Distribution and habitat preferences of marine megafauna in Nosy Be, Madagascar
- Authors: Lubbé, Alicea
- Date: 2023-10-13
- Subjects: Uncatalogued
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/424710 , vital:72177
- Description: Marine megafauna have important ecological roles including the top-down regulation of lower trophic levels and the transport of nutrients. They are also charismatic species of socioeconomic importance, due to their public appeal and focus for ecotourism. However, these taxa face numerous anthropogenic threats including bycatch, habitat-loss, noise disturbance, prey reductions, pollution, and vessel traffic. Northwest Madagascar has been identified as an important marine mammal area and whale shark aggregation site, yet limited information on megafauna within Madagascar exists. In this thesis, I investigated the community structure, distribution trends, related environmental factors (Chapter 2) and predicted habitat suitability (Chapter 3) using a species distribution modelling approach for marine megafauna around Nosy Be, North-west Madagascar. Data collection consisted of opportunistic boatbased surveys conducted while looking for whale sharks, mobulid rays and cetaceans. These activities were performed by dedicated ‘swim with whale shark’ tourism operators, Les Baleines Rand'eau and Safari Baleine during tourism activities directed by the Madagascar Whale Shark Project over four years (2016-2019). A total of 1792 sightings were reported with 13 identified species, including elasmobranchs: whale shark, Rhincodon typus; spinetail mobula ray, Mobula japonica; giant oceanic manta ray, Manta birostris; shortfin devil ray, Mobula kuhlii; balaenids: humpback whale, Megaptera novaeangliae; Omura's whale, Balaenoptera omurai, delphinids: Indo-Pacific bottlenose dolphin, Tursiops aduncus; spinner dolphin, Stenella longirostris; pantropical spotted dolphin, Stenella attenuata; Indian ocean humpback dolphin, Sousa plumbea; false killer whale, Pseudorca crassidens; melonheaded whale, Peponocephala electra and one cheloniid species: leatherback sea turtle, Dermochelys coriacea. Sightings were spatially analysed at a community level according to feeding guilds (i.e., filter feeders vs. predators), and distribution of the megafauna groups v were compared to selected remotely sensed physico-chemical data (sea surface temperature and chlorophyll-a concentration) and bathymetry charts (depth and slope) using generalized linear models (Chapter 2). All four variables had a significant effect on filter feeder distribution. Spatial distribution of filter feeder sightings peaked at an area of steep underwater topography (18-50m; x slope=0.5%) located on the west coast of Nosy Be, where foraging aggregations of large filter feeders (primarily R. typus and B. omurai) were frequently observed. Only SST had a significant effect on the relative abundance of animals in the predator feeding guild, which demonstrated a more longitudinal distribution along the continental shelf. The maximum entropy model (Maxent) was used for predicting habitat preference for the most frequently sighted species (Chapter 3). The distribution of whale sharks, mobulid rays and Omura’s whales were significantly overlapped as they were abundant in two main hotspots, Grand banc de l’entrée (~10 km of the west coast) and near Nosy Mitsio island, (~50 km northeast of Nosy Be). Suitable habitat for the dolphins was in relatively shallow waters in temperatures <28ºC near areas of steep changes in bathymetry. Bottlenose and spinner dolphins had a more widespread distribution across the continental shelf and humpback dolphin was closely associated with inshore reefs and occur in waters <10m. The coastal and inshore region of Nosy Be is an important habitat for a variety of megafaunal species, and the continental shelf provided essential areas for feeding and breeding. The results were similar to those presented for these species in other regions of the Indian Ocean (i.e., Eastern and Southern Africa; Western Australia; Seychelles) as well as outside of the Indian Ocean (i.e., Indonesia; Caribbean; North Atlantic). Thus the results from this thesis could be considered for management recommendations for marine conservation in the region. The value and limitations of data collected by citizen science were outlined (Chapter 4). , Thesis (MSc) -- Faculty of Science, Zoology and Entomology, 2023
- Full Text:
- Date Issued: 2023-10-13
Native-range studies on insect herbivores associated with African Lovegrass (Eragrostis curvula) in South Africa: prospects for biological control in Australia
