Quantitative evaluation of the evolution of the ecohydrological regime and its drivers in the Han River Basin

The river ecohydrological regime determines the interactions between various aspects of river ecosystems, such as material cycling, energy transfer, habitats, and organisms, and thus impacts river ecosystem health (Guo et al. 2018; Jiang et al. 2021a, 2021b). In recent decades, with economic and social development and population growth, intense anthropogenic disturbances have altered the hydrological cycle processes in the watershed (Liang et al. 1996; Oki & Kanae 2006), changing the ecohydrological regime of the river to varying degrees. These large-scale human activities are bound to seriously disrupt the hydrological regime of the basin.

In order to describe changes in the hydrological regime, Richter et al. (1996) proposed the indicators of hydrological alteration (IHA), a system of 33 indicators classified into five categories (flow, event, frequency, duration, and rate of change). These indicators are closely related to ecosystem functioning and health and contribute to the understanding of the impacts of hydrological changes on ecosystems. Later, in order to better measure changes in the hydrological regime, Richter et al. (1998) proposed a univariate integrated hydrological change assessment, the range of variation (RVA) method. The RVA approach quantifies the range of variability of hydrological indicators and provides specific values to assess hydrological variability. Ju et al. (2022) conducted a comprehensive assessment of the degree of change in the hydrological regime in the Weihe River Basin using the RVA method. Guo et al. (2018) assessed and analyzed the degree of hydrological change in the lower reaches of the Three Gorges and the impacts of hydrological change on the habitat conditions of important fish spawning sites using the RVA method. However, the correlation between the 33 indicators needs to be better resolved and reflects the specific changes in ecological flows in the river. To address this issue, Vogel et al. (2007) assessed river flow surpluses and deficits based on dimensionless ecological index analyses (ecological surpluses and ecological deficits) based on flow duration curve (FDC), with targets and thresholds for each index (25th percentile and 75th percentile). FDC is able to provide comprehensive information from low to high flows, not just a few statistical indicators. This makes it more comprehensive and accurate in dealing with variability and extremes of flow. Gao et al. (2009) analyzed that ecological surplus and ecological deficit can better describe the changes in the hydrological regime, and Wang (2021) comprehensively evaluated the impacts of the construction of reservoirs on the hydrological regime in the Taizi River Basin based on the ecological flow index of FDC. Meanwhile, for the analysis of the driving force of runoff change, in recent years, scholars have mainly used various hydrological models, elasticity coefficient methods, and other methods to analyze the influencing factors of runoff change in different rivers. Choudhury (1999) and Yang et al. (2008a, 2008b) proposed a coupled equation of water-heat balance in the watershed based on the Budyko hypothesis and classified the climate change impacts into precipitation and potential evapotranspiration. The method has been widely applied in important river basins worldwide (Liu et al. 2019; Yan et al. 2020).

Although the Han River is one of the major tributaries of the middle reaches of the Yangtze River, with large runoff and precipitation, and has a great potential for water resources development (Deng et al. 2015), there is still a lack of research on its ecohydrological regime. Previous studies on the Han River were mostly confined to the annual scale, with a single evaluation method. However, due to the combined effect of climate change and human activities, the changes in hydrological processes in the basin are no longer reflected only in a single time scale or a single direction but are more three-dimensional and complex. Zhang et al. (2020) found that hydrological changes in changing environments span different spatial and temporal scales. Therefore, this study comprehensively assessed the hydrological regime of the Han River Basin through various ecohydrological indicators, analyzed the hydrological characteristics of the basin from different perspectives and aspects, and further comprehensively considered the impacts of climate change and human activities on the hydrological regime, which can provide a scientific basis for the future management of water resources in the Han River.

In order to address the shortcomings of the current studies on the changes in the hydrological regime of the Han River Basin, this study is divided into the following three main steps: (1) using the Mann-Kendall (M-K) test and the cumulative distance level method, combined with the double cumulative curve method, to analyze the characteristics of long-term series changes of hydrometeorological elements in the Han River; (2) using the IHA-RVA method, ecological surplus, and ecological deficit, and kernel density estimation (KDE), to provide a comprehensive quantitative evaluation of the degree of changes in the hydrological situation in the Han River; (3) analyzing the influencing factors of runoff changes on an annual scale, including precipitation, potential evapotranspiration, and underlying surface based on the Budyko hypothesis.

