Use of geographic information system for the vulnerability assessment of landscape in upper Satluj basin of district Kinnaur, Himachal Pradesh, India

ABSTRACT

The present study was conducted in the upper river Satluj basin of district Kinnaur, Himachal Pradesh, India. The vulnerability assessment was done on the basis of selected parameters like slope, slope profile, slope aspects, relative relief, curvature, soil texture, lithology, river morphometric, precipitation, land use and land cover, mass movement, flood, geological elements, earthquake occurrences, and anthropogenic activity (hydroelectric projects). Here, a quantitative and qualitative approach was used to generate physical vulnerability assessment index and vulnerability-based maps. Vulnerability assessment of the landscape was done to highlight the risks and sensitivity of the region due to prevailing hazards and anthropogenic activities.

KEYWORDS:

• Geographic information system
• vulnerability assessment index
• hazards and Kinnaur

1. Introduction

The vulnerability is a multidimensional (i.e., physical, social, economic, environmental, institutional, and human), dynamic (changes over time), scale dependent (from individuals to countries), and site specific (each location might need its own approach) (Bankoff, 2003). Vulnerability assessment of landscape includes that how much system is vulnerable as internally and externally due to the risk of climate change, hazards, and human intervention (Kanwar, Kuniyal, & Kumar, 2017). During the last two decades, vulnerability assessment has become essential mainly in the regions where developmental activities for energy or any other activities are continued. Yet, the Hyogo framework recognizes as a key activity to develop a “system of indicators of disaster risk and vulnerability at national and subnational scales” (UNEP, 1992). The term vulnerability is used diversely; therefore, scientists with various disciplines have an ongoing debate regarding its definition. Vulnerability assessment generates the fruitful results and provides a base for decision-making process (Bogardi & Birkmann, 2004). Physical aspect also covered the basic parameters such as slope, slope aspect, slope profile, curvature, land use, soil texture, lithology, and river morphometry (Birkmann, 2006). A slope is a dominant and controlling factor for the incidences of mass movement. The slope aspect also affects settlements, vegetation, agriculture, and geo-hazards processes. The understanding of aspect is important for forest management and planning because it affects the growth and productivity of forest. The slope profile basically influences the soil erosion and mass movement of the area (Tseng, Lin, & Hsieh, 2015). In the Himalayan region, the summital concavity or convexity of a slope is very common (Savinder, 2004). High relative relief indicates the high vulnerability as a geo-hazards point of views. The high relative relief indicates the region as complex and unstable topography (Gansser, 1964). High deviation of curvature influences the surface run-off and soil erosion. Land-use land cover also indicates the region’s ecological status and distribution of the resources and their balances in the region (Anderson et al., 1976). Soil texture and lithology of the area affect the human settlement, agriculture, vegetation, soil erosion, and mass movement of the area. All exogenes is processes are controlled by the particular climatic condition of that region. Indicators are dependent on each other and also influence each other. But some time, human interference affects the indicators and also changes the possible impacts of these indicators.

In the Satluj basin, constructions of hydroelectric projects have been increased after the 1990s in Satluj basin (Gupta & Sah, 2008a). The physical and social environment of the Satluj valley has been adversely affected by hydroelectric projects. Degradation of a physical environment and natural resources of this high-altitude valley is due to haphazard development, which exploits the Agenda 21 (UNEP, 1992). The Satluj valley is also known for its multi-facet hazards, namely, landslides, flood, avalanches, and earthquakes (Gupta & Sah, 2008b). The vulnerability of landscape is high in this valley due to its vulnerable parameters such as steep slope, high relative relief, structural discontinuities, lithology, and unhealthy land cover (Birkmann, 2006; Papathoma, Kappes, & Keiler, 2011).

2. Study area

The study area of the upper Sutlej basin in district Kinnaur, Himachal Pradesh, India, extended from 31° 19′ 04″N to 32° 05′ 59″N latitudes and 77° 82′ 01″E to 78° 83′ 05″E longitudes. Its altitude varies from 455 to 6735 m. The total area of the district is 6401 km² (Figure 1). Zanskar and Himalayas are high mountain ranges that enclose valleys of Sutlej, Spiti, Baspa, and their tributaries. The slopes of the region are covered with thick wood, orchards, fields, and picturesque hamlets. Temperate climate is common due to its high elevation (455–6735 m), with long winters from October to May, and short summers from June to September. The lower part (Satluj and Baspa valley) of the district receives monsoon rain. The upper areas above the Reckong Peo of the valleys fall mainly in the rain shadow area and considered as arid region similar to the Tibet. Alpine species such as juniper, pine, fir, cypress, and rhododendron can be found at the elevations between 3500 and 5000 m. Oak, chestnut, maple, birch, alder, magnolia, apple, and apricot are temperate trees found at lower altitude. According to the 2011 census, Kinnaur district has a population of 84,298, roughly equal to the nation of Andorra and population density was only 13 person/km².

Figure 1. Study area district Kinnaur, Upper Satluj Basin, Himachal Pradesh.

