K-40 
Summary (cont.) 
• Under the null hypothesis, 5% of the sites (P 0 ) would be expected to have 
low IBI scores, even if all sites in a segment were in good condition (i e, no 
low DO, contaminant, nutrient enrichment problems) 
f ** ^ 
: ' V /. - \ 
• Segments declared impaired if P greater than expected under the null 
hypothesis 
P - P > 0 (with 95% confidence) 
Summary (cont.) 
• Variance components in P added 
• Variance in P due to estimating thresholds - from bootstrap 
• Sampling variation within segment - binomial 
• Confidence interval of P - P 0 = 
P-P 0 ± 1.96(SE P + SE P ) = P - P 0 * 1 96*SQRT(Var p * Var p J 
i ( p _ ~p\- 
Var,, = Variance from bootstrap = V -r Plus variance 
1 5CXK) — l 
from segment = (pq/N-1) 
appendix k • 2006 303(d) Assessment Methods for Chesapeake Bay Benthos 
