Classification of small scale landforms , its significance : a case study of the middle Ichamati river , India

INDIAN JOURNAL OF POWER & RIVER VALLEY DEVELOPMENT Landforms are the core concept of geomorphology. The definition of landforms, their characterization and classification are the core subject of geomorphology. But all these become complex when it seems to difficult to identify the landforms, especially when the area is plain land and highly modified by human activities. This paper has examined the characters of the landforms of the middle basin of the Ichamati river, the important distributary in the district of North 24 Parganas, India. It has been primarily taken an attempt to classify the landforms with the help of the satellite image, IRS P6 LISS II and LISS III. The DEM is not enough to identify the micro scale landform. To overcome this difficulty a series of field works have been conducted (2002, 2004, 2012 and 2015). The landforms have been classified according to second order derivative (Wood, 1996) method. Then ANOVA test has been applied to justify the classification. The F-statistics have indicated the effort is satisfying. The changing character of different landforms denote the river is going to be deteriorating from downstream to upward.

Landforms are the core concept of geomorphology. The definition of landforms, their characterization and classification are the core subject of geomorphology. But all these become complex when it seems to difficult to identify the landforms, especially when the area is plain land and highly modified by human activities. This paper has examined the characters of the landforms of the middle basin of the Ichamati river, the important distributary in the district of North 24 Parganas, India. It has been primarily taken an attempt to classify the landforms with the help of the satellite image, IRS P6 LISS II and LISS III. The DEM is not enough to identify the micro scale landform. To overcome this difficulty a series of field works have been conducted (2002, 2004, 2012 and 2015). The landforms have been classified according to second order derivative (Wood, 1996) method. Then ANOVA test has been applied to justify the classification. The F-statistics have indicated the effort is satisfying. The changing character of different landforms denote the river is going to be deteriorating from downstream to upward.

Introduction
T he analysis of landform is a lost art and seems to have become part of the drop-out concept of geomorphic thinking (Fryirs 2016) though it is a relevant issue for mapping and hazards analysis (Darmis 1991). Landforms are the 4D forms and 3D forms can be defined as a specific geomorphic features on the earth surface (Goudie 2004), ranging from large-scale to minor features (Slaymaker et al. 2011, Blaszezynski 1997. The study of landforms appears as the central or core concept of geomorphology (Thorn 1978), deal with specific geomorphic process (Crevenna et al. 2005). Landforms reflect the combine interaction of geology, change of climate (Garavaglia et al. 2009), time and biota also (Brierley et al. 2013;Dietrich and Perron 2006;Piacente 2003).
Geomorphology is concerned with the study of evolution of erosional landforms (Small 1978), interpretation of the forms rather than the process (Wooldridge and Morgan 1959) because geographically-trained geomorphologist are not well qualified to work in the field of process (Strahaler 1969). However, in the broader sense the constructional or depositional landforms have been included within the sphere of geomorphology (Small 1978). In the study of landform, there are two connotations, geomorphology and geomorphometry with sharp distinction. Geomorphology concerns with the study of form 'morphos' of the land (Thorn 1978), but geomorphometry is a sub discipline of geomorphology having an object of quantitative and qualitative description of landforms (Pike 2000), the geometric interpretation of landform (Angillieri 2008, Mesa 2006, Thorn 1978, Strahler 1964, 1957.
The Geomorphologist vastly studied about the multifaceted classifications of landforms summarized are, based on forms, i.e. convex, concave, rectilinear etc. (Chorley et al. 1984;Thorn, 1978), prototype landforms (Dawkins 2004 (Bloom 1991, Summerfield 1991. Geomorphic process being the cascading system, consisted of mass and energy which are directed through different systems (Chorley et al. 1984), among them river is an important molding element (Davis 1899). Nature represents two types of river systems, i.e. contributing river system and distributing river system (Chorley et al. 1984, Sen 1993 or detachment-limited and transport limited (Whipple and Tucker 2002) might have two different types of systemoperation. The philosophy behind the contributing systemoperation is to attain a quasi-equilibrium condition Leopold 1964, Mackin 1948), the equilibrium condition (Bloom et al. 2017); a minimum variance (Langbein and Leopold 1966), the concept of entropy in landscape evolution (Leopold andLangbein 1962, Leopold andWolman 1960). All these models concerns with decreasing height, though errosional process. On the other hand, in the distributing system-operation, rivers always try to increase its energy by depositing its load that represents the negative feedback system indeed (Parkaer and Sutherland 1990, Wilcock andSoutard 1989, Andrews andParker 1987) and the resultant landforms are the signatures of such effort.
The Ichamati river is a meander river in the Ganga-Brahmaputra Details (GBD) environment (Mondal et al. 2018, Mondal andSatpati, 2019). After the formation of GBD plain the Ichamati river superimposed on it (Mondal 2010) and produced its associated landforms by lateral accretion, island formation and channel abandonment (Chorley et al. 1984, Rice 1931. The Ichamati river has been bifurcated from the Mathabhanga river, a distributary of the Padma river (Rudra 2014) and subsequently the river has been anthropogenically delinked from the Padma-Mathabhanga-Churni river system in 1970s (Sarkar 2004). So the river gets hardly upstream discharge from its source stream, except rainy season. The Ichamati river has been intensively investigated by Mondal,  (Mondal 2016). The upper portion of the river is dead and the lower portion of the river is dynamic . The middle portion of the river belongs to the transition zone between these two flow regimes. Therefore, this study can explain how the landforms change spatio-temporal with the change of hydrodynamic of the river.
Traditionally geomorphology is concerned with the mesoscale landforms (Thorn 1978) and these landforms are unexplainable in respect of human lifetime (Chorley et al. 1984;Thorn, 1978). They have a little landform sensitivity (Chorley et al. 1984, Thorn 1978) and a small trigger of geomorphic process is not capable to alter their form immediately. So, it needs a reference of past episode to explain the form of meso-scale landforms (Thorn 1978). The micro-scale land forms have a significant role to overcome such problem because they played a delicate role to balance between form and process and reducing emphasis on the past (Thorn 1978). The objectives of this article are, (a) general discussion of the landforms, (b) the area is highly modified by the human activities. So DEM is not effective to classify the land form in micro level. A detail field survey and statistical analysis have attempted to classify the landforms in micro level, (c) to discuss about the spatio-temporal change of the landforms with the basis of hydrodynamic of the river. This paper is an attempt to discriminate the landforms mainly in the highly modified low-relief area, with a twin-venture of intensive fieldwork and statistical analysis, when DEM is not enough doing it.

