Spatiotemporal hydrochemical variations in river water in the Qilian Mountains and their sources: a case study of the Binggou River Basin

The chemical characteristics of river water reflect not only the water quality of the river but also the characteristics of and changes in the ecological environment of the river basin (Qiao et al. 2017; Ramesh et al. 2018). Studies of the chemical composition and distribution patterns in rivers can more fully reveal the sources of chemicals in the river water, the relationship between the river and the natural environment, and the influence of human activities on the river (Pazi 2008; Tripathee et al. 2014). As global environmental problems intensify, scholars have increasingly focused on river chemistry problems (Connor et al. 2014; Nazri et al. 2016; Rodolfo et al. 2018). Research on river hydrochemistry first began in developed countries. Piper (1984) used trilinear charts to analyze the chemical composition of river water, and Gibbs (1970) studied the main sources of ions in rivers through the concentration relationships between total dissolved solids (TDSs) and anions and cations. With the turn of the twenty-first century, academicians in developed countries began concentrating on the relationship between the ion content in the water of large catchments and regional geological lithology and climatic conditions, such as the Amazon basin (Mortatti & Probst 2003), the Mackenzie River of North America (Millot et al. 2003), and the Lena River Basin in Asia (Huh et al. 1998). The water chemistry characteristics of rivers can reflect water quality, and the analysis of river water quality based on water chemistry parameters is also a focus of water chemistry research (Aminiyan et al. 2016).

In China, studies of river water chemistry began in the early 1960s. Le & Wang (1963) analyzed the temporal and spatial variation in the chemical properties of rivers in China and provided a general picture of the chemical characteristics of rivers. Subsequent research on river water chemistry developed quickly in China. Studies of the ion chemical characteristics of river water in the Yangtze River, the Yellow River, and the Xiangjiang River found that ions are most obviously affected by rock weathering (Chen et al. 2006; Li et al. 2017). Some scholars have studied the concentration and characteristics of inorganic ions in different water bodies in the Hengduan Mountains based on the study of the hydrological relationships of different water bodies (Zhu et al. 2015). Related studies in the Qilian Mountains have mainly been concentrated in the western Qilian Mountains (Li et al. 2015; Liu et al. 2016). The characteristics of the temporal and spatial distributions of water chemistry and control factors have been studied extensively in inland river basins, including the Tarim River (Xiao et al. 2016), the Heihe River (Wen et al. 2004), and the Shiyang River (Yang et al. 2018). Although river chemistry research in China is relatively mature, further strengthening of the understanding of small watersheds in source areas of inland rivers is needed.

The basin of the Shiyang River, which is one of the three largest inland rivers in the Hexi Corridor of Gansu, is located in China between longitudes 102°22′ and 104°16′E and latitudes 36°29′ and 39°27′N (Shi et al. 2012). This basin belongs to the typical continental climatic zone, with little precipitation, strong evaporation, and drought (Zhou et al. 2012; Zhang 2015). The watershed system originates from the southern Qilian Mountains, and melt-water and precipitation in mountain areas are the main sources of supply for the river. As the main river in the region, the Shiyang River supports not only the regional social and economic development and people's livelihoods but also the regional ecological security. However, due to the unreasonable utilization of water resources, ecological, social and economic problems have become increasingly serious, including the deterioration of water quality, reduced groundwater levels, land degradation, loss of biological resources, unsustainable economic development, and intergenerational conflict in water use (Ren 2017). Despite some achievements in recent years, the outlook remains grim. Consequently, better solutions to the water environmental problems in the Shiyang River Basin are urgently needed.

The Binggou River, which is located in the core water source area of the Shiyang River Basin, is one of the National First-grade Drinking Water Conservation Areas. Thus, this paper analyzes the chemical composition of the river water of the Binggou River Basin in the Qilian Mountains and the upper reaches of the Shiyang River using measured data. In addition, the spatial and temporal distribution and sources of different ions and water quality and the influence of human factors on the river are discussed. We anticipate that the findings from this study will help basin water resource managers optimize the management measures.

The Binggou River is a tributary of the upper reaches of the Shiyang River in the Qilian Mountains. The elevation of the Binggou River Basin ranges from 1,928 to 4,736 m, and the height difference in the whole basin is 2,808 m (Figure 1). The annual average temperature, precipitation, and evaporation are 5 °C, 200–400 mm, and 1,300 mm, respectively, with a relatively dry climate. The rock type is mainly granite, and there are massive sulfide deposits of volcanic origin (Wang et al. 2011). The basin area is approximately 326 km². As the altitude increases, the main land types are desert, grassland, woodland, alpine meadow, and permanent glacier.

