Long-Term Impacts of Runoff and Coastal Reclamation on Tidal Bore Variations in the Qiantang River Estuary, China

Abstract

Tidal bore dynamics in estuarine environments are influenced by both natural hydrological changes and human activities, such as coastal reclamation. This study focuses on the Qiantang River estuary, assessing the impacts of runoff variations and reclamation on tidal bores over the past five decades. By employing statistical and time-frequency methods, including the Mann–Kendall test, ordered clustering, and wavelet analysis, the relationships between tidal bore height, river discharge, and reclamation activities are examined. The results indicate that increased freshwater discharge reduces bore intensity over short timescales of 0.3 to 1.2 years, while decreased runoff amplifies it. Over longer periods of 4.1 to 8.3 years, a positive correlation emerges, with changes in runoff preceding variations in tidal bore height. Coastal reclamation, particularly the narrowing of channels, has significantly reduced the bore height at Yanguan, especially in the years following the 2000s. Additionally, the long-term interactions of other factors influencing tidal bores are explored. These findings reveal a delayed estuarine response to human modifications, highlighting the necessity of long-term monitoring and adaptive management strategies to mitigate these impacts. The study provides valuable insights into the complex interplay of natural and human factors, offering guidance for future estuarine management and conservation efforts.

1. Introduction

Tidal bores are rare and fascinating natural phenomena that occur in specific tidal estuaries or bays around the world [1]. It is estimated that over 400 estuaries globally may be influenced by tidal bores [2]. Recent field observations have identified the presence of tidal bores in at least 117 rivers distributed across 25 countries [3]. These phenomena are characterized by a sudden and powerful surge of the incoming tide, forming a wave that travels upstream against the ebb flow [4]. Depending on the hydrodynamic conditions and tidal forces, this wave can range from a smooth, undular bore to a turbulent wall of breaking water. Tidal bores not only captivate people due to their dramatic appearance but they also attract considerable scientific attention because of their complex hydrodynamic behavior [5,6]. They play an essential role in sediment transport, nutrient distribution, and ensuring ecological connectivity within estuarine systems [7,8].

One of the most renowned tidal bores in the world occurs in the Qiantang River estuary (QRE) in China [9]. Known as the “Silver Dragon”, the Qiantang River tidal bore is distinguished by its impressive size and power, making it one of the largest and most dynamic tidal bores globally [10,11]. This tidal bore can reach heights of over 3 m and travel at speeds exceeding 9 m/s [12], creating a breathtaking spectacle while also posing serious challenges for navigation and tide control. This natural phenomenon draws millions of tourists each year, contributing substantially to the local economy, and holds deep cultural importance, with festival celebrations dating back over two thousand years in historical records [13]. However, the safety concerns associated with observing the tidal bore are increasingly critical. As shown in Figure 1, since 1993, more than 700 people have been swept into the river or injured by bores surging onto the shore, with over 110 fatalities reported. One of the most tragic incidents occurred on 3 October 1993 (the 18th day of the eighth lunar month), when a tidal bore swept 86 people from the riverbanks into the water, resulting in 19 deaths, 27 injuries, and 40 missing persons [14]. As a form of marine and coastal hazard, macro-tidal bores pose significant risks to coastal communities, ecosystems, and infrastructure in the QRE.

The formation and amplification of the tidal bore in the Qiantang River are primarily influenced by the unique geomorphology of the estuary [15]. The funnel-shaped QRE, combined with the sandbar at its mouth, enhances the incoming tidal wave as it moves upstream [16,17]. The interaction between tidal forces and river hydrodynamics produces this dramatic tidal bore. Understanding the factors that influence this phenomenon is crucial for the effective management of the estuarine environment and for mitigating the risks associated with its powerful energy [18]. The existence of the tidal bore is the result of a delicate balance between tidal wave propagation, river flow discharge, and riverbed evolution within the estuary. This balance is highly sensitive to boundary conditions, particularly those influenced by human activity [19]. For example [14], the regulation of the Seine River in France nearly led to the disappearance of its tidal bore, while dam construction on the Colorado River in Mexico significantly reduced its tidal bore. Similarly, upstream damming in the Petitcodiac River in Canada almost eliminated its tidal bore from 1968 to 2010. Although these interventions improved navigation and flood control, the weakening of tidal bores has had a significant impact on the ecological environments of these estuaries [2].

Since the 1960s, the QRE has experienced significant human-induced changes that may affect the behavior of its tidal bore [20]. Dam construction, water withdrawals for agricultural and industrial purposes, and climate variability have altered the freshwater discharge into the estuary. These modifications to river flow have the potential to influence estuarine hydrodynamics by changing the hydraulic conditions under which the tidal bore forms [21]. At the same time, extensive coastal reclamation projects have been carried out to support urban expansion, industrial development, and port construction [22]. These activities have reshaped the estuarine morphology [23], altering channel geometry, reducing cross-sectional areas, and potentially impacting both the propagation and characteristics of the tidal bore.

These anthropogenic modifications raise critical concerns about their potential impact on the tidal bore phenomenon in the Qiantang River [24]. Changes in the behavior of the tidal bore could have significant implications for water safety, as the force of the bore poses risks to vessels and infrastructure along the river [25]. Altered sediment transport patterns may affect ecological habitats, influencing the distribution of estuarine species and the overall health of the ecosystem [26]. Furthermore, any reduction to the spectacle of the tidal bore could attract public attention, given its strong connection to government management and the tourism industry centered around the tidal bore.

Despite the recognized significance of these impacts, comprehensive statistical analyses quantifying the combined influence of runoff variations and coastal reclamation on tidal bore variation in the QRE remain limited. Bonneton et al. [27] demonstrated that freshwater discharge affects tidal wave transformation and the occurrence of tidal bores, while field data indicate that bore intensity is mainly controlled by the dimensionless tidal range, which reflects local wave nonlinearity. As runoff discharge weakens the tidal wave during propagation, it reduces the dimensionless tidal range, thereby limiting tidal bore development in the estuary [28]. Pan et al. [29] examined the relationships between tidal range, river discharge, and riverbed volume, establishing a link between tidal bore height and tidal range based on field data and analyzing post-flood-season variations during wet and dry years. Lu et al. [30], using measured data and a mathematical model, analyzed the impact of large-scale river narrowing on tidal patterns, riverbed changes, and tidal bore behavior in the QRE. Li et al. [31] studied the hydrodynamic changes in the coastal section outside the QRE (i.e., Hangzhou Bay) from 1962 to 2015, showing that the maximum tidal range increased by more than 2 m due to the combined effects of increased shoaling and tidal choking, based on satellite data and numerical model simulations.

