Community Based Ecological Mangrove Rehabilitation and Subsequent Development of Adaptive Collaborative Mangrove Ecosystem Management ABSTRACT

Community Based Ecological Mangrove Rehabilitation and Subsequent Development of Adaptive Collaborative Mangrove Ecosystem Management Ben Brown and Yusran Nurdin Massa, Restoring Coastal Livelihoods ABSTRACT CIFOR reports that 90% of reforestation projects fail to meet their restoration goals (source) and Lewis reports While great potential exists to reverse the loss of mangrove forests worldwide, most attempts to restore mangroves often fail completely, or fail to achieve the stated goals (Erftemeijer and Lewis, 2000; Lewis, 2000, 2005, 2009). An independent review of three post-tsunami mangrove restoration projects revealed that International NGO s and the Indonesian Forestry Department experienced 70-100% mortality of planted mangroves within 6 months at the majority of their sites (Brown and Yuniati, 2009). A pair of sites in Sulawesi were planted 4 and 6 times in partnership between government and communities experiencing 100% mortality over a 6 and 9 year period respectively (Melati, 2012 and Brown and Yuniati, 2007). The major cause of failure in all of these instances is the oversimplification of the process of restoration, reliance on direct planting of a low diversity of mangrove species into a substrate either not appropriate or unprepared hydrologically for mangrove establishment. Inadequate public consultation and involvement has also been cited as a major reason for failure. Ecological Mangrove Restoration and other methods which prioritize hydrological amendment and human assisted natural regeneration have been trialed and promoted now for several decades around the world. These methods are now being trialed by NGO s, governments and communities, in Indonesia, and are meeting with success of restoration and management goals. This paper discusses five such collaborative EMR projects, and their success in promoting the establishment and growth of diverse mangrove stands with healthy growth. This paper also looks at social-economic interventions (sustainable livelihood development, gender focused approaches, etc.) that along with successful mangrove rehabilitation, provide enabling conditions for community participation in development of adaptive collaborative management of both rehabilitated and existing mangrove stands. Key words: adaptation; biodiversity; collective action; ecology; forest management; gender; hydrological restoration; institutions; management; resilience; sustainability

Abstract 1.0 Introduction 1.1 Background 1.2 Improved Practices 1.3 Understanding Natural Recovery 1.4 Ecological Mangrove Rehabilitation 2.0 Methods 2.1 Case Study Sites 2.1.1 Tiwoho Village, North Sulawesi 2.1.2 Bengkalis Island, Riau 2.1.3 Simeulue Island, Aceh 2.1.4 Jaring Halus, North Sumatera 2.1.5 Tanakeke Island, South Sulawesi 2.2 Survey Design 2.3 Data Analysis 3.0 Case Studies 3.1 Tiwoho Village, North Sulawesi 3.2 Bengkalis Island, Riau 3.3 Simeulue Island, Aceh 3.4 Jaring Halus, North Sumatera 3.5 Tanakeke Island, South Sulawesi 4.0 Results 4.1 Ecological Results 4.2 The Development of Formal and Informal Community Forest Governance and Adaptive Collaborative Management 5.0 Discussion 5.1 Conditions for the Establishment and Long-term Sustainability of Community-based Mangrove Management The Development of Adaptive Collaborative Management 5.1.1 Damage and Successful Rehabilitation 5.1.2 Short-term Economic Benefits 5.1.3 Women s Participation 5.1.4 Access, Control and Ownership 5.1.5 Monitoring and Sanctions 5.1.6 Social Capital 5.1.7 External Assistance 6.0 Conclusions 1.0 INTRODUCTION 1.1 Background Mangrove forests are ecologically important coastal ecosystems composed of one or more of the 69 species of salt tolerant trees and shrubs. These ecosystems currently cover 146,530

square kilometers of the tropical shorelines of the world, which represents a decline from 198,000 km² of mangroves in 1980 to 157,630 km² in 1990. (Lewis, 2009). Thus, loss of mangrove area has occurred at a rate of 1 to 2 percent per year during the past three decades. Achieving no net-loss of mangrove forests would therefore require a minimum of 150,000 hectares of successful mangrove forest restoration per year at the current rate of loss. In Indonesia, mangrove planting projects are the norm, where government agencies, NGO s and disaster relief agencies support the rearing of mangrove seedlings in nurseries followed by direct plantings. Cases of near total failure (>95% mortality over 2-3 years) are common including; 9 separate planting projects on Simeulue Island, Aceh after the tsunami supported by Australian Red Cross (7) and District Forestry Department (2) (Brown and Yuniati, 2007), one planting site in Maros District, South Sulawesi which experienced total mortality on three separate occasions, and is being funded by Green PNPM for a fourth planting this year (Melati, 2011), and a case where BRLKT and the village government of Tiwoho in North Sulawesi planted an area 6 times over 9 years experiencing total failure each time (Brown, Personal Observation). This failure was attributed to three assumptions regarding restoration: 1) Mangroves can only be restored by planting, 2) roots causes of the failure of natural recruitment to take place are not important to investigate and 3) fluvial mud-flats (between LGT and MSL) are suitable for planting, when in fact they likely never supported a mangrove forest in the first place. 1.2 Improved Practices The five images in Figure 1 illustrate the restoration of a portion of a 500 ha (1236 acres) mangrove forest restoration project in Hollywood, Florida, USA, within the Anne Kolb Nature Park. The interesting part of the story is that none of the mangroves you see were planted! They are all what we call volunteer mangroves (Fig. 2) ones that colonized the site on their own after appropriate biophysical conditions were established. Such conditions include appropriate tidal hydrology to support natural colonization by the millions of floating propagules (seeds or seedlings) of mangroves naturally produced in this area every year. A three year pilot project prior to this major restoration effort had established the target topographic elevation that would facilitate natural recruitment of mangroves. The construction of a tidal creek in the center of the project was essential to replicate the natural tidal creeks

found in all mangrove forests and to allow for the movement of fish and invertebrates into and out of the forest as tidal levels rose and fell. A common goal for such a project, aside from healthy mangrove growth might be the restoration of 75% of a functional fisheries equivalent compared to a natural reference site within five years of constructing the project.

