Fish community structure and diversity in the Ningxia section of the main stream of the Yellow River in China

Access to data

Fisheries resources and environmental surveys were conducted in the Ningxia section of the Yellow River mainstream in July 2022, as well as in February, May, and September 2023. Survey methodologies were performed in accordance with the Manual of Investigation for Fisheries Resources in Inland Waters (Zhang & He, 1991). Following field investigations and integration of site visit findings, 15 sampling sections were finally chosen for examination (Table 1) (Fig. 1). Three types of fishing nets were deployed: ground cages (mesh size 0.5 cm, dimensions 20 m × 0.4 m × 0.4 m), fixed composite gillnets (mesh sizes 4, 6, 8 cm, with a height of 2 m and length of 20 m), and drift gillnets (mesh sizes 2, 4, 6 cm, with a height of 2 m and length of 30 m). The ground cages and fixed gillnets were typically set for 24 h, while the drift gillnets were deployed over a drifting distance of 2 km. Refer to relevant materials (Wu & Wang, 2008; Cai, Zhang & Wang, 2013; Li, 2017) to identify the species of fish collected. Fish samples unable to be identified in situ were fixed in 5–10% formalin solution, with capture time and location recorded, and subsequently transported to the laboratory for species identification. For biological measurements of the catches, body length was measured to the nearest 1 mm, and weight to the nearest 0.01 g. During the survey, water temperature, dissolved oxygen (DO), pH, chlorophyll a (Chl-a), and ammonia nitrogen (NH₃-N) were measured using a YSI portable multi-parameter water quality analyzer. Additionally, 1 L water samples were collected from each section, cryopreserved, and subsequently analyzed for NH₃-N and Chl-a concentrations. All sampling procedures strictly adhered to the guidelines provided by Heilongjiang River Fisheries Research Institute of Chinese Academy of Fishery Sciences for Laboratory Animal Welfare and Ethical Review (No: 20220420-004).

Data processing and analysis

Fish species composition, ecological type and dominant species

The study results showed that a total of 42 fish species were collected from the Ningxia section of the Yellow River mainstem, belonging to six orders, 11 families, and 34 genera. The Cyprinidae family was the most species-rich, with 24 species accounting for 57.14% of the total. Among the 42 species recorded, 25 were non-native species (59.5%), while the remaining 17 were native species (40.5%). According to feeding habits, the fish were classified into different ecological types: carnivorous species dominated with 22 species (52.4%), followed by omnivorous species with 16 species (38.1%), and herbivorous species with four species (9.5%). In terms of habitat, benthic fish were the most prevalent, comprising 25 species (59.5%), followed by pelagic fish with nine species (21%) and semi-pelagic fish with eight species (19%) (Table 2).

In the Ningxia section of the Yellow River, Gobiobotia huanghensis emerged as the overall dominant species. Temporal analysis revealed distinct seasonal and interannual variations in dominant species composition across the four sampling campaigns, with Carassius auratus, Cyprinus carpio, and G. huanghensis maintaining consistent dominance throughout all surveys. The July 2022 survey recorded additional dominance by Pseudorasbora parva and Squalidus lanzhouensis, while February 2023 was characterized by Leuciscus chuanchicus and Abbottina schankaensis. By May 2023, A. schankaensis and P. parva (the latter continuing its dominance from previous months) predominated, followed by L. chuanchicus, A. schankaensis, and Rhinogobius giurinus in September 2023. Spatial analysis demonstrated significant geographic heterogeneity in dominant species distribution across sampling stations, with G. huanghensis and C. carpio showing widespread dominance. Notably, G. huanghensis maintained dominance across all 15 surveyed sections, demonstrating remarkable spatial consistency (Table 3).

