Table 6. Sensitivity of power to within-site variance (τ₁²), between-site initial variance (τ₂²), and between-site trend variance (τ₃²). For each parameter, the power to detect -3% trends over 20 years was estimated at maximum and minimum values of the parameter as estimated across species and effort levels, while keeping the other parameters constant at their average value across species and effort levels and keeping the number of sites and the sampling frequency at minimum levels (10 sites sampled every five years). Sensitivity was then calculated by subtracting power at the maximum value from power at the minimum value of that variance parameter.

Variance parameter	τ12	τ22	τ32
Sensitivity of power	0.551	0.116	0.055