Abundance and Catch Composition of Three Fishing Gears (Hook-and-Line Trap and Spear) in a Coral Reef. Santiago Island, Cape Bolinao, Philippines

The catch rate and species composition of a multispecies reef fishery around Santiago Island, Cape Bolinao, Philippines, was studied from June to December 1986. Three fisheries were monitored, hook-and-line, spear and trap fishing. A total of 144 species were caught by the three gears. One hundred twenty-seven species were represented in the catches of the trap fishery. One hundred one were present inside the reef lagoon and 90 outside the reef lagoon.The family Siganidae contributed 42.2% of the total catch outside the lagoon and Family Labridae contributed 26.5% inside the lagoon. The hook-and-line fishing was dominated by lethrinids (69.8%), specifically Lethrinus rodopterus. Spear fishing was dominated by Siganus canaliculatus, contributing 83.7% of the total catch. The catch rates for the trap fishery were 0.129 kg/haul, 0.131 kg/haul inside and 0.127 kg/haul outside the reef lagoon. The catch rate for the hook-and-line and spear were 0.589 kg/line hr. and 1.33 kg/man hr respectively. The size composition and the lengthweight relationships for the most abundant species of the three fisheries are discussed. The fish yield of a reef area of 9.06 sq. km to the 15-m isobath was 2.46 mt/sq km during the rainy season. The annual fish yield was estimated to be around 4.17mt/sq km yr.