The upstream area is mainly divided into two tributaries. Compared with the west branch, the east branch is shorter in mileage but broader with numerous small tributaries and a wide basin area. As a result, the west branch is sparsely populated, with only a small amount of farmland along the river banks and relatively few human activities. By contrast, along the east branch, there are many towns and villages and large areas of farmland along the riverside.

In total, 72 water samples were collected from eight sampling sites (Table 1) in the river basin once a month between September 2016 and May 2017. Due to a road collapse, samples were not collected from June to August. The water samples were collected in clean 120-mL polyethylene bottles. Prior to sample collection, each sample bottle was washed with deionized water and then rinsed twice with the river water to be collected. After sample collection, the bottle was immediately sealed, and a label with the corresponding time and place of the collection was attached. In addition, GPS was used to record the coordinates of the sampling points, and the geographical environment near the sampling point was recorded. All samples were frozen and transported to the State Key Laboratory of Cryospheric Science, Cold and Arid Regions Environment and Engineering Research Institute and Chinese Academy of Sciences for testing and analysis. Electrical conductivity (EC) and pH values were measured by an EC detector (LT-EC) and a pH instrument (PHS-3D), respectively. The pH meter was calibrated before each test, and the instrument was washed with distilled water before measuring each sample. TDSs and salinity (SAL) were tested by HANNA (HI3512). Before testing ion content, all samples were filtered using 0.45-μm polycarbonate microfiber filters. Major cations (Na⁺, K⁺, Mg²⁺, and Ca²⁺) were analyzed by a Dionex-600 ion chromatograph, and major anions (Cl⁻, SO₄²⁻, and NO₃⁻) were analyzed by a Dionex-3000 ion chromatograph. Bicarbonates were determined according to the charge balance as described by Trower (2009). Comparison testing showed that the contamination of the samples during sampling, transportation, and treatment could be ignored.

Land use and cover data were acquired by a visual interpretation combining a long-term field survey with remote sensing images such as Landsat 8, Sentinel, and Google. The accuracy of the interpretation was 92.5%. The track number of the TM image data was 132–33, and the time of image generation was 10 August 2016. The geometric correction was carried out based on control points collected from the map at a scale of 1:10,000,000, and the position error was within 2 pixels. Based on the spectral information and texture characteristics of the remote sensing images in combination with the characteristics of the study area, the land use and cover in the study area were divided into eight types. Temperature and precipitation data were grid data from the same period obtained from the China Meteorological Data Service Center (CMDC). Runoff data were acquired from the Management Bureau of Shiyang River Basin. Evaporation data were calculated using the Penmanship–Menteith model and the H formula.

Mathematical and statistical analyses of the meteorological and water chemistry data were performed in the software SPSS 21, and the results of the analyses were plotted using the software Origin 8.0. The analysis, processing and mapping of the land use data were performed on the ArcGIS 10.2 platform. The Piper diagram method for water chemistry data was implemented in the software GW-chart.

As shown by the results of the sample analysis in Table 2, the average pH of the river water is 7.83, indicating neutral water, with slight variations among the different sampling sites. SAL ranges from 0.06 to 0.15 (ppt), and the average value is 0.1 (ppt). The average EC is 208.45 (μS/cm), with a range of 132.71–318.00 (μS/cm). The differences in EC and TDS values among the different sampling points are obvious, with corresponding ranges of 185.29 (μS/cm) and 125.42 (mg/L), respectively.

The average concentrations (mg/L) of the major cations exhibit the following pattern of dominance: Ca²⁺ > Na⁺ > Mg²⁺ > K⁺. The dominant cation in the Binggou River is Ca²⁺, accounting for 44%, followed by Na⁺ and Mg²⁺ at 36% and 18%, respectively. The average contents of the main anions decrease in the following order: SO₄²⁻ > HCO₃⁻ > NO₃⁻ > Cl⁻. The dominant anion is SO₄²⁻, accounting for 53%, followed by HCO₃⁻ at 24%. The chemical type of the river water in this area is Ca–Na–SO₄–HCO₃. The variation coefficients of Na⁺, Mg²⁺, Cl⁻, and NO₃⁻ are highest, with values of 0.75, 0.66, 0.69, and 0.74, respectively, while the variation coefficients of Ca²⁺ and SO₄²⁻ are lower, 0.19 and 0.34. In addition, the contents of Na⁺, Mg²⁺, SO₄²⁻, NO₃⁻, and TDS are slightly higher than the global averages for rivers (Meybeck 2003).

