Effects of sandfish (Holothuria scabra) removal on shallow-water sediments in Fiji

Introduction

Marine coastal ecosystems are among the most productive and diverse on earth (Poore & Wilson, 1993). Coastal communities are reliant on the various ecosystem functions and services these areas provide for their livelihoods (Conservation International, 2008). Recent studies have identified significant ecological impacts when benthic marine organisms were removed from these ecosystems (Lohrer, Thrush & Gibbs, 2004; Solan et al., 2004). A loss in bioturbation (the biogenic mixing of sediment) has been demonstrated following the reduction in abundance and diversity of marine benthic fauna (Solan et al., 2004). This loss is of particular concern as bioturbation has a substantial influence on the rate of organic matter (OM) decomposition and nutrient recycling (Meysman et al., 2006). Efficient processing of OM and nutrients in marine coastal ecosystems is crucial to the health and productivity of these ecosystems, and as such is vital to the coastal communities dependent on them.

The majority of OM in marine coastal ecosystems is trapped in permeable reef sands and degraded by the infaunal community, particularly high densities of sand-associated microbes (Wild et al., 2004). Due to the high surface area of permeable carbonate reef sands and advective flow into and within the sediment, these sandy sediments promote efficient degradation of OM (Wild et al., 2004; Wild, Laforsch & Huettel, 2006). Such sandy sediments have consequently been referred to as a biocatalytic filter system, promoting critical recycling processes (Wild, Tollrian & Huettel, 2004; Huettel, Wild & Gonelli, 2006). Benthic-pelagic coupling implies that changes in the OM concentration of overlying water are integrated and reflected in sediments, and furthermore implies that changes in sediment OM composition affect the overlying water quality (Wild, Tollrian & Huettel, 2004). Thus, the efficient functioning of this biocatalytic filter system provides coastal ecosystems—which are increasingly stressed by OM enrichment from anthropogenic inputs (Halpern et al., 2008; Rabalais et al., 2009)—with buffering capacity. Without the capacity of the system to buffer increasing OM loads, the health of coastal ecosystems could be compromised, resulting for example in shifts from hard coral- to algal-dominated reefs (Fabricius, 2005; D’Angelo & Wiedenmann, 2014).

Coral reefs are a critically important component of coastal ecosystems for the ecosystem services and goods they provide, such as fisheries, coastal protection, and nutrient cycling (Moberg & Folke, 1999). Inshore reefs and their associated ecosystems receive a considerable amount of OM from rivers and marine sources. The quantity of OM that reaches inshore reefs increases following heavy rainfall and flooding events (Briand et al., 2015), delivering a ‘pulse’ of OM that places further stress on the ecosystem. The efficiency of the system to absorb this stress and act as a biocatalytic filter is dependent on a number of factors including temperature, OM supply, light, water currents, and bioturbation (Kristensen, 2000). Bioturbation is particularly important for the efficiency of this filter system (Kristensen, 2000) as it increases the surface area of the sediment, breaks down geochemical gradients, and maximizes advective flow into the sediment (Thibodeaux & Boyle, 1987; Kristensen, 2000; Lohrer, Thrush & Gibbs, 2004; Meysman, Middelburg & Heip, 2006).

Deposit feeding sea cucumbers mainly of the order Holothuriida potentially play a significant role in enhancing OM mineralization through bioturbation as well as through feeding and excretion (Uthicke, 1999; Uthicke, 2001a; Uthicke, 2001b; Purcell, 2004; Wolkenhauer et al., 2010; MacTavish et al., 2012). Whilst sea cucumbers were once a ubiquitous component of many coastal marine benthic communities, growing demand from Asian markets for their dried form—known as bêche-de-mer—has driven their overexploitation, leading to population collapse and local extinctions throughout the Indo-Pacific (e.g., Toral-Granda, Lovatelli &, Vasconcellos, 2008; Friedman et al., 2009; Friedman et al., 2011; Purcell et al., 2013). Sandfish, Holothuria scabra (Jaeger, 1833), is one of the highest valued species in the bêche-de-mer trade (Purcell, 2014). Individuals of this species are easily accessible to fishers as they generally inhabit low energy environments behind fringing reefs or within protected bays and shores (Hamel et al., 2001), rendering the species particularly vulnerable to overexploitation. H. scabra exhibits a natural diurnal burying cycle and ingests a considerable amount of sediment during feeding, thus playing a key role in bioturbation (Mercier, Battaglene & Hamel, 1999; Purcell, 2004). Historically this species was found in densities of up to 2 individuals m⁻² (Hamel et al., 2001), implying that its bioturbation impact on inshore reef habitats is likely to have been substantial.

