In this paper, the reason of horizontal angle gross error occurring in a connecting traverse was discussed firstly. The relationships of coordinates between forward and backward direction calculation were derived and analyzed. Furthermore the utility of the method is investigated using the results of an example. Finally some suggestions to avoid gross error were proposed.

Theoretical backing of the similarity laws obtained from the dual dimensionless parameters of the bed-form classification diagram is given by solving the equations of similarity conditions that are obtained from the 1-dimensional equation of motion, the continuity equation, the bed deformation equation, and the formula of bed-load transport. Furthermore, a distorted model experiment was conducted to predict the bed variation during a flood in the Chubetsu River (a tributary of the Ishikari fiver in Hokkaido, Japan). The usefulness of the similarity laws of a distorted model was ascertained, and it was found that the maximum scour depth occurs during a recession of the flood hydrograph.

By using the . . Method, Return Period Method, Bankfull Discharge Method and the Effective Sediment Transport Discharge Method to the hydrological data of Ba River, the bed-building discharge has been calculated and made as flow conditions of river model experiment. Finally, it was applied to the model experiment about Ba River and the results was compared with the one of balance riverbed from the normal river model experiment made by Northwest Water Conservancy Science Research Institute in 2001. The result shows that 6 years working life of the reservoir of Ba-river No.B rubber dam being regime channel is agreed with that of the prototype. It indicates furthermore that the bed-building discharge used in this river model experiment is correct, and the . . Method, Return Period Method, the LuoMinxun formula of Bankfull Discharge Method has a better applicability to calculate bedbuilding discharge for Ba River.