1.3 Understanding Natural Recovery Mangroves exist in a variety of intertidal situations, with the common denominator of substrate elevation. Nearly all true mangrove are found established and growing between Mean Sea Level, and Highest gravitational tide. Individual species and associations change based on environmental gradients, which may express themselves as ecotones. One pre-requisite for the re-establishment of mangroves is the availability of open substrate, at the appropriate elevation. This will influence both the frequency and duration of tidal inundation, which are key determinants to the natural establishment and subsequent growth of mangroves. In addition to the need to understand the existing hydrology as it relates to topography of an adjacent mangrove forest reference area, it is also important to understand the natural recovery processes in damaged mangrove forest areas, also known as secondary succession.

The natural secondary succession process in damaged mangrove forests often begins with the appearance of a nurse species (Fig. 2), which is typically herbaceous plant species such as Smooth Cordgrass, Spartina alterniflora, or Saltwort, Batis maritima. These species, which are associated with mangrove forests around the world, appear to facilitate the recolonization and eventual natural restoration of damaged mangrove areas. Although not well studied, facilitation may work by improving edaphic conditions and/or physical trapping of floating mangrove seeds. Such knowledge is important in planning and conducting mangrove restoration projects, because mimicking the natural secondary succession may lead to a more successful restoration in the long-run. (Lewis, 2009; Friess, 2011). 1.4 Ecological Mangrove Restoration There is enough evidence that over-simplified restoration of mangroves does not work, while Ecological Mangrove Rehabilitation trials in Florida and Indonesia have resulted in success, and led to increased community protection and development of adaptive collaborative management systems. Six Principles Successful mangrove forest restoration should be routinely successful if a few basic ecological restoration principles are applied at the early planning stages. These principles have been taught now at 3-5 day workshops around the world (India, Sri Lanka, Thailand, Malaysia, Indonesia, Cambodia, United States, Cuba, Nigeria, Brazil, Ecuador and Honduras), targeting mangrove managers, as well as community fisherfolk as active, hands-on participants. 1. Understand both the autecology (individual species ecology) and the community ecology of the mangrove species at a particular location, in particular the patterns of reproduction, propagule (seeds and seedlings) distribution, and successful seedling establishment. 2. Understand the normal hydrologic patterns that control the distribution and successful establishment and growth of targeted mangrove species. 3. Work together with local communities to assess the modifications of the previous mangrove environment that occurred and that may prevent natural secondary succession. 4. Select appropriate mangrove restoration sites through application of steps 1-3 above that are both likely to succeed in restoring a sustainable man-grove forest ecosystem and are cost effective, given the available funds and labor.

5. Design the restoration program at appropriate sites selected in Step 4 to initially restore the appropriate hydrology and utilize natural recruitment of propagules or seeds for plant establishment unless other-wise determined likely not to succeed. 6. Only utilize actual planting of propagules, collected seedlings, or cultivated seedlings after determining through steps 1-5 that natural recruitment will not provide the quantity of mangroves desired (or the desired rate of stabilization or growth of seedlings) established as quantitative goals for the project. Ecological Mangrove Rehabilitation courses include a field component during which participants examine pre-existing mangrove restoration projects as well as newly proposed sites for restoration. Learning from past mistakes, in order to improve techniques is often a transformative process. The following section of this paper discusses five specific case studies in Indonesia where community learning around restoration failure, resulted in community based implementation of Ecological Mangrove Rehabilitation (in projects now 1 10 years old). In each case, EMR was augmented by additional interventions including; the development of sustainable livelihood alternatives out of existing coastal resources, and community organizing towards the development of mangrove management plans. The following findings have been notable, and indicate both acceptance of the EMR method as an appropriate tool for community based coastal resource management; 1) EMR methods were understood by communities, practiced both as part of the activity and replicated on their own, and promoted by some portion of the community as a recommended practice when subsequent government or other planting projects visited their area. 2) Growth of EMR plots always met targets of at least 1250 seedlings per hectare growing healthy after year three of the intervention 3) Even in recent interventions where success is not yet evident, the logic of the EMR approach, in conjunction with associated programs (livelihoods development, coastal field school, etc.) has been effective in catalyzing the organization of communities to protect (formally or informally) their mangrove resources and advocate for mangrove conservation and sustainable utilization to some level of government.

