nvestigating the influences of tidal inundation and surface elevation on the establishment and early development of mangroves for application in understanding mangrove rehabilitation techniques 1 2

Chapter – Introduction 1.1 Mangroves are highly threatened Mangroves provide a suite of highly valued ecosystem services (Barbier, 2007; Rönnbäck et al., 2007), which can be classified into supporting (Walters et al., 2008), provisioning (Rönnbäck, 1999; López-Hoffman et al., 2006; Rönnbäck et al., 2007), regulating (Gilman & Ellison, 2007; Bosire et al., 2008; Gilman et al., 2008; Kristensen et al., 2008; Donato et al., 2011; Rivera-Monroy et al., 2013) and cultural services (Millennium Ecosystem Assessment, 2005; Rist & Dahdouh-Guebas, 2006; Walters et al., 2008). Despite providing a wide range of ecological and socio- economic benefits, mangrove ecosystems are generally over-exploited, undervalued, and poorly managed, with only 6.9 % of total mangrove area protected under the existing protected areas network (IUCN protected areas categories I – VI) (Giri et al., 2011). Human exploitation and conversion of natural ecosystems is causing widespread ecosystem loss and degradation, with estimates of global mangrove losses ranging from 33 % (Alongi, 2002) to 50 % (Burke et al., 2001). Mangroves are additionally vulnerable to climate change such as rising sea levels that expose mangroves to longer and more frequent inundations which may surpass their natural inundation thresholds, resulting in the die-back of mangroves (Nicholls et al., 1999; Nicholls & Cazenave 2010). The drivers of mangrove degradation are diverse. They include deforestation (Macintosh et al., 2002; Gilman & Ellison, 2007), urban development (Ellison & Farnsworth, 1997; Lai et al., 2015), agriculture (Chen et al., 2007; Chen et al., 2009; Webb et al., 2014), commercial aquaculture (Hossain et al., 2001; Barbier & Cox, 2002; Barbier, 2006) and modification of hydrology (Kairo et al., 2001; Ren et al., 2009). Of these, development-related activities such as urbanisation, aquaculture and deforestation have been identified as the top three activities contributing to most of mangrove degradation (Dale et al., 2014). In Asia, deforestation and aquaculture activities were responsible for huge expanses of derelict land. For example, Philippines lost 67% of their mangroves between 1951 and 1987, of which conversion of mangroves to aquaculture ponds accounted for approximately half of the loss (Primavera, 1993). Additionally, the lifespan of most intensive shrimp ponds are restricted to – 10 years and end up abandoned due to self-pollution and disease problems (Flaherty & Karnjanakesorn, 1995; Stevenson, 1997). 1.2 Past attempts at mangrove rehabilitation Mangroves have the potential to recover from degradation to produce self-sustaining ecosystems resilient to normal periodic stresses (Borja et al., 2010). Through secondary succession, degraded ecosystems can be rehabilitated to the pre-existing functional condition. Mangrove rehabilitation may: (i) occur naturally through secondary succession with no human intervention; or (ii) be engineered through human-aided secondary succession (SER, 2004; Elliott et al., 2007; Simenstad et al., 2006; Stein & Cadien, 2009). As some mangroves have been degraded to the extent where natural recovery is no longer possible, there is an increasing need to catalyse the recovery of mangroves through human-mediated mangrove rehabilitation projects (Kaly & Jones, 1998). 1.2.1 Defining mangrove rehabilitation In scientific literature, the terms “restoration” and “rehabilitation” of mangrove ecosystems have frequently been used interchangeably. Some papers adopt broad definitions (i.e. Macintosh et al., 2002; Biswas et al., 2009; Kamali & Hashim, 2011; Ye et al., 2013) whereas others adopt narrower definitions where restoration is reported but the term rehabilitation is not mentioned (i.e. Hoang Tri et al., 1998; Chen et al., 2009; Ren et al., 2011; Chen et al., 2012; Rovai et al., 2012). Primarily, this study will use “rehabilitation”, defined as the act of partially or fully replacing the structural or functional characteristics of an ecosystem that have been diminished or lost, or the substitution of alternative qualities or characteristics than those originally present such that they have more social, economic or ecological value than existed in the disturbed or degraded state (Field, 1999; Ellison, 2000; Chen et al., 2007). The term “functional” refers to ability of rehabilitated mangroves to stabilise shorelines, trap sediments, improve coastal protection, offer suitable animal habitat, provide timber, and heighten the aesthetic value of coastal areas, in a way comparable to natural mangroves (Bosire et al., 2008). Any inevitable use of the term “restoration” would also point to “rehabilitation” as it is almost impossible to “restore” and return any ecosystem back to its original condition (Field, 1999). 1.2.2 A lack of mangrove rehabilitation success The lack of rehabilitation success has been widely reported (Field, 1999; Elster, 2000; Bosire et al., 2008; Primavera & Esteban, 2008) and attributed to the lack of knowledge and experience. For instance, most rehabilitation efforts in Southeast Asia have followed a trial and error method without any explicit framework, baseline ecological information, or proper consideration of community involvement (Biswas et al., 2007; Biswas et al., 2009). Some efforts meet with immediate failures while others fail several years after initiation (Ellison, 2000), with costs of a failed rehabilitation attempt ranging from a few thousand to millions of dollars. For instance, the outcome of two decades of immense mangrove rehabilitation efforts in Philippines has cost millions of dollars, with only 10 – 20% long-term survival rates (Primavera & Esteban, 2008). Topography governs mangrove distribution, with physical processes playing a dominant role in their formation and function (Kjerfve, 1990). Hence, knowledge of appropriate surface elevations, and its inherent control on inundation periods, are arguably the critical determinants of mangrove rehabilitation success (Stevenson et al., 1999; Lewis, 2005; Lewis, 2009; Gilman & Ellison, 2007; Friess et al., 2012). The relationship between surface elevation/inundation regime and mangrove establishment and physiological development has been studied widely, with the consensus that