Fish diversity

According to the survey results, the Shannon-Wiener diversity index (H) exhibited values ranging from 1.189 to 2.441 across sampling sites, with an average of 1.791. The lowest index was recorded at Station S1, while Station S14 showed the highest. For the Margalef species richness index (D), values varied between 2.112 and 3.805 (mean = 3.066), with the lowest richness observed at Station S2 and the highest at Station S12. The Pielou evenness index (E) spanned 0.4505–0.7897 (mean = 0.5911), with the minimum and maximum values recorded at Station S1 and Station S14, respectively. The Simpson dominance index (C) ranged from 0.4536 to 0.8802 (mean = 0.7083), with the smallest value at Station S1 and the largest at Station S14.

From a temporal perspective, the Shannon-Wiener diversity index (H) fluctuated between 2.012 and 3.132 across the four surveys, with an average of 2.474. The Margalef species richness index (D) ranged from 2.213 to 3.588 (mean = 2.89), while the Pielou evenness index (E) varied between 0.5214 and 0.7356 (mean = 0.6456). The Simpson dominance index (C) showed the widest range, spanning 0.62 to 0.9644 with an average of 0.7177 (Table 4).

Abundance biomass comparison curve

The ABC curve indicates that, over the course of 2023, the W values in May and September were less than zero. During these months, the abundance and biomass curves intersected, suggesting that the fish community experienced moderate disturbances. Conversely, the W values in July 2022 and February 2023 exceeded zero, with the biomass curve positioned above the abundance curve, pointing to a stable state in the fish community during these periods (Fig. 2). From a spatial perspective, the W values at Nanchangtan (S1), Shapotou (S2), Jinshawan (S7), Meijiawan (S9), Linhe (S11), and Taole (S13) were all below zero. Typically, the abundance dominance curve was above the biomass dominance curve, with the starting point of the former higher than that of the latter, indicating significant disturbances within the fish communities at these six stations. Conversely, the W values at Baima (S4), Kushou (S8), Wanghong (S10), and Wayaocun (S15) were greater than zero, where the biomass dominance curve intersected with the abundance dominance curve, signaling moderate disturbances at these four stations. Meanwhile, the W values at Yuding (S3), Mingsha (S5), Niushoushan (S6), Yuyahu (S12), and Hongyazi (S14) also exceeded zero, with the biomass dominance curve consistently above the abundance dominance curve, and the starting point of the biomass dominance curve surpassed that of the abundance dominance curve, reflecting a relatively stable fish community structure at these five sites (Fig. 3).

Community cluster analysis

Cluster and NMDS analyses were conducted on fish communities, as depicted in Fig. 4. Cluster analysis revealed that fish community structures in the Ningxia section of the Yellow River mainstem were grouped into four clusters at a Bray-Curtis similarity level of 48.62%. Notably, visual inspection of cluster dendrograms suggested limited differentiation among sampling sections. One-way analysis of similarities (ANOSIM) tests further indicated no significant differences in community structure among the four clusters (R = 1, P = 0.1), supporting the hypothesis of homogeneous community composition. The NMDS ordination (stress = 0.07) aligned with cluster results, demonstrating moderate explanatory power for inter-sample relationships.

Relationship between fish community and water environmental factors

RDA was performed to explore the relationships between seven dominant fish species (e.g., G. huanghensis, C. carpio) and five water environmental variables (water temperature, DO, pH, Chl-a, and NH₃-N monitored across the Ningxia section of the Yellow River mainstem. Random permutation tests uncovered significant temporal variations in how water environmental factors influenced fish community structure in the Ningxia section of the Yellow River mainstem. Specifically, water temperature and dissolved oxygen were significant factors in July 2022 (P < 0.05); NH₃-N and Chl-a in February 2023 (P < 0.05); NH₃-N and dissolved oxygen in May 2023 (P < 0.05); and only water temperature in September 2023 (P < 0.05). It is evident that water temperature and NH₃-N are the primary environmental factors affecting the fish community in this section of the Yellow River (Fig. 5).

Note: a is July 2022, b is February 2023, c is May 2023, d is September 2023. WT is water temperature, DO is dissolved oxygen, AN is ammonia nitrogen, Chla is chlorophyll a. The red solid line indicates a significant impact, while the blue dashed line indicates no significant impact.