2.0 METHODS 2.1 Case Study Sites 2.1.1 Tiwoho Village, North Sulawesi (01 35 00 N; 124 50 21.06 E) This study site is located on the mainland of North Sulawesi, and is at the Eastern-most boundary of Bunaken National Marine Park. EMR was undertaken in 25 hectares of cleared

mangrove situated within > 100 hectares of coastal, fringing mangroves with a natural diversity of 33 species of true mangroves. EMR implemented in 2004. 2.1.2 Bengkalis Island, Riau Province (01 27 00 N; 102 30 20 E) This study site is located on the Western Coast of Bengkalis Island, in the Province of Riau along a pair of river estuaries, the Jangkang and the Kembung. The main cause of mangrove degradation was due to clear-felling for export-oriented charcoal making. This system has a high level of peat formation, yet also experiences mangrove degradation along the coast due to changing current patterns coupled with potential effects of sea-level rise. EMR was trialed in a 33 hectare section ceded to the community, and included planting and natural recruitment areas. EMR implemented between in 2006-2007. 2.1.3 Simeulue Island, Aceh Province (02 30 58 N; 96 21 07 E) This study site includes 7 villages in NE Simeulue Island and 7 villages in Teluk Dalam, a large bay in north-central Simeulue. Mangroves disturbance on this island was natural, due to tectonic uplift ranging from 75 150 cm in the NE coast and 25-50 cm in Teluk Dalam as a result of the 2004 tsunami and 2005 Nias earthquake along the Sunda Megathrust. This relocated the intertidal zone, requiring an equivalent seaward migration of the island s diverse, fringing mangrove. Large parts of the island are propagule limited due to uplift of adult trees out of the range of tidal influence. EMR not yet implemented. 2.1.4 Jaring Halus, NE Langkat Wildlife Sanctuary, North Sumatera Province (03 56 34 N; 98 33 54 E) Jaring Halus is a 33 hectare village adjacent to a 40 hectare highly biodiverse village mangrove forest located at the mouth of the Wampu River. The system is an estuarine delta on a vast alluvial floodplain. EMR was undertaken in a 10 ha abandon shrimp pond complex in the adjacent, 9000 hectare wildlife sanctuary, and was followed up by enhancement planting by communities in bare patches in sporadic sections of the sanctuary which was largely logged over for illegal charcoal development. EMR implemented between 2007-2008. 2.1.5 Tanakeke Island, South Sulawesi Province (05 29 38 S; 119 18 28 E) This island lies off the SW tip of South Sulawesi province, a one hour small boat ride from the mainland. 1200 ha of mangroves were converted to aquaculture ponds in the 1990 s of an original 1776 ha of mangroves. This system is known as a mangrove over wash system, situated on an atoll with proportionately little terrestrial area compared to sub-tidal and intertidal systems. EMR implemented between 2010-2012

Each of the above sites have experienced mangrove disturbance due to aquaculture development or clear-felling for industrial charcoal production. Stakeholders at each site have tried to address this disturbance initially with direct planting of only a single or narrow range of mangrove species. In each case a high degree of planting failure ensued. Subsequently, community groups were engaged in Ecological Mangrove Rehabilitation, and have succeeded in adequately restoring their forests in older projects (5-10 years old), or are confident that EMR will work where the initiative is still new (<3 years). In each case, success or the acceptance of EMR as a superior method, has led to an increased degree of protection (formal or informal), and a change in attitudes and behaviors, which are presented anecdotally. 2.2 Survey Design To gather information on mangrove restoration experiences, conservation and utilization of mangrove resources; key informant interviews, as well as formal and informal group discussions were conducted over time (from 1999 to 2012) with communities, local NGO s, academicians and government agents. In addition, resilience assessments have been perfomed in Tanakeke Island, South Sulawesi and Simeulue Island, Aceh using a method (Skema Kajian Risilian) adapted from the Resilience Alliance (Resilience Alliance, 2007 & Sonjaya, 2011). The interviews and discussions focused on numerous aspects of the villages relationships with the mangroves, particularly the history of villages, traditional uses of the mangroves, conflicts with outside interests, history and mechanism of mangrove restoration, results of mangrove restoration, and the development of formal and informal mangrove management and protection practices, and rules governing those practices. Forest conditions have been monitored using a standard monitoring methodology, which has been adapted by the Mangrove Action Project Indonesia from various sources including English et al. (1994) and Duke (2011). The Tanakeke Simeulue and Riau sites were compared with nearby reference forests, as well as any existing chronoseres as benchmarks for success. Twelve (12) to twenty two (22), 100m 2 vegetation plots (10 m 10 m or 20 m x 5 m ) were established at random locations in each forest. In each plot, the girth at breast height (GBH at 1.3 m height) of each stem with a GBH 4 cm (1.27 cm diameter at breast height (dbh)) was measured, and height was calculated using either a clinometer or extendable rod. Species identification was done with assistance from the Handbook of Mangroves in Indonesia, (JICA, 1999), with the exception of Tiwoho where previous surveys by PhD. Rignolda Djamalludin

were also consulted. Complete data sets for Tiwoho, has been collected by PhD. Rignolda Djamalludin and his students from University of Sam Ratulangi but was not accessible at the time of this writing. 2.3 Data Analysis Analysis of recruitment and growth of natural recruits and planted material has not been consistently performed across all sites. Abundance, basal area and frequency for each species in reference forests and naturally regenerating chronoseres were computed at Tanakeke Island and Simeulue Island, following the procedure of English et al. (1994). Analysis has not yet been performed for data from Tiwoho or Bengkalis Island. Because of inconsistent methodologies and durations of monitoring, comparisons between sites, in this paper, will be made qualitatively. Social information has been compiled and analyzed by MAP Indonesia and a multi-disciplinary team from the Asia Pacific Study Center of the University of Gadjah Mada and are available in numerous project reports. 3.0 CASE STUDIES In order to get a clearer understanding of the village-level dynamics and relationship with the mangroves, especially the influence of community-based management, the reader is asked to refer to more detailed accounts of the site histories and management practices found in Brown, 2007; Sonjaya,2007; Sonjaya et Al, 2011, and Brown, 2012. The following accounts have been consolidated from those sources, as well as field experiences, informal interviews and discussion group data. Following the contextual descriptions, we summarize socio-economic, and ecological outcomes of EMR and associated interventions. 3.1 Tiwoho Village, North Sulawesi In 1990, the Bureau of Land Rehabilitation and Conservation (BRLKT) cleared 25 hectares of old growth mangroves (including 300+ year old trees), within the boundary of the Bunaken National Marine Park. Ten (10) hectares were developed as extensive shrimp ponds, slightly excavated with low dike walls and surrounding canals. These 10 hectares were operated

