Experimental Study to Estimate Hyporheic Velocity Using Wavelet-Hybrid Soft-Computing Model

The hyporheic zone represents the interface between surface and subsurface flow. An experimental investigation was conducted in a laboratory flume featuring a maximum flow rate of 50 L per second, utilizing artificial grass completely submerged in water. Two sets of experiments were carried out, one with vegetation cover and the other without. The findings revealed that vegetation cover led to a reduction in hyporheic velocity, whereas the absence of vegetation increased hyporheic velocity. The study also noted that the absence of a hyporheic zone in vegetation, compared to gravel, could be attributed to the formation of a separation zone. Additionally, it was observed that vegetation cover facilitated the supply of more nutrients around the divide line, owing to upwelling flows from both upstream and downstream directions. Given the limited dataset, Soft Computing (SC) techniques, namely Wavelet-GEP (WGEP) and Gene Expression Programming (GEP), were employed to formulate mathematical equations for estimating hyporheic velocity under steady-state conditions for a given discharge. These equations, incorporating hydraulic and geomorphic variables, proved effective in estimating hyporheic velocity under stable bed conditions during steady flow. The models were trained and tested using 70% and 30% of the collected data, respectively. The performance of the models was assessed using statistical criteria. The results indicated a strong correlation between noise reduction in data and improved performance of the WGEP model compared to the GEP model in estimating velocity.