Relationship between the Distribution of Broodstock and Vorticity of Spawning Grounds of Four Major Chinese Carps in the Middle Reaches of the Yangtze River during Ecological Operation of the Three Gorges Dam

: Hydrodynamic characteristics of spawning grounds are important factors affecting the spawning of four major Chinese carps ( Mylopharyngodon piceus , Ctenopharyngodon idella , Hypophthalmichthys molitrix , and Aristichthys nobilis ). To investigate the relationship between the preferred hydrodynamic characteristics of spawning sites and the response of ﬁsh spawning behavior, we monitored the ﬂow ﬁeld of spawning sites during ecological operation of the Three Gorges Dam (i.e., man-made ﬂood regulation) in 2014 and 2015. We used the data to explore the correlation between vorticity changes in spawning grounds and the spawning amount. Pearson correlation coefﬁcients of the average vorticity in all cross-sections of the Yidu spawning ground and spawning amount in 2014 and 2015 were 0.730 and 0.822, respectively, indicating a signiﬁcant positive correlation between vorticity and spawning activity. In some speciﬁc regions, this correlation was even stronger (Pearson correlations of the regional vorticity and egg production were >0.95). To further corroborate and analyze the relationship between these speciﬁc regions and the distribution of broodstock during the breeding season, an ultrasonic telemetry test of broodstock was conducted in the Yidu spawning ground in 2016. The results showed that the broodstocks were concentrated in the reach near the Quantong Pier (~76 km from the Three Gorges Dam). These regions were consistent with areas of river characterized by highly correlated vorticity and egg production levels, suggesting that these regions are areas preferred by four major Chinese carps for spawning.


Introduction
Black carp (Mylopharyngodon piceus), grass carp (Ctenopharyngodon idella), silver carp (Hypophthalmichthys molitrix), and bighead carp (Aristichthys nobilis) are collectively referred to as  Research conducted combined monitoring of flow field and early-stage fish resources for the Yidu spawning ground. The measurement was consistent with the Three Gorges ecological operation. Flow field measurements were carried out every 500 m, covering the Chenjiahe-Yunchi section, for a total of 24 sections (Figure 1).

Flow Field Measurement of Spawning Sites of FMCC
The flow field measurement system comprised a walking acoustic Doppler velocity profiler (ADCP, manufactured by RDI, USA, 600 kHz, depth of measurement range 0.7-75 m, suitable for topography and flow measurement in the middle reaches of the Yangtze River), a GPS satellite positioning system (China CHCNAV Company, i80 GNSS receiver, plane accuracy: ±(8 + 1 × 10 -6 × D) mm), a recording computer, ADCP matching Win River (version 1.06) measurement software, and Huace navigation software. During each measurement, the ADCP and GNSS receivers and recording computer were mounted on a boat, which navigated a cross-section of the river. Data, including the profile velocity, bed topography, and GPS position, were collected continuously ( Figure 2). The ADCP measurement divides the trajectory section into microsections, each of which is further divided into several bins in a vertical direction; the data, including velocity, depth, and positional coordinates, are stored in each bin.  Research conducted combined monitoring of flow field and early-stage fish resources for the Yidu spawning ground. The measurement was consistent with the Three Gorges ecological operation. Flow field measurements were carried out every 500 m, covering the Chenjiahe-Yunchi section, for a total of 24 sections (Figure 1).

Flow Field Measurement of Spawning Sites of FMCC
The flow field measurement system comprised a walking acoustic Doppler velocity profiler (ADCP, manufactured by RDI, USA, 600 kHz, depth of measurement range 0.7-75 m, suitable for topography and flow measurement in the middle reaches of the Yangtze River), a GPS satellite positioning system (China CHCNAV Company, i80 GNSS receiver, plane accuracy: ±(8 + 1 × 10 -6 × D) mm), a recording computer, ADCP matching Win River (version 1.06) measurement software, and Huace navigation software. During each measurement, the ADCP and GNSS receivers and recording computer were mounted on a boat, which navigated a cross-section of the river. Data, including the profile velocity, bed topography, and GPS position, were collected continuously ( Figure 2). The ADCP measurement divides the trajectory section into microsections, each of which is further divided into several bins in a vertical direction; the data, including velocity, depth, and positional coordinates, are stored in each bin.

