Assessment of Stock Enhancement Efficacy for Hypophthalmichthys molitrix and Aristichthys nobilis in the Xixi of Jiulong River Basin

Abstract

Stocking and replenishing fish are crucial for the ecological restoration of aquatic biological resources. Since 2017, a long-term stocking program of Hypophthalmichthys molitrix and Aristichthys nobilis has been underway in the Xixi River basin of the Jiulong River. To understand the status of fishery resources following this long-term stocking program, field surveys were conducted every two months from October 2023 to October 2024. Traditional netting, resource assessment and environmental DNA (eDNA) analysis methods were used to conduct a comprehensive assessment of resource abundance, stocking contribution and ecological adaptability. The research revealed that the annual survival rates for H. molitrix and A. nobilis were 40.25% and 48.19%, respectively. The current numerical ratio of H. molitrix to A. nobilis stands at 1.97:1, indicating that the survival number of H. molitrix is better than that of A. nobilis. No mature gonads were observed in any sampled individuals, demonstrating that the current population is highly dependent on artificial replenishment. This study provides valuable data support for aquatic resource restoration and ecological management in the Jiulong River Basin.

1. Introduction

To mitigate the decline of fishery resources, augment recruitment, and improve aquatic ecosystems, various fishery conservation measures have been implemented. [1,2,3]. Among these, stock enhancement [4] has emerged as a primary technical approach for increasing fishery resource abundance and restoring aquatic environments. Stock enhancement constitutes a complex systematic project. Its principal objectives include: increasing fishers’ income, restoring depleted populations, purifying water bodies, protecting endemic fish species, developing recreational fisheries and maintaining biodiversity [5]. This practice involves the staged, continuous release of live aquatic organisms—such as fry, fingerlings, and broodstock—into public waters (rivers, seas, lakes, etc.) harboring conspecific native populations. Releases are conducted through various methods including transplantation, introduction, and direct stocking, with the ultimate goals of increasing population size and improving environmental conditions [6,7]. Currently, the actual effectiveness of stock enhancement initiatives and their potential ecological impacts are receiving increasing attention. In practice, issues such as improper selection of released species, mismatch between the scale of release and the carrying capacity of habitats, poor quality of released individuals, and lack of effective genetic management and adaptability assessment may lead to suboptimal stocking outcomes, limited restoration of target populations, and even potential ecological risks. These risks include genetic introgression, disease transmission, and disruption of the structure and function of native ecosystems [8,9,10]. Assessment of stock enhancement efficacy constitutes the core component of resource augmentation and provides the scientific foundation for subsequent enhancement planning and implementation strategies [11,12]. Scientific evaluation not only quantifies immediate fishery yield increases but also reveals the potential of released cohorts as renewable “residual stocks” and their sustained contribution to future harvests [12]. Successful stock enhancement generates continuous ecological benefits through recurring fishery production that supports river ecosystem restoration. Consequently, comprehensive assessment of stock enhancement benefits requires establishing a multidimensional indicator system, typically encompassing economic benefits (e.g., catch increment), ecological benefits (e.g., population structure optimization, ecosystem function recovery), and social benefits (e.g., fisher livelihood improvement, public satisfaction enhancement) [12,13]. Scientific and comprehensive assessment of the integrated outcomes of stock enhancement, particularly the restoration of ecological efficacy, has emerged as a current research priority.

Traditional research methods for assessing the recovery of ecological benefits from stock enhancement include “mark-release-recapture,” ecological survey monitoring, and socio-economic evaluation. For example: Zhang et al. [14] successfully monitored the growth of Liza haematocheila juveniles using otolith marking. Riley et al. [15] effectively monitored salmonids using Passive Integrated Transponder (PIT) tag technology. Tang et al. [16] achieved successful results marking Spinibarbus sinensis Bleeker using fluorescent marking. However, these traditional marking methods and direct fishing methods have disadvantages such as being time-consuming and labor-intensive, requiring large economic investment, causing physical damage, and interfering with population dynamics [17]. With advancements in modern biotechnology and molecular genetics, genetic markers and pedigree analysis have been applied to monitor stock-enhanced populations [18]. Another technique is environmental DNA (eDNA), which refers to DNA that organisms shed into the surrounding environment, from sources such as shed tissues or cells, metabolic waste, and decaying corpses. This technology collects these substances without capturing or processing the organisms themselves [19,20].

