Abstract
River morphology describes a river's cross-sectional shape, sedimentation, and erosion. The meandering parameters, oxbow formation and decadal land usage land cover (LULC) fluctuations of Barak River were investigated using 21 meandering spans to measure river morphological changes. The decadal meandering parameters were calculated reach-wise and section-wise to characterize river morphological changes. It was observed from the paired t-test that the river width significantly changed during the study period (1990–2020). Strong inter-relationships between the meandering parameters are shown from the regression analysis. The morphological investigation found a reduction in the centerline distance due to variations in the radius of curvature caused by the internal arc's reduction. As a result, the average sinuosity has decreased over time. The current work used SVM and ML techniques for LULC classification, and a comparison of ML and SVM techniques was also done. The SVM technique performs better. The decadal LULC analysis suggests that between 1990 and 2020, the areas of water bodies, forests, and bare land types declined. Whereas, agricultural and settlement areas increased. River morphology is substantially impacted by agriculture and urbanization, particularly in areas where oxbows occur simultaneously, since this work may apply to other similar meandering river management along the alluvial flood plain.
HIGHLIGHTS
River morphology is affected by meandering parameters, land resource changes, and oxbow development.
A minor change in meandering parameters like sinuosity, radius of curvature, etc., can affect river width, migration rate, and morphology. A statistical analysis can estimate the river's meandering parameters.
Oxbow formation influence morphology. Future oxbows can be identified through morphological study, GIS technique.
INTRODUCTION
The flow and sediment regimes combine with the landscape's physiographic features and vegetation cover to give alluvial floodplain rivers their dynamic nature (Ward & Stanford 1995). In the alluvial stretch, the rivers meander frequently and take different courses. Subsequently, the river morphology changes on a spatial and temporal scale. Several variables influence river morphology (Lancaster & Bras 2002). These phenomena represent the hydro-geomorphologic forms caused by the river's lateral movement (Yousefi et al. 2016). The river morphology mainly describes the cross-sectional shape of the river channel and the process under which the river section changes due to sedimentation and erosion from banks and beds (Manjusree et al. 2013). Meanders are significant features that can alter the morphology of floodplains (Lagasse et al. 2004; Guneralp et al. 2012). The reason and process of the changing nature of morphological traits have been investigated by various researchers over the last few decades (Kotoky et al. 2005; Sarkar et al. 2012; Sinha & Ghosh 2012; Gogoi & Goswami 2014; Verma et al. 2021). Inner banks erode, and outer banks accumulate to keep a river balanced (Debnath et al. 2017) to avoid lateral and vertical instabilities in the river (Resmi et al. 2018). Bank erosion and channel migration can occur on different timescales, depending on the environment (days, years, decades).
Riverbank erosion has long-term impacts on natural disasters (Das et al. 2014). Erosion along riverbanks affects several nations, negatively affecting the landscape and having an effect on the ecosystem and society. Over two million people in Bangladesh lost their homes between 1970 and 2000 due to bank erosion in the Padma and Jamuna rivers (Islam & Rashid 2011). The Himalayan foothill region's society benefits from an integrated investigation of erosion, its causes, and a calculated move toward modifying mitigation procedures in fluvial dynamicity (Prokop et al. 2020), and bank erosion would have an impact on the River Mahananda in Sub-Himalayan North Bengal (Chakraborty & Saha 2022). Space Application Centre (SAC), Ahmedabad, India, and the Brahmaputra Board, India (1996) studied river erosion on Majuli Island to identify and delineate island portions that had changed along the bank line due to the river's dynamic behavior.
Many studies have been conducted on meander parameters and the variables that affect them (Hooke 2013; Nabegu 2014). The sinuosity index is one of the most crucial factors for determining and measuring river morphological changes’ spatial and temporal direction (Ozturk & Sesli 2015; Sapkale et al. 2016). The sinuosity index and riverbank erosion have an opposite relationship (Nath & Ghosh 2022). Flood risk increases as stream sinuosity decreases, and riverbanks fail when driving forces exceed resisting forces (Mohammed-Ali et al. 2021). In addition to having an irregular meander belt with variable width and wavelength, they also have distorted geometric patterns. The channel planform characteristics of a meandering system, such as its width, wavelength, and radius of curvature, exhibit clear correlations (Leopold 1994). In contrast, meandering streams in alluvial zones frequently form oxbows. One of the primary causes of oxbow lakes’ formation is soil erosion. A new flood plain was generated by an oxbow (Ijafiya & Yonnana 2018), resulting in significant local or regional effects (Hickin & Nanson 1984). Hence, river morphology studies and geomorphic risks are required for sustainable river basin management (Arnaud-Fassetta et al. 2009; Papini et al. 2011). Oxbow lakes and their subsequent sediment filling are significant for channel self-confinement and meander belt evolution dynamics. In this regard, several river morphological factors such as radius of curvature, centerline distance, sinuosity, axis length, river width, and the impact of anthropogenic activities or changes in land use/land cover (LULC) and affect due to oxbow were investigated to identify morphological changes over the river.
Meandering behavior has been linked to severe issues for human life, such as the destruction of residential and agricultural lands in floodplain zones. Several researchers have investigated the consequences of urbanization on the hydrological, geomorphologic, and biological processes of the fluvial system (Clark & Wilcock 2000; Kondolf et al. 2007; McCann 2013; Das et al. 2014). Assessing bank erosion hazards for the detailed study of fluvial erosion is the most crucial aspect of correlating decline and land use changes (Deng et al. 2019; Twisa & Buchroithner 2019; Neupane et al. 2020). Human intervention alters river courses by focusing on artificial barriers that regulate flow rates (Gregory 2006). Aerial images can be used to compare and evaluate channel geometry changes, particularly river width and urbanization's impact on the land because significant land use changes occur over short enough time scales (Galster et al. 2008). Urbanization and channel morphology are closely linked in many aspects. For instance, Leopold (1970) showed that a river's channel width in the United States was reduced by 20 years due to urban areas. Urbanization has many effects, such as making channels wider and more homogeneous (O'Driscoll et al. 2009; O'Driscoll et al. 2010). This rise cut down on the length of the flow and made the drainage less intense. Urbanization affects the stability of channel banks, the flow of streams, the transport and deposition of sediment, and the widening of channels (McCann 2013). Stream flow and storm runoff in Baltimore, Maryland, have increased due to urbanization in some basin regions (Nelson et al. 2006). Minor land use changes dramatically influenced regional soil erosion rates and sediment transfer to China's Lushi river basin (Wang et al. 2012). Hazarika et al. (2015) found that river dynamics in the upper Brahmaputra plains affected land use and negatively impacted the livelihoods of those living in floodplains. Assessing the impact of agricultural land conversion on soil erosion (Papini et al. 2011; Zhu et al. 2012) is essential. The growing human population is one of many factors that affect how agricultural land is used over time. The soils along riverbanks are vulnerable and erode quickly due to changes in farming practices and vegetation clearing for agricultural use. As a result of harvesting, crops frequently have lower soil organic matter (SOM) than forests (Zhu et al. 2012). Land use and management processes influence soil's chemical and physical composition (Bhattacharya et al. 2016). These land use changes reduce precipitation infiltration, forest cover interception, and drainage efficiency. Flooding and river meandering are impacted by deforestation (Barasa & Pereram 2018; Adnan et al. 2019). Pesticides and herbicides lose effectiveness with heavy use and intensive farming. In response, parasites become more effective and frequent. Intensive farming destroys soil by interfering with natural processes. Pesticides kill soil microorganisms that control composting and organic matter assimilation, reducing soil stability and erodibility. Tree clearance, slash-and-burn techniques, and clearing of forest areas for agriculture have caused deforestation and soil erosion (Sial et al. 2022). Aside from human intervention, changes in the meander characteristics can also be linked to local management operations such as urbanization expansion and the construction of levees, riprap, a dam, and a road (Nelson et al. 2013).
