Effects of Biotic and Abiotic Factors on the Spatiotemporal Distribution of Round Scad (Decapterus maruadsi) in the Hainan Island Offshore Area

Abstract

Fishery resource and environmental data from four surveys conducted in the Hainan Island offshore area from March 2021 to June 2022 were used to investigate the spatiotemporal distribution of round scad (Decapterus maruadsi). A generalized additive model was applied to explore the relationships among abundance and distribution, also biotic and abiotic factors (i.e., temperature, salinity, chlorophyll-a concentration, water depth, phytoplankton abundance, zooplankton abundance, and jack mackerel (Trachurus japonicas) abundance). Round scad abundance (average 67.17 kg/km²) showed distinct spatial and seasonal differences around Hainan Island, with highest abundance in summer (171.72 kg/km²) and lowest abundance in spring (3.06 kg/km²). The optimal model revealed that jack mackerel abundance, sea bottom temperature, sea surface temperature, and latitude very significantly (p < 0.01) affected round scad distribution. Jack mackerel abundance (mainly in the range 0–50 kg/km²) was positively correlated with round scad distribution. The distribution showed a dome-shaped relationship with bottom water temperature in the range 18–30 °C, with maximum abundance at 24 °C. The distribution was negatively correlated with surface water temperature in the range 22–30 °C. Sea surface chlorophyll-a concentration and longitude were significantly correlated (p < 0.05) with round scad distribution. The results provide theoretical support to further investigate the formation mechanism of round scad fishing grounds and to enrich knowledge of pelagic fish abundance in the continental-shelf waters of the northern South China Sea.

1. Introduction

Pelagic fish generally inhabit the middle or upper layers of the water column for most of their lives [1]. Such fish are of considerable economic, social, and ecological importance in marine fisheries owing to their unique life-history characteristics (mainly including small individuals, short life cycles, high reproduction, high recruitment, and occupation of a lower trophic level) [2]. Pelagic fish production accounts for around 25% of the total global fish production, which shows significant interannual and interdecadal variation in catches, caused not only by overfishing but also strongly influenced by oceanographic environmental changes [3,4,5]. An important commercial fish species in the continental shelf waters of the northern South China Sea (NSCS), round scad (Decapterus maruadsi), is widely distributed on the vast continental shelf from the Taiwan Shoal to the Beibu Gulf [6]. The fish are mainly caught by pelagic trawls and light seines, and play a crucial role in the ecology of the NSCS [6,7]. Historically, round scad has been among the most productive species in the NSCS, with production once exceeding 300,000 tonnes in the mid-1990s and declining in recent years, but still remaining at approximately 200,000 tons [8].

The spatiotemporal distribution of round scad in the NSCS shows distinct variability [9,10]. A study in the waters off western Guangdong showed that the catch per unit effort (CPUE) of round scad is significantly and positively correlated with longitude, and that the sea surface chlorophyll-a concentration, sea surface height, and water depth each have a pronounced impact on the distribution of round scad [9]. In addition, the distribution of round scad in the eastern coastal waters off Hainan Island is significantly correlated with longitude, sea surface temperature, sea surface chlorophyll-a concentration, and wave direction [10]. Further analysis showed that the variation in abundance and migration routine of round scad in the eastern coastal waters off Hainan Island was associated with monsoon-related oceanographic environmental change [10]. In the context of global anomalous climatic events, pelagic fish are among the most sensitive stock [11]. Zhang et al. [12] reported that abnormal blooms of small pelagic fishes, especially jack mackerel (Trachurus japonicus) and round scad, followed four La Niña events, in 2007/2008, 2010/2011, 2011/2012, and 2020/2021, in the Beibu Gulf marine ecosystem. The impact of anomalous climatic events and climatic regime shift on abundance is usually directed through effects on water temperature and primary productivity, which further affect the recruitment of fish stock [5,13]. Many previous studies have focused on the effect of abiotic factors on the spatiotemporal distribution of fishes, however, the impact of biotic factors is ignored. Previous studies have shown that round scad feed on zooplankton and small fish, with a change in diet as the individual grows. Meanwhile, jack mackerel fed on pelagic crustaceans and small fish, and there was also a change in diet [14]. Both round scad and jack mackerel are crucial pelagic fish in the NSCS, occupying similar ecological niches and competing for food to some extent in the Beibu Gulf. Additionally, there is a noticeable overlap in their respective life history areas [14]. Therefore, it is important to explore whether the distribution of round scad abundance is also associated with these biotic factors (i.e., zooplankton, jack mackerel abundance).

In this study, we applied a generalized additive model (GAM) to analyze the relationship between the distribution of round scad abundance and biotic and abiotic factors. The main objectives of the present study were to analyze the spatiotemporal variability in the distribution of round scad in the Hainan Island offshore area, and to quantify the biotic and abiotic factors that affect its distribution. Our research is focused on investigating the formation mechanism of the round scad fishing grounds and enriching knowledge of the variation in pelagic fish abundance in the NSCS. The present results provide basic information for subsequent resource assessment, conservation, and rational utilization, and for formulation of management frameworks for pelagic fish to withstand fishing pressure and climate change.