The eDNA technique involves extracting specific DNA fragments from environmental samples (such as water, soil, and sediment), followed by PCR amplification and high-throughput sequencing. This process enables the identification of species present and estimation of their abundance within the sampled environment, achieving the goal of species monitoring [21,22]. This technology has gradually been applied to monitor various aquatic organisms, including fish [23,24,25]. Some studies have employed eDNA to analyze fish diversity, demonstrating the feasibility of eDNA technology in resource monitoring [26,27,28]. This study employs eDNA as a simple, non-invasive method to gather additional information. eDNA samples were collected from release areas using the same sampling locations as fish catches. Relative quantitative analysis was performed on the number of sequences obtained via eDNA technology. Data from the eDNA method was used to supplement traditional analyses, comparing it with biomass from traditionally caught species to understand the distribution of dominant species. This approach was integrated with mortality rates and contribution rates derived from traditional analyses for comprehensive evaluation. Compared to traditional fishery resource surveys, eDNA technology offers simpler, more efficient methods with broader analytical dimensions, while remaining non-destructive to sampling sites and fish populations [29]. Consequently, eDNA technology holds broad development prospects for applications including biodiversity assessment, species monitoring and conservation, stock enhancement assessment, ecosystem health assessment, and water quality monitoring [30].

The Jiulong River is an important ecological waterway in southern Fujian Province. It has been designated by the state as a key mangrove biodiversity protection area and a Class III water conservation functional area on the southeast coast. Its complex estuarine environment provides spawning and nursery grounds for numerous fish species, playing an indispensable role in maintaining biodiversity. Simultaneously, it functions as a major fishing base for areas including Xiamen and Longhai, boasting abundant fish resources. The Xixi River is the largest tributary of the Jiulong River, with a watershed area of approximately 3940 km² and a total river length of 172 km, accounting for about one-quarter of the entire watershed. It serves as the primary drinking water source and fishery resource reservoir for three cities [31]. According to the relevant information released by the Fujian Provincial Bureau of Ocean and Fisheries, with the surrounding economic development, reservoir construction, hydropower projects, and overfishing, the river has undergone severe “lake-like transformation” and water quality deterioration [32]. This has altered the integrity of the river ecosystem and its biodiversity, placing pressure on fishery resources. Some economically important species, such as Konosirus punctatus and Mugil cephalus, are already overexploited. The catch is mainly smaller fish and juveniles, and the resource amount is declining [33,34]. Therefore, from 2017 to 2024, the Zhangzhou Municipal Bureau of Ocean and Fisheries has released nearly 3 million H. molitrix and A. nobilis into the Xixi River basin within its jurisdiction to replenish fishery resources. And comprising 2.0475 million H. molitrix and 0.6825 million A. nobilis at a 3:1 ratio.

H. molitrix and A. nobilis are both members of the Cyprinidae family within the Cypriniformes order. They are typical semi-anadromous fish species [35] that lay drifting eggs and have a long lifespan (7–10 years) [36,37,38]. They possess advantages such as fast growth rates, strong adaptability, high breeding output, and a typical filter-feeding diet [39,40], enabling them to increase surface fishery yields within a short cycle and improve the economic income of surrounding fishermen. Therefore, they are often selected as stocking fish species in efforts to improve aquatic ecosystems and replenish biological resources. However, their reproduction is limited and influenced by factors such as water temperature, dissolved oxygen levels, and hydrological conditions. They require the stimulation of flood peaks caused by river water fluctuations to produce eggs [41]. Additionally, the eggs produced need a certain water flow velocity to remain suspended in the water and drift downstream for hatching. Sections of the river with swift currents, large fluctuations in flow velocity gradients, and complex flow patterns are the optimal spawning sites for mature adult fish [42,43]. The construction of hydroelectric power plants and flood control projects in the Xixi River Basin has altered the original hydrological conditions, slowing water flows. This has prevented some fish species from accessing faster-flowing waters. Consequently, during the breeding season, fish lack the necessary hydrological stimulation, resulting in poor reproduction and low hatching rates. Therefore, it is an ideal research subject for evaluating the benefits of stocking, providing important guidance for optimizing stocking strategies.

The main objective of this study was to investigate resource dynamics following long-term stocking. Traditional sampling methods are used to analyze and statistically evaluate the biological data of H. molitrix and A. nobilis, including survival parameters, and to estimate population abundance. In this study, in order to assess the changes in H. molitrix and A. nobilis stocks resources, eDNA data were used to analyze population composition. This will provide a reference and basis for future stocking and management measures in the Xixi River basin of Jiulong River.

2. Materials and Methods

2.1. Study Area and Sampling Stations

The primary release site is located at the Xihu Ecological Park, which is situated in the middle section of the Xixi River basin. Therefore, the sampling sites for this study are centered on the Xihu Ecological Park, with one station each upstream at the intersection and downstream at the confluence with the Jiulong River, designated as upstream A, midstream B, and downstream C, respectively. Considering the overall habitat of this fish species, the distribution of the fish population after release, and the actual fishing areas of fishermen, and since H. molitrix and A. nobilis are mid-to-upper layer fish species that are more commonly found in areas with river bends and slow-moving water, station A is located at a bend in the Xixi River, station B is the core release area with calm water surfaces, station C is located before the estuary. Since H. molitrix and A. nobilis are freshwater fish, selecting a more upstream location as the site is more appropriate. Additionally, the water flow in this stocking area is gentle and cannot generate rapid currents, preventing effective stimulation of gonadal development. Consequently, no mature fish specimens with well-developed or mature gonads were collected.

