IMPACT OF FISHING AND HABITAT DEGRADATION ON THE DENSITY OF BANGGAI CARDINAL FISH ( Pterapogon kauderni , Koumans 1933 ) IN BANGGAI ARCHIPELAGO , INDONESIA

Banggai cardinal fish (Pterapogonkauderni, Koumans 1933) is uncommon example of a marine fish with distributed in small range area while being in highly exploited. This fish is in high demand as an ornamental fish. However, the information on the number of density is limited. An underwater visual fish census survey was conducted in June to July 2010 at 18 fishing sites around Banggai archipelago to estimate the density of the stock and assess the impact of fishing and habitat on density. The areas are divided into three main islands, namely Banggai Island, Peleng Island, Toropot-Tumbak-Labobo Island. The lowest density index of the P. kauderni recorded at Kindandal village on Peleng Island, 0.014 fish/mwhile the highest abundance index of 3.0 fish/m found at Toropot village at Toropot Island. In three survey sites (Bonebaru and Toropot villages) where the fishing activities are still ongoing, the density has declined compared to the survey conducted in 2004. Majority of the villages in Peleng Island have lower density compared with the other islands probably due to the degradation of microhabitat of P. kauderni. In many cases, microhabitat degradation might be as a result of collection of sea urchins and sea anemone for consumption by local community.


INTRODUCTION
The endemic Banggai Cardinalfish (Pterapogonkauderni) in habits a small area in the Banggai-Sula platform Archipelago, Eastern Indonesia.It is mainly found in shallow sheltered bays and in harbours, on silty reef flats with sandy bottoms and sea grass beds (Allen, 2000).Depth distribution generally ranges between 0.5 and 6 m, but the species is most commonly found between 1.5 and 2.5 m (Vagelli, 2005).First described from specimens collected in the Banggai Islands of Sulawesi in 1920 by Koumans (1933), it was forgotten by the Western scientific community until 1991 when a Bali tour operator chanced upon the species and brought it to the attention of a taxonomist (Lunn & Moreau 2004).Since its rediscovery, P. kauderni became heavily exploited as it is highly-prized in the aquarium trade (Lunn & Moreau, 2004).
Fishing pressure may result a negative impact on P. kauderni population, such as affecting the density, group size and the density of its preferred associate fauna (sea urchins and sea anemones).Vagelli & Erdmann (2002) reported that in Bangkulu Island, approximately 10.000 individuals of Banggai Cardinalfish per month were caught and between 50.000-60.000P. kauderni were received for export each month at North Sulawesi buyer alone.The study carried out by Vagelli & Erdmann (2002) also reported that density in three different sites (Bokan, Limbo, and Masoni) islands was "cropped" due to heavy fishies pressure, namely 0.029, 0.031, and 0.027 individual fish/m 2 respectively.Moore & Ndobe (2005) also reported that density of P. kaudernihad varied from 0.31 to 11.99 individual/m 2 at several sites in Banggai Island, Peleng and Toropot Island.
The aims of this study are to analyse the underwater visual fish census survey data conducted in 2010 and compare the findings to a previous survey conducted in 2004(Moore& Ndobe, 2005).The analysis will then focus on the effects of fishing and habitat on the observed densities of P. kauderni at 18 sites.

MATERIALS AND METHODS Underwater Visual Fish Census Survey
The underwater visual fish census surveys were conducted from June to July 2010 in 18 sitesin the Banggai Archipelago area, identified as habitat of

