Keywords

1 Introduction

The mighty Himalayas politically shares a considerable area of north India, Nepal, and Bhutan and forms an arc-shaped series of high mountains, showing the highest peak (Mount Everest—8848.86 m a.s.l.) of the world and several peaks that have elevation above 7000 m above sea level [a.s.l.] (e.g. Kanchenjunga, Nanga Parbat, Nanda Devi, Makalu etc.) (Fig. 1). A large number of geological-geomorphic processes (both endogenetic and exogenetic) led to shape and reshape the landscape of the Himalayas, what we see presently. Created by continent–continent collision, the Himalayas is one of the best classic examples of the orogenic system (Dewey and Bruke 1973). To understand the geomorphic processes and diversity in the Himalayas, it is essential to look back in time and understand the continental framework that led to the rise of such massive and spectacular high ranges.

Fig. 1
figure 1

a DEM is showing variation in elevation along the Himalayas and surrounding regions with major drainage networks. b Spatial distribution of rainfall along the Himalayas (Source https://www.worldclim.org/)

Full size image

About 180 Ma ago, the Indian continent started breaking from the Gondwana land and drifted northwards at rates varying from ~4 to 20 cm/yr (Seeber et al. 1981; Patriat and Achache 1984; Pichon et al. 1992). The relatively fast, inexorable drift of the South Asian continent then nearly 50 Ma ago collided with the Eurasian plate and resulted in uplifting of huge compressed seafloor (the Tethys) and formation of an extraordinary gigantic range of fold mountains named the Himalayas that stands above all other places in the world (Patriat and Achache 1984). The abundant distribution of fossils of ancient sea creatures, for example, ammonites on the Himalayan sedimentary rocks indicates that the region once was beneath the ocean.

Based on the range’s orientation and structure, the Himalayas can be divided into four major broad physiographic divisions (Fig. 2). The southern section comprises an Indo-Gangetic plain made of alluvium brought by the Himalayan Rivers (Singh et al. 2015). The low, frontal sub-Himalayas form the foothill region. These foothills have an average height of about 900–1500 m (Whitehouse 1990). North of the sub-Himalayas, the hill ranges within the lesser Himalayas show elevation between 700 and 2500 m. The higher Himalayas forms the highest relative relief (~6000 m) with several peaks above 8000 m. The Tibetan marginal range forms a 6000–7000 m high Tibetan plateau north of the higher Himalayas.

Fig. 2
figure 2

Generalized broad physiographic divisions and variation in precipitation in the Himalayas (Alaknanda catchment) from south (a) to north (b). See Fig. 1. For the swath line. IG: Indo-Gangetic plain, SH: Sub-Himalayas, LH: Lesser Himalayas, HH: Higher Himalayas, TP: Tibetan Plateau

Full size image

The Indian Summer Monsoon (ISM) modulates the process of erosion and its rates along the southern margin of the Himalayan topographic front (Bookhagen and Burbank 2006). The rainy season in the Himalayas starts from early June and continues till September; therefore, this orographic rainfall has a significant role in the erosional mechanism (Fasullo and Webster 2003). The topographic and rainfall distribution was poorly understood until the work of Bookhagen and Burbank (2006). They showed the relationship between rainfall and topographic (elevation) distribution across the Himalayas through a number of swath profiles. The monsoons, while moving northward, experience the first significant topographic barrier due to the Lesser Himalayas, where intense orographic rainfall occurs. Once the cloud crosses Lesser Himalayas, rainfall intensity becomes less until it once again reaches the Higher Himalayas and the second phase of rainfall occurs in front of this massive topographic front. The Higher and Tethyan Himalayas receives the majority of its precipitation from the Western Disturbances responsible for the winter snow accumulation (Rees and Collins 2006; Dimri et al. 2015).

Landscapes of the Himalayas are mainly formed by the collision of continental blocks that uplifted the landmass; and the intense and variable summer monsoon precipitation shaped the modern topography in Lesser and Outer Himalayas (Abrahami et al. 2016; Sinha and Sinha 2016). The Himalayas has become a natural laboratory to understand the complex relationship between tectonic uplift and climate on erosion for the past several decades. While a large number of studies only focus on the role of uplift (e.g. Kothyari and Luirei 2016; Bhakuni et al. 2017; Sharma et al. 2018), several studies indicate a combination of climate and tectonics to play a crucial role in differential erosion rates (e.g. Vence et al. 2003; Thide et al. 2005) through geological time. Since the antecedent river response is simultaneous to both tectonic and climatic variability across the cross-sections of the Himalayas, they are ideal locales for detailed studies.

