Effects of the Water Diversion Project from the Three Gorges Reservoir to the Hanjiang River on diatom blooms in Middle and Lower reaches of the Hanjiang River

Over the development of human society, the phenomenon of water eutrophication has become increasingly prominent, which is commonly followed by algal blooms. Algal blooms have become a global aquatic eco-environmental problem, which seriously threatens water safety (Vörösmarty et al. 2010; Xia et al. 2019). River blooms were less common than lake and reservoir blooms (Mitrovic et al. 2008; Jung et al. 2009; Yang et al. 2011). However, due to urbanization and rapid economic development, algal blooms appear in many rivers both in China and abroad (Yang et al. 2011; Xue et al. 2023). Furthermore, because of the dense interweaving of river systems and strong hydrodynamic force, river blooms are often characterized by wide impacts, complex outbreak causes, and great difficulties associated with their control (Li et al. 2020); usually, river blooms have more serious negative impacts.

The Hanjiang River is one of the most important tributaries of the Yangtze River. Danjiangkou Reservoir (DJK) in the upper reaches of the Hanjiang River is the water source of the middle route of the South-to-North Water Diversion Project (SNWDP). Algae blooms have occurred frequently in the Middle and Lower Reaches of the Hanjiang River (MLHR) in recent years, which has become the most important aquatic eco-environmental problem and has attracted much attention because of the sensitivity and uniqueness of the region. It was found that river blooms are dominated by diatoms with low-temperature tolerance and occur mostly in winter and spring (Yang et al. 2011). Outbreaks of algal blooms in large rivers are usually closely related to changes in the hydrological regime (Ferris & Lehman 2007).

Many researches have been conducted about the driving factors of algae blooms in the Hanjiang River. Generally high nutrient concentration is not regarded as a restricted factor that results in algal blooms in the Hanjiang River. The blooms are more apparently affected by external factors such as hydrology and meteorology (Li et al. 2020). Cheng et al. (2019) identified the water level of the Yangtze River estuary, total phosphorus (TP) levels, and water temperature as the main influencing factors of algal blooms. Xin et al. (2020) identified phosphorus as a key limiting factor for algae growth, and velocity threshold values of 0.74 m/s or a flow rate of 800 m³/s were proposed for the Xiantao section. The water level difference between Xiantao and Hankou was identified as an important factor and the response time of blooms to driving factors was found to be about 10 days (Xia et al. 2020; Liu et al. 2022).

Mitigation measures for the spreading of algae blooms in the Hanjiang River have been promoted in recent years. Xie et al. (2005) examined the impacts of SNWDP on algal blooms in the MLHR based on a model to calculate the blooms occurrence probability, and their results showed that the probability increased after water diversion. Yang et al. (2012) established a generalized additive model to analyze the dynamic responses of algae to environmental changes. Li et al. (2022) proposed an ecological scheduling scheme for the prevention of blooms by increasing the discharge from DJK and regulating fluctuations in the water level.

At present, research on the mechanism underlying explosive algae growth, as well as simulations of algae growth dynamics are rare. Yang et al. (2012) proposed a more water-saving discharge scheme for the DJK to prevent algal blooms. Xia et al. (2022) added hydrological factors into a traditional riverine algae growth model and combined it with a one-dimensional distributed hydrological model; algae growth under different SNWDP schemes was simulated and found that the SNWDP would significantly increase the possibility of bloom outbreaks.

The Water Diversion Project from the Three Gorges Reservoir to the Hanjiang River (WDP-TH) is the follow-up water source project of the SNWDP. Operating with cascade reservoirs of the Hanjiang River and other water diversion projects, WDP-TH will further change the hydrodynamic force, water quality, water temperature, and other conditions that restrict the growth and propagation of algae in the MLHR, affecting both the occurrence and development of algal blooms. This paper focuses on the impacts of WDP-TH operation. A mathematical model is constructed to quantitatively study the spatio-temporal distribution and variation of algal density under different scenarios. Ultimately, the effect of WDP-TH on diatom blooms in the MLHR is described. The results of this study provide a reference for the coordinated dispatch of water projects to both prevent and control algal blooms in the MLHR.

In recent years, the water quality of the MLHR was generally fine, and the annual water quality basically reached Class II of ‘Environmental quality standards for surface water’. In certain parts and at specific times, the water quality was Class III, and TP and ammonia nitrogen was identified as the main limiting factors. At present, algae blooms are the main eco-environmental problem in this basin. Since the first large-scale algal blooms were reported in February 1992, there have been 11 large-scale algal blooms outbreaks until 2021, most of which occurred from January to March during dry seasons in late winter and early spring. Recently, because of the influence of multiple factors, such as water depletion, cascade reservoir development, water diversion, and backwater in the estuary, the blooms outbreak frequency has shown an increasing trend. Moreover, the occurrence range has also moved from the lower reaches of the Qianjiang to the Xinglong Reservoir. In addition to water pollution, algae can easily block the filters of water plants along the river, affecting the normal water supply.

