Keywords

13.1 Introduction

Soil or land degradation under rills and gullies is developed by concentrated flow erosion, resulting in the detachment and displacement of topsoil particles (Govers 1991; Govers et al. 2007). The soil loss includes (1) loosening and detachment of soil particles from the soil mass through the process of rilling and gullying, (2) removal and transport of eroded soils downslope and downstream by overland flow, and (3) soil slumping under the impact of the increased volume of detached soils (Singh and Dubey 2002). Gully erosion is one of the forms of accelerated soil erosion. Its occurrence often indicates an extreme form of land degradation warranting special attention (Singh and Phadke 2004).

In India, an area of approximately 4 million ha is affected by rill and gully erosion (Sharda et al. 2007). The Ministry of Agriculture, Government of India (2000) reported the 12 Indian states that were most seriously affected by rill, gully, and ravine erosion. Gully erosion in India has mainly focused on the morphological characteristics (Bandyopadhyay 1987, 1988; Das and Bandyopadhyay 1996; Sen et al. 2004; Ghosh and Guchhait 2015; Shit et al. 2012), hydrological processes (Shit and Maiti 2013), and controlling measures of hillslope gully and on the development process (Singh and Phadke 2004), features, controlling factors, and field monitoring methods of ephemeral gully channels (Singh and Dubey 2002; Shit et al. 2011, 2012), and the delineation and monitoring of gully erosion lands (Pani and Mohapatra 2001; Ghosh 2015; Shit et al. 2015).

Moreover, relatively few studies have examined differences in the causes and influences of gully erosion in a subhumid lateritic belt in West Bengal, India (Shit et al. 2013). In this region, geomorphological features, controlling factors, and monitoring methods for ephemeral gullies have been studied (Sen et al. 2004; Shit et al. 2014; Ghosh 2015). Additionally, Shit et al. (2011) examined headward erosion of gully heads under different discharge condition, and Shit et al. (2013) studied the distribution and influencing factor of vegetation (above biomass and ground biomass) on gully erosion in this region.

Several studies have been carried out on rill and gully erosions, but only a few studies have concerned utilizing natural vegetation to control gully erosion as well as headward erosion. Vegetation has an obvious effect in reducing concentrated flow and controlling soil loss, affecting sediment yield of the overland flow in gully basins (Rey 2003; Spaan et al. 2005; Li et al. 2006; Su et al. 2014; Shit et al. 2012, 2013). However, most of the previous studies of vegetation in soil erosion were mainly conducted on hill slopes with few examining the effects on vegetation in gully erosion. Numerous researchers reported that the vegetation cover has a significant influence on surface runoff volume and sediment (Braud et al. 2001). Spaan et al. (2005) and Rey (2003) reported that vegetation could effectively slow the velocity of the flow, trap sediment, build up backwater, and promote sedimentation to reduce gully erosion. However, few studies focus on rill-gully erosion mechanisms and particularly the impact of vegetation cover on hydraulic properties of surface runoff in gully channel and beds, because of the complex development processes and the many influential factors compared with sheet erosion.

In this chapter we investigated the role of a vegetation belt on gully erosion. For this aim, six ephemeral gully channels were selected to (1) examine the dynamic changes of gully hydraulic properties and sediment yield in response to the grass cover in the gully beds and (2) analyze the relationship between runoff properties and sediment content.

13.2 Materials and Methods

13.2.1 Study Area

All experiments were conducted at the Rangamati badland area (22°23′ to 22°24′ N latitude and 87°17′ to 87°18′ E longitude), which is a typical dissection of the landscape by dense and deep rills and gully erosion in the western part of West Bengal (Fig. 13.1). The average slope of this area is between 25% and 35%. The most frequent landforms are complex slopes and gullies. The average annual precipitation and temperature are 1500 mm and 23.5 °C, respectively. The dominant soils at the surface of the study area are dry red soil and vertisols (USDA soil taxonomy). Dry red soils often have high sand content by iron and silica; vertisols often have high clay content with a mean bulk density of 1.34 to 1.86 cm−3. According to a previous study (Shit and Maiti 2014), the region is dominantly herb species with Eragrostis cynosuroides, Lantana camara, and Andropogon aciculatus vegetation species in the gully bed (Shit and Maiti 2014). Eragrostis cynosuroides is a local species that can grow rapidly on gully beds in a subhumid tropical environment (Shit et al. 2013).

