The Three Gorges Dam (TGD), located in the main stream of the Yangtze River, is the world's largest hydroelectric station in terms of installed power capacity. It was demonstrated that the TGD had caused considerable modifications in the downstream freshwater discharge due to its seasonal operation mode of multiple utilisation for flood control, irrigation, and power generation. To understand the impacts of the freshwater regulation of the TGD, an analytical model is adopted to explore how the operation of the TGD may affect the spatial–temporal patterns of tide–river dynamics in the Yangtze River estuary. We evaluated the effect of the TGD by comparing the changes in major tide–river dynamics in the post-TGD period (2003–2014) with those in the pre-TGD period (1979–1984). The results indicate that the strongest impacts occurred during the autumn and winter, corresponding to a substantial reduction in freshwater discharge during the wet-to-dry transition period and slightly increased discharge during the dry season. The underlying mechanism leading to changes in the tide–river dynamics lies in the alteration of freshwater discharge, while the impact of geometric change is minimal. Overall, the results suggest that the spatial–temporal pattern of tide–river dynamics is sensitive to the freshwater regulation of the TGD, so that the ecosystem function of the estuary may undergo profound disturbances. The results obtained from this study can be used to set scientific guidelines for water resource management (e.g. navigation, flood control, salt intrusion) in dam-controlled estuarine systems.

Estuaries are transition zones where the river meets the ocean (Savenije, 2012). Tide–river interactions, a result of both hydrologic drivers and geomorphic constraints, are highly dynamic in estuaries (Buschman et al., 2009; Sassi and Hoitink, 2013; Guo et al., 2015; Cai et al., 2016; Hoitink and Jay, 2016; Hoitink et al., 2017; Du et al., 2018). In natural conditions, they usually experience a wide range of temporal variations, on timescales ranging from a fortnight to season (e.g. Zhang et al., 2018). Human intervention, such as dam construction in the upstream parts of a river and the growing number of water conservancy projects built along large rivers (such as freshwater withdrawal), has caused seasonal changes in downstream freshwater discharge delivery, leading to adjustments in the function of fluvial and estuarine hydrology (e.g. Lu et al., 2011; Mei et al., 2015a; Dai et al., 2017). Consequently, it is important to understand the impacts of large-scale human intervention, such as flood control, navigation, salt intrusion, and freshwater withdrawal, which are relevant not only to tide–river dynamics and riparian ecology but also to sustainable water resource management in general.

River discharge generally fluctuates following a wet–dry cycle due to the
seasonal variation of precipitation in the upstream river basin. For
instance, the Yangtze River, the largest river in China in terms of mean
discharge, which flows into the East China Sea, has a maximum river discharge
during summer in July and a low value during winter in January, with a
maximum discharge difference of approximately 38 000 m

The Yangtze River estuary, located near the coastal area of East China Sea,
is one of the largest estuaries in Asia. In the mouth of the Yangtze River
estuary, bifurcation occurs, and the characteristics of tides have been
broadly investigated in previous studies (e.g. Zhang et al., 2012; Lu et al.,
2015; Alebregtse and Swart, 2016). However, in these studies, river
influences are usually neglected. In recent years, the processes of nonlinear
interactions between tidal wave and river flow in the Yangtze River estuary
have received increasing attention (e.g. Guo et al., 2015; M. Zhang et al.,
2015a, b, 2018; Cai et al., 2016; Kuang et al., 2017; F. Zhang et al., 2018). However,
recent studies on tidal properties, such as asymmetry, changes near the mouth
area, and seasonal variations in tidal wave propagation and fluvial effects
over the entire 600 km of the tidal river, up to the tidal limit of the
Datong hydrological station, have been limited. In addition, the operation of
the Three Gorges Dam (TGD), the largest dam in the world, has substantially
affected the downstream river hydrology and sediment delivery. There is a
variety of debate regarding the potential impacts of the TGD on the downstream
river morphology, hydrology, and ecology, since the underlying mechanism of
the impact of the TGD is not fully understood. Specifically, the TGD
operation has altered the downstream fluvial discharge and water levels on
the seasonal scale, directly following the reservoir seasonal impounding and
release of water volume (e.g. Chen et al., 2016; Guo et al., 2018). However,
the impacts of seasonal freshwater regulation by the TGD on the
spatial–temporal tide–river dynamics in the downstream estuarine area have
not been systematically investigated. For example, during the dry season, TGD
operation increased the multi-year monthly averaged river discharge at Datong
station from 9520 to 12 896 m

