Introduction

The evolution of a landscape by the formation, modification, or destruction of the landforms is often associated with the structures and processes related to the deformation of the Earth’s crust (i.e. tectonics) controlled by endogenic processes (Bull and McFadden 1977). Often referred to as tectonic geomorphology or morphotectonics, describing the relationship between landforms and tectonic activities such as faulting, folding, tilting, upliftment or subsidence and erosion processes (Burbank and Anderson 2001), it is immensely important for identifying and understanding the seismic hazards and for quantifying active and geologically recent tectonic deformations (Ramírez-Herrera 1998; Bahrami 2013). Various studies suggest that morphometric indices can help in ascertaining the geological and geomorphological characterization of an area (Bull and McFadden 1977; Rockwell et al. 1985; Wells et al. 1988; Keller and Pinter 1996; Azor et al. 2002; El Hamdouni et al. 2008; Dehbozorgi et al. 2010).

Quantification of morphometric indices has proven to be useful in the tectonically active regions like the Himalaya by succoring in understanding of the basin characteristics related to the shape, slope, rock hardness, diastrophism and other factors (Strahler 1964; Jackson et al. 1998; Sung and Chen 2004; Ramsey et al. 2008). These indices have been extensively used by gathering the necessary information from the topographic maps, and aerial photographs and continue to be in vogue due to the advancements in the use of satellite remote sensing for topographic mapping (Horton 1945; Keller and Pinter 1996; Wells et al. 1988). Accordingly, seven morphometric indices that include drainage pattern/stream order (U), the bifurcation ratio (Rb), drainage basin asymmetry (AF), stream length-gradient index (SI), valley floor width to valley height ratio (Vf), Basin relief (H) and, longitudinal river profiles (LRP) (Bull and McFadden 1977; Pandey and Dubey 2003; Ahmed et al. 2010; Thomas et al. 2010), were assessed using multisource DEMs. Studying these indices for inferring the tectonic activity has enhanced the scientific understanding of various seismically active areas around the world (Rockwell et al. 1985; Wells et al. 1988; Cox 1994; El Hamdouni et al. 2008) inclduing Kashmir Valley (Ahmad and Bhat 2012; 2013; Ahmad et al. 2014, 2018; Dar et al. 2014a). These indices are calculated from the topographic information or can be generated by several techniques like extensive ground surveys, Digital Elevation Models (DEMs) produced from topographic maps, aerial photographs, laser scanning, and Interferometric Synthetic Aperture Radar (InSAR) (Seferciket al. 2007; Sharma et al. 2010).

The morphometric characterization of tectonically active areas has become easier with the emergence of freely available DEMs, advancements in GIS platforms, and the development of advanced GIS algorithms (Verrioset al. 2004; Hayakawa and Oguchi 2006; Font et al. 2010; Ahmad et al. 2018). Moreover, it has been studied that the resolution of DEMs and the technique for acquiring the topographic data significantly affect the accuracy of the processes being simulated (Jenson 1991; Wolock and McCabe 2000; De Vente et al. 2009). Many researchers have identified that the elevational differences in the DEMs are more pronounced in high-relief areas as compared to lower elevations which may lead to misinterpretation of the analysis (Toutin 2002; Hirano et al. 2003; Orvis 2003). Many researchers have also ascribed false heights to landscape elements owing to the lack of knowledge about the datum and/ or ellipsoid/ geoid on which the geographic data is projected (Perez et al. 2006; Apollo et al. 2020). Therefore, it is imperative to have DEMs with similar projection characteristics before going for any comparative quality assessment.

This study evaluates the robustness of three similar-resolution DEMs using seven morphometric characteristics and field-based measurements for the first time over tectonically active Kashmir Himalaya. Although there is a plethora of literature related to the morphotectonics of the region, the suitability of DEMs for tectonic characterization has never been attempted. In this paper, ASTERGDEM, CARTODEM, and SRTMDEM, all with a spatial resolution of 30 m, were used over Kashmir Himalaya, India to ascertain their robustness for tectonic characterization of the region.