3. Methodology

It was a selective indicator’s assessment-based approach in which we score every indicator on the basis of its characteristics and visible counted impact in terms of physical loss on selected indicators. The physical loss (affected area) was calculated in terms of landslides, floods, and construction activities, which was extracted from the satellite images and validated through ground survey (Table 1). Total 13 indicators (hydropower project types, slope, aspect, slope profile, relative relief, land use land cover, curvature, channel morphology, lithology, landslide, flood, earthquakes, and soil texture) were selected; the selective indicators have high importance in study region. The presence of these indicators also determined the vulnerability and risk of a landscape. However, these indicators have heterogeneity, but their evaluation decided the degree of vulnerability and risk. The vulnerability assessments were done in terms of low, moderate, high, and very high. Finally, the low vulnerability is explained as a less affected area between 0 and 0.05 km², investigation hydroelectric projects (no construction activities), gentle slope, summital convexity, high relative relief >4000 m (in-habitant area), low curvature value, flat slope aspect, low channel gradient, low stream frequency, low bifurcation ratio, hard rock strata: quartzite, grey shale, and hard rock. Moderate vulnerability of region is explained as a affected area between 1 and 5 km², moderate slope, slope profile of rectilinear section, high relative relief >4000 m, affected land use of barren land and waterbodies, moderate curvature value, less affected slope aspect of north, north east, east, northwest, moderate channel gradient, moderate stream frequency, moderate drainage density, moderate frequency earthquake magnitude (5.0), hard rock strata: quartzite, basic volcanic, sandstone, and less affected soil texture of medium type. High vulnerability was explained as where affected area varies 5–10 km², under construction type hydroelectric projects, moderate curvature, steep slope, basal convexity of slope profile, high relative relief of 2000–4000 m, high degradation of land use such as settlements, agricultural, and forestland. The high curvature value, highly affected slope aspect of south east, high bifurcation ratio, high stream frequency, weak lithology, poor soil texture, and high earthquake magnitude >6 were some responsible factors for high vulnerability. The very high vulnerability was explained on the basis of some observed parameters: very highly affected/degraded area of <10 km², very steep slope to vertical slope, free face, R_R of 1244–2000 m, highly valuable land use land cover such as settlements, agricultural land, forest cover, high curvature value, south and southeast aspect of slope, high channel slope gradient, (S_f) high stream frequency, drainage density (D_d), high bifurcation ratio (R_b), very weak lithology of limestone, siltstone, shale rocky/bad land, and soil of coarse texture. The geographic information system (GIS) and remote sensing were the major tools, which were used to highlight the vulnerability through digital map and index. Finally, all indicators’ impacts were recorded and observed during field visit and impact marked as value “1” and no impact marked as “0.” The landscape vulnerability area is also explained on the basis of parameters characteristics. Finally value means (x⁻) were taken and hazards vulnerability and high value of indicators classes were identified in terms of comprehensive vulnerability. All selected indicators were analysed and geo-processed in GIS environment. The overlay analysis was done and final physical landscape map was prepared, which indicated a highly vulnerable area (Figure 7) (Tables 1 and 2). Indicators analysis process and their data sources were well explained in Table 1.

Table 1. Parameters selection, data sources, and analysis.

Parameters	Data sources	Analysis Process
1. Hydropower projects	Field observation; GPS, Google Earth	First, the HEPs were categorized into four types on the basis of previous EIA study, then the location of hydroelectric projects were collected through field survey and the buffers of 1 km^2 were created from the location of dam sites and affected area was identified in terms of landslide, flood, earthquake, and construction activities. The satellite imageries of Landsat ETM^+ 30 m and Digital Glove were acquired. Validation was done through people interaction and field-based survey
2. Slope	DEM 23 m; www.usgs.com	The slope analysis of area was done on the basis of DEM at the resolution of 30 m, Young classification was used to reclassify the raster data of slope into six types. The slope map was generated and shape files of affected area were overlapped. After that, pixels were counted that how many pixels of different slope types got affected and counted pixel values were converted into km^2
3. Slope profile	DEM 23 m; Google Earth	Slope profile classification was done on the basis of Wood (1942) classification, after that, the slope profile was identified in the field. Terrain analysis and measurement were done in Google Earth. Then the raster file overplayed on the slope profile and pixels were counted. The loss area was counted on different identified segments of slope
4. Relative relief	DEM 23 m; Habitation points	Relative relief of study area was calculated through the DEM and classified into three categories after that the habitation, degraded area layers were overplayed. The affected area was measured in all three classes of relief after that the vulnerability analysis was done and relief-based vulnerability map was generated
5. LULC	Landsat7 (ETM^+) (resolution 30 m)	Land-use land-cover study was classified into eight classes with the help of satellite imageries at the resolution of 23 m. The satellite imageries were geo-processed in Erdas Imagine 2018 and area were calculated in square kilometres. The raster layers were overlapped on land use land cover and affected area of the land use land covered was measured with the help of ArcGIS 10.3
6. Curvature	DEM 23 m	Slope curvature was calculated through ArcGIS; simply pixel values were classified into concavity and convexity and affected area pixels were counted on the concavity and convexity of surface curvature
7. Aspect	DEM 2 3m	Slope aspect was classified into eight classes on the basis of eight cardinal direction points. In ArcGIS, the surface analysis tool was used and slope aspect was generated. The affected area layers were overlapped and affected pixels were counted on every classified aspect
8. Channel morphology	DEM 2 3m Topographic map	Channel slope gradient identification was done in Global mapper 11, the channel slope gradient measured in degree
9. Morphometric	Google Earth DEM 23 m Topographic map 1:50,000	The streams of different order were digitized on the Google Earth image, after that, the Horton morphometric-based analysis was done and stream frequency, stream density, and bifurcation ratio were calculated. Flood vulnerability assessment was done on the basis of different stream orders. The buffer of 30 m was taken from the different stream orders, and in this buffer, the total affected area was calculated. The flood incidences locations were also identified in field. The points were also overlapped on the stream orders area and the flood-affected area was calculated then flood vulnerable area was extracted
10. Lithology	GSI map Topographic map 1:50,000	Lithology map was prepared on the basis of secondary Geological Survey India map of Himalayan region at the scale of 1:250,000 and geological structures were identified from the secondary map. The degraded area layer was overlapped and lithology types were also identified through GPS survey
11. Soil texture	Soil map(NBSS) Topographic map 1:50,000	Soil texture map was prepared on the basis of prepublished map of NBSS. This map geo-referenced and soil texture types digitized, soil sampling points were taken in the field and then overplayed on the map and soil texture area was validated. After that, the affected areas layered were overplayed and affected area under different texture was calculated
12. Landslide	Field observation, Google Earth, satellite images	Landslide analysis was done on the basis of field survey and visual interpretation method. The landslide incidence points were collected in the field through GPS and affected area was digitized through Google Earth and with Landsat7 (ETM^+) satellite image
13. Earthquakes	Field observation, IRIS, Google Earth, satellite images	The earthquakes studies were done on the basis of secondary data, the previous earthquake incidence points were collected. The geological element (Fault, MBF, and MCT) was identified through the lithological and geomorphological maps. The buffers of 1 km from the epicentre were taken for study and all degraded areas were measured. The people perception method was also taken to validate the data
Compressive vulnerability	Overlay analysis was done by adding vector and raster layers	Finally all vulnerability maps were prepared and then all vector and raster data were superimposed on each other and vulnerable area identified in the study region. The vulnerability indexes were generated on the basis of calculated/observed value for each dependent parameters and independent parameters