privately by an inexperienced owner for only 6 months, before the venture was abandoned. Over time, five of the ten hectares closest to the sea experienced significant natural revegetation, as their seaward dike walls crumbled and propagules from the adjacent forest (containing A. marina, A. alba, S. alba, S. ovata, S. caseolaris, B. gymnorhizza, B. paviflora, C. tagal, R. stylosa, R. apiculata and. granatum) established appropriate substratum. The remaining 20 hectares of cleared forest experienced low to zero re-growth (between 0 400 seedlings per hectare over 10 years), as dike walls and canals remained relatively intact. This area was planted 6 times over the 9 year period between 1995 (four years after abandonment) and 2004, by BRLKT and later the Forestry Department, but only working with a small number of local villagers (8 paid members of the Village Head s family). Bare root propagules of two species (R. apiculata and C. tagal) were planted, but without regard for ecological or hydrological needs. Total mortality occurred during each of the first five plantings. Before the 6 th government planting event, local NGO Yayasan KELOLA, along with MAP- Indonesia, PhD Rignolda Djamaluddin of the University of Sam Ratulangi, 50 community members and 120 school children undertook Ecological Mangrove Rehabilitation, including strategic breaching of dike walls, filling of drainage canals, and limited enrichment planting of 17 locally occurring species. Enrichment planting occurred between 2001 2005, with hydrological work taking place in 2004. By 2012, 23.5 out of 25 hectares of the entire complex had regrown at > 2500 seedlings per hectare. Areas of 4000-8000 stems per hectare are common, with canopy heights of 8-11 meters. 27 of the 30 recorded species of true mangrove occurring in the area have been observed in the rehabilitation site (see table 1). In addition to EMR, communities have been engaged in a wide variety of livelihood programs (organic farming, bamboo preservation for furniture making, improved cookstove production, processing of non-timber forest products, etc). Environmental education programs have been run for 5 years with elementary school children, and over 80 community members were involved in creating a village ordinance for mangrove protection, ratified in an open community meeting. No mangrove felling takes place in the entire village, and subsistence gathering of fisheries products is a daily activity in both the old mangrove and rehabilitation site.

Fig 4. Before and After hydrological restoration of 25 hectares of abandoned shrimp ponds in Tiwoho, North Sulawesi. This work was carried out with hand tools and community labor. Mid term corrections are required to repair 3-4 hectares of the area. The majority of the area,however, is successfully restored.

3.2 Bengkalis Island, Riau Province In 2003, MAP-Indonesia worked with local NGO Yayasan Laksana Samudera to create a GIS Atlas of the condition of mangroves in the Bengkalis Islands. The atlas revealed that 600 hectares of mangroves were being cut per year to fire a charcoal industry. The charcoal concessions were permitted, but often cut mangroves beyond their boundaries, both in terms of space and time. Replanting was a requirement of the permit; and although mangroves regrew, generally, after initial logging, in the 1990 s logging practiced had changed (stumps and rootballs previously left alone were now dug up and carbonized) and the forest remained bare after logging. The excavation of the root ball led to the pock-marking of the landscape, with deep waterlogged holes scattered amidst high mounds created by crabs and the mud lobster Thalassina. This irregular substrate no longer supported growth of mangrove trees (Rhizophora, Brugueira and ylocarpus), but rather Acrostichum ferns on the high mounds, and no vegetation at all in the holes. The construction of a 30 kilometer long dike wall worsened the situation. Communities built the wall to separate the mangrove area from their coconut plantations, which were suffering from salinization after the clearing of the mangroves (which had acted in a buffering capacity). The dike wall, however, excluded the flow of fresh surface water into the mangroves area, maintaining the brackish nature of the water. Increasingly salty water in the mangroves, eventually permeated the long dike wall through numerous crab holes and micropores, killing many coconut trees. Direct planting of disturbed areas, by government and concerned communities ensured but was not successful. Things changed in 2004-5, as community interest in the mangrove area peaked, catalyzed by a page in the GIS Atlas depicting areas for potential community stewardship. These stewardship areas (totaling 300 hectares) were formalized due to support from the ADB Co-fish program and Yayasan Lakasana Samudera. One by one, these community groups have been attempting EMR (to date four groups have practiced the methods). They reconnected fresh water flows into the mangroves, razed the Acrostichum mounds, filled in pot-holes, and dedicated areas for both natural regeneration and planting. Six years later after initial EMR, a pair of community groups have already thinned their plots by selective logging, leaving behind 12 meter tall trees for future thinning. Groups are also engaged in a variety of livelihood projects, such as clam fattening in the mangroves and virgin coconut oil making, which augment proceeds from selective cutting.