Study site description

Jhelum, an asymmetric basin, draining through Kashmir Valley is an oval-shaped basin that lies in the North-West Himalaya between the Pir Panjal Range in the SW and the Greater Himalayan Range in the NE (Romshoo et al. 2017) in the Indian state of Jammu and Kashmir. The Pir Panjal Range comprises a complex faulting pattern including several thrusts (Thakur et al. 2010). The valley comprises of Archean to Recent stratigraphic record of rocks of all ages and is filled with up to 1300 m thick ‘Karewa Group’ formations which are Plio-Pleistocene fluvio-glaciolacustrine sediments (Bhatt 1975; Kotlia 1985; Dar et al. 2014a). The valley is characterized by numerous springs and 24 main tributaries that drain into the Jhelum River forming the Jhelum basin. The present analysis has been carried out in one of the watersheds of Jhelum, the Ferozpora watershed (Fig. 1). The watershed drains an area of ~ 450 km2 lying between 33° 54′ 26′′–34° 18′ 23′′ N lat and 74° 18′ 40′′–74° 42′ 25′′ E lon with the altitude extremes of 1455 m and 4538 m asl (Rashid et al. 2016). Ferozpora Nallah, a tributary of Jhelum drains the upper snow-clad slopes of the Pir Panjal (Dar et al. 2014b) between the Jamianwali Gali and the Apharwat ranges (Rashid et al. 2016). On entering the relatively plain area, Ferozpora Nallah gets divided into two branches emptying into the Wular, Asia’s largest freshwater lake (Meraj et al. 2018). The Ferozpora Nallah has several streams forming a well-developed dendritic drainage pattern in the upper portion of the catchment and more or less parallel drainage pattern in the lower portion (Lawrence 1895; Wadia 1975). Geologically, the area is dominated by the basic volcanic rocks of Panjal traps, Triassic Limestone, Quaternary sediments of Karewa deposits, and alluvium (Middlemiss 1910; Bhatt 1989).

Fig. 1
figure 1

Location of the study area. Top panel: topographic variability of Ferozpora watershed using CARTODEM along with sub-watersheds, Bottom right: Ferozpora watershed in Kashmir valley. Top right: location of study area with respect to Indian mainland. Bottom panel: zoomed-in view of sub-watersheds

Full size image

Materials and methods

Datasets and data quality

The datasets used in this research could be broadly classified into four categories that include multisource DEMs, high-resolution basemap imagery of ArcMap 10.1, Google Earth images, and ground data collected from Global Positioning System (GPS). CartoDEM v3 R1 (Cartosat-1 DEM Waterbody Flattened) is an openly accessible National DEM provided by the Indian Space Research Organization (ISRO) with a spatial resolution of 30 m (Baral et al. 2016). Shuttle Radar Topography Mission DEM (SRTMDEM) is a product of a collaborative effort by the National Aeronautics and Space Administration (NASA), the National Imagery and Mapping Agency (NIMA), the German Aerospace Centre (DLR) and Italian Space Agency (ASI). The data has been collected using the InSAR technique at 1 arcsecond (~ 30 m) and 3 arcseconds (~ 90 m) (Rabus et al. 2003; Sharma et al. 2010). Another DEM, ASTER GDEM v2 (Advanced Spaceborne Thermal Emission and Reflection Radiometer) synthesized from the ASTER stereopairs from NASA’s Terra spacecraft is available at 30 m (Abrams 2000). Further, drainage was manually delineated from high-resolution basemap imagery of Arc Map 10.1. Handheld Trimble Juno T41/5 with a positional accuracy of ± 8 m was used for in-situ data collection. The logging interval for coordinate data collection was set to 10 s for at least 5 min at a single location. Table 1 provides details of the datasets used in this study.