The physical and chemical parameters of the river water in different seasons are shown in Figure 2. Compared to spring, the contents of TDS and EC are higher in winter and lower in autumn. The pH, SAL, and Mg²⁺ are relatively stable among the three seasons. The concentrations of Ca²⁺ and HCO₃⁻ are highest in spring, whereas the concentrations of the other major ions are highest in winter, with obvious seasonal variations of Cl⁻, NO₃⁻, Na⁺, Ca²⁺, and HCO₃⁻ in the three seasons. As shown in the Piper diagram (Figure 3(a)), the variations in the anions are greater than those of the cations, with the largest change in the content of SO₄²⁻. The changes in the cations are less obvious, except that the value of Ca²⁺ is lowest in winter, whereas the other cations are highest in winter. In all three seasons, the concentrations of the cations in the river decrease in the following order: Ca²⁺ > Na⁺ > Mg²⁺ > K⁺. The concentrations of the anions vary in the different seasons and exhibit the order SO₄²⁻ > HCO₃⁻ > Cl⁻ > NO₃⁻ in spring and winter and SO₄²⁻ > HCO₃⁻ > NO₃⁻ > Cl⁻ in autumn. The type of hydrochemistry is SO₄–Ca in all three seasons.

The spatial variations in the concentrations of major ions in the Binggou River are shown in Figure 4. The values of SAL and pH are stable among all sampling sites. Other major ions and physical parameters are abnormal in the Qilian River and Qilian spring water sampling points, including extremely low Ca²⁺ and HCO₃⁻ concentrations and very high concentrations of the other ions, particularly SO₄²⁻, Mg²⁺, and Na⁺. The average values of EC, TDS, SAL, and pH are much higher in the east branch than in the west branch upstream and the main stream, with the values of 291 (mg/L), 194.5 (mg/L), 0.14 (ppt), and 8.02, respectively. The concentrations of Ca²⁺ and HCO₃⁻ are lower in the east branch than in the west branch upstream and the main stream, but all other major ions are higher in the east branch, among which SO₄²⁻, Na⁺, NO₃⁻, Cl⁻, and Mg²⁺ account for 44.5%, 60.67%, 64.1%, 59.9%, and 59.8% of the whole basin, respectively. The contents of EC, TDS, SAL, and other major ions are higher in the west branch than in the main stream, but the pH is the same as that in the main stream (7.77) (Table 3). The hydrochemical type of the east branch is SO₄–Na, and the hydrochemical types of the west branch and main stream are both SO₄–Ca (Figure 3).

The correlation analysis of the ions in the Binggou River Basin is presented in Table 4. There are significant correlations (p < 0.01, R² = 0.935) between Cl⁻ and Na⁺ in the river water, indicating that these two ions have the same source. However, the concentration ratio of the two ions (Na⁺/Cl⁻) is 3.03, suggesting substantial redundant Na⁺. Na⁺ is also highly correlated with NO₃⁻ and Mg²⁺ at the test level of p < 0.01, with R² values of 0.819 and 0.813, respectively. These results indicate that the weathering of magnesium carbonate rock contributes greatly as an ion source in the river. NO₃⁻ shows a strong positive correlation with Mg²⁺ at the test level of p < 0.01 (R² = 0.966). In addition, there is a slight correlation between HCO₃⁻ and Ca²⁺ at the test level of p < 0.05 (R² = 0.509). In this study, the concentration ratio of the two ions (Ca²⁺/HCO₃⁻) is 1.68, suggesting a contribution of sources other than rock weathering to Ca²⁺ and HCO₃⁻.

As shown in the Gibbs diagram (Figure 5), the sampling points are mainly concentrated on the left side of the graph, with a small number of points distributed to the right and beyond the range of the dotted lines. This distribution shows that rock weathering is the main source of ions in the river water, whereas evaporation and precipitation have limited influence. Granite is the main rock in the basin, primarily in the form of O-Type adakite (sodium-rich), which is the main source of Na⁺, Mg²⁺, and K⁺. The shale and conglomerates distributed in the basin are mainly carbonate rocks such as calcareous shale and limestone, which are the main sources of Ca²⁺ and HCO₃⁻. In addition, the massive sulfide deposits distributed in the basin are among the sources of SO₄²⁻ (Wang et al. 2011).

In this inland area of Northwest China, the monsoon season lags slightly, and precipitation is the main source of the supply of river water in autumn. In winter, there is very little precipitation, and the temperature is low; consequently, the groundwater supply is the main source of river water. In the spring, as the temperature rises, melt-water becomes the main source of supply, along with a small amount of precipitation (Figure 6).

