Characteristics of Ions Composition and Chemical Weathering of Tributary in the Three Gorges Reservoir Region: The Perspective of Stratified Water Sample from Xiaojiang River

Abstract

River water chemistry offers information on watershed weathering and responds to the global carbon cycle. Watershed weathering processes and water chemistry in stratified water are still unclear in Xiaojiang River, as a major tributary of the Three Gorges Reservoir (TGR) which is the largest reservoir in the world. Major ions of river water at different depths were measured to reveal the ionic composition and chemical weathering properties by principal component analysis and stoichiometry in Xiaojiang River. Ca²⁺−HCO₃⁻ dominated the hydrochemical facies of river. Surface river water had the lowest total dissolved solid (146 mg/L) compared to other layers of water. According to principal component analysis, the major ions were divided into two principal components. PC1 was the weathering end-member of rocks, including the main ions except K⁺ and NO₃–N, and PC2 may be the mixed end-member of atmospheric input and anthropogenic input. From stoichiometry, carbonate weathering dominated the cationic composition, with a contribution ratio of 56.7%, whereas atmospheric input (15.2%) and silicates weathering (13.9%) had similar extent of contribution. Compared with other major tributaries of TGR, Xiaojiang had more intense chemical weathering processes. The weathering rates of carbonates and silicates were 19.33 ± 0.68 ton/km²/year and 3.56 ± 0.58 ton/km²/year, respectively. Sulfuric acid as a proton may have participated less in the weathering processes of Xiaojiang River. The CO₂ consumption budgets for silicates and carbonates weathering were 0.8 ± 0.2 × 10⁹ mol/year and 2.8 ± 0.2 × 10⁹ mol/year, respectively. These results enrich the watershed weathering information of TGR tributaries and provide data support for understanding the global carbon cycle.

1. Introduction

Rivers are an important part of the Earth’s critical zone because they connect the terrestrial and oceanic ecosystems [1]. Because of a close relationship with the global carbon cycle, the migration and transformation of terrigenous material from rivers to oceans have attracted extensive attention [2,3,4]. Chemical weathering of rocks is a key source of river solutes and affects CO₂ levels throughout geological time [5,6,7]. CO₂ from the atmosphere and soil respiration takes part in weathering processes and enters rivers as dissolved inorganic carbon (DIC) [8,9]. Therefore, the chemical mass balance of dissolved and particulate phases can provide critical information about the carbon storage potential of the river basin. According to recent research, in addition to atmospheric input and local lithology, human activities (e.g., agricultural irrigation and damming) can influence ionic composition and interfere with chemical weathering processes [10,11,12,13], further affecting the global carbon cycle [14,15,16]. Hence, describing the geochemical processes of river solutes at local and regional scales is critical to the carbon cycle and long-term global climate change.

The Three Gorges Reservoir (TGR) is located in Chongqing and Hubei province in southwest China, regulating 56% of Asia’s largest river (the Yangtze River) [17]. Because the TGR is the largest reservoir in the world, and more than half of the basin is covered by carbonates, it is suitable for quantifying the impacts of dams on stream geochemistry and carbon cycle [18,19,20]. The reservoir impoundment management can change the residence time, flow direction, and level of river water, and further affect the processes of water–rock interaction [21,22]. The major ions composition of river water contains important information about chemical weathering [23,24,25], water supply [26,27], and human activities [28,29,30], all of which are essential for understanding the damming influences on the material cycle and climate change in the basin [13,31]. Previous studies have shown that the major ionic composition of TGR’s main channel was dominated by rock weathering processes [32,33,34], and the CO₂ consumption of silicates and carbonates in TGR account for about 1% of the world’s CO₂ consumption of silicates and carbonates, respectively [29,35]. Some tributaries in TGR also have high weathering rates [36,37,38], e.g., Chetelat et al. [29] reported that the cationic carbonate weathering rate of Wujiang and Jialingjiang were about 10 ton/km²/year and 6 ton/km²/year, respectively, which were close to the cationic carbonate weathering rate of the Yangtze River (10 ton/km²/year) at TGD. These results indicate that the carbon sink capacity of tributaries cannot be ignored. Hence, research into hydrochemistry and weathering processes of tributaries in the TGR requires more attention.

