The role of power (β) in restoration monitoring

The goal of restoration is to create, enhance, or restore a site into a self-sustaining, functional ecosystem. Success standards are developed to determine if a site is trending towards this ultimate goal. Quantitative monitoring is then used to collect data to determine if these success standards are being met. Although there are larger fundamental questions in restoration about what constitutes a functioning ecosystem and do success standards approximate a natural ecosystem, a more basic question is: Is the monitoring design capable of detecting true change (high power, low Type II error rate) in the variables of interest to determine success at a restoration site? This is a very simple question, however, power is not regularly considered when designing quantitative monitoring for restoration sites. A power analysis was applied to 3 restoration sites in San Diego County, California, two wetland sites and one upland site. Alpha was set at 0.10, Beta (power) was set at 0.90, and the minimum detectable change varied from 10 – 20 depending on the success standards for the three sites. The number of samples used in the first year of quantitative monitoring (serving as a pilot study) was found to give low power (< 0.90) at all three sites. Sample size will be increased in the remaining years of quantitative monitoring to raise power to 0.90. Including power in sample size analyses for restoration monitoring allows for collection of data that can detect true change and ultimately determine if success has been achieved at a site.