2. Materials and Methods

2.1. Data Collection

2.1.1. Round Scad Abundance

Data were obtained from the bottom trawl survey of nekton conducted by the South China Sea Fisheries Research Institute, Chinese Academy of Fisheries Sciences, in spring (March 2021), autumn (September 2021), winter (December 2021), and summer (June 2022) in the Hainan Island offshore area. The survey area was 16°45′–20°15′ N, 107°15′–112°15′ E, and comprised a total of 54 survey stations (Figure 1). The Qiongzhou Strait, located in the northern part of Hainan Island, is deemed unsuitable for trawl survey due to topographical and hydrological factors. Consequently, sampling stations are evenly distributed throughout the surrounding sea area of Hainan Island with the exception of Qiongzhou Strait (Figure 1). The station settings are the same for all seasons. The survey vessel was Guibeiyu 69068, a steel fishing vessel with a gross tonnage of 590 t, overall length 53.8 m, width 8.2 m, depth 4.6 m, and main engine power 436 kW. The length of the net used was 60.5 m, the perimeter of the net mouth was 80.4 m, the width of the net mouth was 37.7 m, and the mesh of the net bag was 40 mm. Based on the substrate condition and surrounding hydrographic environment in the survey area, each station was trawled once for 1 h at an average trawling speed of 3 nmile/h. The total catch of round scad at each station was recorded, which was divided by the trawling time, and standardized to the catch per unit time as the catch rate (in kg/h). The following formula (Equation (1)) was used to calculate the fish stock density of round scad at each station [15,16], which represents the round scad abundance for subsequent study:

(1)$P=\frac{C}{q\times A}$

where P is the fish stock density and represents the abundance in this study, C is the catch per hour in the sampling area, q is the gear capture rate, taken as 0.5, and A is the area of sea swept by the net per hour. The units for P, C, and A are kg/km², kg, and km², respectively.

2.1.2. Biotic and Abiotic Factors

The biotic factors considered were phytoplankton abundance, zooplankton abundance, and jack mackerel stock density. Phytoplankton and zooplankton were collected using a shallow water type III plankton net (mesh aperture 77 μm, net mouth inner diameter 37 cm, and net length 270 cm) and a shallow-water type I net (mesh aperture 0.505 mm, net mouth inner diameter 50 cm, and net length 145 cm), respectively, from the bottom to the surface of the vertical trawl sampling. Samples were fixed with 5% formalin solution and transported to the laboratory for analysis and identification. The jack mackerel catch samples were obtained simultaneously with round scad sample collection. The sample collection stations are shown in Figure 1. Jack mackerel stock density was calculated using Equation (1).

Environmental data, such as sea surface temperature, sea surface salinity, sea bottom temperature, sea bottom salinity, and water depth, were measured simultaneously at each station using conductivity–temperature–depth (CTD) sensors. Chlorophyll-a (Chla) concentration was measured by filtering 1 L of seawater through Whatman (45 mm) GF/F filters. The filters were frozen in the dark for 24 h before extraction in 90% acetone. The Chla concentration was measured with a fluorometric method in accordance with Yentsch and Menzel [17] using a Turner Designs 10-Au Fluorometer.

2.2. Data Analyses

2.2.1. Generalized Additive Model

In fisheries research, the effects of independent variables, such as temperature and salinity, on dependent variables, such as CPUE and fishing effort, are often nonlinear. Therefore, for the present data analysis a GAM was used, which was first proposed by Hastie and Tibshirani [18] as a nonparametric generalized multiple nonlinear regression method that can handle nonlinear relationships between response variables and explanatory variables in high-dimensional data. A GAM is a semi-parametric extension of the GLM, which smooths the predictor variables independently and calculates the degree of response change in an additive manner, thus enabling representation of both linear and nonlinear relationships between variables. The general expression of a GAM is [18]:

(2)$Y=\alpha +{\displaystyle \sum }_{i=1}^k{f}_i\left(x_i\right)+\epsilon$

where Y is the response variable (round scad abundance), and x_i is the explanatory variables, that is, the biotic and abiotic factors at each station. In this study, eight abiotic factors, namely longitude (Lon), latitude (Lat), sea surface temperature (SST), sea bottom temperature (SBT), sea surface salinity (SSS), sea bottom salinity (SBS), water depth (Depth), and sea surface chlorophyll-a concentration (Chla), were selected. Based on relevant previous studies, the density of phytoplankton, zooplankton, and jack mackerel may have direct or indirect effects on the distribution of round scad [14,19]. Therefore, phytoplankton abundance, zooplankton abundance, and jack mackerel stock density were selected as three biotic factors. In addition, seasonal variables were added to the model for analysis as factor variables. In Equation (2), f_i, a, and ϵ are a smoothing function (with k ≤ 3 to avoid overfitting), intercept, and error terms, respectively. The error functions for the model analysis were all normally distributed and the link function was the natural logarithm. Round scad abundance was log transformed with increasing constant terms before addition to the model. All explanatory variables used in this study are listed in Table 1.

2.2.2. Factor Screening and Model Selection

Pearson correlation analysis and the variance inflation factor (VIF) were applied to screen for multicollinearity characteristics among the explanatory variables added to the model. A value of VIF > 3 is generally considered to represent multicollinearity [20]. The screening of factors was conducted in conjunction with the biological properties of round scad.