This study was divided into two scenarios: one with WDP-TH and another without WDP-TH. The study year is 2035, which is the year designated for WDP-TH completion. Considering the inflow and discharge of the DJK as well as the water recharge process of the WDP-TH, six calculation schemes were developed in both scenarios by selecting a typical median water year, a typical dry year, and a typical extremely dry year. The frequency analysis of the natural annual inflow of DJK from 1956 to 2016 showed that the water inflow frequency of 50, 90 and 98% responded to 1979, 1995 and 1999, respectively. Then the representative years of the typical median water year, typical dry year and typical extremely dry year were 1979, 1995 and 1999, respectively. Each scheme includes Danjiangkou, WangFuzhou, Xinji, Cui Jiaying, Yakou, Nianpanshan, and Xinglong reservoirs, which should all be in operation by 2035. The study area covers 652 km of the MLHR mainstream, reaching from the Danjiangkou Dam to the Hanjiang River estuary. The calculation time is from January to April during the period of the Hanjiang algae bloom.

Algal density is a key index that reflects algal blooms. In reference to the algal blooms occurrence in the MLHR, the most dominant algal group in algal blooms in the Hanjiang River over the years is diatoms (Li et al. 2020); therefore, in this paper, diatoms were selected as representative species for simulation analysis. Based on a model of hydrodynamic force, water temperature, and water quality, algal density was predicted in combination with a mathematical model of the growth dynamics of algae, thus providing a reference for the effects of algal blooms.

In the model, algal growth, metabolism, predation rate, temperature, light exposure, nutrient limit, and other parameters were mainly obtained through calibration. The critical flow velocity was determined according to the relationship between bloom occurrence and flow velocity in the MLHR over the years. In reference to the simulation results of velocity distribution during previous bloom occurrences (Li et al. 2022), the average flow velocity at Xiantao in the early stage of bloom outbreaks ranged from 0.33 to 0.51 m/s, and the average flow velocity was 0.67 m/s when blooms subsided. Considering the least favorable condition, the critical velocity v* in Hanjiang River was 0.67 m/s.

Daily monitoring data of flow rate, water quality, algae density, and chlorophyll-a at Shayang and Xiantao in MLHR during the diatom outbreak period from February 13 to March 7, 2018, were selected to calibrate the algae growth model. Daily monitoring data of algae density from January 19 to January 30, 2021, were selected for verification. In addition, as initial values of algae growth, metabolism, and predation rate, default values of the model were used, and initial values of temperature, light exposure, and nutrient limits were estimated based on monitoring data of bloom occurrence years in the MLHR from 2004 to 2018. Parameters were obtained as shown in Table 2.

The effect of water diversion to the MLHR on algae density in typical years was considered under the maintenance of the current water quality level of the Three Gorges Reservoir. Table 3 shows a comparison of peak values of algal density for January to April at five main control sections of the MLHR under two scenarios. The peak values at each section along the MLHR increase gradually from upper to lower reaches under both scenarios. This result is mainly caused by the gradual increase of nutrient concentrations along the MLHR (such as TP), as well as by the deceleration of flow velocity caused by backwater from the Yangtze River and the Xinglong Reservoir storage. Taking the scenario with WDP-TH as an example, the peak values gradually increased from 75 × 10⁴ to 1,128 × 10⁴ cell/L in a typical median year, and from 67 × 10⁴ to 863 × 10⁴ cell/L in a typical dry year, showing an increase of about 12–14 times from upper to lower reaches.

After the implementation of the WDP-TH, the peak values increased in the typical median water year but decreased in typical dry and extremely dry years. Following water diversion in a typical median water year, the peak value at Xiangyang increased slightly (by 2.7%), and peak values at Huangzhuang, Shayang, Qianjiang, and Xiantao (downstream of Xiangyang) increased by 14.1–26.9%. Increases were especially pronounced at Shayang and Qianjiang. Under the scenario with WDP-TH, the peak values at Qianjiang and Xiantao reached 1,048 × 10⁴ and 1,128 × 10⁴ cell/L, respectively. Following the water diversion in typical dry and extremely dry years, the peak values at Xiangyang, Huangzhuang, Shayang, Qianjiang, and Xiantao decreased by 21.2–45.9%. These decreases were relatively large downstream of Huangzhuang in the typical dry year, all of which were about 40%. In the scenario with WDP-TH, the peak values at Qianjiang and Xiantao in a typical dry year and at Qianjiang in an extremely dry year decreased to less than 1,000 × 10⁴ cell/L.

