates above and below Changzhou Dam, but SLs at ages 1 and 2 were significantly different between the two populations.The different SLs at age 1 and 2 can explain variation in SL composition between upstream and downstream populations, because age structure was not significantly different.Similarly, the long-lived shovelnose sturgeon in the Williston-Yellowstone had larger fork-length for the first 10 years than those in the Bismarck segment (Everett et al. 2003).Siniperca kneri had a faster growth at the first year in the Three Gorges Dam relative to other populations at similar latitudes (Zhao et al. 2019).Upstream of the dam, grass carp had a lower density that may have mitigated intraspecific competition (Shuai et al. 2017).Furthermore, larvae and juveniles of grass carp would grow and develop in the impoundment of Changzhou Dam and experience better feeding conditions.Consequently, grass carp upstream displayed more rapid growth for the first two years (Zhao et al. 2019;Broaddus and Lamer 2022).Some fish species in the scope of small scales also display distinct history life traits due to different environmental variables.Due to different water temperatures, Pseudogobio esocinus at the upper and lower reaches of the Nakagawa River with a length of 35 km showed different SL-BW relationships, age group distributions, condition factor values and growth rates (Nakajima and Onikura 2016).Sarotherodon melanotheron in Nakwa lagoon and Brenu lagoon in Ghana experienced variable salinity and suffered from different fishing pressure and thus showed differences in size structure, the length-weight relationship, the condition factor and growth performance (Zuh et al. 2019).However, there are no data concerning the distribution and feeding habitats of grass carp larvae and juveniles above Changzhou Dam, and thus the reasons for the faster growth at ages 1 and 2 need further analysis and verification.
The spatial patterns of age and growth between upstream and downstream of Changzhou Dam were not the same for grass carp and silver carp.Silver carp had very similar population structure and growth rate above and below Changzhou Dam, and thus could be treated as a single management unit.Grass carp showed variation in size structure, weight-at-length and SL-at-ages 1 and 2 between upstream and downstream samples.This indicates that there is species-specific variation in life history traits when potamodromous fish adapt to newly formed habitats caused by the construction of dams.These differences may determine their invasive ability and population size.Both species have successfully invaded the US, silver carp have reached extremely high populations, whereas grass carp are abundant in some locations and never reached anywhere near the abundances reached by silver carp (Ridgway and Bettoli 2017;Broaddus and Lamer 2022).Upstream of Changzhou Dam, grass carp and silver carp may have different migratory distances and stay in different micro-habitats, and thus may experience different environmental conditions.Additionally, compared to grass carp, silver carp is more likely to make upstream and downstream passage through the Changzhou Dam, and silver carp above and below the dam may have some mixture (Tan et al. 2015;Turney et al. 2022).As a result, age and growth of grass carp and silver carp showed different degrees of change upstream relative to downstream.This situation has also been found for other fish species such as the Longnose sucker, Shorthead Redhorse and Silver Redhorse at upstream and downstream sites of the Generating Station in the Saskatchewan River, Canada.There was variation in age and growth among three species of Pelteobagrus in lotic, transitional and lentic zones in the Three Gorges Reservoir, China, with changes in reproductive characteristics (Enders et al. 2017;Liao et al. 2018).
Life history traits generally vary with geographical area and latitude (McNicholl et al. 2018;Wright et al. 2020).In this study, SL-at-age and growth of grass carp and silver carp showed latitudinal and regional changes.SL-at-age of grass carp and silver carp at high latitude (Heilong River and Songhua River) were lesser than those at middle and low latitudes.The growth rate of silver carp was lower at high latitudes than at middle and low latitudes.SL-at-age was greater for grass carp and silver carp in the Yangtze River than in the Pearl River and Dianchi Lake.These results suggest that SL-at-age of grass carp and silver carp and growth of silver carp adhere to Bergmann's rule to some extent, i.e. populations at higher latitude grow more slowly than those at lower latitude (Nakajima and Onikura 2016;McNicholl et al. 2018).The smaller SL-at-age and slower growth of grass carp and silver carp at high latitude (in the Heilong River and Songhua River) are related to lower water temperature and a shorter period of growth in northern China (Huang et al. 2022).However, the larger SL-at-age and the faster growth of grass carp and silver carp for the first five years at the middle latitude relative to the low latitude were opposite to the rule.Most populations of grass carp and silver carp have similar growth rates among middle and low latitudes.The spatial patterns indicate that instead of water temperature, prey availability and other environmental pressures (e.g.intraspecific competition and fishing pressure) are the main factors influencing the growth of grass and silver carp at later juvenile and early adult stage at middle and low latitudes (Adams et al. 2018;McNicholl et al. 2018).Water temperature and prey availability at middle latitudes may ensure relatively rapid growth of grass carp and silver carp during the growing season (summer and autumn), which may offset a long period of growth at low latitudes.The lack of a relationship between growth and latitude is not uncommon for freshwater fish.For example, the growth rate, condition and mortality rate of golden perch were similar between the northern and southern Murray-Darling Basin regions; Hemiculter leucisculus showed invariant growth rate and mean SL-at-age throughout the distribution range in China (Wright et al. 2020;Huang et al. 2022).The mechanisms underlying these phenomena are complex and are related to fish species and regional environmental conditions (Adams et al. 2018;Wright et al. 2020).
In conclusion, we have documented population structure, mean SL-at-ages and growth rates of grass carp and silver carp above and below Changzhou Dam, and have analyzed the variation with latitude and region.Grass carp showed differences in some of these life history traits between upstream and downstream populations, and mean SL-at-age varied with latitude to some degree, whereas silver carp had almost no differences in age or growth between upstream and downstream areas but showed variation in SL-at-age and growth with latitude and among regions.Age and growth of grass carp changed with regional and small-scale ecological processes (e.g. the upper and low reaches of the same river) or large-scale environmental processes (e.g.latitude).Silver carp showed growth changes with latitude and large-scale geographical range.The different responses to small-scale and large-scale geographical range may suggest the distribution and invasive ability of grass carp and silver carp.These results provide some guidelines for fisheries management and conservation of the commercially exploited fish that grass carp should be regarded as two groups and silver carp is a group in the middle and lower reaches of the Pearl River.The comparison may have additional value in the management and understanding of elements controlling the growth of these important fishes in other locations.

Figure 3 .
Figure 3. standard length (sl, cm) frequency distributions of grass carp and silver carp collected upstream and downstream of changzhou Dam.

Figure 4 .
Figure 4. the relationships of standard length (sl, cm) and body weight (Bw, g) of grass carp (a) and silver carp (b) collected upstream and downstream of changzhou Dam.

Figure 5 .
Figure 5. age compositions of grass carp and silver carp collected upstream and downstream of changzhou Dam.

Figure 6 .
Figure 6. the relationship between age (years) and standard length (sl, cm) of grass carp (a) and silver carp (b) fitted to von Bertalanffy growth function at upstream and downstream areas of changzhou Dam.

Figure 7 .
Figure 7. estimated standard length (sl) at age of the populations for grass carp (a) in Bositeng lake, the upper and lower yangtze river, and the Pearl river; and for silver carp (b) in heilong river, songhua river, Biliu reservoir, the yangtze river, Dianchi lake and the Pearl river in this study.

Table 1 .
Mean ± standard deviation (sD) and range of back-calculated standard length (cm) at each age class for grass carp and silver carp collected downstream and upstream of changzhou Dam.