Bhagirathi and Alaknanda are two major rivers that flow across the Himalayas covering the major section of Lesser and Outer Himalayas and forming the mega river Ganga in the foothills. In the past few decades, these two rivers and the areas within the catchment have become the interests of geomorphologists to understand the complex geomorphic processes (e.g. Valdiya 1988; Tyagi et al. 2009; Ray and Srivastava 2010; Rana et al. 2016). The geomorphic processes in this region are genetically related to structural and tectonic complexities within Kumaun Lesser Himalayas as described in classical literature (Valdiya 1988). Nonetheless, the influence of climate in the formation of river terraces and stratigraphic development is very well described (e.g., Ray and Srivastava 2010). Rana et al. (2016) earlier elaborated the significance of tectonic geomorphology within the Alaknanda valley through morphometric analysis. Tyagi et al. (2009) made a remarkable discussion by identifying of differentially uplifted areas using the steepness index of the Alaknanda River profile. These workers (Tyagi et al. 2009) divided the longitudinal profiles into several reaches to compute the normalized steepness index (ksn). Although considering a single longitudinal profile merely provides some instances of differential uplift and erosion at local scale, for the regional tectonic framework, it is essential to estimate ksn for all the available drainages as well, and correlate the average ksn with erosion rate for a robust conclusion. Therefore, the aim of this present study is set to understand the role of tectonics in the formation of anomalies on the longitudinal profiles, variation in steepness and concavity index and estimation of erosion rates in between major Himalayan thrust belts in the Alaknanda-Bhagirathi catchment. In the event of an emerging climate change scenarios marked by heavy rainfall, cloud burst triggered flash-floods or the Glacial Lake Outburst Floods (GLOF), an assessment of the longitudinal profiles across these Himalayan states is essential to anticipate their response.

2 Study Area

2.1 Location

The Alaknanda and Bhagirathi Rivers (Fig. 3) flow over the active orogeny of the Himalayas (Ray and Srivastava 2010). These two rivers cross the Himalayan thrust belts in the northern section while the southern section is characterized by the vast Ganga plain (Singh et al. 2007, 2015). River Alaknanda originates from the Satopanth glacier at an elevation of 3641 m, and the Bhagirathi originates from the Gangotri glacier near Gaumukh at an elevation of 3892 m a.s.l. These two rivers meet near Devprayag and form the sacred river Ganga. Geomorphological breaks that are marked by a series of knick points in Bhagirathi and Alaknanda show deep incision and gorges in the upper and middle reaches. Geomorphologically, these two rivers show three distinct reaches: (a) the glaciated, semi-arid U- shaped valley in the upper section; (b) steep V-shaped deep gorges in the middle reach and this section forms the main orographic barrier of the ISM and shows intense precipitation marked by numerous geomorphic processes (Bookhagen and Burbank 2006; Wasson et al. 2008); and (c) gentle slope with sediment-filled wide valley in the lower reaches.

Fig. 3
figure 3

Location of the Alaknanda-Bhagirathi catchment. a Major rivers and locations in the study area. See Fig. 2 for the swath profile of A-B. b Simplified geological map of the study area (reproduced after Singh 2018). Note that the thrusts are not limited as shown in this figure but continuous features along the Himalayas from west to east. c The confluence of Alaknanda and Bhagirathi near Devprayag. d Landscape configuration at the upper Alaknanda catchment

Full size image

2.2 Geology

The Alaknanda-Bhagirathi catchment shows two major geomorphic divisions: (i) the Himalayan active terrain and (ii) the Ganga foreland plain (Ray and Srivastava 2010). Himalayan thrust belt, located in the northern section of Alaknanda- Bhagirathi catchment shows several parallel thrusts marked by distinct lithology groups. The major lithology in this catchment from north to south are (i) Tethyan sedimentary, (ii) upper higher Himalayan crystalline (Vaikrita group), (iii) lower higher Himalayan crystalline (Munsiari formation), (iv) inner lesser Himalayan sedimentary, (v) outer lesser Himalayan sedimentary, and (vi) Quaternary sediment. While flowing over the Himalayas, Alaknanda river crosses the South Tibetan Detachment (STD) in the northernmost region of catchment, followed by Main Central Thrust (MCT) II and I, Main Boundary Thrust (MBT), and Main Frontal Thrust (MFT) in the south before reaching the vast Gangetic Plain.

