Introduction

The annual global suspended flux into the oceans is estimated to be ~13.5 × 1012 kg, out of which Asian rivers constitute ~80 % of the flux and the Himalayan rivers viz Ganga and Brahmaputra contribute ~25 % (Berner and Berner 1987). Quantifying sediment flux dynamics is a fundamental goal that relates to the geomorphologic evolution of river channels, deltas and estuaries. Meade 1996 stated that flux calculated at seaward gauging stations may not exactly represent the true flux into oceans as most of the sediments could be deposited inland, and therefore, headwater streams of the major river systems can give a precise estimate. Alaknanda River is considered as a headwater stream of river Ganga. Previous studies of Abbas and Subramanian (1984), Sarin and Krishnaswami (1984), Bhatt and Saklani (1996), Singh and Hasnain (1998), Pandey et al. (2002), Bickle et al. (2003), Chakrapani (2005), Chakrapani and Saini (2009), Yadav and Chakrapani (2011) have cited Alaknanda as a major contributor of dissolved and sediment load to river Ganga. These studies have dealt with water and sediment geochemistry on seasonal basis and the impact of lithologies and glacier melt. The unique feature of the river Alaknanda is the high physical weathering rate of ~863 t/km2/year, which is five times the average global physical denudation rate and a chemical erosion rate of ~153 t/km2/year, which is six times of the global average of 24 t/km2/year (Chakrapani et al. 2009; Chakrapani and Saini 2009; Yadav and Chakrapani 2011). In recent years, owing to the large hydropower potential in the Alaknanda basin, erosion processes and fluvial transport of sediments have become a focus of revived attention. Studying sediment size and its composition is important in understanding the river processes related to hydraulics, geomorphology, flow dynamics, heavy metal pollutants, siltation in reservoirs and contamination by pathogens (Slattery and Burt 1997; Xu 1999; Pandey et al. 2002; Zhang et al. 2006; Wang et al. 2008). The major factors responsible for grain size characteristics and load vary widely including climatic conditions, lithology, discharge, relief, basin area, river energy and anthropogenic activities (Milliman and Meade 1983; Lindholm 1987; Gaillardet et al. 1999; Xu 1999; Chakrapani 2005; Goni et al. 2006; Wang et al. 2008; Feng et al. 2011; Xu et al. 2012). In the mountainous terrains, river energy is the leading factor controlling grain size and sediment load which vary both spatially and temporally depending on discharge, slope, rainfall variations and other related factors. Grain size also exerts strong control on the chemical composition of sediments leading to the preferential enrichment of specific materials in certain grain size fractions (Lindholm 1987; Xu 1999; Weltje and von Eynatten 2004). Warren et al. (2003), Walling (2005) have included the importance of fine sediment in the transport of nutrients, contaminants, pathogens and heavy metals through fluvial systems. Walling (1988), Galy and France-Lanord (2001), Walling et al. (2001, 2002), Singh (2009), Yan et al. (2012), Li et al. (2013) stated the importance of sediment source for understanding the sediment dynamics and the sediment budget of a catchment. Singh and Hasnain (1998), Galy and France-Lanord (2001), Chakrapani (2005), Singh (2009), Tripathy et al. (2010), Singh (2010) have comprehensively applied geochemical methods using major, trace, rare earth elements (REEs) and isotopes composition of river load to find out the source of Ganga River sediments and for sediment budgeting. The incompatibility of the REE makes them an excellent tracer for sediment fingerprinting. The concentration of REE varies markedly with rock type and source area. As shown by many studies, the river sediments are rich in normalized LREE than the HREE (Byrne and Kim 1990; Yang et al. 2002; Goni et al. 2006; Yang et al. 2009; Xu et al. 2012; Yan et al. 2012; Li et al. 2013). In this paper, we preliminarily evaluate the grain size of suspended and bank sediments; source of suspended and bank sediment was inferred using the REEs and major oxide composition. Since Alaknanda is the major contributor of dissolved load to river Ganga, an attempt has also been made to quantify the source of major elements using the forward modeling (Krishnaswami and Singh 1998, 2005; Moon et al. 2007; Zhu et al. 2013). Major elements of the dissolved load in rivers owe their source to precipitation, chemical weathering of rocks and sediments and anthropogenic effects in the catchment area (Gaillardet et al. 1995, 1999; Krishnaswami and Singh 1998, 2005; Moon et al. 2007; Galy and France-Lanord 1999, 2001; Zhu et al. 2013). This integrated study dealing with sediment and water aims to understand the geochemical nature of water and sediments of the Alaknanda River and its tributaries.