Status of fish resources

The fish collected in this survey were predominantly cyprinid species, a prevalence that aligns with the common characteristic of freshwater fish faunas across China (Chen, Chen & He, 2002). Compared with historical records of 18 species (He, Wang & Song, 1963) and 26 species (Hu et al., 2022) in the Ningxia section of the Yellow River, the fish species composition identified in this study exhibited higher diversity. This higher diversity may be attributed to more comprehensive survey coverage and more diverse fish sampling methods employed in this study. The shifts in fish ecological types are closely linked to river flow velocity, nutrient levels, and the distribution of aquatic organisms (Li et al., 2013a). In this survey, carnivorous and omnivorous fish dominated the assemblage, comprising 90.5% of the total catch—a notable increase compared to historical records. It is speculated that the reasons for these changes may be water nutrient input from irrigation return flows in the Ningxia section of the Yellow River, loss of fish habitats due to changes in hydrological conditions caused by the construction and operation of the Qingtongxia Reservoir, and continuous decline in water quality, food resources, and habitat conditions caused by high sediment content in the Ningxia section of the Yellow River (Qi, 2017). These factors collectively result in a sustained decline in sensitive fish populations and a corresponding increase in tolerant species. Carnivorous and omnivorous fish, characterized by broader ecological niches and enhanced adaptability to disturbed environments, exhibit greater resilience under these altered conditions (Meng et al., 2022).

Studies indicate that exotic fish generally exhibit higher adaptability than indigenous species, conferring them significant advantages in food acquisition, habitat utilization, and survival-reproduction strategies (Ba & Chen, 2012). Chen, Yang & Li (1998) documented that invasive exotic fish species represent a key driver of endangerment and even extinction of indigenous fish populations on the Yunnan-Guizhou Plateau. In this survey, 25 exotic fish species were recorded, comprising a proportion exceeding that of indigenous species. Among them, top predators like S. lucioperca and C. argus which occupy high trophic positions in the food web (Li, 2017) exert significant predation pressure on indigenous fish, potentially threatening their populations and biomass. Due to religious practices and artificial stocking, certain non-native species such as H. olidus and Protosalanx hyalocranius has established naturalized populations in the mainstem of the Yellow River, China (Wang, 2021). Therefore, greater attention should be paid to the potential risks posed by non-native fish species, and rigorous risk assessments should be proactively implemented.

Fish miniaturization has emerged as a widespread phenomenon in inland rivers, reservoirs, and lakes across China (Liu et al., 2021). The results of IRI analysis indicated that the dominant fish species in the Ningxia section of the Yellow River mainstem were small-bodied fishes, including G. huanghensis, C. auratus, and P. parva, the predominance of Cyprinus carpio in the Ningxia section of the Yellow River primarily results from extensive stocking programs conducted by local fishery management agencies. Potential drivers of this pattern include overfishing, aquatic ecological degradation, and invasion of alien species (Xu et al., 2012; Chen et al., 2012).

Characteristics of fish communities

Xu et al. (2007) demonstrated a significant positive correlation between water temperature and fish diversity indices. In this study, fish diversity in the Ningxia Yellow River mainstem exhibited marked spatial and seasonal variations. Specifically, biomass, species richness, Shannon-Wiener diversity index, and Margalef richness index in July 2022 were significantly higher than those in the other three surveys, which may be linked to the higher water temperatures in July. On a spatial scale, fish community metrics in the Ningxia Yellow River mainstem showed that the Shannon-Wiener diversity index, Pielou evenness index, and Simpson dominance index were lowest at station S1 and highest at station S14, which may be due to the gentle water flow, suitable habitats, and well-developed coastal shallow waters at station S14 providing optimal spawning and foraging grounds for fish, while station S1 is located in a canyon area with turbulent currents and low water transparency that create harsh conditions for diverse fish assemblages; higher fish species diversity, richness, and evenness indices collectively reflect greater complexity and stability of fish community structures in aquatic environments, where elevated richness and evenness indicate a more balanced species distribution and enhanced community resilience to environmental perturbations (Cao, Wang & Zhang, 2003). Overall, the relatively high average richness index (3.066) and evenness index (0.59) of fish in the Ningxia Yellow River mainstem suggest a complex and stable fish community structure in this section.

