PRISTINE MANGROVE COMMUNITY IN WONDAMA GULF , WEST PAPUA , INDONESIA I

Papua shoreline has been considered as the most extensive mangrove area in Indonesia and great size of their trees. It implies that the forest could be a perfect habitat for biota food web due to its high productivity. Wondama mangrove, part of Cendrawasih National Park – West Papua, had a promising mangrove forest to be discovered. Nevertheless, the forest was rarely studied due to limited access and costly, so that the scientific data was difficult to be found scientifically. A preliminary study was carried out to identify the condition of mangrove in this area. The objectives of this study was to analyze mangrove community health and structure. Canopy coverage and vegetation structure data were collected from 107 circular plots included three plot areas (radian: 5m, 2.5m and 1m) which were distributed to determine three plant classes i.e. tree, sapling and seedling, respectively. Present study showed that Wondama mangrove is a pristine mangrove community. It had large size of tree trunk diameter by 19.77± 6.55cm averagely. Its diameter size affected on low tree density which was less than 1000 tree/ha. Those two parameters were strongly negative-correlated each other. On the other hand, community canopy was highly covered mangrove area by 82.46±6.43%. Rhizophora has mostly dominated in the forest. Mangrove regeneration in Wondama was excellent referring to density and species composition of sapling-seedling levels.


INTRODUCTION
The Papua shoreline is the heaven for mangrove community.Data of Geospatial Information Bureau (2012) and Ministry of Forestry (2013) showed that approximately 49.6% of mangrove area in Indonesia is found in Papua.It interpreted that Papua islands (only in Indonesia territory) has the most extensive area of mangrove in the world since Indonesia is well-known as the largest mangrove habitat on earth (Giri et al., 2011).Slightly sloped and wide coastal geomorphology, large number of wide estuarine and tropical condition provide appropriated environment for mangrove distribution.In addition, Papua mangrove has a low anthropogenic pressure which is implied by low degradation rate of mangrove area in Papua until 2012 (Ilman et al., 2016).
Mangrove is one of the most habitats along its coastline behind seagrass and coral reef who has played the strategic role in maintaining ecology, physics and chemistry function balance in the coastal ecosystem.It is also a potential for mitigating climate change by reducing sea level rise impact and sequestrating carbon dioxide (CO 2 ) (Alongi 2012;Donato et al., 2011dan Murdiyarso et al., 2015).Papuan uses mangrove for alternative food sources; construction materials; energy alternatives, traditional medicines and ecotourism services (Arobaya & Pattiselanno, 2010).
Wondama regency is located along semi close estuarine and small gulf that is a part of Cendrawasih Gulf National Park.It has a complete coastal ecosystem to maintain biological and social needs.Limited exploration and well-published information of Papua mangrove, especially in Wondama Regency would be impacted on careless of its existing.Costly research operational and limited cozy accommodation in Papua are considered as some obstacles for conducting research in Papua mangrove.For this study, we would like to present the recent condition of mangrove community in Wondama regency.The objective of this study is to analyze its mangrove community health and structure as an initial information for public policy and future sophisticated research.

Study Area
Wondama Gulf Regency is located in west part of Cendrawasih Gulf, approximately 200 km from Manokwari City.Pristine terrestrial vegetation is vast-growth in hilly land topography.As many as five rivers provide freshwater and nutrients on estuarine habitat where mangrove is existed.The largest river flows from the southeast edge of Wondama gulf.Tidal type is mixed semidiurnal, which presents twice high and low tide daily.Substrate types are varied from sand, mud and sandy mud.The study sites were focused in mangrove area along closer gulf system from Wamesa village until Aisandami village (Table 1; Figure 1).

Community Structure Assessment
Sampling location were determined by purposive stratified random sampling on each site based on mangrove species composition structure and time availability.As many as, 107 circular plots were distributed to represent mangrove area in Wondama gulf.There were three different area size of plots to determine different classes among mangrove vegetation, i.e : 5m-rad plot (tree class, DBH>4cm); 2.5 m-rad plot (sapling, DBH<4cm) and 1m-rad plot (seedling, height<1m).The class determination was followed Asthon & Macintosh (2002).On each plot, girth of trunk on breast height (GBH) was measured from each identified mangrove species.Species identification was based on Giesen et al. (2006).

Canopy Coverage Analysis
Mangrove canopy coverage was analyzed using hemispherical photography method which was developed in mangrove community health by Dharmawan & Pramudji (2014).As many as 98 photo samples from 12 sites were captured by Himax Polymer X 5MP front camera.Community canopy coverage is calculated as the ratio between number of canopy pixel and number of total pixel in a photo hemisphere and multiflied by 100 to get the percentage of canopy coverage.

Data Analysis
Univariate and normal distributed data (plant density from each class; trunk diameter/DBH, canopy coverage) were analyzed using ANOVA to identify a variance among sites and followed by Tuckey test.Correlation among sites was identified and clustered using Eucladian distance and pearson analysis.Multidimensional scaling (MDS) based on tree's important value index (IVI) was generated and combined with Pearson rank correlation to determined sites group tendency.