Maps of the Yangtze River basin

In this study, for the first time, the spatial–temporal variations in the hydrodynamic processes due to the interactions of tidal flow and fluvial discharge in the Yangtze River estuary caused by natural forcing and human intervention were studied, with specific focus on the effect of TGD seasonal regulation. Here, we adopted a well-developed analytical model proposed by Cai et al. (2014a, 2016) to investigate the spatial–temporal patterns of tide–river dynamics in the entire Yangtze River estuary and quantify the impacts of the TGD operation. In the following sections, we introduce the study site of the Yangtze River estuary. This is followed by a description of the available data and analytical model of tide–river dynamics in Sect. 3. Subsequently, we applied the model to the Yangtze River estuary, where the TGD has been operating since 2003 (Sect. 4). In particular, we explored the alteration of the tide–river dynamics after the TGD closure and summarised the impacts of the TGD on the spatial–temporal patterns of tide–river dynamics. The impacts of channel geometry and river discharge alterations on tide–river dynamics as well as the implications for sustainable water resource management were then discussed in Sect. 5. Finally, some key findings were addressed in Sect. 6.

The Yangtze River, flowing from west to east in central China, is one of the
world's most important rivers due to its great economic and social relevance.
It has a length of about 6300 km and a basin area of about
190 000 km

Apart from river flows, tidal waves are also recognised as the major sources
of energy for hydrodynamics in the Yangtze River estuary, which is
characterised by a meso-tide with a tidal range of up to 4.6 m and a mean
tidal range of

To quantitatively investigate the relationship between freshwater discharge regulation caused by the TGD operation and the tide–river dynamics, monthly averaged hydrological data for both pre-TGD (1979–1984) and post-TGD (2003–2014) periods of tidal range and water level from the above-mentioned six tidal gauging stations along the Yangtze River estuary were collected. They were published by the Yangtze Hydrology Bureau of the People's Republic of China. The monthly averaged tidal amplitude is determined by averaging the daily difference between high and low water levels and dividing by two. To correctly quantify the residual water level along the Yangtze estuary, locally measured water level at different gauging stations are corrected to the national mean sea level of the Huanghai 1985 datum.

In tidal rivers, the tidally averaged water level (i.e. residual water
level) depicts a steady gradient, which usually increases with freshwater
discharge (e.g. Sassi and Hoitink, 2013). The key to deriving the dynamics
of the residual water level lies in the one-dimensional momentum equation,
which can be expressed as (e.g. Savenije, 2005, 2012)

It was shown by Cai et al. (2014a, b, 2016) that the tide–river dynamics is
dominantly controlled by four dimensionless parameters (see their definitions
in Table 1). They include the dimensionless tidal amplitude

Definitions of dimensionless parameters used in the analytical model.

In this study, we used the analytical solutions proposed by Cai et
al. (2014a, b, 2016), in which the solutions of the major tide–river dynamics
are derived by solving a set of four implicit equations for the tidal
damping, the velocity amplitude, the wave celerity, and the phase lag (see
details in Appendix B). The major dependent parameters can be described by
the following four variables (see also Table 1):

Changes in monthly averaged

It is worth noting that the analytically computed tide–river dynamics (

To quantify the impacts of TGD operation on the downstream tide–river
dynamics, we divided the time series into two periods, including a pre-TGD
period (1979–1984, representing the condition before the operation of the
TGD) and a post-TGD period (2003–2014, after the closure of the TGD with an
operating TGD). Figure 2 shows the changes in the observed tidal range

Changes in tidal damping rate

Changes in residual water level slope

In Fig. 2a we observe an increasing trend in tidal range for the post-TGD period at the six gauging stations (see also Table S1 in the Supplement), except for the marked decrease at the ZJ station in the first half of the year (i.e. January–June). On average, the maximum increase (0.20 m) in tidal range occurs in October, which is mainly due to the substantial reduction of river discharge caused by the TGD operation. This indicates a consistent enhancement of tidal dynamics along the Yangtze estuary, except the reach near the ZJ station. The exceptional case at the ZJ station is likely due to the fact that the ZJ station is located near the position of the tidal current limit during the dry season (Guo et al., 2015; Zhang et al., 2018). The shallow and narrow geometry around the ZJ station impedes the tidal wave propagation when river discharge increases due to the TGD operation during the dry season (Chen et al., 2012), leading to a remarkably decreasing tidal range in the first half of the year. For the residual water level, Fig. 2b clearly shows that the change in the residual water level directly follows that of the river discharge due to the stable relationship between these two parameters (see also Table S2 in the Supplement). In particular, we see that the residual water levels increased by 0.26, 0.30, and 0.16 m, respectively, in January, February, and March, while they significantly decreased by 0.72, 1.17, and 0.70 m, respectively, in September, October, and November. In addition, the decrease trend in residual water level is more significant at upstream stations when compared with those in the downstream areas.

Comparison of multi-year monthly averaged river discharge

Characteristics of geometric parameters in the Yangtze River estuary.

