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.

Recovery of fish and crustacean communities during remediation of tidal wetlands affected by leachate from acid sulfate soils in north-eastern Australia.

In the 1970s, acid sulfate soils (ASS) distributed within about 720 ha of predominantly mangrove and salt pan wetlands at East Trinity in north Queensland were developed after the area was isolated from tidal flooding by a surrounding seawall and the installation of tidal gates on major drainage creeks. Following drainage and oxidation of these estuarine acidic sediments, resultant acid leachate caused considerable, ongoing environmental problems including regular fish kills. A rehabilitation program covering much of these former tidal wetlands commenced in 2000 using a lime-assisted tidal exchange management regime. Changes in the established populations of estuarine fish and crustaceans were monitored in the two creeks (Firewood and Hills Creeks) where tidal flows were reinstated. In Firewood Creek between 2001 and 2005, there was a progressive increase in fish species richness, diversity and abundance. The penaeid prawn Fenneropenaeus merguiensis was a major component of the cast net catches in the lower sections of both Firewood and Hills Creeks but its relative abundance decreased upstream of the tidal gates on the seawall. Well established stocks of predominantly juvenile, male Scylla serrata resident upstream of the tidal gates indicated suitable habitats with acceptable water and sediment quality and adequate availability of food. The regular fish kills that occurred prior to the management regime abated and, overall, the implementation of the rehabilitation program is yielding positive benefits for the local fisheries.

Fisheries resources of the Bohle River, Queensland.

The fisheries resources of the Bohle River and its small catchment area adjacent Townsville, north Queensland, were investigated through available literature, scientific research surveys and analysis of commercial and recreational catch and effort data. Research surveys produced a total of 4383 fish from the waters of the Bohle River during 1997-1998. These were classified into 104 fish species from 49 families. Gillnetting, cast netting, fish trapping and crab potting techniques were used in the estuarine waters of the Bohle River with freshwater reaches in the upper catchment surveyed by electrofishing. This range of survey techniques was used to estimate the relative abundance of ten commercially and recreationally important species: Barramundi (Lates calcarifer), king threadfin (Polydactylus macrochir), blue threadfin (Eleutheronema tetradactylum), mangrove jack (Lutjanus argentimaculatus), banded and spotted grunter (Pomadasys kaakan and Pomadasys argenteus), pikey and yellowfin bream (Acanthopagrus berda and Acanthopagrus australis), tilapia (Oreochromis spp.), jungle perch (Kuhlia rupestris) and mud crab (Scylla serrata). The results of each survey method are discussed with a focus on spatial and temporal patterns in diversity and catch rate.

Fish mediate high food web connectivity in the lower reaches of a tropical floodplain river

High levels of hydrological connectivity during seasonal flooding provide significant opportunities for movements of fish between rivers and their floodplains, estuaries and the sea, possibly mediating food web subsidies among habitats. To determine the degree of utilisation of food sources from different habitats in a tropical river with a short floodplain inundation duration (similar to 2 months), stable isotope ratios in fishes and their available food were measured from three habitats (inundated floodplain, dry season freshwater, coastal marine) in the lower reaches of the Mitchell River, Queensland (Australia). Floodplain food sources constituted the majority of the diet of large-bodied fishes (barramundi Lates calcarifer, catfish Neoarius graeffei) captured on the floodplain in the wet season and for gonadal tissues of a common herbivorous fish (gizzard shad Nematalosa come), the latter suggesting that critical reproductive phases are fuelled by floodplain production. Floodplain food sources also subsidised barramundi from the recreational fishery in adjacent coastal and estuarine areas, and the broader fish community from a freshwater lagoon. These findings highlight the importance of the floodplain in supporting the production of large fishes in spite of the episodic nature and relatively short duration of inundation compared to large river floodplains of humid tropical regions. They also illustrate the high degree of food web connectivity mediated by mobile fish in this system in the absence of human modification, and point to the potential consequences of water resource development that may reduce or eliminate hydrological connectivity between the river and its floodplain.

The effects of human induced disturbances on the sediment geochemistry of Baltic Sea embayments with a focus on eutrophication history

Historical sediment nutrient concentrations and heavy-metal distributions were studied in five embayments in the Gulf of Finland and an adjacent lake. The main objective of the study was to examine the response of these water bodies to temporal changes in human activities. Sediment cores were collected from the sites and dated using 210Pb and 137Cs. The cores were analyzed for total carbon (TC), total nitrogen (TN), total phosphorus (TP), organic phosphorus (OP), inorganic phosphorus (IP), biogenic silica (BSi), loss on ignition (LOI), grain size, Cu, Zn, Al, Fe, Mn, K, Ca, Mg and Na. Principal component analysis (PCA) was used to summarize the trends in the geochemical variables and to compare trends between the different sites. The links between the catchment land use and sediment geochemical data were studied using a multivariate technique of redundancy analysis (RDA). Human activities produce marked geochemical variations in coastal sediments. These variations and signals are often challenging to interpret due to various sedimentological and post-depositional factors affecting the sediment profiles. In general, the sites studied here show significant upcore increases in sedimentation rates, TP and TN concentrations. Also Cu, which is considered to be a good indicator of anthropogenic influence, showed clear increases from 1850 towards the top part of the cores. Based on the RDA-analysis, in the least disturbed embayments with high forest cover, the sediments are dominated by lithogenic indicators Fe, K, Al and Mg. In embayments close to urban settlement, the sediments have high Cu concentrations and a high sediment Fe/Mn ratio. This study suggests that sediment accumulation rates vary significantly from site to site and that the overall sedimentation can be linked to the geomorphology and basin bathymetry, which appear to be the major factors governing sedimentation rates; i.e. a high sediment accumulation rate is not characteristic either to urban or to rural sites. The geochemical trends are strongly site specific and depend on the local geochemical background, basin characteristics and anthropogenic metal and nutrient loading. Of the studied geochemical indicators, OP shows the least monotonic trends in all studied sites. When compared to other available data, OP seems to be the most reliable geochemical indicator describing the trophic development of the study sites, whereas Cu and Zn appear to be good indicators for anthropogenic influence. As sedimentation environments, estuarine and marine sites are more complex than lacustrine basins with multiple sources of sediment input and more energetic conditions in the former. The crucial differences between lacustrine and estuarine/coastal sedimentation environments are mostly related to Fe. P sedimentation is largely governed by Fe redox-reactions in estuarine environments. In freshwaters, presence of Fe is clearly linked to the sedimentation of other lithogenic metals, and therefore P sedimentation and preservation has a more direct linkage to organic matter sedimentation.

