Dynamic Bayesian network inferencing for non-homogeneous complex systems

Dynamic Bayesian Networks (DBNs) provide a versatile platform for predicting and analysing the behaviour of complex systems. As such, they are well suited to the prediction of complex ecosystem population trajectories under anthropogenic disturbances such as the dredging of marine seagrass ecosystems. However, DBNs assume a homogeneous Markov chain whereas a key characteristics of complex ecosystems is the presence of feedback loops, path dependencies and regime changes whereby the behaviour of the system can vary based on past states. This paper develops a method based on the small world structure of complex systems networks to modularise a non-homogeneous DBN and enable the computation of posterior marginal probabilities given evidence in forwards inference. It also provides an approach for an approximate solution for backwards inference as convergence is not guaranteed for a path dependent system. When applied to the seagrass dredging problem, the incorporation of path dependency can implement conditional absorption and allows release from the zero state in line with environmental and ecological observations. As dredging has a marked global impact on seagrass and other marine ecosystems of high environmental and economic value, using such a complex systems model to develop practical ways to meet the needs of conservation and industry through enhancing resistance and/or recovery is of paramount importance.

Predicting the temporal response of seagrass meadows to dredging using Dynamic Bayesian Networks

Predicting temporal responses of ecosystems to disturbances associated with industrial activities is critical for their management and conservation. However, prediction of ecosystem responses is challenging due to the complexity and potential non-linearities stemming from interactions between system components and multiple environmental drivers. Prediction is particularly difficult for marine ecosystems due to their often highly variable and complex natures and large uncertainties surrounding their dynamic responses. Consequently, current management of such systems often rely on expert judgement and/or complex quantitative models that consider only a subset of the relevant ecological processes. Hence there exists an urgent need for the development of whole-of-systems predictive models to support decision and policy makers in managing complex marine systems in the context of industry based disturbances. This paper presents Dynamic Bayesian Networks (DBNs) for predicting the temporal response of a marine ecosystem to anthropogenic disturbances. The DBN provides a visual representation of the problem domain in terms of factors (parts of the ecosystem) and their relationships. These relationships are quantified via Conditional Probability Tables (CPTs), which estimate the variability and uncertainty in the distribution of each factor. The combination of qualitative visual and quantitative elements in a DBN facilitates the integration of a wide array of data, published and expert knowledge and other models. Such multiple sources are often essential as one single source of information is rarely sufficient to cover the diverse range of factors relevant to a management task. Here, a DBN model is developed for tropical, annual Halophila and temperate, persistent Amphibolis seagrass meadows to inform dredging management and help meet environmental guidelines. Specifically, the impacts of capital (e.g. new port development) and maintenance (e.g. maintaining channel depths in established ports) dredging is evaluated with respect to the risk of permanent loss, defined as no recovery within 5 years (Environmental Protection Agency guidelines). The model is developed using expert knowledge, existing literature, statistical models of environmental light, and experimental data. The model is then demonstrated in a case study through the analysis of a variety of dredging, environmental and seagrass ecosystem recovery scenarios. In spatial zones significantly affected by dredging, such as the zone of moderate impact, shoot density has a very high probability of being driven to zero by capital dredging due to the duration of such dredging. Here, fast growing Halophila species can recover, however, the probability of recovery depends on the presence of seed banks. On the other hand, slow growing Amphibolis meadows have a high probability of suffering permanent loss. However, in the maintenance dredging scenario, due to the shorter duration of dredging, Amphibolis is better able to resist the impacts of dredging. For both types of seagrass meadows, the probability of loss was strongly dependent on the biological and ecological status of the meadow, as well as environmental conditions post-dredging. The ability to predict the ecosystem response under cumulative, non-linear interactions across a complex ecosystem highlights the utility of DBNs for decision support and environmental management.

Testing the sediment-trapping paradigm of seagrass: Do seagrasses influence nutrient status and sediment structure in tropical intertidal environments?

Seagrass meadows are considered important for sediment trapping and sediment stabilisation. Deposition of fine sediments and associated adsorbed nutrients is considered an important part of the chemical and biological processes attributed to seagrass communities. This paradigm was based on work in temperate regions on Zostera marina and in tropical regions on Thalassia testudinum, two species that maintain relatively high biomass, stable meadows. The current study investigates this concept for three species of intertidal tropical seagrass meadows in northeastern Australia. Sediment structure and nutrient status did not differ between vegetated and unvegated habitats in intertidal areas within the central region of the Great Barrier Reef World Heritage Area. The 'trapping' functions that have been attributed to seagrasses need to be re-assessed for a variety of locations and species before they can be accepted as dogma. In tropical Australia, intertidal meadows are predominantly ephemeral and comprised of structurally small species of low biomass. Consequently, sediment trapping within these meadows is largely insignificant.

