Planform Adjustments of Sub Himalayan Mahananda Balason River Over Five Decades in Response To Anthropogenic Interventions

Arabian Journal of Geosciences (2023) 16:676
https://doi.org/10.1007/s12517-023-11809-z
ORIGINAL PAPER
Planform adjustments of sub‑Himalayan Mahananda‑Balason River

over five decades in response to anthropogenic interventions
Suman Mitra1 · Mehebub Mondal1 · Debajit Datta2 · Lakpa Tamang1
Received: 10 May 2023 / Accepted: 23 November 2023

© Saudi Society for Geosciences and Springer Nature Switzerland AG 2023
Abstract
This study aims to analyze the impacts of anthropogenic interventions on channel planform adjustments of the sub-Hima-
layan Mahananda-Balason River system for 52 years (1968–2020). In the studied period, Siliguri, a small urban sprawl, has
largely grown to the 3rd largest in West Bengal, India, just after Kolkata and Asansol and, alongside with this the studied
river system, flowing through Siliguri urban region, has adjusted its planform largely. Now, this study is chiefly framed to
study the interrelationship between these. A total of 6 satellite images (1968, 1979, 1989, 1998, 2008, and 2020) have been
used to measure anthropogenic interventions that include embanking, sediment extraction, land use changes, longitudinal
obstructions, and channel planform properties comprising channel width, area, length, bar coverage, and channel pattern.
All these quantifications are done reach-wise as both the river systems, that is, Mahananda and Balason, are sub-divided
into 5 equidistant reaches each. Considering the nature of the generated data structure, initially, a two-tailed “t”-test was
performed to measure the statistical significance of the change rate of the indicators followed by the panel data regression
using the fixed effect model (FE) to measure the intensity of ongoing alterations on channel planform adjustments. Results
showed that, with significantly increasing urbanization, embanking and in-channel sediment extraction, channels get narrowed
by > 50%. Alike to channel width, the channel length and bar coverage were also reduced in reaches 2–5, and reaches 7–9.
Ultimately, the inherent braided nature of the channel planform had adjusted largely and even transformed sinuous (reaches
3, 4, 9, and 10). Obtained R2 values inferred that the anthropogenic interventions were highly responsible for such intensive
channel adjustments. Relatively, upper reaches (reaches 1 and 6), having fewer human interventions, are still managing their
pre-existing channel morphology, although undersupply of sediments at the lower reaches and gradual engulfment of rural
areas adjacent to upper reaches for urbanization is driving these reaches towards adjusting their channels accordingly. An
economically viable and robust assessment like this is needed to prevent the possible undersupply of riverine resources and
to develop policies on its sustainability. Such studies are immensely suitable in data-scarce regions also where investigations
do not take place in the past and secondary data are also inadequate.
Keywords Channel adjustment · Human intervention · Sediment extraction · Sub-Himalayan foothill · Urbanization
Introduction
Channel reconfiguration or channel planform adjustments

are the ultimate outcomes of the complex interaction
between sediment and water flux which are eventually con-
trolled by natural channel-forming factors such as floods,
Responsible Editor: Amjad Kallel
tectonic movements, landslides, volcanic eruptions, and
* Lakpa Tamang forest fires (Benda et al. 2003; Surian et al. 2016; Righini
ltgeog@caluniv.ac.in et al. 2017; Nafchi et al. 2021; Ghosh 2022). However, in
this era of Anthropocene, channel adjustments can not be
1
Department of Geography, University of Calcutta, discussed without considering the anthropogenic altera-
Kolkata 700019, India
tions (Paul & Meyer 2001; Chin 2006; Kiss & Blanka 2012;
2
Department of Geography, Jadavpur University, Scorpio & Rosskopf 2016; Bandyopadhyay and De 2018),
Kolkata 700032, India
13
Vol.:(0123456789)
676 Page 2 of 22 Arab J Geosci (2023) 16:676
as it is becoming increasingly dominant to such an extent to these, the channels are getting completely altered from
that it may overwhelm the natural channel-forming factors their inherent channel pattern, and the shortage of riverine
(Brown et al. 2017; Macklin & Lewin 2019). Among the resources brings hydro-biological as well as socio-hydrolog-
anthropogenic alterations, longitudinal and lateral obstruc- ical imbalances (Simon & Rinaldi 2006).
tion in the form of dams, barrages, and bridge pillars and “Terai” or the adjacent plains of Siwalik Himalaya are
in-channel sediment extraction apply direct anthropogenic high in topographical and hydro-biological distinctive-
stress to the channels which result in rapid transformation ness which becomes fragile with minimal human interfer-
of the channel planform as it actively modifies the water and ence and can lose its inherent characteristics. Despite such
sediment influx (Gregory 2006; Kondolf 1997). Conversely, importance, the “Terai” of West Bengal remains neglected
other human interventions such as land use transformation regarding studies related to human interventions on rivers
of the adjoining floodplain in the form of urbanization and and quantifying its responses. The rivers flowing through the
deforestation reshape the planform slowly by altering the sub-Himalayan foothills or “Terai” of West Bengal, namely,
overland flow which contributes large volumes of sediment Mahananda, Balason, Teesta, Karala, Lish, Gish, Chel, Jald-
and in the absence of which channel planform adjusts inevi- haka, and Torsa, and many others characterized by heavy
tably (Kang and Marston 2006; Bandyopadhyay & De 2017; seasonal discharge, wide gravelly riverbeds, braided chan-
Yousefi et al. 2019; Bhattacharya et al. 2019). The coupled nel patterns, are experiencing massive adjustments in their
effect of these eventually drives the entire river system on platform due to increasing anthropogenic pressure. Not only
the verge of irreversible degradation from where attainment the direct modifiers such as dams, barrages, embankments,
of its pre-existing equilibrium condition becomes hardly large-scale sediment extraction, and obstruction created by
achievable in quick succession and ultimately brings enor- bridge pillars altering the platform but also rapid urbaniza-
mous multifaceted hazards in the socio-hydrological as well tion with vegetation removal from the floodplains is taking
as hydro-biological domain (Kondolf et al. 2002; Gregory part in reconfiguring the channel morphology of these rivers
2006; Surian et al. 2009a, b; Moretto et al. 2014; Talebmo- (Wiejaczka et al. 2014; Tamang and Mandal 2015; Ayaz
rad & Ostad-Ali-Askari 2022). et al. 2018; Biswas & Banerjee 2018; Saha & Bhattacharya
In case of longitudinal obstructions, river water, stored 2019; Dhali et al. 2020; Hasanuzzaman et al. 2021). Being
at barrages and reservoirs, being free from bedload, when transboundary rivers, alterations in the upstream Indian
gets discharged, vigorously scours its bed with excess energy part can cause serious hydromorphological alterations in
(Kondolf 1997; Grable & Harden 2006). Besides this, lat- downstream reaches flowing through Bangladesh, and lack
eral disconnections such as embankments do not allow the of availability of hydrological data ends up having very
stormwater flux to be spilt in the floodplains, so to reallocate few paleo-hydrological analyses of these rivers. The reason
the excess water the river starts down-cutting (Booth 1990; behind choosing this river system is the presence of the larg-
Rinaldi & Simon 1998). It was estimated that approximately est urban agglomeration of northern West Bengal, i.e., Silig-
32 to 50 billion tons of sand are required to meet the global uri at its adjacent floodplain. A priori investigations have
need, the majority of which are extracted from riverine shown that this system can be distinctly separated into two
sources, even exceeding the global fossil fuel extraction parts, the confined hilly region and the unconfined flood-
(Bendixen et al. 2019). Massive extraction of well-sorted, plain, and the hilly part poses no such alteration activities
granular, coarse, round-shaped, and seasonally replenishable since topographic hindrances discourage landscape altera-
riverine sediments lowers the channel bed, changes the bar tions and riverine resource extractions from there. Therefore,
morphology and channel pattern, and inevitably disturbs the only the unconfined part of this river system is selected and
sediment balance, fish habitats as well as the groundwater examined here. Researchers have recently tried to assess
table (Liébault & Piégay 2002; Rinaldi et al. 2005; Surian the human role in modifying channel planform and hydro-
et al. 2009a, b; Best 2019; Hajdukiewicz and Wyżga 2019; morphology of the rivers of the Terai region (Wiejaczka
Ostad-Ali-Askari & Shayannejad 2021). Changes in the et al. 2014; Mitra et al. 2020), but an indicator-based, robust
impervious surface and vegetation cover mostly reconfig- assessment for a long period, is still due.
ure the channel as the presence of canopy cover provides Considering these research gaps, this study is designed
interception storage for precipitation and thus increases the on the basis of the following objectives:
lag time between precipitation and influx; however, highly
impervious and less vegetated catchments or floodplains A. To assess the nature and extent of the anthropogenic
add bed load-free runoff in quick succession after a spell of intervention and channel adjustments experienced by the
precipitation, which increases the stream energy and trig- unconfined reaches of the Mahananda-Balason system.
gers the down-cutting of the channel (Paul & Meyer 2001; B. To analyze the impact of such interventions on channel
Vanacker et al. 2005; Keen-Zebert 2007; Galster et al. 2008; adjustment in the studied system and the effect of that in
Pal & Talukdar 2020; Khaleghi & Surian 2019). In response the socio-hydrological and bio-hydrological domains.
13
Arab J Geosci (2023) 16:676 Page 3 of 22 676
Therefore, the novelty of this work lies in estimating the on ongoing changes in the river planform. Along with
ongoing anthropogenic alterations and the channel responses this, data related to sediment extraction, embankments,
using several indicators along with obtaining the impact of and urban infrastructure were collected from the District
such alterations on channel responses at the unconfined Land Reforms Office (DLRO) of Darjiling district and
Mahananda-Balason River for the last 52 years (1968–2020). from the office of the Irrigation and Waterways Depart-
A robust assessment like this, using multiple pressures and ment, Siliguri division, Government of West Bengal. The
responses related to human intervention and channel plan- previous research works (Sarkar 1989; De 1998; Tamang
form properties, can be helpful in such data-scarce regions 2013) were also consulted to develop a prior understand-
to prevent channel morphological transformations and take ing of the past condition of the study area. Thus, this study
curative measures which can help to regain the inherent offers a noble hybrid methodology which consisted of the
channel planform properties, quickly to maintain a sustain- measurement of several indicators using a geospatial data-
able flow of riverine resources between two nations. This base and validation of the results from field verification,
study also puts forward some necessary site-specific resto- data procured from public agencies and personal inter-
ration strategies to bring down the extent of anthropogenic views. Field validation of the derived results has proved to
stress-induced alteration. be immensely useful in the understanding of the channel
evolution of this river system since due to its transbound-
ary nature, discharge-related database is absent, and some
Database and methodology of the geospatial images (1989, 1998, and 2008), used to
quantify the human intervention and channel planform
Data procurement and pre‑processing properties, were coarser regarding their spatial resolution.
It also provided key insights into how such interventions
This multi-temporal study incorporated several multi- are reshaping the inherent properties of that fluvial sys-
spectral and panchromatic satellite images with varying tem. However, in many circumstances, field visits cannot
spatial and temporal resolutions. Initially, an ALOS PAL- precisely portray the past conditions of a fluvial regime;
SAR DEM of 12.5 m spatial resolution, dated 11/02/2009, in that case, the secondary datasets (e.g., procured from
was downloaded from https://asf.alaska.edu to delineate public agencies) and personal interviews play an impor-
the unconfined floodplain region. Thereafter, 4 multispec- tant role in validation. For this study, public agencies such
tral LANDSAT images (1989, 1998, 2008, and 2019) and as DLRO and the Irrigation and Waterways Department
two panchromatic CORONA images (1968 and 1979) were provided precise information on sediment extraction,
downloaded from https://earthexplorer.usgs.gov/ (Table 1). embankment building, and bridge establishment, which
In addition to this, Google Earth images with very became beneficial in validating the measured anthropo-
high spatial resolution were used, and several extensive genic interventions. The interviewees were mostly asked
field surveys were carried out from July 2019 to February about the past condition of the studied rivers (e.g., avail-
2020 to ground verify the results derived from satellite ability of fishes, groundwater, and sediments) from where
image analysis. As a part of the field survey, 117 persons, the changes detected at the time of measuring the channel
mostly sediment extraction workers, fishermen, people morphological properties were validated. Thus, the adop-
living at the banks of the rivers, and people who used to tion of such a hybrid methodology ultimately increased the
come regularly to the riverside for their household and efficiency and effectiveness of this study.
religious work, were interviewed to have their perception
Table 1 Geospatial dataset used Date of acquisition Satellite ID Sensor Path/row Resolution
in this study
22/12/1968 CORONA - - 6 ft. (~ 1.83 m)
19/05/1979 CORONA - - 6 ft. (~ 1.83 m)
04/12/1989 Landsat_5 TM 139/041 30 m
29/12/1998 Landsat_5 TM 139/041 30 m
21/10/2008 Landsat_5 TM 139/041 30 m
05/01/2019 Landsat_8 OLI 139/041 30 m
26/01/2020* Landsat/Copernicus - - -
*
Acquired from Google Earth platform
13
Delineation of the study area and reach and conversely extensive urban areas, and fallow lands in the
segmentation downstream (Table 2).
Mahananda and Balason, the two rain-fed tributaries of

