formalin-seawater solution stained with rose bengal. If stored 

 for a period of time greater than 3 months, the benthic samples should 

 be transferred to 70 percent isopropyl alcohol. After identification 

 and numeration, reference samples should be stored in 70 percent 

 isopropyl alcohol . 



(d) The number and the locations of stations should be chosen 

 carefully before the project begins. Addition and deletion of stations 

 should be avoided as much as possible. The number of stations should 

 be adequate to address spatial variability of fauna. 



(e) Replications should be adequate (1) to account for variability 

 within station fauna, and (2) to collect the majority of the species 

 inhabiting the operation site. Refer to Hurme, Yancey, and Pullen 

 (1979) to determine the number of replicate samples required. 



(f) There should be a sufficient temporal frequency of sampling to 

 address seasonal variation in the physical and biological parameters; 

 at least seasonal sampling is recommended. 



(g) Sampling methods for "pre," "post," and "during" operations 

 should be consistent and comparable. 



(h) Identification of organisms should be confirmed by an expert 

 for each taxonomic group. New or undescribed species should also be 

 sent to a specialist for further confirmation or identification. 



(i) Consistency in all data analysis (methods, designs) should be 

 maintained. 



( j ) Planning of the fieldwork, timing of the field trips, and 

 devising the sampling plan should be carefully done (Hurme, Yancey, 

 and Pullen, 1979) . 



4. Population Analysis . 



The level of reliability of population analysis depends on the quantitative 

 accuracy of the samples. Good quantitative data collected before, during, and 

 after beach nourishment can be analyzed for changes in species diversity, 

 abundance, and biomass using valid statistical approaches. This type data can 

 be used in evaluating recolonization after a disturbance. Several commonly 

 used statistics are discussed by Hurme, Yancey, and Pullen (1979). 



5. Manpower Requirements . 



Manpower estimated for collecting, processing, and analyzing benthic data 

 varies depending on the location of sampling, site conditions and size, number 

 and type samples to be taken, the size of animals collected (macrobenthos 

 or meiobenthos) , and the level of taxonomic identification. As a general rule, 

 project time for an assessment can be prorated as follows: field time, 10 to 

 25 percent; sample processing, 50 to 75 percent; data analysis, 5 to 10 per- 

 cent; and preparation of an assessment document, 10 to 20 percent. Sorting 

 (or picking) macrohenthic samples generally takes 1 to 4 hours per sample 

 depending on whether or not the sediment ia fine or coarse. Processing time, 