NBSS: National Bureau of Soil Science.

Table 2. Vulnerability assessment on the physical landscape on the basis of selective indicators.

Parameters	Classification of parameters	Affected area (km^2)	Indicator natural processes as anthropogenic					Explanation of the parameter classes	Empirical study
			Landslide	Flood	Earthquakes	Hydroelectric projects	Vulnerability
1. Hydropower projects	Under construction	13.87	1	1	1	1	VH	Under construction project has high-degree loss of physical landscape as comparison to other type of categories	Lata et al. (2017) and Gupta and Sah (2008a)
	Commissioned	3.63	0	0	0	0	L
	Obtaining clearance	0.89	0	0	0	0	L
	Under investigation	0.81	0	0	0	0	L
2. Slope	Gentle slope (0^°–5^°)	3.02	0	1	0	0	L	In mountainous region, the high degree of slope >30° has high vulnerability, which triggers the incidence of landslides, avalanches slope failure, and impact of earthquake is high on high degree of slope	Webb et al. (2011), Lee et al. (2004), and Slosson & Krohn (1982)
	Moderate slope (5^°–10^°)	6.24	1	1	0	0	M
	M. steep slope (10^°–18^°)	18.56	1	0	0	1	H
	Steep slope (18^°–30^°)	10.27	1	0	1	1	VH
	V. steep slope (30^°–45^°)	9.71	1	0	1	1	VH
	Vertical slope (45^°–90^°)	7.91	1	0	1	1	VH
3. Slope profile	Summital convexity	8.16	1	0	1	0	L	Landscape degradation is high on free face profile and basal convexity, gravity pulls snow, soil mass, rock mass into ground or lower layers	Schoeneberger, Wysocki, and Benham (2012) and Wood (1942)
	Rectilinear section	9.15	1	0	1	0	M
	Free face	11.15	1	0	1	0	H
	Basal convexity	14.16	1	1	1	1	VH
4. Relative relief (RR)	RR 1244–2000	56.17	0	1	1	1	VH	High degree of slope and higher relative relief will be vulnerable and is adroit to physical loss	Hooijer, Klijn, Pedroli, and Van Os (2004)
	RR 2000–4000	27.98	0	0	0	0	H
	>RR 4000	10.9	0	0	0	0	L
5. LULC	Settlements	0.31	1	1	1	0	H	The loss of settlements, agriculture land, forestland, and barren land is high in region of high degree slope/high relative relief because the hazards (flood, landslide, avalanches, earthquakes, and anthropogenic) impact intensity is always high. Land-use changes have a major effect on large floods in large catchment areas of river	Disse & Engel (2001), Hooijer et al. (2004), and Pinter, Van der Ploeg, Schweigert, and Hoefer (2006)
	Agricultural land	1.89	1	1	0	1	VH
	Forest cover	7.52	1	1	0	1	VH
	Barren/Wasteland	38.57	1	1	0	0	H
	Grass/Grazing	6.28	1	1	0	1	H
	Scrubland	5.97	1	1	0	1	L
	Waterbodies	0.96	1	1	0	0	L
	Snow and glacier	14.2	0	0	0	0	L
6. Curvature	Concavity (97.81)	16.6	1	0	0	0	L	High convexity triggers more physical losses in hilly to mountainous region	MacMillan, Pettapiece, Nolan, and Goddard (2000)
	Convexity (−100.4)	59.1	1	1	1	1	VH
7. Slope aspect	Flat	0.01	0	0	1	0	L	A reasonable explanation of this pattern may be that slope aspect affects the density of shallow debris slides by limiting the development and thickness of drier slopes	Dearman and Fookes (1974)
	North	2.11	0	0	0	1	L
	Northeast	3.96	0	0	0	1	L
	East	3.72	1	0	0	0	L
	Southeast	15.03	1	1	1	1	VH
	South	17.08	1	1	1	1	VH
	Southwest	10.37	1	0	0	1	VH
	West	10.26	1	0	1	1	H
	Northwest	3.71	1	0	0	0	L
8. Channel morphology	Channel slope gradients 13.45	12.11	0	1	0	1	M	High channel slope gradient is directly proposal to soil erosion, high flood impacts and encourage hydro development	Gupta and Sah (2008b) and Ligon et al. (1995)
9. Morphometry	(Sf) Stream frequency 4/0 (Dd) Drainage density4.2 (Rb) Bifurcation ratio 1.63	14.64	1	1	0	1	H	Medium drainage, density, frequency, and high bifurcation ratio indicate high degree of physical loss and risk	Horton (1945))
10. Lithology	Pt1 regionally metamorphosed	3.98	0	0	0	0	L	In the case of limestone, dolomite, slate terrain, the effects of weathering can be very severe, due to a combination of physical alteration and limestone solution. The latter, in particular, may play a significant role in enlarging joints and fractures, and in favouring detachment of rocks along the outer ridges. It may happen prior to some seismically induced rock falls	Dearman & Fookes (1974), Kowalski (1991) and Zellmer (1987) Palmström (1995) and Selmer-Olsen (1971)
	Pt3e greenish grey sandstone	18.84	1	0	0	1	VH
	Y granite and granitoid	1.39	0	0	0	0	L
	Pg3o boulder conglomerate, sandstone, shale, and clay	0.12	1	1	0	0	L
	OC limestone, siltstone, and shale	7.36	1	0	0	1	H
	Pt23 slate, phyllite, quartzite, and grey shale	0.36	1	0	0	0	L
	Pt2 ortho quartzite, basic volcanic, and limestone/dolomite	0.39	1	0	0	1	L
11. Soil texture	Coarse texture	18.75	0	0	0	0	VH	Coarse texture soil types have high erosional capacity in cool temperature with humidity fine texture soil has high vulnerability type has dry, extremely cold condition	Jenny (1941) and NRSA (1997)
	Fine texture	14.35	0	0	0	1	H
	Medium texture	11.97	1	0	0	1	L
	Rocky/Bad land	36.47	1	0	0	1	VH
	Total vulnerability mean (X)	11.64	0.67	0.36	0.30	0.55

