Rendering Lineament Induced Stream alignment in Upper Krishna Basin, India: A Geospatial Approach

Lineaments play an important role in drainage development, stream alignment and groundwater recharge. These are traced as surface or subsurface features. A linear feature that is associated with dislocation and deformation is known as lineament. The present study aimed to identify the in�uence of lineaments extracted from satellite images in the Upper Krishna River basin in India. Remote sensing datasets are considered the best option when using image enhancement techniques for extracting lineaments. Lineament studies are useful for groundwater and mineral exploration and also in the �eld of engineering geology. The availability of high-resolution satellite data and image processing techniques have rendered it further convenient to map lineaments. In the present study, lineaments were extracted from Sentinel–2 images with a resolution of 10 m, using various image processing techniques. A total of 1314 lineaments were extracted from the study area with a total length of 3983.44 km. The analysis of the extracted lineaments revealed that the lineament density was higher in the upper reaches of the basin, where the undulating hilly region is located in the Western Ghats. This �nding implied that these regions have a high structural deformation and a higher groundwater in�ltration potential. Moreover, approximately 15% of the total stream length was observed to be in�uenced by the lineaments. The maximum in�uence of lineaments was observed in the source region. The lineament extraction results of the present study would assist in understanding the geomorphology of this region and the structural control on the streams and groundwater potential zones, particularly as a contribution to water resource management in this region.


Introduction
Lineament extraction was performed manually until the introduction of the digital elevation model (DEM).
However, lineament extraction from hardcopy geological maps or from aerial photo interpretation like observing linear features from the aerial photographs and interpreting visually was a time-consuming process and suffer from a high degree of subjectivity like it needs to be interpreted manually which may create an extra subject matter and it might get obstructed only up to the visible range of spectrum because of human eyesight limitation.The advent of remote sensing technology enabled the extraction of geological features, such as lineament, dike, and fracture zone from satellite images.The advancements in satellite technology, data processing, and data visualization techniques rendered it further convenient to extract lineaments.Lineament extraction and mapping are useful in various elds of scienti c research and in-demand applications such as geographical information system (GIS) and remote sensing (RS) technology.Accordingly, remote sensing data also have wide applicability in geological mapping for various scienti c research elds, such as geomorphology, geology, mineral exploration, etc.
The present study is signi cant in understanding the geological setting and the structural deformation caused by lineaments, which would help to understand the in uence of lineaments on the alignment of the drainage network.The present study is also important for improving the understanding of the level of structural control on landscape evolution in general and stream alignment in particular.Such studies also reveal the level of changes that have occurred in the drainage network due to structural deformation caused by lineaments.

Study Background:
Hopkins mapped lineaments in 1841 and was a pioneer in describing the relationship between lineaments and topographical characteristics (Tiren, 2010).The term lineament was rst used by Hobbs in 1904 to describe a linear characteristic that represented the 'signi cant lines of a landscape'.
Lineaments may include linear, lineation, geo-fracture, suture, mega fracture, and shear zones formed by several tectonic processes and the associated activitie (Rakshit et al., 1985), Lineament magnitude is a scale-dependent parameter it may range in the length from a few centimetres to kilometres and may vary from small as a cleavage in minerals to large as the inter-plate boundaries.The term lineament is, however, most often used for describing Earth's unidirectional features of larger magnitude (Bhave et al., 1989).Lineament patterns might be correlated to the structural, geomorphic events and results of tectonic deformation.A lineament represents the features such as faults, fractures, master joints, axial plane fractures, dyke systems, long and linear lithologies, straight courses of streams, vegetation alignment, or topographic linearity (Basavarajappa et al., 2015).
Until the last decade, researchers used aerial photogrammetry and topographic maps to extract lineaments manually.Since the availability of aerial photos is very limited and the process of extracting lineaments through photogrammetry is tedious, the satellite data such as the Digital Terrain Model (DTM) or high-resolution satellite images are also used when the objective of the study was limited to the In the present study, lineament extraction was performed using remote sensing data and GIS technology to understand the in uence of lineaments on the drainage network in the upper Krishna River basin.The objective of the study was to signify the spatial distribution of the lineaments and their in uence on the drainage network in the study region.

Study Area:
The Upper Krishna River basin is located in the Deccan Volcanic Province of the Maharashtra plateau, on the eastern side of theWestern Ghats.The study area extended from 73°50´ E to 75°05´ E longitudes and 16°02´ N to 18°07´ N latitudes (Fig. 1).The total area of the Upper Krishna River basin is approximately 19331 km 2 .River Krishna and all its tributaries originate in the Western Ghats region, from where these ow in the North-West to South-East direction.The relief variation of the study area is 505 m to 1440 m above the MSL, which indicates steep to moderate relief.The Upper Krishna River basin is characterized by three major physical divisions: i) the Ghats, hills, and plateaus, ii) the foothill zones and iii) the plains.

