Rocky intertidal communities support a high biodiversity. However, they are also easily affected by disturbances, especially human induced disturbances like marine pollution. The recovery process would be very long even after abatement actions have been taken. Also, the outcome of this recovery process cannot be readily predictable. Marine environments in Hong Kong are seriously exposed to pollution problems. The rocky intertidal areas along Victoria Harbour have been suffering from marine pollution impacts for a long time. Recently, the Hong Kong Government has implemented the Harbour Area Treatment Scheme (HATS) aimed at improving the water quality of Victoria Harbour. While water quality of this Harbour was reported to have improved after the scheme, whether the natural rocky shores along the Harbour have recovered from the pollution impact remain unclear. Responses of rocky intertidal community assemblages, as well as those of some particular intertidal species, i.e. the barnacle Tetraclitra japonica, in the light of HATS, were the main focuses of the present study. / From June 2004 to May 2007, the community assemblages of six selected rocky shores along the Victoria Harbour were monitored. From west to east, these six natural shores were Peng Chau, Green Island, Causeway Bay, Hoi Xum Park, Heng Fa Chuen and Tung Lung Chau. Results showed that Hoi Xum Park was completely different from the other five sites with the lowest diversity index . In contrast, the diversity indices of the other sites ranged from 2.13 0.37 to 2.25 0.21. During the study period, only 11 species were recorded in Hoi Sum Park. This was far lower than that in the other shores (number of species ranged from 32 to 38). In addition, the community structure of Hoi Xum Park was only 25% similar to that of the other sites. This high dissimilarity was caused by high coverage of the barnacle Balanus amphitrite and cyanobacteria Chroococcus sp., and the absence of the limpet Petalloida saccharina in Hoi Xum Park. Except for Hoi Xum Park, a decreasing similarity of the community assemblages was also observed from west to east (from about 85% to 70% similarity) along the Victoria Harbour, with Tung Lung Chau in the east being least similar with Peng Chau in the west. The community structural differences between sites were found to be related to the gradients of total phosphorus level along the six rocky shores, resulting in a shift in their algal compositions that, with a bottom up effect, eventually altered their whole community structure. Recovery has not taken place in most of the sampling sites after the cessation of pollution input into the Victoria Harbour. However, a weak sign of recovery was recorded in Tung Lung Chau. Temporal increase of relative abundance of the brown pseudo-perennialalgae Sargassum hemiphyllum by 10% and 20% decrease of grazer numbers were observed, suggesting that the water quality in this particular site was improving, and was better than those of the other sites. The location of this site and the direction of seasonal water currents in Victoria Harbour could have contributed to the sign of recovery observed. The physiological or ecological changes of some particular species of marine organisms may provide an indication of the water quality of its habitat. In order to test whether the locally common barnacle T. japonica is a good environmental indicator, its population dynamics and reproductive biology were examined. Starting from August 2005，a 14-month monitoring was conducted on samples collected from five of the six sites within the Victoria Harbour, as no T. japonica was found in Hoi Xum Park. On the other hand, samples from Hoi Ha Wan，a site away from Victoria Harbour, were also collected to serve as additional control. Oocytes of the samples collected were examined, their size and density were