Explanation: “0” indicates “not observed impacts,” “1” indicates “adverse impacts.” L: Low vulnerability; M: medium vulnerability; H: high vulnerability; VH: very high vulnerability; LULC: land use land cover.

4. Results and discussion

4.1. Vulnerability due to physical aspects

4.1.1. Relative relief (R_R)

Melton (1958) relative relief is used for the overall assessment of morphological characteristics of terrain and degree of dissection. The relative relief of study area varies from R_R = 1244 to 6755 m. The dissection value varies between 0 to 1 and increasing value from 0 to 1 indicates the high degree of dissection/erosion of basin. Dissection index (DI = R_R/A_R) has high value (0.64–0.87) which made cleared that basin has high vulnerability as a mass movement, soil erosion, and flood (Savindra, 2004). The hazards incidences such as landslides, floods, earthquakes, avalanches, and maximum human habitation were recorded under the elevation of 1244–2000 m. The maximum counted loss (56.17 km²) was recorded under the relative relief of 1244–2000 m, which indicates high vulnerability and risk. The probability of landscape and human settlement loss was very high from 2000 to 4000 m. The landscape vulnerability recorded low above 4000 m (Figure 2(a) and Table 2).

Figure 2. Relative relief and slope-based vulnerability.

4.1.2. Slope

Slope is one of the most important parameters from stability consideration viewpoint (Lee, Choi, & Min, 2004; Webb, Yin, & Harrison, 2011). The highest affected area (18.56 km²) of physical landscape was recorded under the slope types of moderate steep slope (100–180) (Young, 1964). The landslide impact was highly observed on moderate slope to vertical slope in terms of slump, slide, subsidence, fall, and crap. Flood impact was high on different segments of slope like gentle slope to moderate steep slope. The vulnerability of physical landscape is very high on slope segments of steep slope to vertical slope. Very less affected area (6.24 km²) was recorded under the gentle slope (00–50). Maximum settlement concentrations were found on the gentle to moderate type of slope. Tangling, Apka, Lippa, Asrang, Chakra, Spillo, Poo, Khab, and Nako village areas had very steep slope, which had poor soil texture, less vegetation covered, high gully erosion, and dry climate condition. People of the study region also revealed that the loss of vegetation due to anthropogenic activities (road and dam construction) was common. Slope stability was also affected due to the anthropogenic activities (road and dam construction) in the area of Karcham Wangtoo. Gentle slope to moderate slope was found in the region of Chaura, burang, and Wangtu. Above the region of Yashang Dhar, Punag Khas, Urni, Chooling, Merru Khas, Rurag, Kibla, Karcham, and Rali have moderate steep to steep slope where forest cover is moderately sparse above the Rali Ranrang, Shongtong, Barang, Tangling, Kalpa, Pangi, Khadura, Akpa, Khab having very steep slope with sparse vegetation. The highest vulnerability was obtained by the moderate steep slope, high vulnerability on steep slope, while gentle having low vulnerability (Figure 2(b) and Table 2).

4.1.3. Slope profile

The distinctive segments such as profile are called slope elements or slope segments (Savindra, 2004a). These types of slopes profile (summital convexity, basal concavity, rectilinear section, and free face of slope profile) are commonly found in the Satluj valley (Melton, 1958a). Free face slope profile cliff bare rock vertical slope profile was commonly observed in the basin Reckong Peo to Khab. The concave segments of the slope profile were commonly affected by the mass movement. Talus accumulation was commonly observed at this segment of the slope profile. The moderate-to-high vulnerability was found on all segments of the slope profile. Forest area losses were very commonly observed on the free face segments of the slope. The loss of river morphology was found on basal concavity. Human property’s loss was found on summital concavity, rectilinear, and basal concavity. The flood and soil erosion highly affected the basal concavity segments of the slope profile. High and very high vulnerability was found under the basal concavity (Table 2).