Field data and numerical models estimated meandering and its morphological changes (Xiao et al. 2019). Unfortunately, most alluvial river lengths are unreachable by field technique. Remote sensing is often used to map surface phenomena such as sinuosity index, land use change, and river meandering parameters (Rawat et al. 2013). Very few literatures have documented the effects of LULC transitions, particularly concerning the morphology of rivers and how much of an impact they have. Given the importance of LULC change in channel alterations, the current study attempts to link those changes to understand the phenomenon.
Study area
The River Barak, which flows through the Cachar district, is one of North East India's most meandering rivers. After the Brahmaputra and Chambal rivers, it is the third-largest river in NE India. In several places, the valley is influenced by the river's high meandering and shifting behavior (Das et al. 2007; Das 2012). Agricultural fields, tea garden complexes, rural and urban inhabitants, and abandoned and uncultivated areas compose the floodplain areas, mixing multiple LULCs (Das & Das 2014). Increased urbanization in recent decades has also been a concern for the valley. Human development and its major effects on LULC changes are seen in the Barak River (Annayat et al. 2022). Several environmental problems have arisen as a result of the rapid conversion of open space and vegetation to built-up a region, which is impeding Silchar's ability to grow sustainably (Devi et al. 2019). Land usage, land cover, and river morphology are all linked, according to literature reviews. LULC investigations, on the other hand, are critical for land and water resource management, particularly in flood-prone, highly meandering regions such as Barak. The Barak River's channel behavior was analyzed using satellite imagery and some field data to find river segments that remained steady between 1910 and 1988 (Bardhan 1993). To anticipate channel migration over Barak, an attempt is made to explain and assess the empirical approach and time sequence extrapolation method. It is discovered that Nanson and Hickin technique predicts meander migration better than other empirical methods (Annayat & Sil 2020a). Between 1918 and 2003, a quantitative investigation of the Barak River found an increasing trend in erosion and deposition (Laskar & Phukon 2012). Furthermore, considerable bank shifting in the Barak River was investigated, with the study predicting that it will soon have severe consequences for the economy and people's livelihood (Annayat & Sil 2020a). As a result of the study area's significant sinuosity variation, which causes erosion, the formation has a continual tendency to cut off (Nath & Ghosh 2022).
The study is especially important for this form of highly meandering river management because it indicates how high meandering river morphology has changed over the period and identified some major causes of the changes. This study deals with how the Barak River's meander parameters are interrelated to each other and how it influences river morphology. To the literature, it has been demonstrated that the River Barak has had a significant impact over time and causes regional morphological vulnerability. The links between land use land cover and river morphology are complex. But LULC studies are vital for both land and water resource management. Although there was some research on river morphology (Das 2012; Laskar & Phukon 2012; Annayat & Sil 2020a) and some research was conducted on land use changes over the basin (Das 2012; Devi et al. 2019, Annayat & Sil 2020b). But, these studies were carried out LULC analysis using the ML approach (Das 2012; Devi et al. 2019, Annayat & Sil 2020b; Annayat et al. 2022). However, a literature review revealed that using the SVM technique for LULC classification offered more accuracy. As a result, SVM and ML techniques were utilized in the current work for LULC classification, and a comparison between ML and SVM techniques was also made. Which are provided more preciously quantify the LULC changes in this region, which is essential for related meander river morphological studies.
For statistical analysis over meander parameters, there was some study conducted using the GIS platform that analyze the significance of meander parameters, but very few studies quantify the inter-statistical relationship between each parameter concerning others. In this study, a regression analysis is performed to identify the inter-relationship among meandering parameters. Future trends in river morphology and meander evolution can be predicted by investigating the meander parameters. Better river management and damage reduction may be possible by having an understanding of the evolving meander parameters. One of the key elements responsible for regional and global morphological alterations has been recognized as oxbow formation. It is important to identify future oxbow-prone zone to limit the morphological changes. According to the literature, river bank erosion and its impacts are one of the disasters that have a major influence on some of the world's main flood plains. Barak is a highly meandering river that has undergone significant morphological changes over time. This study provided a better understanding of the dynamics and actual meandering characteristics of alluvial rivers along with morphological causes of vulnerability in the context of the significance of this river and its catchment area.
METHODOLOGY
Data collection
The multispectral remote sensing and Landsat data used in this study were collected from the United States Geological Survey (USGS). Landsat images were taken every ten years from 1990 to 2020 (Table 1).
Satellite data used in the study
Spacecraft ID . | Landsat 5 . | Landsat 7 . | Landsat 7 . | Landsat 8 . |
---|---|---|---|---|
Date | 16/01/1990 | 29/02/2000 | 08/02/2010 | 04/02/2020 |
Sensor ID | TM | ETM + | ETM + | ETM + |
WRS Path | 146 | 146 | 146 | 146 |
WRS Row | 43 | 43 | 43 | 43 |
Resolution(m) | 30 × 30 | 30 × 30 | 30 × 30 | 30 × 30 |
Spacecraft ID . | Landsat 5 . | Landsat 7 . | Landsat 7 . | Landsat 8 . |
---|---|---|---|---|
Date | 16/01/1990 | 29/02/2000 | 08/02/2010 | 04/02/2020 |
Sensor ID | TM | ETM + | ETM + | ETM + |
WRS Path | 146 | 146 | 146 | 146 |
WRS Row | 43 | 43 | 43 | 43 |
Resolution(m) | 30 × 30 | 30 × 30 | 30 × 30 | 30 × 30 |
Technique
The meandering parameter, LULC changes, and river migration assessment of the study were all determined using layer-stacking Landsat images. Landsat data and digitized river pathways were used to track the shifting of river bank lines along the river. In morphological assessments based on digitized river paths, oxbows’ formation and effects are useful. LULC mapping with a supervised classification technique was utilized to identify land use trends across the research area. This study focused on morphological changes in the river, as well as changes in LULC characteristics and the identification of potential oxbow formation zones.
Data used
(a) selection of location for morphological analysis. (b) Meandering parameters of study meanders: axis length (A), meander neck length (L), river width (W), radius curvature (R), Centerline distance (S). (c) Identification of oxbow formation.
(a) selection of location for morphological analysis. (b) Meandering parameters of study meanders: axis length (A), meander neck length (L), river width (W), radius curvature (R), Centerline distance (S). (c) Identification of oxbow formation.
River path digitization
For every satellite image between 1990 and 2020, the river bank has been located and defined. The river bank line may be easily identified using the NDWI image. The computation of river channel migration, changes in river channel width owing to erosion and accretion, and identification of erosion and accretion sites is followed by the preparation of the classified images, vector profile, and reference points. The distance between the reference point and the river banks on either side was digitized as a line feature to compute river channel migration. Following that, the attribute table's computed geometry options were used to determine the lengths of these lines. As the distance between the vector line and channel bank increases, positive values show the channel shifting toward the inner side of the river, which caused accretion. The channel shifted toward the outer banks, causing erosion as the distance between the vector line and the channel bank decreased, as indicated by the negative values. Through area calculation using GIS software tools for polygon areas with shifting banklines over the study period, the erosion and deposit area have been estimated.
Meander parameters
Measurement tools in ArcGIS 10.2 and AUTOCAD 2020 were used to measure morph metric parameters for all meander loops between 1990 and 2020.
River migration and oxbow formation
One of the most influential factors in changing river morphology is river migration and oxbow formation (Johnson & Paynter 1967; Sharma et al. 2010; Tal & Paola 2010; Micheli & Larsen 2011). Unfortunately, most alluvial river lengths are unreachable by field technique.