Calculation of Mean Vorticity in Spawning Ground Section
Vortex is a special form of velocity gradient. There are various scales and forms of vortices under the combined effects of river terrain and water flow. Here, we only focused on vortices in a cross-section of the river, defining this type of vortex as a section vortex, the axis of which was parallel to the flow direction ( Figure 3). In natural rivers, it is difficult to directly measure vorticity. This study used VB.NET to independently program code and read the flow velocity vector, depth, and GPS of each measurement section and bin through the flow field data of ADCP. The coordinates and other data were used to calculate the average vorticity of the cross-section based on the velocity component of the adjacent bin. Details of the calculation method can be found in [9,10].
If the unit vorticity of a region contains both positive and negative vorticity, the simple superposition of the regional vorticity will obscure the complexity of the flow field itself. Therefore, to avoid the cancellation of the positive and negative vorticity in the calculation process, we summed the regional vorticity integrals after taking the absolute value.
The mean vorticity of the section was calculated using Equations (1)-(3): where x is the flow direction; y is the transverse direction; z is the vertical direction; Ω is the vorticity at a certain position; x Ω is the vorticity along x-direction (the vortex axis was parallel to the flow direction); ABS Ω is the average vorticity of the cross section; ABS Γ is the absolute circulation of cross-section; TOT A is the area of the cross section; u Δ , v Δ , w Δ are the flow rates

Calculation of Mean Vorticity in Spawning Ground Section
Vortex is a special form of velocity gradient. There are various scales and forms of vortices under the combined effects of river terrain and water flow. Here, we only focused on vortices in a cross-section of the river, defining this type of vortex as a section vortex, the axis of which was parallel to the flow direction ( Figure 3). In natural rivers, it is difficult to directly measure vorticity. This study used VB.NET to independently program code and read the flow velocity vector, depth, and GPS of each measurement section and bin through the flow field data of ADCP. The coordinates and other data were used to calculate the average vorticity of the cross-section based on the velocity component of the adjacent bin. Details of the calculation method can be found in [9,10].
If the unit vorticity of a region contains both positive and negative vorticity, the simple superposition of the regional vorticity will obscure the complexity of the flow field itself. Therefore, to avoid the cancellation of the positive and negative vorticity in the calculation process, we summed the regional vorticity integrals after taking the absolute value.
The mean vorticity of the section was calculated using Equations (1)- (3): where x is the flow direction; y is the transverse direction; z is the vertical direction; Ω is the vorticity at a certain position; Ω x is the vorticity along x-direction (the vortex axis was parallel to the flow direction); Ω ABS is the average vorticity of the cross section;Γ ABS is the absolute circulation of cross-section; A TOT is the area of the cross section; ∆u, ∆v, ∆w are the flow rates of discrete cells in x, y, and z directions, respectively; ∆y∆z is the area of cells in the cross-sectional direction.
Water 2018, 10, x FOR PEER REVIEW 6 of 15 of discrete cells in x, y, and z directions, respectively; y z Δ Δ is the area of cells in the cross-sectional direction.