4.1.4. Slope aspect

The slope aspect category of north (901.88 km²) has high area, followed by southwest (896.14 km²), east (889.68 km²), and northeast (889.29 km²). The lowest area was recorded under the slope aspect of northwest (820.44 km²). The slope aspect is affected by the sunray angle and all geomorphic processes controlled by solar energy, which is the main driving force (Dearman & Fookes, 1974). In general, the slope aspect influenced the distribution and density of mass movement by controlling the concentration of soil moisture or orientation of tectonic fracture. In northern hemisphere above 33 latitudes, the maximum sunny area was found under the south, southwest, and southeast aspects of a slope. The south (17.08 km²), southeast (15.03 km²), and southwest (3.72 km²) slope aspects have the highest area under the landslide, soil erosion, barren and wasteland; southeast (15.03 km²) and south (17.08 km²) have very high vulnerability and risk; northeast, flat, north, northwest have a low vulnerability (Table 2).

4.1.5. Curvature

The profile curvature affects the acceleration and deceleration of flow across the surface. A negative value (−10.415) indicates that the surface was upwardly convex at that cell, and flow will be decelerated. A positive profile (10.415) indicates that the surface was upwardly concave at that cell, and the flow will be accelerated (Rautelal & Lakhera, 2000). Talus accumulation was very common in the concave segments of the slope. Concavity indicates the superiority of concavity and rugged surface. The convexity of the slope has a highly affected area under the mass movement/soil erosion/avalanches, which indicates high vulnerability. The convexity surface area had highly affected area of 59.11 km². The risk factor of human settlements was very high at the convexity segments of the slope (Table 2).

4.1.6. Lithology

This was generalized and modified after preparing the geological map of the Himalaya at 1:1000,000 (Plate-1) by Geological Survey of India, Government of India (1989) (Melton, 1958). In Satluj basin of district Kinnaur, the maximum area 1866.13 km² (29%) was covered under the geology types of Pt3e: greenish grey sandstone, quartzite, grey and dark shale–sandstone, a band of limestone, and phosphorite. The affected area (3.98 km²) was recorded under the lithology type of Pt1 (regionally metamorphosed katazonalmeta sediments), while the lowest area (0.12 km²) was observed under the Pg3o (boulder conglomerate, sandstone, shale, clay, soft sandstone, and shale reddish and sandstone), occupying total 0.01 km² under the Pt23 lithology categories (slate, phyllite, quartzite, grey shale, siltstone, limestone, gypsum, metamorphosed in proximity of granite) which have area 0.36 km² and Pt3e had highest affected area (18.84 km²) of greenish grey sandstone, highly degraded under the process of landslides and avalanches. The Pt1, Pt23, Pt2, and Pg3 have a low vulnerability (Kumar, Gupta, Jamir, & Chattoraj, 2018). The forest area of lithology type limestone, siltstone, shale was highly affected due to the landslide, flood, and soil erosion (Gupta, 2003) (Figure 3(a) and Table 2).

Figure 3. Vulnerability of soil texture and lithology.

4.1.7. Soil texture

Coarse, medium fine, and very fine categories of soil texture were found in the study zones. The study region had maximum area under the very fine texture (5135.4 km²) 81%, followed by fine texture (579.71 km²) 9%, medium texture (455.71 km²) 7%, coarse texture (158.16 km²) 2%, and rocky land had (36.12 km²) 1%. The very fine texture area was highly vulnerable for the soil erosion and mass movements in dry temperate (Jenny, 1941). The sandy soil had low infiltration of water, above the Karcham Wangtoo area which was barren having less vegetation cover. The soil was sandy clay and sandy loamy. The water retention capacity was very low and soil erosion was high due to wind, where an area was well covered with the vegetation and water infiltration rate was very high (NRSA, 1997). But in the lower zone, the climate is humid; this region has sandy loamy soil and good moist salty soil. The highest vulnerable area 36.47 km² was found under the rocky/bad land type topography and vulnerability was recorded high. However, fine texture type soil had high degraded area (14.35 km²) due to landslides and avalanches which had a very high vulnerability (Figure 3(b) and Table 2).

4.1.8. Morphometric

The erosion work of the river depends on a channel gradient, a volume of water, velocity, water discharge, and river load (Horton, 1945). The high channel slope gradient has an adverse and supportive impact on the river morphology, soil erosion, and flood (Figure 4). The channel sinuosity was tortuous, irregular, and slope gradient was high. Exposed bedrock were commonly seen in the upper basin of Satluj. The stream frequency was moderate (S_f = 4). Horton (1945) defined drainage as the ratio of total length of all stream segments in a given drainage basin to the total area of that basin as follows: D_d = L_k/A_k, where L_k indicates total length of all stream segments of the basin and A_k indicates the total area of the basin. Drainage density was found moderate (D_d); the bifurcation ratio (R_b) relates to the branching pattern of the drainage network which is defined as the ratio of a number of streams of a given order (N_u) to the number of streams of the next order (N_u+1). It is expressed with the equation as R_b = N_u/N_u+1 (Giusti & Schneider, 1965; Horton, 1945). Bifurcation ratio within the present region tends to decrease with increasing stream orders (Horton, 1945). A high value (1.79) of bifurcation ratio indicates that the region was very much dissected and very sensitive to physical loss (Singh, 2004). The low values of bifurcation ratio indicate that the Satluj basin had very controlled bedrock strata. Trellised drainage pattern was found because here stream pattern was well adjusted to the regional slope and geological structure (folds, faults) (Smith & Pain, 2009). Many streams were developed on both the flanks of the ridges. Few parts of region have hard exposed bedrocks, which control the development of the drainage network. High drainage density, high drainage frequency, and high bifurcation ratio indicate high vulnerability in terms of physical landscape loss in the upper basin of Satluj in district Kinnaur (Table 2).