Multi-temporal high-resolution data allows us to track river configuration changes, new channels or oxbow lakes, and riverbank erosion/deposition (Rawat et al. 2013). The shifting of river bank lines along the Barak River was traced using Landsat data from 1990 to 2020. River migration and oxbow formation were studied using Landsat images to digitize river paths. For referring river migration rate, the position of river banks migration was checked. A U-shaped meander in a river is occasionally cut off from the actual river channel that caused it known as an oxbow. When the gap between two river sections is considerably reduced, it is referred to as a high oxbow-prone zone (Figure 4(c)). To further understand how oxbows form, Figure 4(c) provides an example where, in 1990, the gap between two meander necks of a segment was marked by X, and in 2010, the gap had constantly decreased to become Z. The segment's necks coming together between 2010 and 2020 caused the oxbow to form (X > Y > Z > Oxbow formation). So, future oxbows can be introduced by continuously reducing the space between sharp curve necks.
Land use mapping between 1990 and 2020 and their change detection
Changes in land use have influenced hydrology and sediment transport, which have a considerable impact on soil erosion (Brierley & Fryirs 2005; Wang et al. 2012; Mahapatra et al. 2014; Hegazy & Kaloop 2015; Hazarika et al. 2015). Hydrology, climate, land use, and human involvement affect riverbank erosion and deposition (Arohunsoro et al. 2014; Barman 2016; Barasa & Pereram 2018, Adnan et al. 2019; Raj & Singh 2020).
The flowchart shows the sequential land use mapping processes (Figure 2). There are two subtypes of machine learning techniques: supervised and unsupervised techniques (Wu et al. 2019). Supervised categorization uses user-developed spectral signatures of established categories. The maximum likelihood (ML) and SVM algorithms were used in supervised classification. An overview of the two algorithms is given in the subsections that follow (Taati et al. 2015).
D = weighted distance (likelihood)
C = particular class
X = measurement vector of the candidate pixel
Mc = mean vector of the sample of class c
ac = the likelihood that any candidate pixel belongs to class c as a percentage
Covc = the pixels in the sample of the class c covariance matrix
= Covc's determinant (matrix algebra)
Covc − 1 = inverse of Covc
T = transposition function (matrix algebra).
More recent techniques, such as artificial neural networks (ANN), SVM, RF, decision trees, and other models, have drawn a lot of interest in remote sensing-based applications, such as LULC classification, over the past decade. As a result, many machine learning techniques have been used in multiple studies on LULC modeling (Teluguntla et al. 2018; Abdullah et al. 2019; Zhang et al. 2019; Talukdar et al. 2020). Additionally, a few types of research have been conducted to determine which machine learning classifier is the most appropriate and accurate for LULC mapping (Camargo et al. 2019; Jamali 2019). It has been discovered that ANN, SVM, and RF generally offer more accuracy compared to other conventional classifier algorithms, while SVM and RF are the best machine-learning techniques for LULC classification (Ma et al. 2017). The most extensively utilized algorithms in the research community for multispectral and hyperspectral image classification tasks are RF and SVM (Belgiu & Draguţ 2016; Georganos et al. 2018; Zhang et al. 2019; Talukdar et al. 2020).
In this aspect, several factors such as river meandering parameters, oxbow lake formation, and decadal LULC changes are analyzed to quantify river morphological changes (Hickin & Nanson 1984; Hooke 2013; Gogoi & Goswami 2014; Yousefi et al. 2016; Nath & Ghosh 2022).
RESULTS
The findings of the studies are presented, including changes in LULC in the study area over time, particular changes in meander characteristics through time, river migration analysis, as well as the effect of oxbow formation on river morphology.
LULC analysis
LULC analysis
LULC classes . | 1990 . | 2005 . | Rise/loss of area 1990–2005 (sq. km) . | 2020 . | Rise/loss of area 2005–2020 (sq. km) . | Area covered by classes (1990) . | Area covered by classes (2020) . |
---|---|---|---|---|---|---|---|
Water | 59.19 | 57.56 | −1.6 | 53.22 | −1.34 | 10.76% | 9.68% |
Forest | 151.08 | 127.89 | −23.19 | 114.83 | −13.06 | 27.48% | 20.88% |
Agriculture | 48.33 | 64.29 | +15.96 | 74.39 | +10.1 | 8.79% | 13.53% |
Bareland | 270.01 | 259.61 | −13.4 | 242.8 | −15.81 | 49.11% | 44.16% |
Settlement | 21.16 | 40.42 | +19.26 | 64.53 | +24.11 | 3.84% | 11.73% |
LULC classes . | 1990 . | 2005 . | Rise/loss of area 1990–2005 (sq. km) . | 2020 . | Rise/loss of area 2005–2020 (sq. km) . | Area covered by classes (1990) . | Area covered by classes (2020) . |
---|---|---|---|---|---|---|---|
Water | 59.19 | 57.56 | −1.6 | 53.22 | −1.34 | 10.76% | 9.68% |
Forest | 151.08 | 127.89 | −23.19 | 114.83 | −13.06 | 27.48% | 20.88% |
Agriculture | 48.33 | 64.29 | +15.96 | 74.39 | +10.1 | 8.79% | 13.53% |
Bareland | 270.01 | 259.61 | −13.4 | 242.8 | −15.81 | 49.11% | 44.16% |
Settlement | 21.16 | 40.42 | +19.26 | 64.53 | +24.11 | 3.84% | 11.73% |
(a) LULC mapping 1990. (b) LULC mapping 2005, (c) LULC mapping 2020 (d) LULC Analysis 1990–2020.
(a) LULC mapping 1990. (b) LULC mapping 2005, (c) LULC mapping 2020 (d) LULC Analysis 1990–2020.
Settlement increase over time was one of the most influential parameters in the study area. In 1990, settlement land covered 21.16 sq. km, 40.42 sq. km in 2005, and 64.53 sq. km in 2020. Other land use parameters are also affected by settlement or urbanization, such as agricultural land, which covered 48.33 sq. km of the study area in 1990, 64.29 sq. km in 2005, and 74.39 sq. km in 2020. Increased urbanization was also one of the key influencing factors for deforestation. In 1990, forestry covered over 151.08 sq. km, which decreased to 127.89 sq. km in 2005 and 114.83 sq. km in 2020, according to the LULC mapping. The area matrices of the various LULC types indicate how the Bareland has been converted into other land classes, particularly settlement, and how this has influenced the gradual increase in agricultural land area. Agricultural land and built-up area have shown a significant upward trend, whereas forestation and barren areas have shown a declining trend. As a result of the disturbance caused by settlement, agricultural land changes, and a significant decline in forestation, some of the region's important influencing factors were founded.
The findings are comparable to those of Gebrehiwot et al. (2014) in the Blue Nile basin's Birr and Upper Didesa basins, as well as Gashaw et al. (2017) in northeastern Ethiopia's Dera area, where urbanization and agricultural land have increased while forest land has decreased.
Changes in meander parameters
The study reach's channel centerlines were digitized, and 21 meander loops were chosen for the time interval between the study period and the end of the study period (Figure 6). From 1990 to 2020, morphological parameters for all meander loops were measured, and the morphological change of meander loops was estimated (Table 3).