Fig. 13.1
figure 1

Location of lateritic belt area (i.e., rill-gully-prone area) and experimental site (Shit and Maiti 2014)

Full size image

13.2.2 Field Monitoring and Measurement

Six gully beds were selected for simulation runoff experiments. The gully beds were 18 m long with width varying from 0.25 to 0.50 m, with an average slope of 18–20° on the gully bed. The six gully beds have similar topographical conditions, and headcut height was 0.3 m. The grain size distribution of soil is mainly gravel (30%), sand (40%), and silt and clay (30%). Dry soil bulk density ranges from 1.34 to 1.78 g/cm−3.

Eragrostis cynosuroides was growing naturally in four gully beds with similar density and an average height of vegetation of 20–50 cm during the experiments, and the average vegetation cover of buffer strips was 67%. However, vegetation buffer strips were selected at the bottom end of the gully beds extended up to headward direction. The only bare gully bed was selected for control measurement and also compared with the buffer gully beds. Six strip widths were established on gully beds: bare (0), 3, 6, 9, 12, 15, and 18 m, which were named Gully-A, Gully-B, Gully-C, Gully-D, and Gully-E, respectively (Fig. 13.2).

Fig. 13.2
figure 2

Sketch of experimental simulation concentrated flow and vegetation distribution of gully beds. The length of each section is 3 m, gully width varies from 0.2 to 0.25 m, and average slope of the gully bed is 18–20°

Full size image

The experiment on each gully bed consisted of five simulation runoff tests in the dry season (November 2018), during which no natural rainfall occurred. Clean water pumped through a 2-HP diesel pump from the nearby pond and a nylon water pipe was used to connect the adjustable opening to a flowmeter to measure the flow discharge during the experiment (Su et al. 2014). The flow scouring tests on each gully bed at lasted 60 min with flow discharge of 78.5 l min−1. The discharge was adjustable according to average precipitation in the region. During these experiments, the hydraulic parameters and sediment yield were measured. The hydraulic parameters included the flow velocity (v in ms−1), flow width (w, in meters, m), flow depth (d, in meters, m), and temperature (in °C). The flow velocity was measured using the color tracer method (potassium permanganate solution); flow width and depth were measured with a ruler and measuring tapes (Shit and Maiti 2014). Water temperature was measured with a centigrade thermometer. All parameters were measured three times for each gully bed section at 10-min intervals during the experiments (Fig. 13.3). Additionally, hydraulics parameters, including Reynolds number (Re), Froude number (Fr) , hydraulic share stress (τ), Manning roughness coefficient (N), and Darcy-Weisbach friction factor (f) were calculated according to Eqs. (13.1)–(13.4):

Fig. 13.3
figure 3

Layout of experimental plots: (a) simulation of water supplied by 2-hp diesel pump; (b) Gully-A; (c) Gully-C; (d) Gully-B; (e) Gully-E, (f) Gully-D

Full size image
$$ \operatorname{Re}=\frac{vr}{\mu } $$
(13.1)
$$ \mathrm{Fr}=\frac{v}{\sqrt{gh}} $$
(13.2)
$$ \tau =\rho grs $$
(13.3)
$$ f=\frac{8 grs}{v^2} $$
(13.4)

Where v is the flow velocity (m s−1), r is the hydraulic radius (m), s is the surface slope angle (m m−1), g is the gravitational acceleration (m s−2), h is the flow depth (m), ρ is the water density (kg m−3), and μ is the water kinematic viscosity coefficient (m2 s−1).

