Bedforms--their measurement, analysis and application
General Material Designation
[Thesis]
First Statement of Responsibility
H. M. Mehrdad
Subsequent Statement of Responsibility
K. Mahmood
.PUBLICATION, DISTRIBUTION, ETC
Name of Publisher, Distributor, etc.
The George Washington University
Date of Publication, Distribution, etc.
1989
PHYSICAL DESCRIPTION
Specific Material Designation and Extent of Item
314
DISSERTATION (THESIS) NOTE
Dissertation or thesis details and type of degree
D.Sc.
Body granting the degree
The George Washington University
Text preceding or following the note
1989
SUMMARY OR ABSTRACT
Text of Note
Bedforms constitute the most distinguishing feature of mobile boundary channels. Yet, the measurements of bedforms are not common and methods for quantifying their dimensions have not been standardized. This study is based on bathymetric, hydraulic and sedimentation surveys carried out by the writer on four study reaches on Jamrao and Rohri canals in Pakistan under--ACOP, a GWU-WAPDA cooperative research project, sponsored by NSF and the Government of Pakistan. To add geographical diversity, the data from the Missouri River collected by U.S. Army Corps of Engineers and from three laboratory studies were also analyzed and the following conclusions are drawn: (1) For length and height dimensions, geometric methods like zero crossing and crest trough methods are superior to statistical methods and give realistic results. (2) For the determination of bedform width, cross spectral and crest trough methods can be used; however, the crest trough method gives more consistent results. (3) The height of bedforms in various study segments is normally distributed and the ripple-index is lognormally distributed. The bedform length, however, shows a Weibull distribution. (4) The trough length of bedforms is about one-third of the bedform length. (5) Among the methods available to predict form friction factor, Vanoni-Hwang consistently overestimates with an average error of 55%; Engelund consistently underestimates with an average error of 40%; Einstein-Barbarossa and Haque-Mahmood show average error of 12 and 8%, respectively. It is concluded that for sandbed canals where bedform dimensions are available, Haque-Mahmood method gives the best results of all available methods of predicting form friction factor. (6) Bedload in sandbed channels can be measured from the analysis of sequential profiles by down crossing and cross spectral methods. Both of these methods give very similar results. The down crossing method is computationally simpler and has other advantages. (7) The Meyer-Peter Muller formula and Einstein's bedload function both provide a close estimate to the bedload measured from sequential profiles. Bedload by extrapolation of suspended sediment concentration and velocity profiles to the channel bed is not feasible due to the large gradients in these quantities close to the channel bed.