- Authors: Yell, Liam Dougal
- Date: 2023-10-13
- Subjects: Uncatalogued
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/424446 , vital:72154
- Description: Eragrostis curvula (Schrad.) Nees. (African Lovegrass) is an African native species of grass that was intentionally introduced for pasture in Australia. It has since escaped cultivation and has become widespread and highly invasive. Eragrostis curvula has been recorded in every state and territory in Australia where it has altered fire regimes, disrupted nutrient cycles and can reduce livestock carrying capacity by up to 50%. The Centre for Biological Control at Rhodes University and the New South Wales Department of Primary Industries have been working in collaboration to identify and screen herbivorous insects as biological control agents for E. curvula in Australia. Native-range surveys were conducted between 2021 and 2022 on E. curvula at twenty-two sites across South Africa to identify herbivorous natural enemies associated with it. Species accumulation curves were generated to ensure adequate sampling was performed to identify all the insects associated with E. curvula. Twenty-nine non-target grass species were surveyed simultaneously to determine the field-host range of the natural enemies associated with the target weed. Herbivorous natural enemies were prioritised as possible biological control agents against E. curvula in Australia based on field-host range, predicted efficacy and climatic suitability. Four insect species were consistently found on E. curvula, two of which were herbivorous, as well as a parasitoid and a detritivore. Species accumulation curves show that the insect community was adequately sampled in South Africa. The two herbivorous insects were identified to the lowest taxonomic level using COI barcoding. Both species are undescribed phytophagous wasps in the genus Tetramesa (Hymenoptera: Eurytomidae). Because Tetramesa species have been shown to be host specific and highly damaging in previous biological control programs for other invasive grass weeds, we assessed their suitability as candidate biological control agents for use on E. curvula in Australia. Both Tetramesa species (“sp. 4” and “sp. 5”) were found on several native congeners under field conditions in South Africa. Congeneric South African-native non-target grass species were used as phylogenetic proxies to assess the risk posed to Australian native Eragrostis species. This highlighted three non-target Australian native Eragrostis species, namely: E. parviflora (R. Br.) Trin., E. leptocarpa Benth. fl., and E. trachycarpa Benth., that are at risk of being attacked by the two candidate agents based on their phylogenetic proximity to E. curvula. Predicted efficacy trials were conducted at five long-term repeat survey sites and revealed that Tetramesa sp. 4 does not reduce the probability of E. curvula tiller survival or reproduction, while Tetramesa sp. 5 does not reduce the probability of tiller reproduction but does increase the probability of tiller survival. This result was unexpected and may be a plant compensatory response to herbivory. The sites where both Tetramesa species were collected in South Africa are climatically similar to the invaded range of E. curvula in Australia, and as such, the Tetramesa spp. are likely to be suitably adapted to the climate where they would be released in Australia. These results suggest that both Tetramesa species associated with E. curvula may have too broad a host range to be used as biological control agents in Australia. However, further quarantine-based host-range assessments on Australian native Eragrostis species are recommended to confirm this. The field-based methods used in this study have reduced the number of insect and plant species that host-range assessments will be required to be performed on, thus preventing wasted resources. , Thesis (MSc) -- Faculty of Science, Zoology and Entomology, 2023
- Full Text:
- Date Issued: 2023-10-13
Phytoplankton and aquatic macroinvertebrate assemblages from coastal and inland lakes of South Africa
- Authors: Nkibi, Esethu
- Date: 2023-10-13
- Subjects: Uncatalogued
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/424435 , vital:72153