Runoff samples were collected at the bottom part of each gully bed using a 0.5-l plastic bottle during the simulation runoff at 5-min intervals. These samples were deposited for 24 h and the clear water decanted; then, the sediment was oven dried at 105 °C for 12 h and weighed to determine the sediment loss (Kodamatani et al. 2017). Gully bed topography was monitored by paired photographs with field measurements based on terrain height (points) per square meter. Gully bed topography changes, including width, depth, volume, and elevation, were analyzed by subtracting the volumes of the digital elevation model (DEM) at 10 m spatial resolution before and after scouring using Raster math tools and spatial analyst tools of ArcGIS software. All statistical analysis was carried out using MS Excel and Origin 6.1 software.

13.3 Results and Discussion

13.3.1 Vegetation Effects on Concentrated Flow Hydraulics in Gully Beds

Summaries of the statistics of all experimental results are presented in Table 13.1. Hydraulics parameters (i.e., v, Re, Fr, F, t, and N) for vegetation buffer strips were clearly shown to be smaller than those for bare gully beds. The average parameter values for the vegetated buffer strips varied between 35% and 50%, which is lower than those for bare gully beds with a minimum value (74.53%) of τ (Table 13.1). Concentrated flow resistance values were presented by Darcy-Weisbach friction factor (f) and Manning roughness coefficient (N). The results showed that these values for vegetated buffer sites were 1.2 and 1.4 times higher than those for bare gully sites, respectively (Table 13.1).

Table 13.1 Summary of the statistics of hydraulic parameters on bare sites and gully beds with vegetation buffer strips
Full size table

13.3.2 Relationship of Different Hydraulic Factors Between Vegetated and Bare Sites

Figure 13.4 represents a relationship between Reynolds number (Re) and several hydraulic factors in vegetated and bare gully beds. The increasing regression lines of Re-v were illustrated for both bare soil and vegetated gully beds, but the Re-v trends line for bare soil has a steeper gradient than that of vegetated buffer gully beds (Fig. 13.4a). The equal value was also calculated for Re-Fr (Fig. 13.4b). The increasing trends lines with values of flow shear stress for bare sites of gully were significantly higher than that for vegetated buffer strips (Fig. 13.4c). Figure 13.4d shows a negative relationship between Re-f and indicated that bare sites have a steeper slope than vegetated sites. However, these results revealed that vegetated buffer strips usually have higher values of hydraulic flow resistance and lower values of flow hydraulics parameters (i.e., Re, Fr, t) than bare gully beds.

Fig. 13.4
figure 4

Relationship between vegetated sites and bare sites for different hydraulic factors: (a) Reynolds number (Re), (b) Froude number (Fr) , (c) flow shear stress (t), and Darcy-Weisbach friction factor (f) (d)

Full size image

13.3.3 Role of Vegetation in Soil Erosion of Gully Beds

Table 13.2 illustrates the erosion processes of gully beds during concentrated flow experiments. All gully sections studied were clearly eroded after six experiments: average gully erosion volume was about 0.33 m3 and 0.27 m3 for bare soil gully beds and vegetated gully beds, respectively. These results indicated that sedimentation was 1.24 times higher than that in vegetated buffer strips in respect to bare soil gully beds. The vegetated buffer strips in gully beds indicate that total gross deposit of sediment after experiments ranged between 0.02 and 0.1 m3 with an average of 0.037 m3, which was 1.6 times higher than that of bare soil gully beds. The net erosion volume ranged from 0.22 m3 to 0.55 m3 via an average of 0.31 m3 in bare soil gully beds, whereas vegetated buffer strip beds varied from 0.05 m3 to 0.15 m3 by an average of 0.23 m3 (Table 13.2). These results showed that vegetated buffer strips significantly reduce flow velocity, control gully incision, and trap sediment in gully beds.