Changes in meander parameter
Section . | 2020 . | 2010 . | 2000 . | 1990 . | 2020 . | 2010 . | 2000 . | 1990 . | 2020 . | 2010 . | 2000 . | 1990 . |
---|---|---|---|---|---|---|---|---|---|---|---|---|
Radius of curvature | Center line distance | Meander neck length | ||||||||||
P1 | 864 | 1,019 | 881 | 927 | 5,873 | 6,015 | 6,128 | 5,961 | 3,684 | 3,741 | 3,762 | 3,658 |
P2 | 1,906 | 1,833 | 1,525 | 1,802 | 5,100 | 5,697 | 5,506 | 5,373 | 3,677 | 3,984 | 4,183 | 4,044 |
P3 | 3,838 | 3,501 | 2,993 | 3,707 | 9,200 | 8,836 | 9,228 | 9,184 | 6,870 | 6,859 | 6,830 | 6,915 |
P4 | 3,265 | 3,015 | 2,982 | 3,903 | 9,059 | 8,497 | 8,500 | 8,443 | 7,413 | 7,530 | 7,541 | 7,599 |
P5 | 2,579 | 2,573 | 2,324 | 1,444 | 7,032 | 6,862 | 6,976 | 6,860 | 4,831 | 4,731 | 4,728 | 4,689 |
P6 | 895 | 1,267 | 755 | 1,223 | 9,224 | 10,119 | 10,050 | 19,485 | 4,416 | 4,231 | 4,324 | 4,334 |
P7 | 637 | 1,447 | 1,241 | 1,114 | 3,754 | 3,418 | 3,201 | 3,203 | 1,947 | 1,877 | 1,871 | 2,154 |
P8 | 592 | 563 | 351 | 384 | 3,332 | 2,899 | 3,026 | 4,412 | 1,891 | 1,762 | 1,884 | 1,800 |
P9 | 1,105 | 1,089 | 795 | 612 | 4,100 | 4,146 | 4,018 | 3,983 | 2,199 | 2,221 | 2,266 | 2,228 |
P10 | 876 | 499 | 355 | 296 | 4,686 | 4,702 | 4,919 | 4,820 | 1,677 | 1,538 | 1,681 | 1,640 |
P11 | 849 | 910 | 775 | 848 | 4,963 | 4,843 | 4,830 | 4,867 | 1,441 | 1,443 | 1,451 | 1,500 |
P12 | 573 | 578 | 536 | 570 | 4,882 | 4,928 | 4,885 | 4,902 | 1,103 | 1,116 | 1,107 | 1,049 |
P13 | 907 | 711 | 846 | 790 | 3,996 | 3,772 | 3,760 | 3,792 | 1,861 | 1,770 | 1,780 | 1,769 |
P14 | 327 | 262 | 297 | 469 | 2,996 | 3,275 | 3,155 | 3,225 | 932 | 1,096 | 1,053 | 972 |
P15 | 962 | 717 | 889 | 759 | 3,976 | 3,909 | 3,941 | 3,984 | 1,779 | 1,759 | 1,715 | 1,789 |
P16 | 843 | 651 | 805 | 693 | 7,886 | 8,036 | 7,835 | 7,787 | 1,348 | 1,379 | 1,278 | 1,365 |
P17 | 1,287 | 794 | 1,208 | 671 | 6,679 | 6,738 | 5,700 | 6,662 | 3,156 | 3,094 | 3,063 | 3,046 |
P18 | 420 | 475 | 509 | 775 | 5,873 | 5,789 | 5,695 | 5,558 | 1,321 | 1,285 | 1,268 | 1,280 |
P19 | 886 | 917 | 833 | 712 | 4,548 | 4,296 | 4,279 | 4,202 | 1,875 | 1,848 | 1,890 | 1,843 |
P20 | 1,798 | 1,435 | 1,335 | 790 | 3,724 | 3,475 | 2,944 | 2,563 | 2,793 | 2,784 | 2,640 | 2,425 |
P21 | 630 | 691 | 1,097 | 614 | 3,120 | 3,223 | 3,093 | 3,020 | 2,316 | 2,265 | 2,437 | 2,498 |
Axis length | Sinuosity | River width | ||||||||||
P1 | 1,789 | 1,820 | 1,894 | 1,843 | 1.59 | 1.60 | 1.62 | 1.63 | 161 | 152 | 156 | 160 |
P2 | 1,447 | 1,507 | 1,552 | 1,451 | 1.38 | 1.42 | 1.31 | 1.32 | 163 | 153 | 152 | 158 |
P3 | 2,844 | 2,689 | 2,694 | 2,651 | 1.33 | 1.29 | 1.35 | 1.32 | 169 | 154 | 141 | 140 |
P4 | 1,311 | 1,371 | 1,408 | 1,501 | 1.22 | 1.13 | 1.12 | 1.11 | 152 | 117 | 114 | 125 |
P5 | 2,207 | 2,145 | 2,213 | 2,201 | 1.46 | 1.45 | 1.47 | 1.46 | 133 | 113 | 108 | 106 |
P6 | 3,989 | 3,978 | 3,875 | 4,400 | 2.08 | 2.39 | 2.32 | 4.5 | 140 | 122 | 118 | 115 |
P7 | 1,254 | 1,295 | 1,268 | 921 | 1.92 | 1.82 | 1.71 | 1.48 | 148 | 119 | 115 | 109 |
P8 | 1,309 | 1,142 | 1,181 | 1,797 | 1.76 | 1.64 | 1.60 | 2.45 | 145 | 110 | 114 | 121 |
P9 | 1,586 | 1,698 | 1,641 | 1,564 | 1.86 | 1.87 | 1.77 | 1.78 | 179 | 117 | 115 | 125 |
P10 | 1,427 | 1,687 | 1,623 | 1,455 | 2.79 | 3.05 | 2.92 | 2.93 | 143 | 133 | 119 | 133 |
P11 | 1,727 | 1,901 | 1,744 | 1,871 | 3.44 | 3.55 | 3.32 | 3.24 | 131 | 113 | 121 | 118 |
P12 | 2,031 | 2,093 | 2,140 | 2,063 | 4.42 | 4.41 | 4.41 | 4.67 | 148 | 110 | 115 | 113 |
P13 | 1,539 | 1,495 | 1,452 | 1,374 | 2.15 | 2.13 | 2.11 | 2.14 | 135 | 107 | 108 | 111 |
P14 | 1,362 | 1,523 | 1,496 | 1,497 | 3.21 | 2.98 | 2.99 | 3.31 | 169 | 158 | 150 | 147 |
P15 | 1,478 | 1,554 | 1,467 | 1,516 | 2.23 | 2.22 | 2.29 | 2.23 | 146 | 127 | 131 | 124 |
P16 | 2,844 | 2,906 | 2,973 | 2,741 | 5.85 | 5.82 | 6.13 | 5.70 | 131 | 109 | 121 | 115 |
P17 | 2,210 | 2,254 | 2,234 | 2,166 | 2.11 | 2.17 | 1.86 | 2.18 | 135 | 121 | 110 | 113 |
P18 | 2,670 | 2,580 | 2,590 | 2,532 | 4.44 | 4.50 | 4.49 | 4.34 | 129 | 115 | 117 | 119 |
P19 | 1,891 | 1,877 | 1,720 | 1,543 | 2.42 | 2.32 | 2.26 | 2.27 | 153 | 140 | 135 | 125 |
P20 | 877 | 818 | 463 | 217 | 1.33 | 1.24 | 1.11 | 1.05 | 135 | 127 | 130 | 127 |
P21 | 656 | 806 | 538 | 278 | 1.34 | 1.42 | 1.26 | 1.20 | 143 | 135 | 125 | 123 |
Section . | 2020 . | 2010 . | 2000 . | 1990 . | 2020 . | 2010 . | 2000 . | 1990 . | 2020 . | 2010 . | 2000 . | 1990 . |
---|---|---|---|---|---|---|---|---|---|---|---|---|
Radius of curvature | Center line distance | Meander neck length | ||||||||||
P1 | 864 | 1,019 | 881 | 927 | 5,873 | 6,015 | 6,128 | 5,961 | 3,684 | 3,741 | 3,762 | 3,658 |
P2 | 1,906 | 1,833 | 1,525 | 1,802 | 5,100 | 5,697 | 5,506 | 5,373 | 3,677 | 3,984 | 4,183 | 4,044 |
P3 | 3,838 | 3,501 | 2,993 | 3,707 | 9,200 | 8,836 | 9,228 | 9,184 | 6,870 | 6,859 | 6,830 | 6,915 |
P4 | 3,265 | 3,015 | 2,982 | 3,903 | 9,059 | 8,497 | 8,500 | 8,443 | 7,413 | 7,530 | 7,541 | 7,599 |