Study area

The study has been carried out in the Alaknanda River, a major tributary of the river Ganga, which originates from the snout of the Satopanth and the Bhagirathi Kharak glaciers, which rises from the Chaukhamba Mountain in the Garhwal Western Himalayas (Fig. 1). At Devprayag, the confluence of Alaknanda with the river Bhagirathi gives river Ganga its formal name. The Alaknanda basins occupy an area of 11.8 × 103 km2, which is much larger than the companion river Bhagirathi (Singh and Hasnain 1998; Chakrapani and Saini 2009). The basin experiences a tropical monsoon climate with almost 75 % of the rainfall occurring in the months of June–September (CWC 2007). The river flows through the Central Crystalline zone composed of medium- to high-grade metamorphic rocks such as gneisses, migmatites, crystalline schist and quartzites near its source and Lesser Himalayas in its lower reaches draining through carbonates, massive quartzites, slates, phyllites and graywackes. The tributaries of river Alaknanda also flow through a varied lithology comprising of quartzites, limestone, shale and slates (Valdiya 1980; Sarin et al. 1992).

Fig. 1
figure 1

Sample location sites on the Alaknanda River and its tributaries

Full size image

Methodology

In total, seventeen sediment samples were collected from the Alaknanda River and its tributaries including two samples from river Ganga during October 2012. The sample location sites are shown in Fig. 1. Approximately 5 l water samples were collected and filtered through 0.45-µm cellulose nitrate membrane filter papers to separate the suspended sediments. Bank sediments were collected from a depth of 10–15 cm using a stainless steel spatula and stored in clean polythene bags. Particle size analyzer (model Microtrac S3500) was used to analyze the grain size of both suspended and bank sediments. The statistical parameters of grain size distribution were determined by following Folk and Ward (1957). Total carbon (TC) was estimated by using TOC analyzer without decarbonating the sediments. River energy was calculated using the approach followed by Bagnold (1960, 1977). Elemental composition of sediments (Chakrapani 2007, unpublished) was used to correlate the variability of chemical composition and grain size, and provenance of sediments. In the absence of chemical composition data for the present samples, we preferred to use previously unpublished data, in view of the fact that significant variation in the composition of sediment generally does not alter annually. Mass-balancing approach (Gaillardet et al. 1995, 1999; Krishnaswami and Singh 1998, 2005; Galy and France-Lanord 2001; Singh et al. 2006) was used to find out the different pools contributing to dissolved load. Based on mass-balancing equations and forward modeling, contribution from the atmosphere, rock weathering and anthropogenic sources were estimated (Moon et al. 2007; Zhu et al. 2013). The data of major ions (published in Chakrapani et al. 2009) were used to perform forward modeling. For atmospheric correction, the data of rain water composition of Rudraprayag (Chakrapani et al. 2009) was used.

Results and discussion

Suspended sediments

Sediment load

There are many variables such as climate, tectonics and anthropogenic disturbance that may significantly control the spatial and temporal variation of sediment load in rivers. Figure 2 shows the spatial variation in total suspended sediments (TSS) from Birahi to Rishikesh during the month of October 2012. The TSS is observed to increase downstream of location A8 which is due to change in lithology (Lesser Himalayas comprise of carbonates, phyllites and quartzites, whereas upper part of the basin has high-grade metamorphic rocks) and due to construction activities and urbanization in the lower part of the Alaknanda basin.