29-36
Banggai Cardinal Fish.This wasbased on information obtained from local fishers.Four survey sites were located at BanggaiIsland, nine sites were located at Peleng Island, three sites atToropot Island, one site at Tumbak Island, and one site at Labobo Island (Fig. 1).Table 1 gives an overview of the survey sites and the exploitation history of the sites.
At each site, transect was set in shallow water close to shore at a depth between 30 cm to 150 cm.In areas where the habitat iswide (shallow area stretching far from the coast line), transect was started at a maximum of 200 m distance frombeach line.This was mainly done to avoid having transects going through villages which are built into the water.All transects were 50 meters long (Figure 2) and it is assumed that the diver can observe fish up to 10 meters along the transect (5 meters to each side) (Labrosse et al., 2002).The location of transect at each site was set in the main fishing areas as identified by local fishers.At each transect the following data were recorded: 1.The number of fish observed.
2. The habitat type along transect was estimated as percentage of the total surveyed area.The habitat types identified were a. Coral reef b.Sea grass bed c.Mangrove d.Sea urchin e. Anemone 3.After the transect had been surveyed specimens for biological sampling were caught at the transect using a small scoop net with a

Density Estimates from the Underwater Visual Fish Census Survey
Density varied among the 18 different sitesfrom 0.014 fish/m 2 to 3 fish/m 2 .The lowest density was found in Kindandal village at Peleng Island while the highest density recorded was in the Toropot village at Toropot Island.In the two villages where fishing is still ongoing, the average density decreased from 1.96 fish/ m 2 to 1.49 fish/m 2 at Bonebaru village, and 11.99 fish/ m 2 to 3 fish/m 2 at the Toropot village in Toropot Island (Table 2).At two sites (Monsongan and Tinakinlaut) where fishing stopped in 2006, the density increased from 0.48 fish/ m 2 to 2.63 fish/ m 2 at Monsongan village but from 0.31 fish/ m 2 in 2004 to 2.41 fish/ m 2 in the 2010 survey at the Tinakinlaut site.This means that density increased from 5 to 8 fold after harvesting stopped (Table 2).

29-36
Likewise to Monsongan and Tinakinlaut the density in Matanga village were fishing also stopped in 2006, has still decreased.The density was 64% lower than estimated in the 2004 survey.Most of the survey sites in Peleng Island have low density estimates compared to the other islands.The lowest density estimate of 0.014 fish/ m 2 is from the Kindandal village while the highest density of 0.498 fish/ m 2 was from the Bobu village as shown in Table 2.

Habitat types in relation to the density index of P. kauderni
Table 3 presents lists the relative vegetation and substrate coverage along transects at the 2010 survey sites.In Figure 3 the relationship between the percentage of habitat type and density is plotted.There is significant between the percentage covered by a given habitat type and density.The results from a correlation test are summarised in Table 4, in all cases the tests do not show significant correlation between density and the coverage of a given habitat type.