P5 | 2,579 | 2,573 | 2,324 | 1,444 | 7,032 | 6,862 | 6,976 | 6,860 | 4,831 | 4,731 | 4,728 | 4,689 |
P6 | 895 | 1,267 | 755 | 1,223 | 9,224 | 10,119 | 10,050 | 19,485 | 4,416 | 4,231 | 4,324 | 4,334 |
P7 | 637 | 1,447 | 1,241 | 1,114 | 3,754 | 3,418 | 3,201 | 3,203 | 1,947 | 1,877 | 1,871 | 2,154 |
P8 | 592 | 563 | 351 | 384 | 3,332 | 2,899 | 3,026 | 4,412 | 1,891 | 1,762 | 1,884 | 1,800 |
P9 | 1,105 | 1,089 | 795 | 612 | 4,100 | 4,146 | 4,018 | 3,983 | 2,199 | 2,221 | 2,266 | 2,228 |
P10 | 876 | 499 | 355 | 296 | 4,686 | 4,702 | 4,919 | 4,820 | 1,677 | 1,538 | 1,681 | 1,640 |
P11 | 849 | 910 | 775 | 848 | 4,963 | 4,843 | 4,830 | 4,867 | 1,441 | 1,443 | 1,451 | 1,500 |
P12 | 573 | 578 | 536 | 570 | 4,882 | 4,928 | 4,885 | 4,902 | 1,103 | 1,116 | 1,107 | 1,049 |
P13 | 907 | 711 | 846 | 790 | 3,996 | 3,772 | 3,760 | 3,792 | 1,861 | 1,770 | 1,780 | 1,769 |
P14 | 327 | 262 | 297 | 469 | 2,996 | 3,275 | 3,155 | 3,225 | 932 | 1,096 | 1,053 | 972 |
P15 | 962 | 717 | 889 | 759 | 3,976 | 3,909 | 3,941 | 3,984 | 1,779 | 1,759 | 1,715 | 1,789 |
P16 | 843 | 651 | 805 | 693 | 7,886 | 8,036 | 7,835 | 7,787 | 1,348 | 1,379 | 1,278 | 1,365 |
P17 | 1,287 | 794 | 1,208 | 671 | 6,679 | 6,738 | 5,700 | 6,662 | 3,156 | 3,094 | 3,063 | 3,046 |
P18 | 420 | 475 | 509 | 775 | 5,873 | 5,789 | 5,695 | 5,558 | 1,321 | 1,285 | 1,268 | 1,280 |
P19 | 886 | 917 | 833 | 712 | 4,548 | 4,296 | 4,279 | 4,202 | 1,875 | 1,848 | 1,890 | 1,843 |
P20 | 1,798 | 1,435 | 1,335 | 790 | 3,724 | 3,475 | 2,944 | 2,563 | 2,793 | 2,784 | 2,640 | 2,425 |
P21 | 630 | 691 | 1,097 | 614 | 3,120 | 3,223 | 3,093 | 3,020 | 2,316 | 2,265 | 2,437 | 2,498 |
Axis length | Sinuosity | River width | ||||||||||
P1 | 1,789 | 1,820 | 1,894 | 1,843 | 1.59 | 1.60 | 1.62 | 1.63 | 161 | 152 | 156 | 160 |
P2 | 1,447 | 1,507 | 1,552 | 1,451 | 1.38 | 1.42 | 1.31 | 1.32 | 163 | 153 | 152 | 158 |
P3 | 2,844 | 2,689 | 2,694 | 2,651 | 1.33 | 1.29 | 1.35 | 1.32 | 169 | 154 | 141 | 140 |
P4 | 1,311 | 1,371 | 1,408 | 1,501 | 1.22 | 1.13 | 1.12 | 1.11 | 152 | 117 | 114 | 125 |
P5 | 2,207 | 2,145 | 2,213 | 2,201 | 1.46 | 1.45 | 1.47 | 1.46 | 133 | 113 | 108 | 106 |
P6 | 3,989 | 3,978 | 3,875 | 4,400 | 2.08 | 2.39 | 2.32 | 4.5 | 140 | 122 | 118 | 115 |
P7 | 1,254 | 1,295 | 1,268 | 921 | 1.92 | 1.82 | 1.71 | 1.48 | 148 | 119 | 115 | 109 |
P8 | 1,309 | 1,142 | 1,181 | 1,797 | 1.76 | 1.64 | 1.60 | 2.45 | 145 | 110 | 114 | 121 |
P9 | 1,586 | 1,698 | 1,641 | 1,564 | 1.86 | 1.87 | 1.77 | 1.78 | 179 | 117 | 115 | 125 |
P10 | 1,427 | 1,687 | 1,623 | 1,455 | 2.79 | 3.05 | 2.92 | 2.93 | 143 | 133 | 119 | 133 |
P11 | 1,727 | 1,901 | 1,744 | 1,871 | 3.44 | 3.55 | 3.32 | 3.24 | 131 | 113 | 121 | 118 |
P12 | 2,031 | 2,093 | 2,140 | 2,063 | 4.42 | 4.41 | 4.41 | 4.67 | 148 | 110 | 115 | 113 |
P13 | 1,539 | 1,495 | 1,452 | 1,374 | 2.15 | 2.13 | 2.11 | 2.14 | 135 | 107 | 108 | 111 |
P14 | 1,362 | 1,523 | 1,496 | 1,497 | 3.21 | 2.98 | 2.99 | 3.31 | 169 | 158 | 150 | 147 |
P15 | 1,478 | 1,554 | 1,467 | 1,516 | 2.23 | 2.22 | 2.29 | 2.23 | 146 | 127 | 131 | 124 |
P16 | 2,844 | 2,906 | 2,973 | 2,741 | 5.85 | 5.82 | 6.13 | 5.70 | 131 | 109 | 121 | 115 |
P17 | 2,210 | 2,254 | 2,234 | 2,166 | 2.11 | 2.17 | 1.86 | 2.18 | 135 | 121 | 110 | 113 |
P18 | 2,670 | 2,580 | 2,590 | 2,532 | 4.44 | 4.50 | 4.49 | 4.34 | 129 | 115 | 117 | 119 |
P19 | 1,891 | 1,877 | 1,720 | 1,543 | 2.42 | 2.32 | 2.26 | 2.27 | 153 | 140 | 135 | 125 |
P20 | 877 | 818 | 463 | 217 | 1.33 | 1.24 | 1.11 | 1.05 | 135 | 127 | 130 | 127 |
P21 | 656 | 806 | 538 | 278 | 1.34 | 1.42 | 1.26 | 1.20 | 143 | 135 | 125 | 123 |
Between 1990 and 2020, mean river width estimates reflect an increasing trend in river width. The average river width in 1990 was 125.09 meters, but by 2020 it had increased to 147.07 meters. The average sinuosity of the sections decreased as the centerline distance between time spans decreased. According to the study, the average centerline distance between sections was 5,823.14 meters in 1990 and 5,428.71 meters in 2020. Therefore, during the two decades, the average sinuosity (C) declined. According to sinuosity analysis, P6 indicates the highest content in sinuosity across the research period. According to the research, the sinuosity of the sector was 4.5 in 1990 and 2.08 in 2020. This is because, between 1990 and 2000, an oxbow formed near Sonaimukh (Kaptanpur), cutting off a substantial stretch of the river route from the main Barak. As a result, in this section of the river, a significant decrease in centerline distance (10.26 km) was found between study periods. The highest radius of curvature was discovered in section P3, and the lowest radius of curvature was discovered in section P18. Within the study period, the most significant variation in radius of curvature was observed in P5, P6, P7, P10, and P17. Paired t-tests were used between 1990 and 2020 to determine the significance of variance in the river meandering parameters.