Coastal and estuarine systems worldwide are profoundly influenced by both natural forces and human activities, such as land reclamation, which alter hydrodynamics, sediment transport, and ecological balance [32,33]. In Liaodong Bay, China [34], reclamation has transformed 85% of natural coastal wetlands, causing fragmentation and accelerating landscape succession due to rapid urbanization and industrialization. Similarly, the Bohai Sea has experienced significant hydrodynamic shifts over the past 30 years, driven by large-scale reclamation projects [35]. The Red River Delta is also undergoing noticeable coastal morphological changes, shaped by increasing natural pressures and human interventions [36]. Human activities have been instrumental to the Thames Estuary’s development [37], and rising sea levels are predicted to lower intertidal zones by up to 0.5 m relative to high water, indicating potential long-term ecological impacts. These cases highlight the widespread and profound effects of reclamation and other human activities on coastal environments, underscoring the need for their sustainable management. While previous studies have largely focused on theoretical models or localized observations, they often lack the integration of long-term data on human activities and hydrological changes in the QRE. A systematic investigation combining hydrological data, records of reclamation activities, and tidal bore measurements is needed to fully understand the interplay of these factors.

This study seeks to statistically analyze the impact of runoff variations and coastal reclamation on tidal bore dynamics in the QRE over the past five decades, from 1974 to 2023. By investigating temporal patterns and periodicities in tidal bore height, river discharge, and other key hydrodynamic variables, this research explores how changes in river discharge and estuarine morphology influence tidal bore behavior at different timescales. It also assesses the effects of coastal reclamation on tidal bore height and broader estuarine dynamics. These findings contribute to advancing our knowledge in coastal hydrology and estuarine engineering, particularly regarding how human activities shape tidal processes. Moreover, this study offers practical insights for policy decisions, sustainable development, flood risk management, and the preservation of natural phenomena and coastal hazards. This research not only deepens our understanding of the long-term interactions between natural and human-induced changes but also provides guidance for integrated estuarine management and directions for future research.

This paper is organized as follows: The next section describes the materials and methods used in this study, including details of the study area, data collection procedures, and statistical analysis techniques used. The Results section presents the findings of these statistical analyses, highlighting the impacts of runoff variations and coastal reclamation on tidal bore characteristics; the long-term variations and interactions of other factors associated with tidal bore have also been thoroughly investigated. The Discussion interprets the results in the context of the existing literature, explores the implications for coastal management, and addresses the limitations of this study. Finally, the Conclusions summarize the key findings, discuss their significance for the field, and offer recommendations for future research and policy considerations.

2. Materials and Methods

2.1. Study Area

As shown in Figure 2, the QRE is located in the southeastern part of China, ultimately draining into the East China Sea. The Qiantang River is the longest river in Zhejiang Province, with a total length of approximately 688 km and a drainage area of 55,558 km² [38]. The estuary is geographically positioned between 29°43′ N and 30°53′ N latitude and 119°39′ E and 122°00′ E longitude, encompassing the lower reaches of the river as it flows through the coastal city of Hangzhou and empties into Hangzhou Bay. This estuarine system is characterized by its funnel-shaped mouth, which significantly amplifies tidal waves, resulting in a dramatic tidal bore phenomenon.

The QRE plays a crucial role in regional flood control and water resource management. Based on variations in hydrodynamic conditions and riverbed evolution, the estuary is divided into three distinct sections. The upstream river section, extending 77 km from the Fuchun River Hydropower Station (FRHS) dam to Wenjiayan, is primarily governed by runoff, resulting in a relatively stable riverbed; the transitional section, spanning 116 km from Wenjiayan to Ganpu, is influenced by both runoff and tidal forces, leading to significant erosion and deposition along the riverbed; the downstream coastal section, stretching 98 km from Ganpu to Nanhui, is part of Hangzhou Bay and is primarily shaped by tidal forces [38].

The riverbed of the QRE is composed of a mixture of sandy and muddy sediments, with sediment transport largely driven by the interplay between tidal forces and freshwater runoff. The interaction between these forces results in intense erosion and deposition processes, particularly in the transitional section of the estuary. The region experiences a semi-diurnal tide, with two high and two low tides each day. The tidal range varies considerably, with the maximum range exceeding 9 m during spring tides, further intensifying the tidal bore phenomenon. Seasonal freshwater inflows from the upstream catchment also play a crucial role in shaping the hydrodynamics of the estuary. Peak discharges occur during the wet season, typically between April and October, driven by monsoon rainfall and occasional typhoon events [39]. During this period, the monthly average runoff discharge can exceed 5000 m³/s, influencing both sediment transport and tidal bore characteristics.

Figure 2. Study area: (a) Zhejiang Province in China; the base map is from the standard map service system of the Ministry of Natural Resources of China and has the approval number GS(2016)1554 [40]. (b) The Qiantang River estuary (QRE) in Zhejiang Province. (c) Satellite image of the QRE, created from Landsat imagery. Landsat imagery courtesy of the U.S. Geological Survey. SH: Shanghai; HZ: Hangzhou; WY: Weijiayan; YG: Yanguan; GP: Ganpu; NH: Nanhui; FRHS: Fuchun River Hydropower Station. The red dots in (c) represent the locations of tide gauges T01 to T09.

Figure 2. Study area: (a) Zhejiang Province in China; the base map is from the standard map service system of the Ministry of Natural Resources of China and has the approval number GS(2016)1554 [40]. (b) The Qiantang River estuary (QRE) in Zhejiang Province. (c) Satellite image of the QRE, created from Landsat imagery. Landsat imagery courtesy of the U.S. Geological Survey. SH: Shanghai; HZ: Hangzhou; WY: Weijiayan; YG: Yanguan; GP: Ganpu; NH: Nanhui; FRHS: Fuchun River Hydropower Station. The red dots in (c) represent the locations of tide gauges T01 to T09.