Inshore reef ecosystems are very dynamic environments in which conditions exhibit a large variation of temperature, salinity, turbidity, and wave energy, all of which influence sediment function and sea cucumber behavior (Mercier, Battaglene & Hamel, 1999; Mercier, Battaglene & Hamel, 2000). Initial studies investigating the ecological roles of sea cucumbers and sediment function have mainly been conducted ex situ or using mesocosms (e.g., Uthicke, 1999). Given the nature of inshore reef ecosystems, in situ studies are required to obtain more realistic results (see Wolkenhauer et al., 2010; Namukose et al., 2016).

Sedimentary O₂ consumption (SOC) and sediment O₂ penetration depth (OPD) provide a proxy for the function of sediment as a biocatalytic filter system. SOC reflects the respiration of the entire sedimentary community contained in a core and integrates OM deposited from the overlying water column (Nickell et al., 2003). O₂ penetration depth (OPD) determines the REDOX reactions occurring within the sediment, and thus the volume of sediment effectively participating in aerobic OM decomposition (Glud, 2008). Furthermore, sea cucumber feeding can alter grain size distribution through dissolution of calcium carbonate via acidity and potentially abrasion processes in their gut (Hammond, 1981; Schneider et al., 2011). As the composition of sediment grain size influences sediment O₂ dynamics (Urumović & Urumović, 2014), changes in grain size distribution caused by sea cucumbers may account for changes in SOC and OPD.

The current study aimed to understand how the removal of sea cucumbers affects sedimentary function through in situ experimental manipulations of H. scabra densities, thus providing information on the poorly understood ecological implications of sea cucumber fisheries. We hypothesized that high densities of H. scabra would facilitate the efficient degradation of OM. Thus, in areas with H. scabra, SOC rates were expected to decrease or remain relatively lower, and OPD was expected to increase compared to those areas devoid of H. scabra. Finally, sediment turnover—the amount of sediment ingested by H. scabra—was quantified as a proxy for the bioturbation potential of the animal.

Methods

Study site

The experiment was conducted between August 2015 and February 2016 on an extensive reef flat in front of Natuvu village, Wailevu East District, Vanua Levu, Fiji (16°44.940′S 179°9.280′E) (Fig. 1). The reef flat was part of the locally-managed marine area (LMMA) of the local villages. The entire reef flat covered an area of roughly 1.2 km², and contained a tabu area (periodically harvested closure). This tabu area supported a relatively high density of H. scabra compared to the wider region, which was closer to regional reference values (Pakoa et al., 2013). The remoteness of the study site and lack of electricity and infrastructure unfortunately prohibited reliable OM measurements.

A pilot survey of the site using 100 m × 2 m belt transects was conducted various times throughout the day and night, and during various tides in order to optimize estimates of the population density, and burying and feeding behavior of H. scabra throughout a 24 h cycle.

Results

As indicated above, primary sediment sampling was conducted during the first week of each month. Sampling began in September 2015 (T = 0) and continued each month until February 2016 (T = 5). As enclosures were checked daily and restocked if necessary to maintain stocking densities, the effect of loss of H. scabra between restocking was deemed minimal; overall, restocking was in the range of two to three animals in the high treatment enclosures and one to two animals in cages of the natural treatment. H. scabra would occasionally have to be removed from exclusion treatment enclosures.

Environmental parameters

Water temperature as quantified during initial SOC measurements showed a steady increase between September (T = 0; 23.8 °C ± 0.23 mean ± SE) and December 2015 (T = 3; 26.6 °C ± 0.19), followed by a severe increase between December 2015 and February 2016 (T = 5; 31.2 °C ± 0.07) (Fig. 2A). Heavy rain 10 days prior to sampling in November (T = 2; Fig. 2C) caused coastal flooding, and a tropical depression (Tropical Disturbance 05F; Fiji Meteorological Service, 2015) five days prior to sampling in January (T = 4; Fig. 2C) caused coastal flooding and storm surge at the site.

Discussion

The current study explored the impact of holothurian removal on sedimentary function in a unique in situ setting. Despite the considerable ‘background noise’ from inherent and uncontrollable factors such as waves, wind, currents and marine benthos, several parameters provided clear evidence that overharvesting of sea cucumbers has a potentially strong effect on the ability of sediment to function as a biocatalytic filter system. The estimated rates of sediment turnover observed here for Holothuria scabra alone are substantial, and are more than twice the turnover recorded for a mixed population of H. atra and Stichopus chlorontus (4,600 kg dry wt yr⁻¹ 1,000 m⁻²), which was estimated to be equivalent to turning over the upper 5 mm of sediment in that area annually (Uthicke, 1999).