Table 13.2 Landscape changes at different sites of gully beds during experiments
Full size table

13.3.4 Relationship Between Experimental Time (Minutes) and Soil Erosion

Figure 13.5 illustrates the experimental gully beds, both vegetated and bare soil, sediment concentration, and net erosion volumes against concentrated flow times. The net erosion volume shows a negative relationship with time for all the experimental gully beds. Higher erosion at the beginning time was observed in Gully-A. All gully beds were relatively stable during the experimental time, except Gully-A (Fig. 13.5a). The presence of the vegetation buffer did not significantly reduce soil erosion during the first 30 min. Similarly, a negative relationship could be observed on sediment concentration against time (Fig. 13.5b). The maximum total erosion volume was observed at Gully-A, whereas vegetation cover had zero percentage (i.e., bare gully beds), and minimum total erosion volume was observed at Gully-E, whereas vegetation cover was 66.67% (Table 13.3). The vegetation buffer strips showed that erosion rate is 3.1 times less than in bare soil gully beds (i.e., Gully-A). These results indicated that the percentage of vegetation buffer significantly controls erosion rate and sediment deposit in gully beds (Table 13.3).

Fig. 13.5
figure 5

Relationship between experimental time (min) and soil erosion: (a) net erosion volume; (b) sediment concentration (g l−1)

Full size image
Table 13.3 Net erosion volumes and erosion rate of each gully beds under flow discharge (78.5 l min−1) during 1 h
Full size table

13.3.5 Effects of Vegetation on Hydraulic Processes and Soil Erosion of Gully Beds

Figure 13.6 represents the relationship between hydraulic parameters and net erosion volume on gully beds. Runoff velocity (v), Reynolds number (Rn), and Froude number (Fr) showed negative trends with net erosion volume. A similar relationship between hydraulic parameters and soil erosion resulted from vegetation buffer strips , but the relationships were not statistically significant (Fig. 13.6). A similar poor relationship was found for friction factor (f) and erosion loss (Fig. 13.6d). Our result was in line with a previous study by Gimenez and Govers (2008), who showed that a poor correlation existed between flow hydraulic parameters and rill detachment rate in buffer strips. Figure 13.7a clearly represents that during the rainy season abrupt gully incision occurs without vegetation cover, but Fig. 13.7b shows that vegetation is growing along gully beds. These photographs indicate that increasing vegetation cover also reflects the increasing root system (i.e., length, density, mass) in gully beds. Moreover, the area covered with dense vegetation had reduced surface runoff , sediment traps by stems, and reinforcement of soil erosion and stabilization of the gully (Fig. 13.7c).

Fig. 13.6
figure 6

Relationship between hydraulic parameters and net erosion volumes: runoff velocity (v) (a), Reynolds number (Re) (b), Froude number (Fr) (c), and Darcy-Weisbach friction factor (f) (d)

Full size image
Fig. 13.7
figure 7

Influence of vegetation cover at different location of Rangamati badland area : (a) gully incision without vegetative cover , (b) gully stabilized by gully beds vegetation, and (c) reduced gully erosion by growing the vegetation along the gully channel and beds

Full size image

13.4 Conclusion

Our field experimental results showed that a vegetation buffer on the gully floor significantly influences hydraulics parameters, and it could increase resistance to concentrated flow. The consequences of sediment entrainment rates are reduced, and sediment is trapped by vegetation roots in gully beds. Results of the hydraulics parameters (velocity, Reynolds number , Froude number , and Darcy-Weisbach friction factor ) of bare soil gully beds showed the vegetation buffer strips in gully beds are 35–50% smaller and have 1.2 to 1.4 times higher concentrated flow resistance in Darcy-Weisbach friction factor (f) and Manning roughness coefficient (N), respectively. The total sediment deposit and net erosion volume on five gully beds produced a significant negative exponential, decreasing with increasing experimental time and vegetation cover. The vegetation cover shows that the erosion rate is 3.1 times less than in bare soil gully beds. Thus, this study showed that vegetated buffer strips significantly reduce flow velocity, control gully incision, and sediment loss in gully beds.