PRIMARY DATA SOURCE
Land survey was conducted with a Dumpy level and beside these the cross profiles of river were taken with the help of eco-sounder in the lower reach to prepare a bedcontour map. For the purpose of Dumpy level survey regional Bench Mark (BM) has been selected at Swarupnagar (Tentulia: 4m), Baduria (near Block Development Office: 3.14m) and Basirhat (near Basirhat Irrigation Office: 3.14m) (Ogdahl et al, 2014). The tidal velocities (low tide and high tidal velocity) were measured by an AA price current meter in a calm condition at two stations i.e. Basirhat (15/08/2015) and Tentulia (16/08/2015). All the cross sections were not confined in the river channel from bank to bank, rather extended 500m on both sides of the channel.

MATHEMATICAL AND STATISTICAL TECHNIQUES
For classification of landforms I have used the second order derivative (Wood, 1996), to calculate the rate of change of the form, such as the length of the river (x), the width of the river or channel (y) and height of the landforms (z). I have classified the land forms as, meander scar or meander scroll (d 2 z/dx 2 >0, d 2 z/dy 2 =0), channel (d 2 z/dx 2 < 0, d 2 z/dy 2 =0) and also flood plain (d 2 z/dx 2 =0, d 2 z/dy 2 =0) (Wood, 1996)  For justification of the classification of the landforms, I have used the ANOVA one way to calculate the F factor, which is based on the difference between the data group {n 1 (x 1 -x) + n 2 (x 2 -x) + n n (x n -x)} and within the data group { (n 1 -1)s 1 2 + (n 2 -1)s 2 2 + (n n -1)s n 2 }. The abbreviations of the monthly rainfall is 959.4mm). In Basirhat region, there is an average yearly rainfall is 231.8 mm. The area is highly densely populated. People are by and large agriculturist here.

Result
The study area belongs to the deltaic plain of West Bengal. So, there are little variations of landform in the study area. The northern portion of the study area is under the mature deltaic plain, whereas the middle and lower portion is under the active deltaic plain. The mature part of the delta is going to be hydrological moribund because of the less active tidal action of the rivers and the rivers become choked with shoals and islands indeed. The degeneration of many of the channels has been progressed by natural maturity and human interferences (Bagchi and Mukherjee, 1978).

