The non-unique relation of concentration–discharge \((C\) - \(Q)\) has been studied in past but still not completely understood how it evolves with time. Current study discusses the relationship between the power law coefficient \((f)\) and past average sediment concentration \({(C}_{PN})\) during recession where \(N\) represents number of past days. The advantage of the proposed model is that it can predict future suspended sediment concentration at a river cross-section during recession. We applied the proposed model to 80 US Geological Survey basins and found the 75th, 50th and 25th values of NSE to be (0.39, 0.80), (0.2, 0.74), and (− 0.04, 0.61) for rating curve and proposed model, respectively. Main objective here is to provide a clear understanding of sediment concentration relation with dominant basin parameters during recession. Principal component analysis along with multiple linear regression has been performed between NSE and principal component analysis resulting with R2 as 0.31 and 0.52 during calibration and validation, respectively.