In this study, daily flow data from the Xiangyang Hydrological Station from 1975 to 2019 and daily meteorological information from 16 different meteorological stations were used, as well as land use data and DEM data for 5 periods: 1980, 1990, 2000, 2010, and 2020. The daily flow data were obtained from the Yangtze River Water Conservancy Commission (http://www.cjw.gov.cn/). The meteorological information was obtained from the National Meteorological Science Data Centre (https://data.cma.cn/). The land use data were obtained from the Environmental and Resource Science Data Centre of the Chinese Academy of Sciences (https://www.resdc.cn). DEM data were obtained from the Hydrological Basin (https://www.hydrosheds.org).

The process of hydrological change in the Han River Basin is very complex. This study used the M-K test and the cumulative distance level method to determine the year of change of annual mean runoff data and validated it using a double cumulative curve. In addition, trends of runoff, precipitation, and potential evapotranspiration were analyzed.

The cumulative distance level method is a non-linear statistical method that reflects a sequence change through a curve (Usman Liaqat et al. 2020). From the fluctuation of the cumulative anomaly curve, the evolutionary trend and change of the sequence can be judged, and according to the inflection point of the cumulative anomaly curve, the mutation point can be judged. The cumulative distance level method can better distinguish the inter-annual changes of runoff and precipitation, which helps to calculate the annual runoff and precipitation anomalies and accumulate them year by year to get the cumulative anomaly curve.

The double cumulative curve method (Ran et al. 2010) describes the consistency of the relationship between two variables by accumulating them at the same time step, with the successive cumulative values of the two variables as the horizontal and vertical coordinates, respectively, and plotting the relationship between the two in a right-angle coordinate system. Under normal circumstances, the relationship is a straight line with little fluctuation; if the double cumulative curve deviates significantly at a point, indicating a change in the relationship, influenced by other factors, thus determining the point of mutation.

Richter et al. (1996) proposed a hydrological regime change index (IHA) to reflect the river's ecohydrological regime. The index includes five indicators of flow, event, frequency, duration, and rate of change (Table 1). It can quantitatively analyze the degree of impact of climate change and human activities on the Han River's hydrological regime and provide a basis for estimating the river's ecological flow series.

Annual and seasonal FDC curves can also be constructed using daily flow data. Gao et al. (2012) proposed 75% quantile and 25% quantile as thresholds. The critical river ecological health protection range was identified, and the pre-hydrological change period was used as the baseline. The area between a given annual or seasonal FDC and an FDC above the 75th percentile was defined as an ecological surplus, and the area between a given annual or seasonal FDC and an FDC below the 25th percentile was defined as an ecological deficit. A river health system is considered to be in the healthy range if the river ecosystem lies between the 25th and 75th percentiles (Wang et al. 2017). In this study, ecological surplus and ecological deficit indicate that the actual flow is below or above the runoff value required by the river ecosystem, respectively, and both are referred to as ecological flow indicators.

Despite the wealth of research on ecological flows nowadays, there needs to be a harmonized definition and a standard method of calculating them. Several methods for measuring ecological flow criteria exist, such as the Tennant and Smakhtin methods, which are susceptible to extreme events and unequal annual distributions (Tan et al. 2018). Therefore, in this study, the river flow corresponding to the extreme value of the monthly probability density curve was taken as the most suitable ecological flow for that month. Then, the corresponding threshold of the most suitable ecological flow was determined.

KDE is a non-parametric method for estimating the probability density function of the data. KDE is based on the kernel function and estimates the probability density of the flow for each month by weighted averaging the kernel function near each data point with certain bandwidth parameters. The flow value corresponding to the maximum KDE is the optimum ecological flow for the river ecosystem for a given month (January–December) (Jiang et al. 2021a, 2021b). The annual-scale and seasonal-scale optimum ecological flow thresholds were obtained based on the 75% quantile and 25% quantile proposed by Gao et al. (2012) as the high- and low-flow thresholds.