VARIATIONS OF THE HEIGHT
More than half of the study area is lying below 3-m contour line. The general slope of the study area is uniformly different terms of ANOVA are, standard deviation (s), mean of each group (A -), mean of the whole sample (X), sum square between group (SSB) and sum square within the group (SSE/SSW), degree of freedom (df), mean square (MS), total sum square (SST), a number of groups (K) and number of samples (n).

LOCATION
The study area, geographically includes the Indian portion of the middle Ichamati river (Kalanchi to Basirhat bridge) and covers an area of about 1884.96 sq m. Geographically the area is situated between 22 o 10/N to 23 o 11/N of Latitudes and 88 o 37/E to 89 o E of Longitude ( Fig.1 and 2)

GEOLOGY
The study area belongs to the district of North 24 Parganas of West Bengal, situated in the southern part of Bengal Basin. The Bengal Basin is a percartonic basin (Basu and Sil 2000). The Bengal Basin was formed by a chain of fluvio-tectonic actions (Sengupta 1996). The subsurface geology of the study area is completely blanketed by the quaternary sediments comprising a succession of silty-clay and sand of various grades, and sand mixed with occasional gravels and thin intercalations of silty-clay (Sikdar and Sahu 2009).

CLIMATE
The study area is under tropical climate. The area receives maximum rainfall during the monsoon period (average Meander scrolls or meander scars (Thornbury 1998) are the signature of the channel shifting. Another way, it denotes the threshold condition, river exceeded the old flood plain and meander scrolls occur (Nansen 1986; Magilligan 1992). The 3D DEM picture does not indicate any significant variation of the landform (Fig.3(a), (b)).
A meander scar consists of two elements: (a) plain floor: variable width, and (b) backside steep wall. In the study area, the meander scars appear as a pseudo-terrace like landform. The older meander scar is situated on the higher elevation and demarcated by the older tree line. In Bhekutia village (J.L. No.24, Swarupnagar P.S.) there are two successive meander scars on both sides of the Ichamati river (Plate 1 and Fig.4). The river has shifted its position in different times, keeping its footprint as a meander scar, at different elevations (Fig.5). In Kabilpur, there are three successive meander scars on the  It is a comparatively low land and highly modified by human activities and followed by intensive paddy cultivation (Fig.3). Different cropping pattern, practiced on the meander scrolls indirectly exhibit several hydrodynamic regimes. The upper meander scars are composed of comparatively coarse sand (medium permeability) and predominated by Rabi crop. The lower scrolls are composed of fine sand and predominated by paddy cultivation throughout the year (Fig.4). The river regimes shift from coarse sand to finer one. These indicate that the river lost its carrying capacity with time.

LANDFORMS IN THE CHANNEL (ON CHANNEL LANDFORMS) 4.2.2.1 Point bars
Convex bank of the river is characterized by point bar, mainly in the lower reach of the river. The area of the point bar has been decreased from the upstream downwards. The linear function between the area of the point bar and the distance along the river is y = 5411x-13689 (R 2 = 0.264) (Fig.6). But the correlation of coefficient is high (r = 0.514). During high tide these are submerged, but come into being with ebb period. Towards the upstream, the point bars are not well demarcated, but in the lower reach of the river the point bars are prominently developed. There are basic differences between the point bars of the upstream and the downstream reaches. First: the area of the point bars in the upstream is much lesser than those of the downstream. Second: the point bars in the upstream are composed of fine sand (photograph 1) but the point bars in the lower portion are composed of coarse sand (photograph 2).

Ripples
The point bars of the lower reach of the river are covered with ripples. The length of the ripple varies from 8cm to 130cm. Ripples are mostly asymmetric in form (Fig.7). The downstream slope of the ripples are stepper than the upstream one (photograph 3 to 6). The ripples are mostly modified by two flow dynamics, e.g. (a) flow dynamic during high tide, and (b) flow dynamic during ebb period. As the ripples submerge during tides, its modification is beyond the observation.