Xiaojiang River, as the largest tributary in the middle of the TGR, has attracted extensive attention due to water quality problems [39,40]. However, a description of the major ions composition and chemical weathering processes in the Xiaojiang River is still lacking. Previous studies have shown that the hydrological process and geochemical cycle of elements in Xiaojiang River were affected by the storage regulation resulting from the construction of the Three Gorges Dam (TGD), including hydrodynamic characteristics, nutrient dynamics, and CO₂ flux [21,41,42]. These studies also showed that river water at different depths had different geochemical condition. Therefore, the major ions composition of stratified river water may contain additional information about chemical weathering processes, and this has not been described in previous studies. Hence, this study investigated the ionic composition of river water at different depths of Xiaojiang River, combined with principal component analysis and chemometrics, with the goal of revealing (1) the content and spatial distribution of major ions in stratified river water, (2) source apportionment of ions, and (3) chemical weathering rates and CO₂ consumption in the basin. This study fills the gap in the understanding of weathering in the Xiaojiang River and provides data support for understanding the impacts of the damming on global climate change.

2. Study Area and Methods

2.1. Location of Study Area

The Xiaojiang River straddles the longitude of 107°44′50″–108°53′56″ E and the latitude of 30°39′35″–31°41′19″ N. It is the largest tributary in the north bank of the middle TGR, and the total area of the basin is 5225 km². The Xiaojiang River basin belongs to the subtropical humid monsoon climate zone, with an average annual temperature of 18.2 °C [43]. The annual precipitation and discharge are 1300 mm and 127 m³/s, respectively, and the flood season (from May to September) accounts for about 75% of the annual discharge [44]. The Xiaojiang River is 182 km long and flows from north to south into the main stream (the Yangtze River) of the TGR. Mainly carbonate and clastic rocks constitute the geology of the Xiaojiang River basin, including limestone, sandstone, mudstone, siltstone interlayered with shale, and coal seams [45]. Due to the geological background and human activities (slope tillage and urbanization), soil erosion is a serious problem in the basin [46].

2.2. Sample Collection and Treatment

As shown in Figure 1, in August 2020, water samples were collected from six sites in the downstream of the Xiaojiang River. Global Positioning System was applied to measure the longitude and latitude of each site. Water samples were collected at three depths in the center of the river channel: surface (50 cm depth), middle (1/2 water depth), and bottom. Then, all samples were filtered through the cellulose acetate membranes on the day of sampling (0.22 μm), and they were stored in high-density polyethylene bottles. Highly purified nitric acid was dropped into the samples used for cationic test to keep the pH < 2. All parameters of samples were measured within one week after being filtered.

2.3. Measurement of Parameters

Total dissolved solids (TDS), electrical conductivity (EC), and temperature measurements were performed in the field at the time of sampling with a portable multiparameter instrument (YSI Inc., Yellow Springs, OH, USA). PH values were measured with a pH meter (Rex., Shanghai, China). Ion chromatograph (Dionex ICS-600, Thermo Fisher Scientific Inc., Sunnyvale, CA, USA) and inductively coupled plasma optical emission spectrometer (Optima 5300DV, PerkinElmer Inc., Waltham, MA, USA) were used to determine the concentrations of major anions (F⁻, NO₃–N, Cl⁻, and SO₄²⁻) and cations (Na⁺, K⁺, Mg²⁺, and Ca²⁺), respectively. During measurement, blanks, duplicate samples, and national standard materials (GBW08606 and GBW082051, National Research Center for Certified Reference Materials, China) were used to correct the test to ensure the accuracy of the data (the deviation was <5% throughout the analysis). The total alkalinity of water samples was determined by titration with hydrochloric acid. According to Hunt et al. [47], the content of HCO₃⁻ can be calculated by linear relationship between pH and alkalinity. The calculation formula is as follows:

wHCO₃⁻ = −0.5 × pH + 3.79

(1)

2.4. Measurement of Contribution Ratio

Due to changes in the water level caused by dam regulation or the passage of flood peaks of main channel, water from the main stream may recharge into its tributaries [27]. The area where the tributaries are affected by the main stream is called the confluence zone [26]. In this confluence zone, the major ions composition of the tributary is not only affected by upstream inflow, but also contributed to by the main stream. Therefore, the contribution ratio refers to the proportion of discharge from main channel to the confluence zone and can be quantified by the concentration of main ions in conservative behavior. The formula is [26]

δ_M = xδ_T + (1 − x)δ_U

(2)

where x is the contribution ratio of discharge from TGR main channel; 1 − x is the contribution ratio of the discharge from upper reach in Xiaojiang River. δ_T, δ_U, and δ_M represent ionic concentration of the TGR main channel, the upper reach, and confluence zone of Xiaojiang River, respectively.