Positioning of Broodstock during Spawning Based on Ultrasonic Telemetry
Ultrasonic telemetry is an advanced technique for tracking and locating aquatic animals and has been widely used in fish behavior monitoring [15][16][17], fish habitat studies [18,19], and for determining the relationship between aquatic organisms and their environment [20,21]. The ultrasonic telemetry system comprises a transmitter device and a monitoring device. The launching device is an ultrasound tag that is implanted in, or attached to, an aquatic animal. The ultrasound tag used in this study was V9-2X (77 tags, serial numbers 55127-55203, diameter 9.0 mm, length 29.0 mm, weight 2.9 g, battery life 910 days, nominal delay of 180 s; Mai Vision Technologies Co., Ltd., Shanghai, China) and V9-4X (30 tags, serial numbers 34540-34569, diameter 8.0 mm, length 20.5 mm, weight 0.9 g; battery life 210 days, nominal delay of 120 s; Mai Vision Technologies Co., Ltd., Shanghai, China). The emission signal frequency was 69.0 kHz. The monitoring device is an ultrasonic signal receiver with the animal position and other information (e.g., water depth, temperature, etc.) transmitted from the ultrasonic signage plate. The model of the ultrasonic signal receiver used in this study was VR2W (Vemco Ltd., Shad Bay, NS, Canada). The effective detection distance of the receiver in the Yangtze River was 490 m.
To verify the validity of the calculated spawning grounds of FMCC, the distribution of, and preference for, spawning grounds were further determined. From June to July, 2016, based on the results of resource monitoring during early 2014-2015, ultrasonic telemetry tests of broodstock fish was conducted within the Yidu spawning ground ( Figure 1). A total of 107 broodstock of FMCC were marked during the trial (20 M. piceus, 29 C. idella, 37 H. molitrix, and 22 A. nobilis). The mean body length of the marked fish was 78.76 ± 6.83 cm, and the average body weight was 12.95 ± 2.46 kg. Ultrasound tags were implanted following the surgical methodology described by [22]. After the operation, the broodstock was kept in the laboratory temporarily (retention time 33-35 days). On 1 June 2016, the broodstock was released in the Zhijiang River section, ~60 km downstream of Yunchi. Nine ultrasonic signal receivers were placed in the Yidu spawning ground ( Figure 1).

Positioning of Broodstock during Spawning Based on Ultrasonic Telemetry
Ultrasonic telemetry is an advanced technique for tracking and locating aquatic animals and has been widely used in fish behavior monitoring [15][16][17], fish habitat studies [18,19], and for determining the relationship between aquatic organisms and their environment [20,21]. The ultrasonic telemetry system comprises a transmitter device and a monitoring device. The launching device is an ultrasound tag that is implanted in, or attached to, an aquatic animal. The ultrasound tag used in this study was V9-2X (77 tags, serial numbers 55127-55203, diameter 9.0 mm, length 29.0 mm, weight 2.9 g, battery life 910 days, nominal delay of 180 s; Mai Vision Technologies Co., Ltd., Shanghai, China) and V9-4X (30 tags, serial numbers 34540-34569, diameter 8.0 mm, length 20.5 mm, weight 0.9 g; battery life 210 days, nominal delay of 120 s; Mai Vision Technologies Co., Ltd., Shanghai, China). The emission signal frequency was 69.0 kHz. The monitoring device is an ultrasonic signal receiver with the animal position and other information (e.g., water depth, temperature, etc.) transmitted from the ultrasonic signage plate. The model of the ultrasonic signal receiver used in this study was VR2W (Vemco Ltd., Shad Bay, NS, Canada). The effective detection distance of the receiver in the Yangtze River was 490 m.
To verify the validity of the calculated spawning grounds of FMCC, the distribution of, and preference for, spawning grounds were further determined. From June to July, 2016, based on the results of resource monitoring during early 2014-2015, ultrasonic telemetry tests of broodstock fish was conducted within the Yidu spawning ground ( Figure 1). A total of 107 broodstock of FMCC were marked during the trial (20 M. piceus, 29 C. idella, 37 H. molitrix, and 22 A. nobilis). The mean body length of the marked fish was 78.76 ± 6.83 cm, and the average body weight was 12.95 ± 2.46 kg. Ultrasound tags were implanted following the surgical methodology described by [22]. After the operation, the broodstock was kept in the laboratory temporarily (retention time 33-35 days). On 1 June 2016, the broodstock was released in the Zhijiang River section,~60 km downstream of Yunchi. Nine ultrasonic signal receivers were placed in the Yidu spawning ground (Figure 1).