Mechi, after being originated from Mahaldiram hills and Selecting and measuring the indicators
Lepcha Jagat, debouched at Gulma and Dudhia respec-
tively and thereafter flow for several km through the larg- A total of 10 indicators, divided into two sets with 5 indica-
est urban agglomeration of northern West Bengal, i.e., tors each, were formulated to estimate the extent of human
Siliguri, before they coalesce at Naukaghat. Geomorphi- interventions and the dynamicity of the planform attributes.
cally, the Mahananda-Balason system can be divided into These indicators were selected from previous studies consid-
two segments: the hilly rugged confined region extending ering the ongoing anthropogenic interventions in the study
between the sources to debouching points and the gently area and alteration of the channel morphology. Due to the
slopped unconfined floodplain at the southern portion of lack of previous field measurements, change in change bed
the debouching points. Anthropogenic activities revolve elevation could not be accommodated in the indicators pool.
around these topographic hindrances, and a majority of The fluctuations of discharge also could not get measured
human interventions are confined in the unconfined flood- due to the unavailability of the dataset. All these indicators
plain. Therefore, only the floodplain region, extending were measured reach-wise for 6 decades, i.e., 1968, 1979,
between the debouching points to Naukaghat, is selected 1989, 1998, 2008, and 2020, using the geospatial datasets
for the study. Delineation of this region initiated with the and thoroughly verified afterwards from field investigations,
demarcation of the mountain front using ALOS DEM to personal interviews, works of previous researchers, and
make a clear dividing line between the confined and uncon- governmental reports and using the high-resolution Google
fined areas (Ashmore & Gardner 2008; Bowen et al. 2014). Earth images.
Following that, the floodplain, having an aerial coverage of Quantification of indicators measuring the intensity of
244.2 km2, extending between 26° 40′ N to 26° 50′ N and ongoing anthropogenic activities (Fig. 2) comprised four
88° 12′ E to 88° 28′ E with the highest and lowest altitude major aspects regarding lateral disconnection, sediment
of 300 m and 114 m, was delineated using flow direction extraction, land use transformation, and longitudinal altera-
analysis in ArcGIS (Fig. 1). This unconfined region, cov- tion. The concretized large embankments largely disconnect
ered with thick quaternary sedimentary deposits of gravel, the river from its floodplain. In flood season, it hinders the
sand, and silt with occasional pits, can be morphologically flood water from being spread in the floodplains and pro-
described as a gently slopped alluvial floodplain with wide, tects banks from erosion which eventually prevents widening
multithreaded, braided rivers, namely, Mahananda, Balason, and encourages bed lowering (Gregory 2006). Therefore, to
Rakti, Rohini, and small streams such as Chamta, Pancha- measure the extent of lateral disconnectivity, the lengths of
nai, and Gulma Khola. The region receives heavy rainfall concreate embankments were measured from the geospatial
of more than 2500 mm annually, making the region prone dataset, and thereafter, percentage of riverbanks protected
to annual floods with high sediment yields (Gansser 1964; by embankments (EM) was calculated (Huang et al. 2019)
Sarkar 1989; De 1998; Tamang 2013; Roy 2011). using the below-mentioned formula:
After the demarcation of the unconfined floodplain region Length of the embanked river channel
for both rivers, segmentation into 10 equidistant reaches (5 EMi = ∗ 100 (1)
Total length of the banklines
reaches for each river), viz., Mahananda (R1-R5) and Bala-
son (R6-R10) with 4.5 km and 5.2 km lengths, respectively where EMi is the measured EM for ith reach.
(Goswami et al. 1999). The segmentation method has been Sediment extraction is another type of human interven-
preferred over other morphological aspect-based segmentation which can be detrimental not only the inherent chan-
tion methods primarily to have a larger number of reaches nel pattern but also to the river biota. Extraction, trapping,
as well as uniform reach sizes. If these channels were seg- or mining of extremely useful construction aggregates, i.e.,
mented based on morphological aspects, then the number riverine sediments, can lead to channel narrowing, transfor-
of reaches would be inadequate to perform any statistical mation of channel pattern, and change in habitat condition
analysis, and varied lengths of assessment units would also (Kondolf 1994). In the case of measuring this, both the areal
bring ambiguity to the assessment. Following the reach seg- coverage of extraction and the volume extracted are usually
mentation, their adjoining floodplain was delineated using a taken into account by previous researchers (Wishart et al.
1 km buffer from each bank, since it was observed that most 2008). However, in the absence of previous data related to
of the anthropogenic activities were confined to this region the volume of extracted sediment, only the overall extent
(Mitra et al. 2020). Demarcated adjoining floodplains were of sediment extraction is measured and assessed. Now, to
characterized by dense forest, tea gardens in the upstream calculate the percentage of a reach under sediment extraction
13
Fig. 1 Location map of the study area with segmented river reaches
13
Table 2 Length and adjoining land use of the segmented reaches of

∑
Numbers of extracted sub − reaches
Mahananda-Balason River SEi = ∗ 100 (2)
Total number of sub − reaches
Reach ID River Length (km) LULC
where SEi is the measured SE for ith reach.
Reach 1 Mahananda 4.5 V Land use transformation acts as an indirect modifier
Reach 2 Mahananda Fl, V of the channel planform as it slowly alters the sediment
Reach 3 Mahananda Ba, Fl and water discharge which ultimately impacts the chan-
Reach 4 Mahananda Ba nel planform (Paul & Meyer 2001). Land use transfor-
Reach 5 Mahananda Ba, Agri, Fl mation towards urbanization ultimately makes the runoff
Reach 6 Balason 5.2 Tg, V bedload-free and sediment-free “hungry water” actively
Reach 7 Balason Tg, V, Fl engage in vigorous downcutting which can lead to the nar-
Reach 8 Balason Fl, Tg rowing of the channel and transformation of the channel
Reach 9 Balason Ba, Fl pattern. Here, in this study, the aspect of land use transfor-
Reach 10 Balason Ba, Fl, Tg mation was addressed by measuring two separate indica-
Agri, agricultural land; Ba, built-up area; Fl, fallow land; Tg, tea gar-
tors, namely, the percentage of built-ups in the adjoining
den; V, vegetation floodplain (BA) and the percentage of vegetation cover
in the adjoining floodplain (VA) (Chin & Gregory 2005;
Comiti et al. 2011; Del Tánago et al. 2015), and to meas-
(SE), firstly, each reach was subdivided into 10 smaller seg- ure these, single-class image classifications using support
ments, and thereafter, the extraction activities were traced vector machine (SVM) algorithm were performed for the
sub-reach-wise. Finally, the percentage of a reach with active images of 1989, 1998, 2008, and 2020 (Singh et al. 2014;
in-channel extraction was calculated (Rinaldi et al. 2005) Prasad et al. 2017). However, the datasets before 1989
using the following formula: were panchromatic and therefore could not be classified
Fig. 2 Ongoing anthropogenic

