Mainland Shelf 



205 



LaFond, 1940). An extension of this work 

 was made by eight separate surveys between 

 April 1949 and May 1950 aboard amphibi- 

 ous vehicles in an area off 0.6 km of beach 

 between Scripps and La Jolla Submarine 

 Canyons and extending seaward 1 km to 

 depths of about 30 meters (Shepard and 

 Inman, 195 la, 1953). Later, checks were 

 made of the surveys by repeated observation 

 of the sand level against stakes driven into 

 the bottom by divers (Inman and Rusnak, 

 1956). Comparison of the surveys showed 

 definite cuts and fills to the outermost limits 

 of the study, but with the greatest changes 

 in depths shallower than 10 meters. Changes 

 between the surveys in water depths greater 

 than 3 meters averaged 8 cm and locally 

 were ten times this figure. Estimates of the 

 volumes and areas of change indicate that 

 during times of large waves sand is lost from 

 the shelf, presumably by southward lateral 

 transport and accumulation in the head of 

 La Jolla Canyon. During times of smaU 

 waves the sand is believed to be replenished 

 by southward lateral transport near the 

 beach and around the head of Scripps 

 Canyon. Support for the inferred move- 

 ment of beach sand around the head of 

 Scripps Canyon and into the intercanyon 

 slielf area is provided by the similarity of 

 grain size and mineralogy of the sediments 

 on the shelf and the beaches north of the 

 canyon (Inman, 1953). Sediments on the 

 outer half of the shelf north of Scripps Can- 

 yon (Wimberly, 1955) contain a much larger 

 percentage of silt and clay than those south 

 of the canyon, indicating that southward 

 transportation results in their being trapped 

 by the canyon barrier. 



In the absence of a submarine canyon 

 trap, sediment might be expected to move 

 downcurrent in a broad zone spanning the 

 inner part of the shelf. The fastest move- 

 ment and the coarsest sediment being moved 

 should be at the shore because of greater 

 turbulence there. Farther offshore the 

 movement should generally be less rapid 

 and involve progressively finer maximum 

 sizes until at some point the greater depth 

 and slight turbulence reduce the movement 

 to neghgible proportions. Longshore move- 



ment past rocky coastal barriers such as 

 Point Conception and Point Dume has been 

 shown to exist by Trask (1952, 1955) on the 

 basis of mineralogy, grain size, continuity 

 of subsea slopes below 10 meters in front of 

 these rocky points, and the presence of 

 ripple marks on sand bottom to depths 

 greater than 20 meters (Inman, 1957). If 

 any rocky point extends to depths of 30 

 meters, there would be considerable doubt 

 that sediment could move past it until the 

 sea floor became built up enough by sedi- 

 ment to cover the deep barrier and thus 

 permit subsequent passing of sediment. 

 Observations by divers (Trask, 1955; others) 

 show that in water deeper than about 30 

 meters the bottom is covered by a thin layer 

 of fine brown sediment that could not be 

 present if currents were appreciable. Simi- 

 lar material has been observed temporarily 

 present after rainstorms, as though it were 

 in the process of very slow seaward trans- 

 port, bypassing sediments that form the 

 more permanent sea floor mantle. In sup- 

 port of this hypothesis of movement is the 

 fact that multilevel sediment traps placed on 

 the shelf generally trap sediments that are 

 finer than those constituting the bottom 

 surface. 



In addition to periodic changes in the 

 nature of sediments on the shelves there are 

 secular ones. Ancient changes are recorded 

 by variations in grain size along the length 

 of cores of shelf sediments. A more recent 

 change is shown by a comparison of grain 

 sizes of 200 samples collected from Santa 

 Monica Bay by Shepard and Macdonald 

 (1938) and 580 samples collected from the 

 same area during a study of sewage pollu- 

 tion 20 years later (Terry, Keesling, and 

 Uchupi, 1956). Almost the entire area of 

 comparison is deeper than 20 meters. Con- 

 tours of median diameter show that a defi- 

 nite decrease in grain size occurred between 

 the dates of the two surveys that is greater 

 than might be expected from differences in 

 field collection and laboratory analysis tech- 

 niques. This comparison of contours is 

 made on the basis of whole suites of sam- 

 ples, few of which are at identical positions 

 in the two surveys. For nine pairs that are 