Relationship between Spawning of FMCC and Cross-Section Mean Vorticity of Spawning Ground
According to the monitoring data of the Yangtze River Fisheries Research Institute from 2014 and early 2015, there was a response relationship between the spawning amount of FMCC and discharge of Gezhouba from 1 to 10 June 2014, and 21 to 30 June 2015 (Figures 4 and 5). The water level of the river rise continuously from 4 to 7 June 2014 and from 25 to 28 June 2015, as a result of the Three Gorges ecological operation. Spawning started from day 2 of the increasing water level and peaked on day 3 ( Figure 4). On day 4 of the increasing water level, the spawning amount started to decrease, although it remained at a relatively high level. Two days after the ecological operation (8-9 June 2014), the flood dropped, and spawning amount decreased significantly, although it did not stop completely. On 21-22 June 2015, the water volume fluctuated, and small-scale spawning behavior of the fish started ( Figure 5). On 26-27 June, the water level had been increasing for 2 days, and the flood had reached its peak. Large-scale spawning behavior occurred on 27 June.
In total, the vorticities of 24 cross-sections were calculated daily using Equation (3). The mean value was defined as the average vorticity for all cross-sections of the river. During the monitoring periods of the flow field (5-8 June 2014 and 26-28 June 2015), there was a response relationship between the spawning amount of FMCC and the average vorticity of all cross-sections of the river (Figures 6  and 7). Figure 6 shows that, from 5 to 6 June 2014, the average vorticity of all cross-sections of the river increased from 0.39/s to 0.46/s, which significantly increased the corresponding spawning amount of FMCC. From 6 to 8 June, the average vorticity of all cross-sections of the river was stable at a relatively high level (0.45-0.46/s), and the corresponding fish spawned continuously for 2 consecutive days, with a high spawning amount. Figure 7 shows that the average vorticity in 2015 reached the maximum (0.48/s) on 27 June, and the spawning volume of the FMCC also peaked. However, no spawning amount was observed on 26 or 28 June.
Studies have shown that many biological behaviors and parameters are subject to, or approximate to, a normal distribution [23]. Therefore, we assumed that the spawning behavior of FMCC under certain flow conditions was also subject to a normal distribution. On the basis of this hypothesis, the Pearson correlation analysis method was applied to analyze the relationship between the spawning amount of FMCC and vorticity. The Pearson correlation coefficients of the spawning of FMCC from 5 to 8 June 2014 and from 26 to 28 June 2015 and the average vorticity of all cross-sections of the river were 0.730 and 0.822, respectively, indicating a strong positive correlation between the spawning of FMCC and the average vorticity of all cross-sections of the river. The cross-section vorticity also had a stimulating effect on the spawning behavior of the fish. From a biological point of view, a large vorticity of the cross-section is conducive to fertilization of the eggs, maintaining the position of the fertilized eggs in the water so that they do not sink and die. Therefore, it is biologically significant that FMCC chose to continue spawning when the vorticity of cross-section in the spawning ground was relatively large.
Water 2018, 10, x FOR PEER REVIEW 7 of 15 peaked on day 3 ( Figure 4). On day 4 of the increasing water level, the spawning amount started to decrease, although it remained at a relatively high level. Two days after the ecological operation (8-9 June 2014), the flood dropped, and spawning amount decreased significantly, although it did not stop completely. On 21-22 June 2015, the water volume fluctuated, and small-scale spawning behavior of the fish started ( Figure 5). On 26-27 June, the water level had been increasing for 2 days, and the flood had reached its peak. Large-scale spawning behavior occurred on 27 June. In total, the vorticities of 24 cross-sections were calculated daily using Equation (3). The mean value was defined as the average vorticity for all cross-sections of the river. During the monitoring periods of the flow field (5-8 June 2014 and 26-28 June 2015), there was a response relationship between the spawning amount of FMCC and the average vorticity of all cross-sections of the river (Figures 6 and 7). Figure 6 shows that, from 5 to 6 June 2014, the average vorticity of all cross-sections of the river increased from 0.39/s to 0.46/s, which significantly increased the corresponding spawning amount of FMCC. From 6 to 8 June, the average vorticity of all cross-sections of the river was stable at a relatively high level (0.45-0.46/s), and the corresponding fish spawned continuously for 2 consecutive days, with a high spawning amount. Figure 7 shows that the average vorticity in 2015 reached the maximum (0.48/s) on 27 June, and the spawning volume of the FMCC also peaked. However, no spawning amount was observed on 26 or 28 June.
Studies have shown that many biological behaviors and parameters are subject to, or approximate to, a normal distribution [23]. Therefore, we assumed that the spawning behavior of FMCC under certain flow conditions was also subject to a normal distribution. On the basis of this hypothesis, the Pearson correlation analysis method was applied to analyze the relationship between the spawning amount of FMCC and vorticity. The Pearson correlation coefficients of the spawning of FMCC from 5 to 8 June 2014 and from 26 to 28 June 2015 and the average vorticity of all cross-sections of the river were 0.730 and 0.822, respectively, indicating a strong positive correlation between the spawning of FMCC and the average vorticity of all cross-sections of the river. The cross-section vorticity also had a stimulating effect on the spawning behavior of the fish. From a biological point of view, a large vorticity of the cross-section is conducive to fertilization of the eggs, maintaining the position of the fertilized eggs in the water so that they do not sink and die. Therefore, it is biologically significant that FMCC chose to continue spawning when the vorticity of cross-section in the spawning ground was relatively large.

