since similar shoaling tests run with different water temperatures often 

 give significantly different results. In general, finely ground gilson- 

 ite is used to simulate suspended sediments, and granulated plastic, 

 nylon, or other similar material is used to simulate bedload sediments. 

 Gilsonite is an asphaltic base material with a specific gravity of about 

 1.03; it is usually graded to pass a Tyler No. 24 screen (0.8 millimeter), 

 and is retained on a Tyler No. 35 screen (0.4 millimeter). No attempt is 

 made to model the characteristics (e.g., fall velocity) of a particular 

 suspended sediment. Commonly used granulated materials (with specific 

 gravities) include polystyrene (1.03 to 1.09), nylon (1.13 to 1.15), 

 Tenite Butyrate (1.18 to 1.20), coal (1.4), and naturalite (1.7). These 

 materials are available in various regular shapes such as cubes and cyl- 

 inders or in irregular crushed shapes. The selected material is usually 

 injected into the model in a slurry as either a point or line source, 

 although it is occasionally spread over the entire problem area before 

 starting the model. After completion of the injection of the shoal mate- 

 rial, the model is normally operated for several tidal cycles to allow 

 enough time for the material to be dispersed by the currents and deposi- 

 ted. Waves should only be generated when the shoaling problem area is 

 in the estuary entrance area and thus subject to ocean wave action. Be- 

 cause of the distorted model scales, waves in the model that represent 

 prototype conditions cannot be reproduced; rather, the model waves are 

 adjusted to simulate the degree of agitation required so the model sedi- 

 ments can be moved and deposited by the tidal currents. 



Because the shoaling test technique is developed by trial and error, 

 the validity of the shoaling verification is highly dependent on the 

 quality and quantity of the available prototype data. Surveys of the 

 problem area should be available for a period of at least 2 and prefer- 

 ably 3 or more years, in order that average annual conditions can be de- 

 termined. The problem area is subdivided into sections, and the average 

 annual prototype shoaling rate is determined for each section. These 

 rates are then converted to percentages of the shoaling rate for the 

 entire problem area, and this is the percentage distribution that is 

 reproduced in the model. When an acceptable reproduction of the distri- 

 bution pattern has been achieved, the volume of material recovered from 

 the problem area in the model can be equated to the prototype shoaling 

 rate to establish an approximate shoaling volume scale. The duration of 

 the model test can also be equated to the prototype period for which the 

 shoaling rate was developed to determine the shoaling test time scale. 

 Examples of shoaling verifications are presented in Figures 3-37 to 3-39. 



Since all conditions cannot be exactly duplicated between model and 

 prototype, an exact duplication of the shoaling distribution pattern can- 

 not be expected; e.g., the effects of overdepth or advance maintenance 

 dredging may be difficult to simulate in the model unless the dredging 

 practice is consistent from year to year. Because the models are fixed 

 bed, the effects of local scour or nearby deposition, and the changes in 

 cross section resulting from scour or unusual dredging cannot be simu- 

 lated. At the termination of a model shoaling test, all material in 

 motion deposits immediately in place, resulting in some model shoaling 



13 