The Budyko hypothesis (Sellers 1976) is based on the watershed water balance equation. On multi-year time scales, Budyko assumed that the actual evapotranspiration from the basin was influenced by both water and heat conditions, leading to the development of the coupled water-heat balance theory, which has been validated and developed over the years to derive numerous improved equations. For the analysis of the drivers of runoff change, the Choudhury–Yang formula (Equation (6)) is used in this study. The Choudhury–Yang formula is physically explicit and simple to compute. It considers the effect of spatial heterogeneity of the catchment compared with the mathematical–statistical methods commonly used in hydrological analyses (Choudhury 1999). The calculation principle consists of determining the elasticity coefficients of the respective influencing factors and then using the elasticity coefficients to calculate the runoff changes due to precipitation, potential evapotranspiration, and underlying surface features (Roderick & Farquhar 2011). Given the basin drought index ⁠, the elasticity coefficients corresponding to the underlying surface, precipitation and potential evapotranspiration correspondences can be calculated using the full differential form of the coupled water-energy balance equation according to the definition of the elasticity coefficient (Equations (7)–(9)). From the elasticity coefficients, the change in runoff due to each factor can be calculated (Equation (10)). The difference in runoff depth change is obtained by summing (Equation (11)). Then, the contribution of each factor to the runoff change can be calculated (Equation (12)). The above equations are shown as follows:

Among the highly variable hydrological indicators, the Han River Basin had the highest degree of variability in the June median flow (90.8%). Significant changes in the June median flow may affect fish reproductive behavior, juvenile survival, and fish population decline. June is usually an important period for fish spawning and juvenile growth. With an annual average 30-day maximum flow variability of 72.41%, extreme high-flow events can lead to streambank erosion and habitat destruction. Long-term high flows can also alter streambed structure, affecting fish refuges and foraging grounds. The degree of variability in the number of high-flow pulses was also 90.8%, and changes in the number of high-flow pulses can lead to the destabilization of aquatic ecosystems, increasing the stress on fish survival and possibly causing the decline or disappearance of certain sensitive species, as well as affecting predators dependent on these fish. Hydrologic indicators of moderate variability include May, August, September, and October median flows, annual mean 1-day, 3-day, and 7-day maximum flows, baseflow indices, the number of low-flow pulses and the duration of high-flow pulses, the average rate of flow decline, and the number of inversions.

The construction and operation of the middle and lower reaches of the Han River terraced hydropower stations, which blocked mature parental fish in the middle and lower reaches of the Han River and even in the mainstream of the Yangtze River from traveling upstream to the middle reaches of the Han River and the spawning grounds of the tributaries of the Tangbai River. It was difficult to satisfy the needs of the hydrological conditions for fish spawning and incubation, and the breeding population of fish-producing drifting spawns declined significantly, with a sharp decline in the scale of their spawning grounds (Li et al. 2005; Xie et al. 2009). In addition, the degree of change in the overall hydrological indicators and the degree of change in the hydrological indicators for each group was calculated. Groups 1, 2, 4, and 5 have a medium change of 43, 42, 53, and 56%, respectively, while group 3 has a low change of 18%. The overall hydrological indicators showed a change of 42%, which is a medium change. Overall, the Han River Basin has experienced moderate hydrological changes, which have challenged the stability of the ecosystem and led to changes in the structure of the food web.

On an annual scale, the annual ecological surplus has a small change and a decreasing trend, while the annual ecological deficit has a large change and an increasing trend. This directly affects the habitat of aquatic organisms in the basin. Reduced water resources may lead to higher water temperatures and lower dissolved oxygen levels, which in turn may affect the survival and reproduction of fish and other aquatic organisms. Before and after the mutation, the ecological surplus was larger than the ecological deficit in the natural flow state before the mutation (1975–1990), while the ecological deficit was larger than the ecological surplus after the mutation (1991–2019) due to a variety of factors such as climate change and human activities. Figure 7(a) shows that the ecological surplus and precipitation distance level change relatively consistently on an annual scale. However, the ecological deficit and precipitation distance level do not agree perfectly. This event is also well reflected in the change in their correlation coefficients in Table 3 (from 0.5 before the mutation to 0.3 after the mutation). Climate change after the mutation led to changes in precipitation's spatial and temporal distribution, with some years and seasons becoming more and more arid. In some dry years, such as 1991–2002, the ecological surplus is zero; in some dry years, the ecological surplus is also zero in some water-rich years. On the one hand, warmer temperatures change the vegetation cover and soil moisture, affecting the water cycle process, and the amount of evapotranspiration is increased, further decreasing the available water resources and reducing the ecological surplus. On the other hand, water storage in the reservoirs is affected because, with economic development and population growth, the water supply demand in the basin increases yearly, and the reservoirs prioritize meeting industrial and domestic water supply needs.