Fig. 2
figure 2

Spatial variation in total suspended sediments (TSS)

Full size image

To better understand the relation between the different variables and sediment load (SL), we used the temporal data of total suspended sediments (TSS) of Srinagar and Devprayag town. The rainfall is the dominant factor controlling the sediment load as shown in Fig. 3a. With an increase in rainfall, an increase in the discharge and specific stream power (SSP) is observed (Fig. 3b, c). The increase in rainfall thus contributes in increasing the SL (Fig. 3d, e). In this part of Himalaya, where there is high monsoon rain during June to September; rainfall is the most leading factor controlling the river power and thus the sediment erosion and yield. Chakrapani et al. (1995); Chakrapani (2005) studied the mineralogical composition of the Alaknanda, Bhagirathi and Ganga at Devprayag. The dominance of quartz and mica/illites in the suspended sediment composition indicates the high intensity of physical weathering in the present study area. The erosion rate is basically very high during the monsoon months due to increase in mass wasting events and increase in discharge which are directly related to the intensity of rainfall.

Fig. 3
figure 3

a Simultaneous increase in rainfall and TSS total suspended sediments, b Relationship of rainfall with Q discharge, c Relationship of rainfall with SSP specific stream power, d Relationship of rainfall with TSS total suspended sediments, e Relationship of rainfall with SL sediment load

Full size image

Grain size specifications of suspended sediments

The grain size distribution depends on the lithological composition, weathering intensity, climate, basin physiography and river energy (Gaillardet et al. 1999; Xu 1999; Pandey et al. 2002; Zhang et al. 2006; Chakrapani and Saini 2009). The statistical method developed by Folk and Ward (1957) was applied to interpret the particle size characteristics (Table 1 and Fig. 4a–d). The mean size (M 1) varies from 8.79 to 56.34 µm (4.17–7.01 Φ). There is no significant pattern in the grain size distribution of upstream and downstream samples (Fig. 4a). Mostly coarse to fine silt size fraction is present in the waters of Alaknanda River and its tributaries. T3 shows the smallest mean size of 8.79 µm. The smallest grain size can be related to the inability of the river to carry bigger grains. The dam construction activities across Alaknanda River and its tributaries seem to reduce the coarse size fractions by sedimentation due to sluggish flow of river. Sorting (S 1) varies from 1.23 to 3.16 Φ, and the sediments are poorly sorted.

Table 1 Statistical parameters of suspended and bank sediment size
Full size table
Fig. 4
figure 4

Bivariate plots showing granulometric characteristics of suspended sediments a mean size variation, b mean size versus standard deviation (sorting), c mean size versus skewness, d mean size versus kurtosis

Full size image

This behavior of grains can be attributed to the short distance of transport from source, textural immaturity and the variable source of the sediments such as glacial origin, mass wasting and different lithology along the river course (Pandey et al. 2002). In the dammed river, change in the river energy can also lead to the poorly sorted nature of sediment. Skewness (Sk1) quantifies asymmetry of distribution and varies from −0.41 to 0.23. Most of the sediments belong to fine to symmetrically skewed category. Negatively skewed sediments are present in tributaries T1, T2 and T3 and at A3; negative skewness indicates the dominance of coarse-grained sediments. High gradient, high river energy, glacial contribution and lesser distance of transport can be the reason of coarse sediments in these tributaries. Most of the samples are mesokurtic ranging from 0.94 to 1.07. Figure 4b–d explains the bivariate relationship between mean size (M 1) and sorting (S 1), skewness (Sk1) and kurtosis (Kg1). No good correlation was established between M 1, S 1 and Kg1.