Only the meander width showed a significant change within a 5% significant level between the research periods, according to the samples paired t-test (Table 4). (W). The radius of curvature (R), centerline distance (S), sinuosity (C), meander neck length (L), and meander axis length (A) do not change significantly between periods. Where R-squared (R2) is a statistical measure that measures the amount of variation explained by other independent variables in a regression model for a dependent variable. The degree of marginal significance in a statistical hypothesis test is referred to as the p-value. Acceptance of the null hypothesis is indicated by the ‘p’ value or probability value where strong significance variation is indicated by a p-value less than 5%. The outcomes of variable correlation are summarized in Table 5.
Performance of paired t-test
Meander parameter . | Average difference between variable . | Standard deviation . | T value . | Degree of freedom . | Sig. (* significance) . |
---|---|---|---|---|---|
Radius of curvature | 139.8 | 941.54 | −1.45 | 20 | 0.08 |
Center line distance | −394.42 | 2,898.09 | 0.79 | 20 | 0.22 |
Meander neck length | −3.19 | 1,794.12 | 0.09 | 20 | 0.46 |
Sinuosity | −0.09 | 1.27 | 0.75 | 20 | 0.22 |
Axis length | 41.23 | 817.70 | −0.72 | 20 | 0.23 |
River width | 21.95 | 18.22 | −8.25 | 20 | 3.56 × 10−8* |
Meander parameter . | Average difference between variable . | Standard deviation . | T value . | Degree of freedom . | Sig. (* significance) . |
---|---|---|---|---|---|
Radius of curvature | 139.8 | 941.54 | −1.45 | 20 | 0.08 |
Center line distance | −394.42 | 2,898.09 | 0.79 | 20 | 0.22 |
Meander neck length | −3.19 | 1,794.12 | 0.09 | 20 | 0.46 |
Sinuosity | −0.09 | 1.27 | 0.75 | 20 | 0.22 |
Axis length | 41.23 | 817.70 | −0.72 | 20 | 0.23 |
River width | 21.95 | 18.22 | −8.25 | 20 | 3.56 × 10−8* |
Variable correlation and regression analysis
Variable correlation | R2 | R(p) | S(p) | L(p) | A(p) | W(p) | Intercept |
C | 0.92 | 0.08 | 5.63 × 10−6 | 1.14 × 10−7 | 0.02 | 0.16 | −0.10 |
Variable correlation | R2 | C(p) | S(p) | L(p) | A (p) | W(p) | Intercept |
R | 0.85 | 0.07 | 0.21 | 0.008 | 0.80 | 0.60 | −5.38 |
Variable correlation | R2 | C(p) | R(p) | L(p) | A (p) | W(p) | Intercept |
S | 0.97 | 5.63 × 10−6 | 0.21 | 2.14 × 10−7 | 2.03 × 10−5 | 0.04 | 901.20 |
Variable correlation | R2 | C(p) | R(p) | S(p) | A (p) | W(P p) | Intercept |
L | 0.98 | 1.14 × 10−7 | 0.008 | 2.14 × 10−7 | 0.005 | 0.06 | −469.61 |
Variable correlation | R2 | C(p) | R(p) | S(p) | L (p) | W(p) | Intercept |
A | 0.86 | 0.02 | 0.80 | 2.03 × 10−5 | 0.005 | 0.12 | −692.6 |
Variable correlation | R2 | C(p) | R(p) | S(p) | L (p) | A(p) | Intercept |
W | 0.34 | 0.15 | 0.60 | 0.04 | 0.06 | 0.12 | 135.13 |
Regression equation | C = −0.10 + 0.00,043R + 0.0001S − 0.001L + 0.0007A + 0.01W | ||||||
R = −5.38 + 453.5 C −0.38S +0.96L + 0.08 A −4.13W | |||||||
S = 901.20 + 737.1C − 0.26R + 1.18L + 0.88 A − 12.52W | |||||||
L = −469.61–606.73 C +0.39 R + 0.71S − 0.52A + 8.99W | |||||||
A = −692.6 − 449.19C + 0.05R + 0.80S − 0.79L + 9.35W | |||||||
W = 135.13 + 11.98C − 0.004R − 0.02S + 0.02L + 0.01 A |
Variable correlation | R2 | R(p) | S(p) | L(p) | A(p) | W(p) | Intercept |
C | 0.92 | 0.08 | 5.63 × 10−6 | 1.14 × 10−7 | 0.02 | 0.16 | −0.10 |
Variable correlation | R2 | C(p) | S(p) | L(p) | A (p) | W(p) | Intercept |
R | 0.85 | 0.07 | 0.21 | 0.008 | 0.80 | 0.60 | −5.38 |
Variable correlation | R2 | C(p) | R(p) | L(p) | A (p) | W(p) | Intercept |
S | 0.97 | 5.63 × 10−6 | 0.21 | 2.14 × 10−7 | 2.03 × 10−5 | 0.04 | 901.20 |
Variable correlation | R2 | C(p) | R(p) | S(p) | A (p) | W(P p) | Intercept |
L | 0.98 | 1.14 × 10−7 | 0.008 | 2.14 × 10−7 | 0.005 | 0.06 | −469.61 |
Variable correlation | R2 | C(p) | R(p) | S(p) | L (p) | W(p) | Intercept |
A | 0.86 | 0.02 | 0.80 | 2.03 × 10−5 | 0.005 | 0.12 | −692.6 |
Variable correlation | R2 | C(p) | R(p) | S(p) | L (p) | A(p) | Intercept |
W | 0.34 | 0.15 | 0.60 | 0.04 | 0.06 | 0.12 | 135.13 |
Regression equation | C = −0.10 + 0.00,043R + 0.0001S − 0.001L + 0.0007A + 0.01W | ||||||
R = −5.38 + 453.5 C −0.38S +0.96L + 0.08 A −4.13W | |||||||
S = 901.20 + 737.1C − 0.26R + 1.18L + 0.88 A − 12.52W | |||||||
L = −469.61–606.73 C +0.39 R + 0.71S − 0.52A + 8.99W | |||||||
A = −692.6 − 449.19C + 0.05R + 0.80S − 0.79L + 9.35W | |||||||
W = 135.13 + 11.98C − 0.004R − 0.02S + 0.02L + 0.01 A |
For each variable, a regression equation was developed using a regression analysis between the variable. Except for river width (W), all dependent variables had R2 statistics that were satisfactory.