Geology:
The Upper Krishna River basin is formed on the Deccan trap lava ows of the Late Cretaceous Paleocene period.In certain regions of the river basin, extensive alluvium depositions of the Quaternary period are observed along the main river channel (Bondre et al., 2006).It is in these regions that agricultural practices are dominant.
The broad classi cation of the geology of the Upper Krishna River basin is provided in Table 1 below.

Alluvium
Alluvium formations normally lay over the Deccan basalt and comprise loose to semi-consolidated material, such as sand, silt, gravels, clayey, etc. Certain isolated patches of recent alluvium depositions, which vary from 2 to 20 m in thickness, are also formed along the banks of the Krishna River and its major tributaries (Table 1).In total, there is 2% alluvium soil in the Upper Krishna River basin (GSDA, 1975).

Laterite
Laterite crusts are observed in the study area on the plateau top and the form of tableland with a reddishbrown colour and coarse texture.The laterite formation in the Panchgani region is well exposed (Fig. 2).This has been considered to have been developed on a large plain surface with a gentle gradient in the direction of southerly to southwesterly established on the sequence of the Cretaceous-Eocene Deccan lava ows (Sahasrabudhe and Deshmukh 1981).

Deccan Basalt
The maximum of the study area belongs to the Deccan trap geological formation of the Late Cretaceous-Paleocene period and comprises basalt rock.In the Deccan basalt region, two types of basalt rocks are present -one massive and the other vesicular basalt.In the study area occurrence of vesicular basalt is limited while massive basalts are common.The Deccan Basalt comprises various thicknesses, which are differentiated based on the presence of a red-bole, a weathered layer of basalt covered with dense basalt, and a vesicular basalt layer covered with dense basalt.In the study area, generally NNE-SSW lineaments are observed (Figs. 4 and 5).Similar observations are made by [20] .

Climate:
The climate of the Upper Krishna River basin is of a tropical and sub-tropical monsoon type.The rainfall in the sub-basin region varies from 500 mm to 6208 mm as per IMD rainfall data for the period 1951 to 2015.The average annual rainfall recorded in the Upper Krishna River basin is 1300 mm.The temperature of the study area varies from 14°C in winter to 36°C in summer.The average annual mean temperature in this region is 28°C (IMD, 2020).

Database and Methodology
The study focused mainly on lineament extraction and visualization of the in uence of lineaments on the drainage network of a large river basin area.Therefore, satellite-derived data were required to obtain the output.Remote sensing technology serves as a concrete source for spatial data to be used as an input for GIS and generate a detailed map using other collateral data derived from different sources.A systematic analysis of a satellite image generally involves consideration of the image and terrain elements (Basavarajappa., 2014).In the present study, satellite images and DEM were used for the extraction of lineaments and the drainage network.These datasets were also utilized to visualize the terrain of the region and map the feature such as lineaments.
The details regarding the data and methods used are as below (Fig. 3).

Lineament Extraction:
In previous studies use of aerial photogrammetry was a common approach to identifying and extracting lineaments.Satellite images such as multispectral or digital elevation models and aerial photography are Sentinel 2 Image data were downloaded, and atmospheric correction was performed on the retrieved data.Afterwards, the image data were processed for layer stacking and then converted to raster .tiffle format.The raster .tiffformat les were then processed using the ENVI 5.0 software to generate PCA imageries, which were then exported with 8-bit grey-scale resolution in the GeoTIFF/TIFF le format.The automatic lineament extraction was performed using the PCI Geomatics 2016 software.The line module in the algorithm library was used for processing the nal data product by running the lineament extraction command.The generated lineament map, depicted in Fig. 3, was veri ed using the lineament data available on the Bhuvan portal of NRSC, India.

Lineament Orientation:
The lineaments range in length from a few meters to tens of kilometres.Generally, they appear as a rectilinear feature based on the dip of the structural plane.It has been observed that all the lineament in the study area either meet almost perpendicular at large lengths or intersect each other or are parallel to each other.In total 1314 lineaments are identi ed (Fig. 4).The orientation of maximum lineaments is observed in the direction of NE 45° to SW 225° and minimum orientation in the direction of NW 345° -SE 165° (Fig. 5).