2.2. Traditional Resource Investigation Methods

The Xixi River basin has conducted continuous stocking and release operations since February 2017. This sampling represents the first sampling effort, primarily targeting fish species. Sampling commenced in October 2023, with intervals of two months between each sampling session. A total of seven sampling sessions were conducted, with the final sampling date occurring in October 2024. Sampling gear and methodology were established in accordance with the People’s Republic of China Fishery Industry Standard “Specifications for Freshwater Fishery Resources Investigation” (SC/T 9249-2019) [44]. Sampling was conducted using three-layer gillnets (mesh sizes of 4 cm, 7 cm, and 10 cm, net length 100 m, net height 2–5 m). At each sampling station, three nets were set parallel to each other at 50 m intervals perpendicular to the shoreline or main current direction. Soak time was standardized at 6–12 h (typically dusk to dawn). Across the three study stations, three nets were deployed per station (representing triplicate sampling), totaling nine three-layer gillnets. All catches were separately labeled and preserved by station and corresponding replicate. On-site identification of H. molitrix and A. nobilis, and other fish species was conducted referencing Fauna Sinica: Osteichthyes and Fish Records of Fujian Province, with all specimens transported to the laboratory for biological measurement and subsequent analysis.

2.3. eDNA Analysis Method

This strategy was designed to account for vertical variation, minimize spatial heterogeneity of eDNA distribution, exclude potential contamination outliers, and improve detection reliability. Water sampling was conducted in the Xixi River in October 2023 and April 2024 at three stations (A, B, and C) (Figure 1). At each station, three simultaneous sampling points approximately 1 m apart were selected. From each point, 1 L of surface water and 1 L of bottom water were collected using a water sampler. All samples were stored in pre-treated 1 L sterile wide-mouth bottles, transported to the laboratory under refrigeration at 4 °C, and processed within 24 h. In the laboratory, the 6 L of water collected from the three simultaneous sampling points at each station (including both surface and bottom layers) were pooled and thoroughly mixed. From this composite, 3 L was used for filtration. Each liter of water was filtered through one mixed cellulose ester (MCE) (Tianjin Jinteng, Tianjin, China) membrane (50 mm diameter, 0.45 μm pore size) using a glass fiber vacuum filtration system (Tianjin Yuze, Tianjin, China), resulting in three membranes per station. All membranes were subsequently stored in 2 mL sterile tubes at −80 °C and processed individually for eDNA extraction, providing three biological replicates.

Total DNA adsorbed on the membranes was extracted using the sodium dodecyl sulfate (SDS) lysis method, with the extracted eDNA stored at −4 °C for subsequent analysis. Amplification was performed using primers and methods adapted from Ji et al. [45], with the reaction volume increased to 30 μL. Subsequent data processing included quality control, assembly, and removal of chimeric sequences. Using the UPARSE algorithm [46], operational taxonomic units (OTUs) were clustered at a 97% similarity threshold. Non-fish sequences were excluded, and non-redundant sequences were extracted after removing singletons. Figure 2 shows a partial display of the sample plots, samples and tools we collected.

2.4. Data Analysis

2.4.1. Estimation of Growth and Death Parameters

The natural mortality coefficient (M) and the total mortality coefficient (Z) were estimated using the formula proposed by Pauly [47]. The total mortality coefficient (Z) and fishing mortality (F) were estimated using the Length-Converted Catch Curve (LCCC) method in FiSAT II (Version: 1.2.0.2), with reference to the approach described by Gayanilo et al. [48]. Pauly’s formula provides a benchmark estimate under general environmental conditions. Since temperature is more stable, reliable, and readily available than other factors, it is a suitable predictive variable. This method is applicable to most temperate, subtropical, and tropical fisheries studies where water temperatures range from 5 to 30 °C and is not limited to previously studied species. In this study, the released species remained in a natural growth state without artificial intervention. The Xixi water area is located in a subtropical region with an average temperature of 24.7 °C. Water temperature is primarily influenced by climate change, which makes this method suitable for analysis. It is important to note that this method provides indirect estimates and that these values may therefore carry uncertainty.

The respective formulae are as follows:

Ln (N_i/dt_i) = a + bt_i

(1)

LnM = −0.0066 − 0.279lnL_∞ + 0.6543lnk + 0.4634lnT

(2)

A = 1 − e^−Z

(3)

F = Z − M

(4)

In Formula (1), N_i is the number of groups i in the body length frequency distribution; t_i is the median age of group i calculated by VBGE (von Bertalanffy Growth Equation); dt_i is the difference between the upper and lower limits of body length in group i.

In Formula (2), L_∞, K are the parameters of the von Bertalanffy growth equation; M is the natural mortality coefficient; T is the average water temperature of the fishing area.