activities in the floodplain of
Mahananda and Balason
13
rather a patch-wise digitization was done and the percent- images and, for 2008 and 2020, Google Earth images are
age of overall patch area was calculated. The percentages with finer resolution. The CW and CCB were calculated
were estimated using the following formulae: using the following formulae:
The are aunder built_up of a reach
∑
Width measured
BAi =
Total area of the adjoining f lood plain that reach
∗ 100 CWi = (6)
Total number of observation
(3)
The are aunder vegetation cover of a reach
∑
Area covered by the barin the channel of a reach
VAi = ∗ 100 CCBi = ∗ 100
Total area of the adjoining f loodplain that reach Total area of the channel of that reach
(4) (7)
where BAi and V Ai are the measured BA and VA for ith where CWi and C CBi are the measured CW and CCB for
reach. ith reach.
Longitudinal disconnections are immensely important Lastly, the dynamicity of the channel pattern was quanti-
in studying human intervention on rivers, as it is the single fied through the Braiding Index after Brice (1964) (BI) using
most dominant factor which can disturb the entire riverine the following formula:
system from its morphology to its ecology. However, in ∑
2( Length of all the islands and bars in a reach)
this study area, no such significant longitudinal obstruc- BIi =
tion was observed; however, structures altering longitudi- Total length of the reach
nal connectivity in the form of bridge pillars were often (8)
found. Hence, the longitudinal alteration was measured in where BIi is the measured BI for ith reach.
terms of the occurrence of crossing structures on the river The lengths of the reaches were measured midway
channel per km (CS) (Rinaldi et al. 2013). The rationale between the banks in the channel belt, and similarly, the
behind the inclusion of this indicator was that these cross- lengths of the islands and bars were also measured follow-
ing structures alter the channel in two ways; firstly, its pil- ing its longest axis (Friend & Sinha 1993; Goswami et al.
lars and piers obstruct and deflect the flow, and secondly, 1999). During all of these measurements, no data gap was
it provides easy access between two banks, influencing observed. The changes measured through these indicators
the growth of settlements or increase in impervious land- were further statistically tested using a two-tailed t-test with
scapes. Initially, the crossing structures were identified a 95% confidence interval.
from the satellite images; then, the occurrences of such
were measured per km using the following formula: Assessing the relationship between anthropogenic
alteration and planform change
Number of crossing structures in a reach
CSi = (5)
Total length of that reach Regression analysis was performed to assess the influence of
anthropogenic interventions on channel planform and since
where CSi is the measured CS for ith reach.
the structure of the data had spatial as well as temporal reso-
Measurements of the anthropogenic interventions were
lution, panel data regression was opted. Panel data are also
succeeded by the estimation of planform attributes which
known as longitudinal or cross-sectional time-series data
addressed two major aspects of channel morphology and
having observations of the same spatial units for different
channel pattern; however, due to the lack of previous data,
periods (Chamberlain 1982; Hu & Schiantarelli 1998; Naudé
channel hydrology was not included in this study. Firstly,
& Saayman 2005; McManus 2015). It is found useful as it
the channel area bounded between two banks (CA) was
deals with heterogeneous datasets and gives a standard effect
quantified by demarcating the channel banks from the
in longitudinal data. For this study, a fixed effect model (FE)
satellite images. Thereafter, using the same bank lines,
was used for regression analysis (Galvao 2011; Makuta &
the bank-to-bank channel width (CW) was measured at
O’Hare 2015) using the following equation:
every 50 m and averaged reach-wise (Rinaldi et al. 2013;
Yousefi et al. 2019). After these, two other indicators of yit = 𝛽1 Xit + 𝛼i + uit (9)
channel planform were measured as the total length of the
channels in km (CL) and areal coverage of bars in percent- where αi (i = 1…. n) is the unknown intercept for each entity
age (CCB) (Hajdukiewicz & Wyżga 2019). Demarcation (n entity-specific intercepts); Yit is the channel planform
of the bars from coarse resolution Landsat images (30 m properties, where i = entity and t = time. Xit represents one
spatial resolution), especially for the years 1989 and 1998, indicator of anthropogenic intervention; β1 is the coefficient
was accompanied by a Bare Soil Index (BSI). For the rest for that IV; uit is the error term.
of the years, i.e., 1968 and 1979, high-resolution Corona The Hausman specification test was performed to choose
the effects between FE and RE (random effects). The results
13
13
◂Fig. 3 Boxplots showing a temporal change in anthropogenic activi- embanked with ≥ 70% of their channel banks having large
ties and channel planform adjustments. a1 Percentage of riverbanks concretized embankments.
embanked. a2 Percentage of sediment mining in a reach. a3 Percent-
age of built-up area in the adjoining floodplain. a4 Percentage of veg-
Similarly, the measurements of SE showed a significant
etation cover in the adjoining floodplain. a5 The number of crossing increase for both rivers by 7400% (from nil in 1968 to 74%
structures per km. b1 Channel area in sq. km. b2 Channel width in m. in 2020) at Mahananda and 9600% (from nil in 1968 to
b3 Channel length in km. b4 Percentage of a reach covered by bars. 96% in 2020) at Balason. Therefore, it was evident from the
b5 Braiding Index
measurements that, like EM, for SE, the temporal change
was noticeably higher in Balason than that in Mahananda.
backed the suitability of the FE model over RE which For the initial 10 years (1968 and 1979), all the reaches of
explores the relationship between predictor and outcome both rivers were completely devoid of sediment extraction
variables within an entity. It helps to understand the rela- activities; however, in the next 10 years, i.e., 1979 onwards,
tionship between the multiple dependent and independent almost the entire unconfined Balason was engulfed with
variables within an entity on a longitudinal scale. As indi- active in-channel sediment extraction (from nil in 1979 to
vidual variables are unique in characters, it is quite difficult 85% in 1989) in comparison with a much lesser increase of 0
to understand every relationship among variables. Here, FE to 20% during 1979–1989 for Mahananda. However, all the
model gives a standardized result to understand the overall reaches of Mahananda did not follow similar trends regard-
relationship among the indicators, and the coefficients of ing SE, as the uppermost reach of Mahananda, i.e., R1, was
determinants (R2) would directly infer the intensity of the completely free from any extraction activities for the entire
effects of anthropogenic activities on every channel plan- study period, and the R4 and R5 were also free from any
form property. The analysis was performed primarily for extraction activities until 1989. In recent years, from 1998
the whole system, but since that comprises two major river onwards, a slight decrease in SE (from 80% in 1998 to 74%
systems, i.e., Mahananda and Balason, therefore, the analy- in 2008 and 2020) was seen exclusively in lower Mahananda
sis was further repeated for individual systems to achieve an at R4 and R5. However, withdrawal from the entire reaches
inter-system comparative assessment. was ongoing at Balason.
It was estimated that throughout the period, BA continu-
ously increased, and conversely, VA steadily decreased from
the adjoining floodplains of both rivers. The 15.8% of the
Results built-up area adjoining Mahananda in 1968 rose to 45.6%
in 2020; conversely, the floodplains of Balason were com-
Measurement of the indicators paratively less urbanized, where only 3.8% of the impervi-
ous area in 1968 increased to 27.8% in 2020. However, the
Quantification of the indicators related to anthropogenic increase for Balason was an overwhelming 613% compared
interventions and channel planform attributes (Fig. 3a) to a much lesser 188% for Mahananda (Table 3). The flood-
showed marked changes over the entire period. Measure- plain of Mahananda was already inhabited before the 1970s,
ments of EM for Mahananda and Balason portrayed a (15.8% in 1968); hence, the scope of expansion was limited
continuous increase from 10.1 to 60.6% and from 2.2 to there in comparison with that of Balason, which was almost
44.1%, respectively, during the entire study period (1968 free from any noticeable built-up patch until 1989 (8.8%).
to 2020). Although the channel banks of Balason were less Among the reaches, in 2020, the uppermost reaches of both
embanked than that of Mahananda, the change was nearly rivers, i.e., R1 and R6, were measured with minimum imper-
4 times higher at Balason (1904%) than that of Mahananda viousness (2.1%), and conversely, R3, R4, R5 and R9, and
(500%). It was observed that most of the reaches were devoid R10 had > 45% of their area covered with built-ups.
of any lateral obstruction till 1988 except the areas adjoining As a consequence of the massive expansion of the urban-
the Gulma railway station at R1, R4, and R5 along which ized landscape, 26.3% of vegetation cover at the floodplains
the Siliguri urban region is situated. But scenarios started of Mahananda, in 1968, was reduced by 20% and left with
changing from 1989 onwards, and the EM had increased 21.1% in 2020. The reduction was more rapid at Balason,
by > 135% at Mahananda and > 58% at Balason during the where 21.4% of vegetation cover in 1968 was reduced by
next 10 years (1989 and 1998). In 2020, there was hardly any 68% and only left with 6.9% in 2020 (Table 3). Among
reach found with the negligible presence of embankments, the reaches, only R3 and R7 were found with continuous
even the lowest embankment cover found in R6 was > 20% reduction of vegetation; otherwise, at all other reaches, it
of the entire channel banks of that reach, while reaches such fluctuated over time. Reaches such as R2, R4, R6, R8, and
as R1, R6, R7, and R9 showed relatively lower embankment R10 showed a slight increase in vegetation cover in 1989;
coverage (< 45%) on their respective channel banks, that of however, declined afterwards. Although the overall VA was
all other reaches such as R3, R4, R8, and R10 were vastly reduced significantly, from 1998 onwards, slight though,
13
the continuous increase in vegetation cover was noticed at
0.0002**
0.0098**
0.0010**
0.0001**
0.0001**
0.0000**
0.0057**
0.0173*
R5 and R9. Overall, in 2020, the R1 was largely covered
p value
0.5000
0.0694
with vegetation (> 80%), and except for this, no other reach
had > 20% vegetation cover with the lowest vegetation cover
measured at R10 with just 1%. The overall change was high-
(1968–2020)
Change in %
est at R8 (88%) and lowest at R1 (6%).
9600.0
1904.5
− 48.3
− 13.8
− 49.6
− 53.2
− 53.6
− 67.8
Estimated CS showed no change in Balason for the entire
0.0
613.2
study period; however, at Mahananda, changes occurred
from 2008 onwards due to the establishment of 2 new
6.2
73.6
79.0
306.0
7.8
0.2
6.9
27.1
96.0
44.1
2020
bridges (Table 3). The highest CS was observed at R4 with

1.1 bridges/km followed by 0.7 bridges/km at R5. Con-
versely, R2, R3, R7, R8, and R9 had no crossing structures
7.3
71.5
87.7
378.8
9.7
0.2
8.1
18.7
95.0
39.6
2008
over there. CS of R1, R6, and R10 remained constant as no

new crossing structures were added there. Other than this,
7.5
72.2
95.6
556.5
14.3
0.2
12.0
11.5
95.0
31.0
the CS of R4 and R5 had increased largely, owing to the

1998
addition of crossing structures on those reaches.

The estimated channel planform indicators (Fig. 3b)
7.8
70.8
114.0
644.8
16.5
0.2
15.4
8.8
85.0
19.5
1989
depicted that all the channel planform parameters for almost

all reaches had reduced nearly by 50% during the entire stud-
ied period. While the CA of Mahananda was reduced by
11.5
81.6
157.8
588.7
15.0
0.2
14.5
7.8
0.0
14.2
1979
41.5% (9.4 k m2 in 1968 to 5.5 k m2 in 2020) (Table 3), the

reduction was more intense at Balason by 53% (16.8 k m2 in
Balason
12.0
85.4
156.8
653.7
16.8
0.2
21.4
3.8
0.0
2.2
1968
1968 to 7.8 k m2 in 2020) from 1968 to 2020. In 2020, the

largest channel area was observed at R10 (1.9 km2) followed
by R7 (1.8 km2) and R8 (1.7 km2), and conversely, the small-
0.0092**
0.0022**
0.0001**
0.0377*
0.0286*
p value
0.2248
0.0799
0.2453
0.4121
0.0661
est channel area was measured at R4 (0.6 km2) followed

by R3 (1 k m2) and R9 (1.1 k m2). Overall the Balason has
relatively larger areal coverage than Mahananda and suffered
(1968–2020)
Change in %
heavier losses regarding CA also. Among the reaches, except

R1, CA of which increased by 28.3%, all other reaches expe-
7400.0
− 55.2
− 10.4
− 54.4
− 45.5
− 41.5
− 19.8
100.0
188.6
500.0
rienced a reduction of nearly 50% and reaches such as R9

followed by R4, R2, R8, R6, R7, and R3 even exceeded this
(Supplementary Table 1). Notably, R5 and R10 depicted a
Table 3 Estimated indicators, their temporal change, and test of significance
3.9
69.2
37.8
247.6
5.5
0.4
21.1
45.6
74.0
60.6
2020
much lower rate of reduction of their areal coverage owing

to lesser encroachment of channel banks there. The nature
5.1
68.0
55.0
398.4
8.8
0.3
22.8
39.2
74.0
43.2
of change fluctuated in most of the reaches except R3 and

2008
R4, where the CA were continuously declining for the entire

52 years. Other than this, except at R1, R2, and R7, a sharp
5.4
70.1
57.1
400.5
8.9
0.2
23.2
32.3
80.0
40.0
1998
decline of CA followed by a slight increase was observed

during 1968–1979 and 1979–1989, respectively. However,
from then onwards, all the CA had sharply declined in most
5.3
71.5
52.4
372.8
8.3
0.2
24.7
26.9
20.0
17.0
1989
of the reaches. The R1, with overall increased CA, and R2,
Significant at 95%; **significant at 99%
with increasing channel area till 1998 (slight narrowing

8.0
77.5
83.1
437.2
9.0
0.2
23.9
24.3
0.0
14.9
1979
observed between 1979 and 1998), were also significantly

Mahananda
narrowed during the last decade (2008–2020).