mangroves exhibit species-specific thresholds to inundation periods, and hence surface elevations (McKee, 1995; Kitaya et al., 2002; Chen et al., 2005; He et al., 2007; Chen et al., 2013). Yet, rehabilitation attempts commonly move immediately into manual planting of mangrove propagules and/or seedlings (Salmo et al., 2007) without first rehabilitating the physical environment (i.e. surface elevation) and processes that govern the distribution and maintenance of mangroves (Lewis, 2005; Primavera & Esteban, 2008; Samson & Rollon, 2008; Friess et al., 2012). Some projects further avoid dealing with long-drawn land tenure issues in areas where mangroves naturally establish by “rehabilitating” natural mudflats (Moberg & Rönnbäck, 2003). Species selection in particular, has failed to consider the biological inundation thresholds of planted species. This is most clearly seen in rehabilitation projects that often support the planting of commercially attractive but non-pioneer species such as Rhizophora species on low-elevation mudflats (i.e. the pioneer zone) where mangroves did not previously exist because of natural environmental constraints (Lewis, 2005; Samson & Rollin, 2008). A more appropriate species would be Avicennia species, which has been shown to be the most widespread genus of mangrove pioneer trees that colonise bare tidal flats of tropical regions around the world (A. marina in Africa – Osborne & Perjak, 1997; A. alba for Southeast Asia – Lee et al., 1996; Panapitukkul et al., 1998; A. germinans in South America – Proisy et al., 2009; Allemen et al., 2011; A. marina in Australia – Clarke, 1993). 1.3 Ecological Mangrove Rehabilitation (EMR) EMR is a rehabilitation approach that seeks to produce functional and self-sustaining ecosystems through the removal of barriers and/or stressors that impede natural recovery. The approach advocates identifying site-specific causes of ecosystem degradation before manipulating the structural and compositional components of the physical environment to optimise it for sustainable regeneration via natural recruitment and establishment (Lewis, 2005). Structural manipulation encompasses hydrological (i.e. re-establishing a suitable tidal regime) and substrate engineering (to achieve suitable elevation ranges). This are essential components which offer reference for planning mangrove rehabilitation as mangroves generally not establish below mean sea level (Stanley & Lewis, 2011) and generally exhibit species-specific tolerances to abiotic conditions such as inundation period and salinity. Compositional manipulation includes seeding and/or planting multiple species (Biswas et al., 2009). However, any planting should only be used as a last resort. The approach also advocates periodic post-rehabilitation monitoring and the inclusion of multiple stakeholders in order to achieve rehabilitation success and long-term management of a rehabilitated site. 1.4 Aims and objectives The aim of this thesis is to contribute to the understanding of hydrologic management in the success of mangrove rehabilitation projects. The focus is on how surface elevation and tidal inundation influences the establishment, survival and early development of mangroves. The broad question is examined through a large scale field study and further through a controlled mesocosm experiment. Specifically, the field study examines the influence of surface elevations on mangrove establishment and early development while the mesocosm experiment was designed to provide complementary knowledge through examining the influence of inundation durations on mangrove development and survival. Additionally, the mesocosm experiment also functioned to control for confounding factors in the field study that may affect observed field results. In this thesis, there are two specific objectives to be achieved: 1. Investigate how hydrologic restoration contributes to the successful establishment of mangrove vegetation in rehabilitated sites. In this study, the effects of two aspects of hydrologic restoration on mangrove rehabilitation were studied. The study site was located in Makassar, Sulawesi, Indonesia and comprises of abandoned aquaculture ponds that were formerly forested with mangroves. The study focused on (i) tidal inundation regime – by restoring a gradient of tidal regime in the abandoned ponds to study the effects of inundation and (ii) surface elevation – regrading the site to study the effects of differing surface elevations relative to sea level. The above two were carried out via strategic breaching of dike walls and regrading of selective areas to produce substrate at appropriate surface elevations for mangroves, relative to sea level. Quantification of rehabilitation success was achieved via vegetation surveys which mapped and quantified the extent of recolonisation of mangrove vegetation. 2. Quantify the inundation thresholds impacting survival and growth for two selected mangrove species – Rhizophora mucronata and Avicennia alba. There exist inter-specific variations on the thresholds of mangroves to inundation durations. Yet, this frequently goes unacknowledged given that inappropriate species are planted under sub-optimal conditions. Rhizophora propagules are favoured for planting in “rehabilitation projects” on low elevation mudflats, when a more appropriate choice would be the pioneer genus, Avicennia. As such, a controlled experiment was used in this study to test the effect of inundation (i.e. inundation threshold) on seedling survival and growth of these two species – R. mucronata and A. alba. The experimental design used a mesocosm set-up that exposed seedlings to increasing inundation durations. The objective was to observe if mangrove species exhibited natural inundation thresholds which when exceeded, had negative effects on survival and growth rates of seedlings. . 6 elevation and tidal inundation influences the establishment, survival and early development of mangroves. The broad question is examined through a large scale field study and further through. through examining the influence of inundation durations on mangrove development and survival. Additionally, the mesocosm experiment also functioned to control for confounding factors in the field. derelict land. For example, Philippines lost 67% of their mangroves between 19 51 and 19 87, of which conversion of mangroves to aquaculture ponds accounted for approximately half of the loss