Table 1 Details of the datasets used in the study
Full size table

Methods

In the present study, morphometric analysis of seven parameters (U, Rb, AF, SI, Vf, H and LRP) has been carried out. The results from morphometric analysis were then compared with the field-based measurements to study the morphotectonics of this region and to check the efficacy of the DEMs for tectonic studies. The automatically derived drainage patterns from coarser-resolution DEMs were validated with the manually delineated drainage patterns from high-resolution satellite data (1:5000 scale). The manual delineation with the cognitive inputs from the analyst was employed given its potential in mapping the geomorphic features in a topographically rugged terrain (Rashid et al., 2016, 2020a, b; Majeed et al. 2021). The DEMs (ASTERDEM, CARTODEM, and SRTMDEM) were used in the GIS environment to generate the boundaries of the sub-watersheds and the drainage pattern by defining a pour point at the outlet of the Ferozpora watershed (Garbrecht and Martz 1997; Tarboton and Shankar 1998; Altaf et al. 2013). The pour point is a location where water drained from the whole of the watershed flows into the main river (Antonić et al. 2001). After that, the drainage network of the Ferozpora watershed was manually digitized using the high-resolution basemap imagery of ArcMap 10.1.

Stream order (U)

For stream ordering, Strahler’s Law (Strahler 1952; Schumm 1956; Singh 1980; Romshoo et al. 2012) was followed by designating an unbranched stream as the first-order stream which after joining another first-order stream forms a second-order stream just below the junction. Two second-order streams join together to form third order and so on. The number of streams of each order were counted and recorded. The calculations were carried out for both the automated and manually delineated drainage networks. Therefore, the drainage pattern generated from DEMs was validated with that from the manual digitization to check the closeness of patterns between the two in the Ferozpora watershed.

Bifurcation ratio (Rb)

It is defined as the ratio between the total number of stream segments of one order (Nu) to that of the next higher (Nu + 1) order (Strahler 1957). It is a dimensionless property and was calculated as per the Eq. (1) as:

$$Rb\; = \;\frac{Nu}{{Nu\; + \;1}}$$
(1)

The mean bifurcation ratio (Rbm), which is the mean of the Rb of all the orders, is an indicator of structural complexity and permeability of the terrain and is thus negatively correlated with the permeability of a watershed (Horton 1945). High Rbm indicates high structural complexity and an early hydrograph peak with a potential for flash flooding during the storm events which results in the degradation of topsoil (Howard 1990; Rakesh et al. 2000). This Rbm was calculated from all the sources and was later on compared with the calculations from manual digitization to check the accuracy of the DEMs.

Drainage basin asymmetry (AF)

AF allows determination of the general tilt of the basin landscape, irrespective of whether the tilt is due to local or regional tectonic deformation (Hare and Gardner 1985) and is expressed as per the Eq. (2) as:

$$AF\; = \;\left( {\frac{Ar}{{At}}} \right)\; \times \;100$$
(2)

where AF is the drainage basin asymmetry, At is the total area of the drainage basin, and Ar is the area of the drainage basin to the right side of the trunk channel when facing downstream.

AF was developed for detecting tectonic tilting as the active tectonic deformations form drainage with distinct patterns and geometry. For a stable stream network, AF should be equal to 50. AF greater than 50 suggests that the channel has shifted towards the left side of the channel while the vice versa is true for the values lesser than 50 (Hare and Gardner 1985; Gardner et al. 1987; Ahmad et al. 2014). AF was calculated from the three DEMs for the Ferozpora watershed which was then compared with the AF results from a digitized stream network.