Provenance of suspended sediments

To find out the source of suspended sediments, rare earth elements (REEs) data of Chakrapani 2007 (Unpublished) have been used (Table 2 and Fig. 1). REEs are considered to be a good provenance tracer because they behave conservatively during sediment formation and within different grain sizes also, they resemble their source rock REE pattern (Taylor and McLennan 1985; Goldstein and Jacobson 1988; Singh and Rajamani 2001; Xu et al. 2012; Liu et al. 2014). Normalization to the NASC (Normalized American Shale Composites) is a widely accepted method of comparing the REE contents of various geological materials (Haskin et al. 1968; Taylor and McLennan 1985). Similar to most world rivers, Alaknanda River also shows the enrichment of LREE because of high adsorption of LREE onto clay particles and lower content of heavy minerals on suspended sediments (Aubert et al. 2001; Singh and Rajamani 2001; Yang et al. 2002; Shouye et al. 2003; Li et al. 2013); the ratio of (La/Yb)diss/(La/Yb)susp for Alaknanda, Bhagirathi and Ganga rivers at Devprayag is 0.13, 0.35 and 0.26, respectively, which shows HREEs go into solution more easily. HREEs are preferentially transported in solution as they form more soluble bicarbonate and organic complexes than the LREE (Balashov et al. 1964; Byrne and Kim 1990; Yang et al. 2002). LREE/HREE and La/Yb ratio shows that samples are highly enriched in LREE (Table 2; Fig. 5), pointing to the association with high silicate fraction, dominantly due to enhanced physical weathering. La/Yb and Gd/Yb ratio shows that fractionation in the streams is less due to faster rate of transportation. The suspended sediment shows significant negative Eu anomalies (Eu/Eu*), indicating highly differentiated felsic source. The positive Eu/Eu* is only observed at site A3 (river Alaknanda at Birahi). Under the reducing condition, Eu3+ can be reduced to soluble Eu2+. The Deoban formation is enriched in limestone and dolomite, and Eu2+ are substituted for Ca2+, resulting in the positive Eu/Eu* in the sediments of Birahi Ganga. Thus, it can be inferred that positive Eu/Eu* at A3 is because of confluence of Birahi Ganga with Alaknanda. Bhandari et al. (1992) in the Spiti Valley section at Lalung observed positive Eu/Eu* values in the limonitic deposits. There is no consistent pattern in the Eu/Eu* from upstream to downstream regions (varies from 0.18 to 3.23), maybe because of joining of tributaries with the Alaknanda River and the change in lithology. The negative Eu/Eu* of the sediments shows their derivation from felsic source. Similar to the Alaknanda River, suspended sediments of Bhagirathi River (BS1) are also found to be derived from the felsic source. Bhagirathi also flows through the same lithology as Alaknanda River. Ramesh et al. (2000) observed negative Eu anomalies in most of the sediments from the Ganga–Brahmaputra river systems. Thus, based on REE indices, Central Crystalline rocks rich in Gneisses, granites and schist are considered to be a probable source of suspended sediments for the Alaknanda River and Ganga River at Devprayag.

Table 2 REE concentration in the suspended sediments (µg/g) (Source: Chakrapani 2007 unpublished)
Full size table
Fig. 5
figure 5

Normalized REE pattern in the suspended sediments of Alaknanda, Bhagirathi and Ganga River

Full size image

Bank sediments

Grain size specifications of bank sediments

Large variations are displayed in the grain sizes of the bank sediments from Birahi to Rishikesh. The distribution of bank sediment size ranges from 23.7 to 563.2 µm and is mostly coarse silt to medium sand as shown in Table 1 and Fig. 6a. Coarse silt is only found at A5 (sample from Kalyasaur) owing to the low flow condition due to hampering of river for construction of hydropower project at Srinagar. The larger grain size is at A1 (Chamoli). Results show that all the sediments are moderately sorted (Fig. 6b) because of dominance of fine-medium sand carried by the river. Due to presence of finer fraction, fine to symmetrically skewed sediments are found. Most of the samples are mesokurtic, but at the upstream site, A1 sediment is platykurtic which may be because of channel morphology or fluctuation in glacial contribution. No good correlation was established between the M 2 and S 2, Sk2 and Kg2 as moderately sorted and fine to nearly symmetrically skewed sediment behavior exist for a wide range of mean size values as shown in Fig. 6b–d.

