Systematic errors due to condition of the level, rod out of plumb, 

 temperature of tape, slope of tape, and tape not on line were considered 

 insignificant and had no great effect on the data collected. Bad turning 

 points undoubtedly resulted in some error, but since the leveling was not 

 closed back to the bench mark, there is no definite method of determining 

 specifically when an error might have occurred or to what extent. Another 

 source of systematic error results from the sag of the tape and wind effects 

 on taping. The magnitude of this error is assumed to be an average maximum of 

 -O.i foot per 200 feet of tape length. 



Taking into account these error possibilities and various other errors due 

 to human and environmental causes, the data were considered "accurate" if 

 every point on the profile was within ±0.05 foot vertically and ±0.5 foot 

 horizontally of the actual values. The data were also considered "dependable" 

 if sufficient checks on the survey data were performed to ensure that no per- 

 sonal errors affected the data. Based on these criteria, it was concluded 

 that the data obtained were of acceptable accuracy and dependability. 



5. Data Reduction and Quality Control. 



Until 1968, survey data were recorded in field notebooks, reduced and 

 hand-plotted by the surveyors, and then forwarded to CERC. These plots were 

 later digitized and placed in a punchcard format. After 1968, the survey data 

 were still recorded in fieldbooks, but the data were then transferred to 

 optical scanning forms before being sent to CERC. At CERC the data were 

 logged and scanned with an optical mark page reader (OMPR) to produce punch- 

 cards. The cards were then read into a computer where the data were processed 

 using an editing program which plotted profile points. From these plots, 

 apparent errors were identified and returned to the surveyors for correction 

 or comment. A final edit check was made and the data were stored in a 

 magnetic-tape format when all detectable errors were satisfactorily corrected. 



A quality control study by Czerniak (1973) indicated a 25 percent proba- 

 bility that there would be an error of ±0.1 foot in the recorded elevation of 

 a surveyed point due to rounding by the survey party in the field. Because of 

 the improbability of this rounding error occurring numerous times on the same 

 profile, this error, if present, should have no adverse affect on any data 

 analysis. 



Figure 24 diagrams the basic steps taken throughout the BEP program from 

 the initial observation in the field to the final computer output. 



Appendix B provides a tabulation, by profile, of all the survey data 

 collected during the study. 



6. Data Analysis . 



Two primary parameters calculated from the profile data are (a) the change 

 in MSL shoreline (AS) and (b) the change in unit storage volume (AV). The 

 first parameter, AS, is the horizontal change, between surveys, of the posi- 

 tion of MSL at a profile line. If the beach at MSL prograded during the time 

 between surveys, a positive number would result for AS; a negative value 

 would result if the beach receded. The second parameter, AV, is the change 

 in volume above MSL between two surveys for a unit width parallel to the 

 shoreline at a profile line. If accretion occurs between surveys, AV will 

 have a positive value, and if erosion occurs, AV will be negative. 



30 



