Abesamis et al How much does the fishery at Apo Island benefit from spillover from a marine reserve? 



373 



On the other hand, the highest maximum monthly 

 CPUE and IPUE were invariably found near Apo Re- 

 serve (Fig. 5, Table 3). This result is consistent with 

 the occasional spillover of bigger fish from the reserve. 

 Such an effect may have a positive influence on the at- 

 titudes of fishermen toward reserves (Russ and Alcala, 

 1996). However, in the case of the artisanal fishery at 

 Apo Island, any psychological impact of large catches 

 near the reserve on fishermen is probably attenuated 

 by the importance of the northern fishing grounds. In 

 other fisheries, however, occasional spillover of large 

 adults may be important. Recreational "trophy" fish- 

 eries, for example, may benefit directly from such an 

 effect of no-take reserves (Bohnsack, 1998; Johnson et 

 al., 1999; Roberts et al., 2001). In New Zealand, Kelly 

 et al. (2002) showed that lobster catch rates (kg/trap 

 haul) were similar close to and far from Leigh Marine 

 Reserve, although catches around the reserve consisted 

 of fewer individuals. The lobsters caught near the re- 

 serve were bigger. Furthermore, the amount of money 

 made per trap haul close to the reserve was similar to 

 sites far from the reserve. 



The findings at the scale of a few hundred meters 

 from the boundary of the reserve provide little evidence 

 to indicate that spillover from Apo Reserve is present. 

 In fact, the most informative result at this spatial scale 

 was that fishermen seemed to avoid fishing very close 

 to the reserve (i.e., within 100 m from the boundaries) 

 (Table 5). Furthermore, catch rates seemed to be low- 

 est closest to reserve boundaries for hook-and-line and 

 spear fishing (Fig. 6, A and C). However, gillnet fishing 

 seemed to have a pattern of decreasing catch rate away 

 from reserve boundaries (Fig. 6B). Gear selectivity in- 

 teracting with the spatial distribution of target species 

 may explain the differences in spatial patterns of CPUE 

 among the three gears. Gillnet fishing is probably less 

 selective than hook-and-line and spear fishing. 



The pattern of fishing effort found very near to the 

 reserve can be interpreted in two ways. Firstly, local 

 fishermen may be well aware of the distribution of catch 

 rates near Apo Reserve, and they adjust their fishing 

 effort accordingly. This interpretation is supported to 

 some degree by the correspondence between the pattern 

 of CPUE and the pattern of fishing effort for hook-and- 

 line and spear gun gear. Both CPUE and fishing effort 

 for these gear were highest at intermediate distances 

 (100-200 m), but lower at the closest (0-100 m) and 

 farthest distances (200-300 m) from the reserve (Fig. 6, 

 A and C, Table 5). Fishermen may know from experi- 

 ence that hook-and-line and spear gun catch rates are 

 low nearest the reserve (0-100 m), hence they tend to 

 avoid fishing in this area. Experimental fishing with 

 hooks and line, but specifically targeting A^. vlamingii, 

 supports the contention that hook-and-line CPUE is low 

 very close to the boundary of Apo Reserve (Abesamis 

 and Russ, 2005). Hook-and-line CPUE for N. vlamingii 

 was higher at intermediate distances (150-200 m) than 

 at the closest (50-100 m) and farthest (250-300 m) 

 distances from the reserve (Abesamis and Russ, 2005). 

 Although movement of fish from Apo Reserve to sites 



at intermediate distances (100-200 m) from the re- 

 serve has never been demonstrated directly, research 

 in the last two decades has shown that fish populations 

 (Acanthuridae, Carangidae, Serranidae, Lutjanidae, 

 and Lethrinidae) have increased outside but close to 

 the southern boundary of the reserve (about 200-250 m 

 from that boundary) after about eight years of reserve 

 protection (Russ and Alcala, 1996; Russ et al. 2003, 

 2004). This increase in populations may indicate that 

 the present spatial distribution of hook-and-line and 

 spear gun effort at the local scale may reflect the re- 

 sponse of a small number of fishermen to spillover from 

 the reserve. 



Secondly, local fishermen may avoid fishing very close 

 to the boundaries of Apo Reserve in order to prevent 

 being accused of poaching inside the reserve. The in- 

 consistent result of a higher catch rate but lower fishing 

 effort nearest the reserve for gillnet fishing supports 

 this idea (Fig. 6B, Table 5). Furthermore, there may 

 be tremendous motivation for resident fishermen not 

 to be implicated in poaching inside the reserve. The 

 ultimate reason for this probably lies in the smallness 

 of the community at Apo Island. Almost everyone has 

 known each other for most of their lives. The no-take 

 reserve was established by the community for their own 

 benefit (Russ and Alcala, 1999). It has been guarded 

 and maintained by community members, many of them 

 fishermen themselves, for most of the two decades of its 

 existence (Russ and Alcala, 1999). It has an important 

 role in tourism on the island, from which the local com- 

 munity has benefited considerably in many ways (Alcala, 

 1998; Russ and Alcala, 1999). It is the principal factor 

 that has made Apo Island a nationally and internation- 

 ally recognised model for successful, community-based 

 resource management (Alcala, 1998; Russ and Alcala 

 1999). Many residents are probably aware of most, if not 

 all, of these achievements. For these reasons, it is clearly 

 against the best interests of a resident fisherman, and 

 his family, to be labeled by fellow community members 

 as a threat to the security of their reserve. 



In conclusion, this study has shown that spillover 

 yield from the no-take reserve at Apo Island probably 

 contributes much less than 10% of the overall yield to 

 the local fishery. Fishing effort was often lowest near 

 the reserve, despite higher catch rates there. The find- 

 ing of low fishing intensity near reserve boundaries 

 despite indications of spillover contrasts with most em- 

 pirical evidence collected so far (e.g., Gell and Roberts, 

 2003) and recent theoretical studies (e.g., Salomon et 

 al., 2002) indicate that spillover will increase fishing 

 effort near reserve boundaries. Fishing effort adjacent 

 to the reserve appears to be limited by 1) weather deter- 

 mined by the monsoons, 2) the traditional importance 

 of the northern fishing grounds, 3) high variability of 

 catch rates and income rates, 4) lower value of target 

 species found near the reserve, and 5) social pressures 

 related to the history of community management of the 

 reserve. However, the present study has no informa- 

 tion on how fishing effort, yield, and catch rates near 

 the reserve have changed over the past 20 years since 