The Akaike information criterion (AIC) was used to test the fit of the model by stepwise regression. The two-factor prediction model with the smallest AIC value was obtained by stepwise addition of other factors as explanatory variables. The number of factors in the model was increased in turn until the AIC value of the model did not decrease with the addition of new factors. The model with the smallest AIC value is the model with the best fit [21].

The formula used (Equation (3)) was:

AIC = 2 k − 2 ln L(3)

where k denotes the number of parameters and L is the likelihood function.

The fitting and validation of the models was performed using R version 4.1.0. The construction and testing of the GAM were conducted with the R package ‘mgcv’.

3. Results

3.1. Spatiotemporal Distribution of Round Scad

The catch rate of round scad was calculated as the stock density, according to Equation (1), to represent the variation of round scad resource abundance in the Hainan Island offshore area. Round scad abundance showed distinct seasonal differences in the Hainan Island offshore area. The average abundance was 67.17 kg/km² and the highest seasonal abundance was observed in summer (171.72 kg/km²), followed by autumn (58.98 kg/km²), winter (34.94 kg/km²), and spring (3.06 kg/km²). In spring, round scad was recorded at only 20 stations, and the area of highest abundance was located off the northern coast of Hainan Island, whereas, in the Beibu Gulf and Beibu Gulf mouth waters, round scad was only sporadically recorded at stations and the abundance was low. The abundance in summer was substantially higher than in the other seasons. Round scad was caught at 34 stations in summer. The waters off the northwest coast of Hainan Island, that is, the central waters of the Beibu Gulf, had the highest abundance of round scad, which was mainly distributed in the mouth of the Beibu Gulf and the eastern coastal waters off Hainan Island. In autumn, round scad was caught at 42 stations; the species was widely distributed in the waters around Hainan Island, and the catch was more evenly distributed among stations. Round scad was recorded at 41 stations in winter. The winter catch rate was lower than that in autumn and summer, and the resource distribution was more extensive, but the main distribution area was located in the eastern and northeastern coastal waters off Hainan Island (Figure 2).

3.2. Selection of Variables

Pearson correlation analysis was conducted on 12 explanatory variables in the Hainan Island offshore area. The variables SST and SBT were significantly correlated with other abiotic factors, and SBS and depth were strongly correlated with other explanatory variables (Figure 3). The VIF values for all variables were within the range of 1 to 3 except those for depth and SBS (Table 2). Therefore, only depth and SBS were excluded and the remainder of the selected variables were suitable for inclusion in the GAM.

3.3. Model Selection

A GAM was constructed to evaluate the effects of biotic and abiotic factors on the distribution of round scad abundance in the Hainan Island offshore area. The process of testing stepwise the fit of the model is shown in Table 3. The results showed that Lon, Lat, SST, SBT, SSS, Chla, JM, and seasonal factors were suitable for addition to the model (Table 4). The optimal model revealed that JM, SBT, SST, and Lat had highly significant effects on the distribution of round scad, and that Chla and Lon were significantly correlated with the distribution of round scad. However, SSS did not significantly affect the distribution of round scad. These results indicated that the GAM satisfactorily explained the relationship between each impact factor and the distribution of round scad abundance.

3.4. Effect of Explanatory Variables on Distribution of Round Scad Abundance

The effect of each explanatory variables on the distribution of round scad abundance is shown in Figure 4. The most significant factors influencing the distribution of round scad abundance were JM, SBT, SST, and Lat. The effect of JM on the distribution of round scad was mainly in the range of 0–50 kg/km²; the variation in round scad abundance increased but gradually tended to level off with increase in JM. Round scad was mainly distributed within the SBT range of 18–30 °C and showed a dome-shaped relationship, attaining maximum abundance at approximately 24 °C. Round scad was mainly distributed in the SST range of 22–30 °C and abundance decreased with increase in SST. The round scad abundance showed an inverse dome-shaped relationship with latitude, attaining minimum abundance at 18.5°. The main distribution of round scad abundance was observed in the Chla range from 0 to 1 mg/m³. Variation in abundance was negligible in the Chla range from 0 to 1 mg/m³ and round scad abundance declined with increase in Chla. The round scad abundance was lowest at 110°, and had an inverse dome-shaped relationship with longitude (Figure 4).

4. Discussion

4.1. Spatiotemporal Distribution of Round Scad

The bottom trawl survey in the Hainan Island offshore area revealed that the distribution of round scad showed significant seasonal and spatial differences. The abundance was substantially higher in summer than in the other three seasons, for which there may be two main reasons. First, in this study, the summer survey period was from late May to early June. This period is the spawning season for round scad migrating to the northeastern coast of Hainan Island and the Beibu Gulf [6]. Under the influence of the southwest monsoon in summer, round scad juveniles drift to shallow coastal waters and bays with the wind and sea currents [6,10]. Therefore, more juvenile fish are distributed in these areas, which may become the target of offshore seine and set net fisheries together with other pelagic fish [6,22]. Second, a fishing moratorium applies in Chinese coastal waters in summer, when fishing pressure is accordingly low [12]. Therefore, summer is conducive to the reproduction and growth of fish in large numbers, which may result in the higher abundance surveyed during this period than in the other seasons [12].