In addition to natural hydrological factors, human activities have increasingly altered the hydrodynamic characteristics of the QRE. The construction of upstream dams and reservoirs, such as those on the Xinan River and the Fuchun River reservoirs, has played a key role in regulating river flow. While these structures effectively reduce flood peaks, they also disrupt the natural flow regime, impacting the seasonal variability of the freshwater discharge into the estuary. As illustrated in Figure 3, coastal reclamation projects have further transformed the estuarine morphology by narrowing the estuary and modifying the patterns of tidal wave propagation. These anthropogenic changes have had a significant impact on both the physical structure of the estuary and the dynamics of the tidal bore, altering its intensity and behavior.

2.2. Data Collection

The runoff data used in this study were primarily sourced from a hydrological station downstream of the FRHS (Figure 2). This station has provided continuous and reliable discharge measurements since 1930, facilitating the assessment of long-term changes in estuarine runoff. As a key monitoring point for upstream inflows, the FRHS data are crucial for understanding the hydrological processes that affect the estuary. In this study, the runoff data used span 50 years, from 1974 to 2023, providing a comprehensive perspective on river discharge variations. Collected monthly from Zhejiang Provincial Hydrology Bureau, the data have a high temporal resolution, capturing both seasonal and annual trends in river flow. Runoff is primarily determined by the tailwater discharge from power generation at the FRHS, with its uncertainty estimated to be within ±10%, which is mainly influenced by environmental factors such as extreme weather conditions during the measurement period.

The coastal reclamation data used in this study were primarily sourced from official records, including documents published by the Zhejiang Provincial Department of Water Resources and the QRE bathymetric survey reports from the Zhejiang Institute of Hydraulics and Estuary. These records provide detailed descriptions of the scope, completion dates, and specific locations of reclamation projects in the QRE. Since the 1950s, the Zhejiang Institute of Hydraulics and Estuary has conducted bathymetric surveys of the QRE every April, July, and November, using a sectional scale of 1:2000. This study has compiled annual reclamation areas, with a focus on the relationship between major reclamation activities and changes in the estuarine system. The bathymetric surveys were carried out using echo sounders, which provide highly precise water depth measurements. To account for tidal influences during data collection, post-processing corrections were applied. The uncertainty in the bathymetric data is estimated to be ±0.10 m, primarily due to water-level fluctuations during the survey period.

This study focuses on the variation characteristics of the tidal bore in the QRE, primarily utilizing height data from Yanguan (Figure 2), where the tradition of observing the tidal bore has continued for over 400 years, since the Ming Dynasty (1368–1644 AD). Measurements were conducted through field observations and instrumental monitoring to capture water level changes as the bore reached Yanguan, with tidal bore height defined as the difference between water levels before and after the bore front. Historical data were sourced from archival records maintained by the Qiantang River Administration, which provide detailed daytime tidal bore height observations dating back to 1974. To reflect river conditions at the time of the bore’s arrival, low water levels recorded at Yanguan prior to the bore, as documented by the Zhejiang Hydrology Bureau, were collected for the corresponding periods. All measurement data conformed to the China National Height Datum 1985, maintaining a vertical measurement accuracy of 0.1 m. Tidal bore height was measured using a combination of staff gauges, tide gauges, and real-time water level sensors, all of which were regularly calibrated to ensure accurate measurements. The uncertainty in these measurements is estimated to be ±0.05 m, mainly due to the rapid fluctuations in water levels during the passage of the bore.

The tidal range data at Ganpu (Figure 2), near the point where the tidal bore first forms, are essential for understanding the hydrodynamic conditions that contribute to the development of the tidal bore in the QRE. Historical data were sourced from archival records maintained by the Zhejiang Provincial Hydrology Bureau, documenting several decades of tidal range observations. Recent data were collected through real-time monitoring systems established by the hydrological authority, providing more accurate and frequent measurements. The measurement uncertainty of the tidal level data is estimated to be ±0.05 m.

2.3. Data Analysis

Before conducting the analysis, a thorough data processing procedure was followed to ensure the accuracy and integrity of the hydrodynamic and tidal-bore-height data. This process involved cleaning and preprocessing the raw data, as well as aligning the time series for consistency. Any extreme values caused by recording errors were flagged for further analysis. Missing data were handled using linear interpolation for small gaps, while larger gaps were imputed using model-based approaches or excluded if imputation was not feasible.

2.3.1. Wavelet Analysis

In this study, wavelet analysis was used to examine the periodic variations in the tidal bore height and associated hydrodynamic data. The Morlet wavelet was selected as the basis for the analysis due to its capability to detect both time and frequency characteristics in non-stationary signals. The Morlet wavelet function, expressed as [41]

$$\psi (t)={\pi }^{-1/4}{e}^{i{\omega }_{0}t}{e}^{-t^2/2},$$

(1)

where ω₀ = 6, was employed for its ability to balance time and frequency localization effectively. The continuous wavelet transform was applied to the time series data x(t), producing wavelet coefficients W(a,b) using the formula [42]

$$W(a,b)={\displaystyle {\int }_{-\infty }^{\infty }x(t)\frac{1}{\sqrt{a}}}\psi \left({\displaystyle \frac{t-b}{a}}\right)dt,$$

(2)

where a represents the scale parameter (related to frequency) and b is the translation parameter (related to time). These wavelet coefficients provide insights into how periodic components of the time series evolve over different timescales. In this study, the modulus of the Morlet wavelet coefficients |W(a,b)| was used to show the distribution of the coefficients across different timescales.

Additionally, wavelet variance provides insight into the energy distribution of the time series across different timescales [43]. The total variance V(a) for each scale a was calculated as

$$V(a)={\displaystyle \frac{1}{T}}{\displaystyle {\int }_0^T{\left|W(a,b)\right|}^{2}db},$$

(3)

where T represents the total length or duration of the time series being analyzed and the square of the continuous wavelet transform (|W(a,b)|²) corresponds to the power of the signal [44].

2.3.2. The Cross Wavelet Transform

The cross wavelet transform is a signal analysis technique that integrates wavelet analysis with cross-spectral analysis, enabling the study of the relationship between two wavelet-variance time series across multiple timescales in both the time and frequency domains [41]. This method helps identify regions in time and scale where two time series exhibit common power and phase relationships, which is crucial for understanding their interactions and dependencies over time.

The mathematical expression of the cross wavelet transform of two time series x(t) and y(t), each with their respective wavelet transforms W_x(a,b) and W_y(a,b), is defined as [42]

$$W_{xy}(a,b)=W_x(a,b)W_y^{\ast }(a,b),$$

(4)

where W_x(a,b) and W_y(a,b) are the continuous wavelet transforms of x(t) and y(t), respectively, and W_y*(a,b) is the complex conjugate of W_y(a,b).