Stream length-gradient index (SI)

This parameter is calculated for a particular stream of interest and correlates to the stream power or differential rock erodibility (Hack 1973). Hydrologically, total stream power available at a particular reach of the channel is important because it is related to the ability of a stream to erode its bed and transport sediment (Azor et al. 2002). The SI is very sensitive to changes in the channel slope and this sensitivity allows the evaluation of relationships among possible tectonic activity, rock resistance, and topography (Keller and Pinter 1996). The stream gradient index (SI) calculated for a particular reach of interest (stream segment) can be computed as per the Eq. (3) as:

$$SL\; = \;\frac{\Delta H}{{\Delta L}}\; \times \;L$$
(3)

where ΔH/ΔL is the channel slope or gradient of the reach, and L is the total channel length from the point of interest where the index is being calculated upstream to the highest point on the channel. For the ΔH, the maximum and minimum elevations were calculated from the highest point of the stream and the center of the ΔL respectively. ΔL is the length of the reach of interest where the index is being calculated. The calculations were carried out for all the DEMs and digitized stream network to check the accuracy of the DEMs.

Valley floor width to valley height ratio (Vf)

This index differentiates between broad-floored canyons associated with relatively high values of Vf, and V-shaped valleys associated with relatively lower values (Bull and McFadden 1977; Keller and Pinter 2002). High values of Vf are associated with low uplift rates so that streams cut broad valley floors. Low values of Vf reflect deep valleys with streams that are actively incising, commonly associated with uplift (Keller and Pinter 1996). This index is based on the observation that areas undergoing rapid upliftment are marked by incised streams with narrow valley floors and V-shaped valley profiles. Vf is expressed as per the Eq. (4) as:

$$Vf\; = \;\frac{2Vw}{{\left[ {\left( {Eld\; - \;Esc} \right)\; + \;\left( {Erd\; - \;Esc} \right)} \right]}}$$
(4)

where Vw is the width of the valley for given profile at fixed length, Erd and Eld are the elevation of right and left divides for a given section line respectively facing downstream and Esc is the elevation of the valley floor. The Vf calculations were carried on a sub-watershed SW3 of the Ferozpora watershed to validate the results from DEMs with the field-based measurements. For the field measurements, 30 elevation values (GCPs) were gathered using the GPS across a transect from one to another ridge of SW3.

Basin relief (H)

The elevation difference between the highest and the lowest points on the valley floor of a watershed is referred to as total relief (Schumm 1956). There is a strong correlation between hydrological characteristics and the H of a drainage basin (Dodov and Foufoula-Georgiou 2005). It is an index of overall steepness of a drainage basin as well as of the intensity of erosion processes operating on the slopes of a basin (Dodov and Foufoula-Georgiou 2005) and is expressed as per the Eq. (5) as:

$$H\; = \;H_{\max } \; - \;H_{\min }$$
(5)

where Hmax is maximum elevation and Hmin is the maximum elevation of the basin. The calculations were carried out on SW2 sub-watershed of the Ferozpora watershed using the three DEMs, and the field measurements. A continuous stream of points was acquired along the main stream of SW2 using the GPS for the field-based measurements. Then the basin relief was calculated using the maximum and minimum elevation values from all the sources to check the efficacy of DEMs in representing the H.

Longitudinal river profile (LRP)

Rivers that are not tectonically perturbed typically develop a smoothly changing, concave longitudinal profile (Zuchiewicz 1980). Departures of the river gradient from this ideal smooth shape may reflect variations in the lithology of the river bed, or tectonic activity (Mackin 1948; Turowskiet al. 2009). Rivers that are tectonically disturbed are predicted to approach a gradient profile rapidly (Snow and Slingerland 1987) once such disturbance ceases. Thus, perturbations in river profiles may be interpreted as a response to ongoing tectonism. The plotting of longitudinal profiles shows altitude against distance downstream. It represents the channel gradient of the river from its source to the mouth. LRP was generated by plotting a histogram between the river length and the elevation of the Ferozpora watershed to study the tectonic signatures in the area. The results were validated with the field-based measurements wherein the LRPs of the two sub-watersheds SW1 and SW2 of the Ferozpora watershed were generated using the GPS surveys. A continuous stream of 180 GCPs for the elevation measurements were taken along the main streams draining the SW1 and SW2 which were then plotted against the length of these rivers. The GPS-based LRP was then compared with the LRPs of SW1 and SW2 generated from the DEMs for the same river segment.