Since the TGD operation affects tide–river dynamics primarily through the
alteration of the freshwater discharge, it is worth exploring the patterns
of trends in the relationship between the freshwater discharge and gradients
of the main tidal parameters with respect to distance (i.e. the tidal
damping rate and the residual water level slope). Here, we estimated the
tidal damping rate

Comparison of monthly averaged values for

The analytical model presented in Sect. 3.2 was subsequently applied to the
Yangtze River estuary, with the seaward boundary using the tidal amplitude
imposed at the TSG station and the landward boundary using the river
discharge imposed at the DT station. The computation length of the estuary is
470 km, covering the entire estuary from TSG to DT. The adopted geometric
characteristics (including the tidally averaged cross-sectional area, width,
and depth) are the same for both pre- and post-TGD periods, which were
extracted from a digital elevation model (DEM) using Yangtze River estuary
navigation charts surveyed in 2007. The geometric characteristics, calibrated
by fitting the observed values using Eqs. (4) and (5), are presented in
Table 3, where a relatively large cross-sectional area convergence length (

Longitudinal variability of simulated tidal damping number

Longitudinal variability of simulated velocity number

With the significant seasonal discharge variations resulting from the TGD
regulation, an understanding of the seasonal impacts on tide–river dynamics
along the estuary has become increasingly important. In Figs. 6 and 7, we see
how the TGD operation impacts the longitudinal variation of the main tidal
dynamics in terms of the four dependent parameters (

Figure 6a, c, e, g show the comparison of the analytically computed tidal
damping number

Overall, in the seaward reach of the estuary, the effect of freshwater
discharge alteration by the TGD operation on the major tide–river dynamics
(i.e.

Longitudinal variability of simulated estuary shape number

Longitudinal variability of simulated residual water level slope

Dam operations, which dramatically modified downstream flow and sediment
regimes, are becoming an increasingly important factor controlling the
morphological evolution. Previous studies show that, as a result of the
trapping of sediments by the TGD, considerable erosion occurred in the first
several hundred kilometres downstream of the TGD, considerably coarsening the
bedload (Yang et al., 2014). In particular, the river bed immediately
downstream was eroded at a rate of 65 Mt yr

Cumulative distribution function (cdf) estimated by using the kernel
smoothing function

Temporal variation of the position of the tidal limit relative to the TSG station for both the pre-TGD and post-TGD periods. The vertical error bar at each data point indicates the standard deviation of the analytically computed time series.

The water conservancy of the TGD has multiple purposes, in which the seasonal
discharge regulation and their impact on the ecosystem are well documented
(e.g. Mei et al., 2015a, b; Chen et al., 2016; Guo et al., 2018). However,
the actual influence of discharge regulation on the tide–river dynamics in
the estuarine area is not fully understood. With the analytical reproduction
of tide–river dynamics for pre- and post-TGD periods, it is possible to
quantify the extent of the changes in the major tidal dynamics, including the
estuary shape number

Figures 8 and 9 show that during the wet season (summer–autumn), the estuary
shape number

The construction of the TGD is the largest hydro-development project ever performed in the world, having multiple influences on downstream water resource management, including navigation, flood control, tidal limit variation, and salt intrusion.

The navigation condition is mainly controlled by both high water and low water levels. Figure 10 shows the estimation of the cumulative distribution function (cdf) for both the high water level (Fig. 10a) and the low water level (Fig. 10b) at the six gauging stations along the Yangtze River estuary for both the pre- and post-TGD periods. The results indicate that navigation conditions during the non-flood season are generally improved, because percentages of both high water and low water levels are increased due to the additional freshwater discharge released from the TGD. On the other hand, during the flood season, the reduction in the freshwater discharge by TGD impounding tends to exert a negative impact on navigation. However, the reduced freshwater discharges in the late summer and autumn are not of sufficient magnitude to cause any navigation problems. This is due to the fact that the mean water levels during the flood season are relatively high; hence, the regulating flow quantity and regulating capacity are relatively small (e.g. Chen et al., 2016). In general, due to the staggered regulation in freshwater discharge, seasonally, the actual navigation condition is improved due to the significant increase in the percentage of low water levels.