Similarly, the estimation of CW also showed that except
8.7
77.2
82.9
454.7
9.4
0.2
26.3
15.8
0.0
10.1
1968
for R1, which had widened by 14.3%, all other reaches expe-
rienced a considerable narrowing of 45.5% (454.7 m in 1968
CS (structures/km)
to 247.6 m in 2020) at Mahananda and 53.2% at Balason

(653.7 m in 1968 to 306 m in 2020) during 1968–2020. The
CA (sq. km.)
remarkable narrowing was observed at R9 followed by R4,

Indicators
CCB (%)
CL (km)
CW (m)
EM (%)
SM (%)
VA (%)
BA (%)
R2, R8, R6, R7, and R3, measuring > 50% in comparison
with R5 and R10 which exhibited relatively lower narrowing
BI
*
13
of 40.5% and 34.2%. Only R1 was measured with a slight (Supplementary Table 1). Alike to other planform indica-
widening of the channel by 14.3%. However, the CW of R1 tors, the changing pattern was not unidirectional, but rather
was also narrowed by 44% in the last decade (2008–2020). highly fluctuating in nature. Reaches such as R3, R4, R7, R8,
In 2020, the widest reach was R10 (380 m) followed by R9, and R10 had experienced mostly a continuous declin-
R7 (344.2 m) and R8 (343.7 m), conversely, the narrowing trend of bars until 1998–2008, where it had increased
est reach was R4 (156.7 m) followed by R9 (222.7 m) and slightly and then again declined during 2008–2020. Con-
R3 (224.3 m). Overall the Balason has wider channels than versely, R1 and R2 noticeably declined during 1968–1979,
Mahananda. In-depth analysis of temporal change regard- increased till 1998–2008, and then, again, declined in
ing CW revealed several distinct narrowing and widen- the later years. Interestingly, the lowermost reaches of
ing phases. The CW of the reaches such as R3, R5, R6, Mahananda and uppermost reach of Balason, i.e., R5 and
R8, R9, and R10 was narrowed during the initial 10 years R6, passed through the same trend of initial increase till
(1968–1979) which was somehow reversed during the next 1979–1989, a slight decrease during 1989–1998 and then
10 years (1979–1989), and after this slight widening phase, again increase in the later years. Noticeably, despite the
all reaches followed the same narrowing trend for the rest overall reduction, R4, R5, R6, R7, and R10 had gained CCB
of the study period. Only R4 was measured with continu- in the last decade (2008–2020).
ous narrowing for 52 years and R1 and R7 with continuous Changes in CL and CCB ultimately altered the BI for both
widening until 2008 and 1989 respectively. The R2 was also rivers which resulted in its lowering by 55% (8.7 in 1968 to
observed with multiple phases of widening (1968–1979 and 3.9 in 2020) for Mahananda and 48% (12 in 1968 to 6.2 in
1989–2008) and narrowing (1979–1989 and 2008–2020). 2020) for Balason, respectively; and the reach-wise, maxi-
Following this, the CL was also reduced for both the mum, and minimum reductions were observed at R2 (77%)
rivers, and the overall reduction for Mahananda and Bala- and R1 (8.3%). Notably, all 10 reaches portrayed a reduction
son was 54.4% (82.9 km in 1968 to 37.8 km in 2020) and of their respective BI, and except for R1 (reduced by only
49.6% (156.8 km in 1968 to 79 km in 2020), respectively. 9.5%), the CCB of all other reaches of both river systems
Unlike CA and CW, in this case, the loss of channel length is were reduced by > 40%. In 2020, the highest BI was found
higher in Mahananda than in Balason. Except for R1 and R5, at R8 (7.2, after reducing by 43.7% from 1968) followed by
which showed an increase of 13.7% and 41.5% respectively, R9 (7.1), R7 (5.8), and R10 (5.7). Conversely, the lowest
all other reaches had lost their channel length by > 35%. BI was observed at R4 (2.4, after reducing by 53% from
Reaches, such as R2, were measured with highest reduction 1968) followed by R5 (2.6), R3 (3.8) and R2 (4). Therefore,
in channel length of > 80%, followed by R3, R9, R7, R4, R6, it was evident that the Balason was more braided than the
and R10 with > 50% loss. Overall in 2020, the highest CL Mahananda; however, their overall reduction was more or
was observed at R8 (22.3 km) followed by R7 (16.9 km) and less similar (< 50%). Analysis of the temporal dynamics of
R7 (15.7 km), and the lowest was observed at R4 (4.8 km) BI revealed that R3, R4, R7, and R10 experienced a continu-
followed by R2 (6.6 km) and R3 (7.2 km). The dynamics of ous reduction of BI. Other reaches such as R2, R5, R6, and
CL however was immensely fluctuating in nature and except R8 also followed a similar trend with an occasional increase
R6, all other reaches had experienced several phases of loss during 1989–1998. The remaining reaches R1 showed
and gain regarding CL. Surprisingly, the R4 and R9, after highly fluctuating BI during these 52 years, and in the case
a continuous decline from 1968 to 2008, suddenly expe- of R9, it was observed that the BI increased initially during
rienced an increase in 2020. Reaches such as R1, R5, R7, 1968–1979 but sharply declined afterwards. Therefore, it
and R8 after a significant increase during the initial decade can be concluded that the reaches such as R2, R3, R5, R9,
(1968–1979) steadily lost their channel length in later years. R4, and R7, having declined BI of > 50%, were in a process
However, between 1979 and 2020, slight increases were also of reconfiguring from multi-threaded braided patterns to
visible in those reaches during 1989–1998. single-threaded sinuous ones.
The status of CCB was found almost opposite to the CL, Results of the T-test showed that (Table 3) for both rivers,
and where ever the channel length was lost, the bar cover- changes in EM and SE were statistically significant indicat-
age increased. That is why the overall lowering of CCB was ing a profound increase in lateral obstruction and sediment
not significant as the overall lowering was 10.4% (77.2% in extraction. Besides this, the change in built-up area was also
1968 to 69.2% in 2020) and 13.8% (85.4% in 1968 to 73.6% found statistically significant but only for the reaches of
in 2020) for Mahananda and Balason, respectively. In 2020, Balason. Since the floodplains of Mahananda were already
reaches such as R1, R7, R10, R8, R6, and R2 was meas- urbanized from the beginning of the study period, there-
ured with > 70% bar coverages, whereas R4, R9, R3, and R5 fore, the change became insignificant. However, changes
had < 70% covered with bars. Overall, all the reaches, except in VA and CS remained insignificant for both river sys-
R1 and R5 where the CCB was increased by 8.75% and tems. In response to these increased human interventions,
50.46%, respectively, witnessed the loss of bars by > 25% except CL and CCB, all other channel planform attributes
13
exhibited significant change for both these rivers. Changes intervention. Therefore, it was affirmed that the dynamic-
that occurred regarding CL and CCB remained insignificant ity of channel planform was profoundly explained by the
at Mahananda. selected variables of anthropogenic intervention. It is well
established that the multiplied human interference over riv-
Relationship between anthropogenic intervention ers had profoundly impacted the channel planforms of both
and channel planform unconfined Mahananda and Balason.
It was noticeable from the calculated R2 values that all the