measured. Both the oocytes size and density were found to vary among sites. Samples from Green Island brooded the largest oocytes (70.50 12.49 μm) and with the highest density (76.08 33.30 eggs per mm2). The smallest mean egg diameter was aobserved in Hoi Ha Wan samples (57.57 18.86 μm) and the lowest mean oocyte density was observed in Peng Chau samples (64.20 38.63 eggs per mm2). There was a decreasing trend of egg diameter across the sampling sites from west to east of the Victoria Harbour. Such phenomenon was closely associated with the water quality variations (silica level) of the sampling sites，which may alter the plankton community structure and thus the effectiveness with which nutrients are transferred as food to T. japonica. This could result in the difference in the sizes of the eggs brooded by the barnacles among sites. Such relationship, however, was not found for oocyte density. The total number of eggs may thus be a more accurate way to reflect the real brooding situation of the barnacles. Oocytes were found to be brooded all year round, which was different from that found in previous studies. Reproductive development of oocytes as a temperature sensitive process maybe triggered by the elevated water temperature in recent years, due to global warming, resulting in prolonged reproductive period. The spawning period, however, remained limited in summer. / For the population dynamics study, the basal diameter and the population density of T. japonica were measured. Results showed that there were spatial variations in these two parameters examined. The mean basal diameter was the smallest for samples collected at Hoi Ha Wan (14.99 7.03 mm) and the largest at Causeway Bay (22.47 4.69 mm). In contrast, Hoi Ha Wan recorded the highest value of mean ( SD) barnacle density (605.11 428.75 individuals / m2) while the lowest density was found at Peng Chau (82.46 26.23 individuals / m2). Regression analysis showed that there was a significant relationship between the spatial monthly mean barnacle basal diameter and density, thus the densest population was found in Hoi Ha Wan with the smallest mean size. Some curvilinear or exponential relationships were detected between the population parameters of T. japonica and the water quality. The level of suspended solids, which maybe a food source of barnacles, was found to be associated with the spatial variations in growth and settlement of this species. Mass settlement event was recorded during early or mid summer. However, new settlers were also observed during mid winter at Hoi Ha Wan, Green Island, Causeway Bay and Tung Lung Chau when this species was not reproductively active. These recruits may have come from populations outside of Hong Kong with temporal variation in their reproductive output. / The present study examined the responses of the rocky intertidal communities along the Victoria Harbour after pollution abatement. Generally, spatial variations of these responses were more obvious than that of temporal variations, for both the community structural changes and the reproductive and recruitment dynamics of T. japonica, the indicator species. A longer duration of monitoring should be continued as recovery from pollution impact in nature is a long term process. More detailed studies on the population dynamics of T. japonica is suggested，which may provide a better insight of the responses of this species to environmental changes and the effectiveness of this species as an indicator of water quality. The knowledge obtained in the present study could provide the baseline information for future studies. A good development plan incorporating conservation strategies would certainly be a much better approach than post pollution actions, since alteration of the community structure maybe irreversible, and the original habitat could be lost permanently. