Characteristics of stable isotope and hydrochemistry of surface water, and their significance on the Huaibei plain, China

Surface water systems like rivers, lakes and swamps are important in the human environment, serving as major stores of water resources. Many studies have examined the hydrochemical characteristics and evaluated water quality, based on major and trace element, and stable isotope determinations in surface waters (Bertrand et al. 2014; Li et al. 2014; Martinez et al. 2015). Such studies provide supporting information for efficient surface water development schemes and help to reduce water quality deterioration. Isotope technology development since the 1950s and 1960s has led to its successful use in research into the water cycle. The main applications of hydrogen and oxygen isotope determination in water research are to follow the water cycle, determine water sources, and the components of surface- and ground- water flow, and separate water sources (West et al. 2014; Shen et al. 2015).

Since the 1980s, droughts have occurred frequently, causing serious water shortages, in North China. Human activities, runoff generation and recharge conditions on the plain have also changed substantially. The Huaibei Plain, in northern Anhui province, is the main component of the North China Plain, which has significant coal resources. The region is at the southern end of the warm temperate zone, with an average annual temperature of 14 or 15 °C. The main crops are wheat, corn, soybean and sorghum, as well as cotton and peanuts. The depth to which coal is exploited on the Huaibei Plain also shows up the water resource problem. So, studies providing hydrochemical information on the surface waters are needed. To date, studies focused on the major ion and isotopic contents of the surface waters have not been undertaken on the Plain.

The objectives of this study were to discuss the hydrochemical characteristics of the major elements, and hydrogen and oxygen stable isotopes, by sampling the river- and ground- water of the plain, determining the major element and isotope concentrations, to reveal the hydrochemical distribution characteristics and their influencing factors.

According to data collected during the period 2005 to 2009, mean annual precipitation on the plain is between about 600 and 1,400 mm. During a typical year, about 70% of rainfall occurs from June to September, known as the rainy season. Most of the Plain is covered with Quaternary strata, with a small amount of exposed bedrock in the northeast. The basement of Huaibei Plain comprises Archean and Proterozoic metamorphic rocks, covered by late-Proterozoic to Permian sediments. Groundwater resources in the region have been abundant historically, the shallow groundwater providing the main irrigation and domestic supplies for rural residents of the Plain.

To reveal the hydrochemical character of the river water, eighteen representative samples were collected of river water and two of shallow groundwater (Figure 1). The latter were collected from artesian wells, approximately 30 m deep, and the sampling sites are shown in Figure 1. To ensure the accuracy of the determinations, every fifth sample was taken in duplicate. The samples were put into 2.5 L plastic barrels, which had been rinsed three times with the water, and the sampling location (longitude and latitude), temperature, pH, conductivity, and TDS were all recorded on site. All surface water samples were analyzed for the major ions, and stable hydrogen and oxygen isotope concentrations, while the shallow groundwater samples were tested only for the hydrogen and oxygen isotope concentrations. The major ions were analyzed in the Department of Coal Geology's Anhui Province testing center, China. K⁺ + Na⁺ were analyzed by atomic absorption spectrometry, SO₄²⁻ and Cl⁻ by ion chromatography, Ca²⁺ and Mg²⁺ by EDTA titration, and alkalinity by acid-base titration. It is noted that K and Na were determined together, as if a single species (Ussing 1959). The isotopic compositions were determined in the National Engineering Research Center of Coal Mine Water Hazard Control Laboratory and are reported relative to Standard Mean Ocean Water, and the analytical precisions are ±0.5‰ and 0.1‰ for δD and δ¹⁸O.

The results from analyses of the river water samples are presented in Table 1. The river water samples were mainly Na-SO₄ and Na-HCO₃ types, with a few Na-Cl and Ca-HCO₃, indicating that Na⁺ + K⁺, SO₄²⁻, and HCO₃⁻ are the dominant ions in the river water.

Previous studies have shown that multivariate, statistical analyses–e.g., the correlation and principal component analyses (PCA)–were good ways to reveal the complex relationships between variables, as well as their roles (Yidana et al. 2008; Tiri et al. 2014). Both correlation and PCA analyses were used in this study, and the results are presented in Tables 2 and 3.

It can be seen in Table 2 that there is a certain degree of positive correlation between Ca²⁺ and Mg²⁺, with a correlation matrix of 0.55. This is thought to indicate dolomite dissolution. No correlation was found between any of the anions reported. Positive correlations exist between Na⁺ + K⁺ and SO₄²⁻, and Na⁺ + K⁺ and Cl⁻, with correlation matrices of 0.88 and 0.64, respectively, indicating that these ions came from the same source. In addition, HCO₃⁻ and Ca²⁺ showed negative correlations with D and ¹⁸O, with correlation matrices of −0.66, −0.66, −0.57 and −0.60, respectively. This suggests that the hydrogen and oxygen isotope fractionation in the water samples could be influenced by carbonate-type minerals.