Stream Network Generation:
Water tries to remain in equilibrium and follows the easiest way through the lineaments and fracture zones.In this process, lineaments act as conduits.This causes the water to follow a lineament course that is generally linearly aligned.To better understand the in uence of lineament on stream behaviour, the pattern followed by the streams has to be analyzed, which is achieved by generating a stream network.
In the present study, the stream network was generated using data from ASF-DEM, which had a spatial resolution of 12.5 m (Fig. 4).ASF-DEM provides high-resolution elevation data and therefore, results in better accuracy as well as better surface details in the outcomes of data analysis.The GIS software was then employed to process these data and generate the stream network in the following steps -'DEM lls', ' ow direction', ' ow accumulation', 'stream order', and 'raster calculation.In the 'stream order' step, Strahler's (1952) method of 'stream ordering' was used.Next, the stream order raster was converted to the vector format for the analysis of the linear morphometric properties of the drainage network.

Streams In uenced by Lineaments:
The streams in uenced by lineaments were identi ed using a buffer tool.A buffer of 100 m was rst drawn along the lineaments.The streams located within the buffer limits were considered as streams in uenced by lineaments (Fig. 7A and 7B).These streams were selected and extracted for further analysis.The respective lengths of the in uenced streams were calculated and separated according to the stream order.

Results and Discussion
Lineament mapping and analysis were signi cant components of the present study.The parameters associated with lineaments and the number of streams in uenced by lineaments, along with their respective lengths, are discussed below in detail.

Lineament analysis:
The lineament analysis mainly included the categorization of lineaments based on their respective lengths, lineament density, and in uence.

Lineament Categorization:
A total of 1314 lineaments were identi ed for the study area in the present study.The total length of these lineaments was 3983.44 km.The identi ed lineaments were categorized into three groups based on their length.Among all lineaments identi ed, 998 lineaments were categorized into the dominant group of minor lineaments, for which the length ranged from 0.64 km to 3.55 km, thereby covering a total length of 1983.82km (49.80%).The second dominant group of lineaments was the intermediate lineaments, with a length range of 3.59-8.88km, covering a total length of 1386.94 km (34.82%).A total of 269 lineaments were categorized into this second dominant group.The third group was that of major lineaments, with a length range of 9.31-36.0km and a total length of 612.68 km (15.36%).A total of 47 lineaments were categorized into this third group.The major lineaments are straight and control the stream ow.The intermediate lineaments are abundant along the regional strike.The minor lineaments are higher in number and smaller in magnitude, thereby forming a network across the study area (Table 2; Fig. 4).The orientation of lineaments and their impact on the orientation of streams is also an important outcome of this study.The majority of streams are in uenced by the lineaments oriented in NNE to SSW and WNW to ESE directions.The second important direction of the lineaments that in uenced the streams is NE to SW (Fig. 8 ).Other important direction of stream orientation is NW to SE. 3.3 Analysis of the In uence of Lineaments on the drainage network: The study area was drained by streams of one to seventh order.The total length of all streams in the subbasin region was 11145 km, of which approximately 1642.81 km of stream length of all orders was structurally controlled, i.e., it was in uenced by lineaments (Fig. 7B).Among all the streams in uenced by lineaments, the rst-, second-, and third-order streams received the maximum in uence of (51.04%), (28.95%) and (11.09%) respectively (Fig. 6).This was mainly because of the loss of stream power, because of which the streams were not able to cross the structurally controlled threshold limit set by the lineaments.The streams from the fourth order to the seventh order received the minimum in uence of (5.11%), (3.19%), (0.61%), and (0.003%) respectively.The fourth-, fth-, and sixth-order streams exhibited minor lineament control, while those of the seventh-order exhibited negligible control by lineament (Table 3).This might be because of the increased water volume and stream power.Therefore, it was inferred that rather than following the structural control, these streams followed the slope of the terrain.
The results of the present study distinctly indicated that the maximum in uence of lineaments was exerted on the streams of the rst, second, and third order all of which are located in the source region (Fig. 9).The in uence of lineaments on the streams of the fourth order and higher was relatively lower and the in uence became negligible for the streams of the seventh order.The structural control of lineaments on the streams could be observed distinctly by overlaying the extracted lineaments on the Koyna Dam region using the Google Earth Pro software (Fig. 10).