Flood control is one of the most important functions of building dams and reservoirs in large rivers. Before the construction of the TGD, the Yangtze River basin suffered from frequent and disastrous flood threats. For instance, the floods of 1998 in the Yangtze River were reported to have killed 3656 people, destroyed 5.7 million homes, and damaged 7 million more. Many studies have examined the flood control capacity of the TGD over the past two decades (Zhao et al., 2013; Chen et al., 2014). In particular, the capability of the TGD flood control is influenced by multiple factors (e.g. Huang et al., 2018), particularly in the estuarine area, which is strongly influenced by tides from the ocean. During the flood season, the reduced freshwater discharge by TGD impounding benefits the flood control by reducing the peak flood discharge. However, as the tidal influence is enhanced, both the percentages of high water and low water levels for the post-TGD period are considerably increased, as shown in Fig. 10, indicating a decreased flood control capability. For instance, at the WH gauging station located in the upstream part of the Yangtze River estuary, the 8 m high water level increased by approximately 10 % after the TGD closure during the wet season. The corresponding flood prevention standard, therefore, is reduced due to the increased high water level (see also Nakayama and Shankman, 2013).

It is important to detect the position of the tidal limit (corresponding with
the position where the tidal amplitude to depth ratio is less than a certain
threshold, e.g.

The operation of the TGD changed the location of tidal limit, which, in turn, directly influences the intensity of saltwater intrusion, especially during the dry season, when the freshwater discharge is low and saltwater intrusion is important (e.g. Cai et al., 2015). The analysis of tide–river dynamics shows that the tidal dynamics are considerably enhanced during the autumn due to the substantial decline in freshwater discharged into the estuary, which may lead to enhanced saltwater intrusion. However, with supplemented discharge after the TGD during the winter, saltwater intrusion tends to be significantly suppressed, and the isohalines are pushed seaward by additional river discharges (e.g. An et al., 2009; Qiu and Zhu, 2013). In contrast, during the wet season, the TGD operation slightly extended the timing of saltwater intrusion and increased its intensity by impounding freshwater. Since the total river discharge rate during the wet season is the largest during the year, the influence of saltwater on freshwater reservoirs along the coastal area is limited. Therefore, the operation of the TGD is overall favourable for reducing the burden of freshwater supplement in the tidally influenced estuarine areas. However, to quantify the potential impacts of the TGD's operation on salt intrusion and related aquatic ecosystem health in general, it is required to couple the hydrodynamic model to the ecological or salt intrusion model (e.g. Qiu and Zhu, 2013; Cai et al., 2015).

An analytical approach was used to examine the potential impacts of TGD operation on the spatial–temporal patterns of tide–river dynamics along the Yangtze River estuary. It was shown that the freshwater regulation caused by the TGD, on a seasonal scale, exerts significant impacts on the tide–river dynamics, with the maximum influence occurring in autumn and winter. This generally corresponds to a dramatic decrease in freshwater discharge during the wet-to-dry transition period and a slight increase in discharge during the dry season. The analytical results indicate that the discharge regulation by the TGD drives the alterations in the tide–river dynamics instead of the geometric change. In particular, the change in the freshwater discharge changes the estuary shape number (representing the geometric effect), the residual water level slope (representing the effective frictional effect), and hence the tide–river dynamics. This study, using the Yangtze River estuary as an example, provides an effective yet simple method to quantify the seasonal regulation in freshwater discharge by large reservoirs or dams on hydrodynamics in estuaries. The results obtained from this study will, hopefully, shed new light on aspects of water resource management, such as navigation, flood control, and salt intrusion.

All data and results in this paper are available upon request from the corresponding author.

Assuming a periodic variation of flow velocity, the integration of Eq. (1)
over a tidal cycle leads to an expression for the residual water level slope
(e.g. Cai et al., 2014a, 2016):

The analytical solutions for the dependent parameters

the tidal damping/amplification equation, describing the tidal amplification
or damping as a result of the balance between channel convergence

the scaling equation, describing how the ratio of velocity amplitude to tidal
amplitude depends on phase lag and wave celerity:

the celerity equation, describing how the wave celerity depends on the
balance between convergence and tidal damping/amplification:

and the phase lag equation, describing how the phase lag between HW and HWS
depends on wave celerity, convergence, and damping:

The supplement related to this article is available online at:

All authors contributed to the design and development of the work. The experiments were originally carried out by HC. XZ and LG carried out the data analysis. MZ built the model and wrote the paper. FL and QY reviewed the paper.

The authors declare that they have no conflict of interest.

All the authors thank Dr. Du Jiabi and Dr. Matt Lewis for their constructive comments and suggestions, which have greatly improved the quality of this paper.

This research has been supported by the National Key R&D of China (grant no. 2016YFC0402600), the Open Research Fund of State Key Laboratory of Estuarine and Coastal Research (grant no. SKLEC-KF201809), the National Natural Science Foundation of China (grant no. 51709287), the National Natural Science Foundation of China (grant no. 41701001), the China Postdoctoral Science Foundation (grant no. 2018M630414) and the Guangdong Provincial Natural Science Foundation of China (grant no. 2017A030310321).

This paper was edited by Neil Wells and reviewed by Jiabi Du and Matt Lewis.