The hydrochemical composition of river water is controlled mainly by precipitation, evaporation, and water-rock interactions, as well as by human activity (Song et al. 2006). The PCA was conducted to obtain detailed statistical information. The rotated PCA loadings are presented in Table 3, where it can be seen that there are three principal components, with initial eigenvalues of 5.06, 2.36, and 1.84, respectively. More than 84.22% of cumulative variance was found, and parameters like TDS, Ca²⁺, SO₄²⁻, Na⁺ + K⁺, Cl⁻ and Mg²⁺ all have high values in PC1, showing that this arises mainly from ionic sources controlled by precipitation. PC2 accounts for 21.48% of the total variance with an eigenvalue of 2.36. Combined with the high loading values for TDS, D, and ¹⁸O, this suggests that PC2 could arise from evaporation. Evaporation usually leads to enrichment of deuterium and heavy oxygen isotopes in river water, as the lighter isotopes volatilize. The pH values and CO₃²⁻ have high values in PC3, indicating carbonate dissolution and acidification, which also affects the proportions of D and ¹⁸O in river water.

Other data–e.g., the global meteoric water line (GMWL), local meteoric water line (LMWL) and local surface water line (LSWL)–were required to obtain more information about the heavy isotope characteristics of the Huaibei Plain river water samples. The GMWL is characterized as δD = 8*δ¹⁸O + 10.56 (Craig 1961); the LMWL as δD = 7.9*δ¹⁸O + 8.2, summarized from stable isotope data (Zhang 1989), and the LSWL as δD = 6.74*δ¹⁸O − 3.33 (Gui et al. 2005). All these lines, and the isotopic signatures of the river and ground- water samples are shown in Figure 3. The river water samples all plotted below the LSWL, LMWL, and GMWL, suggesting that the source of the river water was rainfall with varying degrees of evaporation. The two shallow groundwater samples plot on or near the LSWL.

There is a strong positive correlation between δD and δ¹⁸O, with a correlation matrix of 0.99, described by the equation δD = 5.32*δ¹⁸O − 16.54. Compared to the GMWL and LMWL, the value of the equation slope (5.32) is lower than that obtained from the two previous equations, indicating that hydrogen and oxygen isotope fractionation could be occurring in the surface waters. However, the slope value is similar to that of previous data from the Yellow River (Su et al. 2003; Gao et al. 2011); the slopes are 5.29 and 4.71, respectively. Thus, the hydrogen and oxygen isotopic ratios in river waters from the Huaibei Plain are controlled by similar factors to those on the Yellow River. Previous studies have shown that the proportions of δD and δ¹⁸O in river water are controlled mainly by evaporation, so that the river water data are described by the local evaporation line, especially in the dry season. Because of this, the equation δD = 5.32*δ¹⁸O − 16.54 describes both the local evaporation line and the relationship between δD and δ¹⁸O.

TDS reflects the total concentration of major ions in water, indicating the total salinity. In general, TDS increases gradually along the river's flow path. As would be expected, the concentration of TDS in riverine headwaters is less than that downstream. EC, which also reflects water-rock interaction, etc., has similar characteristics.

Comparison of TDS with δD and δ¹⁸O concentrations shows that the stable isotope–δD and δ¹⁸O–concentrations in the river water generally decrease as TDS increases. In other words, a negative correlation exists between TDS, and δD and δ¹⁸O; its correlation matrix is −0.4 (Table 2). The concentrations of δD and δ¹⁸O in river water are not controlled only by evaporation. Previous studies (Chen et al. 2008) have shown that isotope exchange reactions between water and surrounding rocks could lead to light isotope enrichment. The concentration of HCO₃⁻ also showed negative correlation with δD and δ¹⁸O (Table 2).

Eighteen river- and two shallow ground- water samples were collected on the Huaibei Plain, China, the hydrochemistry and isotopic characteristics were studied, and statistical analyses carried out:

All waters sampled were alkaline, with relatively high TDS contents. Na⁺ + K⁺, determined as a single species, are the dominant cations, while the dominant anions are SO₄²⁻ and HCO₃⁻. The river waters are mainly Na·K-SO₄ and Na·K-HCO₃ types, with a very few Na·K-Cl and Ca-HCO₃. EC and TDS in the headwaters are lower than downstream, and variations in their levels reflect the rivers’ flow paths.

Three Principal Components, representing three influencing factors controlling the ionic sources, were found, accounting for more than 84% of the cumulative variance. PC1 appears to relate to precipitation, as it concerns mainly ionic sources. PC2 could be arise from evaporation, with principal contributions from TDS, and the stable hydrogen and oxygen isotopes. PC3 has high contributions from pH and CO₃²⁻, indicating carbonate dissolution.

The river water samples have high concentrations of δ¹⁸O and δD compared to the shallow groundwaters in the Huaibei Plain, arising from evaporation. The correlation between the proportions of D and ¹⁸O in the river water can be described by the equation δD = 5.32*δ¹⁸O − 16.54, which is that of the local evaporation line.

The relatively high TDS concentrations and low ionic ratios of TDS-(Na⁺ + K⁺)/(Na⁺ + K⁺ + Ca²⁺) and TDS-Cl⁻/(Cl⁻ + HCO₃⁻) in the river waters suggest that the ionic content of the water samples arises from both rock weathering and evaporation. The negative correlations between TDS and HCO₃⁻ with both δD and δ¹⁸O indicate that carbonate dissolution has also had affected the hydrogen and oxygen isotope characteristics.

The study was supported by the National Nature Science Foundation of China (41373095), the Key projects of University young talents in Anhui Province (gxyqZD2016346) and the Postdoctoral research project in Suzhou University (2015BHB01).