The survey showed that round scad inhabited the waters around Hainan Island, but the main distribution area was the northeastern coastal waters of Hainan Island and the mouth of the Beibu Gulf. Previous studies have shown that the Beibu Gulf and northeastern Hainan Island may be important spawning areas for round scad in the NSCS [6,22]. The spawning period can extend from late winter to summer, the main spawning season is concentrated in April to June, and the water depth of the main spawning area ranges from 60 to 150 m [22]. Round scad migrate from the southern part of the Beibu Gulf to the coast of the Beibu Gulf for feeding during winter, and then spawn in shallow sedimentary waters in late winter and early spring, and gradually move to deeper waters after spawning with the arrival of the southwest monsoon in summer [6]. Fan et al. [23] used a habitat suitability index to study round scad in the NSCS, showing that the Beibu Gulf is a superior habitat for the species. In addition, there is a strong seasonal upwelling in the eastern waters of Hainan Island; therefore, seasonal fishing floods form readily during the summer and autumn [10,24]. Upwelling of nutrient-rich waters has a positive effect on pelagic fish aggregation [25]. The strong upwelling along the coast causes the water temperature to decline and draws a large amount of nutrients from deep water layers to the surface layer, thereby promoting the growth of phytoplankton and increasing the chlorophyll-a concentration and primary productivity in the area, and thus providing suitable conditions for fish spawning, nursing, and feeding [26,27]. Based on the above upwelling, productivity and other aspects of the impact, resulting in a high density of round scad in the northeastern waters of Hainan Island. The vertical distribution of phytoplankton is more uniform in the eastern waters off Hainan Island and the contribution of upwelling attains more than 90% [28]. The low-salinity and nutrient-rich environment created by continental runoff from the Pearl River and other water systems into the sea, along the Qiongzhou Strait and Guangdong coastal waters, plays an important role in the reproduction and hatching of fish eggs [29], and provides favorable conditions for fishing in the northwestern part of the South China Sea from the eastern waters off Hainan Island to the western end of the Guangdong coast. Therefore, both biotic and abiotic factors may be important factors affecting the distribution of round scad. In addition, the interaction between factors is critical, and their specific effective mechanisms require further investigation.

4.2. Effects of Biotic Factors

In this study, the biotic factors phytoplankton abundance, zooplankton abundance, and jack mackerel stock density were added during construction of the GAM. However, only the density of jack mackerel stock had a significant strong correlation with the distribution of round scad. Jack mackerel and round scad are both important pelagic fish in the NSCS. Their distribution areas strongly overlap, and fish of both species are often caught together in light seine and bottom trawl catches [30,31]. Li et al. [14] showed that round scad and jack mackerel occupy similar trophic levels and show a high degree of overlap in ecological niches in the NSCS, indicating that they may compete for food, which may therefore result in a correlation between their distributions and variation in their abundance. At some stages of their life histories, they feed on zooplankton, crustaceans, and small fish, but there are shifts in feeding habits [14]. This competition for feeding will probably have an impact on the early recruitment stage of the population, thus resulting in fluctuations in population dynamics. Both round scad and jack mackerel, as economically important species in the NSCS, have shown distinct interannual to decadal fluctuations in catches in the South China Sea in recent years, and the variation in trends is increasingly pronounced (Figure 5). Previous studies have demonstrated the apparent nonsynchronized interannual to decadal variation in the abundance of pelagic fish with similar ecological niches [32,33,34]. In the case of sardine (Sardinops sagax) and anchovy (Engraulis japonicus) in the Northwest Pacific, completely opposite trends in the same interdecadal resources are observed in the same sea area [35,36]. Nakayama et al. [32] showed a clear interspecific interaction exists between sardine and anchovy, which may be one reason for the opposite patterns of variation of the two species. The optimal temperature for early growth differs between sardine and anchovy in the same habitat, and hence when the temperature changes, the early growth stage of the two fish will have completely different adaptations [36,37]. This may further lead to differences in the percentage survival of juvenile fish in the early life-history stage, thus affecting the recruitment of the species. In the case of fishes with strongly overlapping distribution areas, when environmental and other factors are more favorable to one species, the survival space of the other species is inevitably affected, resulting in variation in its distribution and abundance [36]. The four surveys performed in the present study showed that the occurrence of round scad and jack mackerel at the survey stations overlapped considerably, indicating that their distribution areas strongly overlap. Simultaneously, the GAM results showed that the distribution of round scad was significantly correlated with the stock density of jack mackerel. Both round scad and jack mackerel were able to feed on zooplankton at a certain life history stage, but the zooplankton did not show a significant correlation with their distribution, which may be related to the fact that in this study we selected the total zooplankton abundance rather than the abundance of specific species that both species preferred to feed on. In addition, both species had feeding transitions, which could also show nekton feeding at a later stage. Therefore, the distribution of the two species needs to be explored in more depth to determine how they interact with each other.