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / 岩質潮澗帶群落有著很高的生物多樣性，然而，這些群落的結構很容易受到 騷擾而改變，特別是當受到人爲騒擾時，例如海洋污染。就算減污措施經已實行， 這些群落之復元往往需要很長的時間，後果也不能一槪如所預料。在香港，很多 海洋環境都受到污染，而位於維多利亞港的岩岸亦同樣受海洋污染的威脅有著相 當長的時間。直至近年，香港特區政府實施了「淨化海港計劃」，旨在改善維港 兩岸之水質。計劃實施後，維港水質已見改善，然而，關於沿岸之岩質潮澗帶復 元的狀況，刖仍然很不清楚。因此，是次硏究的重點便集中於測量岩岸生態，包 括整個群落結構及個別品種譬如日本笠藤壺(Tetraclita japonica)對這項措施的 反應。 / 於二零零四年六月至二零零七年五月期間，六個沿著維多利亞港的岩岸被挑 選作群落結構考察之用。由西至東，這些天然岩岸分別爲：坪州、青州、銅鑼灣、 海心公園、杏花村以及東龍州。結果發現海心公園跟其餘岩岸有很大的分別，其 多樣性指數（H´ة）只有0.36 ±0.40，是眾岩岸中最低的，其餘的岩岸的多樣性指 數則徘徊於2.13 ±0.37至2.25 ±0.21之間。在整個硏究期間，只有11種岩岸 生物在海心公園內被發現，遠低於其他岩岸（物種數量於32至38之間）。此外，海心公園的岩岸群落結構跟其他岩岸只有25%之相似度，而這高度的不相似主要是因爲來自紋藤壺與色球藻於海心公園有著很高的覆蓋率，同時也沒有在 該處發現鳥爪擬帽貝。撇除海心公園的資料，其餘岩岸顯示出一個沿著維多利亞 港由西向東的相似度下降趨勢（由85%降至70%相似度），位於最東面的東龍 州之群落結構跟最西面的坪州最爲不相似。硏究發現各考察點之間群落結構上的 不同，是跟六個岩岸水域中的磷含量坡度有關，以致海水中藻類之群落結構有所 轉移，從而透過上行效應改變整個岩岸之群落結構。大部份的考察點於污染物被 停止投入維多利亞港後仍未有復元的跡象。不過，有一個微弱的復元徵兆刖在東 龍州岩岸上出現。結果顯示，僞全年生的褐藻-半葉馬尾藻的相對數量於硏究期 間有10%之增長，而岩岸食植者的數量亦於此期間有20%之下降，此現象顯 示出東龍州的水質有所改善，而且亦相對比其餘的考察點好。這結果可能與此岩 岸的位置及季節性的水流方向有關。 / 個別物種的生理或生態變化可能反映出相應水質的狀態。爲了測試本地常見 的藤壺，日本笠藤壺，是否可以用於此目的，是次硏究也重點地測試其種群動態 和生殖生物學。從二零零五年八月開始，在維多利亞港五個岩岸（海心公園除外) 進行一個爲期十四個月的監測，而離維多利亞港較遠的海下灣的樣本亦被收集以 充當對照。日本笠藤壺的卵母細胞於硏究期間亦被採集，以量度牠們的大小及密度。而這些卵母細胞的大小和密度均有著空間性的變化。青州樣本孵著最大（70.50 ± 12.49微米）與最高密度的卵（每平方毫米76.08 ±33.30顆卵)。而最小的平均卵 母細胞直徑於海下灣樣本中發現(57.57 ± 18.86微米)，最低的平均卵母細胞密度 則於坪州樣本中發現（每平方毫米64.20 ± 38.63顆卵）。從維多利亞港以東至西 面的樣本有著卵母細胞直徑下降之趨勢。而這現象與其空間性的水質差異（矽 含量）有關，而此差異可能會改變浮游生物的群落結構，因此影響到這些浮游生 物被轉換爲日本笠藤壺之食物的營養成效，導致卵母細胞在不同岩岸有著空間性 的大小差異。然而，這種關係並沒有在卵母細胞密度與水質之間被測到，因此， 使用卵母細胞的總數也許更能準確地反映出藤壺的育雛情況。此外，不同於以往 的硏究，採集回來的樣本被發現全年皆有育雛情況。這可能是由於近年來全球暖 化，提升了海水溫度，觸發了卵子發育這個對於溫度比較敏感的生理變化。 / 至於日本笠藤壺之種群動態硏究，其基部直徑和族群密度亦被測量。結果顯 示兩項參數均有著空間性之差異。海下灣樣本之平均基部直徑爲最小（14.99 ± 7.03毫米），最大的則是銅鑼灣的樣本（22.47 ± 4.69毫米）。相反，海下灣記錄 得最高的平均藤壺密度（每平方米605.11 ±428.75個體），而最低的密度則被發 現於坪州（每平方米82.46 ± 26.23個體）。回歸分析顯示每月平均空間性藤壺基 部直徑與其族群密度有著重大的關係，因此最密集的族群被發現在海下灣，但平 均基部直徑卻是最小的。日本笠藤壺之種群動態被檢測到與水質具備了一些曲線或指數的關係。懸浮物ー也許是藤壺的食物來源，被發現到跟此品種的生長及附 著之空間性差異有密切的關係。此品種大規模附著事件於本港之初夏或夏季中段 期間錄得。但是，此品種的新個體亦於中旬冬季在海下灣、青州、銅鑼灣及東龍 州被發現，唯該期間並不是此品種的繁殖活躍期，這些新個體可能來自香港以 外，有著不同繁殖期族群的區域。 / 本硏究展示了沿維多利亞港的岩質潮澗帶群落於污染消減後的反應。普遍 地，整個岩岸群落結構和日本笠藤壺在空間性的反應，比起時態的反應更爲明 顯。更長時間的監測應持續進行，以確認復元是否會真正展現，或是此生態系統 經已達到其新的平衡階段，以致復元並不可能。至於日本笠藤壺，更詳細的硏究， 當可提供更多資料作爲水質指標的參考。因此，在目前的硏究所得的結果可以作 爲基線資料以提供將來硏究之用。然而，這些污染對群落帶來的改善也許是不可 逆轉的，而這些原來的棲息地可能會永久失。因此，一個好的保育計劃或發展 計劃，應該比污染後的減污行動更應被優先的考慮。 / Chiu, Yat Ming. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2009. / Includes bibliographical references (leaves 210-222). / Abstract also in Chinese. / Acknowledgements --- p.i / Abstract (English) --- p.iv / Abstract (Chinese) --- p.x / Contents --- p.xiv / List of Tables --- p.xvii / List of Figures --- p.xix / Chapter Chapter 1 --- General Introduction / Chapter 1.1 --- Intertidal Areas and Rocky Shores --- p.1 / Chapter 1.1.1 --- The Importance of Rocky Shores --- p.2 / Chapter 1.1.2 --- Factors Affecting the Communities of Rocky Shores --- p.3 / Chapter 1.1.3 --- Effects of Pollution on the Rocky Shore Communities --- p.6 / Chapter 1.2 --- Hong Kong Rocky Shores --- p.9 / Chapter 1.2.1 --- Victoria Harbour and the Harbour Area Treatment Scheme (HATS) --- p.10 / Chapter 1.2.2 --- Hong Kong Rocky Shores and HATS --- p.13 / Chapter 1.3 --- The Use of Rocky Intertidal Organisms as Pollution Indicators --- p.14 / Chapter 1.3.1 --- Tetraclita japonica --- p.15 / Chapter 1.4 --- Objectives --- p.17 / Chapter 1.5 --- Thesis Outlines --- p.17 / Chapter Chapter 2 --- Comparison of the Community Structure of the Rocky Intertidal Shores along Victoria Harbour / Chapter 2.1 --- Introduction --- p.24 / Chapter 2.2 --- Materials and Methods --- p.28 / Chapter 2.2.1 --- Sampling Sites and Timing --- p.28 / Chapter 2.2.2 --- Detailed Descriptions of the Characteristics of Each Study Sites --- p.29 / Chapter 2.2.3 --- Community Structure Monitoring --- p.32 / Chapter 2.2.4 --- Statistical Analysis --- p.33 / Chapter 2.3 --- Results --- p.36 / Chapter 2.3.1 --- Species Diversity --- p.36 / Chapter 2.3.2 --- Spatial and Temporal Comparison of Rocky Intertidal Community Assemblages --- p.37 / Chapter 2.3.3 --- Correlation of Water Quality with Community Structure --- p.42 / Chapter 2.4 --- Discussion --- p.43 / Chapter 2.5 --- Summary --- p.50 / Chapter Chapter 3 --- Effect of Eutrophication on the Reproduction of Tetraclita japonica / Chapter 3.1 --- Introduction --- p.84 / Chapter 3.2 --- Materials and Methods --- p.87 / Chapter 3.2.1 --- Sampling Sites and Timing of Collection --- p.87 / Chapter 3.2.2 --- Sample Collection and Preparation --- p.89 / Chapter 3.2.3 --- Histological Studies of the Ovaries of Tetraclita japonica --- p.89 / Chapter 3.2.4 --- Statistical Analysis --- p.91 / Chapter 3.3 --- Results --- p.92 / Chapter 3.3.1 --- Histological Studies of the Ovaries of Tetraclita japonica --- p.92 / Chapter 3.3.2 --- Correlation of Water Quality with the Reproductive Biology of Tetraclita japonica --- p.94 / Chapter 3.4 --- Discussion --- p.96 / Chapter 3.5 --- Summary --- p.100 / Chapter Chapter 4 --- Population Dynamics of Tetraclita japonica / Chapter 4.1 --- Introduction --- p.137 / Chapter 4.2 --- Materials and Methods --- p.142 / Chapter 4.2.1 --- Sampling Sites and Timing --- p.142 / Chapter 4.2.2 --- Population Dynamic Studies of Tetraclita japonica --- p.142 / Chapter 4.2.3 --- Statistical Analysis --- p.143 / Chapter 4.3 --- Results --- p.145 / Chapter 4.3.1 --- Population Dynamic Studies of Tetraclita japonica --- p.145 / Chapter 4.3.2 --- Relationship between Water Quality and the Population Structures of Tetraclita japonica --- p.154 / Chapter 4.4 --- Discussion --- p.154 / Chapter 4.5 --- Summary --- p.158 / Chapter Chapter 5 --- Summary and Perspectives --- p.203 / Reference --- p.210

Coastal transition zones form important interlinking regions where marine ecosystems, rivers and estuarine environments significantly influence each other. Coastal rocky shores are key habitats that sustain a variety of primary producers and invertebrates and due to the dynamic nature of coastal ecosystems, suspension-feeders on rocky shores can be influenced by an array of autochthonous and allochthonous food sources. Fatty acid and stable isotope trophic markers were employed to distinguish between regional and temporal changes in the potential food sources to rocky shore suspension-feeders in the Eastern Cape Province of South Africa. The primary aim was to assess the spatial and temporal influences of contrasting river flows on the available food sources to three indigenous coastal suspensionfeeders, namely the volcano barnacle Tetraclita serrata, brown mussel Perna perna and tubebuilding polychaete Gunnarea capensis. This was done by examining the intra- and interspecific changes in the fatty acid and stable isotope signatures of the barnacles, mussels and polychaetes in the adjacent marine environment of a freshwater-restricted (Kariega) and freshwater-dominated (Great Fish) estuary during austral summer and winter. Multivariate and Kruskal-Wallis analyses of variance on the fatty acid and isotopic signatures, respectively, identified significant regional changes in the barnacles and mussels, while only stable isotopes distinguished between the Kariega and Great Fish polychaetes (P < 0.05). In addition, significant temporal changes were observed in consumer fatty acids and isotope values in both regions (P < 0.05). Bacterial sources, detritus and phytoplankton assemblages, which are influenced by hydrology and vegetation, differed between regions and were mainly responsible for the regional and temporal separations. Principal component analyses on the consumer fatty acid signatures distinguished between animals situated upstream (i.e. north) and downstream (i.e. south) of the Kariega Estuary mouth in summer. The north/south separation