The modulus of the cross wavelet transform, |W_xy(a,b)|, indicates areas where both time series share significant common power, revealing the strength of their correlation across different timescales. The phase difference between the two time series can be calculated from the argument of W_xy(a,b), offering insights into their lead–lag relationship at specific timescales.

2.3.3. Wavelet Coherence

Wavelet coherence is a technique used to measure the local correlation between two time series in the time–frequency domain. Unlike the cross wavelet transform, which reveals regions of shared power, wavelet coherence normalizes this relationship, offering a more precise measure of how closely the two signals are correlated across various timescales. This is particularly useful for identifying significant relationships, even in regions where the cross wavelet power is low [45].

The mathematical expression for wavelet coherence between two time series, x(t) and y(t), is [42]

$$R^2(a,b)={\displaystyle \frac{{\left|S(a^{-1}{W}_{xy}(a,b))\right|}^2}{S(a^{-1}{\left|W_x(a,b)\right|}^2)S(a^{-1}{\left|W_y(a,b)\right|}^2)}},$$

(5)

where S is a smoothing operator applied to both time and scale.

The resulting wavelet coherence values, R²(a,b), range from 0 to 1, where values close to 1 indicate a strong local correlation between the two time series at a given time and scale, while values near 0 indicate weak or no correlation.

2.3.4. The Mann–Kendall Test

The Mann–Kendall test is a non-parametric statistical test used to identify trends in time series data, which is particularly useful for detecting monotonic (increasing or decreasing) trends over time. Unlike parametric tests, it does not require the data to follow a normal distribution, making it well suited for analyzing environmental and hydrological datasets, where non-normality and nonlinearity are common. For n variables, x₁, x₂,…, x_n, the test statistic S_k is defined as [46]

$$S_k={\displaystyle \sum _{i=1}^k{\displaystyle \sum _{j=1}^{i-1}{\alpha }_{ij}}}\begin{array}{cc}& (k=2,3,\cdots n)\end{array},$$

(6)

$${\alpha }_{ij}=\left\{\begin{array}{c}1\begin{array}{cc}& x_i>x_j\end{array}\\ 0\begin{array}{cc}& x_i\le x_j\end{array}\end{array}\right.\begin{array}{cc}& 1\le j\le i\end{array},$$

(7)

and the statistic index UF_k is defined as follows:

$$U{F}_k={\displaystyle \frac{S_k-E(S_k)}{\sqrt{Var(S_k)}}}\begin{array}{cc}& k=1,2,\cdots ,n\end{array},$$

(8)

where

$$E(S_k)={\displaystyle \frac{k(k-1)}{4}},$$

(9)

$$Var(S_k)={\displaystyle \frac{k(k-1)(2k+5)}{72}}.$$

(10)

A backward sequence, UB, is calculated using the same formula but applied to the reversed data series. The null hypothesis, which posits no significant change point, is rejected if the UF values exceed the bounds of the confidence interval. The approximate timing of the mutation is determined by the intersection of the UF and UB lines within the confidence interval. If this intersection lies outside the interval, another method must be used. In this study, an ordered clustering analysis was used to re-evaluate the variability of the time series. By combining both methods, a more precise identification of mutation points is achieved.

2.3.5. Ordered Clustering Analysis

The ordered clustering analysis method is a statistical estimation technique commonly used in hydrological time-series analyses to detect potential mutation points, or significant changes, within the data. This method identifies whether there are abrupt changes in the sequence by dividing the sample. The original sequence {x_i} (i = 1, 2, …, n) is split into two samples: {x_i} (i = 1, 2, …, τ) and {x_i} (i = τ + 1, τ + 2, …, n). The change points in the sequence are then identified based on the statistical characteristics of these samples.

Assuming that the sum of the squared deviations of the two samples after splitting is [47]

$$V_{\tau }={\displaystyle \sum _{i=1}^{\tau }(x_i-}{\overline{x}}_{\tau }{)}^2,$$

(11)

$$V_{n-\tau }={\displaystyle \sum _{i=\tau +1}^n(x_i-}{\overline{x}}_{n-\tau }{)}^2,$$

(12)

where ${\overline{x}}_{\tau }={\displaystyle \frac{1}{\tau }}{\displaystyle \sum _{i=1}^{\tau }{x}_i}$ and ${\overline{x}}_{n-\tau }={\displaystyle \frac{1}{n-\tau }}{\displaystyle \sum _{i=\tau +1}^n{x}_i}$, the total sum of the squared deviations is then expressed as

$$S_n(\tau )=V_{\tau }+V_{n-\tau }.$$

(13)

The value of τ that satisfies the condition $S_n=\min \{S_n(\tau )\}$ is considered the most probable split point, or the change point.

3. Results

3.1. Time–Frequency Analysis

A tidal bore is a natural phenomenon that occurs during the propagation of tidal waves, resulting from the nonlinear distortion of a tidal wave. A typical tidal-level process in the tidal bore section of the QRE is shown in Figure 4. The tidal bore reaches its maximum strength in the river section between the Caoejiang Sluice (T01) and Linjiang Wharf (T02), where the water level rapidly rises upon the arrival of the bore, with a height exceeding 3 m. Yanguan (T04) is a traditional observation point for the tidal bore, where the tidal front forms an S-shape. Due to the curvature of the shoreline at Laoyancang (T05), a secondary reflected bore, caused by the bore impacting the seawall, can be observed. Upstream of Cangqian (T06), the bore height significantly decreases, and by Wenjiayan (T09), the breaking front becomes less pronounced, transitioning into a undular bore. Historically, Yanguan has the most detailed records of tidal bore height. This study, combined with hydrological data, analyzes changes in the tidal bore height at Yanguan from 1974 to 2023.

Figure 5 presents the long-term observed data of the monthly maximum tidal bore height (MTBH) and corresponding low-tide level (LTL) at Yanguan, along with the upstream average runoff discharge (ARD) and downstream average tidal range (ATR) at Ganpu in the QRE, based on monthly data from 1974 to 2023. The figure illustrates the temporal evolution and seasonal variations in these four key parameters, emphasizing both intra-annual (monthly) patterns and long-term changes over the 50-year observation period.