Fig. 6
figure 6

Bivariate plots showing granulometric characteristics of bank sediments a mean size variation, b mean size versus standard deviation (sorting), c mean size versus skewness d mean size versus kurtosis

Full size image

Relation between bank sediment size and chemical composition

Table 3 and Fig. 7 show the major oxide composition of the bank sediments. Si and Al are the dominant elements accounting for more than 90 % of the bank sediment composition. SiO2 is very resistant to weathering, and Al2O3 is generally very less mobile and locked in aluminosilicate rocks. Na2O and K2O concentrations vary 0.94–2.23 and 1.86–2.95 wt%, respectively, whereas CaO and MgO have a concentration of 1.78–8.09 and 1.81–7.16 wt%, respectively. Not such strong correlation was found between the major oxides and mean grain size (M 2) (Fig. 8a–g), which indicates the chemical immaturity of these sediments (Pandey et al. 2002; Dalai et al. 2004; Singh 2010). Figure 8h also shows a poor correlation between the total carbon (TC) and M 2.

Table 3 Major oxide composition (wt%) of bank sediments (Source: Chakrapani 2007, unpublished)
Full size table
Fig. 7
figure 7

Spatial variation of major oxide (wt%) in bank sediments

Full size image
Fig. 8
figure 8

Correlation of mean grain size with a–g major oxide composition of bank sediments, h TC total carbon

Full size image

Chemical index of alternation

Feldspars are the most abundant of the reactive (labile) minerals; chemical weathering involves the degradation of feldspars and concomitant formation of clay minerals. Calcium, sodium and potassium are generally removed from the feldspars by aggressive soil solutions, resulting in increased proportion of alumina to alkalis in the weathered product. A good approach to quantify the extent of chemical weathering is through the calculation of chemical index of alteration, CIA (Nesbitt and Young 1982), defined as:

$${\text{CIA}} = \left[ {{\text{Al}}_{2} {\text{O}}_{3} /\left( {{\text{Al}}_{2} {\text{O}}_{3} + {\text{Na}}_{2} {\text{O}} + {\text{K}}_{2} {\text{O}} + {\text{CaO}}^{*} } \right)} \right] \, \times 100$$

The CIA values are calculated using the molar concentrations of the oxides, where CaO* being the concentration derived only from silicate fraction of the rock. Following McLennan (1993), a correction is made for the carbonate and apatite content. The resultant value is a measure of the proportion of Al2O3 versus the labile oxides in the analyzed sample. As weathering progresses, the more labile cations such as Na, Ca and Mg would be released from the source rocks, thereby depleting their concentrations in the residual material. In contrast, the less mobile elements such as Al, Fe and Ti concentrations would increase in the residue. Thus, with increase in silicate weathering, the ratio would increase, ultimately reaching the limiting value of 100. The changes in CIA reflect change in proportion of feldspar and the various clay minerals in the sediments. Table 3 gives the concentrations and CIA values of various oxides in bank sediments at different locations. Figure 9 is a bar diagram of CIA value of sediments analyzed in this study with the reference line that of the un-weathered granites/granodiorites.

Fig. 9
figure 9

Bar diagram of CIA values of sediments in Alaknanda, Pinder, Bhagirathi and Ganga Rivers

