Abstract
Shoreline armouring replaces natural shorelines with artificial shorelines such as vertical seawalls and boulder ripraps, forming sea defence structures. They are generally used for land reclamation or revetment, as breakwaters, or for their practicality for ports and docks. The construction of shoreline armouring is expected to increase exponentially with coastal city populations and as an adaptational response to sea level rise brought by climate change. This replacement of natural shorelines with artificial seawalls leads to a multitude of negative impacts on the intertidal ecosystem.The quantity and quality of habitat provided on artificial shorelines is dependent on the complexity of the substrate and therefore differ between vertical seawalls and ripraps. The smooth, highly homogenous surface of vertical seawalls, often built with quarried rock and/or concrete blocks, provides no refuge from heat and desiccation stress during low tide or predation during high tide. The lack of available suitable habitat (such as crevices or rock pools) leads to extreme competition between resilient species for any habitable space. These issues are also seen in ripraps of quarried rocks or concrete, the surfaces of which also lack crevices and water retaining pools. Although intertidal epibiota can take refuge underneath/behind these boulders, the exposed surfaces lack appropriate habitat and water retention. Lower alpha (α) diversity of marine organisms has, therefore, been found on artificial shorelines globally.
This reduction in biodiversity is further exacerbated when shoreline armouring is scaled up to a landscape-scale, for example, in Victoria Harbour of Hong Kong, with 95% artificial shoreline. Homogenisation of the shoreline habitat is then reflected in a reduction in within-site beta (β) diversity. Furthermore, in areas with high proportions of artificial shoreline, the negative impacts of artificial shorelines are often combined with other associated anthropogenic impacts of cities and ports, such as storm drain run off, sedimentation, and pollution. Considering the goals of the United Nations Decade on Ecosystem Restoration (2020-2030) to prevent, halt and reverse the degradation of ecosystems worldwide, rehabilitation of artificial shorelines into biodiverse, functional ecosystems has become a pressing issue.
Ecological engineering (eco-engineering) is a relatively new approach to environmental rehabilitation that focuses on creating habitats within artificial seawalls to compensate for the habitat loss. Eco-engineering has been categorised into hard (prefabricated artificial structures such as complex panels or rock pools), soft (addition of natural habitats such as oyster shells or mangroves) and hybrid (a combination of natural and artificial components) techniques. Retrofitting suitably complex habitats with crevices, holes, shading and water retention to vertical seawalls and ripraps was demonstrated to be successful in enhancing marine biodiversity in different parts of the world, but such trials are generally constrained to temperate regions and shorter time scales. Most documented experimental studies were also restricted by fabrication constraints to create simple designs, and by permitting and funding that limit large-scale trials over longer time scales.
The biodiversity and assemblage of a habitat is known to be affected by the habitat variability and complexity. However, the combination of multiple microhabitat types within a single eco-engineered feature is rarely seen due to difficulties with fabrication. In Chapter 2, two types of eco-engineered panels were retrofitted to a vertical seawall in highly developed and nutrient-rich Tsuen Wan Typhoon Shelter, Hong Kong. One internationally available panel (Imported Commercially Available Design - ICAD) had a repetitive pattern of short extrusions. The other was a locally designed, double-sided panel with deep triangular intrusions creating a varying, heterogenous pattern and providing shade on the reverse side - the Hong Kong Design (HKD). Taxonomic richness and relative abundance on the treatment panels were compared to flat, concrete panels and scraped seawall plots (n = 4) for 18 months. At the conclusion, the two closest unmanipulated seawalls were compared to the trial site to reveal within-site β diversity. The more heterogenous, double-sided HKD was colonised by the highest epibiotic taxon richness, cover of bivalves, and species richness within key functional groups. At the habitat-scale level (tens of metres), the HKD and ICAD contributed to an increased within-site β diversity and ecosystem functioning compared to nearby vertical seawalls. The addition of the more heterogenous HKD had an overall positive effect. These findings indicate that increasing topographic complexity and incorporating multiple types of microhabitats will be more effective in promoting intertidal marine biodiversity than a single microhabitat type.