3 Data and Methodology

3.1 Data Processing

High-resolution digital elevation models are often considered in quantitative landscape evolution studies around the world. In this study, we utilized the Shuttle Radar Topographic Mission (SRTM) digital elevation model (DEM) of 90 m spatial resolution (http://srtm.csi.cgiar.org). The DEM was processed in ArcGIS 10.1 to fill data gaps, generating Alaknanda-Bhagirathi catchment and streams, utilizing hydrology tools. We also delineated sub-catchments by carefully selecting all the tributaries that have streams of 4th order while joining the trunk stream. Then we computed relief, slope, and hypsometric integral [mean elevation-min elevation/max elevation- min elevation] for all the sub-catchments.

We plotted longitudinal profiles in MATLAB (https://www.mathworks.com) based stream profiler tool (https://geomorphtools.geology.isu.edu). Then normalized steepness index (ksn) for the entire stream network was estimated in TopoToolbox (Schwanghart and Kuhn 2010).

Pixel-based TRMM precipitation data (www.geog.ucsb.edu/~bodo/TRMM) was utilized for swath profiling and rainfall-ksn correlation.

3.2 Longitudinal Profile and Steepness-Concavity Analysis

The effect of spatially varying rock uplift rates on the bedrock river profiles can be understood from the steepness index (Howard et al. 1994; Kirby and Whipple 2001). The river profile evolution can be expressed as an equation through the incision modeling of the power-law function of channel gradient and drainage area (Howard and Kerby 1983; Howard 1994; Kirby and Whipple 2001):

$$\frac{dz}{dt}=U\left(x,t\right)-K{A}^{m}{S}^{n}$$
(1)

where dz/dt indicates the changes in the bedrock elevation through time, U indicates the rate of bedrock uplift while the base-level is constant, S is the local channel gradient, A is the upstream drainage area and K is the coefficient of erosion. The constants m and n are positive and are closely related to erosion, hydraulic geometry, and hydrology (Howard et al. 1994; Whipple and Tucker 1999). For equilibrium condition, the above equation can be rewritten as:

$${S}_{c}={\left(\frac{U}{K}\right)}^{1/n}{A}^{-\left(\frac{m}{n}\right)}$$
(2)

and

$$S={k}_{s}{A}^{-\theta }$$
(3)

where S denotes the slope, ks is the steepness index and θ indicates the concavity index. Since past decades, several studies have suggested a close relationship between erosion rate and the steepness index. As differential uplift has different erosion rates along the channel, as a proxy, this index can be used to determine the variable uplift of a catchment. Although the steepness index is a good indicator of varying uplift, ks can significantly be affected by lithological variation (Das 2020). Moreover, departures from the normal concavity (θ = 0.5 ± 0.2) may indicate tectonic uplift by disrupting the longitudinal profile (Burbank and Anderson 2012). In this study, the ks values are normalized using a reference concavity (θ) of 0.45 for comparison of the steepness in different reaches, tributaries and sub-tributaries.

4 Neo-Tectonic Movement and Channel Evolution

4.1 Longitudinal Profiles of Bhagirathi-Alaknanda

Figure 4 shows longitudinal profiles of the Bhagirathi and Alaknanda rivers. It can be clearly observed that characterized by the glacier domain, first few kilometers of Alaknanda show a very steep gradient. Later, the river shows a series of knickpoints in between the Himalayan thrusts (STD and MCTs). These thrusts closely match with the knickpoints in Alaknanda (Fig. 4). From the source to the MCT I, the Alaknanda forms the knick zone. After crossing the MCT I, the gradient becomes very low. Compared to the Alaknanda, the Bhagirathi river shows a lesser gradient and fewer knick points. While Alaknanda shows prominent knicks in between MCT I and MCT II (MCT II is also known as Vaikrita thrust), such knicks are not very common in the Bhagirathi. Most of the major steep sections on the longitudinal profile are located above the MCT II.

Fig. 4
figure 4

Longitudinal profiles and the steepness-concavity index in between thrusts and major steep reaches. a Alaknanda; b Bhagirathi

Full size image

4.2 Steepness-Concavity Index and Influence of Major Thrusts

In the present study, ksn and θ are calculated for reaches in between major thrusts or distinct steeper topography. The longitudinal profile of Alaknanda is divided into four sections for the steepness concavity analysis. The normalized steepness index (ksn) above STD is 121 (θ = 0.23 ± 0.076), followed by 792 (θ = −0.54 ± 0.3) in between STD and MCT II, 446 (θ = −17 ± 40) between MCT II and MCT I, and 99.1 (θ = 0.71 ± 0.33) from MCT I to Ganga plain, respectively, from north to south. A careful observation indicates that most of the knickpoints in the study area are distributed between MCT I and STD.