Introduction
The Philippines is an archipelago composed of more than 7000 islands. It's coast line stretches approximately 34,417 km and is bordered by 27,000 km of coral reef surface. This unique geographical feature gives the Philippine people a strong orientation toward the sea and fishing. The significance of the reef ecosystem to the Filipinos is it's productivity. Seafood is a source of relatively cheap protein food and livelihood in coastal communities (Bolanos and Alino, 1984). While about 60% of the total fisheries catch is obtained from the coastal waters in general (Gomez et.al,1981), at least 10 to 15% of fish production is associated with coral reefs (Carpenter, 1977). Smith (1978) calculated that coral reefs and associated communities of depths of less than 30m cover about 600,000 kmA2 on a global° basis. 30% of these communities are found in the area bounded by Indonesia to the west, Northern Australia to the south, the Philippines to the east, and mainland Asia to the North. 30% is in the Indian Ocean, Arabic Gulf and Red Sea, 13% in the South Pacific, 12% in the North Pacific, 14% in the Caribbean and North Atlantic, and 1% in the South Atlantic. The most heavily exploited areas are believed to be in parts of the Caribbean and the Philippines, and localized in areas close to major urban centers throughout the tropics (Munro and Williams, 1985).
Coral reefs, seagrass beds and mangrove swamps form a widespread, highly productive complex of shallow marine communities in the tropics. The total production of fish in these communities has never been calculated but it is undoubtably substantial (Johannes, 1980). In tropical regions, coral reefs have one of the highest fish yields. They may rank second to estuaries and ponds in 20 tons/ Sea-grass fish productivity, having a range of 0.09 to sq km/yr (Marten and Polovina, 1982).
beds are often interposed between intertidal mangrove communities and offshore coral reefs. No measurements of fish production are available for sea-grass beds but it is known that primary production is exceptionally high (Odum et al. 1973;Zieman, 1975).
Where these communities occur together, their faunas overlap considerably (Olsen et al. 1973). Many species of fish which spend their adult lives in the reef, live as juveniles in the adjacent seagrass beds or mangrove swamps. Many species venture out on the seagrass beds to feed at night, returning to the safety of the reef during the day (Austin, 1971;Munro, 1974). Harvesting gillnets (Munro and Williams, 1985).
other fishing gears such as spear guns, seine nets, and a variety of other devices are used in specific fisheries.
Size selection and species selection differ with each fishing method. Hook-and-line is usually used to catch predatory fishes and it is highly selective in  Many of the standard techniques for sampling fish populations can only be used with great difficulty in reef environments (Sale, 1982). The choice of a suitable method is in large determined by the type of reef, the limitations and biases of the gear, and the constraints confronted by the investigator, such as time and manpower. Russell et al. (1978) listed the techniques commonly used in the collection of coral reef fishes. They described a number of methods widely used in stock assessment, and noted the advantages and disadvantages of each method. Among the methods used were fish traps, handlines, handspears, Hawaiian slings, spear guns, gillnets, and others. These methods are highly selective in terms of size or species, and are usually suitable only for certain reef types or topographies (Sale, 1982;Gomez, 1981). This is not to imply that traditional methods have not been used in the study of reef ecology. Munro (1976), Munro et al. (1973), and Thompson and Munro, (1974b) have used fish traps to produce highly creditable examples of catch composition and fish yields using "traditional" approaches (Sale, 1982). Relatively good assessments of fish abundance and species compositon can be obtained  (Munro and Williams, 1985). Marten and Polovina (1982) found that fish yields from coral reefs were similar to those for other continental shelf fisheries despite the higher primary productivity of coral reefs. Stevenson and Marshall (1974) attempted a generalization as to the fisheries potential of coral reef and adjacent shallows, and suggested that coralline shelves with good cover of actively growing coral reef, seagrass beds, and algae usually produced around 4 to 5 mt/kmA2/year.
More recently, Alcala (1981), Alcala and Luchavez (1982), Marriott (1984), and Wass (1982) reported estimates ranging from 15 to 27 Tons/KmA2/yr. It seems that conflicting reports on yield from coral reefs have resulted from comparison of areas with greatly different mean depths and physiographic features of the coral reef environments (Alcala and Gomez, 1985;Munro and Willians, 1985). Reef fishing is generally done in a patch of coral reef (which is highly productive) and in sandy bottoms (which are not so productive). The yield per unit area that is calculated for a reef depends very much upon the size of the area and the percentage of that area which is actually covered by coral or other hard substrate (Marten and Polovina, 1982). Saila and Roedel (1980) recommended that research to develop methods for rapid assessment of stock based on enviromental indices be initiated. For example, a morpho-edaphic index could be developed for coral areas in which the index was based upon the relative proportions of different habitats on the shelf (Munro and Williams, 1985). A habitat approach emphasizes attention to species composition (including a classification of fish communities), how different fish communities are associated with different habitats, and how community composition responds to human activities (including fishing) which impinge upon the fishery (Smith et al. 1973;Marten, 1981;Marten and Polovina, 1982). This perspective does not mean that massive amounts of quantitative data are required in more detail than before, but that the data must be sensitive to species composition.