- Description: Freshwater lakes are generally defined as permanent natural standing water bodies, with some of them having a direct and indirect connection with groundwater, rivers, and the ocean. Freshwater lakes provide essential socio-economic and ecological goods and services including recreation, aesthetic, support aquatic biodiversity, food in a form of fisheries and water for domestic use. Given their critical role in sustainability in providing socio-economic services, freshwater lakes are among the most threatened ecosystems globally due to intense human impacts over the last decades. South Africa has limited freshwater lakes, which according to the National Biodiversity Assessment (NBA 2018), we know little about their current biology except historic aquatic biodiversity studies conducted in the early 1940s. There are no management strategies in place to protect and conserve freshwater lake biodiversity and important ecosystem services. This thesis aims to: (1) produce a biodiversity inventory of phytoplankton and aquatic macroinvertebrate species and, (2) investigate important environmental drivers responsible for phytoplankton and aquatic macroinvertebrate species composition from six South African freshwater lakes. It was hypothesized that the three lake types will show different phytoplankton and aquatic macroinvertebrates species composition attributed by the geographical region (coastal and inland lake) and related physico-chemical parameters. Study sites consisted of two Northern KwaZulu-Natal coastal lakes (hereafter Coastal Lakes, CL) i.e., Lake Sibaya, Lake Mzingazi; two fresh inland lakes (hereafter Fresh Inland Lakes, FIL) i.e., Lake Banagher fresh and Lake Tevrede Se Pan; and two inland salt lakes (hereafter Salt Inland Lakes, SIL) i.e., Lake Banagher salt and Lake Chrissiesmeer, all inland lakes are situated in Mpumalanga province, together with other Pans making up the Mpumalanga Lake District of South Africa. The study sites were categorized based on their geographical position i.e., coastal vs inland and physico-chemical characteristics i.e., the presence and absence of aquatic vegetation, dominate substrate, salinity and different physico-chemical concentration i.e., Temperature, Dissolved Oxygen. Aquatic macroinvertebrates were collected from four littoral zone sites (< 1-meter depth) around each lake, whereas phytoplankton samples were collected from four water column sites (> 5-meters depth) and (0.5-meter depth) from the water surface at each lake during summer and winter season. The results were consistent with our hypothesis that both phytoplankton and aquatic macroinvertebrate species composition were influenced by physico-chemical parameters and that the differences in salinity concentration and aquatic vegetation between CL, FIL, and SIL were the driving factors for phytoplankton and aquatic macroinvertebrate species composition. In summary, one hundred and twenty-two phytoplankton taxa were collected and identified during this study, belonging to seven Phyla which included Chlorophyta, Bacillariophyta, Cyanophyta, Chrysophyta, Dinophyta, Euglenophyta, and Cryptophyta. The most abundant phytoplankton groups were Bacillariophyta and Chlorophyta. Phytoplankton relative taxa abundance, Pielou’s evenness, taxa richness, and Shannon diversity were significantly different between lake types. Aquatic macroinvertebrates, on the other hand, summed up to 10 orders, 67 families, and 80 taxa. The most abundant group were the order Coleoptera, Hemiptera, Odonata, and Gastropoda. Aquatic macroinvertebrate relative taxa abundance, taxa richness, and Shannon diversity were also significantly different between lake types. Aquatic macroinvertebrate relative taxa abundance, Pielous evenness, and Shannon diversity index were not significant between seasons, and only taxa richness was significant. Canonical analysis of principal coordinates (CAP) results further showed unique and distinct phytoplankton and aquatic macroinvertebrates community composition between lake types. The present study provides baseline biodiversity inventory (or species list) for important lake ecosystems biological indicators i.e., phytoplankton and aquatic macroinvertebrates and species composition in relation to lake type for six freshwater lakes in South Africa. Furthermore, the study provides empirical evidence that will inform policy and the development of management strategies for freshwater lakes in South Africa which is currently missing. The current study will also contribute to the next National Biodiversity Assessment Report (2024), concerning the freshwater lakes biological data deficiency noted in the previous NBA (2018) report. The study will also fill up the gaps to better understand species composition in lake systems and how they function which is currently limited. , Thesis (MSc) -- Faculty of Science, Zoology and Entomology, 2023
- Full Text:
- Date Issued: 2023-10-13