Eco-engineering implementation in ripraps often includes units with a large, water retaining pool. However, such designs were often tested in environmental conditions sufficient for good colonisation and may not always be effective in different environmental conditions. In Chapter 3, four hard eco-engineered units and an oyster shell reef were deployed in a riprap in Tsuen Wan Typhoon Shelter, Hong Kong, for 18 months. Two units were internationally designed and commercially available, both with crevices, shading, and water retention. The other two units were designed specifically for the conditions of the site, combining multiple microhabitat types of different widths and depths, one with a deep, shaded pool. The oyster shell reef consisted of two layers of cured, locally acquired Magallana hongkongensis shells. After 18 months, the oyster shell reefs harboured significantly higher biodiversity and more species within key functional groups than comparative riprap boulders. Of the hard eco-engineering units, the site-specific unit with the deep, shaded rock pool attracted the highest species diversity. The internationally available unit with a wide, shallow pool was colonised by algal turf, due to the nutrient rich site conditions. At the experiment conclusion, the two nearest unmanipulated ripraps were compared against the trial site for epibiotic and fish community assemblage, and epibiotic β diversity. All four hard eco-engineered designs generally increased within-site β diversity and fish biodiversity compared to nearby unmanipulated ripraps. Suspension-feeding species and more species within key functional groups colonised eco-engineered units at both patch and site scales. The overall results show that tailored, site-specific eco-engineering has great potential to rehabilitate degraded ripraps into functional, novel ecosystems. The site conditions should, therefore, be considered when designing for or choosing prefabricated eco-engineered features, particularly for degraded and polluted sites.
Field experiments are generally restricted by permitting and funding, and therefore, most eco-engineering research currently focuses on testing a design concept with a small scale and a short timeframe, and without consideration of recovery from future disturbances. However, coastal ecosystems are experiencing major disturbance events at greater frequencies as climate change driven extreme weather and anthropogenic impacts increase. This study tested the hypothesis that the augmented epibiotic community on eco-engineered features would recover to higher biodiversity and more complex community composition than conventional seawalls. In Chapter 4, established intertidal communities on vertical seawall and riprap eco-engineering features/controls and natural rocky shores in three locations in Hong Kong that had varying levels of anthropogenic impacts were physically disturbed to initiate secondary succession and recovery. Secondary succession of the disturbed quadrats was monitored for 18 months, alongside nearby intact quadrats on the same eco-engineered feature or control/rocky shore plot. Species richness of the riprap features and natural rocky shore quickly recovered to pre-disturbance states, however, the recovery on vertical seawall features varied among different locations. Species richness on the traditional riprap/seawall controls either quickly recovered to a lower pre-disturbance richness or never recovered over the 18 months. Community succession was significantly different among the eco-engineered features and controls in ripraps and vertical seawalls in all three locations. In general, the eco-engineered features recovered to a higher species richness and a more complex community with higher cover of sessile taxa than the respective controls. The results showed that the eco-engineered features have a greater capacity for recovery to a more biodiverse community following a physical disturbance, even in locations with multiple anthropogenic stressors.
As a relatively new field of environmental rehabilitation, most eco-engineering studies are also restricted to shorter study periods of one to two years, such as those conducted for Chapters 2-4 in this thesis. Documented eco-engineering studies longer than two years remain rare in the literature. In Chapter 5, annual monitoring of vertical seawall and riprap eco-engineering sites and nearby rocky shores was conducted to study the temporal trend of marine intertidal biodiversity in three locations in Hong Kong for five years. After five years of monitoring, two artificial shorelines and one rocky shore for each eco-engineering site were also surveyed. Species richness on the eco-engineered features and controls within each site was significantly higher than controls across all vertical seawalls and ripraps, with more species accumulated over the five years and more complex epibiotic communities. Within-site β diversity was higher on the eco-engineering sites than unmanipulated, artificial sites, but eco-engineered vertical seawalls generally showed lower within-site β diversity than natural rocky shores. After five years, within site α and β diversity was often but not consistently higher in the eco-engineering sites than artificial and natural sites across the three locations. However, the epibiotic community composition showed that the eco-engineered features were providing habitat that was not available on the unmanipulated artificial shorelines, with riprap eco-engineered features colonised by epibiotic communities close to that of natural rocky shores.
Although this thesis presents a wide range of studies with many findings, there were multiple conclusions that are important for the rehabilitation of future artificial shorelines. In Chapters 2-3, a vertical seawall and a riprap sloping seawall within a highly developed and polluted area were successfully rehabilitated into novel, functional ecosystems. However, the biodiversity was dependent on the design of the eco-engineered feature, as both studies showed that highly heterogenous and site-specific designs were necessary for this nutrient-rich site. The effect of eco-engineering compared to controls was also extended to the recovery of an established epibiotic community from a physical disturbance (Chapter 4). The riprap eco-engineered features recovered to more biodiverse communities with higher cover of invertebrates and more complex assemblages than controls, regardless of experimental location within Hong Kong. Finally, five years of monitoring showed that eco-engineered features continued to provide important habitats with high α and β diversity (Chapter 5). These overall results show that eco-engineering at any site, including highly developed and degraded cities and ports, can have both immediate and long-term benefits for biodiversity of the epibiotic community. Eco-engineering can then support conservation strategies for sustainable cities through the rehabilitation of existing artificial shorelines, thereby supporting regional biodiversity and ecosystem functioning.
| Date of Award | 7 Aug 2025 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Mei Yee Kenneth LEUNG (Supervisor) |
Cite this
- Standard