Results

This study involved use of seven morphometric parameters and field-based GPS observations to evaluate the quality of 3 multisource DEMs.

The stream order (U) of the Ferozpora watershed was delineated from the three DEMs and a high-resolution image as shown in Fig. 2. The drainage network showed that the stream order going to 5th order streams. The highest synonymity in the pattern was observed in the streams generated from the CARTODEM when compared to the manually digitized streams especially the 5th order. The streams showed a somewhat similar pattern up to the 4th order in all the DEMs and deviation in the 5th order. The 5th order stream in all the DEMs and high-resolution imagery were observed in the lower plains of the watershed. It is hence obvious that the rivers are best represented by the CARTODEM.

Fig. 2
figure 2

Stream order of Ferozpora watershed as delineated from (a) ASTERDEM; b CARTODEM; c SRTMDEM; aʹ–cʹ Comparison of manually digitized drainage (in Red) with the DEM derived drainage patterns at higher region; aʹʹ–cʹʹ Comparison of manually digitized drainage (in Red) with the DEM derived drainage patterns at planar region

Full size image

The bifurcation ratio (Rb) generated from the streams of high-resolution imagery and DEMs showed close resemblance in all the stream orders. The mean bifurcation ratio (Rbm) showed the closest resemblance in streams generated from CARTODEM compared to manually digitized streams followed by SRTMDEM. Rbm in the Ferozpora watershed is 4.12, 5.44, 4.07, and 4.39 for the digitized streams and streams derived from ASTERDEM, CARTODEM, and SRTMDEM respectively (Table 2).

Table 2 Bifurcation ratio (Rb) values of the Ferozpora watershed
Full size table

The parameters used for the calculation of AF for the Ferozpora watershed are shown in Fig. 3. The analysis of drainage basin asymmetry (AF) for the Ferozpora watershed indicated ASTERDEM and SRTMDEM show values less than 50 i.e. 31 and 27 respectively, indicating that the channel downstream has shifted towards the right side of the basin. On the contrary, calculations from the CARTODEM (AF = 63) reveal that the channel has shifted towards the left side. The calculations cohere with the outputs from the digitized stream network generated from high-resolution imagery. Thus, indicating that the results from CARTODEM are closer to reality than the other two DEMs (Table 3).

Fig. 3
figure 3

Parameters used for calculation of AF of the Ferozpora watershed. At—Total basin area and Ar—Area of drainage basin to the right side of trunk channel when facing downstream

Full size image
Table 3 Drainage basin asymmetry (AF) of the Ferozpora watershed
Full size table

The higher values of SI from the SRTMDEM (863.4), CARTODEM (632.2), and ASTERDEM (423.9) indicate higher tectonic activity in the Ferozpora watershed (Table 4). Our calculations from the GPS data also revealed high tectonic activity (SI = 742.9) which demonstrated the closest resemblance to the values from CARTODEM followed by SRTMDEM indicating that the CARTODEM shows better results than the other two DEMs (Fig. 4).

Table 4 Stream gradient index (SI) of the Ferozpora watershed. ΔL = 2009 m
Full size table
Fig. 4
figure 4

Parameters used for calculation of SI of the Ferozpora watershed. ∆L—portion of the channel where the index is being calculated and L—total channel length from the point of interest where the index is being calculated upstream to the highest point on the channel

Full size image

The Vf calculations carried out for sub-watershed SW3 using the valley cross-section are shown in Fig. 5a, b. The results revealed that all the DEMs indicate SW3 is a U-shaped with lesser tectonic activity. On contrary, field-based GPS measurements showed that SW3 is V-shaped with higher tectonic activity (Table 5).