Fishing Activity and Habitat Type and its Impact to the Density Index of P. kauderni
In the two villages (Bonebaru and Toropot village) where fishing activities were still ongoing at the time of survey (2010), average density index decreased from 1.96 fish/m 2 in 2004 to 1.49 fish/m 2 in 2010 in Bonebaru and from 11.99 fish/m 2 to 3 fish/m 2 in Toropot.Fishing activity probably caused the decrease in the density of P. kauderni over the period.
Ndobe & Moore (2009) reported a significant coral reef degradation at five sites around Banggai archipelago including at Bonebaru from survey/ monitoring data collected in 2004 to 2006 compared to the latest survey conducted in 2011 by Ndobe & Moore (2012).According to this data, area covered by coral had reduced from 25% to 11%.The latest survey conducted by Wijaya (2010) who reported that the densities of P. kauderni at Bonebaru were 0.65 fish/m 2 , Mbato bato 0.42 fish/m 2 , Tolokibit 0.31 ind/ m 2 , and Bandang Island 0.87 fish/m 2 , respectively.The survey results reported by Wijaya (2010) were similar with these of current work.The correlation tests showedweak positive correlation between density of P. kauderni with the size area covered by other fauna such sea urchins and sea anemones.However the tests were not significant.Sea urchins and sea anemones are known as microhabitat of P. kauderni (Lilley, 2008).Ndobe et al. (2008) reported that massive extraction of sea anemones was first observed in a survey in 2007.This resulted in a drastic decline of P. kauderni after the sea anemones, that were numerous in 2004 and 2006, had all disappeared.Moore & Ndobe (2012) also stated that all recruits groups of P. kauderni of more than 3 individuals were associated with sea anemones, often also inhabited by clownfish.Sea anemone seems to be a particularly important microhabitat for newly released P. kauderni recruits and small juveniles.
In two sites (Monsongan and Tinak from laut) where fishing stopped in 2006, the density has increased from 0.48 fish/ m 2 to 2.63 fish/ m 2 in Monsongan village while 0.31 fish/ m 2 to 2.41 fish/ m 2 in Tinakinlaut site.It means thatdensityincreased 5 to 8 fold since fishing stopped in 2006.It also implies that the population of P. kauderni has recovered to some extent in the four years after fishing stopped in 2006.
Most of the survey sites in Peleng Island likely have lower density compared to the other islands.In Peleng Island, the lowest density of 0.014 fish/ m 2 was recorded in Kindandal village while the highest density of 0.498 fish/ m 2 was observed at Bobu village.Unfortunately, no survey data were available for comparison from 2004.Fishers and local resident in Peleng Island collect sea urchins for consumption, as bait for bubu (trap), or as feed for Napoleon fish.The fishers catch Napoleon fish and rear the fish temporarily in a small pond before they sell it to the buyer.The fish is fed on sea urchins collected by the fishers.Sea urchin (Deademasetosum) and sea anemone are the main shelter of P. kauderni and as one of important habitat in their lifecycle stage.Study related to the importance of microhabitat was conducted by Ndobe & Moore(2012) stated that over 80% of new P.kauderni recruits were assosiated with sea anemones and sea urchin.This may explain the low density of P. kauderni at Peleng Island.

CONCLUSION
P. kaudernii density increases when fishing activity no longer conducted.Similarly at sites where fishing is on-going density is lower in 2010 than was recorded in 2004.Majority of the villages in Peleng Island have lower density compared to other islands due to the degradation of microhabitat of P.kauderni as a result 29-36 of collection of sea urchins and sea anemone for consumption by local community.The habitat types appeared to be not significant to influence the density of fish according to the data presented here since the existing data have been very limited in scope compared to the total habitable area of the population.The results suggest a better monitoring of the fishery for P. kauderni would be essential in the future including log-books of daily record of catches by village and biological sampling (length measurements, maturity data, etc).

Figure 2 :
Figure 2: A schematic drawing of the underwater visual census survey design.mesh size of 4-5 mm and the length of net used was 60-80 cm.The material of the net is nylon monofilament.The material sampled was then analysed at the laboratory (See next section).4. The following environmental parameters were measured at each transect: temperature, salinity, dissolved oxygen, pH, and depth.The depth was measured by depth sounder, while temperature, salinity, pH, and dissolved oxygen were measured by Lutron Digital Oxygen Meter with a polarographic type probe.5. Fish density for each transect was calculated according to the formula: D = n/w, where D is density (individuals per m 2 ), n is the number of fish counted at a transect and w is the size of the transect area in m 2 .

Figure 3 :
Figure 3: The correlation plot of habitat areas to the index density of P. kauderni in 500 m 2 transect (Monsongan and Bonebaru-b were excluded in sea grass and fauna habitat type).

Table 1 :
Sites surveyed for Banggai Cardinalfish density from June to July 2010 on the Banggai Archipelago, Indonesia.Longitude and Latitude are on a decimal form.X means that the availability of density estimated from 2004

Table 2 :
Moore & Ndobe (2005)survey with density estimates and for comparison the estimates from the 2004 survey.Abundance is the density multiplied with the size of the area as estimated in table 2. The aesthetic (*) means survey data conducted byMoore & Ndobe (2005)

Table 3 :
Estimates of habitat types at transect sites in the 2010 P. kauderni survey and the associated density estimate

Table 4 :
Value of pearson's correlation (r) and probability value (p-value) obtained from linear model analysis Impact ofFishing and Habitat Degradation on The Density….inBanggai Archipelago, Indonesia (Kasim, K., et al)