The MTBH (Figure 5a) displays a distinct seasonal pattern, with higher values predominantly observed from August to October. These autumn peaks follow the flood season, when favorable river and tidal conditions allow the tidal bore to propagate further upstream and reach greater heights. While no consistent long-term trend is evident over the observation period, several notable peaks are seen in the mid-1990s. From 2010 onwards, the tidal bore height appears to have stabilized. The corresponding LTL (Figure 5b) shows a seasonal pattern opposite to that of the MTBH, with higher levels in the spring and lower levels in the autumn. The long-term trend reveals notable variation, particularly with an increase in the LTL around 2008. The upstream ARD (Figure 5c) follows a clear seasonal cycle, peaking in the summer and declining through the autumn and winter. Significant inter-annual variability is observed, especially in the mid-1990s and late 2010s, when values increased. However, in the past three years, there has been a decrease in the ARD. The downstream ATR (Figure 5d), recorded at Ganpu, also shows a seasonal pattern, with larger tidal ranges occurring from August to October, corresponding to periods of larger tidal bores. The long-term trend indicates an increase in the tidal range, particularly after 2010. In the past five years, the ATR has shown signs of stabilization.

The time–frequency characteristics of the monthly MTBH and corresponding LTL at Yanguan, the upstream ARD, and the downstream ATR at Ganpu, covering the period from 1974 to 2023, were analyzed using a wavelet transform. The modulus of the wavelet coefficients for these variables is shown in Figure 6, which provides insights into the temporal evolution of significant frequencies, highlighting both short-term variations and long-term trends across the studied parameters.

The MTBH (Figure 6a) exhibits significant variations in intensity across different frequency scales, with central oscillation frequencies occurring around 1977, 1987, 1997, 2007, and 2018, corresponding to dominant frequencies of 0.093, 0.053, 0.049, 0.045, and 0.044 year⁻¹, respectively. The frequency scale of 0.093 to 0.109 year⁻¹ persists throughout the entire study period, showing a strengthening trend over the past 20 years. The LTL (Figure 6b) exhibits a periodicity similar to the MTBH, with central oscillation frequencies occurring around 1982, 1999, and 2015, corresponding to dominant frequencies of 0.032, 0.031, and 0.030 year⁻¹, respectively. The frequency scale of 0.159 to 0.199 year⁻¹ persisted throughout the first 20 years, while the frequency scale of 0.074 to 0.089 year⁻¹ has dominated the 20 most recent years. The ARD (Figure 6c) exhibits a strong seasonal cycle with a frequency of approximately 1 year⁻¹, driven by monsoon-induced runoff. The most intense periods of high ARD occurred from the mid-1990s to the early 2020s, corresponding to regional hydrological changes. Its central oscillation frequencies occurred around 1977, 1987, 1997, 2007, and 2018, with subdominant frequencies of 0.053, 0.052, 0.051, 0.048, and 0.041 year⁻¹, respectively. The ATR (Figure 6d) shows dominant oscillations centered around 1982, 1999, and 2016, with a frequency of approximately 0.03 year⁻¹, and subdominant oscillations centered around 1979, 1989, 1999, 2009, and 2019, with a frequency of approximately 0.05 year⁻¹.

Figure 7 presents the wavelet variance analysis of the monthly MTBH and corresponding LTL at Yanguan, as well as the upstream ARD and downstream ATR at Ganpu. This analysis highlights the dominant frequencies and long-term trends of each hydrological variable. The wavelet variance of the MTBH reveals three distinct peaks at 32.9 years, 21.0 years, and 10.6 years, with the 21.0-year period corresponding to the largest peak, indicating the strongest periodicity and serving as the primary cycle. Similarly, the wavelet variance of the LTL exhibits three peaks at 33 years, 12.7 years, and 6.3 years. In contrast, the wavelet variance curve for the upstream ARD is less smooth, displaying higher local variability, which reflects greater short-term fluctuations in runoff. The wavelet variance for the upstream ARD shows three notable peaks, at approximately 33.8 years, 20.4 years, and 1 year. On the other hand, the wavelet variance of the downstream ATR presents only two prominent peaks at 34.1 years and 20.2 years, indicating that tidal range fluctuations are primarily driven by long-term tidal dynamics and estuarine morphological changes. This latter curve is the smoothest of all, suggesting a stable and continuous influence from long-term processes, with minimal short-term variability.

3.2. Mutation Analysis

Figure 8 presents the results of a mutation analysis using the Mann–Kendall test for the monthly MTBH and corresponding LTL at Yanguan, as well as the upstream ARD and downstream ATR at Ganpu. The MTBH (Figure 8a) began to exhibit an upward trend around 1988, which became significantly more pronounced after 1992, surpassing the critical significance level of 0.05. The UF and UB curves intersect at 1988, which is almost outside the critical interval, indicating that a potential mutation in the MTBH may have occurred around that time. Similarly, Figure 8b shows that the UF and UB statistical curves for the LTL intersect in 1978, 1983, and 2005, but only the 2005 intersection falls within the confidence interval. This suggests that a mutation in the LTL occurred in 2005. In Figure 8c, the UF statistical curve for the upstream ARD fluctuates throughout the observed period. The UF and UB curves intersect within the confidence interval around 1978, 1998, and 2009, indicating potential mutation points during these years. The downstream ATR at Ganpu (Figure 8d) began to show an upward trend around 1992, which became significantly more pronounced after 1995, surpassing the critical significance level of 0.05. The UF and UB curves intersect in 2011, though outside the critical interval, suggesting that a potential mutation in the ATR may have occurred around that time.

The Mann–Kendall test did not fully identify a clear mutation point, so an ordered clustering analysis was employed for further detection. In this method, the year with the smallest sum of squared deviations is considered the mutation year. Figure 9 presents the results of the mutation analysis using ordered clustering analysis for the monthly MTBH and corresponding LTL at Yanguan, as well as the upstream ARD and downstream ATR at Ganpu. For the MTBH (Figure 9a), the analysis identifies 1988 as the mutation year, where the sum of squared deviations reaches its minimum. Similarly, the corresponding LTL (Figure 9b) shows a mutation in 2005, marking a notable change in the LTL at Yanguan. The mutation year for the upstream ARD (Figure 9c) is identified as 2010. Finally, the downstream ATR (Figure 9d) reveals a mutation in 2012, indicating a shift in tidal range dynamics. Combined with the results of the Mann–Kendall test, the mutation years for the monthly MTBH, corresponding LTL at Yanguan, upstream ARD, and downstream ATR at Ganpu can be identified as approximately 1988, 2005, 2010, and 2012, respectively.