Potential interactions between humans and non-breeding shorebirds on a subtropical intertidal flat

The primary aim of this study was to investigate whether bait harvesting, with all its inherent effects, occurring in the intertidal zone of a subtropical estuary, had an impact on a migratory shorebird, the eastern curlew Numenius madagascariensis. In a large-scale manipulative study (units of experiment were 1 ha plots), callianassid shrimp Trypaea australiensis populations were harvested simulating the technique (manual pumping) and the levels of harvesting intensity per unit area (347 shrimp per hectare per harvesting event) exhibited by bait-collectors in SE Australia and South Africa. It was found that at present levels of harvesting intensity per unit area (approximately 1% of standing stock removed per harvesting event) there is no threat to the stocks of Trypaea exploited by the curlews in Moreton Bay, Australia. However, the results show that the curlews themselves apply a considerable predation pressure on Trypaea. Based on the birds' foraging rates and densities, it was estimated that they would consume up to 100% of the initial Trypaea stock over the course of a non-breeding season (October to March). However, the stable seasonal trend in the density of the size-cohort of Trypaea preyed upon by the curlews indicates that the existing rates of predation are easily counterbalanced, e.g. through continuous density-dependent recruitment of these crustaceans. We suggest that this mechanism will provide for a stable foraging environment for both the shorebirds and bait collectors.

A broad-scale analysis of links between coastal fisheries production and mangrove extent: A case-study for northeastern Australia

The paradigm that mangroves are critical for sustaining production in coastal fisheries is widely accepted, but empirical evidence has been tenuous. This study showed that links between mangrove extent and coastal fisheries production could be detected for some species at a broad regional scale (1000s of kilometres) on the east coast of Queensland, Australia. The relationships between catch-per-unit-effort for different commercially caught species in four fisheries (trawl, line, net and pot fisheries) and mangrove characteristics, estimated from Landsat images were examined using multiple regression analyses. The species were categorised into three groups based on information on their life history characteristics, namely mangrove-related species (banana prawns Penaeus merguiensis, mud crabs Scylla serrata and barramundi Lates calcarifer), estuarine species (tiger prawns Penaeus esculentus and Penaeus semisulcatus, blue swimmer crabs Portunus pelagicus and blue threadfin Eleutheronema tetradactylum) and offshore species (coral trout Plectropomus spp.). For the mangrove-related species, mangrove characteristics such as area and perimeter accounted for most of the variation in the model; for the non-mangrove estuarine species, latitude was the dominant parameter but some mangrove characteristics (e.g. mangrove perimeter) also made significant contributions to the models. In contrast, for the offshore species, latitude was the dominant variable, with no contribution from mangrove characteristics. This study also identified that finer scale spatial data for the fisheries, to enable catch information to be attributed to a particular catchment, would help to improve our understanding of relationships between mangroves and fisheries production. (C) 2005 Elsevier B.V. All rights reserved.

Patterns of digenean parasitism of bivalves from the Great Barrier Reef and associated waters

Digenean parasites of marine bivalves are relatively poorly known, particularly in Australia. We surveyed 2256 bivalve individuals ( 47 species, 17 families) from Queensland marine waters incorporating south-east Queensland, Heron Island ( southern Great Barrier Reef) and Lizard Island ( northern Great Barrier Reef). Infections of trematode species from three families, Bucephalidae, Gorgoderidae and Monorchiidae, were found. Overall prevalence of infection was 2.3%. The Bucephalidae was the most commonly found family; 11 species were found in Tellinidae, Ostreidae, Isognomonidae and Spondylidae - the latter two previously unknown as hosts for bucephalids. A single gorgoderid infection was found in a venerid, Lioconcha castrensis. Five species of monorchiids were found from Tellinidae and Lucinidae. All infections are new host/parasite records. No infections were found in 35 of the 47 bivalve species sampled. The generally low prevalence of infection by digeneans of bivalves suggests that it is unlikely that any of the species reported here are seriously damaging to bivalve populations in these waters. We deduce that, at best, we have some life-cycle information but no actual identifications for 10% of the species of trematodes that infect bivalves of Queensland marine waters.

Infracommunity structure of parasites of Hemigymnus melapterus (Pisces : Labridae) from Lizard Island, Australia: The importance of habitat and parasite body size

This Study describes the community of all metazoan parasites from 14 individuals of thicklip wrasse, Hemigymnus melapterus, from Lizard Island, Australia. All fish were parasitized, and 4,649 parasite individuals were found. Twenty-six parasite species were identified although only 6 species were abundant and prevalent: gnathiid isopods, the copepod Hatschekia hemigymni, the digenean Callohelmis pichelinae, and 3 morphotypes of tetraphyllidean cestode larvae. We analyzed whether the body size and microhabitat of the parasites and size of the host affected understanding of the structure of the parasite community. We related the abundance, biovolume, and density of parasites with the host body size and analyzed the abundances and volumetric densities of some parasite species within microhabitats. Although the 2 most abundant species comprised 75% of all parasite individuals, 4 species, each in similar proportion, comprised 85% of the total biovolume. Although larger host individuals had higher richness, abundance, and biovolume of parasites than smaller individuals, overall parasite volumetric density actually decreased with the host body size. Moreover. parasites exhibited abundances and densities significantly different among microhabitats; some parasite species depended on the area available, whereas others selected a specific microhabitat. Parasite and habitat size exhibited interesting relationships that should be considered more frequently. Considerations of these parameters improve understanding of parasite community structure and how the parasites use their habitats.