Fig. 5
figure 5

a Transect used for calculation of Vf; b Vf calculations for the transect (where Vw width of the valley for given profile at fixed length; Erd and Eld the elevation of right and left divide; Esc elevation of the valley floor)

Full size image
Table 5 Valley floor width to valley height ratio (Vf) of SW3
Full size table

H was specifically used to check the robustness of DEMs in capturing the highest and lowest elevation of a mountainous watershed. For this parameter, the analysis was carried out on SW2 sub-watershed over a continuous stream of 65 points using a GPS. These points were further compared with the elevation values of DEMs at the respective locations. The results showed that CARTODEM (H = 39) showed the closest resemblance to the GPS-acquired relief (H = 35) values (Table 6).

Table 6 Basin relief of SW2
Full size table

As shown in Fig. 6, the LRP of Ferozpora showed a steep slope at the source followed by the decreasing gradient in all the DEMs which converges to a convex depression particularly within the resistant lithology of the Panjal trap from 1–3 km. The river has a steep slope up to 10 km which becomes gentle from 10–30 km and planar beyond 30 km in all DEMs. To check the accuracy of the DEMs, 2 sub-watersheds of the Ferozpora watershed, SW1 and SW2 shown in Fig. 7a, b respectively, were taken into account. It was observed that the elevation values collected using GPS showed the closest resemblance to the elevation values of the CARTODEM in both the sub-watersheds.

Fig. 6
figure 6

LRP of Ferozpora watershed based on different DEMs: a ASTER; b CARTO; c SRTM

Full size image
Fig. 7
figure 7

a LRP for SW1; b LRP for SW2

Full size image

Discussion

The DEM quality in the tectonically active Kashmir Himalaya (Bhat 1982; Schiffman et al. 2013; Mir et al. 2017) was assessed using morphometric parameters and field data. While all the 7 morphometric parameters, were specifically analyzed to identify the best-fit DEM, the 5 parameters that include Rb/Rbm, AF, SI, Vf, and LRP provided some insights into the tectonic behavior of the Ferozpora watershed. The analysis shows that streams up to 5th order could be delineated from both the DEMs as well as manual digitization from the high-resolution satellite data. It is pertinent to mention that the properties of the stream networks are very important to study basin characteristics (Strahler 2002). It has been demonstrated that higher the number of stream segments in a particular basin higher will be the order of streams (Shreve 1966; Saran et al. 2010). The higher order of the streams infers high drainage density (Godsey and Kirchner 2014) controlled by the rock types (Tucker et al. 2001; Sangireddy et al. 2016), vegetation cover (Istanbulluoglu and Bras 2005; Hooshyar et al. 2019), and slope (Tucker and Bras 1998; Schneider et al. 2017). It was observed that the stream pattern is best represented by CARTODEM, however, none of the DEMs highlighted the bifurcation of the streams in the 5th order. Similarly, the parameter H confirms that the elevation values of CARTODEM showed the closest resemblance to the GPS-acquired values also observed by Rawat et al. (2019).

The Rbm from the manually digitized streams, CARTODEM and SRTMDEM indicated values between 4 and 5 indicating tectonic stability compared to that of ASTERDEM (Rbm = 5.44) indicating some tectonic activity. The Rbm values for tectonically stable watershed ranges from 3–5 indicating that the geologic structures do not distort the drainage pattern (Bali et al. 2012; Dar et al. 2014a). The Rbm as calculated from all the four sources (viz., three DEMs, and manual delineation) is less than 5 in all the stream orders except for the 4th and 5th order stream generated from the ASTERDEM. The Rbm in the 5th order of streams from ASTERDEM is 9 because the DEM has delineated nine 4th-order streams which converge into one 5th order thus reflecting the structural disturbance in the area which is contrary to results from manual delineation and the other two DEMs indicating the possible source of error in ASTERDEM (Gesch et al. 2012). Moreover, the parameter AF has been developed for detecting tectonic tilting as the active tectonic deformations form drainage with distinct patterns and geometry. While the results from ASTERDEM and SRTMDEM show that the Ferozpora watershed has shifted towards the right side of the basin, the CARTODEM and digitized stream networks on contrary revealed that the channel has shifted towards the left side. It indicates that the results from CARTODEM are closer to reality than the other two DEMs. Irrespective of the direction of tilting, an AF factor deviating from value 50 suggests the active tectonics or structural control (El Hamdouni et al. 2008) in the Ferozpora watershed.