In Formula (3), A is the total mortality rate and Z is the total mortality coefficient.

In Formula (4), F is fishing mortality.

2.4.2. Biomass Analysis

The formula for calculating the dominance of fish and the formula for counting the catch quantity of different species are, respectively [49]:

IRI = (N + W) × F × 1000

(5)

Y = T_i/T × N_i/N

(6)

$$C_i \%={\displaystyle \frac{N_i}{N}}\times \mathrm{100}\%$$

(7)

In Formula (5), N represents the proportion of the number of a certain fish to the total number of fish, %; W represents the percentage of the biomass of a certain fish to the total biomass, %; F represents the frequency of a certain fish in the entire sampling period. In this paper, species with IRI > 100 are considered dominant species.

In Formula (6), T_i represents the number of sample points where the i-th fish appears; T represents the total number of sample points; N_i represents the sequence abundance of the i-th fish. When Y ≥ 0.02, the species is defined as a dominant species.

In Formula (7), C_i % is the percentage of the number or weight of the i-th fish; N_i is the number or weight of the i-th fish; N is the total number or weight of the catch.

3. Results

3.1. Population Composition

There were 41 fish species caught in the traditional fishing survey, belonging to 1 class, 8 orders and 18 families (

Table 1. List and distribution of catch during the survey period.

Species	Station A	Station B	Station C
Pleuronectiformes
Paralichthyidae
Tephrinectes sinensis			+
Clupeiformes
Engraulidae
Coilia grayii	+	+	+
Clupeidae
Konosirus punctatus	+	+	+
Clupanodon thrissa	+	+	+
Elopiformes
Megalopidae
Megalops cyprinoides	+
Cypriniformes
Cyprinidae
Hemiculter leucisculus	+	+	+
Ctenopharyngodon idella	+	+
Squaliobarbus curriculus	+	+
Spinibarbus hollandi			+
Sinibrama macrops			+
Cyprinus carpiovar.specularis	+
Carassius auratus	+	+
Cyprinus carpio	+	+	+
Hypophthalmihthys molitrix	+	+	+
Cirrhinus molitorella		+	+
Cirrhinus mrigala			+
Megalobrama terminalis		+	+
Culter alburnus	+	+	+
Mylopharyngodon piceus	+	+
Chanodichthys dabryi	+	+	+
Osteochilus salsburyi	+		+
Aristichthys nobilis	+	+	+
Cobitidae
Paramisgurnus dabryanus			+
Perciformes
Clariidae
Clarias fuscus			+
Theraponidae
Therapon oxyrhynchus			+
Gerreidae
Gerres filamentosus			+
Cichlidae
Sarotherodon galilaeus	+	+	+
Oreochromis sp.	+		+
Cichlasoma managuense		+	+
Oreochromis niloticus	+	+	+
Coptodon zillii	+	+	+
Anabantidae
Anabas testudineus		+
Eleotridae
Eleotris oxycephala		+	+
Gobiidae
Glossogobiuss giuris	+	+	+
Mugilogobius			+
Siluriformes
Bagridae
Pelteobaggrus nitidus		+
Tachysurus fulvidraco	+	+
Amblycipitidae
Ictalurus punctatus		+	+
Loricariidae
Pterygoplichthys pardalis	+	+	+
Tetraodontiformes
Tetraodontidae
Takifugu ocellatus		+	+
Mugiliformes
Mugilidae
Mugil cephalus	+	+	+

Note: “+” indicates the presence of this species at the station; A, B, and C, respectively, represent three sampling points.

), and the community composition showed a relatively diverse pattern. Among them, Cypriniformes was the dominant order with 18 species accounting for 43.9% of the total, next is the Perciformes with a total of 12 species and accounting for 29.27%., Siluriformes (9.76%), Clupeiformes (7.32%), Pleuronectiformes (2.44%), Elopiformes (2.44%), Tetraodontiformes (2.44%), and Mugiliformes (2.44%). The species composition of each sampling sites showed that there were 24 species belonging to 6 orders, 9 families in station A; there were 28 species belonging to 6 orders, 12 families in station B; and there were 32 species belonging to 7 orders and 15 families in station C. Overall, among station A, B, and C, station C collected the most species, exhibiting greater diversity in species composition compared to the other stations.

As shown in

Table 2. Number of samples collected each time from 2023 to 2024 (ind.).

Survey Dates	Station A	Station B	Station C	Total
2023/10	117	110	90	317
2023/12	147	39	85	271
2024/2	74	180	159	413
2024/4	183	120	90	394
2024/6	374	186	266	826
2024/8	274	290	280	844
2024/10	202	133	94	429
Total	1371	1058	1064	3494

, a total of 3494 individuals were caught in this survey, with a total weight of 666.85 kg. The results of the fish relative importance index (IRI) calculations are shown in

Table 3. Relative important indices (IRI) of Fish.