Fig 5. Eight of ten community stewardship groups totaling 300 hectares on Bengkalis Island, Riau 3.3 Simeulue Island, Aceh Province Not many people had heard of this island before December 24, 2004, which gained fame due to its long-standing traditional information system announcing the arrival of a tsunami known as smong. Because of this early warning system, the tsunami claimed only three victims even at it s origin. The islanders had also taken good care of their island resources, and mangroves had only been significantly logged and degraded near the small port town of Sinabang, fueling a brick-making industry. Although communities proved resilient against the tsunami, the island s mangroves fared less well. Initially, mangroves on the Eastern edge of the island subsided along with the island, dying within weeks due to water logging. After the subsequent March 2005 mega-earthquake, the island experienced cumulative seismic uplift. Mangroves along the NE and NW coasts of the island were lifted entirely out of the influence of the tidal zone, and subsequently died of desiccation or competition with terrestrial vegetation. Propagule limitation in the absence of

mother trees, has resulted in very lownatural recruitment of newly uplifted substrate even where the substrate is at the appropriate tidal elevation. Nonetheless, this area was planted with seedlings from North Sumatera after the tsunami but large scale failure was recorded, due to a variety of reasons (seedling condition, planting method, planting area, damage due to water buffalo, species selection). The story differs slightly in Teluk Dalam, North Central Simeulue, where seismic uplift (25-50 cm) measured less than the tidal range (78cm) and many adult mangroves survived, able to disperse their propagules to colonize the newly uplifted intertidal zone. Natural recruitment has been recorded averaging 2856 seedlings per hectare growing well and exhibiting the full range of original biodiversity. Although communities planted mangroves as part of post-tsunami activities, they now see that planting is unnecessary and are interested in petitioning sub-district and district government to declare the bay of Teluk Dalam a community based mangrove reserve for the conservation and sustainable use of mangrove resources.

Communities from Eastern Simeulue, having been trained in the methods of EMR, are now interested in human assisted propagule distribution, to promote re-colonization of their coast by mangroves, for fisheries and other benefits. All the while, communities both in Teluk Dalam and Eastern Simeulue are engaged in sustainable livelihood development, from mangrove and other available coastal resources. 3.4 Jaring Halus, NE Langkat Wildlife Sanctuary, North Sumatera Province The 9000 hectare NE Langkat Wildlife Sanctuary is situated at the Mouth of the Wampu river, which flows north out of the mountains separating North Sumatera from Aceh. It is predominantly natural mangrove habitat, yet has undergone severe illegal logging over the past 25 years, largely to feed the charcoal industry. The entire area was patrolled by only 3 staff from the BKSDA (Agency for Conservation of Natural Resources) a division of the National Forestry Department, with minimal patrolling budgets, and little experience in engaging local communities. The village of Jaring Halus is a 33 hectare stilt village, home to 3300 ethnic Malay fisherfolk. The village makes its entire living from mangrove and estuarine resources, predominantly processing fish catches from neighboring isolated villages for shipment upriver to the commercial center of Medan. Historically, the village controlled 2000 hectares of mangrove resources, however with the creation of the National Wildlife Sanctuary, their access to these resources was diminished to a 40 hectare forest adjacent to their village. This forest has been protected since antiquity under strict adat rules. Cutting is allowed, for timber for housing or pilings, as well as fuelwood for male circumcision ceremonies, however only with permission by the traditional council and appropriate replanting. The 40 hectare forest maintains a high biodiversity of 26 species of mangroves of enormous height and girth. Only a hundred meters across a tidal channel, the wildlife sanctuary sports small trees and large areas of clear cuts. The government had enjoined the community to plant mangroves in the sanctuary in 1999, but as the seedlings grew, they were cut again for charcoal making at only 6-8 years of age. Asked to plant mangroves again in 2006, the community declined, suspicious that some of the BKSDA agents were involved in the illegal logging. Through the USAID sponsored Environmental Service Project, mangrove rehabilitation was proposed again to the community, who again initially declined. Instead, a 10 hectare shrimp pond complex in the landward edge of the

Wildlife Sanctuary was restored using EMR techniques of strategic dike wall breaching. Mangroves grew back quickly. In the meantime, community organizers facilitated a group from Jaring Halus to make a presenteation at the office of BKSDA, petitioning for restoration of access and control over a wider area of mangroves. Within the year, BKSDA and the Director General of Forestry had granted the community 500 hectares of collaborative management area, the first case of its kind in a nationally protected mangrove forest in Indonesia. The community then initiated enhancement planting of the 500 hectares, focusing around four major tidal creeks, each managed by a community group from the village. Poorer members of the village were given crab trapping equipment, to increase community vigilance over the area. There has, however, been a degree of back-sliding. With the USAID program over, the community has been left on its own to deal with an agency whose entire staff has been changed out. In addition, 2000 hectares of the sanctuary towards the mainland was reportedly filled for the development of oil palm plantation. In this case, continued partnership is needed, to ensure the community of Jaring Halus long term access to and control over mangrove resources. 3.5 Tanakeke Island, South Sulawesi Province Tanakeke Island is situated 20 km off the SW coast of South Sulawesi in the Takalar District. The mangrove forest is characterized as an overwash system (Whitten, 1985). LandSat photos from 1976 reveal a mosaic containing 1776 hectares of mangroves, primarily low and midmangal, with relatively little upper mangrove due to small proportion of hinterland area on the island, yet a significant amount of sub-tidal lagoon interspersed within the atoll. Recent satellite images depict a highly altered topography. In the 1990 s, the transmigration department cleared 400 hectares of mangroves for aquaculture development, while local landowners, both individually and in partnership with investors converted an additional 800 ha. By the mid 90 s 1200 hectares of aquaculture ponds (milkfish and prawn polycultures) had been developed, but by 2005 nearly all community ponds were disused. The condition of mangroves on the island before aquaculture development was already degraded due to shifting clear-cutting of young trees and low growth potential. Due to lack of fresh water and poor edaphic conditions (little organic matter with a calcareous base - average depth 15cm under organic layer, high interstitial salinities), biomass accumulation is low.