4.3. Effects of Abiotic Factors

The GAM results revealed that the distribution of round scad in the waters around Hainan Island was affected by SST, SBT, and Chla. It is generally considered that temperature and chlorophyll-a are important for the formation and distribution of fishing ground, and that temperature has a crucial influence on fish growth and metabolism from an early growth stage to the adult stage [23,38]. Temperature is an important factor affecting the distribution of fish stock [12]. Fish migration is thought to be in pursuit of the optimal habitat temperature and a superior feeding environment [39]. Zhang et al. [12] observed abnormal blooms of small pelagic fish, especially round scad and jack mackerel, following four La Niña events in 2007/2008, 2010/2011, 2011/2012, and 2020/2021, in the Beibu Gulf marine ecosystem. The main reasons for the blooms included not only fishing pressure, but also variation in the water temperature in the Beibu Gulf caused by the La Niña events, which further affected the abundance of pelagic fish. The present results showed that both SST and SBT had important effects on the distribution of round scad abundance. The SST range of the main concentration of round scad distribution was 22–30 °C, which was similar to the results of Fan et al. [23] and He et al. [10]. Round scad were more inclined to inhabit an area with a higher SST. The resource abundance of round scad varied with SBT in a dome-shaped pattern. The optimal water temperature range was 22–26 °C, which was consistent with the higher abundance in the SST range. Round scad was mainly caught in light seine and bottom trawl catches in the NSCS, and data from daytime bottom trawls were used in the present study. Although round scad is a pelagic fish, owing to its diurnal migratory characteristics and shallow water depth in the NSCS (all survey areas were located at a depth of 200 m or less), in the daytime it occupies the bottom or near-bottom water layers; therefore, the variation in bottom water temperature also has an important impact on the fish [40].

Round scad predominantly feeds on small fish and crustaceans [14]. The Chla concentration affects the distribution of large zooplankton, small fish, and crustaceans. Therefore, Chla may be an important factor affecting the distribution of round scad fishing grounds. The main physical drivers of changes in fish catch in the NSCS include river runoff, the monsoon, and typhoons, whose effects change the primary productivity of an area mainly by controlling the nutrient supply of the water column [29]. The sea surface chlorophyll-a concentration is an important parameter for estimation of marine productivity and can serve as a vital indicator of primary productivity [41]. The present results showed that Chla and round scad distribution were significantly correlated, and that the concentrated distribution area of round scad corresponded with Chla of 0–2 mg/m³. The Chla in the NSCS is generally low, showing a trend for gradual decrease from the nearshore to the offshore. In previous studies, the optimum surface chlorophyll-a concentration range for the distribution of round scad in the NSCS is mostly 0–2 mg/m³, and areas of high round scad abundance coincide with a surface chlorophyll-a concentration range of 0–1 mg/m³, which are consistent with the present results [10,23,42]. Chla concentration, as an important factor affecting the distribution of phytoplankton and zooplankton, did not show any significant correlation with zooplankton in the distribution of round scad in this study, therefore, how chlorophyll-a concentration further affects the distribution of round scad abundance also needs to be explored in more depth.

5. Conclusions

In this study, a GAM was applied to analyze the spatiotemporal variation in round scad abundance, and to investigate the spatiotemporal distribution characteristics of the species and their relationships with biotic and abiotic factors. The stock density of jack mackerel, water temperature, and latitude had highly significant effects on the distribution of round scad in the Hainan Island offshore area. In recent years, intense fishing pressure in the NCSC has caused severe declines in the fishery resources of major commercial species. As a result, the dominant species have gradually changed to small fish with short life-history cycles, especially small pelagic fishes, which are highly sensitive to changes in their environment. Substantial variability in the abundance of such fish has important implications for organisms of the higher trophic levels on which they feed. Therefore, it is crucial to understand the resource variation of the organisms feeding on round scad and the species occupying similar ecological niches to that of round scad, to elucidate the effects of climate change and fishing pressure on the distribution of round scad.

Author Contributions

Conceptualization, L.W., S.M. and D.S.; methodology, L.W. and C.Y.; formal analysis, C.Y. and Y.L.; resources and investigation, B.S. and L.W.; software and data curation, L.W.; writing—original draft preparation, L.W.; writing—review and editing and supervision, S.M. and D.S.; funding acquisition, D.S. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Acknowledgments

The authors would like to express their gratitude to all crew members on the Guibeiyu 69068 for their assistance with sample collection.

Conflicts of Interest

The authors declare no conflict of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Tables

Figure 1. Location of nekton survey stations in the Hainan Island offshore area. The colors and numbers indicate the water depth, based on the color bar.

Figure 2. Distribution of round scad abundance in the Hainan Island offshore area. The colors and numbers indicate the water depth, based on the color bar.

Figure 3. Pearson correlation coefficients among explanatory variables. * p < 0.05, ** p < 0.01. The colors and numbers indicate the correlation coefficients, based on the color bar.

Figure 4. Effect of biotic and abiotic factors on the distribution of round scad abundance in the Hainan Island offshore area.

Figure 5. Variation in the catch of round scad and jack mackerel in the northern South China Sea. Data are derived from Chinese Fishery Statistical yearbooks for the period 1979–2021 [8].

Table 1

Biotic and abiotic factors considered in this study.