Full size image

Provenance of bank sediments using major oxides

The composition of bank sediments of the Alaknanda River shows a positive correlation between the SiO2 and Na2O (R 2 = 0.46); Al2O3 and K2O (R 2 = 0.92); and Al2O3 and TiO2 (R 2 = 0.96), which suggests their association with phyllosilicates and clay size fraction of sediment controlling them. The negative correlation of CaO and MgO with Al2O3 shows their derivation from carbonate rocks (Table 4). Geochemically, SiO2/Al2O3 shows the dominance of silica and immature nature of the bank sediments. K2O/Na2O ratio is often used to signify the provenance, type of feldspars present in the source rock and chemical maturity (Gupta et al. 2012; Das and Haake 2003; Das et al. 2008; Wu et al. 2012). In the Alaknanda catchment area, K2O/Na2O > 1 with moderate SiO2/Al2O3 suggests weathering from K-feldspar enriched terrain (Gupta et al. 2012; Verma et al. 2012). Central Crystallines are enriched in felsic rocks such as schist, gneiss and granite (Valdiya 1980), and therefore, these sediments have their source from them. Thus, Central Crystallines are the main supplier of both the suspended (refer ‘Provenance of suspended sediments’ section) and bank sediments in the Upper Ganga Basin. The result of Galy and France-Lanord (2001), Dalai et al. (2002, 2004) also shows the Central Crystalline as a dominant supplier of sediments to the Himalayan river system.

Table 4 Correlation matrix showing major oxides of bank sediments
Full size table

Forward modeling and dissolved load

Rivers carry the imprints of their compositional source. At Devprayag, river Ganga confirms a dissolved load of 2.74 × 106 tons of which about 1.80 × 106 ton is delivered by the river Alaknanda (Chakrapani et al. 2009). Thus, Alaknanda River is the major supplier of dissolved load to the river Ganga. In this segment, we try to find out the major controller of dissolved load. Major elements (X) in the dissolved load of rivers owe their supply from precipitation, chemical weathering of rocks and anthropogenic sources.

$$\left[ {\text{X}} \right]_{\text{river}} = \left[ {\text{X}} \right]_{\text{atmosphere}} + \left[ {\text{X}} \right]_{\text{carbonate}} + \left[ {\text{X}} \right]_{\text{silicate}} + \left[ {\text{X}} \right]_{\text{evaporate}} + \left[ {\text{X}} \right]_{\text{anthropogeny}}$$

Krishnaswami and Singh (1998), Galy and France-Lanord (1999) and Dalai et al. (2002) have used this mass-balancing approach to evaluate the source of dissolved load. Using the forward model, mentioned by Moon et al. (2007) and Zhu et al. (2013), percentage contribution by each reservoir is determined. The total cations (Tz+ = Na + K+Ca + Mg) from each reservoir (in µm) is calculated by the approach mentioned in Fig. 10 and then converted to the percentage value resulting in quantification of the dissolved source. The values of major ion composition presented  in Chakrapani et al. (2009) were used to find out the share of each reservoir. The anthropogenic influence on dissolved load is cited as nil since this part of the Himalaya is sparsely populated, having a least population density of 49 persons/km2 in Chamoli District and maximum density of 170 persons/km2 in the Tehri-Garhwal District (Census 2011 by Govt. of India).

Fig. 10
figure 10

Framework for the forward model depicting the four reservoirs of dissolved load carried by the River (adapted from Zhu et al. 2013)

Full size image

Precipitation input

Following the approach mentioned by Krishnaswami and Singh (1998); Galy and France-lanord (1999); Moon et al. (2007); Rai et al. (2010); Zhu et al. (2013) for atmospheric correction, the lowest Cl− is supposed to have derived entirely from the rain water. The Cl− concentration was found to be lowest in Pandukeshwar sample. The following equation was used to obtain the concentration of cations derived from rainfall: \({\text{X}}_{\text{atm}}^{ + } = \left( {{\text{X}}_{\text{rain}}^{ + } /{\text{Cl}}_{\text{rain}}^{ - } } \right) \times {\text{Cl}}_{\text{rain}}^{ - }\). The result shows that rainwater contribution to the dissolved load varies from 1.0 to 5.6 %. The highest influence of rainwater was found in the Alaknanda water at Manna during the pre-monsoon month (Table 5).

Table 5 Different reservoirs contributing to dissolved load in the Alaknanda basin (values in %)
Full size table

Lithological input

Rock weathering is the major reservoir for the dissolved load. Atmospheric corrected solute concentration was used to find out the proportion of dissolved cations derived from the evaporite, silicate and carbonate rocks.