2.5. Measurement of Chemical Weathering Rate and CO₂ Consumption

Cationic weathering rate (ton/km²/year) was used to evaluate the degree of rock weathering in Xiaojiang River basin [48]. The calculation formulas were as follows:

Φ_sili = [Na⁺]_sili + [K⁺]_sili + [Ca²⁺]_sili + [Mg²⁺]_sili

(3)

Φ_carb = [Ca²⁺]_carb + [Mg²⁺]_carb

(4)

where sili and carb are silicate weathering and carbonate weathering, respectively. The contribution concentrations of carbonate and silicate weathering to cations were calculated based on the forward model.

According to Chetelat et al. [29], the content of [SO₄²⁻] (Φ_SO4) that is involved in rock weathering should theoretically be the difference between [HCO₃⁻] and the sum of cations derived from silicate and carbonate weathering, and Φ_SO4 should follow the formulas below:

[SO₄²⁻]_sili = 0.5[Na⁺]_sili + 0.5[K⁺]_sili + [Ca²⁺]_sili + [Mg²⁺]_sili

(5)

[SO₄²⁻]_carb = 0.5[Ca²⁺]_carb + 0.5[Mg²⁺]_carb

(6)

If the Φ_SO4 was assumed to be entirely derived from silicate weathering, the cations derived from silicate weathering and CO₂ consumption were calculated as follows:

$$[{\mathrm{Na}}^{+}{{]}^{\mathrm{ssw}}}_{\mathrm{sili}}={\mathsf{\Phi }}_{\mathrm{SO}4} \times [(\frac{[{\mathrm{Ca}}^{2+}]}{[{\mathrm{Na}}^{+}]}{)}_{\mathrm{sili}}+(\frac{[{\mathrm{Mg}}^{2+}]}{[{\mathrm{Na}}^{+}]}{)}_{\mathrm{sili}}+(\frac{[{\mathrm{K}}^{+}]}{[{\mathrm{Na}}^{+}]}{)}_{\mathrm{sili}}+0.5{]}^{-1}$$

(7)

$$[{\mathrm{K}}^{+}{{]}^{\mathrm{ssw}}}_{\mathrm{sili}}=[{\mathrm{Na}}^{+}{{]}^{\mathrm{ssw}}}_{\mathrm{sili}} \times (\frac{[{\mathrm{K}}^{+}]}{[{\mathrm{Na}}^{+}]}{)}_{\mathrm{sili}}$$

(8)

$$[{\mathrm{Ca}}^{2+}{{]}^{\mathrm{ssw}}}_{\mathrm{sili}}=[{\mathrm{Na}}^{+}{{]}^{\mathrm{ssw}}}_{\mathrm{sili}}\times (\frac{[{\mathrm{Ca}}^{2+}]}{[{\mathrm{Na}}^{+}]}{)}_{\mathrm{sili}}$$

(9)

$$[{\mathrm{Mg}}^{2+}{{]}^{\mathrm{ssw}}}_{\mathrm{sili}}=[{\mathrm{Na}}^{+}{{]}^{\mathrm{ssw}}}_{\mathrm{sili}}\times (\frac{[{\mathrm{Mg}}^{2+}]}{[{\mathrm{Na}}^{+}]}{)}_{\mathrm{sili}}$$

(10)

X^csw_sili = X_sili − X^ssw_sili

(11)

CO_{2 sili} = 2[Ca²⁺]^csw_sili + 2[Mg²⁺]^csw_sili + [Na⁺]^csw_sili + [K⁺]^csw_sili

(12)

Similarly, if Φ_SO4 was totally involved in carbonate weathering, then

$$[{\mathrm{Ca}}^{2+}{{]}^{\mathrm{scw}}}_{\mathrm{carb}}={{{\mathsf{\Phi }}_{\mathrm{SO}}}_4}_{ }\times [0.5(\frac{[{\mathrm{Mg}}^{2+}]}{[{\mathrm{Ca}}^{2+}]}{)}_{\mathrm{carb}}+0.5{]}^{-1}$$

(13)

$$[{\mathrm{Mg}}^{2+}{{]}^{\mathrm{scw}}}_{\mathrm{carb}}=[{\mathrm{Ca}}^{2+}{{]}^{\mathrm{scw}}}_{\mathrm{carb}}\times (\frac{[{\mathrm{Mg}}^{2+}]}{[{\mathrm{Ca}}^{2+}]}{)}_{\mathrm{carb}}$$

(14)

X^ccw_carb = X_carb − X^scw_carb

(15)

CO_{2 carb} = [Ca²⁺]^ccw_carb + [Mg²⁺]^ccw_carb

(16)

where SSW and SCW represent silicate and carbonate weathering with sulfuric acid participation, and CSW and CCW represent silicate and carbonate weathering with CO₂ participation, respectively. X is the cationic concentration that is contributed from weathering of carbonates or silicates.