3.3. Wavelet-Based Correlation Analysis

Figure 10 presents the cross wavelet transform results for various hydrological variable pairs in the QRE, illustrating the time–frequency relationships between the time series of the runoff, tidal conditions, and tidal bore height. To minimize boundary effects and avoid high-frequency noise, the thin solid line marks the boundary of the cone of influence (COI), with the valid spectral region confined within this boundary. The thick contour represents the 95% confidence level against red noise, while the color bar at the bottom indicates cross wavelet power levels. The phase angle between two time series, represented by the arrows in the plots, assesses their interaction. When the two series are in phase, the arrow points to the right, indicating synchronous increases or decreases. When out of phase, the arrow points left, indicating opposite movements. If the arrow points downward, the first series is ahead of the second by a quarter of a cycle. Conversely, if the arrow points upward, the first series lags behind the second by a quarter of a cycle.

The correlation between the monthly MTBH and the corresponding LTL at Yanguan (Figure 10a) shows significant wavelet power within the 0.8–1.2 year period, particularly during the late 1990s and early 2000s. Before 1988, leftward-pointing arrows indicate a negative correlation. To further explore this relationship, we compared the Pearson correlation coefficients (r) for the entire study period (1974–2023) and the resonance period (1974–1988). Over the full period, the r value was −0.399, whereas during 1974–1988, it was −0.598. This transition from a stronger to a weaker negative correlation suggests that external factors or long-term changes in estuarine hydrodynamics may have influenced this relationship. The correlation between the monthly MTBH and the upstream ARD (Figure 10b) exhibits significant wavelet power within shorter periods (0.6 to 1.2 years), particularly from the 1990s to the 2010s. The phase arrows generally point to the left, indicating that the ARD is negatively correlated with the MTBH at these shorter timescales.

The cross wavelet analysis between the monthly MTBH and the downstream ATR at Ganpu (Figure 10c) shows strong correlations within the 0.6 to 1.2 year period. After 2005, a predominantly positive phase relationship emerged, suggesting that increases in the ATR are closely associated with increases in the MTBH. However, between 1974 and 1979, the two variables showed a primarily negative correlation. To gain further insights, we compared the Pearson correlation coefficients r of the entire study period and specific resonance periods (1974–1979 and 2005–2023). During the entire study period, the r value was 0.176 (p < 0.05), whereas during 1974–1979 and 2005–2023, the r values were −0.355 (p < 0.05) and 0.342 (p < 0.05), respectively. This comparison reveals a notable shift from a negative correlation in the earlier period to a positive correlation after 2005.

The correlation between the LTL and the ARD (Figure 10d) primarily occurs within shorter time periods (0.6 to 1.2 years). From 1983 to 1994 and 2012 to 2023, the phase arrows point to the right, indicating a positive correlation, suggesting synchronous changes between the LTL and the ARD. For the pair of the ATR and the ARD (Figure 10e), the most significant correlations occur within the 0.7 to 1.3 year period, particularly after 2010. The upward and right-pointing phase arrows indicate a clear positive correlation, with the ATR lagging behind the ARD by 1–3 months. The correlation between the LTL and the ATR (Figure 10f) shows significant power in the short term (0.7 to 1.2 years). The downward and right-pointing phase arrows also indicate a clear positive correlation, with the LTL ahead of the ATR by 0–3 months.

Figure 11 presents the wavelet coherence analysis of various hydrodynamic variables in the QRE, illustrating their interrelationships over time and across different temporal scales. Each subplot presents a different pair of variables, displaying magnitude-squared coherence through color intensity, while arrows illustrate the phase relationships. The thick contours represent the 95% confidence level against red noise. Arrows pointing right indicate in-phase relationships, while arrows pointing left indicate out-of-phase relationships, providing insights into the direction of the influence between the variables.

Figure 11a illustrates the wavelet coherence between the monthly MTBH and the corresponding LTL at Yanguan. Significant coherence is observed over short periods of 0.6 to 4 years. The predominance of upward–left-pointing arrows suggests a general out-of-phase relationship, indicating that increases in the MTBH are associated with a lower LTL during these time periods. Additionally, significant negative correlations are observed in the longer period ranges of 4.6–7.4 and 9.3–16.5 years. Figure 11b illustrates the wavelet coherence between the monthly MTBH and the upstream ARD. Significant coherence is observed over periods of 0.3 to 1.2 years, with phase arrows showing fluctuations indicating negative correlations. In the longer period range of 4.1 to 8.3 years, there is a prominent upward–right-pointing correlation, indicating a positive relationship between the ARD and the MTBH over longer timescales, with the ARD leading the MTBH by 0.3 to 1.5 years.

In Figure 11c, the coherence between the monthly MTBH and the downstream ATR at Ganpu is analyzed. Over the short period of 0.6 to 1.3 years, there are fluctuations between positive and negative phase relationships. However, a significant negative correlation is observed over the longer period of 2.6 to 4.4 years between 1989 and 2002. Figure 11d analyzes the relationship between the LTL and the upstream ARD, with coherence concentrated in periods of 0.5 to 2.3 years. The rightward-pointing arrows indicate an in-phase relationship, suggesting that a higher upstream discharge raises the LTL. However, over the longer period of 3.3 to 8.7 years, a negative correlation is observed, with arrows pointing to the lower left, indicating that large discharges during high-flow periods may lower the LTL due to increased scouring. Conversely, during low-flow periods, the LTL rises as a result of sediment deposition.

Figure 11e focuses on analyzing the coherence between the downstream ATR and the upstream ARD. The most significant coherence occurs over periods of 0.6 to 1.5 years. The upward–right-pointing phase arrows indicate a clear positive correlation between the ATR and the ARD, with the ATR lagging behind the ARD by 1 to 4 months. Figure 11f analyzes the coherence between the LTL and the downstream ATR. Significant coherence is observed over periods of 0.6 to 1.5 years, with downward–right-pointing phase arrows indicating a clear positive correlation between the LTL and the ATR, with the LTL ahead of the ATR by 0 to 4 months. Over the longer period of 2.2 to 5.2 years, a significant positive correlation is observed, as seen between 1991 and 2010.