Relationship between the Local Average Vorticity and Spawning Amount of the Fish at Each Section of the Spawning Ground during Ecological Operation of the Three Gorges Dam
To further study the correlation between the local average vorticity of each section of the Yidu spawning ground and spawning amount of the fish during the ecological operation of the Three

Relationship between the Local Average Vorticity and Spawning Amount of the Fish at Each Section of the Spawning Ground during Ecological Operation of the Three Gorges Dam
To further study the correlation between the local average vorticity of each section of the Yidu spawning ground and spawning amount of the fish during the ecological operation of the Three

Relationship between the Local Average Vorticity and Spawning Amount of the Fish at Each Section of the Spawning Ground during Ecological Operation of the Three Gorges Dam
To further study the correlation between the local average vorticity of each section of the Yidu spawning ground and spawning amount of the fish during the ecological operation of the Three

Relationship between the Local Average Vorticity and Spawning Amount of the Fish at Each Section of the Spawning Ground during Ecological Operation of the Three Gorges Dam
To further study the correlation between the local average vorticity of each section of the Yidu spawning ground and spawning amount of the fish during the ecological operation of the Three Gorges dam, the measured sections were equally divided into three subregions: the left and right bank, and middle of the river (Figure 8). The cross-section average vorticity of each subregion during the ecological operation of the Three Gorges dam was calculated and Pearson correlation analysis was used to calculate the correlation coefficient between variation in the vertical vorticity and changes in spawning amount in the three subregions (Tables 1 and 2 Figure 8b). These results show that the special hydrodynamic conditions (increased vorticity) generated during the ecological operation may stimulate spawning and reproduction of the fish, resulting in a corresponding increase in spawning amount. According to the 2-year flow field monitoring data, there are many areas where the local cross-sectional average vorticity is highly related to fish spawning.