2.6. Statistical and Spatial Analysis

ArcGIS 10.4 GIS (ESRI^®) software (Environmental Systems Research Institute; Redlands, CA, USA) was used to show the spatial distribution of sampling points and geological background. A Piper diagram was generated using the “three-line diagram” function of the Origin 2018 software. The vertical concentration distribution of river major ions was shown using the function of “Cubic Spline Interpolation” by Surfer 15 software. Principal component analysis (PCA) was performed by SPSS 25.0. The software Origin 2018 (Origin Lab., Hampton, MA, USA) and Adobe Illustrator 2018 (Adobe Inc., San Jose, CA, USA) were applied to make figures.

3. Results and Discussion

3.1. Hydrochemical Characteristics of Xiaojiang River

Table 1. The hydrochemical parameters in Xiaojiang River (n = 18).

Parameters	Detection Limit	Range	Mean	±SD	WHO a
[F−] (mmol/L)	0.03	0.01–0.03	0.01	0.01	0.08
[Cl−] (mmol/L)	0.04	0.08–0.65	0.28	0.23	7.05
[NO3–N] (mmol/L)	0.06	0.00–0.09	0.01	0.02	0.18
[SO42−] (mmol/L)	0.10	0.12–0.47	0.26	0.14	3.12
[HCO3−] (mmol/L)	-	1.81–2.51	2.21	0.25	-
[Na+] (mmol/L)	0.03	0.15–0.72	0.37	0.23	-
[K+] (mmol/L)	0.01	0.06–0.08	0.07	0.01	-
[Ca2+] (mmol/L)	0.04	0.82–1.17	1.00	0.12	-
[Mg2+] (mmol/L)	0.01	0.16–0.41	0.27	0.09	-
TDS b (mg/L)	1	118–2 02	156	20	1000
EC c (μs/cm)	1	211–360	277	20	-
pH	0.1	7.50–7.95	7.75	1.8	6.5–8.5

^a WHO representative guidelines for drinking water quality, published by World Health Organization in 2017. -: No values. ^b TDS: Total dissolved solids. ^c EC: Electrical conductivity.

and Table S1 displays information on the hydrochemical parameters of water samples. During the sampling period, the pH values of samples were between 7.50 and 7.95, indicating a weakly alkaline river water. The values of TDS and EC gradually increased along the river. The average TDS value (156 mg/L) was lower than the main channel of TGR (292 mg/L) reported by Wang et al. [32] and some other major tributaries of TGR, such as Daning River (250 mg/L) [27], Jialingjiang River (261 mg/L) [29], and Wujiang River (355 mg/L) [49]. These results may be related to the difference in chemical weathering process and the degree of anthropogenic input. As for major ions, the absolute values of normalized inorganic charge balance of most samples were ≤10%, suggesting that the ions in samples were basically balanced, and the data were reliable. HCO₃⁻ and Ca²⁺ concentrations dominated the anions and cations in river water, respectively. Water quality is closely linked to environmental safety and human health [50,51,52]. The urbanization of the Xiaojiang River basin has reached 39.4%, with a total population of nearly 300,000 [39]. Thus, the quality of drinking water needs attention persistently. The concentrations of major ions in the Xiaojiang River were substantially lower than those reported by the WHO, indicating that the river poses a low health risk to humans.

The hydrochemical types of water in Xiaojiang Rivers was Ca²⁺−HCO₃⁻ (Figure 2). Concentration of HCO₃⁻ dominated the anions, followed by Cl⁻ + SO₄²⁻. The order of cations was Ca²⁺ > Na⁺ + K⁺ > Mg²⁺. The ionic features of the Xiaojiang River were found to be similar to the Jialingjiang River and Wujiang River [29,53]. The solute variations of the Xiaojiang River were also consistent with the Yangtze River reported in other studies. Compared to other big rivers in China, the contents of ions (Cl⁻ + SO₄²⁻ and Na⁺ + K⁺) that related to the dissolution process of evaporites in Xiaojiang River were lower than that of the Jiulongjiang River [54] and Yellow River [55], while the concentration of ions (Ca²⁺ and HCO₃⁻) that related to carbonate weathering were higher. These results were consistent with the geological background of rivers, where evaporites are mostly exposed in Jiulongjiang River basin, and carbonates are predominant in Xiaojiang River. Overall, similar to other rivers, the ionic composition of river water in Xiaojiang River should be mainly controlled by rock weathering during the wet season.