Figure 4. Channel slope gradient of river upper Satluj basin in district Kinnaur.

5. Land use and land cover

Land-use land cover (LULC) is also considered to be the major factor in influencing the mass movement. Barren and sparsely vegetated areas are prone to weathering and slope instability (Anbalagan, 1992; Raghuvanshi, Ibrahim, & Ayalew, 2014; Raghuvanshi, Negassa, & Kala, 2015). The topo-sheet 2010 survey of India, at the scale of 1:50,000 data, were used to know the general status of land use in the Satluj basin of district Kinnaur (open series). The maximum area was found under the categories of snow-cover 2490 km² (39%) and wasteland 2030.63 km² (31%). However, waterbodies (5.3 km²), built up (1.54 km²), and agriculture (52.8 km²) land had very low area. Forest cover was only 484.63 km² (8%) area and grassland cover was 1383.1 km² (20.58%) of total area. The forest cover was only 8%, according to the forest policy, the minimum 33% area must be covered under the forest for the healthy landscape. The percentage of wasteland and snow-covered land was very high, which stands 70%. The 1.89 km² of agriculture land was very low about 1.2% which indicates the tuff terrain and low possibility of agriculture. The built-up area covered only 0.31 km². The landscape has unhealthy land cover and which has high vulnerability of physical loss.

Along the Satluj valley area, Khani Dhar, Burang, Bara Khamba, and Gharsu Nathpa Jhakri areas had better cover of grassland and less recorded landslides. Because the share stress not reduces during the raining season and slope stability was maintained. But in the region of Rarang, Wangtu, Tapri, Chagaon, Cholling, and Karcham area had less vegetation and soil texture/rock strata was commonly exposed in this area. During rainy season, the surface run-off was very high and share stress reduces which triggered the landslides. It was made clear from the field survey that the area of Karcham Wangtoo, Powari, Tangling, Kalpa, Bokta, Pangi, Rarang Khas, Akpa, Rispa, Jangi, Spillo, and Nako had dry condition, very fine soil texture, and broken land topography. The soil compactness was reduced during the snowfall and incidences of landslides increased. About 66.79 km² area was damaged due to landslides and construction activities, which include 1.89 km² of agriculture land, 7.52 km² forestland, 38.37 km² wasteland, 6.28 km² grassland, 1.97 km² scrubland, and 5.97 km² under the waterbodies. The land degradation was also high due to blasting: road widening and construction activities (hydropower projects). About 0.31 km² land of human settlements was damaged. High vulnerability was found under the area of agriculture, forest, and barren land. Low vulnerability was found in the area of scrubland, waterbodies, and snow glacier areas (Table 2).

5.1. Landslide

The economic and loss of life due to landslides were considerably increased in the last century, and most of the landslides are due to global climate change, such as El Niño and human activities (Runqiu & Weile, 2011). There is also high confidence that changes in heavy precipitation will affect landslides in some regions (IPCC, 2012). Precipitation pattern was different in the study region which was influenced by the topographic variance (434–6448 m). Isohyets study made it clear that the annual isohyets varied from 100 to 1400 mm (Singh & Jain, 2002). Annual rainfall in this basin decreased from the lesser to the Greater Himalayan range (Singh & Kumar, 1997). Rotational landslide, transitional, rockfall, topple fall, and debris fall debris flow were commonly found in the noticed places of Nathpa Jhakri, Shongtong, Spillo, Apka Khas. The landslides were due to snowfall which were very common above the Reckong Peo to Khab. Soil creep was common in an upper region because of unconsolidated types of very fine soil texture. Freeze and thaw weathering were very common in the upper areas of Kinnaur districts: Purbani, Ribba, Rarang, Khadura, Apka, Kwangi, Spillo, Nasarg, Samdayan, Puh, Dublin, Chango, and Nako. Slope excavation due to dam construction and road construction was commonly observed in Karcham Wangtoo, Shongtong Reckong Peo, Pangi, Kasang, and Nathpa Jhakri. Total 1541 incidences of landslides were observed out of which 1396 are small (0–0.2 km²), 138 are medium (0.2–0.5 km²), and 7 fall under the category of a large landslide (>0.5 km²). The landslide occurrences were very high on the moderate steep and steep slopes. The highest damaged area (12.13 km²) was found under the small categories of landslides (Gupta, Bartarya, Virdi, 2003; Gupta, Sah, Virdi, 1993). However, the earthquake average intensity in the upper basin varies from 5.0 to 5.5, which was sufficient to generate small landslides. People also believed that landslides were common at the base concavity and steep slope of the basin. According to the Pacific north seismic network, earthquakes of magnitude 4.0 and greater have been known to trigger landslides. The vulnerabilities of the physical landscape and human property were very high (Figure 5(a) and Table 2).

Figure 5. Landslide incidences and flood-based vulnerability.

5.2. Flood

Himalayan region due to the construction activities of dams and road-added sediments had adverse impact on the river system (Ligon, Dietrich, & Trush, 1995). In Satluj basin, the most disastrous floods were experienced in 1993, 1995, 1997, 2000, 2005, 2007, and 2013 (Gupta & Sah, 2008b). The unusual high discharge was observed during the flash floods. The water level in a river had risen to about 15–20 m from normal level and its discharge value was as high as 10–12 times more than a normal discharge during the specified period (Gupta & Sah, 2008b). The lowest area is recorded under the first-order stream which affects the lowest area about 9.77 km². The first-order streams, nallas and khads, have not possible impacts on the risk elements and no direct indirect relation with the hazards like earthquakes and avalanches. But small rill formation and soil erosion were noticed in a field. Second- and third-order streams affect only 12.88 km² area and flood impacts high during the rainy season. Nallas and khads were responsible for the small landslides and soil erosion. The loss of forest and agriculture land was observed during the field observation in the area of third (like Baspa, Shongtong, Karcham, and Bhaba) and fourth (Satluj river) large stream orders. The very high flood vulnerable areas were found within the buffer of 100 m. The high stream order had a highly vulnerable area under the physical loss because of high cross section and steep channel gradient (150). The vulnerability and risk factor were very high at the mainstream and low at the first-stream orders (Figure 5(b) and Table 2).