Conclusion
DEM comprising high-resolution data is an important dataset for analyzing structural landforms.DEM has been proven as a suitable input for lineament extraction.The results of the present study revealed that lineaments guided the ow of streams with low stream power.This was observed mainly for the streams of the lower order, including the streams of the rst and second order, which together accounted for approximately 80% of the total length of the whole drainage network in the study area.The streams of the middle and higher orders were relatively less in uenced by lineaments.The streams of the third, fourth and fth order together accounted for only 20% of the in uenced streams in terms of total length.
In the case of third-order streams, less than 5% of stream length was controlled by lineaments.Negligible control of lineaments was observed on the streams of higher order.This could be attributed to the increased volume of water and steam power in higher-order streams.The water ow and stream power probably overcame the threshold of resistance caused due to the presence of lineaments.Therefore, based on the analysis of lineaments conducted in the present study, it was concluded that the in uence of the structural control set by lineaments was higher on the lower-order streams compared to the higherorder streams.
The present study used image analysis techniques for lineament extraction and the study of the in uence of lineaments on streams.The change in the course of higher-order streams is mainly attributed to the control caused by the lineaments and the subsurface lithology.The structural complexity of a river basin renders it necessary to understand the level of in uence of lineaments.In the present study, the in uence of lineaments on lower-order streams was observed to have manifested mainly in the form of stream routing or alignment according to the orientation of the lineaments.The overall landscape development was also in uenced by the dense network of lineaments.The ndings of the present study are signi cant as they revealed the geological setting and its structural deformation caused by lineaments, which would, in turn, assist in further exploration of the drainage network and also in water resource management.The present study and other similar ones are important for improving the understanding of the level of structural control on landscape evolution in general and stream alignment in particular.Such studies also reveal the changes that have occurred in the drainage network due to structural deformation caused by lineaments.
Page 15/22    The methodology used for generating the stream network of the study area and these steps are usually followed extract drainage network using and also in the delineation of the watersheds.are the steps in the analysis tool.
Page 19/22 The stream order vs. stream length graph illustrates the in uence of lineaments on the length the streams.It is distinctly visible that more streams of the lower order were in by compared to those of the higher order.
geological aspec (Henderson et al. 1996; Chaabouni et al. 2012;Raj et al. 2017; Das and Pardeshi 2018;Das).In the late twentieth century, automatic lineament extraction techniques were developed, which enabled an improved approach to conduct several work (Zlatopolsky 1992;, Majumdar and Bhattacharya 1998;, Mostafa and Bishta 2005;, Costa and Starkey 2001).The automatic extraction process using GIS techniques has enabled the extraction of the geological surfaces and subsurface features and the subsequent integration of the derived information (Das et al., 2018).The major lineaments, faults, and fractures of a region may be identi ed and mapped through visual analysis of satellite images.Since the lineaments are straight and linear in nature, Digital Elevation Model (DEM) and satellite images have been applied widely for the identi cation and extraction of lineaments (Mathew and Ari n, 2018; Das et al 2018).The mapping of the surface features of the Earth has become further convenient with the availability of huge amounts of remote sensing data and several image processing techniques, including the extensive use of various remote sensing databases and the GIS software for data (Qari 1991; Qari and Şen 1994; Chang et al.1998; Leech et al. 2003; Farina et al. 2005 Chandrasekhar et al. 2011; Teixeira et al. 2013; Das and Pardeshi 2018;).Digital Elevation Model (DEM) with the good spatial resolution is the key input for a GIS-based analysis of the lineaments and extraction of drainage networks.Digital Image Processing (DIP) and Geographic information system (GIS) are two important tools that enable a convenient, accurate, and e cient understanding and interpretation of geotectonic activities.

Figure 1 Location
Figure 1

Figure 3 The
Figure 3

Figure 7 A
Figure 7

Table : 1
The age and formation of the geology of the study area.
Source: Geological Survey of India(1975) (Satish 2002;Basavarajappa et al. 2015)ent studies, including those concerned with de ning geological structures and tectonic activities.In several recent studies, DEM has been used for lineament extraction.In other recent studies, researchers have performed lineament extraction to ensure the mapping of the entire set of lineaments available for the study area (Abdullah et al.2010; Tahir et al. 2015; Muhammad and Awdal 2012; Raj et al. 2017; Das et al. 2018).To improve the interpretation and analysis, Digital Image Processing has been used together with various algorithms to enhance the surface or linear features.In the present study, edge detection, threshold setting, curve extraction, and principal component analysis (PCA) were used for image analysis(Satish 2002;Basavarajappa et al. 2015), Edge enhancement is useful for delineating the edges and sharpening the lineaments.Spatial ltering techniques were used to sharpen the images with linear features and edge-highlighting convolution methods were used.Finally, the PCA technique was used in the lineament extraction.

Table 2
Lineament categories are based on stream length.

Table 3
In uence of lineaments on the drainage network in the Upper Krishna River basin.