Species	W	N	F	IRI
Hypophthalmihthys molitrix	0.67044	0.27619	1.00	946.63
Coptodon zillii	0.03008	0.12822	1.00	158.30
Cichlasoma managuense	0.01622	0.14797	0.86	140.74
Sarotherodon galilaeus	0.03997	0.09101	1.00	130.98
Cyprinus carpio	0.08439	0.02776	1.00	112.15
Coilia grayii	0.00940	0.09130	1.00	100.70
Oreochromis niloticus	0.02514	0.04980	1.00	74.94
Aristichthys nobilis	0.04865	0.02261	1.00	71.26
Hemiculter leucisculus	0.00940	0.05810	0.86	57.85
Pterygoplichthys pardalis	0.01282	0.01631	1.00	29.14
Konosirus punctatus	0.00677	0.01717	0.86	20.53
Culter alburnus	0.00445	0.01460	0.86	16.33
Clupanodon thrissa	0.00509	0.01030	0.86	13.20
Chanodichthys dabryi	0.00137	0.00572	0.86	6.08
Glossogobiuss giuris	0.00075	0.00744	0.71	5.85
Carassius auratus	0.00253	0.00372	0.86	5.36
Squaliobarbus curriculus	0.00562	0.00544	0.43	4.74
Ctenopharyngodon idella	0.00560	0.00200	0.43	3.26
Mugil cephalus	0.00377	0.00172	0.57	3.13
Clarias fuscus	0.00493	0.00143	0.43	2.73
Takifugu ocellatus	0.00119	0.00515	0.43	2.72
Cirrhinus molitorella	0.00114	0.00114	0.43	0.98
Tachysurus fulvidraco	0.00035	0.00172	0.43	0.89
Osteochilus salsburyi	0.00124	0.00172	0.29	0.84
Gerres filamentosus	0.00041	0.00200	0.29	0.69
Ictalurus punctatus	0.00275	0.00057	0.14	0.47
Mylopharyngodon piceus	0.00213	0.00086	0.14	0.43
Oreochromis sp.	0.00046	0.00114	0.14	0.23
Pelteobaggrus nitidus	0.00015	0.00143	0.14	0.23
Mugilogobius	0.00008	0.00057	0.29	0.19
Therapon oxyrhynchus	0.00010	0.00114	0.14	0.18
Megalops cyprinoides	0.00073	0.00029	0.14	0.15
Eleotris oxycephala	0.00014	0.00086	0.14	0.14
Megalobrama terminalis	0.00026	0.00057	0.14	0.12
Spinibarbus hollandi	0.00054	0.00029	0.14	0.12
Cyprinus carpiovar.specularis	0.00049	0.00029	0.14	0.11
Cirrhinus mrigala	0.00018	0.00029	0.14	0.07
Anabas testudineus	0.00010	0.00029	0.14	0.06
Tephrinectes sinensis	0.00008	0.00029	0.14	0.05
Paramisgurnus dabryanus	0.00003	0.00029	0.14	0.05
Sinibrama macrops	0.00002	0.00029	0.14	0.04

. The dominant species in the Xixi River basin were: H. molitrix, Coptodon zillii, Cichlasoma managuense, Sarotherodon galilaeus, Cyprinus carpio, and Coilia grayii. Their IRI values were 946.63, 158.30, 140.74, 130.98, 112.15 and 100.70, respectively. The total catch proportions of the released monitoring species H. molitrix and A. nobilis accounted for 27.62% and 2.26%, respectively. Their IRI values were 946.63 and 71.26.

3.2. eDNA Results

Using a minimum read threshold of 10 for taxonomic classification, 323,716 raw sequences were processed, yielding 36 taxonomically annotated species across 6 orders and 13 families (

Table 4. Fish species and their proportions detected by eDNA.

Species	Station A	Station B	Station C	Ci (%)
Hypophthalmichthys molitrix	+	+	+	38.7167%
Oreochromis mossambicus	+	+	+	34.2677%
Coreoperca whiteheadi	+	+	+	16.0247%
Mugil cephalus	+	+	+	3.1629%
Coilia grayii	+	+	+	2.5585%
Cyprinus multitaeniata	+	+	+	1.3063%
Mylopharyngodon piceus	+	+	+	1.2881%
Chelon haematocheilus	+		+	0.6391%
Hemibagrus wyckioides	+	+		0.5400%
Glyptothorax sinensis	+			0.3824%
Squaliobarbus curriculus	+		+	0.2632%
Pseudolaubuca sinensis		+		0.2474%
Konosirus punctatus	+	+		0.1897%
Sinocyclocheilus anshuiensis	+	+	+	0.1072%
Elopichthys bambusa	+	+		0.0698%
Cyprinus carpio	+	+	+	0.0439%
Spinibarbus denticulatus yunnanensis	+		+	0.0315%
Pseudobagrus medianalis	+	+		0.0247%
Lateolabrax japonicus	+		+	0.0182%
Metzia formosae	+	+	+	0.0130%
Channa striata	+		+	0.0124%
Acrossocheilus monticola	+	+	+	0.0102%
Perccottus glehni	+		+	0.0080%
Barbodes semifasciolatus	+	+	+	0.0074%
Neolissochilus hexagonolepis	+	+	+	0.0074%
Mastacembelus marmatus	+		+	0.0074%
Aphyocypris kikuchii	+	+	+	0.0071%
Carassius auratus	+		+	0.0068%
Microphysogobio fukiensis	+			0.0059%
Microphysogobio brevirostris	+	+	+	0.0056%
Neolissochilus benasi	+	+	+	0.0056%
Psilorhynchus homaloptera	+	+	+	0.0046%
Sarcocheilichthys sinensis	+	+	+	0.0046%
Acipenser ruthenus	+			0.0043%
Puntius sophore	+	+	+	0.0040%
Garra micropulvinus	+	+		0.0031%