5.3. Earthquake occurrences

However, the stronger earthquakes with greater magnitudes of MS7 might yield an even more pronounced effect on the overall sediment flux (Hovius, Meunier, & Lin, 2011). Several studies have also elaborated the link between earthquakes, landslides, and fluvial sediment transport in the seismically active mountain belts (Dadson, Hovius, & Dade, 2003; Meunier, Hovius, & Haines, 2008). The magnitude of earthquake incidences in the Kinnaur district magnitude varies from low to moderate in intensity level 4.0–6.1. The average magnitude of the basin in Kinnaur district was 4.5. As per the mention guideline of  the Mercalli intensity scale the magnitude of earthquake explain that feel in intensity of IV-V, the tremors of earthquake can felt easily, some awakened, disturb window, door disturb, wall making cracking sound, sensation like heavy truck strictly building, unstable object overturned, Pendulum clock may stop. Earthquake magnitudes affect the physical mass movement behaviour. The incidences of landslides increased at the magnitude of 5.0. The active landslide is commonly observed in the Urni and Shudarang Dhaku villages. People also explained that earthquake impacts (cracks in a wall, landslides) were clearly seen in the study region. People also said that geological setting of the area was affected with blasting and construction activities. Due to the earthquakes, the small landslides were generated and property losses were seen in terms of agriculture land and settlements. The problems of soil erosion and avalanches were noticed due to its indirect causes. But huge loss of landscape was noticed during the time of landslides; these impacts were seen and observed by people in the village Urni and Shudarang Dhaku. Sixty villages with population of 25,317 persons were found under the earthquake vulnerability (Figure 6(a) and Table 2).

Figure 6. Earthquake-based vulnerability and project-affected population vulnerability.

6. Vulnerability due to anthropogenic factor

6.1. Hydroelectric projects

In the Satluj basin, a hydroelectric project is one of the main construction activities, which is responsible for landscape destruction. Slope instability was common in and surrounding area of hydropower projects (CEIA, 2014). The topsoil, forestland, threatened faunal species, wildlife sanctuary eco-sensitive zone, and 228 villages were influenced under the hydroelectric projects. Increase in landslide incidences, flash floods, river morphological changes, and water quality deterioration and reduction in agricultural horticultural production, forest degradation, land degradations, inadequate compensation due to construction activities, damage to human health due to dust during construction activities, damage to houses due to blasting and tunnelling activities, and some adverse impacts were noticed within the project-affected area (Lata, Herojeet, & Dolma, 2017; Wang, Du, & Chen, 2012). The four types of hydroelectric projects (Under construction, Obtaining clearance, Commissioned and Under investigation) were identified in the upper Satluj basin of the district Kinnaur. It was made clear that maximum affected areas of landslides were found under the categories of under construction type (13.87 km²). Commissioned type hydroelectrics were less affected. It was observed that under construction hydroelectric project has high vulnerability and risk. Due to these construction activities, the incidences of landslide and flood have been increased. According to the people of the study area, the loss of river morphology and forest (7.82 km²) had been observed (Figure 6(b) and Table 2). Land degradation (75.75 km²) and loss of agriculture land (1.89 km²) were found under the type of construction hydroelectric projects. High vulnerability was found within the buffer of under construction hydroelectric projects.

7. People perception on geophysical setting and sustainable planning of landscape

The questionnaire survey was conducted in the study region to know the status of the physical landscape (Annexure 1). During this survey, the people perception were taken on the different selected parameters. Total 51 people told that hydropower were a major contributor to the loss of physical landscape. 63 respondents believed that high degree of the slope, high local relief are supporting factor for the landslides and soil erosion. 59 respondents were asserted that the maximum area wasfall under the barren land. The forest cover was low. However, 33 respondents also believed that the incidences of earthquakes were high, but actual losses were not observed. During the interview of 61 respondents, it was found that the incidences of flood were very common and people also told that River Satluj was known for its worst floods in 1993, 1995, 1997, 2000, 2005, 2007, 2013, and 2018. The channel morphology was also affected with the construction activities of hydroelectric projects; 57 people also told that there was not proper dumping site and excavated material was directly dumped into the river, which had a adverse impact on the river morphology. Some old experienced group of people (66) believed that siltstone, dolomite topography-based region had incidences of creep and subsidence. People also told that the soil of the region was easily eroded because of less vegetation where soil texture was very fine, loss of soil was very common, and dust problem was very common, which have a serious impact on the vegetation and horticulture. During the rainy season, the mudflow and rock slides were very common from the region Tapri to Reckong Peo (Figure 8).

Figure 7. (a) Lose unconsolidated soil texture near to Akpa village, (b) large landslide in Bara Khamba, (c) muck dumping at the river Satluj side of Karcham Wangtoo HEP, (d) drilling the hill at 100 mw Tidong Hydro., (e) Urni landslide, (f) large landslide at Rekong Peo, (g) questionnaire survey at village Kwangi, (h) houses affected by the landslide at village Nigulseri, (i) strategic environmental assessment meeting at deputy commissioner office Rekong Peo, Kinnaur, on November 2014, (j) construction site of Tidong project, (k) dumping of muck along the river Sainj by Parbati HEP, (l) house cracks in Yulla village because of the tunnel construction of Karcham Wangtoo HEP.