POINT BAR: CLASSIFICATION AND JUSTIFICATION
There is an ambiguity, whether the point bar would be classified as in-channel landforms or surface landform? Because high tidal level (HTL) divides the point bar into two distinct parts, one: a part under the water and another part is on the water. To analyse the classification of point bar, I have used two form elements such as width (y) and height (z) based on their area (x). So, the applicable equation, here, appeared as, d 2 z/dx 2 = 0, d 2 z/dy 2 = 0. For the second order derivation between height and distance, I have applied the polynomial function for more accuracy (error = 0.0001%) rather than linear or other functions. Here the function {f(x)} is 0.  Table 1). The ANOVA test of the three groups of point bars (area: the width of the channel) shows a better result. The value of the SSB and SSE are 795 and 859622 respectively. The F-statistics factor (0.008) ( Table 2) marks no significance in meander classification.

MEANDER SCROLL: CLASSIFICATION AND JUSTIFICATION
The meander scroll has an immense impact on local landuse. The Ichamati is characterized by unpaired meander scroll in its middle course. The different meander scrolls are classified according to different height of their locations. I have also considered their width parameter. The value of the second order derivative of the height factor (z) to the length factor (x) is greater than zero (For first meander scroll: d 2 z/dx 2 >0.012), and d 2 z/dy 2 = 0). For second meander scroll, d 3 z/dx 3 >0.022 and d 3 z/dy 3 = 0. The ANOVA test also justified strongly this classification. The test was applied to the meander scrolls (n = 13) on the left side bank. The value of SSB is 5.28 and SSE is 7.35 (Table 2), successively. The value of the F-statistics is 8>>>0.05 (Table 2), which strongly rejected the null hypothesis (meander scrolls are uniform in character). The relation between height and width of the meander scroll is moderately negative (r = -0.35).

RIPPLE: CLASSIFICATION AND JUSTIFICATION
I have taken into account 117 ripples in three stations, namely Basirhat, Harispur-Bibipur and Baduria region (Fig.8).   for such micro features. For the justification of the above three classes, the value of SSB is 21392.9 and SSE is 55866.2 ( Table 3). The value of the F-statistics is 21.82>>>0.05 (Table  2), which is significantly high to cancel the null hypothesis. The length of the ripple has been decreased downstream upward.
There is also a remarkable local level variation of the ripple-size. The length-width ratio (r = -0.523) shows that the larger size ripple are formed at the margin of the point bar, whereas the ripple on the central point bar comparatively smaller (Fig.9). Besides the above parameters, the slope ratio between two sides slopes (upstream (U s ) and downstream (D s )) of the individual ripple are not satisfactory (r = -0.17), also.  Beyond Baduria, the flow is not competent to form such micro level landforms on point bar. The velocity and specific energy of the river is gradually decreasing downstream upward (Mondal, et al. 2019). At every station I have taken 39 ripples for ANOVA test. Therefore, K will appear as 3, here. The formula d 2 z/dx 2 >0, d 2 z/dy 2 = 0 etc. are not applicable

Conclusions
An explanation for the classification of the landforms in the highly modified by human activities in the low relief deltaic region is proposed. The cross sections of the area, produced by DEM (based on IRS P6 LISS II and LISS III) do not identify any significant variations of the landforms. After the field investigations, the landforms of the area have been classified and ANOVA test has justified the classification. The oxbow lakes, natural levee are highly modified and converted to agricultural land and thus lost their geomorphic significance. Flood plain genetically converted from point bar and the point bars are attached to the flood plain on the convex bank and gradually extend toward the concave bank. So, there is no well defined demarcation line between flood plain, river bank, point bar and channel bed. Therefore, the low value of F-statistics (0.00017) justified such lack of clarification. On the other hand, the high value of the F-statistics (F=8, F =21.86) proves the right classification of meander scrolls and ripple. These classifications have brought the geomorphic truth before us as, (a) The river is enough capable to produce the prominent landforms such as point bars (area and distance relation: r = 0.514), ripple (number and length decrease downstream upward) etc., (b) the agricultural practices on the meander scrolls are different due to sediment texture which indicates the change of river regime. The lower meanders are characterized by paddy cultivation throughout the year.' These landforms are comparatively young and composed of fine sediment, that for they have high water contain capacity. The series of coincidence indicates the river lost its carrying capacity with time. The present paper claims some detail data sources. Sediment character has been analyzed by feel method. An intensive sediment discharge record, bed load data may increase the quality of the paper. This lack may fill up subsequently to enhance the research quality.