Although the production of multispecies fishery can be manipulated to some extent by adjusting the harvesting regime, the yield from the fishery can be 6 reduced immensely by habitat destruction (Gomez et al. 1981;Marten and Polovina, 1982;Munro and Williams, 1985). Destructive fishing practices such as dynamiting, poisoning, seining and trawling disrupt the production of fish food or spawning of fish. Another factor is habitat destruct)on originated by non-fishing activities such as mining, siltation due to run-off, oil drilling and pollution from urban centers (Carpenter and Alcala, 1977;Alcala and Gomez, 1979;Corpuz and Alino, 1983;Hudson et al. 1982;Marten and Polovina, 1982;Bolanos and Aline, 1984).
The fish yield from an ecosystem cannot be inferred from records without reference to the fishing effort behind those records (Marten and Polovina, 1982).
Fishing effort is very important for yield estimates, because such estimates are generally obtained by multiplyng catch per unit effort (from sampling data) by the total effort in the fishery (Munro, 1978;Gulland, 1979;Marten and Polovina, 1982). Determining the fishing effort of coral reef fisheries can be extremely difficult because of the large number of fishermen involved and the different users of the resource. The yield from a multispecies fisheries is not only a matter of how much fishing, but also, what kind of fishing. Marten (1979aMarten ( , 1979b summarized the impact upon the stocks in Lake Victoria due to the amount of fishing and the kind of fishing. He showed 7 that above a certain fishing effort, the total multispecies catch is not affected much by fishing effort per se, but by the kind of gear employed. He also indicated that there was not a gear mix which is optimal for all species in the fishery. What is optimal for one species may under-exploit or destructively exploit another species. Fishing gear also has indirect ecological effects upon fish species that may not even be captured by that kind of gear because of predation and competition, and these effects may lead to successional changes in the species composition and age composition of the fishery (Marten, 1979a;Craik, 1982;Marten and Polovina, 1982;Munro and Williams, 1985).  Munro (1980) and Sainsbury (1982) reached the same conclusions.
Multispecies fisheries management and yield assessment will have to use a more empirical approach, based on observations of how fish community composition and yields change under different circumstances, taking advantage of the work done in different places and with different fishing effort and fishing conditions (Munro, 1980;Marten and Polovina, 1982;Kirkwood, 1982;Sainsbury, 1982;Munro and Williams, 1985).
So far, the studies conducted in coral reef areas have focused mainly on the estimation of fish yield rather than on both the fish yield and species composition (Marten and experimental studies in Polovina, 1982). Most of the reef areas involve small species and are carried out within localized areas (Sale, 1982;Munro and Williams, 1985). The accuracy with which the results of these studies can be expanded into general statements varies with the sophistication of the experimental designs (Munro and Williams, 1985).
Marshall (1980) (Munro, 1974a;Miller, 1986). The fundamental concept is the same in most areas. The fish enter the trap via one or several entrance funnels, generally · 12 designed for easy entry and difficult escape. Munro et al. (1971) and Munro (1974) studied the dynamics of  (Powles and Barans, 1980). High and Beardsley (1970) contended that fish enter the traps for reasons other than the bait. Random movements, use of traps as shelter, curiosity, intrapecicific social behavior, and predator escapement are probably very important factors in trap catches. Munro (1983) assumed that the minimum size retained for a given species is a function of the mesh size and the maximum body depth of the species. Hartsuijker and Nicholson (1981); Luckhurst and Ward (1986) suggested that recruiment to the trap fishery may be more a function of behavioral changes with size than the mesh size used. Munro et al. (1971) suggested that conspecific attraction in increasing the ingress rates of a species was important. Luckhurst and Ward (1986) suggested that conspecific attraction was a major factor affecting the species composition and abundance of fish in traps. Catch rates undoubtedly vary according pattern.
to moon phase and corresponding tidal Craig (1976) and Luckhurst and Ward (1986) observed that greatest catches were usually associated with rough sea conditions.  (Craig, 1976;Boardman and Weiler, 1980). Thompson and Munro (1974a,c) (Munro, 1983). Kawaguchi (1974) reported that experienced line-fishermen usually catch an average of 50% more than less experienced fishermen.
The success of the fishing operation depends largely on the fisherman's ability to locate the fishing ground, based on his knowledge of the preferred habitat and the movements of the different fish species.
The fishing power in hook-and-line fishing is A number of studies have reported substantial bias in length-frequency samples derived from angling (Ralston, 1982). Fry (1949);Frazer (1955)  Catch and effort data were obtained from note-books of middlemen, and from interviews with the fishermen.
The fishing power of each gear was standarized in order to get a reliable index of relative abundance. The unit of effort used for the trap fishery was the haul rather than hours or day assuming that fishermen knew optimum soak time for traps depending on the area. This approach was previously used by other researchers (Munro, Taylor 1974b;Stevenson and Stuart-Sharkey, 1980;and McMichael, 1983 To test if location influenced the size of capture an analysis of variance was done for the trap fishery. (2.5) Length-weight Relationship Weights were measured to the nearest gram whenever possible. Length-weight relationships were calculated for all the species for which adequate data was obtained. For the analysis, the power function: calculated by ordinary least squares on log-transformed data with bias correction was used where W = weight in g, L = length in cm, a and b are constants. (Saila, Recksiek, Prager, and Chen., 1980).
The fish yield (Y) in mt/sq km/yr for the rainy season was calculated using the formula that Alcala and Gomez (1985) Opening Detail description of a fish trap use to catch coral reef fishes in Cape Bolinao, Philippines. Picture of a typical bambo raft used for navigation and fishing operations by trap fishermen around Cape Bolinao, Philippines.