RESULTS
On this study, as many as 18 true mangrove species were distributed in 12 sites (Table 2).WDMM12 has the highest species diversity among sites which was found 13 species.Meanwhile, only five species were found in WDMM10 as the lowest species number.Rhizophora and Sonneratia species were noted at most of study sites.
Tree density in most of study sites were relatively less than 1000 tree/ha (Table 3).It would be the impact of average of trunk diameter size which was implied by Pearson correlation coefficient (r=-0.60,P<0.05) between tree density and trunk diameter.Totally, Wondama mangrove has large size of trunk which was 19.77±6.55cm of those average and maximum at 78.59cm (Figure 1).Canopy of mangrove community covered more than 75% of mangrove area.It represented that mangrove community health in Wondama was totally in pristine condition (Figure 2).
Spearman correlation coefficients showed that tree diameter had similar pattern with canopy coverage among sites, even they were poorly correlated (r=0.04,P>0.05).Diameter average had significant negative-correlation on tree and sapling density at -0.76 and -0.59, respectively.Canopy coverage of mature plants were not related with all density data.On the other hand, sapling and seedling density were likely contributed to seedling cover (Table 3).
According to Multi-dimensional scaling (MDS) ordination (Figure 3), high domination at WDMM01 implied this site was widely separated from other sites which were also dominated by Rhizophora mucronata.Mixed species composition in R. mucronata dominated sites and co-domination of Brugueira gymnorrhiza    Regeneration of mangrove in Wondama was excellent based on sapling and seedling density.Averagely, mangrove in Wondama gulf was growth at 1204±1397ind/ha and 7226±5746ind/ ha at sapling and seedling respectively (Figure 4).Sapling density was ranging from 255-3182 ind/ ha.Meanwhile, the seedling density was varied by 1592-17097ind/ha.WDMM12 had the bestregenerated mangrove due to the highest seedling density among others and its sapling density over 1000ind/ha.On the other hand, the highest sapling community was found in WDMM11 by 3182±3724ind/ha.However, Spearman rank analysis did not find the significant correlation among these densities data (Table 3).
In term of species composition, the regeneration of mangrove community was also doing well represented by species composition of sapling and seedling levels adhered to tree composition.Spearman rank showed that regeneration of sapling and seedling to tree level had the significant correlation at 0.94 and 0.91 (P<0.01)(Table 3).Moreover, seedling regeneration to its sapling species structure was also highly related (r=0.89,P<0.01).In summary, the dominating species at tree level had a significant contribution on sapling and seedling composition.