	Explanatory Variable	Abbreviation
Abiotic factors	Longitude	Lon
	Latitude	Lat
	Sea surface temperature	SST
	Sea bottom temperature	SBT
	Sea surface salinity	SSS
	Sea bottom salinity	SBS
	Water depth	Depth
	Sea surface chlorophyll-a concentration	Chla
Biotic factors	Phytoplankton abundance	PA
	Zooplankton abundance	ZA
	Jack mackerel stock density	JM

Table 2

Variance inflation factor values for each explanatory variable.

Explanatory Variable	Variance Inflation Factor
Lon	2.36
Lat	1.95
SST	1.58
SBT	2.16
SSS	1.90
SBS	3.09
Chla	1.90
Depth	3.12
JM	1.16
PA	1.26
ZA	1.30
Sea	1.22

Table 3

Process of testing the fit of general additive models.

Model	AIC	Deviance Explained (%)	R 2
s(Lon) + s(Lat) + s(SST) + s(SBT) + s(SSS) + s(Chla) + s(JM) + s(PA) + s(ZA) + Sea	800.18	37.90	0.329
s(Lon) + s(Lat) + s(SST) + s(SBT) + s(SSS) + s(Chla) + s(JM) + s(ZA) + Sea	798.67	37.50	0.329
s(Lon) + s(Lat) + s(SST) + s(SBT) + s(SSS) + s(Chla) + s(JM) + Sea	797.66	37.20	0.329

AIC, Akaike information criterion; R², coefficient of determination.

Table 4

Parametric analysis of impact factors in the optimal generalized additive model.

Explanatory Variable	F	p-Value
s(Lon)	3.759	0.033 *
s(Lat)	5.409	0.005 **
s(SST)	7.539	0.005 **
s(SBT)	6.604	0.003 **
s(SSS)	2.183	0.171
s(Chla)	4.881	0.028 *
s(JM)	9.414	0.000 **

* p < 0.05, ** p < 0.01.

References

1. Palomera, I.; Olivar, M.P.; Salat, J.; Sabates, A.; Coll, M.; Garcia, A.; Morales-Nin, B. Small Pelagic Fish in the Nw Mediterranean Sea: An Ecological Review. Prog. Oceanogr.; 2007; 74, pp. 377-396. [DOI: https://dx.doi.org/10.1016/j.pocean.2007.04.012]

2. Tacon, A.G.J.; Metian, M. Fishing for Feed or Fishing for Food: Increasing Global Competition for Small Pelagic Forage Fish. Ambio. J. Hum. Environ.; 2009; 38, pp. 294-303.

3. Chavez, F.P.; Ryan, J.; Lluch-Cota, S.E.; NÑiquen, C.M. From anchovies to sardines and back: Multidecadal change in the Pacific Ocean. Science; 2003; 299, pp. 217-221. [DOI: https://dx.doi.org/10.1126/science.1075880]

4. Sakurai, Y.; Kiyofuji, H.; Saitoh, S.; Hiyama, Y. Changes in inferred spawning areas of Todarodes pacificus (Cephalopoda: Ommastrephidae) due to changing environmental conditions. ICES J. Mar. Sci.; 2000; 57, pp. 24-30. [DOI: https://dx.doi.org/10.1006/jmsc.2000.0667]

5. Wang, L.M.; Ma, S.Y.; Liu, Y.; Li, J.C.; Liu, S.G.; Lin, L.S.; Tian, Y.J. Fluctuations in the abundance of chub mackerel in relation to climatic/oceanic regime shifts in the northwest Pacific Ocean since the 1970s. J. Mar. Syst.; 2021; 218, 103541. [DOI: https://dx.doi.org/10.1016/j.jmarsys.2021.103541]

6. Deng, J.Y.; Zhao, C.Y. Marine Fisheries Biology; China Agriculture Press: Beijing, China, 1991; pp. 485-516.

7. Chen, Z.Z.; Qiu, Y.S. Assessment of the food-web structure, energy flows, and system attribute of northern South China Sea ecosystem. Acta Ecol. Sin.; 2009; 30, pp. 4855-4865.

8. Ministry of Agriculture and Rural Affairs of the People’s Republic of China. Chinese National Fishery Statistics Yearbook; China Agriculture Press: Beijing, China, 1979–2021.

9. Yu, J.; Lin, Z.N.; Chen, P.M.; Yao, L.J. Environmental factors affecting the spatiotemporal distribution of Decapterus maruadsi in the western Guangdong waters, China. Appl. Ecol. Environ. Res.; 2019; 17, pp. 8485-8499. [DOI: https://dx.doi.org/10.15666/aeer/1704_84858499]

10. He, L.X.; Fu, D.Y.; Li, Z.L.; Wang, H.; Sun, Y.; Liu, B.; Yu, G. Spatio-temporal distribution of Decapterus maruadsi and its relationship with environmental factors in the northwestern South China Sea. Prog. Fish. Sci.; 2022; 43, pp. 1-12.