The seasonal variations of the ion concentrations in the river showed that, among the three seasons, the concentrations of Cl⁻, K⁺, Na⁺, and Mg²⁺ are highest in winter. In winter, groundwater is the major source of river water, and thus, the ions in the river water are mainly carried by groundwater. Granite and sulfide deposits are widely distributed in this area, and the long-term erosion of rocks underground is the main factor affecting ion concentrations in the river. The concentrations of Ca²⁺ and HCO₃⁻ are highest in spring. In spring, the water is mainly composed of melt-water, and surface erosion by this abundant melt-water is obvious. As carbonate rocks are distributed on the surface, the weathering and erosion of these surface carbonate rocks are the main sources of Ca²⁺ and HCO₃⁻. The concentrations of NO₃⁻ and SO₄²⁻ are highest in autumn when the concentrations of other ions are generally lower. The effects of falling dust and infiltration of farmland are obvious in autumn, indicating that farmland and the atmosphere are important sources of NO₃⁻ and SO₄²⁻.

In the upper reaches of the basin, there is little difference in land cover and use types between the east and west branches, but there are significant differences in the distribution of farmland and construction land (Figure 7). As shown in Table 5, the farmland area of the east branch accounts for 75% of the total area of farmland in the whole river basin. Construction land accounts for 57% of the total construction land in the basin. These values clearly indicate that human activities are altering the underlying surface in this area. By contrast, the west branch and the main stream are less affected by human activities. People are mainly engaged in seedling cultivation, planting of Ginseng fruit, highland barley, and Chinese yam and other agricultural activities in which nitrogen fertilizer and farmyard manure are important components of the crop planting process. In addition, coal is the main fuel. These materials are important sources of SO₄²⁻, NO₃⁻, and Na⁺ in the river water of the east branch.

Studies of rivers in other alpine regions of China have found similar results, including weakly alkaline river water, Ca²⁺ and HCO₃⁻ as the major ions, obvious spatial and temporal differences in ion concentrations, and rock weathering as the main source of ions in rivers (Pu et al. 2012; Feng et al. 2014; Wu et al. 2017). However, there are some differences: the river water chemical characteristics of the Yulong Mountains are also affected by monsoons (Pu et al. 2012) and the rivers in the Tianshan Mountains are affected by evaporation and have relatively high SO₄²⁻ concentrations (Feng et al. 2014).

The rock weathering is the major ion source in the Binggou River Basin, which is consistent with other regions, but SO₄²⁻ is the major anion in the study area, while the concentration of HCO₃⁻ is relatively low. Because the study area is located at the intersection of the monsoon zone and non-monsoon zone, the climatic conditions are complex. Moreover, the geological conditions of the Qilian Mountains are unique, with differences in rock types compared with other regions and more frequent human activities. Therefore, further studies are necessary.

The drinking water quality standard issued by the World Health Organization (WHO) is adopted in this paper (Table 2; WHO 2011). The results show that the quality of the river water is far below the upper limit of the standard range, indicating that the river water is very suitable for drinking. However, Na⁺, Mg²⁺, SO₄²⁻, NO₃⁻, and TDS levels are all slightly higher than the global average for rivers (Meybeck 2003), which indicate that the suitability of river water for drinking is slightly worse than the world average. In addition, the water quality is closer to the upper limit of the drinking water standard in the east branch than in the other areas (Table 3), mainly due to the obvious human activities in the east branch.

As shown in Figure 8, the Na⁺% values of the Qilian spring and Qilian River sampling points on the east branch are within the range of 40–60 in the monitoring period, whereas the values in the other regions are less than 40. Although these results show that the water of the Binggou River meets the requirements for irrigation, the east branch of the river is approaching the critical value of 60 (Wilcox 1948). Thus, the formulation of countermeasures is urgently needed to prevent further worsening of water quality. The lower the SAR value of a river, the more suitable it is for irrigation. When the SAR value is over 18, the water quality of the river does not meet the requirements for irrigation (Saleh et al. 1999). The SAR value of the Binggou River is less than 10, but the value of the east branch is obviously higher, which is consistent with the results of Na⁺%.

The results revealed that most of the hydrochemical attributes of the Binggou River Basin show obvious spatiotemporal variations due to the distinct lithology, the seasonal differences in precipitation in the inland arid climate, and the complex interactions with anthropogenic components. The variations of major ion concentrations in the river water are higher during winter but lower in summer and higher in the upstream east branch than in the upstream west branch and, in turn, the main stream. The analysis showed that rock weathering is the main source of ions in the river water, followed by human activities such as farmland fertilization and coal burning. The water quality fully meets the standards for drinking and irrigation, but the regional differences are large, and human activities have a certain impact on river water quality.

This research was financially supported by the National Natural Science Foundation of China (41761047, 41661005, and 41261104), the National Natural Science Foundation Innovation Research Group Science Foundation of China (41421061), and the Autonomous Project of State Key Laboratory of Cryosphere Sciences (SKLCS-ZZ-2017).