Note: “+” indicates that the type exists in the position. A, B and C, respectively, represent three sampling stations.

). At the ordinal level, Cypriniformes dominated with 22 species (61.11%), followed by Perciformes (6 species, 16.67%), Siluriformes (3 species, 8.33%), Mugiliformes and Clupeiformes (each 2 species, 5.56%), and Acipenseriformes (1 species, 2.78%). Analysis of eDNA data from autumn (October) and spring (April) revealed distinct seasonal variations in dominant fish species composition, as shown in Figure 3. The dominant species in October were H. molitrix, Oreochromis mossambicus, Coreoperca whiteheadi, Mugil cephalus, and Mylopharyngodon piceus; whereas in April, the dominant species were Oreochromis mossambicus, Coilia grayii, H. molitrix, Cyprinus multitaeniata, and Glyptothorax sinensis. The shared dominant species across both seasons were H. molitrix and Oreochromis mossambicus.

3.3. Survival Rate and Standing Stock Resource Calculations

The ELEFAN I technique in FiSAT was used to estimate the growth parameters of the species as follows: H. molitrix, asymptotic length (L∞) = 577.5 mm, growth coefficient (K) = 0.30; A. nobilis, L∞ = 556.5 mm, K = 0.21. The total mortality rates (Z) of H. molitrix and A. nobilis were obtained by the length-transformed capture curve method in FiSAT I I to be 0.91 and 0.73, respectively. Applying the mortality Formula (3), total mortality rates were calculated as 59.75% and 51.81%. By incorporating the mean habitat water temperature T (24.7 °C) from Xixi River basin sampling, along with parameters L∞, K, and T into Pauly’s empirical equation, natural mortality coefficients (M) and fishing mortality coefficients (F) were derived for both species, as detailed in

Table 5. Mortality rate of released species.

Species	Total Mortality Coefficients (Z)	Total Mortality Rates (%)	Natural Mortality Coefficients (M)	Natural Mortality Rates (%)	Fishing Mortality Coefficients (F)	Fishing Mortality Rates (%)
H. molitrix	0.91	59.75%	0.64	47.15%	0.27	12.60%
A. nobilis	0.73	51.81%	0.51	39.96%	0.22	11.85%

.

From February 2017 to September 2023, a total of 2.73 million individuals of H. molitrix and A. nobilis were stocked in the Xixi basin of Jiulong River (

Table 6. Release numbers and survival contribution rates of H. molitrix and H. nobilis in the Xixi River basin.

Species	Total Mortality Coefficients (Z)	Xixi River Basin
		H. molitrix (×104 Ind.)	Survivors (Ind.)	Contribution Rate (%)	A. nobilis (×104 Ind.)	Survivors (Ind.)	Contribution Rate (%)
2017/2	23	17.25			5.75
2018/12	100	75	69,435	40.3	25	27,710	48.2
2019/11	80	60	329,842	91.5	20	133,831	90.0
2020/2	20	15	374,284	64.5	5	160,876	59.9
2021/11	10	7.5	211,037	28.6	2.5	101,622	23.7
2022/1	30	22.5	115,137	26.2	7.5	61,020	19.7
2023/9	10	7.5	136,913	66.2	2.5	65,550	55.1
2024			85,299	35.4	68.25	43,637	27.6
Total/average	273	204.75		50.4			46.3