Grain size selectivity has not been conclusively demonstrated for H. scabra (e.g., Wiedmeyer, 1982; Tsiresy, Pascal & Plotieau, 2011). However, within the current study, changes to grain size composition in areas of high H. scabra densities and the lack of any changes in their absence indicate that H. scabra are selective to certain grain sizes. This suggests that when sea cucumbers are present on inshore reef flats in high densities, they play a key role in the physical reworking and change of sediment structure in marine ecosystems. Based on the observed non-significant trends, a higher number of samples are suggested for future work. Several sea cucumber species are able to change sediment grain size through dissolution of calcium carbonate via acidity, and potentially abrasion processes, in their gut (Hammond, 1981; Schneider et al., 2011).

Throughout the study SOC rates and OPD exhibited a ‘buffered’ response in high treatment enclosures compared to exclusion enclosures, where they exhibited several large fluctuations. This buffered response was characterized by a predictable pattern with relatively low amplitude changes. Higher SOC rates in the exclusion treatment were consistent with findings by Nickell et al. (2003); these higher SOC rates were likely driven by the increased respiration of benthic fauna, microbially-mediated oxygenation of OM, and reduced inorganic metabolites (Kristensen, 2000; Nickell et al., 2003). Deviations from this trend occurred at T = 2 and T = 4, following coastal flooding ten days prior to T = 2, and storm surge and coastal flooding five days prior to T = 4. Differences between the T = 0 measurements and the strong increase in the uncaged control between T = 0 and T = 1, compared to an absence of changes in the natural treatment over the same time, could be attributed to sampling during different tides, which was rectified for later time-points. Sea cucumbers actively feed on OM, thus reducing its concentration within the sediment (e.g., Uthicke & Karez, 1999; Wolkenhauer et al., 2010). Thus, there should be a higher OM concentration in the sediment of the exclusion treatment compared with that of the high treatment. Relatively higher OM concentrations facilitate increased microbial abundance and activity (MacTavish et al., 2012), in line with the observed higher SOC rates in exclusion treatment enclosures. Similarly, coastal flooding (observed at the study site ten days prior to T = 2 and five days prior to T = 4) likely delivered a strong OM-‘pulse’ onto the reef flat (Briand et al., 2015), explaining the elevated SOC rates even in high treatment enclosures following flooding (Fig. 2C).

The buffering capacity provided by high sea cucumber densities is likely due to the effect of the animals’ feeding, excretion, and bioturbation. Feeding by sea cucumbers acts to ‘clean’ sediments, whereas the subsequent digestion and excretion of waste products plays an important role in nutrient recycling (Purcell et al., 2016). The recycled nutrients stimulate the growth of benthic microalgal communities (e.g., Uthicke & Klumpp, 1998; Uthicke, 2001a), which absorb nutrients and produce O₂ through photosynthesis (Stockenreiter et al., 2016). Bioturbation increases advective flow into and within the sediment directly through burying, burrowing, and bio-irrigation (Meysman, Middelburg & Heip, 2006), and indirectly as it increases bed sediment complexity (bioroughness). Bioroughness creates pressure gradients that can increase advective porewater flow up to seven-fold compared to smooth-bed flows (Thibodeaux & Boyle, 1987; Huettel & Gust, 1992), increasing the supply of degradable material and electron acceptors (such as O₂) to the sediment (Rusch et al., 2006), and thus promoting aerobic degradation. As aerobic degradation of OM in marine sediments is ca. ten times faster than anaerobic degradation (Kristensen, Ahmed & Devol, 1995; Hulthe, Hulth & Hall, 1998), sea cucumbers promote the efficient degradation of OM in a closely coupled benthic recycling system.