DISCUSSION
Mangrove community in Wondama Gulf was categorized at high diversity comparing with other location in Indonesia.Hinrich et al. (2009) found 21 species of mangrove in Segara Anakan which was mixed between true and associate plants.Though tree density number on Wondama was lower than that study, mean diameter size was higher.Another study, in Kaledupa island, Wakatobi Marine National Park was only noted eight true mangrove species which was dominated by B. gymnorrhiza (Jamili et al., 2009).As many as eight species also was marked in Ambon gulf (Suyadi, 2009).Mangrove species in Tanjung Lesung, Banten Province, also had lower species number (13 species) than present study (Ati et al., 2014).
High diversity of mangrove in Wondama gulf was affected by various habitats at area sites such as riverine, estuarine and fringing.Those tree habitat types had different environment settings affecting species adaptation.For instance, in muddy sand, substrate Riverine sites at WDMM12 was the highest species diversity among sites.Salinity tolerant was responsible for species distribution spatially.The fringing site usually has the higher salinity than estuarine and riverine habitat (Ewel et al., 1998).
Mangrove in Wondama had a specific preference for species distribution.Avicennia alba and A. lanata were tended to grow in estuarine habitats.However, A. marina had vast growing in the fringing sites.Joshi and Ghose (2003) also found this tendency, which A. marina tends to grow in higher salinity.Avicennia alba was growth at 20.9 psu as the optimum salinity level.Salinity was impacted on the physiological process.Leaf area and osmotic potential on seedling were decreasing due to salt concentration increasing (Burchett et al., 1984).Salinity had a negative correlation to assimilation rate, stomatal conductance and intercellular carbon dioxide concentration (Clough and Sim, 1989).
Rhizophora stylosa and R. lamarckii would be found only at fringing habitat.In contrary, R. apiculata and R. mucronata were easily found on each habitat.It showed that R. apiculata and R. mucronata had wider range environment tolerant.It was proved by Hoppe-Speer et al. (2011) and Ball et al. (1997).A comparison study between R. apiculata and R. stylosa salt tolerant found better growth rate of R. stylosa at higher salinity (Ball et al., 1997).Salinity response in Rhizophora species could be identified on leaf biomass.Higher exposed salinity stimulated ticker leaves area and water content (Camileri and Ribi, 1983).
Brugueira gymnorrhiza and B. sexangula distribution had no specific habitat preferences.On the other hand, B. parviflora tended to be found in estuarine.Sukardjo et al. (2014) noted that B. parviflora was more distributed on higher salinity and silt concentration on sediment.Weiss et al. (2016) discovered that estuarine mangrove had lower salinity gradient, higher soil organic carbon, and water-extractable phosphate than marine mangrove.Salinity influences biomass carbon storage on B. gymnorrhiza seedling in Sundarbans (Agarwal et al., 2016).
Nypa fruticans was specifically distributed in the riverine site which has lower salinity than other habitats.Some studies found the similar distribution of Nypa, such as Bunt et al. (1982); Hamilton and Murphy (1988); Robertson et al. (1991), Ukpong et al. (1991) and Theerawitaya et al. (2014).Ukpong et al. (1991) represented that Nypa distribution had a significant and negative correlation to salinity.Robertson et al. (1991) found that N. fruticans was growth in 1-10 ppt salinity range.
High tree-sapling community coverage and mean diameter size were represented that mangrove community in study sites was in pristine condition.Previous studies on natural mangrove forest were found high canopy coverage, i.e.Rote island, NTT: 77.55%-88.69%(Dharmawan, 2015a); Jailolo and Dodinga, North Moluccas: 75.78% and 78.93% (Dharmawan, 2014); and Auki island, Papua: 90.73% (Dharmawan, 2016).Mangrove canopy coverage could indicate a natural level of mangrove ecosystem and detect anthropogenic threats (Dharmawan and Pramudji, 2014).For instance, mangrove sites in Sombano and Lewuto villages, Wakatobi which were nearby local residences, were found a lot of logging scars and affected to canopy coverage percentage ranged from 25.48% to 43.40% (Dharmawan, 2015b).
Tree density of Wondama mangrove was relatively lower than other pristine mangrove areas.It was highly related to a large size of tree diameter.Previous studied described a direct negative relationship between diameter and tree density.Mulyadi and Amin (2016) found higher density (2467-3367tree/ha) in Dumai, Riau Province.However, the trunk diameter size was lower ranged from 4-9cm.Similarly, Calegrio et al. (2015) quantified 10,810 plant density per ha on 3.7-7cm mangrove forest structure.Another study by Zhila et al. (2014) on 7.31-22cm diameter range was determined 1679 tree/ha of plant density.
Mean diameter size was another proves how natural Wondama mangrove is.Reanalyzing some mangrove sites in Indonesia found that eastern Indonesia mangrove forest had higher mean tree diameter and lower density (Dharmawan and Pramudji, 2017).Rapid coastal development and high anthropogenic in western Indonesia triggered mangrove area and quality declining (Ilman et al., 2016).Diameter size was related to the age of plant individually.12-years old Rhizophora plantation in Kenya and Malaysia were consisted by 5132ind/ha and 4000ind/ha, respectively (Engo et al., 1998;Kairo et al., 2008).
Regeneration of mangrove in the present study was excellent.It was implied by a number of seedling distribution.Gan (1995) suggested as many as 5000-10000 seedling densities was recognized as a good regeneration of mangrove forest.However, large domination and pristine community of tree level pressured young plant accretion causing the low density of sapling.The favorable regeneration was also represented by significant correlation on species composition at all plant levels.Most domination of Rhizophora at all levels was likely improved correlation coefficient.Rhizhophora was widely distributed in Indonesian mangrove (Spalding et al., 2010).Chapman (1986) described that Rhizhophora had many ways and a wide-tolerant adaptation to cope salinity gradients.

CONCLUSION
Wondama mangrove is one of the pristine mangrove areas in Indonesia.The large size of diameter and canopy coverage indicated the absence of human intervention and proved how natural the forest is.Mangrove species distribution had a specific habitat preference depend on its adaptation tolerance.The regeneration status was excellent both densities and species composition.

ACKNOWLEDGEMENT
The recent study was successfully conducted due to a great teamwork by Marine Life Conservation Unit, LIPI -Biak staff which was coordinated by Andriani Widyastuti, M.Si.This study was funded by annual budget (DIPA) of Marine Life Conservation Unit, LIPI -Biak.Special thanks to Boby Otoluwa and Paul Inggamer for technical assistances during survey.Lastly, a grateful appreciation would be presented to Andrian Wikayanto and Gravinda Putra Perdana, staffs of Information Technology Division-Indonesian Institute of Science (BIT-LIPI), who were joined and well-documented our entire survey.

Figure 2 .
Figure 2. Tree trunk diameter, tree-sapling density and community canopy coverage of mangrove in study sites of Wondama Mangrove (WDMM) and those trend lines

Figure 4 .
Figure 4. Sapling and seedling densities on each sites

Table 2 .
True mangrove species diversity in Wondama Gulf, West Papua

Table 3 .
Pearson correlation rank among tree diameter and density, community canopy coverage, sapling and seedling density.