Growth figures of Tanakeke s Rhizophora apiculata match up with the lowest of five algorithms of a low-growth site determined by JICA for R. apiculata in Java and Bali. (Table 1 and Fig 7). Table 1 Low Growth Site Yield Prediction Table for Rhizophora apiculata (Site L-III) (Summary Table) JICA 1999 Age 10 15 20 25 30 35 Mean Height (m) 3.84 7.10 9.63 11.23 12.12 12.59 Number (trees/ha) 6723 5565 5063 4828 2715 4659 Volume (m3/ha) 7.94 40.33 88.57 130.17 157.45 173.07 Effective Volume 4.76 24.24 53.14 78.10 94.47 103.84 (m3/ha) Fig 7. Site Index Curve, Low Growth Site (Bali Java) There is little firewood on the island s small, dry terrestrial area and transport to mainland South Sulawesi is costly and dangerous in certain seasons. This means that mangrove wood is the major fuelwood source for the 23 island communities. Mangrove areas called borong measure 20m x 20m are individually owned by the villagers. Subsistence fuelwood is gathered by felling trees within a borong. Small-scale commercial logging, for production of charcoal, sale of fuelwood to the mainland, or use in construction of piers, fences, fishing equipment and seaweed mariculture infrastructure is undertaken by clear-felling entire borong, which may be

replanted at a density if 5 trees per every adult tree cut. The age of felling has also decreased from 12-25 year old trees, to 6-8 year old trees. Over time mangroves become overdense, and biomass addition is low due to competition. The low price of mangrove timber (500 rp/meter for 8 year old poles and 25,000 60,000rp for 60kg of charcoal) means that large areas of mangroves need to be cut for a family to make a living out of timber product sale. Islanders prefer Rhizophora spp., wood due to its high caloric value, and thus R. stylosa and R. apiculata have been anthropogenically selected for. Mangroves occurring on the coast, such as Avicennia marina and Sonneratia alba, have been cleared, and replanted with Rhizophora spp., however, lack of success in many areas have reduced the extent of coastal mangroves. In essence, Tanakeke has lost its lower-most mangal. Periodic attempts to replant coastal mangroves by the forestry department and communities are reported to have failed several times. A high diversity of species in the mid mangrove (Brugueira gymnorhizza, A. marina, A. alba, Ceriops tagal, S. alba and S. ovata), have also been cut and replaced with Rhizophora spp, reducing the natural resilience of the remaining forest. Back mangrove species, such as Lumnitzera racemosa and Pemphis acidula occur in the infrequent hinterland fringe, and on the more frequent dike walls which now dote the island. ylocarpus moluccensis and Heritiera littoralis are rare but also occur. Medium-Scale EMR on Tanakeke As discussed above, traditional management calls for replanting of 5 propagules for every adult tree clear-felled. This practice, however, was not applied in disused shrimp ponds, as pond owners maintained hopes to accumulate enough capital from their main occupation (seaweed mariculture of carrageen species), to re-invest in aquaculture. By and large, fish farmers were at a disadvantage with mainland practitioners, due to the added cost of transportation for purchase of external inputs (feed, piscicides and urea) as well as sale of cultured fish and shrimp. Thus, as villagers farmed seaweed, gleaned crabs and prawns in remnant mangroves, and fished on their bombed out coral reefs (with occasional trips to open ocean, primarily engaged in illegal sea-turtle fishery), the disused ponds of poorer communities members degraded for periods of up to 6-8 years, while richer community members maintained their dike walls and their hopes of rejuvenating small aquaculture enterprises.

Assessment teams from Restoring Coastal Livelihoods project, showed up in 2010, and noted the condition of 6-8 year old chronoseres of dilapidated ponds. The findings were encouraging to the team. R. stylosa and R. apiculata had colonized much of the degraded pond area, with occasional additional species (S. alba, A. marina, B. gymnorhizza, C. tagal). This indicated that, if communities could be convinced to free up disused ponds for rehabilitation, simply removing the dike walls would likely result in successful natural revegetation. From 2010-2012 the RCL team and local communities went to work, engaged in assessments, appraisals, EMR trainings, community awareness raising, study tours, and finally implementation and monitoring of Ecological Mangrove Rehabilitation. The work began in 45 hectares associated with one sub-village (dusun Lantang Peo) and spread to cover 228 ha in 2 years. Three additional communities have pledged land for the project to total 400 hectares by the end of 3 years (of 800 ha of community owned ponds on the island). Success, however, is not yet evident. During the first attempt at strategic dike wall breaching in Lantang Peo, overzealous practitioners created 212, 5-meter wide breaches in dike walls. This high number of breaches resulted in relatively low tidal prisms at each breach, reducing the scouring effect desired during tidal creek formation, to keep the creek open (free of sediment) and facilitate flooding and drainage. Pond interiors were still inadequately drained. Also, R. apiculata and R. stylosa, although abundant nearby, are not renowned pioneer species, in comparison with Avicennia spp, and Sonneratia spp. (small diaspore, able to colonize various substrate elevations and types, quick to establish roots and leaf-out, etc.). Although chronoseres evidence successful natural growth of Rhizophora at 6-8 years, it is unclear how long the colonization process took. To speed up the process of colonization, in year two, midcourse corrections were made at Lantang Peo, which included hand-digging a tidal creek to facilitate flooding and drainage, and human assisted propagule distribution, focusing on pioneering species collected at reference forests in South and North Sulawesi. 4.0Results 4.1 Ecological Results In Tiwoho, diversity was highest of all restoration sites, with 27 species of true mangroves recruited, established and growing out of a total of 30 species noted for Bunaken National Park (Djamalludin, 1996). The highest area of diversity for the site occurs in the hinterland margin. Table 2 records all mangrove species noted during surveys in 2007 and 2011.