11. Ma, S.; Cheng, J.; Li, J.; Liu, Y.; Wan, R.; Tian, Y. Interannual to Decadal Variability in the Catches of Small Pelagic Fishes from China Seas and its Responses to Climatic Regime Shifts. Deep-Sea Res. Pt II; 2019; 159, pp. 112-129. [DOI: https://dx.doi.org/10.1016/j.dsr2.2018.10.005]

12. Zhang, K.; Li, M.; Li, J.J.; Sun, M.S.; Xu, Y.W.; Cai, Y.C.; Chen, Z.Z.; Qiu, Y.S. Climate-induced small pelagic fish blooms in an overexploited marine ecosystem of the South China Sea. Ecol. Indic.; 2022; 145, 109598. [DOI: https://dx.doi.org/10.1016/j.ecolind.2022.109598]

13. Ma, S.Y.; Liu, Y.; Li, J.C.; Fu, C.H.; Ye, Z.J.; Sun, P.; Yu, H.Q.; Cheng, J.H.; Tian, Y.J. Climate-induced long-term variations in ecosystem structure and atmosphere-oceanecosystem processes in the yellow Sea and East China Sea. Prog. Oceanogr.; 2019; 175, pp. 183-197. [DOI: https://dx.doi.org/10.1016/j.pocean.2019.04.008]

14. Li, Z.L.; Zhang, W.X.; He, X.B.; Yan, Y.R. Feeding Ecology and Feeding Competition Between Decapterus maruadsi and Trachurus japonicus in Autumn in the Beibu Gulf, South China Sea. J. Guangdong Ocean Univ.; 2019; 39, pp. 79-86.

15. Sparre, P.; Venema, S.C. Introduction to Tropical Fish Stock Assessment Part I. Manual. FAO Fisheries Technical Paper; 2nd ed. FAO: Rome, Italy, 1998.

16. Wang, X.; Qiu, Y.; Du, F.; Liu, W.; Sun, D.; Chen, X.; Yuan, W.; Chen, Y. Roles of fishing and climate change in long-term fish species succession and population dynamics in the outer Beibu Gulf, South China Sea. Acta Oceanol. Sin.; 2019; 38, pp. 1-8. [DOI: https://dx.doi.org/10.1007/s13131-019-1484-5]

17. Yentsch, C.S.; Menzel, D.W. A method for the determination of phytoplankton chlorophyll and phaeophytin by fluorescence. Deep-Sea Res.; 1963; 10, pp. 221-231. [DOI: https://dx.doi.org/10.1016/0011-7471(63)90358-9]

18. Hastie, T.J.; Tibshirani, R.J. Generalized Additive Models; Chapman and Hall: London, UK, 1990.

19. Jiang, R.J.; Xu, H.X.; Jin, H.W.; Zhou, Y.D.; He, Z.T. Feeding habits of blue mackerel scad Decapterus maruadsi Temminck et Schlegel in the East China Sea. J. Fish. China; 2012; 36, pp. 216-227.

20. Sagarese, S.R.; Frisk, M.G.; Cerrat, R.M.; Sosebee, K.A.; Musick, J.A.; Rago, P.J. Application of generalized additive models to examine ontogenetic and seasonal distributions of spiny dogfish (squalus acanthias) in the northeast (US) shelf large marine ecosystem. Can. J. Fish. Aquat. Sci.; 2014; 71, pp. 847-877. [DOI: https://dx.doi.org/10.1139/cjfas-2013-0342]

21. Burnham, K.P.; Anderson, D.R. Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach; Springer Science and Business Media: New York, NY, USA, 2003; 488.

22. Chen, G.B.; Li, Y.Z.; Chen, P.M. A study on spawning ground of blue mackerel (Decapterus Maruadsi). J. Trop. Oceanogr.; 2003; 22, pp. 22-28.

23. Fan, J.T.; Huang, Z.R.; Xu, Y.W.; Sun, M.S.; Chen, G.B.; Chen, Z.Z. Habitat model analysis for Decapterus Maruadsi in Northern South China Sea based on remote sensing data. T. Oceanol. Limnol.; 2018; 3, pp. 142-147.

24. Wang, X.X.; Yu, J.; Li, Y.Z.; Chen, G.B.; Huang, M.F. The relationship between major upwelling and the upwelling fishing grounds in the South China Sea. Mar. Sci.; 2015; 39, pp. 131-137.

25. Cury, P.; Bakun, A.; Crawford, R.; Jarre, A.; Quinones, R.; Shannon, L.; Verheye, H. Small pelagic in upwelling systems: Patterns of interaction and structural changes in “wasp-waist” ecosystem. Adv. Exp. Med. Biol.; 2000; 559, pp. A341-A349. [DOI: https://dx.doi.org/10.1006/jmsc.2000.0712]

26. Zhang, W.C.; Yu, H.Q.; Ye, Z.J.; Tian, Y.J.; Liu, Y.; Li, J.C.; Xing, Q.W.; Jiang, Y.Q. Spawning strategy of Japanese anchovy Engraulis japonicus in the coastal Yellow Sea: Choice and dynamics. Fish. Oceanogr.; 2020; 30, pp. 366-381. [DOI: https://dx.doi.org/10.1111/fog.12523]

27. Ju, P.L.; Chen, M.R.; Cheung, W.; Tian, Y.J.; Yang, S.Y.; Sun, P.; Jiang, C.P.; Lu, Z.B. Modelling the structure and functioning of an upwelling ecosystem in the Southern Taiwan Strait, China. J. Marine Syst.; 2022; 226, 103666. [DOI: https://dx.doi.org/10.1016/j.jmarsys.2021.103666]

28. Xu, Z.T.; Li, S.Y.; Hu, J.T.; Wang, S.Y.; Wang, B.; Guo, M.X.; Geng, B.X. Summer phytoplankton responses to upwelling and river plume in northern South China Sea. J. Trop. Oceanogr.; 2018; 37, pp. 92-103.