Effect of river morphology due to river migration and oxbow development
River migration analysis over the study period
Sec . | Total migration 1990–2020 . | Migration rate (1990–2000) (m/year) . | Avg. m/year . | Migration rate (1990–2010) (m/year) . | Avg. m/year . | Migration rate (1990–2020) (m/year) . | Avg. m/year . | |||
---|---|---|---|---|---|---|---|---|---|---|
Right . | Left . | Right . | Left . | Right . | Left . | |||||
P1 | 30.69 | 1.01 | 0.63 | 0.82 | 0.58 | 0.55 | 0.56 | 0.47 | 0.51 | 0.49 |
P2 | 27.18 | 0.89 | 0.63 | 0.76 | 0.72 | 0.64 | 0.68 | 0.16 | 0.36 | 0.26 |
P3 | 31.26 | 2.24 | 1.06 | 1.65 | 0.68 | 0.36 | 0.52 | 0.21 | 0.35 | 0.28 |
P4 | 67.3 | 3.12 | 1.18 | 2.15 | 3.86 | 1.83 | 2.84 | 2.92 | 1.12 | 2.02 |
P5 | 106.35 | 5.87 | 2.11 | 3.99 | 5.3 | 2.77 | 4.03 | 4.38 | 2.54 | 3.46 |
P6 | 199.48 | 8.17 | 2.92 | 5.54 | 5.02 | 16.32 | 10.67 | 8.16 | 5.26 | 6.71 |
P7 | 393.75 | 13.73 | 7.88 | 10.8 | 12.89 | 10.82 | 11.85 | 6.01 | 9.99 | 9.0 |
P8 | 105.16 | 3.37 | 2.64 | 3.0 | 3.4 | 1.31 | 2.35 | 1.58 | 1.5 | 1.54 |
P9 | 77.73 | 4.9 | 3.21 | 4.05 | 2.42 | 1.63 | 2.02 | 1.22 | 1.98 | 1.6 |
P10 | 34.94 | 1.47 | 2.14 | 1.8 | 1.01 | 1.35 | 1.18 | 0.6 | 0.62 | 0.61 |
P11 | 53.92 | 3.28 | 2.79 | 3.03 | 1.68 | 2.07 | 1.87 | 1.44 | 1.02 | 1.23 |
P12 | 50.91 | 4.75 | 3.55 | 4.15 | 3.65 | 1.94 | 2.79 | 1.53 | 1.33 | 1.43 |
P13 | 37.36 | 1.3 | 1.17 | 1.23 | 1.33 | 1.07 | 1.2 | 0.55 | 0,42 | 0.48 |
P14 | 22.77 | 1.18 | 1.07 | 1.12 | 0.41 | 0.62 | 0.51 | 0.35 | 0.40 | 0.37 |
P15 | 80.62 | 5.76 | 2.74 | 4.25 | 2.61 | 2.11 | 2.36 | 2.63 | 2.27 | 2.45 |
P16 | 55.04 | 1.84 | 1.97 | 1.9 | 2.25 | 2.17 | 2.21 | 1.21 | 1.15 | 1.18 |
P17 | 78.1 | 2.38 | 2.89 | 2.63 | 1.91 | 3.02 | 2.46 | 0.94 | 1.57 | 1.25 |
P18 | 52.62 | 2.84 | 2.21 | 2.52 | 1.58 | 2.22 | 1.9 | 0.73 | 1.32 | 1.02 |
P19 | 83.39 | 9.92 | 5.18 | 7.55 | 5.72 | 4.33 | 5.02 | 4.22 | 2.94 | 3.58 |
P20 | 602.86 | 5.99 | 15.99 | 10.99 | 13.84 | 13.22 | 13.33 | 15.04 | 11.29 | 13.16 |
P21 | 350.68 | 4.15 | 4.63 | 4.41 | 8.54 | 11.65 | 10.09 | 5.73 | 4.83 | 5.28 |
Sec . | Total migration 1990–2020 . | Migration rate (1990–2000) (m/year) . | Avg. m/year . | Migration rate (1990–2010) (m/year) . | Avg. m/year . | Migration rate (1990–2020) (m/year) . | Avg. m/year . | |||
---|---|---|---|---|---|---|---|---|---|---|
Right . | Left . | Right . | Left . | Right . | Left . | |||||
P1 | 30.69 | 1.01 | 0.63 | 0.82 | 0.58 | 0.55 | 0.56 | 0.47 | 0.51 | 0.49 |
P2 | 27.18 | 0.89 | 0.63 | 0.76 | 0.72 | 0.64 | 0.68 | 0.16 | 0.36 | 0.26 |
P3 | 31.26 | 2.24 | 1.06 | 1.65 | 0.68 | 0.36 | 0.52 | 0.21 | 0.35 | 0.28 |
P4 | 67.3 | 3.12 | 1.18 | 2.15 | 3.86 | 1.83 | 2.84 | 2.92 | 1.12 | 2.02 |
P5 | 106.35 | 5.87 | 2.11 | 3.99 | 5.3 | 2.77 | 4.03 | 4.38 | 2.54 | 3.46 |
P6 | 199.48 | 8.17 | 2.92 | 5.54 | 5.02 | 16.32 | 10.67 | 8.16 | 5.26 | 6.71 |
P7 | 393.75 | 13.73 | 7.88 | 10.8 | 12.89 | 10.82 | 11.85 | 6.01 | 9.99 | 9.0 |
P8 | 105.16 | 3.37 | 2.64 | 3.0 | 3.4 | 1.31 | 2.35 | 1.58 | 1.5 | 1.54 |
P9 | 77.73 | 4.9 | 3.21 | 4.05 | 2.42 | 1.63 | 2.02 | 1.22 | 1.98 | 1.6 |
P10 | 34.94 | 1.47 | 2.14 | 1.8 | 1.01 | 1.35 | 1.18 | 0.6 | 0.62 | 0.61 |
P11 | 53.92 | 3.28 | 2.79 | 3.03 | 1.68 | 2.07 | 1.87 | 1.44 | 1.02 | 1.23 |
P12 | 50.91 | 4.75 | 3.55 | 4.15 | 3.65 | 1.94 | 2.79 | 1.53 | 1.33 | 1.43 |
P13 | 37.36 | 1.3 | 1.17 | 1.23 | 1.33 | 1.07 | 1.2 | 0.55 | 0,42 | 0.48 |
P14 | 22.77 | 1.18 | 1.07 | 1.12 | 0.41 | 0.62 | 0.51 | 0.35 | 0.40 | 0.37 |
P15 | 80.62 | 5.76 | 2.74 | 4.25 | 2.61 | 2.11 | 2.36 | 2.63 | 2.27 | 2.45 |
P16 | 55.04 | 1.84 | 1.97 | 1.9 | 2.25 | 2.17 | 2.21 | 1.21 | 1.15 | 1.18 |
P17 | 78.1 | 2.38 | 2.89 | 2.63 | 1.91 | 3.02 | 2.46 | 0.94 | 1.57 | 1.25 |
P18 | 52.62 | 2.84 | 2.21 | 2.52 | 1.58 | 2.22 | 1.9 | 0.73 | 1.32 | 1.02 |
P19 | 83.39 | 9.92 | 5.18 | 7.55 | 5.72 | 4.33 | 5.02 | 4.22 | 2.94 | 3.58 |
P20 | 602.86 | 5.99 | 15.99 | 10.99 | 13.84 | 13.22 | 13.33 | 15.04 | 11.29 | 13.16 |
P21 | 350.68 | 4.15 | 4.63 | 4.41 | 8.54 | 11.65 | 10.09 | 5.73 | 4.83 | 5.28 |
(a) Section-wise river migration analysis. (b) River Migration near Sonabarighat (section P6). (c) Identification of Future oxbow formation location.
(a) Section-wise river migration analysis. (b) River Migration near Sonabarighat (section P6). (c) Identification of Future oxbow formation location.
Total migration was calculated from the river centerline position from 1990 to 2020. According to river migration analysis, the areas P7, P20, and P21 were the most severely impacted. P20 section was the most vulnerable of all the sections, as it was found near 602.86 meters of land that had been impacted by river migration throughout the study period. P1, P2, P3, and P14 were among the most stable sections. In such sections, annual shifting of fewer than 2 meters has been recorded. Whereas for P20, the rate of migration was also considerably high when compared to other sections of the study area, where near 13.16 meters of moving was still causing land degradation every year because to NH53, this portion of the river is now more vital to Pachgram (Section P20). The only highway connecting Silchar to the rest of the country runs through it.
