size analysis of the mixture. Care must he taken in the laboratory to 

 obtain random splits of appropriate size from the samples. These splits 

 must be mixed completely to eliminate bias in obtaining the final sample 

 of the mixture to be analyzed. Sample splits of equal weight are generally 

 mixed if the samples are collected to represent equal parts of the beach 

 or borrow site. Splits of unequal size would be appropriate when a sam- 

 pling plan requires a weighting factor. The major benefit of mixing samples 

 is the reduction of laboratory time and expense for analysis. For example, 

 two independent mixings of the 8 samples over each of five profiles would 

 require the sieving of 10 samples instead of 40 and would allow evaluation 

 of laboratory errors associated with the mixing process. There are impor- 

 tant drawbacks to this approach as well. Perhaps, for engineering reasons 

 it might seem reasonable to omit all samples collected at the -30-foot 

 (-9.1 meters] depth or to ignore samples obtained in a particular part of 

 a borrow area, in these cases and many others, the initial sample mixtures 

 might be inappropriate for later requirements. Foresight and planning can 

 avoid such problems or, at worst, splits from the original samples can be 

 remixed and analyzed to suit future needs. The potential savings in sedi- 

 ment analysis that may result from mixing samples warrant the consideration 

 of this procedure in most situations. 



Appendix A provides an additional discussion of mixing sand samples for 

 composite analysis and an example of the different procedures for obtaining 

 the composite. The results of the experiment in the appendix show that 

 similar composite characteristics are obtained using either mixing or non- 

 mixing methods. 



IV. BEACH-FILL MODELS 



1. Basic Types . 



Two basic types of mathematical models have been proposed to handle 

 beach-fill problems. The first model enables calculation of a filt factor 

 which is an estimate of the volume of a specific fill material needed to 

 create a unit volume of native beach material. In most cases, fill factors 

 exceed one, indicating that the particular borrow material is less than 

 ideal and that winnowing processes will selectively remove unsuitable parts 

 from the fill until it becomes compatible with existing beach sediments. 

 The second model enables calculation of a renourishment factor which is 

 used to estimate how often placement of a particular fill will be required 

 to maintain specific beach dimensions. In all, three "fill factor" and 

 one "renourishment factor" approaches have been published (Krumbein and 

 James, 1965; Dean, 1974; James, 1974, 1975). James (1975) provides a more 

 detailed and formalized discussion of the four methods. 



These methods are based on simplistic models of the extremely complex, 

 interactive littoral zone system. Each uses composite mean and sorting 

 values of the native and borrow materials as basic input. Thus, at best 

 their predictions can only be as good as the basic composite data used. 

 Fill factors and renourishment factors can be useful and revealing calcu- 

 lations, and powerful tools when considered within the entire engineering 

 framework of a problem. 



28 