As grain size distribution showed no significant difference between high and exclusion treatment sediments at T = 5, simultaneous changes of OPD are assumed to be caused by the presence of sea cucumbers and not an alteration in sediment structure following storm surge prior to T = 4. Exceptionally high temperatures were recorded at T = 5 (31.2 °C ±0.07) compared with earlier months (T = 0 to T = 3; 25.74 °C ± 0.16) as a result of the 2015/2016 El Niño (Blunden, 2017). High water temperatures recorded at T = 5 coincided with the substantial reduction in sediment OPD of exclusion treatment enclosures. Warmer water temperatures induce microbial growth and respiration (Nydahl, Panigrahi & Wikner, 2013), and the relatively calm conditions for the same time period resulted in limited wave action and thus limited mixing of bed sediment and the overlying water column. Both these factors contribute to an increased SOC and a reduced OPD (Kristensen, 2000; Friedrich et al., 2014). The high, natural and uncaged control treatments did not experience a rapid reduction in OPD, despite being subject to the same environmental conditions. The abrupt and significant reduction in OPD during T = 5 in the exclusion treatment was likely caused by the combined manifestation of local and global stressors: the removal of sea cucumbers and the coinciding rapid increase in sea surface temperature.

Storm surge five days prior to T = 4 had visibly shifted and re-suspended a considerable amount of sediment at the study site. This disturbance would have functioned similarly to bioturbation (e.g., by oxygenating sediment and redistributing OM), explaining the significantly decreased SOC rate in exclusion enclosures at T = 4 (Fig. 2C). Disturbances such as storms, which are projected to increase in intensity as a consequence of climate change (Emanuel, 2013; IPCC, 2014), may thus be able to induce sufficient mixing of bed sediment capable of reducing SOC. However, the same models predict an increase in the severity, frequency, and duration of extreme precipitation events and heat waves (IPCC, 2014). While the global stressor of increased sea surface temperature in the current study had an effect on sediment function in all treatments, the high densities of sea cucumbers in the high treatment were able to buffer such extreme changes, particularly to OPD.

These results suggest a synergistic interaction between local and global stressors, whereby reductions in bioturbation caused by local overexploitation of a key benthic species (Solan et al., 2004) in combination with elevated temperatures led to an amplified effect. As mean sea surface temperatures, warm temperature extremes, and heavy precipitation events are forecast to increase and occur at higher frequency and intensity (IPCC, 2014), bioturbation becomes increasingly important to maintaining and improving sediment quality. Given the mechanisms of benthic-pelagic coupling (Wild, Tollrian & Huettel, 2004), the increased SOC rate and reduced OPD observed in areas void of H. scabra could contribute to deoxygenation of the overlying water column. Hypoxia affects the distribution, abundance, diversity and physiological state of benthic communities (Meyer-Reil & Köster, 2000), and increases the risk of hypoxia-derived mortality for larval stages of several marine organisms (Vaquer-Sunyer & Duarte, 2008).

High densities (350 g m⁻²) of the sea cucumber H. scabra seem able to enhance the buffering capacity of sediment to stressors such as increased OM load and anoxia. This finding may have particularly strong implications in shallow restricted bodies of water, which are prone to large temperature variability, such as that of Marovo Lagoon, Solomon Islands. In June 2011, the lagoon experienced a large-scale harmful algal bloom, and its subsequent senescence resulted in large-scale hypoxia-derived mortality (Albert, Dunbabin & Skinner, 2012). The authors of the report suggest this event was caused in part by an increase in the catchment nutrient pool, and reduced processing of sediment nutrients and oxygenation of sediment following 15 years of extensive overharvest of sea cucumbers, combined with prolonged warm and calm weather conditions (Albert et al., 2011). As the frequency and severity of record high global temperatures and El Niño events are projected to increase with climate change (Cai et al., 2014; Blunden, 2017), and coastal waters are becoming increasingly eutrophic (Fabricius, 2005), events such as that reported for Marovo Lagoon may become more frequent. As such, the role sea cucumbers play in the bioturbation of inshore sediment becomes increasingly essential to the buffering capacity of tropical coastal marine ecosystems.

Conclusion

This study builds upon previous work investigating the ecosystem role of sea cucumbers. Sandfish (H. scabra) are shown to play a demonstrable function in the role of sediment as a biocatalytic filter system. The combination of global stressors such as elevated sea surface temperatures and the local stressor of sea cucumber removal is of particular concern, as our results suggest potential synergisms resulting in an amplified effect. A consequence of the extensive overexploitation of sea cucumbers is that ecosystem functions and services offered by coastal marine environments in which these animals have been removed is likely compromised. These changes will leave coastal human communities increasingly vulnerable, particularly those of small Pacific islands that rely heavily upon such ecosystem services for their livelihoods and food security. We recommend that management should maintain moderate (in this study 60 g m⁻²) to high densities (350 g m⁻²) of sea cucumbers within reef ecosystems. Such densities should allow sea cucumbers to fulfil their role in maintaining sediment function at a scale that has a measurable impact.