was mainly due to greater contributions of diatoms to northernlocated animals and dinoflagellates and detritus to southern-located consumers. In addition, the south-flowing Agulhas Current on the eastern shores of southern Africa appeared to influence the north/south separation in the Kariega region, as water leaving the estuary was probably entrained into the south-easterly flowing currents, thereby depositing estuarinederived detritus to southern populations. In general, diatoms and detritus were essential food sources to the filter-feeders in summer, and flagellates, diatoms, zooplankton and detritus were important in winter. Coastal macroalgae was a key food source in the Kariega and Great Fish regions during both seasons. Consistently large levels of diatom markers (16:1n-7 and 20:5n-3) and dinoflagellate markers (22:6n-3) in consumer tissues in the Kariega and Great Fish regions identified that phytoplankton was their dominant food source. The barnacles, mussels and polychaetes had similar fatty acid markers and a fairly narrow δ¹³C range (-16.5 to -14.4 ‰), suggesting that they probably consumed similar food. Carbon isotope analyses, however, separated the suspension-feeders into slightly depleted (barnacles; -16.5 to -16.1 ‰), intermediate (mussels; -15.8 to -15.0 ‰) and enriched (polychaetes; -15.0 to -14.4 ‰) consumers, but did not provide conclusive evidence of their preferences for specific phytoplankton. Conversely, fatty acid analyses highlighted that barnacles and mussels had greater proportions of dinoflagellate markers (22:6n-3; 7.0-15.3 % TFA), while polychaetes had larger diatom levels (20:5n-3; 15.1-22.2 % TFA). In addition, all three species had consistently large contributions from bacterial fatty acids (15:0, i-16:0, 17:0 and i-18:0; 4.2-13.6 % TFA) in summer and winter, and large proportions of saturated fatty acids (33.3-53.1 % TFA) including those with 14 to 18 carbons, indicating that bacterial and detritus food sources played an important role in their diets. Barnacles had small levels of terrestrial markers (18:2n-6 and 18:3n-3; <2.5 % TFA) and demonstrated increased omnivorous feeding compared with the other suspension-feeders [increased levels of 20:1n-11 and 20:1n-9, higher 18:1n-9/18:1n-7 ratios at ~2.1, enriched δ¹⁵N values at ~10.6 ‰; zooplankton (potentially including microzooplankton, larvae and protists) contribution of up to 61 % of the diet]. Mussels contained significant proportions of the terrestrial markers (18:2n-6 and 18:3n-3; >2.5 % TFA) and exhibited intermediate omnivory (intermediate levels of 20:1n-11 and 20:1n-9, intermediate 18:1n-9/18:1n-7 ratios at ~1.3, less enriched δ¹⁵N values at ~7.9 ‰; zooplankton contribution of 10-15 % of the diet). The more depleted nitrogen signatures in the mussels relative to the barnacles and polychaetes possibly illustrated a stronger preference for autotrophic food. Polychaetes mainly consumed plant food sources (i.e. microalgae, macroalgae and detritus; high levels of i-18:0, 18:1n-9, 18:4n-3 and 20:5n-3) and displayed little omnivory (low levels of 20:1n-11 and 20:1n-9, low 18:1n-9/18:1n-7 ratios at ~0.4, intermediate δ¹⁵N values at ~9.1 ‰; zooplankton contribution of <10 % of the diet). The barnacles, mussels and polychaetes are all suspension-feeders, originally presumed to consume the same food sources. The variations observed among the species, therefore, may result from differences in the proportional contributions of the various food sources to their diets as well as distinctions in metabolism. The distinct changes in the fatty acid and stable isotope signatures in all three filter-feeders in the Kariega and Great Fish regions are likely influenced by the diversity in regional vegetation and hydrology in the different systems, combined with interspecific differences in resource partitioning among the species.