FIXED-LOCATION HYDROACOUSTIC MONITORING DESIGNS FOR ESTIMATING FISH PASSAGE USING STRATIFIED RANDOM AND SYSTEMATIC-SAMPLING

Five alternative finite sampling designs are compared using 15 d of 24 -h continuous hydroacoustic data to identify the most favorable approa ch to fixed-location hydroacoustic monitoring of salmonid outmigrants. Four alternative approaches to systematic sampling are compared among themselves and with stratified random sampling (STRS). Stratifying sy stematic sampling (STSYS) on a daily basis is found to reduce sampling error in multiday monitoring studies. Although sampling precision was predictable with varying levels of effort in STRS, neither magnitude nor direction of change in precision was predictable when effort was v aried in systematic sampling (SYS). Furthermore, modifying systematic sampling to include replicated (e.g., nested) sampling (RSYS) is furth er shown to provide unbiased point and variance estimates as does STRS . Numerous short sampling intervals (e.g., 12 samples of 1-min duratio n per hour) must be monitored hourly using RSYS to provide efficient, unbiased point and interval estimates. For equal levels of effort, STR S outperformed all variations of SYS examined. Parametric approaches t o confidence interval estimates are found to be superior to nonparamet ric interval estimates (i.e., bootstrap and jackknife) in estimating t otal fish passage.