(2) Pier 13 - Systematic selection of 10 piles in each of 3 pile 

 categories (30 samples) . 



This pier was divided into three strata: (1) batter piles, 

 (2) bearing piles supporting crane rails, and (3) bearing piles 

 not supporting crane rails. A Tandy Model 100 computer was 

 used to generate a single random number within the range of 

 piles in the first skip interval for each stratum. Using the 

 pier plan drawings, a corner of the pier was selected as a 

 starting point. The total number of piles belonging to a 

 stratum was divided by 10 to determine the skip interval for 

 the systematic pile selection procedure. The first pile was 

 located by selecting a random number within the first skip 

 interval. Subsequent piles were located along the length of 

 the pier by passing over the number of piles equal to the skip 

 interval and marking the pile at that location. This procedure 

 continued until all the samples in each stratum were chosen 

 (Figure 18). 



Note: After the samples were taken for Piers 14 and 13, it was 

 determined that there was not a significant difference among 

 the mean values of the three strata. Therefore, it was decided 

 to treat all the bearing piles as a single group for these piers 

 as well as subsequent piers inspected at St. Helena. 



(3) Pier 12 - Random selection of 10 bents and 3 piles within each 

 bent (30 samples) . 



Ten randomly selected bents were chosen. Within each selected 

 bent, three random piles were chosen to be sampled (Figure 19). 

 This pier has a relieving platform and required the divers to 

 use surface-supplied air. Because of the inherent dangers to 

 the divers on this type of pier, the divers were in two-way 

 communication with the personnel topside. This increased the 

 speed of sampling and aided in the recording of the diver 

 observations. 



(4) Pier 11 - Proportional and Variable Sampling. 



A random selection of 96 piles was sampled based upon propor- 

 tional sampling criteria assuming the most difficult proportion 

 to estimate: that 50 percent of the pilings had borer damage. 

 These piles were evaluated for borer damage during the Level I 

 inspection since no measurement data were required (Figure 20) . 



To acquire the measurement for the variable sample, five bents 

 were randomly selected from the pier. Within each bent, from 5 

 to 7 random piles were selected for a total of 30 samples 

 (Figure 21). 



23 