unconfined reaches were taken into consideration, and then, Discussion
all the planform properties showed high correlations with
anthropogenic interventions. But when the two systems, i.e., Understanding the applicability
Balason and Mahananda, were evaluated separately, some of the methodology, its uncertainties,
inter-system differences could be observed (Table 4). For the and the novelty of the work
entire studied area, it was estimated that except for CCB with
an R2 value of 0.624, all other planform attributes showed The studies regarding the channel responses to human inter-
higher correlations with the interventions. CA showed the ventions have a very long history mostly related to the indus-
strongest correlation with the anthropogenic interventions trialization of Europe (Downs and Piégay 2019). Over time,
with an R2 value of 0.831 followed by CW. In the case of the perspective of such studies has shifted from studying the
Mahananda, the highest R2 of 0.784 was calculated for CA, impact of a single factor dominating the flow regime to the
and the lowest of 0.595 for CL. At Balason, all the planform cumulative impact of several factors on channel hydromor-
properties except CCB were highly correlated with the ongo- phology (Gregory 2006). With time, studies dealing with the
ing anthropogenic interventions. The highest and lowest R2 cumulative impact approach have gained acceptance since
values were observed for CA (0.799) and CCB (0.463). For it is a wholesome study of all factors ongoing in the catch-
both the rivers, CA and CW had similar strong correla- ment and registering the responses of channels (Downs and
tions but CL showed better correlations for Balason than Piégay 2019). Following this trend, this study has adopted
Mahananda. Conversely, CCB was not even correlated with this cumulative impact assessment methodology, and after
the anthropogenic interventions for Balason but was found a thorough background study, the ongoing anthropogenic
highly correlated at Mahananda. Similarly, the dynamicity of interventions are identified and correlated with the chan-
BI had stronger correlations at Mahananda than at Balason. nel planform properties using robust statistical techniques.
These correlations indicate that, at both channels, significant However, due to insufficient data, we have failed to ana-
change had happened regarding CA and CW in response to lyze the responses at the catchment scale and only limited
the increasing human interventions. However, that change in it to a reach scale study. It is also noteworthy that most of
CL was only correlated at Balason since, due to narrowing, the anthropogenic alterations ongoing here are localized in
it had lost its low flow channels largely. Contrary to this, the nature, so this study can be designated as a case study repre-
CCB was not found to have a significant correlation with senting the human alterations and related channel responses
human intervention since, in the case of Mahananda, there of the sub-Himalayan rivers. Globally, past studies mostly
were already fewer bar areas, and in the case of Balason, used expert judgements in declaring anthropogenic inter-
although the bar coverage was lost, the percentage cover ventions as the dominant factor responsible for the change
remained similar due to drying up of low flow channels. in channel pattern (Downs and Piégay 2019); however,
Changes in CL and CCB were ultimately reflected in the sig- this study has statistically tested the correlation between
nificant reduction of BI in response to the increasing human synchronization of the intensity of human alterations and
change in the channel over time. Hence, this study has not
followed the long tradition of using inductive reasoning in
Table 4 Estimated R2 of each planform indicators hypothesis formation unlike to majority of the past research
Indicators Mahananda-Bala- Mahananda Balason which used inductive techniques to interpret their results
son system (Baker and Twidale 1991). It is evident that much of the
R2 R2 R2 studies like this are now heavily dependent on using freely
available geospatial databases and GIS solutions (Gregory
CA (sq. km.) 0.831 0.784 0.799
and Lewin 2015), and this tendency is pushing traditional
CW (m) 0.806 0.778 0.790
field-based studies into the backseat. To maintain a bal-
CL (km) 0.745 0.595 0.673
ance between these two types of assessment procedures, in
CCB (%) 0.624 0.734 0.463
this study, rigorous ground verification of results generated
BI 0.737 0.758 0.658
after analyzing the geospatial datasets was performed. Such
13
rigorous field survey has developed a sound understanding Hausman specification test must be performed. Neverthe-
of a direct pressure response approach between the interven- less, this study successfully estimated the ongoing anthropo-
tions and channel adjustments in a data-scarce region. In the genic alterations and channel adjustments of the unconfined
sub-Himalayan region, data related to riverine flow is highly Mahananda-Balason system. It had precisely identified the
restricted due to their transboundary nature; in such cases, impacts of such alterations in channel planform properties.
assessing the anthropogenic interventions from those data Therefore, the limitations of this study lie in the localized
could not be achievable. The use of a geospatial database spatial frame and small temporal extent. Due to the least
gives a chance to overcome this problem by studying the amount of previous data present for this study area, time
channel forms since the planform of a channel is the ultimate frames (1968–2020) cannot be extended in the past. The
result of the complex interplay between sediment and water availability of the geospatial dataset is so coarse that ana-
flow. Unique, fragile sub-Himalayan river systems are facing lyzing with this in hilly regions is also non-viable. Due to
serious anthropogenic threats at present; this study addresses the unavailability of past hydrological conditions, temporal
such threats and analyzed their impact on channel planform analysis of that also becomes unachievable. Future endeav-
by using robust statistical techniques. The regression tech- ors using this methodology would involve assessing the riv-
nique used here was found to be extremely helpful in tempo- ers at the basin scale for a more in-depth understanding of
ral as well as spatial dimensions. The results directly inferred human alterations on channel adjustments. More robust, in-
the relations between anthropogenic alterations and chan- depth studies on the impacts of such adjustments in socio as
nel planform properties throughout the studied period. This well as bio-hydrological domains are required.
can be beneficial in identifying the key factors of channel
adjustment and thus assists river managers in implement- Analyzing the extent of anthropogenic activities
ing river regulatory measures. Therefore, the novelty of and their impacts on river planform
this research lies in three pillars, i.e., assessing the channel
adjustments of two highly intervened sub-Himalayan river As discussed earlier, the studies related to channel responses
systems where data related to flow and past conditions are against human alterations have a strong knowledge domain,
unavailable. The deductive approach adopted in this study and by using that a-prior knowledge, the responses regis-
gives us a vivid picture of the future of these rivers with tered in this study can be linked with anthropogenic altera-
such immense human alterations. Studies related to channel tions. Worldwide studies on this topic have suggested that
responses against human intervention have identified lon- the “great acceleration phase” of anthropogenic interven-
gitudinal discontinuity and sediment extraction as the two tions on rivers lasted from 1955 to 1990; however, in the
most dominant factors which can alter the entire river system Indian scenario, that phase was initiated after the late 1990s,
(Gregory 2006; Downs and Piégay 2019). However, other induced by economic reforms of 1991 after which industri-
factors such as embanking, removal of vegetation cover, and alization and infrastructural development gained pace which
scouring of the bridge pillars are designated as localized directly impacted the dense river network present nationwide
factors which are capable of altering the fluvial and sedi- (Das 2015; Mitra et al. 2020; Rădoane et al. 2017; Gib-
ment regime at the reach scale. In this context, this study has ling 2018). Researchers further identified that in that phase,
analyzed both the dominant factor and localized factor in the channels respond quickly as those were highly sensitive
unconfined plains of Mahananda-Balason. towards any interventions; however, as the great accelera-
The uncertainties or limitations of this study lie in its tion phase is over, a “morphologically novel configuration”
prerequisites which require considerable prior knowledge of channel planform shall evolve (Downs and Piégay 2019).
of the ongoing alterations and channel adjustments based At this stage with continuous human alterations, the feed-
on which indicators are to be set. Selection of the indica- back mechanism of the river system will be disrupted, and
tors without any prior investigations on these could provide the earth’s processes shall be suppressed by the overwhelm-
faulty results. The unnecessary inclusion of factors that are ing anthropogenic interventions (Chin et al. 2016). Follow-
not ongoing in the study area can mislead the entire study. ing this, the evolutionary trajectory of the Mahananda and
Careful selection of the time frames is also required since Balason river systems can be divided into two phases, i.e.,
when we analyze the change in channel pattern for a long the phase before the acceleration phase (1968–1989) and
period, the timings of natural calamities must be taken care the acceleration phase (1998–2020). Since the acceleration
of. Natural calamities with immense power can drastically phase is still on, no post-acceleration phase is evident.
change all the channel planform properties by overwhelm- In the initial years, i.e., 1968 to 1989, the area was mostly
ing the effects of human interventions; therefore, those tim- characterized by the least presence of embankments, the par-
ings are to be avoided. Another potential source of error tial or complete absence of sediment extraction, negligible
can be the selection between the models performing ran- rate of urbanization, higher presence of vegetation covers,
dom effect and fixed effect, and to resolve this issue, the and the existence of crossing structures in lesser numbers.
13
This scenario started getting reversed in the later years, i.e., the 1960s, the landscape transformation, in terms of change
1998 onwards; when large-scale immigration took place in built-up and vegetation cover, was not found significant
as a result of rapid infrastructural development surround- there. But at the adjoining floodplain of Balason, the change
ing Siliguri (Datta 1998). The immigrated populace mostly in built-up was found significant as the Siliguri urban area
chose the distant rural areas to settle than the city center, was rapidly engulfing its rural–urban continuum.
especially the adjoining floodplain of Balason (Ghosh 2018), Before the phase of accelerated human interventions, the
and thus, those areas started experiencing the urbanization channels were highly braided and dynamic. Since flooding
process. Such newly built settlements were mostly seen is an annual phenomenon in the “Terai,” the dynamicity
upstream of Mahananda (R2, R3) and at the banks of Bala- regarding the channel planform was eventual. The imprints
son (R7, R8, and R10). Besides this, with increasing urban- of the largest flood of North Bengal, which happened in
ization, the demand for construction aggregates had also October 1968, were visible throughout this period, but with
increased manifold which fueled the sediment withdrawal the onset of the acceleration phase, such imprints were eradi-
from Balason and Mahananda. As a result, a large portion cated quickly from the plains (Mitra et al. 2023) and chan-
of the immigrant populace had chosen sediment extraction nel planform witnessed a substantial narrowing of > 50%,
as their livelihood and started living adjacent to the rivers lowering of bars and length of low flow channels, and ulti-
mostly upstream of Mahananda and the entire Balason. mately changing the channel pattern from braided to sinuous
Now, to provide a safe allocation to these newly built set- (Fig. 4) (Stecca et al. 2019). During the analysis of channel
tlements, concretized embankments were introduced, and planform dynamics, it came up that there were 3 distinct
with time, more than half of the unconfined channel banks phases of increase and decrease of channel planform prop-
were protected with large concretized embankments even erties. During the pre-acceleration phase or in the initial
exceeding the height of 13 m (observed in R2, R3). As the years (1968–1989), a steady decrease in terms of CA, CW,
extent of urban areas increased, the vegetation decreased, and CL was noticed, followed by a sudden increase in the
and the numbers of permanent as well as temporary cross- middle years (1989–1998) and thereafter steady decrease
ing structures increased, and increasing bridge density in the later years (1998–2020). When a flood occurs, the
became a common demand of all newly settled dwellers. channel-forming processes get very active and reconfigure
Since the floodplain of Mahananda was already urbanized in the channel vigorously (Surian et al. 2009a, b). However,
Fig. 4 The flowchart showing the phases of changes noticed in the study area, the ongoing anthropogenic interventions, and their effect on chan-
nel planform as well as on the hydro-biology and socio-hydrology
13
just after the floods, the river tends to get back to its previ- with far lesser discharge and excessive human alterations,
ous equilibrium condition, and in this process, it loses some the lower reaches of Mahananda have almost lost their inher-
of the planform properties that it had achieved during the ent braided properties and others are also on the verge of
floods (Surian et al. 2009a, b). The insignificant and sporadic losing them (Fig. 5) .
reduction of planform properties in the pre-acceleration
phase can be justified by this notion as when the flood of Perceiving the effects of planform adjustment
1968 was over, both the rivers started regaining their pre- in the Mahananda‑Balason River system
flood condition, and during this, it had become narrower
with reduced channel length (Mitra et al. 2023). However, it Effects of human intervention-induced planform modi-
had not lost its inherent braided nature during this reduction fications, witnessed mostly in the lower reaches of the
phase. After these initial years, a slight increase in channel Mahananda-Balason, are proven detrimental to the resource
planform properties was measured which could be linked potency of these rivers. Not only the resources which are
to medium-scale annual flooding that happened in 1980, directly collected from the river such as riverine sediments,
1993, and 1998 (Roy 2011; Ziliani et al. 2020). However, the fish, and water but also those which are indirectly dependent
effects or widened channels with increased channel length on the riverine flow such as hyporheic exchange, ground-
hardly lasted after 1998, and by 2020, all the flood marks water levels can also be affected by this intense human
had disappeared from the floodplains (Mitra et al. 2022). interference (Safeeq and Fares 2016). During the interview
Notably, the rivers were still flooded annually, but the impact with local sediment extraction workers, it had come up that,
of such could not be perceived from the measurements of mostly, the reaches of the downstream sections of both riv-
channel planform properties (Mitra et al. 2023). Therefore, ers, i.e., R4, R5, R9, and R10 were witnessing a shortage of
it can be summed up that before the acceleration phase, the sediments of the required grade as excessive extraction of
naturalness of the entire system was maintained, which can that might have disrupted the sediment recharge discharge
be perceived by fluctuating width, channel area, channel cycle (Kondolf 1994). In the studied region, extraction of the
length, and Braiding Index since anthropogenic interven- riverine sediment, chiefly used as construction aggregates, is
tions were negligible (Bertoldi et al. 2010; Shampa and Ali the major source of income for a large populace. Now, if the
2019) . Now, as the acceleration phase set in, due to intense channels become devoid of sediments of the required quality,
human interventions, channel started reconfiguring towards mostly pebbles and cobbles, then the existing workers will
narrowing very sharply. When the river system stays in a start losing their jobs or getting underpaid. At present, the
natural state, all such narrowing and reduction of channel workers are getting underpaid at such reaches which in turn
length and BI maintain a cyclic order following floods (Stan- is driving them in search of new possible extraction sites at
ley et al. 2002; Dewan et al. 2017; Sholtes and Doyle 2011) . the upstreams. Excessive extraction of pebbles and cobbles
However, anthropogenic interventions do not act in a cyclic releases finer particles in the water which eventually clog
order, rather they increase with time, and with intensive the river bed and disturb the exchange between channel flow
interventions, the channel planform could not get back to and the hyporheic zone (Cunningham et al. 1987; Schälchli
its pre-acceleration phase, even after regular flooding (Best 1992). In several case studies, it was observed that after
2019; Mitra et al. 2023; Biswas et al. 2020). This system is the removal of alluvial deposits, the bedrock had emerged
a precise example of this phenomenon. (Zawiejska and Wyżga 2010; Huang et al. 2014); however, in
Although both these river systems have undergone similar this case, due to annual replenishment of sediments induced
phases of anthropogenic interventions, they have not reacted by floods and thick quaternary alluvial deposit, the bedrock
similarly. The floodplain of Mahananda was urbanized and still has not emerged and the alluvial deposits of cobbles
anthropogenically intervened from the initial years (the and gravel mixed with finer sand remained predominant.
1960s). However, Balason experienced such interventions The hyporheic zone, the active ecotone between the sur-
only after 1989. Therefore, the evolutionary trajectory of face flow and groundwater flow that helps in maintaining
these rivers could not be similar. Since the floodplain of the biota assemblage of the stream as well as of the soil,
Mahananda was already urbanized, although, with much is, therefore, getting disturbed which in turn will disturb
lesser intensity, the channel planform somehow started the biotic structure of the entire ecosystem (Boulton et al.
reshaping from long before. Conversely, the floodplains of 1998; Banks et al. 2019). Besides this, the population of the
Balason, initially free from urbanization experienced altera- indigenous fish species is decreasing due to their disturbed
tions, and the channel planform responded in sharp narrow- habitat, which is a result of vigorous scouring (Mahapatra
ing and lowering of BI. Balason has a larger catchment than et al. 2014; Sarkar 2021). Consequently, people associated
Mahananda (Mitra et al. 2022); therefore, it eventually car- with fishing are being forced to switch their occupations.
ries a larger volume of water and sediment which also helps The majority of the fish species found in the Mahananda-
Balason maintain its inherent braided properties. However, Balason system are the bottom feeders and mostly prefer
13
Fig. 5 Change in channel planform from 1968 to 2020. Massive river banks show a remarkable increase. a Balason at Tari in 1968.
channel narrowing along with a lowering in the number of low-flow b Balason at Tari in 2020. c Mahananda at Champasari in 1968. d
channels and bars can be seen. Conversely, embankments along the Mahananda at Champasari in 2020
pebbly bottoms (Paul et al. 2009); hence, with the altered they also blamed the pouring of urban sewage of Silig-
silty and even muddy channel bottom, it can be detrimental uri in this river system for this. Mitra et al. (2020) found
to the indigenous fish species (Brown et al. 1998; Osmund- that, in the last 31 years (1987 to 2018), both rivers had
son et al. 2002; Wohl 2006). Paul et al. (2009) observed 49 incised > 2 m on average which was thoroughly supported
fish species in Mahananda-Balason in 2007–08; however, by the observations of Tamang and Mandal (2015). Dur-
with increasing disturbance in their habitat, the population ing the field survey in 2020, it was noticed that the incision
of those species were largely lowered as witnessed by the process had further intensified in the lower parts of Bala-
local fishermen. The interviews with them affirmed that a son and Mahananda may be in response to the accelerated
sharp decline in the fish population of edible fishes in the sediment withdrawal and secondary embankment building,
Mahananda-Balason system in the last 15–20 years was which eventually resulted in the discharge of vadose water
noticed; however, besides in-channel sediment withdrawal, in channels, leaving a shortage of groundwater for the local
13
populace (Bravard et al. 1997, Sophocleous 2002; Biswas uncontrolled, and ill-planned sediment withdrawal from the
2016). Even after a heavy downpour during the flood season, vicinity of the bridge pillars’ and increased water current
people living in the adjoining floodplains are facing a lower- during monsoon destabilizes the bridge pillars by scouring
ing of water level in their wells immediately after the rain their bases (Kondolf 1994; Rinaldi et al. 2005). Not only do
stops. The lower channel bed might have affected the bridge the bases of the bridge pillars get scoured but also the bases
pillars also as those are seen tilted which causes disturbance of embankments get destabilized due to toe erosion and bank
regarding transport connectivity (Fig. 6). Earlier in 2010, failures occur inevitably. During the flood season, this has
such an incident was witnessed as the Matigara-Balason been a regular phenomenon in the studied region making the
railway bridge collapsed which created railway connectiv- populace settled at the channel banks vulnerable to losing
ity issues (Tamang 2013). Very recently, in the monsoon their houses (The Economic Times 2018; Dhali et al. 2020;
of 2022, a similar destabilization of bridge pillars of the The Telegraph India 2022).
Matigara-Balason road bridge immensely disturbed the local Now, with the increasing demand, resource extraction
connectivity (NDTV 2021). The coupled effect of excessive, or other human interventions are becoming profound in
Fig. 6 Condition of
Mahananda-Balason system
in different regions; a unal-
tered channel conditions at the
upstreams, near the debouching
point of Balason at Dudhia;
b similar conditions prevail
near the debouching point
of Mahananda at Gulma; c
embankment with bed stabiliz-
ing structures at Balason; d
13-m embankment at the bank
of Mahananda in Salugara; e
bridge pillars altering the flow
at Airview mode in Mahananda;
f destabilized bridge pillars in
Mahananda at Champasari; g
bar skimming at Balason near
Tarabari; and h massive sedi-
ment extraction is ongoing from
the Balason River at Matigara
13
the upper reaches which may pose a threat to the resource in channel length were observed. Inevitably, the existing
potency of those reaches. Increasing impervious areas with braided pattern is altered, and both the channels are gradu-
decreasing vegetation cover in the reaches such as R1, R2, ally becoming single-threaded sinuous from broad multi-
and R8 are the initial indications that, in the future, it can threaded ones. The upper reaches of both rivers (R1, R6),
attract more new settlements, bank and bed reinforcements, which offered relatively higher hindrances to human acces-
sediment withdrawal, and crossing structures (Mitra et al. sibility, were observed with a lower degree of channel altera-
2020). Now, if the source or the upstream region of a river tion and the rest of the portions, i.e., the lower reaches (R3,
system got severely compromised, then the entire down- R4, R5, R7, R8, R9, and R10) were highly altered owing
stream section could lose its resource potency (Tombolini to the higher accessibility to the riverine resources. These
et al. 2014). Since rivers as an integrated natural system channel adjustments induced by anthropogenic alterations
do not follow the imposed administrative boundaries or any are posing threats to the resource potency of the rivers. Fol-
spatial extent, the present localized incision process soon lowing that, many socio as well as bio-hydrological hazards
will be widespread further downstream leading towards re- could occur in the form of the scarcity of groundwater, mass
attaining a new equilibrium and the effects of such vigor- extinction of indigenous fishes, joblessness among the sedi-
ous incision shall be evident in the form of the scarcity of ment extraction workers, and disruption of transport connec-
sediments, groundwater and fish (Gregory 2006). Ill-planned tivity. To confront this, a target-oriented sustainable resource
resource exploitation strategies can be detrimental to the utilization policy must be adopted. Doing this long-term,
entire river system as soon as the carrying capacity of these an indicator-based multi-temporal study using freely avail-
reaches is overwhelmed. To confront this, an estimation of able satellite images from USGS archives was found cost-
a sediment budget should be executed before permitting the and time-effective than field-based conventional studies of
mining pits to withdraw (Walling and Collins 2008). Restric- riverine geomorphology in a data-scarce region. This will
tions on in-channel sediment extraction and the building of help in regular efficient monitoring of the river system and
new embankments must be imposed. No mining shall be identification of the alterations that need to be regulated and
permitted within proximity of the bridge pillars and other adjustments that need to be restored.
structures built on the banks. Furthermore, an integrated
Supplementary Information The online version contains supplemen-
river restoration strategy incorporating the local level stake- tary material available at https://doi.org/10.1007/s12517-023-11809-z.
holders should be practised to maintain the sustainable flow
of riverine resources within preserved multi-threaded chan- Acknowledgements We express our gratitude towards those who
nel planform (Overton et al. 2014). provided their logistic support during fieldwork. The authors also
acknowledge the two anonymous reviewers as their suggestions helped
in improving the overall quality of the manuscript.
Conclusions Funding This study received financial support (Fellowship Reference

