coring was complete and the apparatus was lifted back onto the scow. The core 

 liner containing the sediment was removed from the barrel and small representa- 

 tive samples were obtained from the top and the bottom of each core. The liner 

 was then capped and sealed, labeled, and a general core description was made. 

 The scow was then moved to the next coring location. While underway, the coring 

 device was reassembled and loaded with a new liner. 



5. Processing of Data . 



After completion of both phases of data collection, all the navigational fix 

 marks, ship trackline positions, core sites, and shore stations were plotted to 

 show the coverage within the survey area (Figs. 2 to 8) . The seismic records 

 were visually examined and marked to establish the primary geologic features such 

 as regional sedimentary reflectors, erosional unconformities, sediment contacts, 

 and buried stream channels. Selected acoustic reflectors were then mapped to 

 provide areal continuity of horizons considered significant because of their 

 areal extent and relationship to the general structure and geology of the study 

 area. Where possible, the topmost reflectors were correlated with cored sedi.- 

 ment to provide a measure of continuity between cores . 



The cores were visually inspected and described in general terms onboard 

 the scow; a more detailed study of the cores was made later. All cores were 

 split longitudinally to show changes in sediment composition, texture, and phys- 

 ical character. Selected intervals of cores were photographed to provide an ar- 

 chive record of the sediment character. The sediments were identified, logged, 

 and described according to textural properties (using the Wentworth Scale in 

 Table 1), gross lithology, color, strength, thickness, fossils, and depth from 

 the lake bottom (top of the core) (see App. A). Representative sediment samples 

 from each core were examined with a plane, light binocular microscope. A total 

 of 291 grain-size analyses were made. Granulometric parameters (e.g., mean 

 grain size, sorting, cvimulative size distribution) were evaluated for- 141 of 

 the samples by using the CERC Rapid Sediment Analyzer (RSA) as described in 

 the Shore Protection Manual (SPM) (U.S. Army, Corps of Engineers, Coastal Engi- 

 neering Research Center, 1977). These RSA data, as well as sieve data from 11 

 samples too coarse to process by RSA, are presented in Appendix B. Cumulative 

 distribution curves are also presented. 



All of the sand sample sizes are described in both millimeters and phi (4>) 

 units where <j) = -log2D. D is the grain-sized diameter in millimeters (see Table 

 1) . In the RSA analysis the sand sample falls through a tube of water and a 

 pressure transducer is used to determine the fall velocity of the sand grains. 

 A computer program is then used to compute moments for converting fall velocity 

 to hydraulic grain-size diameter. The RSA method is fast and reliable, but it 

 is limited to analyzing very fine to medium sands. Any fine-grained material 

 present with the sand often remains in suspension in the tube when the measure- 

 ments are stopped. Thus, the silt and clay fraction in a muddy sand sample is 

 often omitted from the size analysis results, making the sample appear better 

 sorted than it actually is. Most researchers agree that RSA values are consis- 

 tent and slightly coarser than sieve values for identical samples. Ramsey and 

 Calvin (1977) suggest adding 0.33 phi to the RSA mean to obtain the equivalent 

 sieve mean; another formula, with a similar constant, is shown in Appendix B. 



Eleven samples which were estimated to have more than 10 percent gravel 

 were sieved at a 0.5-phi interval. Five sand samples from cores 101, 102, and 



20 