Figure 8. Comprehensive vulnerability.

7.1. Comprehensive vulnerability

Physical landscape losses were highly recorded under the construction type hydroelectric projects (13.87 km²). The high degree of slope >30 had high vulnerability (17.62 km²). Landscape degradation was high on free face profile and basal convexity because gravity pull was very high. The loss of settlements (0.31 km²), agriculture land (1.89 km²), forestland (7.52 km²), and barren land (38.7 km²) was high. The impact intensity of hazards (landslide, flood, anthropogenic activities, and earthquakes) was high. The LULC status plays a major effect on large floods catchment areas of river because the ratio of barren and forestland was very low. High convexity triggered more soil erosion in valley (14.16 km²). A slope aspect affects the density of slides, southeast and south slope aspects received more physical losses (32.11 km²). High channel slope gradient area has high soil erosion. Medium drainage density, medium frequency, and high bifurcation ratio indicate high degree of physical loss and high risk. In the case of limestone, dolomite, and slate, terrain type lithology was highly affected by weathering. Coarse texture and fine texture soil types had high erosional capacity in such cool temperate region. Overlay analysis was done on the basis of different raster and vector dataset layers. The vulnerability area was identified on every parameter such as slope, slope aspect, slope profile, soil texture, land use land cover, morphometric, lithology, geology, and hazards (earthquakes, flood, and landslide). To sum up all the parameters, finally the physical landscape vulnerability map was generated (Figure 7 and Table 2). Very high vulnerability areas (268.07 km²) of the Satluj basin basically fall under the main valley of Satluj. The maximum concentrations (91.1%) of rural settlements were found in the main valley. The 346.45 km² area was found under the very high vulnerability. The high vulnerability area was 268.07 km², moderate vulnerable area was 922.83 km², and 4833.13 km² falls under the low vulnerability. According to the vulnerability index, the highest mean values were found under the hazards of landslides 0.67 and anthropogenic activities had 0.55. The lowest impact value was found under the earthquake 0.30 and flood 0.36 (Table 2). Hydropower development was one of the major concerns in the region, which is responsible for huge physical landscape and human property losses. The vulnerability of 60 villages of population 25,317 was found under the very high vulnerability area of 346.45 km².

8. Conclusion

In the whole study, one thing was found out that the region has high vulnerability, because of its complex geophysical setting. For future development, it is very necessary to keep in mind the complex geophysical setting. High degree of slope area must be avoided from excessive construction activities. Slope profile of free face is highly damaged due to blasting activities. High relative relief, weak soil texture, and high morphometric parameters indicate its high vulnerability. The unhealthy land use land cover is more prone to soil erosion, floods, and landslides. The hydropower development should be in sustainable and control manner. The maximum losses of hydropower project and social issues were counted in the main Satluj valley. For the future study, it is very necessary and important that the detailed study on landslides and slope stabilization should be undertaken by well-known research institute or concerned departments. The incidences of hazards (landslides, flood, earthquakes, and anthropogenic) and their impact were highly observed in main Satluj valley. The main valley patch of 346.45 km² along river Satluj from Rampur to Khab received maximum losses. This area has high indicator value, and 90% district population resides and has high pressure of construction activities. It is made clear from the analysis that the excessive and haphazard developmental activities are increasing the risk for local communities over this sensitive landscape.

Acknowledgments

The authors are heartily thankful to the “Director, G.B. Pant National Institute of Himalayan Environment and Sustainable Development, Kosi-Katarmal, Almora, Uttarakhand – 263 647, India” for providing facilities at Himachal Regional Centre of the Institute. Thanks are also due to different stakeholders (local communities, project authorities, and local government) for their constant support and cooperation during field study. We are also thankful to Dr. Kaser (Scientist C, G.B. Pant National Institute of Himalayan Environment and Sustainable Development Vivek Vihar, Itanagar – 791 113, Arunachal Pradesh, India) for providing constructive suggestion.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This research work was part of institutional research in-house project (Strategic Environmental Assessment of Hydroelectric Project), in Satluj Basin, Himachal Pradesh and related to PhD topic Vulnerability assessment of Satluj basin for sustainable development. During the research period, all funding was assisted by the G.B. Pant National Institute of Himalayan Environment and Sustainable development, Mohal, Kullu: 175101 [In house project no. 8].

Annexure 1. Villages under survey in the Satluj basin of district Kinnaur (H.P).

Sr. No.	Villages Name	Altitude (m amsl)	Longitude	Latitude
1	Khab	2769	78.6448	31.8025
2	Morang	2536	78.4502	31.6012
3	Kutang	2460	78.4256	31.6393
4	Akpa	2454	78.3992	31.5858
5	Spello	2347	78.4437	31.6615
6	Chitkul	3429	78.432	31.3514
7	Rakchham	3133	78.3643	31.3801
8	Ribba	2761	78.357	31.5792
9	Pangi	2647	78.2789	31.5908
10	Sangla	2646	78.2654	31.4261
11	Barang	2467	78.2684	31.5059
12	Purvani	2390	78.3018	31.5891
13	Tapri	2385	78.097	31.516
14	Raang	2252	78.2442	31.5151
15	Rali	1917	78.2122	31.4955
16	Punag	1930	78.1031	31.5124
17	Chooling	1849	78.1388	31.523
18	Karcham	1802	78.1766	31.5017
19	Wangtu	1697	78.0159	31.5419