For the Bhagirathi, ksn was calculated for three reaches (a) the irregular elevated, steeper topography in the upper-middle section above MCT II (VT), (b) between MCT II and MCT I and (c) relatively gentle middle-downstream reach. The ksn value in the upper Bhagirathi above MCT II is 297 (θ = −0.005 ± 0.16), between MCT II and MCT I is 373 (θ = 4.8 ± 5.5) in, and in the gentler plain section below MCT I until Alaknanda confluence the ksn is 153 (θ = 1.2 ± 1.1).

Several workers have reported that during the Pleistocene to Holocene periods, major boundary thrusts and faults (e.g. MBT, North Almora Thrust, MCT Alaknanda fault etc.) were active in this region (Sundriyal et al. 2007). Considering this, it is obvious that the geomorphology of this region is highly influenced by the tectonic perturbation by modifying the paleo-landscape. The active landscape has been directed to form some remarkable geomorphic features such as deep incision, changes in river courses and meanders, deformed and uplifted sequential fluvial terraces etc. (Sati et al. 2008) as a result of the recent mountain building process.

Vance et al. (2003) studied the erosion and exhumation rates in the Bhagirathi-Alaknanda catchment using in-situ 10Be cosmogenic dating. In this study, besides longitudinal profile, a ksn map for all the tributaries and sub-tributaries was prepared for the study area (Fig. 5a) for correlating the erosion rates at different thrust regions using the data produced by Vance et al. (2003). A significant positive correlation between ksn and erosion rate is observed (Fig. 6) and therefore, any conclusion drawn from the ksn may be considered as valid. Vance et al. (2003) showed a very high erosion rate in upstream of Helong (2.4 ± 0.2 mm yr−1), followed by Srinagar (1.5 ± 0.2 mm yr−1), Malari (1.2 ± 0.1 mm yr−1), Bhagirathi and Rishikesh (both 1.0 ± 0.1 mm yr−1). This study shows a very consistent values of average ksn in the upstream of Helong (ksn = 227), Srinagar (ksn = 218), Malari (ksn = 191), Bhagirathi (ksn = 194) and Rishikesh (ksn = 192).

Fig. 5
figure 5

a Variation of the normalized steepness index (ksn) and knickpoint distribution in the Bhagirathi-Alaknanda catchment. Knickpoints are identified mathematically in MATLAB based Topotoolbox algorithm. The mathematical algorithm identified 2466 knickpoints in the study area. b Vasudhara knickpoint in the upper Alaknanda catchment

Full size image
Fig. 6
figure 6

Correlation between normalized steepness index (ksn) and 10Be derived erosion rates at upstream of several locations. 10Be derived erosion rates for the study area are taken from Vance et al. (2003). Note that the data of Bhagirathi and Rishikesh are overlapping on the graph

Full size image

4.3 Estimation of Erosion Rates Between Thrusts

The integrated study of the longitudinal profile, steepness index and concavity index gives good insights into spatial distribution of differential erosion in Bhagirathi-Alaknanda basins. The upstream regions of the Alaknanda-Bhagirathi catchment in between MCTs show deep incisions and steeper gradients compared to the other reaches. Therefore, it may be said that in the study area, differential rock-uplift rate exerts a critical control on the channel gradient and that makes a difference in the steepness and concavity of the channels in different reaches.

Using the linear regression equation derived from the relationship between ksn (this study) and erosion rate (Vance et al. 2003), a further estimation of the average erosion–deposition rates between thrust are calculated by the following equation:

$$Erosion \,rate=\left(0.0314\times {k}_{sn}\right)-5.0075$$
(4)

while the regression showed R = 0.91, and p value = 0.03.