No. 3128/NET-JUNE 2014) extended by the University Grants Com-
mission, India, to the first author.
This study has succeeded in estimating the extent of ongoing
anthropogenic interventions and channel adjustments of the Data availability The authors confirm that the data supporting the find-
Mahananda-Balason system. It has also precisely analyzed ings of this study are available within the article and its supplementary
the impact of such alteration in the planform adjustment materials.
of the studied river system. It is evident from this study
that the assessed unconfined reaches of Mahananda-Bal- Declarations
ason have experienced remarkable reconfiguration during Conflict of interest The authors declare no competing interests.
the studied period of 52 years (1968–2020) in response to
human interventions. The initial 30 years (1968–1998) were
almost devoid of any significant human interference, and
only a few pockets, mostly adjacent to Siliguri urban region, References
were observed with small built-up patches and negligible
embanking. However, the last 22 years (1998 afterwards) Ashmore P, Gardner J T (2008) Unconfined confluences in braided
were found with a rapid increase in anthropogenic alterations rivers. River confluences, tributaries and the fluvial network. In:
Rice SP, Roy AG, Rhoads BL (eds) River confluences, tributaries
in the form of urbanization, vegetation removal, sediment and the fluvial network , pp 119–147. https://doi.org/10.1002/
extraction, and embanking of the channel banks with con- 9780470760383.ch7
crete. The degree of channel adjustments is found in exact Ayaz S, Biswas M, Dhali MK (2018) Morphotectonic analysis of allu-
accordance with human interventions. Channel adjustments vial fan dynamics: comparative study in spatio-temporal scale of
Himalayan foothill India. Arab J Geosci 11(2):1–16. https://doi.
were negligible until the early 1990s, but after 1998, a mas- org/10.1080/24749508.2019.1568130
sive channel narrowing, loss of channel area, and reduction
13
Baker VR, Twidale CR (1991) The reenchantment of geomorphology. Anthropocene: emergence, status and implications. Earth Surf
Geomorphology 4(2):73–100. https://doi.org/10.1016/0169- Proc Land 42(1):71–90. https://doi.org/10.1002/esp.3943
555X(91)90021-2 Chamberlain G (1982) Multivariate regression models for panel data.
Bandyopadhyay S, De SK (2017) Spatio-temporal changes in pollution J Econ 18(1):5–46
status of the Haora River. In: Human interference on river health. Chavez PS Jr (1988) An improved dark-object subtraction technique for
Advances in Asian Human-Environmental Research. Springer, atmospheric scattering correction of multispectral data. Remote
Cham. https://doi.org/10.1007/978-3-319-41018-0_10 Sens Environ 24(3):459–479
Bandyopadhyay S, De SK (2018) Anthropogenic impacts on the mor- Chin A (2006) Urban transformation of river landscapes in a global
phology of the Haora River, Tripura, India. Géomorphologie: context. Geomorphology 79(3–4):460–487. https://doi.org/10.
Relief, Processus, Environnement 24(2):151–166. https://d oi.o rg/ 1016/j.geomorph.2006.06.033
10.4000/geomor phologie.12019 Chin A, Gregory KJ (2005) Managing urban river channel adjustments.
Banks V J, Palumbo-Roe B, Russell C E (2019) The hyporheic zone. Geomorphology 69(1–4):28–45. https://d oi.o rg/1 0.1 016/j.g eomo
In: Salik Javaid M (ed) Hydrology-the science of water. IntechO- rph.2004.10.009
pen. https://doi.org/10.5772/intechopen.85218 Chin A, Li A, Florsheim JL, Laurencio LR, Marston RA, Solverson
Benda L, Miller D, Bigelow P, Andras K (2003) Effects of post-wild- AP, Simon GL, Stinson E, Wohl E (2016) Investigating feed-
fire erosion on channel environments, Boise River Idaho. Forest backs in human–landscape systems: lessons following a wildfire
Ecol Manag 178(1–2):105–119. https://doi.org/10.1016/S0378- in Colorado, USA. Geomorphology 252:40–50. https://doi.org/
1127(03)00056-2 10.1016/j.geomorph.2015.07.030
Bendixen M, Best J, Hackney C, Iversen LL (2019) Time is run- Comiti F, Da Canal M, Surian N, Mao L, Picco L, Lenzi MA (2011)
ning out for sand. Nature 571:29–31. https://doi.org/10.1038/ Channel adjustments and vegetation cover dynamics in a
d41586-019-02042-4 large gravel bed river over the last 200 years. Geomorphology
Bertoldi W, Zanoni L, Tubino M (2010) Assessment of morphologi- 125(1):147–159. https://doi.org/10.1016/j.geomor ph.2010.09.
cal changes induced by flow and flood pulses in a gravel bed 011
braided river: the Tagliamento River (Italy). Geomorphology Cunningham AB, Anderson CJ, Bouwer H (1987) Effects of sed-
114(3):348–360. https://doi.org/10.1016/j.geomor ph.2009.07. iment-laden flow on channel bed clogging. J Irrig Drain
017 Eng 113(1):106–118. https://doi.org/10.1061/(ASCE)0733-
Best J (2019) Anthropogenic stresses on the world’s big rivers. Nat 9437(1987)113:1(106)
Geosci 12(1):7–21. https://doi.org/10.1038/s41561-018-0262-x Das RJ (2015) Critical observations on neo-liberalism and India’s new
Bhattacharya R, Dolui G, Chatterjee ND (2019) Effect of instream economic policy. J Contemp Asia 45:715–726. https://d oi.o rg/1 0.
sand mining on hydraulic variables of bedload transport and 1080/00472336.2014.1003143
channel planform: an alluvial stream in South Bengal basin Datta P (1998) Migration in India with special reference to Nepali
India. Environ Earth Sci 78(10):303. https://doi.org/10.1007/ Migration. Unpublished PhD thesis. Department of Economics.
s12665-019-8267-3 University of Calcutta, Kolkata, West Bengal. http://hdl.handle.
Biswas P (2016) West Bengal: in this village of 70 wells, 65 dry up net/10603/158869. Accessed 21 Dec 2022
in four months. The Indian Express. https://indianexpress.com/ De S K (1998) A study of the fluvial dynamics of the river Balasan.
article/elections-2016/cities/kolkata/west-bengal-in-this-villa Unpublished PhD Thesis. Department of Geography. University
ge-of-70-wells-65-dry-up-in-four-months-2754117. Accessed of Calcutta, Kolkata, West Bengal. http://hdl.handle.net/10603/
20 November 2018 159910. Accessed on 12th December 2019
Biswas M, Banerjee P (2018) Bridge construction and river chan- Del Tánago MG, Bejarano MD, de Jalón DG, Schmidt JC (2015)
nel morphology - a comprehensive study of flow behavior and Biogeomorphic responses to flow regulation and fine sediment
sediment size alteration of the River Chel India. Arab J Geosci supply in Mediterranean streams (the Guadalete River, southern
11(16):1–23. https://doi.org/10.1007/s12517-018-3789-7 Spain). J Hydrol 528:751–762. https://doi.org/10.1016/j.jhydrol.
Booth DB (1990) Stream-channel incision following drainage-basin 2015.06.065
urbanization 1. JAWRA J Am Water Resour Assoc 26(3):407– Dewan A, Corner R, Saleem A, Rahman MM, Haider MR, Rahman
417. https://doi.org/10.1111/j.1752-1688.1990.tb01380.x MM, Sarker MH (2017) Assessing channel changes of the
Boulton AJ, Findlay S, Marmonier P, Stanley EH, Valett HM (1998) Ganges-Padma River system in Bangladesh using Landsat and
The functional significance of the hyporheic zone in streams and hydrological data. Geomorphology 276:257–279. https://d oi.o rg/
rivers. Annu Rev Ecol Syst 29(1):59–81 10.1016/j.geomorph.2016.10.017
Bowen EE, Hamada Y, O’Connor BL (2014) Mapping mountain front Dhali MK, Ayaz S, Sahana M, Guha S (2020) Response of sediment
recharge areas in arid watersheds based on a digital elevation flux, bridge scouring on river bed morphology and geomorphic
model and land cover types. J Water Resour Protect 6(08):756. resilience in middle-lower part of river Chel, Eastern Himalayan
https://doi.org/10.4236/jwarp.2014.68072 foothills zone India. Ecol Eng 142:105632. https://doi.org/10.
Bravard JP, Amoros C, Pautou G, Bornette G, Bournaud M, Creuzé- 1016/j.ecoleng.2019.105632
desChâtelliers M, Tachet H (1997) River incision in south-east Downs PW, Piégay H (2019) Catchment-scale cumulative impact of
France: morphological phenomena and ecological effects. Regu- human activities on river channels in the late Anthropocene:
lated Rivers: Res Manag: An Int Devoted to River Res Manag implications, limitations, prospect. Geomorphology 338:88–104.
13(1):75–90. https://doi.org/10.1002/(SICI)1099-1646(199701) https://doi.org/10.1016/j.geomorph.2019.03.021
13:1%3C75::AID-RRR444%3E3.0.CO;2-6 Friend PF, Sinha R (1993) Braiding and meandering parameters. Geol
Brown AV, Lyttle MM, Brown KB (1998) Impacts of gravel mining on Soc, London, Special Publications 75(1):105–111. https://doi.
gravel bed streams. Trans Am Fish Soc 127(6):979–994. https:// org/10.1144/GSL.SP.1993.075.01.05
doi.org/10.1577/1548-8659(1998)127%3C0979:IOGMOG% Galster JC, Pazzaglia FJ, Germanoski D (2008) Measuring the impact
3E2.0.CO;2 of urbanization on channel widths using historic aerial pho-
Brown AG, Tooth S, Bullard JE, Thomas DS, Chiverrell RC, Plater tographs and modern surveys 1. JAWRA J Am Water Resour
AJ, Murton J, Thorndycraft VR, Tarolli P, Rose J, Wain- Assoc 44(4):948–960. https://d oi.o rg/1 0.1 111/j.1 752-1 688.2 008.
wright J, Downs P, Aalto R (2017) The geomorphology of the 00193.x
13
Galvao AF Jr (2011) Quantile regression for dynamic panel data with Kondolf GM (1994) Geomorphic and environmental effects of instream
fixed effects. J Econ 164(1):142–157. https://doi.org/10.1016/j. gravel mining. Landsc Urban Plan 28(2–3):225–243. https://doi.
jeconom.2011.02.016 org/10.1016/0169-2046(94)90010-8
Gansser A (1964) Geology of the Himalayas. Inter-science Publish- Kondolf GM (1997) PROFILE: hungry water: effects of dams and
ers, London gravel mining on river channels. Environ Manage 21(4):533–551.
Ghosh A (2018) The importance of Being Siliguri: border effect https://doi.org/10.1007/s002679900048
and the ‘untimely’ city in North Bengal. Logistical Asia. In: Kondolf GM, Piégay H, Landon N (2002) Channel response to
Neilson B, Rossiter N, Samaddar R (eds) The labour of making increased and decreased bedload supply from land use change:
a world region, pp 135–154 contrasts between two catchments. Geomorphology 45(1–2):35–
Ghosh S (2022) Geomorphic appraisal of active tectonics and fluvial 51. https://doi.org/10.1016/S0169-555X(01)00188-X
anomalies in Peninsular Rivers of the Bengal Basin (West Ben- Liébault F, Piégay H (2002) Causes of 20th century channel narrowing
gal, India). In: Shit PK, Bera B, Islam A, Ghosh S, Bhunia GS in mountain and piedmont rivers of southeastern France. Earth
(eds) Drainage basin dynamics: an introduction to morphology, Surf Processes Landf 27(4):425–444. https://doi.org/10.1002/
landscape and modelling, pp 503–541. https://doi.org/10.1007/ esp.328
978-3-030-79634-1_23 Macklin MG, Lewin J (2019) River stresses in anthropogenic times:
Gibling MR (2018) River systems and the anthropocene: a late pleis- large-scale global patterns and extended environmental timelines.
tocene and Holocene timeline for human influence. Quaternary Progress Phys Geography: Earth Environ 43(1):3–23. https://d oi.
1(3):21 org/10.1177/0309133318803013
Goswami U, Sarma JN, Patgiri AD (1999) River channel changes of Mahapatra BK, Sarkar UK, Lakra WS (2014) A review on status,
the Subansiri in Assam India. Geomorphology 30(3):227–244. potentials, threats and challenges of the fish biodiversity of
https://doi.org/10.1016/S0169-555X(99)00032-X West Bengal. J Biodivers, Bioprospecting Dev 2(140):2376–214.
Grable JL, Harden CP (2006) Geomorphic response of an Appa- https://doi.org/10.4172/2376-0214.1000140
lachian Valley and Ridge stream to urbanization. Earth Surf Makuta I, O’Hare B (2015) Quality of governance, public spending
Process Landforms 31(13):1707–1720. https://doi.org/10.1002/ on health and health status in Sub Saharan Africa: a panel data
esp.1433 regression analysis. BMC Public Health 15(1):932. https://doi.
Gregory KJ (2006) The human role in changing river channels. Geo- org/10.1186/s12889-015-2287-z
morphology 79(3–4):172–191. https://doi.org/10.1016/j.geomo McManus P (2015) February. Introduction to regression models for
rph.2006.06.018 panel data analysis. Indiana University Workshop in Meth-
Gregory KJ, Lewin J (2015) Making concepts more explicit for geo- ods. https://scholarworks.iu.edu/dspace/bitstream/handle/2022/
morphology. Prog Phys Geogr 39:711–727. https://doi.org/10. 21957/2 015-0 2-1 3_w im_m cmanu s_p anel_s lides.p df?s eque
1177/0309133315571208 nce=2. Accessed 21 Dec 2022
Hajdukiewicz H, Wyżga B (2019) Aerial photo-based analysis of Mitra S, Roy AK, Tamang L (2020) Assessing the status of changing
the hydromorphological changes of a mountain river over the channel regimes of Balason and Mahananda River in the Sub-
last six decades: the Czarny Dunajec, Polish Carpathians. Sci Himalayan West Bengal India. Earth Syst Environ 4(2):409–425.
Total Environ 648:1598–1613. https://doi.org/10.1016/j.scito https://doi.org/10.1007/s41748-020-00160-y
tenv.2018.08.234 Mitra S, Chatterjee A, De SK, Datta D, Tamang L (2022) Assess-
Hasanuzzaman M, Gayen A, Shit PK (2021) Channel dynamics and ing the hydromorphological condition of the Sub-Himalayan