), comprising 2.0475 million H. molitrix and 0.6825 million A. nobilis at a 3:1 ratio. Mortality analysis revealed an annual total mortality rate of 59.75% (survival rate: 40.25%) for H. molitrix and 51.81% (survival rate: 48.19%) for A. nobilis. Consequently, the projected 2024 standing populations are estimated at 85,299 individuals for H. molitrix and 43,637 for A. nobilis, with annual survival trajectories detailed in Table 5 and Table 6. The overall mortality coefficient was calculated to derive the cl value. The confidence intervals for the mortality rates of H. molitrix and A. nobilis were determined to be 45.7–69.9% and 40.5–60.5%, respectively. The total mortality rates for both species fall within these intervals. The annual contribution rates of stocked cohorts to the extant populations averaged 50.4% for H. molitrix and 46.3% for A. nobilis, peaking in 2019, followed by 2020, with lower values observed during 2021–2022 and 2024. H. molitrix and A. nobilis are common economically important fish species. According to market research conducted in October 2024, the market prices of H. molitrix and A. nobilis are 2.5 CNY/kg and 4 CNY/kg, respectively. The average body weight of H. molitrix was 464.01 g in this study, while the average body weight of A. nobilis was 435.26 g. The economic value was calculated by multiplying the estimated population size (derived from mortality rate and stocking quantity) by the average individual weight and the market price. As shown in

Table 7. Annual release quantity and estimated output value.

Years	H. molitrix		A. nobilis
	Number Stocked (Ind.)	Value Estimate (CNY)	Number Stocked (ind.)	Value Estimate (CNY)
2018	750,000	1,871,420	250,000	1,287,008
2019	600,000	1,490,000	200,000	1,025,000
2020	150,000	372,000	50,000	255,000
2021	75,000	195,500	25,000	130,000
2022	225,000	560,800	75,000	374,000
2023	75,000	347,175	25,000	221,680

and historical stocking data, the year with the highest stocking quantity resulted in the highest estimated economic output, and there is a trend of decreasing economic output as stocking quantities decrease.

4. Discussion

4.1. Efficacy Assessment of H. molitrix and A. nobilis Stock Enhancement

This study estimated the current population ratio of H. molitrix to A. nobilis at approximately 1.97:1, diverging from the initial stocking ratio. The surviving population of H. molitrix is significantly higher than that of A. nobilis. The initial stocking ratio of fry itself is skewed, typically at 3:1 for H. molitrix to A. nobilis, indicating a higher survival rate for A. nobilis. The average annual contribution rates of H. molitrix and A. nobilis were assessed at 50.4% and 46.3%, respectively, indicating substantial survival of stocked individuals and effective resource supplementation. The annual economic yields of the two species was estimated through recapture yield analysis, revealing a positive correlation between stocking quantities and economic returns. IRI values confirmed H. molitrix as the dominant species in the Xixi River basin, while A. nobilis maintained a significant though smaller population. The overall quantity of fishery resources in the Xixi River basin has increased, enhancing the status of fish populations. Therefore, it has played a positive role in maintaining the stability of fishery resources and ecosystems. Consequently, it contributes to maintaining the stability of both fishery resources and the ecosystem in the released waters. As filter feeders primarily consume phytoplankton and zooplankton [50], both species enhance the utilization efficiency of basal aquatic productivity and improve water quality. Additionally, we also conducted a questionnaire survey among local residents and fishermen to analyze their perceptions of the effectiveness of stocking. The findings indicate that residents perceive improvements in water quality, heightened ecological conservation awareness, and increased social recognition of stock enhancement activities, yielding positive societal benefits. Concurrently, these efforts have boosted fishermen’s income, aligning with our findings that fishery resources have been partially replenished. Concurrently, stricter regulations imposed by administrative departments have incentivized hatcheries to voluntarily or mandatorily upgrade breeding technologies, establishing a virtuous cycle. In this study, estimates of parameters such as mortality were indirect, so uncertainties related to environmental factors need to be considered. Future research could improve this study by validating mortality parameters through age-based methods, long-term mark-recapture studies, or direct experimental approaches.

4.2. The Differences Between eDNA and Traditional Assessment Methods

Both eDNA methodology and traditional analytical approaches can assess the efficacy of stock enhancement initiatives, yet significant divergences have emerged with technological advancements, revealing pronounced advantages of eDNA techniques. Previous studies demonstrate these benefits: Xin et al. [51] evaluated the feasibility of biomass assessment for Lutjanus erythropterus using eDNA technology, analyzing water samples pre- and post-enhancement. Their findings confirmed eDNA’s accuracy in detecting biomass fluctuations. Similarly, Wu et al. [52] conducted comparative surveys for the Neophocaena phocaenoides asaeorientalis, demonstrating eDNA’s equivalent detection capability to traditional methods while proving superior for distribution mapping and habitat monitoring. These studies collectively highlight eDNA’s core strengths: reduced operational costs, shortened survey cycles, and enhanced detection sensitivity/accuracy. In contrast, traditional approaches exemplified by Zhang et al. [14] (otolith marking), Riley et al. [15] (PIT tagging), and Tang et al. [16] (fluorescent marking)—though yielding scientifically valid results—exhibit inherent limitations: labor-intensive protocols, high operational costs, physical harm to fish organisms, and suboptimal capture efficiency [53].