Compared with changes in ecological flow metrics and precipitation spacing on annual scales, the difference between the two increased on seasonal scales (Figure 7(b)—7(e)). Changes in ecological flow metrics during the summer, which accounts for most precipitation, were most similar to changes on the annual scale. As seen in Table 3, ecological surplus before the autumn mutation was in good agreement with precipitation distance level (correlation coefficient 0.71), while in spring, summer, and winter, ecological flow metrics were in poor agreement with precipitation distance level, with correlation coefficients less than 0.5. This is because this area is densely populated, and converting forested acreage, which is more disturbed by human activities, to construction land reduces the natural vegetation cover. This leads to an increase in the impermeable area of the ground surface, exacerbating the phenomenon of floods and droughts and further affecting the ecological flow indicators. Compared with the natural period (1975–1990), the magnitude and trend of changes in ecological flow indicators and precipitation distance level changed in all seasons during the change period (1991–2019). Ecological surpluses were significantly reduced, and ecological deficits were significantly increased during the summer months, with ecological deficits being more common in spring and summer, which can lead to increased summer droughts, affecting plant emergence and growth, aquatic reproduction, and water quality in habitats; in winter, especially in 1989–2007, the precipitation distance level showed an overall increasing trend, but the ecological surplus was almost zero. This indicates that the correlation between the ecological flow indicators and the precipitation distance level in all seasons after the mutation was generally reduced. This is because large-scale water conservancy projects such as the Danjiangkou Reservoir, while regulating flow and providing irrigation and water supply, also affect the natural flow state, resulting in a decreased ecological surplus, and the impacts of human activities began to intensify gradually.

Determining optimum ecological flow thresholds is critical to maintaining the ecosystem health of rivers, lakes, and other water bodies. Understanding the variation of the optimum ecological flow thresholds can help develop rational water scheduling programs to meet human water needs while protecting ecosystems' hydroecological needs. In this study, we used the KDE method to fit monthly flow series to determine the optimum ecological flow for each month.

The Budyko hypothesis was used to analyze the driving force analysis of Han River's hydrological regime changes. Calculated results were obtained for multi-year average precipitation, average potential evapotranspiration, multi-year average runoff depth, underlying surface parameters, precipitation, evapotranspiration, and elasticity coefficients of human activities in the Han River Basin (Table 4).

Calculations showed that compared with the pre-mutation period, P and R in the Han River Basin decreased after the mutation, with rates of 4.88 and 27.67%, respectively; ET₀ and n increased, with rates of 2.97 and 23.79%, respectively; and the desiccation index (ET₀/P) increased slightly compared with the baseline period, while the runoff coefficient decreased slightly compared with the baseline period. The elasticity coefficients of Han River Basin runoff to precipitation, potential evapotranspiration, and subsurface after abrupt changes were 2.18, −1.18, and −0.995, respectively, indicating that a 1% increase in precipitation, potential evapotranspiration, and subsurface caused a 2.18% increase in runoff depth, a 1.18% decrease in runoff depth, and a 0.995% decrease in runoff depth, respectively; the changes in the runoff depth in the Han River Basin were positively correlated with P, and negatively correlated with ET₀ and n. The runoff depth was also positively correlated with P and negatively correlated with ET₀. Since the absolute size of the elasticity coefficient responds to the degree of runoff sensitivity to the influencing factors, according to the results, runoff depth is the most sensitive to changes in P and the least sensitive to changes in n.

We further investigated the contribution of each influencing factor to the runoff change in the Han River Basin (Table 5). The results showed that precipitation caused a change of −28.85 mm in runoff at Xiangyang station; potential evapotranspiration caused a decrease of 8.53 mm in runoff, and the underlying surface resulted in a reduction of 55.43 mm in runoff. The contribution of the change in the underlying surface to the change in runoff was the largest at Xiangyang station, at 59.43%; the contribution of precipitation to the change in runoff at Xiangyang station was second, at 31.08%; and the contribution of potential evapotranspiration to the change in runoff at Xiangyang station was the smallest, at 9.19%. This is in general agreement with the findings of Peng et al. (2021) on attribution analysis of runoff changes in the Han River Basin. Therefore, the change in the underlying surface is the main factor contributing to the decrease in runoff at Xiangyang station, followed by precipitation, and potential evapotranspiration has the least influence. However, it is worth noting that the runoff change is least sensitive to the underlying surface. Nonetheless, the underlying surface contributes the most to changes in runoff. Therefore, there is a need to examine the influence of the underlying surface on runoff in more detail.

