Background: Clear-water mangroves, characterized by their proximity to coral reefs and minimal terrestrial input, play a critical role within the coastal ecosystem mosaic (CEM). Despite the importance of these habitats, research on their ecological roles in the Indo-Pacific is limited, with most studies drawing from Caribbean systems. This thesis investigates the spatial and temporal drivers of fish assemblages in clear-water mangroves in north Queensland, emphasizing functional and taxonomic diversity to investigate how tidal-influenced habitats in the Indo-Pacific are used by fish.
Objective: This study aimed to assess the spatial distribution, habitat use, and functional trait diversity of fish in Indo-Pacific clear-water mangroves. It specifically examines the influence of environmental variables such as tidal regime, substrate type, water depth, and seasonality on fish community structure and ecological functions within these habitats.
Methods: Data were collected using remote underwater videos (RUVs) across six sites at Orpheus Island, Queensland. Fish observations were recorded in Excel, and high-frequency water level data loggers recorded the tidal regimes. Community composition was analysed using PERMANOVA, and Generalized additive models (GAMs) were employed to evaluate the relationships between fish size, fish size ratio (proxy for age), trophic level, and mangrove habitat use. These models provided insights into how environmental conditions influenced fish behaviour and habitat selection. Functional diversity indices - functional richness (FRic), functional evenness (FEve), and functional divergence (FDiv) - were calculated to assess how species occupied and were distributed within the functional trait space. RLQ and fourth-corner analyses identified the significant relationships between functional traits and environmental factors such as substrate type, depth, and season. The Random Forest models showed how the significant environmental drivers shaped the functional traits. Exclusion analyses further examined the ecological role of rays (Batoidea), particularly within sand habitats, in shaping functional trait diversity.
Results: Taxonomic diversity was significantly influenced by substrate and season, with rocky substrates supporting higher species richness and abundance than sand. Juvenile fish were primarily associated with sand substrates, likely due to greater refuge and feeding opportunities. Sand habitats exhibited the highest functional richness (FRic), largely driven by large-bodied rays (Batoidea) and their benthic feeding behaviour. These rays occupied unique trophic niches at the edges of the trait space, contributing to both high functional richness and divergence (FDiv). Excluding rays significantly reduced functional richness and divergence in sand habitats, while having little impact on rocky substrates, highlighting their specialized ecological role. Rocky substrates showed higher functional evenness (FEve), with species more uniformly distributed across the trait space. RLQ and fourth-corner analyses identified strong links between functional traits and environmental variables such as substrate type and depth. Random Forest models confirmed substrate as the primary driver of functional trait distributions, followed by depth and seasonality. These findings highlight the importance of preserving species that enhance functional trait diversity, particularly for ecosystem resilience.
This dataset consist of:
- One CSV file containing the recorded environmental variables and fish observation data for each individual observation
- One CSV file containing the functional trait diversity species-based trait matrix
- PDF of the R script used for statistical analysis and modelling
- A README.txt file with metadata descriptors
- The HOBOware data logger data in a zip. file with a folder for each deployment site
Software/equipment used to create/collect the data:
Remote underwater cameras for video recording fish behaviour (GoPro HERO3 Silver Edition HD3.02.03.00 or Adventure Kings Action Camera 1080P Full HD)
HOBO data loggers for tidal and water level measurements (HoboWare, HOBO U20L)
Software/equipment used to manipulate/analyse the data:
RStudio for statistical analysis and visualisation
HOBOware Pro for HOBO data loggers
Microsoft 365 Excel for data visualisation