Silicate weathering

Na+ is the main ion derived from the rainfall, halite and silicate rock weathering. Since halite beds are absent in the study area, the main contributors are rainfall and silicate rocks. To determine the share of silicate weathering in the Na+ flux of the river, we use the following equation: \({\text{Na}}_{\text{sil}}^{ + } = {\text{Na}}_{\text{river}}^{ + } {-}{\text{Na}}_{\text{rain}}^{ + }\). In the same way, K+ is also considered to be derived only from silicate and rain water as carbonates and evaporites have a minimum amount of K+. \({\text{Ca}}_{\text{sil}}^{2 + }\) and \({\text{Mg}}_{\text{sil}}^{2 + }\) are estimated using the silicate bed rock data mentioned in Krishnaswami and Singh (1998); Dalai et al. (2002), showing the molar ratios of Ca/Na = 0.7 and Mg/Na = 0.3. \(\% {\text{Tz}}_{\text{sil}}^{ + }\) was determined using the following equation (mentioned in Krishnaswami and Singh 1998).

$$\% {\text{Tz}}_{\text{sil}}^{ + } = \left( {{\text{Na}}_{\text{sil}}^{ + } + {\text{K}}_{\text{sil}}^{ + } + {\text{Ca}}_{\text{sil}}^{2 + } + {\text{Mg}}_{\text{sil}}^{2 + } } \right)/\left( {{\text{Na}}_{\text{riv}}^{ + } + {\text{K}}_{\text{riv}}^{ + } + {\text{Ca}}_{\text{riv}}^{2 + } + {\text{Mg}}_{\text{riv}}^{2 + } } \right) \times 100$$

The results show that average \({\text{Tz}}_{\text{sil}}^{ + }\) is around 33.5 %. Among the tributaries, Mandakini shows the highest \({\text{Tz}}_{\text{sil}}^{ + }\) value and Pindar shows the lowest \({\text{Tz}}_{\text{sil}}^{ + }\) values.

Carbonate weathering

The Ca2+ and Mg2+ content in the river water was supposed to be derived from rainfall, evaporites (anhydrites and gypsum), silicate and carbonate rocks. The Ca2+ and Mg2+ concentrations were determined by these equations:

$$\begin{aligned} {\text{Ca}}_{\text{carb}} = {\text{Ca}}_{\text{river}} - {\text{Ca}}_{\text{rain}} - {\text{Ca}}_{\text{sil}} - \left( {{\text{SO}}_{{ 4 {\text{river}}}} - {\text{SO}}_{{ 4 {\text{rain}}}} } \right) \hfill \\ {\text{Mg}}_{\text{carb}} = {\text{Mg}}_{\text{river}} - {\text{Mg}}_{\text{rain}} - {\text{Mg}}_{\text{sil}} \hfill \\ \end{aligned}$$

Carbonate content was found to be somewhat higher during the post-monsoon (December 2004) month than the pre-monsoon month (April 2005); the reason can be low discharge during the December (due to lower glacial contribution), resulting in concentration of calcium in the water. The total percentage of cations derived from carbonate weathering varies from 22 to 63 %. The \({\text{Tz}}_{\text{carbonate}}^{ + }\) was found to be highest in the Pindar River water at Karnaprayag and lowest in the Alaknanda River water at Pandukeshwar.

Evaporite weathering

As evaporitic weathering does not involve minerals containing K+ and Mg2+ and since no halite beds are present in the Alaknanda basin, the concentration of \({\text{K}}_{\text{eva}}^{ + }\), \({\text{Na}}_{\text{eva}}^{ + }\) and \({\text{Mg}}_{\text{eva}}^{2 + }\) is considered to be nil. Evaluating the contribution from evaporites is generally difficult because \({\text{SO}}_{4}^{2 - }\) present in the dissolved load can have multiple sources such as dissolution of gypsum, oxidation of pyrite, rainout of natural biogenic emissions, cyclic salts, pollution and volcanism (Berner and Berner 1996; Galy and France-Lanord 1999; Dalai et al. 2002; Singh et al. 2006; Mehto and Chakrapani 2013). In the Garhwal Himalayas, the pyrite beds and deposits of gypsum in the Lesser Himalayan formations are considered to be the probable source of SO4 in the river water (Dalai et al. 2002; Chakrapani and Veizer 2006; Chakrapani et al. 2009).