29. Qiu, Y.S.; Lin, Z.J.; Wang, Y.Z. Responses of fish production to fishing and climate variability in the northern South China Sea. Prog. Oceanogr.; 2010; 85, pp. 197-212. [DOI: https://dx.doi.org/10.1016/j.pocean.2010.02.011]

30. Chen, Q.X.; Xiong, Z.Y.; Tan, Z.M.; Shi, W.Q.; Peng, J.Y.; Li, Y.Q. Comparison between the catches (Trachurus japonicus and Decapterus maruadsi) around two LED lamps. South China Fish. Sci.; 2013; 9, pp. 80-84.

31. Li, Y.N.; Yang, B.Z.; Zhang, P.; Li, J.; Wang, T.; Yan, L. Size selectivity of codend mesh size of trawl for Decapterus maruadsi in northern part of South China Sea. South China Fish. Sci.; 2022; 18, pp. 170-176.

32. Nakayama, S.-I.; Takasuka, A.; Ichinokawa, M.; Okamura, H. Climate change and interspecific interactions drive species alternations between anchovy and sardine in the western North Pacific: Detection of causality by convergent cross mapping. Fish. Oceanogr.; 2018; 27, pp. 312-322. [DOI: https://dx.doi.org/10.1111/fog.12254]

33. Sugihara, G.; May, R.; Ye, H.; Hsieh, C.H.; Deyle, E.; Fogarty, M.; Munch, S. Detecting causality in complex ecosystems. Science; 2012; 338, pp. 496-500. [DOI: https://dx.doi.org/10.1126/science.1227079]

34. Takasuka, A.; Oozeki, Y.; Kubota, H. Multi-species regime shifts reflected in spawning temperature optima of small pelagic fish in the western North Pacific. Mar. Ecol. Prog. Ser.; 2008; 360, pp. 211-217. [DOI: https://dx.doi.org/10.3354/meps07407]

35. Oozeki, Y.; Carranza, M.N.; Takasuka, A.; Dejo, P.A.; Kuroda, H.; Malagas, J.T.; Okunishi, T.; Espinoza, L.V.; Aguilar, D.G.; Okamura, H. et al. Synchronous multi-species alternations between the northern Humboldt and Kuroshio Current systems. Deep-Sea Res. Pt II; 2019; 159, pp. 11-21. [DOI: https://dx.doi.org/10.1016/j.dsr2.2018.11.018]

36. Takasuka, A.; Oozeki, Y.; Aoki, I. Optimal growth temperature hypothesis: Why do anchovy flourish and sardine collapse or vice versa under the same ocean regime. Can. J. Fish. Aquat. Sci.; 2007; 64, pp. 768-776. [DOI: https://dx.doi.org/10.1139/f07-052]

37. Takasuka, A.; Oozeki, Y.; Kubota, H.; Lluch-Cota, S.E. Contrasting spawning temperature optima: Why are anchovy and sardine regime shifts synchronous across the North Pacific?. Prog. Oceanogr.; 2008; 77, pp. 225-232. [DOI: https://dx.doi.org/10.1016/j.pocean.2008.03.008]

38. Oozeki, Y.; Okunishi, T.; Takasuka, A.; Ambe, D. Variability in transport processes of Pacific saury Cololabis saira larvae leading to their broad dispersal: Implications for their ecological role in the western North Pacific. Prog. Oceanogr.; 2015; 138, pp. 448-458. [DOI: https://dx.doi.org/10.1016/j.pocean.2014.05.011]

39. Liu, S.G.; Liu, Y.; Fu, C.H.; Yan, L.X.; Xu, Y.; Wan, R.; Li, J.C.; Tian, Y.J. Using novel spawning ground indices to analyze the effects of climate change on pacific saury abundance. J. Mar. Syst.; 2019; 191, pp. 13-23. [DOI: https://dx.doi.org/10.1016/j.jmarsys.2018.12.007]

40. Zou, X.R.; Xu, L.X. A preliminary study on biology of chub mackerel (Scomber Japonicus) and round scad (Decapterus maruadsi) in the Southern East China Sea. Marine Fish.; 2001; 15, pp. 117-121.

41. Xu, W.L. Biooptical Inversion and Environmental Factors of Chlorophyll–A Concentration Profile in the South China Sea; University of Chinese Academy of Sciences: Beijing, China, 2018.

42. Liu, Z.N. Remote Sensing of Environment and Spatiotemporal Distribution of Light Falling-Net Fishing Ground in the Western Guangdong Waters; Shanghai Ocean University: Shanghai, China, 2019.