Forecasting the sediment load in a river is difficult due to different parameters viz., heavy rainfall and precipitation, tropical climate, transportation of sediment, and so on. The wavelet transformations model helps to analyze the time and frequency information to estimate sediment load by decomposing data over several phases. Inspired from this idea, based on extreme learning machine (ELM) and twin support vector regression (TSVR), this work proposes two coiflet wavelet-based models as, coiflet wavelet-based ELM and coiflet wavelet-based TSVR for sediment load estimation. The results are compared with conventional ELM and TSVR. The performances of the algorithms are examined using five performance evaluation techniques i.e. root mean square error, mean absolute error, ratio between sum of squares error and total sum of squares, symmetric mean absolute percentage error and mean absolute scaled error. The experimental outcomes reveal that the hybrid models based on the coiflet wavelet offer good performance.