The higher values of the SI from the GPS-based observations, SRTMDEM, CARTODEM, and ASTERDEM indicated that the streams of the watershed cross hard rock terrain associated with high tectonic activity (Hack 1973; Keller and Pinter 2002). The field-based measurements showed the closest resemblance to CARTODEM and the highest deviation from the ASTERDEM (Table 4). For assessing the shape of the SW3, the Vf results from all the DEMs indicated a U-shaped valley with lesser tectonic activity in the area. The field-based GPS measurements showed SW3 is a V-shaped valley. This could be an artifact of DEM resolution over the mountainous SW3 sub-watershed (Kervyn et al. 2008). The LRP of the Ferozpora watershed shows the convergence of steep slope to a convex depression within the resistant lithology of the Panjal trap at the source indicating signatures of tectonic activity in the region (Ahmad et al. 2014). For the DEM accuracy, elevation values from the GPS showed the closest resemblance to the values of CARTODEM (Fig. 7).

Our analysis suggested that the values of AF, SI, and LRP infer tectonic control prevailing over the study area. It is pertinent to mention that the Pir Panjal range has a complex pattern of faulting with the superposition of several thrusts, one of which crosses through the Ferozpora watershed as shown by Shah (2015). This research suggests that CARTODEM is the best DEM representing the morphometric parameters almost coinciding with the values of field-based GPS measurements. Similar assumptions have been made by various researchers like Evans et al. (2008) for Utah, USA, Mukhejee et al. (2013), and Das and Pardeshi (2018) for various parts of India. However, other research findings suggest that CARTODEM has lesser accuracy than ASTERDEM and SRTMDEM (Krishnan et al. 2016; Singh et al. 2016). The findings of this study need to be upscaled to different geological and topographic regimes of the Himalaya to ascertain the best-fit DEM utilizing robust ground-based GPS surveys.

Conclusion

This study offers one of the unique approaches to ascertain the robustness of freely available DEMs using GPS surveys and morphometric indices as proxies over the mountainous Kashmir Himalaya, India. The study assumes importance given that many researchers have widely been using the DEMs for quantifying the tectonic behavior of the Kashmir region without ascertaining the suitability and appropriateness of DEMs for morphotectonic characterization. This study is a first attempt to ascertain DEM quality using morphometric parameters as proxies together with GPS-based field observations in the topographically complex Kashmir Himalayan region. This study is also relevant since vast areas across the Himalayan arc in general and Kashmir region in particular lie in the highest earthquake risk zones (Zone 4 and Zone 5). The analysis using 7 morphometric parameters indicated the robustness of CARTODEM in closely matching the reality (GPS-based measurements and manual digitization from high-resolution imagery). The Vf factor however showed a pronounced bias which could be attributed to the spatial resolution of DEMs in comparison to point data as collected using GPS. Further, certain parameters like AF, SI, and LRP suggested that the area is tectonically active while Rbm and Vf suggested stable behavior of the watershed. Although this research suggested that CARTODEM showed closest match with the field-based GPS observations for the Ferozpora watershed, many other researchers have pointed out caveats related to the accuracy of this DEM. It is, therefore, suggested that the robustness of these DEMs need to be assessed in contrasting lithological and topographic regimes using extensive ground surveys to find the best-fit DEM and its application in ascertaining the tectonic behavior of seismically active regions across mountain regions in general and the Himalaya in particular.