In reference to the classification of the algal blooms degree and the technical specifications for monitoring, an algae density ≥ 10⁷cell/L is assumed as a critical value for blooms occurrence. Possible situations of algal blooms at five main control sections in the MLHR from January to April were analyzed for both scenarios, and the results are shown in Table 4. In both scenarios, in each typical year, algal densities at Xiangyang, Huangzhuang, and Shayang did not exceed 10⁷ cell/L, and there were basically no algal blooms.

In a typical median water year, algae densities at Qianjiang and Xiantao did not exceed 10⁷ cell/L in the scenario without WDP-TH, while in the scenario with WDP-TH, on six days (February 3–8), the densities exceeded 10⁷cell/L at Qianjiang and on 19 days (February 4–16 and March 4–9) at Xiantao. After initiation of the WDP-TH, if external conditions remain unchanged, the algal bloom occurrence possibility in the reaches downstream of Shayang may increase from January to April in typical median water years.

In the scenario without WDP-TH, in a typical dry year, there may be 10 days (February 28–March 9) when algae blooms (i.e., algae density ≥ 10⁷ cell/L) may occur at Qianjiang and 15 days (February 27–March 13) when it may occur at Xiantao; in a typical extremely dry year, there may be 35 days (January 28–March 3) at Qianjiang and 45 days (January 26–March 11) at Xiantao when algae blooms may occur. In the scenario with WDP-TH, algal densities at Qianjiang and Xiantao in a typical dry year and at Qianjiang in a typical extremely dry year were all less than 10⁷ cell/L. The number of days when algal blooms can occur at Xiantao in a typical extremely dry year was reduced to 10 days (February 6–15). After WDP-TH initiation, if external conditions remain unchanged, the occurrence possibility of algal blooms in the reaches downstream of Shayang may decrease from January to April in both typical dry and extremely dry years.

The Hanjiang River, where algae blooms are much more severe, has attracted much attention, especially in the MLHR, where the water resources development and utilization degree is much higher. On the other hand, the TP concentration from upstream to downstream gradually increases in the MLHR, resulting in an increase in the risk of algal bloom. In recent years, due to the superposition of the SNWDP and the cascade development, the hydrological regime and water morphology of the MLHR have changed significantly. WDP-TH transfers water from the Three Gorges Reservoir to the Hanjiang River, which further changes the hydrodynamic force and nutrients in the MLHR. These changes affect algae growth and the occurrence of algal blooms. Therefore, taking the MLHR as the research object, based on the superposition of water diversion and cascade development, the effects of changes in hydrodynamic conditions and TP conditions on algal density are discussed, which is a typical example in the study of river blooms.

The sensitivity of algal density in response to changes in TP concentration was explored by the comparison of algal density changes under the same hydrodynamic conditions but under different scenarios of TP concentration. S3 and S4 were selected as the base scenarios. In order to compare with S3, Scenario S7 is added, in which the TP concentration of DJK is replaced by the TP concentration of WDP-TH. In order to compare with S4, Scenario S8 is added, in which the TP concentration of WDP-TH is replaced by the TP concentration of DJK.

The results of the relationship between changes rangeability of peak values of algal density and changes rangeability of TP concentration are shown in Figure 8(b). Compared to Scenario S3, in Scenario S7 the changes rangeability of TP concentration in Huangzhuang and Xiantao were 57.2 and 30.9% respectively, and the changes in rangeability of peak values of algal density in Huangzhuang and Xiantao were 27.8 and 27.0% respectively. Compared to Scenario S4, in Scenario S8 the changes in rangeability of TP concentration in Huangzhuang and Xiantao were −17.8 and −18.5% respectively, and the changes in rangeability of algal density peak in Huangzhuang and Xiantao were −18.8 and −19.4% respectively.

The analysis showed that the peak values of algal density increased with increasing TP concentration, under the same other conditions. A positive correlation was evident between the changes rangeability of peak values of algal density and TP concentration. It demonstrated that a high concentration of nutrients is necessary for the occurrence of diatom bloom, one important similarity between river-type diatom blooms and lake-reservoir diatom blooms (Smith 2003).

Researchers pointed out that, unlike lentic ecosystems such as lakes and reservoirs, physical factors such as river hydrological conditions and temperature are often more important than factors such as water nutrition level (Reynolds et al. 1994). In the following content, the combined effects of hydrodynamic conditions with flow rate and nutrient conditions with TP on algal blooms in the MLRH were further analyzed.