In recent years, the development and utilization of hydropower in the Han River Basin have increased significantly (Wang et al. 2015). There are more than 2,700 reservoirs in the basin, with a total capacity of nearly 33 billion cubic meters. Among them, the Danjiangkou Reservoir in the upper reaches has the largest capacity of 29.05 billion cubic meters. During the operation of the Danjiangkou Reservoir, flows decrease and flood storage increases in April and May, and from June to September, the Danjiangkou Reservoir begins to release water downstream for hydroelectric power generation, resulting in increased flows during the flood season. The construction of water conservancy and hydropower projects has a significant impact on basin runoff.

Changes in the hydrological regime of the basin are more complex due to the uneven distribution of climatic factors within the year and the increase in human activities such as water abstraction and transfer. Traditional hydrological studies can no longer meet the planning requirements for the sustainable use of local water resources. We used a systematic and scientific approach to conduct a multidimensional study of the Han River Basin. We found that the overall trend of runoff in the Han River Basin is decreasing, and the overall hydrological situation is changing moderately, which not only negatively affects the reproduction and survival of local organisms but also is not conducive to the ecological restoration of the downstream wetlands. The decrease in ecological surplus and the increase in ecological deficit indicate that the ecological flow has been under tension for a long time. Based on the analysis of ecological flow thresholds, we found that the optimal ecological flow thresholds required to maintain ecological balance varied in different seasons. In addition, we used Budyko to quantitatively analyze the influencing factors of runoff changes on an annual scale and found that human activities were the main driver of its runoff changes, accounting for 59.73%. The analyses developed in this study can be used to study the hydrological regime of watersheds under environmental changes from multiple perspectives and to quantify the drivers of runoff changes in watersheds, thus achieving a comprehensive evaluation of watersheds.

Hydrological changes in the Han River Basin are closely linked to human activities, and the negative impacts of climate change can be mitigated to some extent through appropriate management and intervention measures, such as the construction of water storage dams. The removal of small-scale water conservancy projects with a large area of influence and low efficiency so that their river connections can be restored and natural flow patterns and hydrological rhythms can be restored to provide a reasonable habitat for some migratory fish species. The impact of climate change and human activities on water resources is a global issue. The results of this study contribute to the understanding of the evolutionary characteristics and driving mechanisms of the hydrological regime of the basin under changing environments and provide valuable lessons for other regions around the globe that are facing similar hydrological challenges. The positive measures taken in the Han River Basin can serve as a reference for other regions to cope with the impacts of climate change. However, each basin is unique, and while drawing on the experience, it needs to be adapted and optimized in the light of local realities.

In order to assess the impacts of human activities and climate change on runoff more comprehensively, future studies should combine other hydrological models, such as Soil and Water Assessment Tool (SWAT) and Variable Infiltration Capacity (VIC), to conduct comparative analyses of runoff change attribution to reduce model uncertainty. In addition, human activities were analyzed as a whole in this study, and future studies can further distinguish the impacts of different types of human activities (e.g., agricultural irrigation, industrial water use, urbanization) on runoff, which can provide a scientific basis for developing more accurate water resource management strategies.

H.W: funding acquisition, project administration, resources, investigation, supervision. H.Y: conceptualization, data curation, formal analysis, investigation, methodology, resources, software, validation, visualization, writing – original draft, and writing – review and editing. Q.W.: investigation, formal analysis, methodology, validation, and visualization. S.C.: structure design and method determination. X.B.: in-depth exploration and interpretation of the paper results. J.L.: data collection and processing. Y.M.: academic exchanges and paper discussions. G.W.: revision and review of papers. W.G.: funding acquisition and project administration.

The authors thank their brothers at the North China University of Water Resources and Electric Power for their comments and help with this study.

This study was supported by the Basic Research Project of Key Scientific Research Projects of Colleges and Universities of Henan Province (24ZX007).

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

The authors declare there is no conflict.