Figure 3 shows the vertical and transverse distribution of major ions in river water. F⁻ is mainly affected by rainfall and coal burning in factories [29,56]. The range of F⁻ in Xiaojiang River was similar to the main stream of TGR (0.06–0.13 mmol/L) reported by Chetelat et al. [29]. They suggested that this range of F⁻ concentration was almost unaffected by human activities and could represent the natural background value. NO₃–N was mainly derived from agricultural fertilization and industrial wastewater [54]. The maximum value of NO₃–N in this study was close to the mean value (0.04 mmol/L) of TGR rainfall reported in a recent study [57]. The above results indicate that there was no risk of fluoride and nitrate contamination in the river water of the study area during the wet season.

As for other major ions, samples closer to the Xiaojiang River mouth have higher ionic concentrations, and the concentrations of ions in most surface waters were lower than those in middle and bottom waters [58,59]. Because the river level rises as the flood crest passes through during the wet season, the river water in the main channel can flow back to the tributary and leads to a mixing process of the ions, which may eventually affect the ionic composition of the river water [60,61,62]. Therefore, the proportion of discharge from the main stream to tributaries can reflect the ionic contribution of the main stream to tributary. Previous studies have shown that major ions (such as Cl⁻, Na⁺, Ca²⁺ and Mg²⁺) have conservative behaviors during migration in water [26,27,63], and can be used to trace the proportion of discharge from the main stream to the tributaries. In this study, Cl⁻ was selected to quantify the contribution ratios of the discharge from the main channel of the TGR. The Cl⁻ concentrations at sample S1 and S6 were assumed to be the contribution of the upper reach in Xiaojiang River and the concentration of main channel of TGR, respectively. As calculated by formula 2, the proportion of discharge from the main channel of the TGR to surface, middle, and bottom water in the confluence zone of Xiaojiang River were 46%, 39%, and 43%, respectively. As tributaries close to TGD, the proportion of discharge from the TGR main stream in the confluence zone of Daning River (80.3%) and Xiangxi River (76%) [27] were about twice that of Xiaojiang River, indicating that the confluence zone of tributaries that close to the TGD should be more susceptible to the influence of main stream inflow.

3.2. Source Apportionment of Riverine Ions

3.2.1. PCA Analysis

PCA analysis can help classify ions of similar origin [64]. As shown in Figure 4 and

Table 2. Results of PCA analysis.

Variable	PC1	PC2
[HCO3−]	0.91	0.205
[Cl−]	0.987	−0.055
[NO3–N]	0.27	−0.74
[SO42−]	0.997	−0.049
[Na+]	0.993	−0.012
[K+]	0.92	−0.032
[Mg2+]	0.995	−0.015
[Ca2+]	0.953	0.117
Eigenvalues	6.607	1.247
Variance (%)	73.406	13.885
Cumulative (%)	73.406	87.261

, two principal components were extracted. The variances of PC1 and PC2 were 73.4% and 13.9%, respectively.

PC1 has a positive loading on HCO₃⁻, Cl⁻, SO₄²⁻, Na⁺, K⁺, Mg²⁺, and Ca²⁺; PC2 has a positive loading on F⁻ and negative loading on NO₃–N. Although the source analysis based on these two components was limited, the major ions related to rock weathering were positive loading with PC1. For example, the Ca²⁺, Mg²⁺, and HCO₃⁻ concentration will increase significantly during carbonate weathering [29]. Na⁺, K⁺, and Cl⁻ are mainly derived from the weathering of evaporites and silicates [12]. Thus, PC1 should represent the end-member of the rock weathering. As discussed earlier, F⁻ in the river is mainly from atmospheric input, and NO₃–N is related to human activities (such as agricultural irrigation and industrial pollution). Therefore, PC2 should be a mixed end-member.

