7615 - Laserfiche WebLink

Home Browse Search

Statistical power of the Interagency Standardized Monitoring Program The power of the monitoring program is dependent on the degree of population increase that is to be detected. Relatively small changes (ie a 50% increase) in CPUE are not statistically detectable (Tables 7-8; Appendix B, Tables 18-22). Increased sample sizes did not improve the statistical reliability of the monitoring program to detect changes at that level. However, the reliability improves dramatically when the goal is to detect changes of 100% and greater. Changes of 200% and greater are detectable at powers exceeding 90% in all cases (Tables 7-8; Appendix B, Tables 18-22). The time frame during which the change occurs is important as well. The power of the adult-monitoring program to detect changes in CPUE is much greater for a five-year period than for a two-year period (Table 7; Appendix B, Tables 18- 21). Statistical power is also much greater when CPUE changes are considered river wide rather than for a specific reach. These time-frame and river-wide considerations have important biological meaning as well. It is unlikely that a fish as slow growing and long lived as the Colorado squawfish will experience a significant increase in adult population over a two-year period. A significant change is more likely to occur over an extended period of 5 to 10 years or longer. Also, because Colorado squawfish are migratory, population changes are unlikely to affect only one reach. The change is likely to include several reaches or the whole river. The current monitoring program is very reliable in detecting river-wide changes in CPUE over a period exceeding five years. The power of the age-0 Colorado squawfish monitoring program also improves when comparisons are made river wide and over a five-year period (Table 8; Appendix 16