Integrating traditional methods with environmental DNA (eDNA) revealed that certain species from the Cypriniformes and Perciformes orders, particularly H. molitrix, were the dominant fish populations in the Xixi River basin. However, the eDNA method only detected the target species, H. molitrix, suggesting a suboptimal approach. Therefore, further refinement of the eDNA method for the Xixi River basin is necessary to more accurately compare eDNA results with biomass estimates and provide recommendations for future methodological improvements.

In addition, the eDNA method captured fish species that were not collected by traditional methods, such as Pseudolaubuca sinensis, Microphysogobio fukiensis, and Lateolabrax japonicus. They demonstrate that eDNA technology can detect deeper and broader ranges, confirming that the eDNA method is effective in fish detection [54]. In addition, previous studies have indicated that common fish species in Xixi include Clupanodon thrissa, Culter alburnus, Takifugu ocellatus, Coptodon zillii, and Carassius information on fish specieswas not detected or was detected in small amounts in this eDNA investigation [55]. Concurrently, non-native species including Sinocyclocheilus anshuiensis, Spinibarbus denticulatus yunnanensis, Pseudobagrus medianalis, Acrossocheilus monticola, Perccottus glehni, Barbodes semifasciolatus, Neolissochilus hexagonolepis, Neolissochilus benasi, Psilorhynchus homaloptera, Acipenser ruthenus, Puntius sophore, and Aphyocypris kikuchii were identified. These discrepancies likely arise from: eDNA degradation in aquatic systems, incomplete reference databases, contamination during sampling/extraction/storage, and limitations of universal primers, resulting in certain taxa not being classified to species level [56]. It can be seen from this that eDNA technology has its own shortcomings and cannot completely replace traditional investigation methods.

Overall, traditional fish resource monitoring primarily relies on physical capture gear, this method has significant limitations. In this study, sampling was inefficient and time-consuming, and it was difficult to collect all species [57], and rare species—leading to biased community structure assessments—and frequent physical harm to fish organisms [58]. The eDNA method used is more convenient, efficient, sensitive, non-invasive, and time-efficient monitoring. It greatly saves the actual sampling time. But it confronts inherent challenges: false positives/negatives [59]; uncertain eDNA origins (allochthonous contamination, prolonged degradation cycles, spatial translocation) [60]; database deficiencies; and difficulties in estimating species abundance or biomass [61,62,63], necessitating standardization.

Subsequent resource investigations can combine the two methods to supplement data, conduct a more comprehensive and efficient assessment of biodiversity within the basin, and provide a more comprehensive perspective on the species composition in aquatic ecosystems [64].

4.3. Management Suggestions

Fishery resource enhancement through stocking serves as an effective approach for resource restoration and conservation [4], integrating multidisciplinary practices spanning aquaculture, fisheries science, and emerging technologies. This process encompasses fundamental scientific issues [65]. H. molitrix and A. nobilis stock enhancement in the Xixi River basin has demonstrated significant benefits, achieving integrated outcomes in fishery resource replenishment and ecological function recovery. Future efforts should prioritize multidimensional assessments and unified management. Consequently, the following recommendations are proposed:

• Implement synchronized control over stocking scale and germplasm quality by establishing dedicated germplasm resource bases to ensure high-quality fingerlings. Optimal stocking scales should be selected to maximize post-release survival rates, with appropriate size classes identified to enhance survival rates and economic returns [66].
• Optimize release timing and regional strategies through comprehensive consideration of species’ breeding seasons, water temperature, food resource availability, stocking objectives, predators, and habitat conditions [67]. For instance, selecting fast-growth seasons for releases improves fishery output, while avoiding summer (excessive heat) and winter (extreme cold) periods that compromise transportation and survival [68].
• Establish sustained monitoring and evaluation mechanisms involving multidimensional assessments. Implement long-term ecological monitoring and data analysis for stock enhancement in the Xixi River basin, incorporating novel technologies (eDNA, microsatellite markers) for resource evaluation [69].
• Strengthen surveillance and control of invasive species. This survey detected dominant non-native species including Coptodon zillii, Cichlasoma managuense, Sarotherodon galilaeus, an Pterygoplichthys pardalis, which impact native species composition. Long-term management through physical removal, biological control, and natural predator introduction should be implemented.

5. Conclusions

The stock enhancement and release of fishery resources is an important measure to protect species and increase fishery resources. Subsequent evaluation of restocking effectiveness necessitates optimization of technical strategies. The release of fishery resources in the XiXi River basin has effectively supplemented and enhanced fishery resources, thereby improving the ecological environment of the water area. It is worth mentioning that this study combined two survey methods to improve sampling efficiency and species detection rates. However, the accuracy of fish detection still has certain limitations and room for improvement. In the future, by combining the two methods, optimizing primer design, constructing DNA databases, and continuously improving a multi-level ecological effect evaluation system based on communities and populations, we can ensure the sustainable development of stocking efforts.