Abstract:
The present study deals with the riverbank stability analysis in order to find out the reasons for failures of the selected reach of Jamuna river of Bangladesh. Various scenarios considering the hydraulic and static load condition have been applied in this study. These load conditions are hydraulic shear stress, surcharge load, pore water pressure, toe scouring, river water fluctuations and slope protection load. Limit equilibrium method (LEM) and finite element method (FEM) of slope stability analysis have been applied for stability analysis for various load condition. Two types of bank materials range from very fine clay soil of median size 0.028mm to coarse sand of size 0.167 mm, have been used in this study. The hydraulic load was determined by applying SMS & SRH-2D depth-averaged numerical model using a highly refined mesh. Erosion rate was evaluated from the excess shear stress model, based on critical shear stress, τ_c and erodibility co-efficient, k_d. The stability of the riverbank was analyzed with the limit equilibrium method in the SLOPE/W module and the finite element method in Optum G2. The limit equilibrium method (LEM) used in this study was the Morgenstern-Price technique, while the strength reduction technique was used in FEM. In both cases, Mohr-Coulomb failure criteria were chosen for analysis. A total of seven scenarios of various load conditions have been used to assess the slope stability of the riverbank. Among these, the water level drawdown of water level, bed scouring, pore water pressure etc. and various static loads are incorporated in the model run. Also, the base material of the riverbank consists of clay with mica (30%), silt and sand layers were taken into consideration for a separate scenario. A sensitivity analysis was performed in SLOPE/W to obtain a range of design data that could be used in the design of riverbank protection works. Results from the hydrodynamic model analysis reveal that hydraulic shear stress at the toe of the riverbank was found as 10 to12 Pa compared to that of critical shear stress 6.34 to 8.91 Pa, which clearly indicates the vulnerability of bank stability under the hydraulic action. Therefore, soil erosion at the toe has to be considered as the most critical fluvial riverbank erosion. Erosion rates have been determined from the excess shear stress obtained from the model analysis, which ranges between 20 mm/hr to 160 mm/hr depending on the lower and upper soil layer of the bank. Erodibility co-efficient obtained from the present analysis found to range between 3.1 to 3.64 cm3/N-s of Jamuna riverbank material, which indicates a very erodible soil particle. It is also found that the stability of the riverbank fails when scour depth attained more than seven meters under the given bank slope. The effect of surcharge load was found to be very significant for the slope failure mechanism. It is found that more than 60 KPa surcharge loads made the riverbank unstable without any scour. With an existing scour depth of five meters, only 45 KPa surcharge loads were needed to make the riverbank fail. The layer with clay and mica (30%) showed more resistant to erosion than the layer with sandy silt and silty sand as obtained from the present analysis. Furthermore, a sensitivity analysis was carried out using slope/w to obtain design soil parameters for stable riverbank against surcharge and slope protection loads. Results obtained from LEM and FEM are compared and found the higher safety factor with LEM than FEM. Finally, a field study was carried out with similar data collected from the Jamuna river near Chauhali, Sirajganj. Results from the field study also compared with the results as obtained from previous scenario analysis and found satisfactory as far as slope stability concerned. It is hoped that the study will be helpful for understanding the design methodology of riverbank training work and bank protection works of rivers of Bangladesh.