During the study period, a large section of the river path was cut off from the main Barak near Sonaimukh (Kaptanpur), and an oxbow formed over section P6. One of the main explanations for the largest sinuosity fluctuation across the decade was oxbow formation (Figure 3(c)).
As a result, overall river migration and migration rates were much higher in P6, P7, and P8 than in other sections. As Sonabarighat Pt II (Section P7) has been subjected to significant migration due to river migration. Piers were built here between 2012 and 2013 (Figure 7(b)), which were used to connect the two ends of the river in preparation for constructing a bridge. As per collective information, due to some issues, construction activities were suspended in 2013. Over a decade, the river has moved an enormous amount of land, reaching the last piers of the bridge. Hare, over 150-meter migration was observed within the last decade. During the field visit, it was observed that a roadway near Govindapur (Section P6) had entirely deteriorated in the recent decade due to massive migration activity. Within this zone, there is still around 6.71-meter shifting per year. The upstream river section is still suffering from river migration, and oxbow development is one factor that makes the area vulnerable to river migration. As a result, determining the future oxbow tendency ability zone is crucial for river management, especially in this type of river where the township is close to the flood plain. An oxbow is formed when two meander necks get near together (Figure 3(c)). Following the river's path, some future oxbow tendency-able zones inside the study area were notified (Figure 7(c)).
As a result, the riskiest zone between sections P6 and P7 is Dhamalia, due to future oxbow tendencies. This section of the river (Dhamalia) is more prone to oxbow formation. If the river is not protected immediately, oxbows will form, cutting Kanakpur, Gangapur, and Neairgram off the main Barak (between sections P6 and P7). The distance between the river's two meander neck ends was over 750 meters in 1990 and is now 490 meters. According to migration analysis, the gap is constantly decreasing. Baghadhar (section P10) and Manikpur (section P16) are similar, with the observed gap between the two meander necks decreasing significantly during the study period. If an oxbow formation takes place, it will be affected similarly to the oxbow formed in Sonaimukh (section P6), resulting in downstream instability.
DISCUSSION
River system morphological changes are linked like a chain (Lagasse et al. 2004). Centerline distance (S) decreased due to recorded changes in the radius of curvature (R) induced by decreasing the internal arc. Axis length (A) also increased during this process. The sinuosity of a meander is also affected by changes in other meander parameters. As a result, the average sinuosity (C) has decreased over time. Changes in river width (W) are inversely proportional to changes in centerline distance (S) (Yousefi et al. 2016). If the length of the meander neck (L) remains constant over time, the radius of curvature (R) will decrease, and meander sinuosity (C) will increase (Crosato 2008; Heo et al. 2009). As a result, the radius of curvature and sinuosity were inversely proportional. Here, the average radius of curvature over the sections shows increasing trends and the meander neck length (L) remains almost constant over the study period; therefore, average sinuosity shows a decreasing trend. In this study, the average radius of curvature over the sections has been growing, and the average sinuosity has decreased. In general, the results show a direct link between changes in the radius of curvature and changes in sinuosity, the meander neck length, centerline distance, and meander axis length. The regression analysis is performed to identify the inter-relationship among meandering parameters. The performance of regression analysis is checked by R2 value (Table 5). It shows in Table 5 that the inter-relationship between each parameter with other meandering parameters is fairly good as R2 statistics varying in between 0.85 and 0.98. The regression equation for L concerning other meandering parameters is a good fit. On the other hand, the regression equation for W concerning other meandering parameters is a poor fit as R2 is equal to 0.34.
The study area's water body, forest, and bareland type areas decreased by 6.92 sq. km, 34.36 sq. km, and 22.27 sq. km, respectively, between 1990 and 2020. On the other hand, agricultural and settlement areas increased by 15.4 and 48.15 sq. km, respectively. Land use parameters, such as the rate of settlement and the rate of agriculture practiced, fluctuated significantly over a decade in the current study, with a constant loss of forest land observed throughout the study area. Integrated river basin management strategies, regulating rapid settlement growth and agriculture practices, and reforesting significantly mitigate the rapid rates of LULC conversions at a watershed. This study shows that the upstream section (sections P1 to P5) has the most hilly-dense forestation and has less urbanization than the downstream section of the study area. As a result, river migration in those sections was significantly lower than in other sections. According to LULC mapping, agriculture operations increased at their highest between (sections P6 to P10), with P5 to P8 being some of the most vulnerable sections identified by the migration study. The most vulnerable sections of the research area were sections P20 and P21, in which agriculture has a slightly greater influence near the river bank. One of the primary contributing factors of section P6 for considerable fluctuation in sinuosity is oxbow development. As a result of the river's morphological changes, it lost over 10.26 kilometers of length from the main Barak. Hence, this part and the downstream segment suffered substantially due to river migration. On the other hand, the most probable future oxbow-prone zones identified in this analysis are Dhamalia, Baghadhar, and Manikpur. Deforestation and urbanization are two major contributors to river basin LULC changes, which are linked to river morphology (Woldemichael et al. 2012; Behera et al. 2018; Li et al. 2020). This finding supports the hypothesis that LULC changes affect river morphology. River morphological changes often take decades to become noticeable, although a human intervention can boost channel alterations (Rinaldi et al. 2005). Therefore, the rate of morphological changes, as well as future oxbow formation, will accelerate more if land use changes as significantly as those previously seen. This study suggests that more research to be done to detect prospective changes in terms of LULC changes, as well as a long-term predicted model for predicting future morphological changes in rivers.
CONCLUSION
Understanding geomorphic responses to global land use change requires a historical perspective. Prior conditions, such as previous disturbances to which systems are responding must be considered while making a change. The current study shows that utilizing RS and GIS, multi-temporal satellite data (Landsat) may be used to monitor river morphological activity with LULC and meandering parameter variations. The study's main purpose was to use remote sensing data to analyze and detect morphological changes in a section of the Barak River from 1990 to 2020. The average centerline distance (S) in this analysis decreased over time, which influenced the change in sinuosity of that section. Before 2000, a cutoff was created between sections P6 and P7, reducing the main centerline distance (S) from the river to around 10.26 kilometers. Oxbow formation was one of the key factors for the highest sinuosity fluctuation over the decade. Therefore, identifying the formation zone for future oxbow formation was essential to achieving the aims of this research. Here, the most dangerous zones for future oxbow development were Dhamalia (P6 to P7), Baghadhar (P10), and Manikpur (P16). If the land is not protected immediately, a large portion of the main river will be cut off, making the area more vulnerable downstream. All dependent variables had acceptable R-Square statistics, except river width (W). Where river width (W) was also affected due to changes in river centerline distance(S), therefore average width of the river shows the incremental trend. Then a regression equation was established for each meandering variable.
Observations of geomorphic responses to land use may only be accurately interpreted if a thorough understanding of prior land use changes and patterns was obtained. According to the investigation critically from coarse resolution imageries, water, forest, and bareland type areas decreased in the study area, whereas agricultural and settlement areas increased during the study period. Agricultural activities are influenced by urbanization or settlement rise. Intensive agricultural land was largely afflicted by river migration. Changes in river migration rates substantially impact other river meandering parameters and morphology. LULC changes and meandering river parameters are all linked to river morphology. The parameters of meanders can be used to forecast future trends in river morphology and meander evolution. Understanding how meander parameters are changing could lead to better river management and mitigation of the damage caused by these changes.
LIMITATION
In this investigation, images with a 30 m pixel reconstruction are employed. This is the consequence of colour misunderstanding between Landsat image stacks. In contrast, the colour associated with urbanization resembles bare land, and the colour associated with agriculture resembles forestation. As a result, selecting the correct class using only Google Maps is very much challenging. As a result, the necessary site-specific knowledge should be required for this type of assessment using coarse-resolution images.
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.
CONFLICT OF INTEREST
The authors declare there is no conflict.