Evaluating the biological effectiveness of fully and partially protected marine areas

Table 5 The response to protection ( $\bar{LnRR}$ , [95% CI]) for each ‘exploitation status’ category (target-protected [TP], target-not protected [TNP] and non-target [NT]), using biomass data

	(a) All MPA case-studies included			(b) Sensitivity analysis; habitat confounded studies excluded
	TP	TNP	NT	TP	TNP	NT
	N = 12	N = 5	N = 7	N = 10	N = 5	N = 6
rho = 0	0.35 [0.12 → 0.59]	0.43 [0.07 → 0.79]	0.18 [−0.14 → 0.50]	0.42 [0.18 → 0.67]	0.43 [0.07 → 0.79]	0.23 [−0.10 → 0.56]
rho = 0.2	0.38 [0.12 → 0.64]	0.42 [0.03 → 0.80]	0.22 [−0.16 → 0.60]	0.44 [0.17 → 0.71]	0.42 [0.04 → 0.80]	0.29 [−0.11 → 0.69]
rho = 0.5	0.40 [0.11 → 0.69]	0.40 [−0.02 → 0.82]	0.26 [−0.17 → 0.68]	0.45 [0.15 → 0.76]	0.40 [−0.02 → 0.81]	0.35 [−0.10 → 0.80]
rho = 0.8	0.40 [0.09 → 0.72]	0.38 [−0.07 → 0.82]	0.28 [−0.18 → 0.74]	0.46 [0.13 → 0.80]	0.38 [−0.06 → 0.82]	0.39 [−0.10 → 0.88]
rho = 1	0.41 [0.08 → 0.74]	0.37 [−0.09 → 0.83]	0.29 [−0.18 → 0.77]	0.47 [0.12 → 0.82]	0.37 [−0.09 → 0.82]	0.41 [−0.10 → 0.92]

Results where different correlation coefficients (rho = 0, 0.2, 0.5, 0.8, 1) among individual species responses were used to compute the within-study sampling variance are given on separate rows. The summary effect size and 95% CI are given for analysis using (a) all PPA case-studies, and (b) excluding PPAs with habitat confounded effects. N represents the number of PPAs for each exploitation category. Ln(RR) > 0 means more fish inside the PPA; Ln(RR) < 0 means fewer fish in the PPA. When the CI overlaps 0 it implies a non-significant difference between PPA and non-protected area.