3.4. Coastal Reclamation Effects Analysis

Coastal reclamation in the QRE has a long history. Since the Song Dynasty (960–1279 AD), the northern and southern banks of the estuary have been fertile regions. As tidal flats grew and the coastal economy developed, large-scale seawall construction was undertaken to enclose tidal lands for cultivation. Starting in the 1950s, reclamation projects expanded to include river channel management and the utilization of tidal flat resources on a larger scale. Figure 12 shows the temporal evolution of the annual and cumulative coastal reclamation area (CRA) in the QRE from 1958 onwards, highlighting distinct phases of reclamation activity.

As seen in Figure 12, significant increases in both the annual and cumulative CRA occurred during the 1970s and 2000s, while the period from 1980 to 2000 saw a decline in activity, indicating a slowdown in the pace of reclamation. Since 2018, no new reclamation projects have been initiated, as the shoreline of the QRE has reached its planned boundary. Coastal reclamation in the QRE has had a significant impact on estuarine dynamics, potentially altering natural hydrodynamic processes. The interaction between reclamation activities and key hydrological factors, including the monthly MTBH, LTL, ARD, and ATR, is critical for effective estuarine management and coastal engineering. Figure 13 presents a cross wavelet transform analysis of various hydrodynamic variables and the CRA, providing insights into their temporal relationships and phase dynamics.

Figure 13a depicts the cross wavelet transform between the monthly MTBH and CRA, revealing significant correlations over periods ranging from 1 to 13 years, particularly after 2005. The phase arrows predominantly point to the left, indicating an inverse relationship in which increases in reclamation area correspond to decreases in tidal bore height. The expansion of reclamation activities may have altered the estuarine morphology, thereby reducing tidal bore height. To further explore this relationship, Pearson correlation coefficients (r) were compared for two distinct periods: 1974 to 2005 and 2005 to 2023. During the 1974 to 2005 period, the r value was 0.109 (p < 0.05), while during the 2005 to 2023 period, the r value was −0.305 (p < 0.05), indicating a shift to a stronger negative relationship over the past two decades.

Figure 13b shows the cross wavelet transform between the LTL and the CRA, which demonstrate significant coherence over periods of 3 to 13 years. The phase arrows pointing diagonally upward and to the right suggest a clear positive correlation between the two, with changes in the CRA occurring ahead of changes in the LTL by 0.4 to 0.8 years. Figure 13c analyzes the relationship between the upstream ARD and the CRA, indicating significant coherence over periods of 3 to 8 years. The arrows point downward and to the left, suggesting a negative correlation between the ARD and the CRA. Large-scale reclamation generally occurs during low-flow, dry periods when runoff is reduced.

Figure 13d illustrates the cross wavelet transform between the downstream ATR and the CRA, showing significant coherence over shorter periods of 0.5 to 1.5 years, with both in-phase and out-of-phase cycles. This variability indicates a complex interaction between reclamation activities and tidal range dynamics. However, the overall phase relationship is negative, suggesting a lagged impact of reclamation on tidal range. To gain further insights, we compared the Pearson correlation coefficient (r) over the entire study period (1974–2023) and during a specific resonance period (2005–2023). During the 1974–2023 period, the r value was −0.219 (p < 0.05), whereas during 2005–2023, the r value was −0.473 (p < 0.05), indicating that this lagged effect has become more pronounced in the later period.

Figure 14 presents the wavelet coherence analysis of various hydrological and tidal variables in the QRE, including the monthly MTBH, the corresponding LTL associated with the MTBH, the upstream ARD, and the downstream ATR, in relation to the CRA.

In Figure 14a, the coherence between the monthly MTBH and the CRA is sporadic, occurring over short time periods. The phase arrows predominantly point to the upper left, indicating a negative phase relationship, where increases in the coastal reclamation area are associated with decreases or lags in the MTBH. Figure 14b shows the wavelet coherence analysis of the LTL and the CRA, demonstrating that there is a positive correlation between these two variables only over the long term, specifically within the period range of 3 to 13 years.

Figure 14c shows that the ARD and the CRA exhibit significant coherence within 3–9 year periods, particularly from the mid-1980s to the late 2000s. Their phase arrows generally point to the lower left, suggesting that major reclamation activities tend to occur during low-flow, dry periods when runoff decreases. Figure 14d illustrates that there is almost no significant coherence between the ATR and the coastal reclamation area. However, most arrows in the coherent regions point to the left, indicating a negative phase relationship. This suggests that reclamation activities may impact tidal range dynamics, but changes in tidal conditions exhibit a lagged response.

4. Discussion

4.1. Temporal Patterns and Periodicities

The comprehensive time–frequency analysis conducted in this study provided significant insights into the periodic behavior of hydrodynamic variables in the QRE over the past five decades. Using wavelet transforms, distinct periodicities were identified, particularly in the MTBH at Yanguan, which exhibited dominant frequencies with cycles of approximately 21 years. Such findings are consistent with patterns observed in other estuarine systems, where decadal oscillations are often linked to large-scale climatic cycles like the Pacific Decadal Oscillation [48].

Additionally, periodicities were also evident in the LTL, the upstream ARD, and the downstream ATR, suggesting that these hydrodynamic variables are influenced not only by human activities but also by broader ocean–atmosphere interactions. Previous studies have similarly reported the influence of climatic variability on estuarine dynamics and tidal patterns, indicating that climate-driven oscillations play a pivotal role in shaping the hydrology of coastal systems [49]. Understanding these periodicities and their correlations with climatic events is crucial for developing long-term predictive models that inform adaptive management strategies for the estuary.

While the current analysis offers valuable insights into short-term and long-term correlations, incorporating older historical observations or indirect methods could enhance the overall assessment of estuarine dynamics. Although high-resolution data from earlier periods are limited, historical records or indirect approaches, such as modeling past conditions, could provide useful context for interpreting the observed changes.

4.2. Impacts of Runoff Variations

Runoff discharge is a critical factor influencing tidal bore dynamics, as our cross wavelet analysis reveals. The analysis of the relationship between the monthly MTBH and the upstream ARD demonstrated significant correlations at shorter periods (0.3 to 1.2 years), while their phase relationship varied over time. This variability suggests that while runoff influences tidal bore height, its effect is complex and modulated by other factors such as riverbed morphology, seasonal shifts, and upstream hydrological interventions. This observation is consistent with studies of other estuarine systems, where freshwater inflows have been shown to affect tidal wave transformation and the formation of tidal bores, albeit in a context-dependent manner [2].