$$\begin{aligned} 2{\text{FeS}}_{2} + 15{\text{O}}_{2} + 8{\text{H}}_{2} {\text{O}} \to 2{\text{Fe}}_{2} {\text{O}}_{3} + 2{\text{H}}^{ + } + {\text{SO}}_{4}^{2 - } \hfill \\ {\text{CaSO}}_{4} \cdot 2{\text{H}}_{2} {\text{O}} \to {\text{Ca}}^{2 + } + {\text{SO}}_{4}^{2 - } \hfill \\ \end{aligned}$$

The exact estimation of the fraction of SO4 coming from pyrite beds is difficult to calculate because the oxidation of pyrite generates sulfuric acid that helps in weathering of surrounding silicate and carbonate rocks. In this process, CO2 is not consumed, and it may lead to overestimating CO2 consumed by silicate weathering. Based on major oxides, it is difficult to distinguish the share of gypsum dissolution and pyrite oxidation. Using the stable isotope approach, Chakrapani and Veizer (2006) suggested that the \({\text{SO}}_{4}^{ - }\) in the water of Alaknanda River and its tributaries (Saraswati, Dhauliganga, Birahiganga, Nandakini, Pinder, Mandakini) owes its origin from both the oxidation of pyrite and gypsum dissolution. However, based on the values of δ34S, it was concluded that dissolved \({\text{SO}}_{4}^{2 - }\) is majorly coming from the oxidation of pyrite beds disseminated in the Lesser Himalayas. Using the forward model, combined share of evaporite (gypsum + pyrite) weathering to the dissolved load in the Alaknanda basin is found to be varying from 4.4 to 40.3 %.

Forward modeling results

The results show that carbonate weathering followed by silicate weathering are the dominant processes controlling the dissolved load composition of the Alaknanda River (Table 5). The relative contribution of rainfall, evaporite, silicate and carbonate weathering to dissolve load is 1.9, 19.7, 33.5 and 44.8 %, respectively. The predominant carbonate lithology and the ease of carbonate dissolution over silicate weathering can be the explanation of the dominance of carbonate weathering influencing the dissolved load composition. The results are synonymous with the finding of the previous literature (Singh and Hasnain 1998; Chakrapani and Saini 2009; Chakrapani et al. 2009) that points Lesser Himalayas enriched in carbonate rocks to be the major controller of dissolved load composition.

Conclusions

Recent anthropogenic intrusion in the Alaknanda basin necessitates an integrated study to understand the nature of river Alaknanda. River discharge, lithology, climate and stream power are the major controlling factor that influence the sediment load. The spatial variation observed in both the suspended and bank sediment sizes can be attributed to lithology, slope, discharge, river power and land use. Results show both natural and anthropogenic factors influencing the grain size and sediment load, specially the construction of dams and reservoirs. The grain size in suspended and bank sediments varies from 8.79 to 56.34 and 23.73 to 563.24 µm, respectively. Suspended sediments were found to be poorly sorted, whereas bank sediments were moderately sorted. The correlation between the major oxides and mean grain size of bank sediment indicates the chemical immaturity of the sediments. Since river carries the imprint of their compositional source, REE and major oxide composition were used for sediment fingerprinting. Results show the dominance of Central Crystallines as a major source of sediment. The dissolved load composition has least influence of anthropogenic control in the study area. The results of forward modeling attribute the principal role of carbonate and silicate weathering processes in controlling the dissolved load composition. The individual contribution to the dissolved load as calculated by forward model is carbonate > silicate > evaporite > rain.