Family Species Present in Rehab Site Acanthaceae: Avicenniaceae: Bombacaceae: Combretaceae: Euphorbiaceae: Meliaceae: Myrcinaceae: Palmae: Pteridaceae: Rhizophoraceae: Rubiaceae: Sonneratiaceae: Acanthus ilicifolius Avicennia alba A. rumphiana A. marina A. officinalis Camptostemon philippensis Lumnitzera littorea L. racemosa Excoecaria agallocha ylocarpus granatum. mekongensis Aegiceras corniculatum Nypa fruticans Acrosticum aureum A. speciosum Bruguiera cylindrica B. gymnorrhiza B. parviflora B. sexangula Ceriops decandra C. tagal Kandelia candel Rhizophora apiculata R. mucronata R. stylosa Scyphiphora hydrophyllacea Sonneratia alba S. caseolaris S. ovata Hertiera littoralis x Local Name Gahana, sarimunte Api-api Api-api Api-api Api-api Kayu pelompong Lolang bajo Lolang bajo putih Buta-buta Kira-kira Kira-kira Anting-anting Bobo Paku-pece Paku-pece Ting putih Makurung laut Makurung Makurung darat Ting papua Ting biasa Bido-bido Lolaro merah Lolaro putih Lolaro putih Lemong pece Posi-posi Posi-posi Posi-posi Kolot kambing Sterculiaceae: Total Species = 30 Species Present = 27 Table 2: Natural recruitment versus total potential species at the Tiwoho restoration site, Bunaken National Marine Park. The next highest site, in terms of diversity are the natural recovery sites in Teluk Dalam, northcentral Simeulue, specifically near the lighthouse known as Mercusuar, Ujung Ranup. 13 plots measuring 20m x 5m were monitored, with 39 subplots of 5m x 2m surveyed. This site has experienced natural recruitment and healthy growth of 13 species of true mangroves (see Figure 8) with an average density of 2856, or 408 seedlings being established per year since final disturbance (the 2005 Nias Earthquake). Sites close farther from sea in Teluk Dalam were regenerating at lower rates, of 125-318 seedlings per year. In Eastern Simeulue (Teluk Sinabang), recruitment rates were lower, with average recruitment in two study areas at 96 and 196 seedlings ha/yr recruiting onto the newly uplifted inter-tidal in areas adjacent to disturbed

mangrove forests. Average densities of around 1000 seedlings per hectare were noted here, but several sub-sites had much lower densities of 200, 275, and 475 seedlings per hectare after 7 years, indicating low recruitment rates of only 29-69 plants per hectare per year. Due to the apparent high recruitment rate in Teluk Dalam, community leaders want to forego additional, and likely unnecessary mangrove restoration, to begin talks leading towards areal management for conservation and sustainable use. In Bengkalis Island, Riau, data for a typical group after EMR can be seen at the Belukap site. Within three years, planted Rhizophora averaged 2.6 meters tall with 80% survivorship and a density of 3208 seedlings per hectare. Average natural recruits of 8 species were 3.4 meters tall with a density of 4210 seedlings per hectare. Anecdotal evidence shows that 6-8 years after EMR, selective logging of Rhizophopra poles has taken place twice in a pair of sites. In the NE Langkat Wildlife Sanctuary, the ten hectare shrimp pond complex experienced exceptional growth after dike wall breaching. Four planted species grew well, with R. mucronata ranging from 1.3 2.5 meters of growth 24 months after breaching year exhibiting 54-89 leaves per tree. More impressive was the growth of hundreds of Avicennia marina trees which had inhabited the pond in dwarf form before breaching (<1m) and grew to between 6 and 10 meters in 24 months (Fig. 8). Similarly over 85% survivorship and good growth was reported 24 months after enhancement planting of 12 species in previously logged areas of the 500 hectare collaborative management zone. On Tanakeke Island, low natural recruitment took place in year one. This was augmented in year two by hydrological corrections as well as human assisted propagule distribution; utilizing 4 colonizing species (A. alba, A. marina, S. alba, and S. Figure 8: 1 2 year old planted R. mucronata (foreground) and natural growth of pre existing stunted A. marina (background) Jaring Halus, North Sumatra ovata) and 6 additional species (R. apiculata, R. stylosa, B. paviflora, B. sexangula, B. gymnorhizza, C. tagal, C. decandra,. molucennsis). Monitoring is taking place quarterly in