Relationship between the Distribution of Marked Broodstock during the Breeding Period and Areas with High Correlation between Vorticity and Spawning
After the marked broodstocks were released 60 km downstream of the Yidu spawning ground, some were traced back to the Yidu spawning ground. The nine receivers received signals from 30 marked broodstock, the tracking numbers of which were 34545, 34559, 34567, 55135, 55162, 55164, 55185, and 55195, corresponding to 1 C. idella, 3 H. molitrix, and 4 A. nobilis. The number of fish monitored and the number of times they were monitored during the breeding period are shown in Figure 9. The results showed that the receiver at position yd05 captured 13 signals from marked fish, and the number of fixes exceeded 400. The receiver at position yd06 captured 15 signals from marked fish, and the number of fixes was ~200.

Relationship between the Distribution of Marked Broodstock during the Breeding Period and Areas with High Correlation between Vorticity and Spawning
After the marked broodstocks were released 60 km downstream of the Yidu spawning ground, some were traced back to the Yidu spawning ground. The nine receivers received signals from 30 marked broodstock, the tracking numbers of which were 34545, 34559, 34567, 55135, 55162, 55164, 55185, and 55195, corresponding to 1 C. idella, 3 H. molitrix, and 4 A. nobilis. The number of fish monitored and the number of times they were monitored during the breeding period are shown in Figure 9. The results showed that the receiver at position yd05 captured 13 signals from marked fish, and the number of fixes exceeded 400. The receiver at position yd06 captured 15 signals from marked fish, and the number of fixes was~200.
These results indicated that the broodstock was mainly distributed around areas near receivers yd05 and yd06 during the breeding period, and the corresponding hydrodynamic characteristics were relatively specific. Comprehensive analysis of Figures 1 and 9 showed that yd05 was located on the left bank between sections 16 and 17, and that yd06 was located on the left bank of section 18

Issues with Small Sample Size and Analysis
According to our field experience over several years, each FMCC spawning event lasts for 1-7 days. From 1900 to 1980 (the Yangtze River was a natural river before the Gezhouba dam was built), the average number of times that the water level increased in the middle reaches of the Yangtze River during the FMCC breeding period (May-July) was 7.1, and the average number of days each increase lasted for was 4.5. From 1981 to 2002 (Gezhouba was operated independently before the closure of the Three Gorges Dam), these figures were 7.5 and 4.5, respectively. From 2003 to 2014 (during the period of the joint operation of Gezhouba Dam and Three Gorges Dam), these figures were 6.2 and 4.2, respectively (figures all based on hydrological data from the Yichang hydrological station from 1900 to 2014). If statistics are based on a single day, regardless of whether causes are natural or have been affected by large-scale water conservancy and hydropower projects, then the sample size of the data showing the amount of spawning and vorticity is no more than 10. The relationship between spawning behavior and water flow suffers from a small sample size. Such small sample problems also occur in other research fields (e.g., pathological analysis of rare diseases, DNA, face recognition, chemical sensors, pattern recognition of incomplete information, etc.), [24][25][26][27][28]. Thus, there might be limitations to our analysis of the results of vorticity and spawning amount, based on our small sample size. In fact, small samples could increase the randomness of correlation analysis and reduce the repeatability of analysis results. However, our observations provide information about fish behavior and flow conditions, even if this information is incomplete. Actually, the correlation between the local vorticity and the spawning amount, revealed by one