3.2.2. Stoichiometry of Weathering Processes

The ions in river water usually come from chemical weathering, anthropogenic, and atmospheric input [65,66,67]. As discussed previously, the major ions in Xiaojiang River should be controlled mainly by chemical weathering and be little affected by anthropogenic input. Carbonates, evaporites, and silicates are mainly distributed in the Xiaojiang River basin. The possible weathering processes are as follows [48]:

2Ca_xMg_{(1 − x)}CO₃ + H₂SO₄→2xCa²⁺ + 2(1 − x)Mg²⁺ + 2HCO₃⁻ + SO₄²⁻

(17)

Ca_xMg_{(1 − x)}CO₃ + H₂O + CO₂→xCa²⁺ + (1 − x)Mg²⁺ + 2HCO₃⁻

(18)

2Na_xK_{(1 − x)}AlSi₃O₈ + H₂SO₄→2xNa⁺ + 2(1 − x)K⁺ + SO₄²⁻ + 6SiO₂ + 2AlOOH

(19)

2Na_xK_{(1 − x)}AlSi₃O₈ + H₂O + 2CO₂→2xNa⁺ + 2(1 − x)K⁺ + 2HCO₃⁻ + 6SiO₂ + 2AlOOH

(20)

CaAl₂Si₂O₈ + H₂SO₄→Ca²⁺ + SO₄²⁻ + 2SiO₂ + 2AlOOH

(21)

CaAl₂Si₂O₈ + H₂CO₃→Ca²⁺+2HCO₃⁻ + 2SiO₂ + 2AlOOH

(22)

The sodium normalization method can be used to roughly estimate the relative contribution of evaporites, carbonates, and silicates to the composition of the major ions [29]. According to Figure 5a, water samples of Xiaojiang River were distributed between three end-members, which were similar to other tributaries of the TGR. Ca²⁺/Na⁺ mole ratios ranged from 1 to 10, and the mole ratios of HCO₃⁻/Na⁺ were between 3 and 15. The upstream samples were closer to the carbonate end-member, and the samples near the mouth of Xiaojiang River were closer to the end-members of silicates and evaporites. These results indicated that the major ions composition of the stratified water in Xiaojiang river was mainly derived from carbonate and silicate weathering, and evaporites may contribute to the major ions in the mouth of Xiaojiang River.

Chemical weathering of rocks is known to influence the global carbon cycle and climate change [68,69]. Both CO₂ and H₂SO₄ participate in the weathering of carbonates and silicates [70,71]. According to stoichiometric equations, the relationship between [SO₄²⁻]/[HCO₃⁻] and [Ca²⁺ + Mg²⁺]/[HCO₃⁻] can help clarify the involvement of carbonic acid and sulfuric acid in rock weathering. If [SO₄²⁻]/[HCO₃⁻] is 1 and [Ca²⁺ + Mg²⁺]/[HCO₃⁻] is 1 or 0, only CO₂ is involved in mineral dissolution. If [Ca²⁺ + Mg²⁺]/[HCO₃⁻] is 2 and [SO₄²⁻]/[HCO₃⁻] is 1, the minerals are dissolved by sulfuric acid. The ranges of [SO₄²⁻]/[HCO₃⁻] and [Ca²⁺ + Mg²⁺]/[HCO₃⁻] of Xiaojiang River were 0.1–0.5 and 1.0–1.5, respectively, indicating that the riverine ions originated from the minerals dissolved by both carbonic acid and sulfuric acid. In addition, the ionic ratios of samples were relatively far away from the alkali feldspar end-member, suggesting that the contribution of alkali feldspar was limited.

3.2.3. Contribution Proportion of Weathering Sources

A forward model was used to further clarify the proportion of ionic contribution from different sources. First, the sources of all ions were classified. Usually, Ca²⁺ and Mg²⁺ are mainly controlled by silicate and carbonate weathering [72]. Na⁺, SO₄²⁻, and Cl⁻ are linked to the dissolution of evaporites [12]. According to the previous discussion, the influence of human input associated with agriculture and industrial wastewater on the river can be ignored. The potential sources of the major ions in the TGR were expressed as follows:

[Cl⁻]_evaporites = [Na⁺]_evaporites

(23)

[SO₄²⁻]_evaporites = [Ca²⁺]_evaporites

(24)

[Cl⁻]_river = [Cl⁻]_atmosphere + [Cl⁻]_evaporites

(25)

[K⁺]_river = [K⁺]_atmosphere + [K⁺]_silicates

(26)

[SO₄²⁻]_river = [SO₄²⁻]_atmosphere + [SO₄²⁻]_evaporites + [SO₄²⁻]_pyrites

(27)

[Na⁺]_river = [Na⁺]_atmosphere + [Na⁺]_evaporites + [Na⁺]_silicates

(28)

[Ca²⁺]_river = [Ca²⁺]_atmosphere + [Ca²⁺]_evaporites + [Ca²⁺]_silicates + [Ca²⁺]_carbonates

(29)

[Mg²⁺]_river = [Mg²⁺]_atmosphere + [Mg²⁺]_evaporites + [Mg²⁺]_silicates + [Mg²⁺]_carbonates