geomorphological adjustments of Kaljani River in Himalayan Mahananda-Balason system using the hydromorphological qual-
foothills. Geocarto Int. https://d oi.o rg/1 0.1 080/1 01060 49. ity index. Adv Space Res. https://doi.org/10.1016/j.asr.2022.09.
2021.1882008 037
Hu X, Schiantarelli F (1998) Investment and capital market imper- Mitra S, Mondal M, Khatoon K, Oraon S, Tamang L (2023) Assessing
fections: a switching regression approach using US firm panel human control on planform modification over floods: a study of
data. Rev Econ Stat 80(3):466–479. https://doi.org/10.1162/ lower Mahananda-Balason River system, India. Floods in the
003465398557564 Ganga–Brahmaputra–Meghna Delta. Springer International Pub-
Huang MW, Liao JJ, Pan YW, Cheng MH (2014) Rapid channeliza- lishing, Cham, pp 127–160
tion and incision into soft bedrock induced by human activ- Moretto J, Rigon ELLF, Mao L, Picco L, Delai F, Lenzi MA (2014)
ity—implications from the Bachang River in Taiwan. Eng Geol Channel adjustments and island dynamics in the Brenta River
177:10–24. https://doi.org/10.1016/j.enggeo.2014.05.002 (Italy) over the last 30 years. River Res Applic 30(6):719–732.
Huang X, Liu J, Zhang Z, Fang G, Chen Y (2019) Assess river https://doi.org/10.1002/r ra.2676
embankment impact on hydrologic alterations and floodplain Nafchi RF, Samadi-Boroujeni H, Vanani HR, Ostad-Ali-Askari K, Bro-
vegetation. Ecol Ind 97:372–379. https://d oi.o rg/1 0.1 016/j. jeni MK (2021). Laboratory investigation on erosion threshold
ecolind.2018.10.039 shear stress of cohesive sediment in Karkheh Dam. Environmen-
Kang RS, Marston RA (2006) Geomorphic effects of rural-to-urban tal Earth Sciences, 80, 1–15. https://l ink.s pring er.c om/a rticl e/1 0.
land use conversion on three streams in the Central Redbed 1007%2Fs12665-021-09984-x. Accessed 21 Dec 2022
Plains of Oklahoma. Geomorphology 79(3–4):488–506. Naudé WA, Saayman A (2005) Determinants of tourist arriv-
https://doi.org/10.1016/j.geomor ph.2006.06.034 als in Africa: a panel data regression analysis. Tour Econ
Keen-Zebert A (2007) Channel responses to urbanization: Scull 11(3):365–391
and Mud creeks in Fayetteville. Arkansas Phys Geography NDTV (2021) Roads, bridges damaged as severe rains trigger land-
28(3):249–260. https://doi.org/10.2747/0272-3646.28.3.249 slides in North Bengal. https://www.ndtv.com/india-news/roads-
Khaleghi S, Surian N (2019) Channel adjustments in Iranian rivers: bridges-damaged-as-severe-rains-trigger-landslides-in-north-
a review. Water 11(4):672. https://doi.org/10.3390/w11040672 bengal-2582007. Accessed 22 October 2021.
Kiss T, Blanka V (2012) River channel response to climate-and Osmundson DB, Ryel RJ, Lamarra VL, Pitlick J (2002) Flow–sedi-
human-induced hydrological changes: case study on the mean- ment–biota relations: implications for river regulation effects on
dering Hernád River, Hungary. Geomorphology 175:115–125. native fish abundance. Ecol Appl 12(6):1719–1739. https://doi.
https://doi.org/10.1016/j.geomor ph.2012.07.003 org/10.5367/000000005774352962
13
Ostad-Ali-Askari K, Shayannejad M (2021) Quantity and quality Schälchli U (1992) The clogging of coarse gravel river beds by fine
modelling of groundwater to manage water resources in Isfahan- sediment. Hydrobiologia 235(1):189–197. https://doi.org/10.
Borkhar Aquifer. Environ Dev Sustain 23:15943–15959. https:// 1007/BF00026211
doi.org/10.1007/s10668-021-01323-1 Scorpio V, Rosskopf CM (2016) Channel adjustments in a Mediter-
Overton IC, Smith DM, Dalton J, Barchiesi S, Acreman MC, Stromb- ranean river over the last 150 years in the context of anthropic
erg JC, Kirby JM (2014) Implementing environmental flows and natural controls. Geomorphology 275:90–104. https://doi.
in integrated water resources management and the ecosystem org/10.1016/j.geomorph.2016.09.017
approach. Hydrol Sci J 59(3–4):860–877. https://doi.org/10. Shampa Ali MM (2019) Interaction between the braided bar and adja-
1080/02626667.2014.897408 cent channel during flood: a case study of a sandbed braided
Pal S, Talukdar S (2020) Assessing the role of hydrological modifica- river, Brahmaputra-Jamuna. Sustain Water Resour Manag 5:947–
tions on land use/land cover dynamics in Punarbhaba river basin 960. https://doi.org/10.1007/s40899-018-0269-x
of Indo-Bangladesh. Environment, Development and Sustain- Shit PK, Bera B, Islam A, Ghosh S, Bhunia GS (2022) Introduction to
ability 22(1):363–382. https://link.springer.com/article/https:// drainage basin dynamics: morphology, landscape and modelling.
doi.org/10.1007/s10668-018-0205-0. Accessed 21 Dec 2022 Drainage Basin Dynamics: an Intro Morphol, Landsc Model 1–9.
Paul MJ, Meyer JL (2001) Streams in the urban landscape. Annu Rev https://doi.org/10.1007/978-3-030-79634-1_1
Ecol Syst 32(1):333–365. https://doi.org/10.1146/annurev.ecols Sholtes JS, Doyle MW (2011) Effect of channel restoration on flood
ys.32.081501.114040 wave attenuation. J Hydraul Eng 137(2):196–208. https://d oi.o rg/
Paul M, Gupta S, Banerjee S (2009) Fish fauna of major rivers of 10.1061/(ASCE)HY.1943-7900.0000294
Darjeeling district, with special reference to their conservation Simon A, Rinaldi M (2006) Disturbance, stream incision, and channel
status. Record Zool Surv India 109(4):15–23. https://doi.org/10. evolution: the roles of excess transport capacity and boundary
26515/rzsi/v109/i4/2009/158973 materials in controlling channel response. Geomorphology 79(3–
Prasad SVS, Savithri TS, Krishna IVM (2017) Comparison of accuracy 4):361–383. https://doi.org/10.1016/j.geomorph.2006.06.037
measures for RS image classification using SVM and ANN clas- Singh SK, Srivastava PK, Gupta M, Thakur JK, Mukherjee S (2014)
sifiers. Int J Electr Comp Eng 2088-8708 7(3):1180. https://doi. Appraisal of land use/land cover of mangrove forest ecosystem
org/10.11591/ijece.v7i3.pp1180-1187 using support vector machine. Environ Earth Sci 71(5):2245–
Rădoane M, Perșoiu I, Chiriloaei F, Cristea I, Robu D (2017) Styles 2255. https://doi.org/10.1007/s12665-013-2628-0
of channel adjustments in the last 150 years. In: Radoane M, Sophocleous M (2002) Interactions between groundwater and surface
Vespremeanu-Stroe A (eds) Landform Dynamics and Evolution water: the state of the science. Hydrogeol J 10(1):52–67. https://
in Romania, pp 489–518. https://doi.org/10.1007/978-3-319- doi.org/10.1007/s10040-001-0170-8
32589-7_21 Stanley EH, Luebke MA, Doyle MW, Marshall DW (2002) Short-term
Righini M, Surian N, Wohl E, Marchi L, Comiti F, Amponsah W, changes in channel form and macroinvertebrate communities fol-
Borga M (2017) Geomorphic response to an extreme flood in lowing low-head dam removal. J N Am Benthol Soc 21(1):172–
two Mediterranean rivers (northeastern Sardinia, Italy): analysis 187. https://doi.org/10.2307/1468307
of controlling factors. Geomorphology 290:184–199. https://d oi. Surian N, Mao L, Giacomin M, Ziliani L (2009a) Morphological
org/10.1016/j.geomorph.2017.04.014 effects of different channel-forming discharges in a gravel-bed
Rinaldi M, Simon A (1998) Bed-level adjustments in the Arno River, river. Earth Surf Proc Land 34(8):1093–1107. https://doi.org/
central Italy. Geomorphology 22(1):57–71. https://doi.org/10. 10.1002/esp.1798
1016/S0169-555X(97)00054-8 Surian N, Ziliani L, Comiti F, Lenzi MA, Mao L (2009b) Channel
Rinaldi M, Wyżga B, Surian N (2005) Sediment mining in alluvial adjustments and alteration of sediment fluxes in gravel-bed riv-
channels: physical effects and management perspectives. River ers of North-Eastern Italy: potentials and limitations for chan-
Res Appl 21(7):805–828. https://doi.org/10.1002/r ra.884 nel recovery. River Res Appl 25(5):551–567. https://doi.org/10.
Roy S (2011) Flood hazards in Jalpaiguri district and its management 1002/r ra.1231
(Doctoral dissertation, University of North Bengal). Unpublished Surian N, Righini M, Lucía A, Nardi L, Amponsah W, Benvenuti M,
PhD Thesis. Department of Geography & Applied Geography. Viero A (2016) Channel response to extreme floods: insights on
University of North Bengal, Darjeeling, West Bengal. http://hdl. controlling factors from six mountain rivers in northern Apen-
handle.net/10603/149931. Accessed on 12th December 2019 nines Italy. Geomorphology 272:78–91. https://doi.org/10.
Safeeq M, Fares A (2016) Groundwater and surface water interactions 1016/j.geomorph.2016.02.002
in relation to natural and anthropogenic environmental changes. Talebmorad H, Ostad-Ali-Askari K (2022) Hydro geo-sphere inte-
In: Fares A (ed) Emerging issues in groundwater resources. grated hydrologic model in modeling of wide basins. Sus-
Springer, Cham, pp 289–326. https://doi.org/10.1007/978-3- tain Water Resour Manag 8(4):118. https://doi.org/10.1007/
319-32008-3_11 s40899-022-00689-y
Saha UD, Bhattacharya S (2019) Reconstructing the channel shift- Tamang L (2013) Effects of boulder lifting on the fluvial characteris-
ing pattern of the Torsa River on the Himalayan Foreland Basin tics of lower Balason basin in Darjeeling district, West Bengal.
over the last 250 years. Bullet Geography Phys Geography Ser Unpublished PhD thesis. Department of Geography and Applied
16(1):99–114. https://doi.org/10.2478/bgeo-2019-0007 Geography. University of North Bengal, Darjeeling, West Ben-
Sarkar S (1989) A geo-environmental appraisal of the upper gal. http://h dl.h andle.n et/1 0603/3 3129. Accessed on 12th Decem-
Mahananda basin of the Darjeeling Himalaya West Bengal. ber 2019
Unpublished PhD thesis. Department of Geography and Applied Tamang L, Mandal DK (2015) Bed material extraction and its effects
Geography. University of North Bengal, Darjeeling, West Ben- on the forms and processes of the lower Balason River in the
gal. http://h dl.h andle.n et/1 0603/1 82332. Accessed on 1 2 th Darjeeling Himalayas India. Geographia Polonica 88(3):393–
December 2019 405. https://doi.org/10.7163/GPol.2015.3
Sarkar T (2021) Coldwater fish diversity and their conservation status The Economic Times (2018) Another bridge collapses in West
in the Teesta, Jaldhaka, Torsa, Kaljani and Sankosh Rivers of Bengal.https://economictimes.indiatimes.com/news/politics-
the Dooars region, West Bengal India. Asian J Conserv Biol and-n ation/w atch-a nothe r-b ridge-c ollap ses-i n-w
est-b engal/v ideo
10(1):146–152. https://doi.org/10.53562/ajcb.WIMQ5691 show/65716181.cms. Accessed 22nd June 2022.
13
The Telegraph India (2022) Rain-induced landslides cut off Sikkim for Wohl E (2006) Human impacts to mountain streams. Geomorphology
10 hours. https://www.telegraphindia.com/states/rain-induced- 79(3–4):217–248. https://doi.org/10.1016/j.geomor ph.2006.06.
landslides-cut-off-sikkim-for-10-hours/cid/1870979. Accessed 020
22nd June 2022 Yousefi S, Moradi HR, Keesstra S, Pourghasemi HR, Navratil O, Hooke
Tombolini I, Caneva G, Cancellieri L, Abati S, Ceschin S (2014) Dam- J (2019) Effects of urbanization on river morphology of the Talar
ming effects on upstream riparian and aquatic vegetation: the River, Mazandaran Province Iran. Geocarto Int 34(3):276–292.
case study of Nazzano (Tiber River, central Italy). Knowl Manag https://doi.org/10.1080/10106049.2017.1386722
Aquat Ecosyst 412:03. https://doi.org/10.1051/kmae/2013085 Zawiejska J, Wyżga B (2010) Twentieth-century channel change on the
Vanacker V, Molina A, Govers G, Poesen J, Dercon G, Deckers S Dunajec River, southern Poland: patterns, causes and controls.
(2005) River channel response to short-term human-induced Geomorphology 117(3–4):234–246. https://doi.org/10.1016/j.
change in landscape connectivity in Andean ecosystems. Geo- geomorph.2009.01.014
morphology 72(1–4):340–353. https://doi.org/10.1016/j.geomo Ziliani L, Surian N, Botter G, Mao L (2020) Assessment of the geo-
rph.2005.05.013 morphic effectiveness of controlled floods in a braided river
Walling DE, Collins AL (2008) The catchment sediment budget as a using a reduced-complexity numerical model. Hydrol Earth Syst
management tool. Environ Sci Policy 11(2):136–143. https://d oi. Sci 24(6):3229–3250. https://d oi.o rg/1 0.5 194/h ess-2 4-3 229-2 020
org/10.1016/j.envsci.2007.10.004
Wiejaczka Ł, Bucała A, Sarkar S (2014) Human role in shaping the Springer Nature or its licensor (e.g. a society or other partner) holds
hydromorphology of Himalayan rivers: study of the Tista River exclusive rights to this article under a publishing agreement with the
in Darjeeling Himalaya. Current Science 717–724. https://www. author(s) or other rightsholder(s); author self-archiving of the accepted
currentscience.ac.in/Volumes/106/05/0717.pdf manuscript version of this article is solely governed by the terms of
Wishart D, Warburton J, Bracken L (2008) Gravel extraction and plan- such publishing agreement and applicable law.
form change in a wandering gravel-bed river: the River Wear
Northern England. Geomorphology 94(1–2):131–152. https://
doi.o rg/1 0.1 016/j.g eomor ph.2 007.0 5.0 03. Accessed 21 Dec 2022
13