Variation in mangrove forest structure and sediment characteristics in Bocas del Toro, Panama

Mangrove forest structure and sediment characteristics were examined in the extensive mangroves of Bocas del Toro, Republic of Panama. Forest structure was characterized to determine if spatial vegetation patterns were repeated over the Bocas del Toro landscape. Using a series of permanent plots and transects we found that the forests of Bocas del Toro were dominated by Rhizophora maugle with very few individuals of Avicennia germinans and Laguncularia racemosa. Despite this low species diversity, there was large variation in forest structure and in edaphic conditions (salinity, concentration of available phosphorus, Eh and sulphide concentration). Aboveground biomass varied 20-fold, from 6.8 Mg ha(-1) in dwarf forests to 194.3 Mg ha(-1) in the forests fringing the land. But variation in forest structure was predictable across the intertidal zone. There was a strong tree height gradient from seaward fringe (mean tree height 3.9 m), decreasing in stature in the interior dwarf forests (mean tree height 0.7 m), and increasing in stature in forests adjacent to the terrestrial forest (mean tree height 4.1 m). The predictable variation in forest structure emerges due to the complex interactions among edaphic and plant factors. Identifying predictable patterns in forest structure will aid in scaling up the ecosystem services provided by mangrove forests in coastal landscapes.

A Site Description of the CARICOMP Mangrove, Seagrass and Coral Reef Sites in Bocas del Toro, Panama

Bocas del Toro is located in the western region of the Republic of Panama. It is part of a province of approximately 8917 km(2) with an estimated 68% of its area covered by tropical rainforest. The area receives 2870 mm/year of rainfall. The dry and rainy seasons are not clearly defined. There are two periods each of low and high rainfall, March and September-October, and July and December, respectively. Mangrove forests, seagrass meadows and coral reefs are vast, covering large areas in the shallow waters surrounding the islands of the archipelago and along the mainland coast. The CARICOMP sites were established in 1998-99 and are periodically monitored following Level I protocol. Herein we describe the sites in a regional context and present the baseline data for each site. This paper fulfills the requirements of the formal site description for CARICOMP monitoring sites.

Adaptive management and its role in managing Great Barrier Reef water quality

Adaptive management is the pathway to effective conservation, use and management of Australia’s coastal catchments and waterways. While the concepts of adaptive management are not new, applications involving both assessment and management responses are indeed limited at the national and regional scales. This paper outlines the components of a systematic framework for linking scientific knowledge, existing tools, planning approaches and participatory processes to achieve healthy regional partnerships between community, industry, government agencies and science providers to overcome institutional barriers and uncoordinated monitoring. The framework developed by the Coastal CRC (www.coastal.crc.org.au/amf/amf_index.htm) is hierarchical in the way it displays information to allow associated frameworks to be integrated, and represents a construct in which processes, information, decision tools and outcomes are brought together in a structured and transparent way for adaptive catchment and coastal management. This paper proposes how an adaptive management approach could be used to benefit the implementation of the Reef Water Quality Protection Plan (RWQPP).

The effect of temperature on the size and population density of dinoflagellates in larvae of the reef coral Porites astreoides

Pre-settlement events play an important role in determining larval success in marine invertebrates with bentho-pelagic life histories, yet the consequences of these events typically are not well understood. The purpose of this study was to examine the pre-settlement impacts of different seawater temperatures on the size and population density of dinoflagellate symbionts in brooded larvae of the Caribbean coral Porites astreoides. Larvae were collected from P. astreoides at 14-20 m depth on Conch Reef (Florida) in June 2002, and incubated for 24 h at 15 temperatures spanning the range 25.1 degrees-30.0 degrees C in mean increments of 0.4 +/- 0.1 degrees C (+/- SD). The most striking feature of the larval responses was the magnitude of change in both parameters across this 5 degrees C temperature range within 24 h. In general, larvae were largest and had the highest population densities of Symbiodinium sp. between 26.4 degrees-27.7 degrees C, and were smallest and had the lowest population densities at 25.8 degrees C and 28.8 degrees C. Larval size and symbiont population density were elevated slightly (relative to the minimal values) at the temperature extremes of 25.1 degrees C and 30 degrees C. These data demonstrate that coral larvae are highly sensitive to seawater temperature during their pelagic phase, and respond through changes in size and the population densities of Symbiodinium sp. to ecologically relevant temperature signals within 24 h. The extent to which these changes are biologically meaningful will depend on the duration and frequency of exposure of coral larvae to spatio-temporal variability in seawater temperature, and whether the responses have cascading effects on larval success and their entry to the post-settlement and recruitment phase.