(30)

As mentioned in the previous discussion, the SO₄²⁻ concentration in the Xiaojiang River far exceeds the world average for large rivers. Atmospheric input and weathering of gypsum and pyrite are the main sources of SO₄²⁻ in river water [29]. Based on the geological background, gypsum exposure should be limited in the Xiaojiang River basin. Therefore, in addition to the atmospheric input, the remaining SO₄²⁻ of Xiaojiang River was probably derived from pyrite weathering. Then, the lowest concentration of Cl⁻ in water samples was estimated to be the Cl⁻ values of atmospheric input, and ionic ratio of each source was referenced to previous studies (

Table 3. Ionic ratios in the precipitation and silicates.

	K+/Cl−	Na+/Cl−	SO42−/Cl−	Ca2+/Cl−	Mg2+/Cl−	Ca2+/Na+	Mg2+/Na+	References
Rain	0.55	0.59	-	1.74	0.56	-	-	Wu and Han [57]
	-	-	2.06	-	-	-	-	Zhang et al. [73]
Silicates	-	-	-	-	-	0.35	0.24	Gaillardet et al. [48]

).

Contributions of different sources to cations in Xiaojiang River water are shown in Figure 6. Carbonate weathering dominated the cationic composition of river water, accounting for about 60% of the total contribution, followed by atmospheric input and silicate dissolution. Ionic contribution rates of evaporites in the reach of Huangshi–Shuangjiang (S4, S5, and S6, Figure 1) were significantly higher (about 30%), and the proportions of middle and bottom were about twice that of surface. As discussed before, Xiaojiang River can possibly be fed by the main stream of TGR. If there are no evaporites distributed in this reach, and the contributions of evaporites are all from the discharge from the main stream of the TGR, the proportion of discharge from the main stream to surface, middle, and bottom water in Xiaojiang River should be about 18%, 36%, and 37%, respectively. However, this result contradicts the previously estimated contribution that the main stream recharges the Xiaojiang River. Thus, the mouth of Xiaojiang River should have evaporites outcrops.

3.3. Protons of Chemical Weathering Agent and CO₂ Consumption

Details on chemical weathering rates in the Xiaojiang River basin are presented in

Table 4. Chemical weathering rate and CO₂ consumption of small rivers.

Sample	Silicates				Carbonates				Evaporites	Total Rock Weathering
	Φsil (ton/km2/ year)	Catsil a (105 ton/year)	CO2 ssw b (109 mol/year)	CO2 csw c (109 mol/year)	Φcarb (ton/km2/ year)	TDScarb d (105 ton/year)	CO2 scw c (109 mol/year)	CO2 ccw b (109 mol/year)	TDSevap e (105 ton/year)	TDStotal f (105 ton/year)
S1	3.43 ± 0.18	0.18 ± 0.01	0.70 ± 0.05	0.86 ± 0.04	18.44 ± 1.88	2.75 ± 0.24	2.46 ± 0.33	2.57 ± 0.27	0.01 ± 0.01	2.93 ± 0.25
S2	2.95 ± 0.24	0.15 ± 0.01	0.74 ± 0.05	0.74 ± 0.05	20.25 ± 0.13	3.06 ± 0.03	2.83 ± 0.02	2.83 ± 0.02	0.03 ± 0.01	3.25 ± 0.03
S3	2.95 ± 0.06	0.15 ± 0.01	0.68 ± 0.07	0.72 ± 0.01	16.97 ± 0.37	2.55 ± 0.04	2.38 ± 0.12	2.42 ± 0.06	0.01 ± 0.01	2.72 ± 0.03
S4	4.35 ± 0.12	0.23 ± 0.01	1.06 ± 0.04	1.06 ± 0.04	19.93 ± 0.58	3.21 ± 0.13	2.92 ± 0.15	2.92 ± 0.15	1.27 ± 0.51	4.71 ± 0.64
S5	4.05 ± 0.52	0.21 ± 0.03	0.98 ± 0.15	0.98 ± 0.15	20.04 ± 0.59	3.25 ± 0.11	2.97 ± 0.15	2.97 ± 0.15	1.68 ± 0.79	5.14 ± 0.85
S6	3.65 ± 0.58	0.19 ± 0.03	0.87 ± 0.17	0.87 ± 0.17	20.32 ± 0.68	3.25 ± 0.09	3.02 ± 0.18	3.02 ± 0.18	1.65 ± 0.71	5.10 ± 0.77