FIGURE 5
Picture of atrap cover with coral in order to simulated a coral head.    -    period. It was difficult to tell when fishermen had permanently ceased fishing traps because part-time fishermen sometime did not use then for periods of one to five consecutives days.

Outside reef
On an annual basis, probably more traps are set on the outside reef rather than inside the lagoon. October, increasing to a higher value in December and November. A summary of monthly trap catches, effort and CPUE inside the lagoon is given in Table 4.    - The summary of the monthly total catch and total catch by location have been plotted in Fig        accounting 26.5% of the total catch. This is followed Table 6 Major Tami lies represented in catches OT traps in the outer •dge OT a coral reeT lagoon. Cape 8olinao. Philippines.    -   Table 9. Results from the statistical test between the slope of the two areas for each of the species showed no significant differences (P< 0.05) between these species. Slope (b) values from the length-weight relationship were tested for uniformity by the construction of 95% confidence intervals (Fig 14).

===============================================================
The length-weight relationship for each species are plotted in Appendix 7.  Ninety-five percent confidence intervals for the slope values from the length-weight relationship of eight species from the trap fishery. The fishing operation is carried out as follows: Fishermen usually put out from shore individually or in pairs in an small non-motorized banca (Fig 17). The operation took place from sunset to sunrise. Fishermen spent 2 to 3 hours getting to the fishing grounds. When they reached the fishing grounds a candle was lit and the banca was allowed to drift. The first hour is usually spent fishing for bait. The bait (squid or pieces of fish) was cut in strips and attached to the hook. In addition to bait, fishermen would at times spread the head and the guts in the water in order to increase fishing effectiveness. Fishermen stayed in the 58 FIGURE 15.
Fishing gear used for the hook-and-line fishery in Cape Bolinao. Philippines

FIGURE 16
Hand made fishing hooks from stainless steel.

( FIGURE 17
Traditional non-motorized banca used for the hookand-1 ine fishermen in Cape Bolinao. 61 fishing grounds as long as the biting of fish was good; when the biting was poor they either switched fishing grounds or came back to shore. (3.6) Catch, effort and catch per unit effort.
The fishing effort of 80.5 line-hour yielded a catch of 176 fish weighing 49.2 kgs. The monthly average catch per line-hour (CPUE) was 0.589 kgs (s.d.= 0.187) ranging from 0.324 to 0.935 kgs (Fig 18). Figure   19 shows the monthly catch ranging from 2.34 to 12 kgs with an average of 7 kgs (s.d.=3.6). The monthly total catch estimates are plotted in Figure 20. The average number of fishing days was 16 (s.d.=1.7); the average number of fishermen per month was 9.75 (s.d.=1.5); and the average fishing hour-day was 5 (s.d.=1.8). Effort, total weight, total number of fish, CPUE, and estimate total catch are given in Table 10. Figures 21 and 22 show the relationship between catch and CPUE versus total effort. A regression analysis of the catch versus the effort is given in figure 23. The monthly number of fish is plotted in Figure 24. Catch per unit effort was calculated for the three most abundant families: Lethrinidae, Lutjanidae, and Serranidae (Fig 25).       The monthly abundance of the three major families is given in Table 12. !!_ethrinus ornatus ranged from 18.7 to 33.2 cm with a       Table 14 Relationship OT Total Length <TL) to weight <W> Tor Tive species Tram the hook-and-line Tishery In wherQ a and b are constants.