year one, and annually thereafter for a period of 3-5 years. General ecological results are summarized along with social and economic outcomes in table 3 below. 350 300 250 200 150 100 50 0 Tree Sapling Seedling R. apiculata R. mucronata C. tagal B. gymnorrhiza L. littoria S. alba P. acidula. sps Non Man Figure 9 Representation of Natural Recruitment Data from Mercu Suar Reference Site - Teluk Dalam - Simeulue 4.2 The Development of Formal and Informal Community Forest Governance and Adaptive Collaborative Management The communities above have similar stories. Their mangrove forests, which provided important goods and services to the community as a whole, were disturbed, either by privatizing interest for short-term economic use, or natural causes. Mangrove rehabilitation, at some point, was attempted, in all cases with intervention from outside groups. However, oversimplified techniques led to initial failure. Techniques were then improved, and mangroves recovery became evident in most cases, either directly (successful rehabilitation), or indirectly (community awareness of successful natural revegetation in nearby chronoseres) with resulting confidence by communities in their own abilities to restore their mangrove resources. In all cases, communities were also involved in livelihood and environmental education programs, which, regardless of the degree of economic success, have effectively developed critical thinking skills amongst a large enough population of the community to allow for the informal and formal development of effective mangrove management regulations, and the ability to engage local stakeholder, including government, in increasing access to and control over both rehabilitated mangrove resources as well as adjacent, undisturbed forests. Table 3 looks at social, economic and ecological outcomes after EMR interventions in all five sites.

Table 3: Summary of outcomes and status related to key components of resilience. Site Tiwoho, N.Sulawesi Communities, Social Outcomes Direct Participants, Total Community Members 1, 420, 1004 Excellent non-formal conservation of mangrove area, small community groups for conservation, formal regulations on mangrove protection drafted but not pushed towards ratification Equitable male and female access and control over mangrove resources. Economic Outcomes Many trial alternative livelihood ventures, 2 lasting ones bamboo treatment and furniture making, and improved coosktove production, Very significant subsistence fishery from mangrove area especially accessed by poorer villagers Ecological Outcomes Excellent rehab of 22 out of 25 hectares of pond area 3 hectares require hydrological adjustment High biodiversity of natural recruits - 27 of 30 original species Excellent growth Shellfish and crab population recovery according to community Bengkalis, Riau Jaring Halus, N.Sumatra Zero timber harvest in 9 years 5, 300, 1200 10 stewardship groups of 30 members with formal protection of 300 hectares of area only 4 groups active Family clans manage mangroves for sustainable timber mostly male Island wide support of mangrove protection and sustainable use cessation of mangrove charcoal production Strong women s involvement in fisheries protection, enagement in conflict over fisheries 1, 165, 3300 Consistent long term traditional protection and sustainable utilization of pristine forest behind village (40 ha) 10 year collaborative management agreement with National government (BKSDA) for conservation and sustainable use of 500 ha of wildlife Poles in young rehab areas already thinned for economic use, Clam fattening pens developed by community in mangroves Coconut plantation recovering behind rehabilitated mangrove Significant subsistence gathering by family clans from creeks and mangrove area Poor community members access crabs and fisheries products regularly in mangroves Small enterprise development from mangrove non-timber forest products stagnant Fishery value from mangrove area extremely Approaching normal diversity but still anthropogenic preference for Rhizophora and ylocarpus dominated areas. Good growth of both planted and naturally recruiting mangroves after EMR Planted areas too dense, require thinning. Low biomass per tree in some areas due to pruning of branches for straight timber 40 hectare village forest still in excellent condition high biodiversity 500 ha area recovering much less incidence of illegal logging few bald patches left, high growth. Excellent regrowth in rehabilitated shrimp pond - >5000 trees/hectare 7

Simeulue, Aceh* sanctuary youth groups, fishing associations and female economic groups protecting mangroves lack of continued NGO collaboration 6, 300, 3000 Low incidence of community mangrove damage Awareness of mangrove value high for fisheries value and village protection Strong community push for mangrove conservation in remote areas Prone to mangrove conversion near main economic center (Sinabang) important to local livelihoods, high value Little direct economic benefit from mangroves and mangrove areas. Indirect benefit from offshore fisheries High perceived value for protective function mixed species ion 10 hectares 2000 ha of reported damage to wildlife sanctuary conversion to oil palm (unsubstantiated) Excellent natural regeneration in Teluk Dalam full original biodiversity normal growth more than 4000 seedlings/ha Poor natural regeneration in Eastern Simeulue (and presumably Western) propagule limited. Some incidence of hydrological disturbance but natural will self-repair. Tanakeke, S.Sulawesi* High degree of community resilience to disaster 8, 800, 4000 Lack of memory of healthy, diverse mangrove forest Strong understanding of protective function of mangroves Weak understanding of relation between frequent clear cutting, low biomass, and low overall value of mangrove system Poor and marginal community lack of basic services (water, electricity, education, health) Strong support for EMR in disused shrimp ponds and participatory processes willing to demand participatory Important and valuable crab fishery (currently being studied) Important subsistence prawn fishery Low-value timber products (fuel wood and charcoal) due to lack of processing small diameter of poles at harvest High perceived protective function No non-timber forest product development yet Dominant ecotone on the island, along with sub-tidal seagrass beds (currently supporting large scale seaweed mariculture) Impact on seagrass and fishery uncertain Low growth due to biophysical condition (low organic content, CaCO3 substrate, lack of fresh water input) coupled with anthropogenic pressure for young Rhizophora timber Extreme conversion of mangroves to aquaculture ponds (from 1776 ha mangroves to 500 ha mangroves or 72% loss) but little permanent hydrological damage due to largely unexcavated ponds and low, non compact dike walls.