The standard deviation of each value was the difference between surface, middle, and bottom water. ^a Cations flux of silicates weathering was calculated as the product of Φ_sil and area of Xiaojiang River basin. ^b Budgets of CO₂ consumption under the assumptions of sulfuric acid completely participated in the weathering of silicates. ^c Budgets of CO₂ consumption based on the assumptions of sulfuric acid completely participated in the carbonates weathering. ^d Flux of carbonates weathering derived from the sum of Ca²⁺ + Mg²⁺ based on forward model and their stochiometric equivalent of HCO₃⁻. ^e The TDS values were determined as the sum of ions that estimated by forward model. ^f The values of TDS were calculated as the sum of weathering of silicates, carbonates, and evaporites.

. The total rock weathering rate of Xiaojiang River was 27.93 ± 4.68 × 10⁵ ton/year. Currently, weathering rates in two tributaries (Jialingjiang River and Wujiang River) of the TGR have been reported [29,37], and their drainage areas are more than 10 times that of Xiaojiang River. The cationic weathering rates of silicates and carbonates in Xiaojiang River (3.56 ± 0.58 ton/km²/year, 19.33 ± 0.68 ton/km²/year) were higher than Jialingjiang River (1.2 ton/km²/year, 6.2 ton/km²/year) and Wujiang River (0.69 ton/km²/year, 10 ton/km²/year), suggesting a more active weathering process. These results may be due to high elevation in the Xiaojiang River basin. Higher terrains are often accompanied by steep slopes that can accelerate rock erosion [74]. The rate of rock weathering increased from upstream to downstream, with a range from 2.72 ± 0.03 × 10⁵ ton/year to 5.14 ± 0.85 × 10⁵ ton/year. The weathering rate of evaporites in the reach of Huangshi–Shuangjiang was about 10 times higher than other sites, indicating more intense evaporite weathering processes occur in the mouth of Xiaojiang River.

H₂SO₄ can replace CO₂ to participate the weathering process. For example, as a large tributary of the TGR, weathering processes involving sulfuric acid provide about 50% of the weathering products in the Wujiang River basin [37]. If the effect of H₂SO₄ is ignored, the CO₂ consumption of rock weathering is possible to overestimate [70]. Therefore, H₂SO₄ should be considered when evaluating the CO₂ consumed by rock weathering. In this study, about 27 μmol/L of H₂SO₄ were involved in rock weathering in the Xiaojiang River, and the contribution rate of weathering was less than 1%, showing a lower effect on weathering processes. In the weathering process involving sulfuric acid, the CO₂ consumption of silicates and carbonates in Xiaojiang river were 0.84 ± 0.17 × 10⁹ mol/year and 2.76 ± 0.18 × 10⁹ mol/year, respectively, and account for only about 1% of the CO₂ consumption in the TGR basin reported by Chetelat et al. [29].

4. Conclusions

Based on the major ions concentration of the stratified water of Xiaojiang River in the wet season, the piper diagram was used to analyze the hydrochemical type, and the methods of principal component analysis, ionic mole ratios, and forward model were used to quantify ionic sources. Finally, the chemical weathering rate and CO₂ consumption of the basin were estimated, and the following conclusions were drawn:

(1)

The hydrochemical type of Xiaojiang River at different depths was consistent (Ca²⁺−HCO₃⁻ type), and the concentrations of Ca²⁺ and HCO₃⁻ dominated the total cations and anions of stratified water, respectively. This ionic characteristic was similar to that of other rivers in the TGR during wet season.

(2)

The ionic composition of river water was mainly controlled by rock weathering, and carbonate weathering had the highest ionic contribution (about 60%). Evaporites provided about 35% of the ionic contribution to the reach near the mouth. In addition, the ionic contribution of anthropogenic input to stratified water was not obvious.

(3)

Compared with other major tributaries of TGR, Xiaojiang River had a high rock weathering rate. The average weathering rate of carbonates in Xiaojiang River was 19.33 ± 0.68 ton/km²/year, which was twice higher than the Jialingjiang River and Wujiang River. However, weathering processes in the Xiaojiang River had a limited ability to consume CO₂, which accounted for only about 1% of the CO₂ consumption in the TGR basin.

Overall, this study supplements the gaps in hydrochemistry and chemical weathering of stratified water in Xiaojiang River and helps to understand the geochemical characteristics of tributaries under TGR regulation. It also provides a data source for research on the damming impact on global climate change. In order to realize the sustainable development, periodic monitoring of the hydrochemistry of the Xiaojiang River is required in future studies.