Erosion and sedimentation is one of the most important problems in controlling Iran’s watersheds. In order to implement conservation programs, soil erosion control and reducing sedimentation, it is necessary to estimate the total volume of sediment in a watershed to help the watershed managers in selecting the appropriate methods of erosion control and conserving natural resources. The aim of this study was to investigate the changes in suspended sediment in 12 stations of the Karkhe river basin and presenting the model of region using physiographic variables in the form of multiple regression equations so that in this way, the estimation of the amounts of suspended sediment in non-gauged basins be possible. In this study, 12 deposition and hydrological stations having sufficient statistics and scattered well throughout the basin have been selected and their amount of sediment have been calculated and completed using sediment rating curve and integrating the monthly, quarterly and annual discharge. Besides, in the upstream basins of the selected stations, 16 physiographic variables have been collected as independent variables in sediment production and multivariate regression analysis was performed in the presence of all the variables. In the next step, the most suitable statistical relation between suspended sediment and the basin features along with the error rate of each model have been obtained in 33 temporal models based of NS, R 2 , RMSE, MAE and AME coefficients. The results of this study showed that the suspended sediment prediction models acquired from regression methods give better results with fewer errors based on all variables. Besides, the models which their observation sediment have been completed based on annual sediment show a higher coefficient of determination. According to the evaluation criteria of the regression equations, the suggested models have an acceptable capability for estimating the suspended sediment and regarding the fact that the method used in this research is based on recorded data and statistical methods approved by different scientific sources, the results of this study can be used in basins with no recorded data like Karkhe basin which is very helpful for organizations and executive-research sectors. Key words: Karkhe river basin, sediment rating curve, multivariable regression, annual average sediment