Planform Adjustments of Sub Himalayan Mahananda Balason River Over Five Decades in Response To Anthropogenic Interventions

Uploaded by

Copyright:

Available Formats

Planform Adjustments of Sub Himalayan Mahananda Balason River Over Five Decades in Response To Anthropogenic Interventions

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Planform Adjustments of Sub Himalayan Mahananda Balason River Over Five Decades in Response To Anthropogenic Interventions

Uploaded by

Copyright:

Available Formats

Arabian Journal of Geosciences (2023) 16:676

Planform adjustments of sub‑Himalayan Mahananda‑Balason River

Received: 10 May 2023 / Accepted: 23 November 2023

Channel reconfiguration or channel planform adjustments

Mahananda and Balason, the two rain-fed tributaries of

Table 2 Length and adjoining land use of the segmented reaches of

Fig. 2 Ongoing anthropogenic

the continuous increase in vegetation cover was noticed at

est at R8 (88%) and lowest at R1 (6%).

bridges (Table 3). The highest CS was observed at R4 with

over there. CS of R1, R6, and R10 remained constant as no

the CS of R4 and R5 had increased largely, owing to the

addition of crossing structures on those reaches.

depicted that all the channel planform parameters for almost

41.5% (9.4 k­ m2 in 1968 to 5.5 k­ m2 in 2020) (Table 3), the

1968 to 7.8 k­ m2 in 2020) from 1968 to 2020. In 2020, the

est channel area was measured at R4 (0.6 ­km2) followed

heavier losses regarding CA also. Among the reaches, except

rienced a reduction of nearly 50% and reaches such as R9

much lower rate of reduction of their areal coverage owing

of change fluctuated in most of the reaches except R3 and

R4, where the CA were continuously declining for the entire

decline of CA followed by a slight increase was observed

with increasing channel area till 1998 (slight narrowing

observed between 1979 and 1998), were also significantly

narrowed during the last decade (2008–2020).

to 247.6 m in 2020) at Mahananda and 53.2% at Balason

remarkable narrowing was observed at R9 followed by R4,

It was noticeable from the calculated R2 values that all the

Conclusions Funding This study received financial support (Fellowship Reference

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

41.5% (9.4 k m2 in 1968 to 5.5 k m2 in 2020) (Table 3), the

1968 to 7.8 k m2 in 2020) from 1968 to 2020. In 2020, the

est channel area was measured at R4 (0.6 km2) followed