Issues with Small Sample Size and Analysis
According to our field experience over several years, each FMCC spawning event lasts for 1-7 days. From 1900 to 1980 (the Yangtze River was a natural river before the Gezhouba dam was built), the average number of times that the water level increased in the middle reaches of the Yangtze River during the FMCC breeding period (May-July) was 7.1, and the average number of days each increase lasted for was 4.5. From 1981 to 2002 (Gezhouba was operated independently before the closure of the Three Gorges Dam), these figures were 7.5 and 4.5, respectively. From 2003 to 2014 (during the period of the joint operation of Gezhouba Dam and Three Gorges Dam), these figures were 6.2 and 4.2, respectively (figures all based on hydrological data from the Yichang hydrological station from 1900 to 2014). If statistics are based on a single day, regardless of whether causes are natural or have been affected by large-scale water conservancy and hydropower projects, then the sample size of the data showing the amount of spawning and vorticity is no more than 10. The relationship between spawning behavior and water flow suffers from a small sample size. Such small sample problems also occur in other research fields (e.g., pathological analysis of rare diseases, DNA, face recognition, chemical sensors, pattern recognition of incomplete information, etc.), [24][25][26][27][28]. Thus, there might be limitations to our analysis of the results of vorticity and spawning amount, based on our small sample size. In fact, small samples could increase the randomness of correlation analysis and reduce the repeatability of analysis results. However, our observations provide information about fish behavior and flow conditions, even if this information is incomplete. Actually, the correlation between the local vorticity and the spawning amount, revealed by one survey, might be random. But when a region showing strong correlation, as revealed by multiple survey results, coincides with the frequent activity region of marked parent fish, it is unlikely to be explained simply as a random phenomenon. Our field survey indicates a problem associated with fish spawning behavior and spawning field selection preference, implying that there must be a certain relationship between fish behavior and the flow conditions of the spawning ground. Here, we have also used an analytical method to establish the relationship between the spawning amount and vorticity to reveal the hydrodynamic triggers affecting fish spawning. Although it requires further development, the current method is of use. Given these shortcomings associated with the current study, a clearer relationship between fish behavior and flow conditions would be attainable by: (1) increasing the number of samples by reducing the statistical timescale of each spawning event. The sampling interval of each spawning event on a daily scale could be changed to a sampling interval of 2 or 4 h; and (2) increasing the number of investigations of spawning events and verifying the repeatability of the results.

Conclusions
Hydrodynamic changes resulting from the continuous increase in water levels in spawning grounds are important factors that stimulate the spawning behavior of FMCC. 'Bubbling water' is the macroscopic expression of vortex, and is also a previously used qualitative description of the characteristics of water flow in the spawning of FMCC (used in the 1982 survey of spawning grounds of four major Chinese carp in the Yangtze River [29]). Based on previous reports, average cross-section vorticity was selected as a hydrodynamic characteristic affecting the choice of spawning grounds, and the flow field distribution of the Yidu spawning ground was measured during the 2014 and 2015 ecological operation of the Three Gorges dam. In addition, the average cross-section vorticities of the spawning grounds of FMCC were calculated. Simultaneous monitoring of early-stage fish resources was performed downstream to analyze the response relationship between the spawning amount of FMCC and the average vorticity of the spawning ground during increasing water levels. In 2016, ultrasonic wave telemetry of broodstock was conducted to capture the location information of breeding fish and to reveal the hydrodynamic characteristics of the positions that the FMCC preferred, enabling the following conclusions to be reached: . This indicated that the cross-section vorticity changed in these areas. There was a significant positive correlation between the changes in the spawning amount of FMCC and average vorticity of cross-section. The special hydrodynamic conditions (increased vorticity) generated in these areas during the ecological operation of the dam could stimulate the spawning and reproduction of FMCC, resulting in a corresponding increase in spawning amount.

3.
Ultrasonic telemetry tests of broodstock fish in 2016 showed that the main distribution region and intense-activity areas of the broodstock fish in the spawning ground of Yidu were within a 490-m radius of receivers yd05 and yd06, mainly around the Quantong Piers of the Three Gorges Dam. In line with the high correlation between vorticity and spawning amount in 2014 and 2015, it is inferred that the river section near the Quantong Piers of the Three Gorges dam is a preferred area for FMCC spawning.
Our analysis of the relationship between fish spawning behavior and water flow in this paper suffers from our small sample size. Although the method of analysis used has some limitations, it can be regarded as beneficial. The results of this research indicate that a continuous increase in water level can effectively stimulate the spawning behavior of fish, such as FMCC. The regulation of, and storage by, the reservoir negatively impacts the natural seasonal increases in water level of the river, reduces the stimulating effect of the flow on the spawning behavior of fish. Therefore, reservoirs should be managed to create multiple artificial flood peaks during the fish breeding period.