At shorter timescales (0.3 to 1.2 years), high-runoff periods tend to attenuate tidal wave energy, reducing the intensity of tidal bores, while low-runoff periods enable tidal energy to propagate further upstream, enhancing tidal bore height [28]. Over longer periods (4.1 to 8.3 years), the results indicate a positive correlation, where the ARD changes several months before the monthly MTBH. Such correlations highlight the long-term impact of sustained hydrological changes, possibly induced by climatic events, on tidal bore behavior. Long-term runoff changes can have more complex impacts, occasionally increasing tidal bore height due to alterations in riverbed morphology or tidal dynamics.

These findings align with the general understanding that upstream modifications, such as dam constructions and reservoir operations, alter flow regimes and subsequently influence tidal dynamics [50,51]. Moreover, environmental factors like sediment load can affect riverbed morphology and sedimentation patterns, further influencing tidal bore characteristics over different timescales. Incorporating sediment transport data into future analyses could provide deeper insights into the interplay between runoff variations, sediment dynamics, and tidal bore behavior.

Our mutation analysis indicates a significant shift in the ARD around 2010, primarily related to increased rainfall, marking the onset of a high-flow period. This change could be attributed to climate variability or regional climatic events that altered precipitation patterns, leading to increased surface runoff and higher net discharge levels in the river. Such fluctuations in discharge not only affect the hydrological conditions at the entry of the estuary but also likely have a significant impact on the intensity of tidal bores. This highlights the necessity of considering climatic factors when assessing hydrological changes and their effects on tidal bore behavior [29].

4.3. Effects of Coastal Reclamation

Coastal reclamation in the QRE has dramatically reshaped its estuarine morphology, affecting both tidal dynamics and sediment transport processes [21]. Since the 1950s, reclamation efforts have expanded significantly, with major activities occurring in the 1970s and 2000s to support urban and industrial development. The cross wavelet analysis between the monthly MTBH and the CRA revealed a significant negative phase relationship, with coherence over periods ranging from 1 to 13 years. This suggests that as reclamation activities increased, tidal bore heights decreased, likely due to morphological changes that dampened tidal wave energy.

Reclamation projects, particularly those narrowing the estuarine channel and modifying the tidal prism, disrupt the conditions necessary for tidal bore formation. These structural changes can alter tidal wave propagation, reduce cross-sectional flow areas, and shift the balance between tidal and river forces, resulting in attenuated tidal bore intensities [24]. Such effects have also been documented in other estuarine systems worldwide, where morphological changes due to coastal engineering have led to similar reductions in tidal bore intensity and frequency [2,27].

Our wavelet coherence analysis suggests that the impacts of reclamation on tidal dynamics may not be immediate but develop over several years as the estuarine system gradually adjusts to the new morphology [52]. This lag effect is critical to understanding the temporal dynamics of human interventions in estuarine environments, emphasizing the need for long-term monitoring to assess the cumulative impacts of reclamation activities [53].

4.4. Integrated Management Implications and Future Directions

The interplay between runoff variations and coastal reclamation underscores the complexity of managing estuarine dynamics in the QRE. The negative correlation between the upstream ARD and the CRA suggests that reclamation activities often coincide with periods of reduced runoff, possibly because the exposure of tidal flats under lower river flows provides more favorable conditions for construction. However, this combination of reduced freshwater inflow and altered estuarine morphology exacerbates changes in tidal dynamics, affecting tidal ranges and bore characteristics. This finding highlights the importance of developing integrated management strategies that consider both hydrological variability and anthropogenic interventions [29].

To ensure the sustainable management of the QRE, future research should use high-resolution numerical modeling techniques to simulate various scenarios involving different runoff regimes, reclamation extents, and climatic conditions. Numerical models have been proven effective in predicting hydrodynamic changes and can provide valuable insights for adaptive management planning [16]. Furthermore, long-term monitoring programs that track hydrodynamic, sedimentary, and ecological changes are crucial for enhancing predictive capabilities and informing management practices that balance development with environmental preservation [32]. Moreover, utilizing advanced technologies, such as sensory instrumented particles, for environmental monitoring [54,55] can improve the accuracy of sediment transport measurements, leading to more precise modeling.

A deeper understanding of the interactions between natural and human-induced factors will better equip policymakers and practitioners to mitigate impacts on tidal bore dynamics. Developing strategies that foster sustainable estuarine development while maintaining ecological balance is crucial. Long-term monitoring and adaptive management approaches are essential to balance natural hydrological variations with human activities, such as coastal reclamation, ensuring the resilience of the estuary in the face of ongoing environmental changes.

5. Conclusions

This study examined the impacts of runoff variations and coastal reclamation on tidal bore dynamics in the QRE over the past five decades. Statistical and time–frequency analysis methods, including wavelet transforms, cross wavelet transforms, wavelet coherence, the Mann–Kendall tests, and an ordered clustering analysis, were applied to identify significant patterns and correlations among tidal bore height, river discharge, and reclamation activities. The results demonstrate that both natural hydrological changes and human interventions have substantially influenced tidal bore characteristics, affecting their intensity and behavior at different temporal scales.

The analysis revealed that increased freshwater discharge tends to reduce tidal bore height in the short term (0.3 to 1.2 years) by attenuating tidal wave energy, while decreased runoff allows tidal energy to penetrate further upstream, enhancing the bore. Over longer timescales (4.1 to 8.3 years), sustained hydrological changes, possibly linked to climatic events, show a positive correlation with tidal bore height changes, indicating a delayed but significant impact. Coastal reclamation activities, particularly those involving channel narrowing and morphological alterations, have had a pronounced negative effect on tidal bore height, especially after the 2000s. The lagged effects observed suggest that the estuary adjusts to new morphologies over several years, emphasizing the need for long-term monitoring.

These findings highlight the complex interplay between natural factors and human activities in shaping tidal bore dynamics. Understanding these interactions is essential for developing integrated estuarine management strategies that balance environmental preservation with developmental needs. Future research should focus on using high-resolution numerical models and establishing comprehensive, long-term monitoring programs that track hydrodynamic, sedimentary, and ecological changes. By deepening our understanding of the factors influencing tidal bore behavior, policymakers and practitioners can better mitigate adverse impacts and promote the resilience of the estuary amidst ongoing environmental changes.