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



363 



fishing grounds visited, names and number of fishers, 

 date and times when fishing started and finished, fish- 

 ing gear used, and composition and weights of catches 

 (verified from records offish buyers). Interviewers were 

 able to determine precisely where fishing was conducted 

 because the names and locations of fishing grounds 

 around Apo Island are common knowledge to local people 

 (Fig. 1). However, if fishing was done near Apo Reserve 

 (e.g., at Katipanan, Tumoy, Kanigaran, Ubos, and Kan- 

 uran; Fig. 1), fishermen were also asked to estimate 

 how far away they were from the boundaries of the 

 reserve. Color-coded marker buoys, which fishermen 

 could locate and identify easily while fishing, indicated 

 approximate distances from reserve boundaries. These 

 buoys were moored permanently at 100 m (blue), 200 m 

 (red), and 300 m (yellow) from each of the northern and 

 southern boundaries of Apo Reserve (Fig.l). The buoys 

 were installed with the aid of a GPS receiver and fish- 

 ermen were informed about them one week before the 

 study started. 



Data were collected from 80 full-time and part-time 

 resident fishermen (Baybay-16, Ubos-47, Cogon-17). 

 This group comprised 70-80% of all fishermen at Apo 

 Island. The information in the present study probably 

 reflects accurately the actual patterns of fishing effort 

 and yield of the local market-oriented fishery. The ma- 

 jority of fishermen at Apo Island sell their catch to fish 

 buyers at each of the three fish landing sites monitored 

 (Maypa et al., 2002). The fish buyers usually purchased 

 their fish from a group of "loyal" fishermen. Also, most 

 fishermen live close to the houses of the fish buyers 

 (landing sites). Thus, interviewers were able to collect 

 data from fishermen regularly. Fishing effort and yield 

 data were not adjusted to account for all resident fisher- 

 men (i.e., any fishermen not included in interviews). 



However, data were not obtained from the following: 



1) catch sold to part-time fish buyers on the island, 



2) catch sold directly to the main island of Negros, 3) 

 fishermen visiting from Negros, 4) catch sold dried, and 

 5) catch brought directly to homes for consumption. 

 Items 1-4 are probably minor contributors to the total 

 marketed yield. Maypa et al. (2002) surveyed the same 

 fish landing sites monitored in the present study and 

 estimated that only 10% of the total marketed yield 

 (presumably from items 1-4) did not pass through the 

 three major fish landing sites. On the other hand, yield 

 from subsistence fishing (item 5) may be comparable 

 in quantity to the marketed yield (White and Savina, 

 1987), but lower in monetary value. It is unlikely, how- 

 ever, that data on fishing effort collected in this study 

 would differ much from data on subsistence fishing. It 

 seemed common that fishermen went to sea to catch fish 

 both to sell and keep for personal consumption. 



Data analysis 



Targeted species were classified into five groups accord- 

 ing to Bellwood (1988): reef-associated species (Carangi- 

 dae and Sphyraenidae), reef planktivores (Acanthuridae, 

 Caesionidae, and Pomacentridae), reef species (mainly 



Lutjanidae, Lethrinidae, Scaridae, Serranidae, Kyphosi- 

 dae, and octopus), open water species (Belonidae, Elopi- 

 dae, and Scombridae), and off-reef species (Lutjanidae). 

 Fishing grounds were classified into three groups: the 

 northern fishing grounds (Enas, Ulo, Kasorenyo, Cogon, 

 Punta Cogon), the western fishing grounds (Largahan, 

 Kan-upi/Boluarte, Baybay, Katipanan), and the fish- 

 ing grounds near Apo Reserve (Tumoy, Kanigaran, 

 Ubos. Kan-uran) (Fig.l). Data were included in the 

 third group if fishing was done s300 m from reserve 

 boundaries. These three fishing grounds had roughly 

 similar surface areas to the 20-m isobath (northern 

 fishing grounds, 17.3 ha; western fishing grounds, 14.3 

 ha; fishing grounds near Apo Reserve, 18.9 ha). It was 

 assumed that most fishing was done within or just out- 

 side the reef area enclosed by the 0- and 20-m isobaths 

 (Fig. 1). In addition, data for the fishing grounds near 

 Apo Reserve were classified into the following categories 

 of distance from reserve boundaries: 0-100 m, 100-200 

 m, and 200-300 m. 



Interviews allowed collection of fishing effort data 

 even when fishermen returned from trips without catch- 

 ing anything (i.e., fishermen who used hook and line, 

 gill nets, and spear guns). On average, about 20% of 

 total fishing trips (or 23% of total fishing effort in per- 

 son hours) returned with no catches. This figure varied 

 considerably according to fishing gear (hook and line; 

 48%, gill net: 3%, spear gun: 7%) and months. However, 

 fishing effort data for trips with zero catch were col- 

 lected only beginning in September. Hence, the recorded 

 fishing effort (in person hours) in July and August was 

 adjusted by adding a correction factor in order to ac- 

 count for fishing trips with no catches. Correction fac- 

 tors were calculated from the equation 



EC ^[ERx(EZ / ET)]/[1-{EZ / ET)], 



(1) 



where EC and ER = the correction factor (expressed 

 in person hours) and recorded 

 monthly fishing effort, respectively, 

 for July or August; and 

 EZ and ET = the total fishing effort with zero 

 catch and the total fishing effort, 

 respectively, in September. 



Values for September were used because this month 

 is within the same season as July and August (SW 

 monsoon). Correction factors were calculated per gear 

 (hook and line, gill net, and spear gun) and per group 

 of fishing grounds. However, prior to adjustment, the 

 recorded monthly fishing effort for July was multiplied 

 by three to obtain an estimate for a 30-day period. The 

 yield for July was also multiplied by three to obtain a 

 30-day yield estimate. Catch per unit of effort (CPUE, 

 in kg/person per hour for each month was calculated 

 by taking the average CPUE of all individual fish- 

 ing trips made in a given month. Monthly CPUE was 

 calculated per gear, per group of fishing grounds, and 

 per distance from reserve boundaries (0-100, 100-200, 

 and 200-300 m). July and August CPUE were adjusted 