The ABC curve can effectively determine the degree of disturbance to fish communities (Yemane, Field & Leslie, 2005). In this study, the fish communities surveyed in July 2022 and February 2023 showed no signs of disturbance, with the W value reaching its highest level among the four surveys in July 2022. This may be attributed to the recent conclusion of the fishing ban period in the Ningxia Yellow River section, as the months-long prohibition provided effective protection for local fishery resources. In contrast, February 2023 coincided with winter, when low water temperatures led to reduced fish activity, consequently resulting in undisturbed community structure during this sampling period. From a spatial perspective, cross-sections with varying degrees of disturbance are dominated by small fish species, whose ecological habits are generally adaptable and pollution-tolerant. Large fish species—primarily H. molitrix, A. nobilis, C. idellus and C. carpio are scarce and predominantly consist of cultured and released individuals, coinciding with river sections experiencing varying levels of water pollution. In Yuding (S3), Mingsha (S5), Niushoushan (S6), Yueyahu (S12), and Hongyazi (S14) where fish community structures were relatively stable the water quality was comparatively optimal, and large-bodied fish comprised a high proportion of the catch.

The analysis results of Cluster indicate that, there is no significant difference in the fish community structure of most sections of the Ningxia section of the Yellow River main stream. The structure of fish communities is closely related to the water environment they inhabit (Fang et al., 2023). The water environments they inhabit are similar, leading to a high degree of similarity in fish community composition (Li et al., 2012). Most sections of the Ningxia Yellow River feature gentle water flow, and their substrata include both sediments and gravel beaches, resulting in relatively high habitat heterogeneity (Qi, 2017).

Relationship between fish communities and water environmental factors

RDA analysis can effectively elucidate the relationships between fish communities and water environmental factors, revealing the correlations between fish species and distinct habitat types (Zhang, 2011). The distribution and characteristics of fish community structure are influenced not only by the intrinsic ecological habits of fish but also by water environmental factors that shape their survival conditions (Liu et al., 2021). Water environmental factors influencing fish community structural characteristics are complex and diverse, including water temperature, DO, pH, and NH₃-N, among others (Wang et al., 2013; Wang, 2013; Wu et al., 2014). The results of RDA indicated that water temperature was the primary water environmental factor influencing the fish community structure in the Ningxia section of the Yellow River mainstem during July 2022 and September 2023. As a critical driver of fish community diversity, water temperature significantly affects fish growth, development, population reproduction, and other life-history traits. With increasing water temperature, fish exhibit enhanced metabolism of prey resources, accelerated growth and development rates, and subsequent increases in population size (Jacob et al., 1998). And this result is consistent with the research conducted by Gao et al. (2017b) on the fish community in Beijiang and Li (Li, Li & Jia, 2010) on the fish community in Xijiang, both of which reflect the significant influence of water temperature on the structure of fish communities.

Excessive NH₃-N content in water bodies is an important factor leading to eutrophication (Qian et al., 2021). In water bodies with high eutrophication levels, the proportion of large-bodied fish gradually decreases, while the proportion of small-bodied fish continues to increase (Tammi et al., 1999). The results of this study show that small-bodied fish are the dominant species in the fish community of the Ningxia section of the Yellow River, which may be partially attributed to the excessive ammonia nitrogen content in the water (Qi, 2017). The results of RDA analysisfurther revealed that ammonia nitrogen was the primary water environmental factor influencing the fish community structure in the Ningxia Yellow River mainstem during February and May 2023. This finding is consistent with the research by Fang et al. (2023). On fish communities in the middle reaches of the Yangtze River, which reflects the significant influence of ammonia nitrogen concentration in water on fish community structure.