FIGURE 26
Fishing lamp use in the spear fishery with a modified gas tank and stainless steel lamp shade.

FIGURE 27
Fishing gear use for the spear fishermen, wooden goggles spear gun and wooden flippers.

79
The operation usually starts at midnight and ends at daybreak. In the case of a full moon night it will start just after the moon disappears. There is a certain degree of kinship between the spear fishermen in a crew. It is normal to find father, son and relatives fishing together from the same banca. The monthly number of fish is given in Figure 31.
Effort, total weight, total number of fish, CPUE, and    Figure 34.
As mentioned before spear fishing in Cape Bolinao is mainly a one-species fishery. The family Siganidae accounted for 83.7% of the total catch and 88.7% of the total number. The families comprising more than 1% of the total weight and number caught are shown in Table   16. Siganus canaliculatus accounted for 82% by weight and 87% by number.      Effort (men/hour) Liniar regression between total catch and total effort for the spear fishery. Cape Bolinao. June-December 1986.     Table 18 Relationship of total length «IL>,standard length CSL) and standard length <SL> to total length «IL> for five species from the spear Fishery wh&re a and b are conSUints.
The area of the fringing reef to the 15-m isobath is about 9.06 sq km (Fig 1)   in Apo and Sulimon Island (Alcala and Gomez, 1985).
These researchers also reported that the trap fishery    deeper than 30m. Some lutjanid and lethrinid species were caught in the traps, which indicated that sea-grass beds or shallow waters are used as nursery grounds during their early life stages, while they are associated with the reef as adults. Munro (1974) and Rivas (1970)  The abundance of large size fishes can be due to the large mesh size (4 inches) used for this fishery.
As was mentioned before, one species accounted for 82% of the spear fishery. There were other species but , they only accounted for 18% of the catch. Calvelo and Ginon (1974) (1977) and Tsuda et al. (1974). The differences are probably due to differences in growth rates between the sample areas and to the fact that the fish used in this study were from 9.5-19.5cm (

Mulloidichthys falvolineatus
The smallest M. falvolineatus caught in the spear fishery was 12.lcm (TL) and the largest was 20cm. Rau and Rau (1980) reported that fish in the size range of 20-30cm were common and a maximum size was 40cm. The length-weight relationships for this species showed a allometric growth b<3 (r=.96). These estimates of length distribution and the regression coefficient agree with the estimate reported by De la Cruz (1986) for this species in a fish corral in Guiuan Eastern Samar Philippines.

Siganus guttatus
The minimum size of ~ guttatus caught by the spear fihery was 11.4cm and the largest was 21.Scm (TL). Rau and Rau (1980) reported a common size of s.

Lutjanus fulviflarnrna
The minimum size of Lutjanus fulviflarnrna caught by the hook-and-line fishery was 19.8cm and the largest was 30.Scm (TL). Rau and Rau (1980), reported a common size range between 25-30cm and a maximum of 35cm.
The small sizes observed in the trap fishery indicated that growth overfishing has occurred.
Johannes (1980) reported that this has taken place in many reef and lagoon areas throughout all the tropics. invertebrates. These are not included in this study. Alcala and Gomez (1985) reported fish yields ranging from 5 to 36 mt/sq km yr in the Central Philippines.
Some of these estimates are very high in comparison with the estimate of this study. This is due in part to the fact that only coral reef bottoms were included in their estimates. The fish yield from this study is greater than the fish yield reported by Munro (1977) in 7)The lack of standarization of the area used in fish yield estimates needs to be solved. The stratification of the area to be studied may be an answer to this problem. An obvious division would be between reef, coastal lagoon, seagrass beds and further sub-divisions based on the type of botton substrate.

8)Reports of high fish yields for heavily exploited
reefs, such as Cape Bolinao should be carefully reviewed and monitored. In many cases, It would be preferable to report more conservative estimates. It is better to report underestimates than overestimates, in order to make more rational decisions.