Table 1 shows that the average ksn value above STD (toward Tibet) is 178 while the derived average erosion rate is 0.58 mm/yr. Although the Tethyan sediment is more erodible than Himalayan crystalline rocks, the dominance of glacial activity over the fluvial process is the main reason behind the lower erosion rate above STD. Most of the landscape in this region is covered by Alpine and Valley glaciers that make a thick veneer on the rocky substrate and prevent intensive erosion. Consequently, the region between MCT I and II show the maximum estimated rate of erosion (ksn = 251; erosion rate = 2.87 mm/yr). This region is found to have differential exhumation rates through geological past with a rapid exhumation of garnet rock as high as 9 mm/yr reported after reactivation of MCT during Miocene (~8 Myr) (Harrison et al. 1997; Catlos et al. 2001; Vance et al. 2003). Several studies suggest that in tectonically active landscape, differential ksn values between faults indicate differential uplift (Kirby and Whipple 2001). Tyagi et al. (2009) suggested that the MCT region of the Alaknanda falls under a higher rate of uplift. Hence, based on results from this study, it can be said that the higher erosion rate between MCT I and MCT II may be a result of a higher uplift rate which is reflected by a higher frequency of landslides (Sarkar et al. 2015). Also, the MCT region shows high relief topography compared to the region above STD, which could be possibly another reason behind active incision and higher erosion rate. However, the negative values southwards of MBT (Table 1) indicate dominant depositional regime although, the present inferences are purely based on a statistical correlation between the erosion rate and ksn, demanding further validation. Based on the above information, the average rates of deposition between MBT-MCT I, MFT-MBT and on the frontal plain are estimated as 0.42, 4.07 and 4.59 mm/yr, respectively.

Table 1 Average normalized steepness index (ksn) and erosion/deposition rates at regions in between major thrusts
Full size table

4.4 Correlation of Landscape Variables and Rainfall with ksn

Correlation between geomorphic parameters that are related to erosion (e.g. elevation, relief, slope, hypsometry etc.), rainfall, and lithology as independent factors and the normalized steepness index (ksn) as dependent factor provides critical quantitative information about the potential control of the erosion in a given area (Abrahami et al. 2016). Therefore, although our study mainly focused on understanding differential uplift with respect to varying uplift rates in between major thrusts, in addition, we correlated major potentially controlling factors of the varying erosion rate in the Alaknanda-Bhagirathi catchment (Fig. 7). Among five variables (mean elevation, relief, slope, hypsometric integral, and rainfall) only relief is found to have a significant relation with ksn (R = 0.698, α = 0.05). It implies that, in the Alaknanda-Bhagirathi catchment, regions with a great elevation range experience greater erosion. Moreover, all these high relief catchments were found to have fluvially dominant processes contributing a large area between MCT I and II.

Fig. 7
figure 7

Correlation between major driving factors with ksn. Blue and orange dots indicate sub-catchments in Alaknanda and Bhagirathi-Ganga respectively

Full size image

The evolution of the Indian Summer Monsoon in the Himalayan region is hand in hand with the evolution of mountain building processes which are marked by the thrust reactivations in this region. The absence of any correlation between the spatial distribution of precipitation and ksn indicates the landscape processes and erosion rates are more controlled by tectonics than climate. Abrahami et al. (2016) showed a similar observation in Sikkim Himalayas, where they concluded that while erosion rates are highly correlated to ksn and primarily controlled by tectonic uplift, precipitation only has a secondary control.

In understanding the complex nature of coupled tectonic and climatic influence on the evolution of post-orogenic mountains, rock uplift, and erosion; ksn have been extensively considered as a proxy of erosion rate because of its strong correlation with in-situ based erosion rate data (e.g. Abrahami et al. 2016; Dey et al. 2016; Scherler et al. 2017; Nennewitz et al. 2018; Adams et al. 2020). The ksn data is proven to be efficient in comprehending such external perturbations on the landscape dynamics and varying erosion rates in a systematic way. Based on the outcomes of our study and the previously published studies, this method can be considered as one of the simplest approaches that may be applicable in other regions in the Himalayas to understand the impact of tectonics and monsoon climate on topographic evolution even for the regions where in-situ based erosion rate data are unavailable.

5 Conclusion

River longitudinal profiles and distribution of normalized steepness index (ksn) provide significant insight into the landscape evolution and erosion potential of a region. Our study demonstrates a systematic analysis of river longitudinal profiles and ksn to comprehend the role of varying uplift rate, geomorphic indices and monsoon rainfall in differential erosion rates in between major thrusts in the Alaknanda-Bhagirathi catchment. A significant correlation between 10Be derived erosion rate and ksn indicated high rate of erosion (2.87 mm/yr) between MCT I and II. Fluvial processes may have a dominant role in erosion as the maximum erosion rate is found between MCT I and II where the region is under transition from the glacial to the fluvial regime with intense headward erosion. Though topographic relief has significant control on erosion rate in this region, because of no correlation between rainfall and ksn, we conclude that differential tectonic uplift is the main driving factor in topographic evolution while rainfall exerts a secondary control.