The analysis shows a negative correlation between algal density peak and flow rate. This is particularly evident in the Xiantao section with higher algal density. Because the higher flow rate has a more obvious scour effect on algae, it helps to avoid algae enrichment. Furthermore, a high flow rate may have some inhibitory effect on algal growth. The relation between the peak algal density and TP concentration was weaker than that of hydrodynamic, demonstrating that the dominant factor of blooms in the MLHR is hydrodynamic conditions rather than TP concentration.

From the comparison of S3 to S8 listed in Table 5, it was found that under the condition of fixed low TP input but increased flow rate, the flow rate in Huangzhuang section increased by 63 m³/s (12% increase), and the peak value of algae density decreased by 158 × 10⁴cell/L (56% decrease). The flow rate in the Xiantao section increased by 121 m³/s (24.4% increase), and the peak value of algae density decreased by 722 × 10⁴cell/L (51% decrease). This further indicates that the enhancement of hydrodynamic conditions in the MLHR in dry years has a more obvious inhibitory effect on algae growth than the promoting effect of the increase of TP concentration. Yang et al. (2011) pointed out that hydrological conditions are proved to be the most important factors affecting diatom blooms in rivers, and the operation of water conservancy projects has a particularly significant impact on hydrological conditions and diatom blooms, which is consistent with the conclusion in this manuscript.

Based on the research results, it can be seen that enhancing the hydrodynamic conditions and reducing the TP load are the keys to controlling algal blooms in the MLHR. If measures can be taken to increase the flow rate and reduce the TP concentration at the same time in the algal blooms prone period from January to April, the synergistic effect of inhibiting algal proliferation will be generated and a better control effect will be achieved. According to the scheduling rules of WDP-TH, there is more water recharge in dry years and less water recharge in normal years. If the surplus water volume can be used to recharge the MLHR by WDP-TH during the dry period in the median water year, the problem of increasing algal blooms risk caused by the increase of TP concentration after the water diversion in the median water year can be alleviated. Therefore, it is suggested to optimize the joint scheduling mode of WDP-TH and DJK, which can make use of the surplus water volume to recharge the MLHR from January to March, so as to reduce the probability of blooms occurrence as much as possible by enhancing the hydrodynamic conditions. At the same time, it is important to strengthen the control of internal and non-point source pollution and reduce the TP load from tributaries, which are important measures of algal bloom control for the long term.

By integrating the characteristics of diatom blooms in the MLHR, a mathematical model was constructed to simulate the growth dynamics of diatoms. This model focused on quantifying the effects of light exposure, water temperature, TP, and flow velocity on the growth, metabolism, and predation of diatoms. The spatial and temporal distribution and changes of algae density in the MLHR were simulated in two scenarios (with or without WDP-TH) in typical median water, dry, and extremely dry years. Moreover, the influence of hydrodynamic conditions and TP concentration on the peak values of algal density in MLHR were explored. The main conclusions are summarized as follows:

• (1) The algae density increased gradually from the upper to lower reaches along the MLHR. Compared with the scenario without WDP-TH, the peak values with WDP-TH construction at Xiangyang, Huangzhuang, Shayang, Qianjiang, and Xiantao increased by 2.7–26.9% in typical median water year and decreased by 21.2–45.9% in typical dry and extremely dry years. For both the scenario with WDP-TH and without WDP-TH, the peak values of algal density of each typical year did not exceed 10⁷cell/L at Xiangyang, Huangzhuang, and Shayang sections, and would not result in algal blooms; algal blooms may occur at Qianjiang and Xiantao sections. The time of algal density exceeding 10⁷cell/L was mainly found in mid-February or early March.

• (2) The obvious negative correlation between peak algae density and flow rate was found and a positive correlation was evident between the algal density and TP concentration. However, under the combined impact of hydrodynamic and TP, TP concentration was not as important as that of hydrodynamic condition.

• (3) The results provide valuable reference for diatom bloom control under different conditions. In the dry period of typical median water years, it is helpful to optimize the scheduling joint mode of WDP-TH and DJK, and increasing the flow rate of the water diversion can improve the hydrodynamic conditions, inhibit the growth of algae, prevent algae enrichment, and effectively reduce the probability of algal blooms occurrence. Meanwhile, it is an important measure to control diatom blooms in the long term by strengthening the pollution control of tributaries with large TP loads, and then reducing the TP concentration in the mainstream.

This work was supported financially by the National Natural Science Foundation of China (NO.52109005).

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.