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.

RTK-based GNSS drifters for monitoring in estuaries and rivers: Field experimental results

This paper presents a novel RTK-based GNSS Lagrangian drifter system that is capable of monitoring water velocity, turbulence and dispersion coefficients of river and estuarine. The Lagrangian drifters use the dual-frequency real time kinematic (RTK) technique for both position and velocity estimations. The capsule is designed to meet the requirements such as minimizing height, diameter, minimizing the direct wind drag, positive buoyancy for satellite signal reception and stability, and waterproof housing for electronic components, such as GNSS receiver and computing board. The collected GNSS data are processed with post-processing RTK software. Several experiments have been carried out in two rivers in Brisbane and Sunshine Coast in Queensland. Results show that the high accuracy GNSS-drifters can be used to measure dispersion coefficient resulting from sub-tidal velocity fluctuations in shallow tidal water. In addition, the RTK-GNSS drifters respond well to vertical motion and thus could be applicable to flood monitoring.

Does the age of fine root carbon indicate the age of fine roots in boreal forests?

To test the reliability of the radiocarbon method for determining root age, we analyzed fine roots (originating from the years 1985 to 1993) from ingrowth cores with known maximum root age (1 to 6 years old). For this purpose, three Scots pine (Pinus sylvestris L.) stands were selected from boreal forests in Finland. We analyzed root 14C age by the radiocarbon method and compared it with the above-mentioned known maximum fine root age. In general, ages determined by the two methods (root 14C age and ingrowth core root maximum age) were in agreement with each other for roots of small diameter (<0.5mm). By contrast, in most of the samples of fine roots of larger diameter (1.5-2mm), the 14C age of root samples of 1987-89 exceeded the ingrowth core root maximum age by 1-10 years. This shows that these roots had received a large amount of older stored carbon from unknown sources in addition to atmospheric CO2 directly from photosynthesis. We conclude that the 14C signature of fine roots, especially those of larger diameter, may not always be indicative of root age, and that further studies are needed concerning the extent of possible root uptake of older carbon and its residence time in roots. Keywords: fine root age, Pinus sylvestris, radiocarbon, root carbon, ingrowth cores, tree ring

A Strategy for Conserving the Biodiversity of the Uttara Kannada District in South India

Taking the various values ascribed to biodiversity as its point of departure rather many years ago, the present study aims at deriving a conservation strategy for Uttara Kannada. This hilly district, with the highest proportion of its area under forests in South India, is divided into five ecological zones: coastal, northern evergreen, southern evergreen, moist deciduous, and dry deciduous. The heavily-populated coastal zone includes mangrove forests and estuarine wetlands. The evergreen forests are particularly rich in the diversity of plant species which they support - including wild relatives of a number of cultivated plants. They also serve a vital function in watershed conservation. The moist deciduous forests are rich in bird species; both moist and dry deciduous forests include a number of freshwater ponds and lakes that support a high diversity of aquatic birds.Reviewing the overall distribution of biodiversity, we identify specific localities - including estuaries, evergreen forests, and moist deciduous forests - which should be set aside as Nature reserves. These larger reserves must be complemented by a network of traditionally-protected sacred groves and sacred trees that are distributed throughout the district and that protect today, for instance, the finest surviving stand of dipterocarp trees.We also spell out the necessary policy-changes in overall development strategy that should stem the ongoing decimation of biodiversity. These include (1) revitalizing community-based systems of sustainable management of village forests and protection of sacred groves and trees; (2) reorienting the usage-pattern of reserve forests from production of a limited variety of timber and softwood species for industrial consumers, to production of a larger diversity of non-wood forest produce of commercial value to support the rural economy; (3) utilizing marginal lands under private ownership for generating industrial wood supplies; and (4) provision of incentives for in situ maintenance of land-races of cultivated plants - especially evergreen, fruit-yielding trees - by the local people.It is proposed that this broad framework be now taken to the local communities, and that an action-plan be developed on the basis of inputs provided - and initiatives taken - by them.

Mining Land Cover Information Using Multilayer Perceptron and Decision Tree from MODIS Data

Land cover (LC) changes play a major role in global as well as at regional scale patterns of the climate and biogeochemistry of the Earth system. LC information presents critical insights in understanding of Earth surface phenomena, particularly useful when obtained synoptically from remote sensing data. However, for developing countries and those with large geographical extent, regular LC mapping is prohibitive with data from commercial sensors (high cost factor) of limited spatial coverage (low temporal resolution and band swath). In this context, free MODIS data with good spectro-temporal resolution meet the purpose. LC mapping from these data has continuously evolved with advances in classification algorithms. This paper presents a comparative study of two robust data mining techniques, the multilayer perceptron (MLP) and decision tree (DT) on different products of MODIS data corresponding to Kolar district, Karnataka, India. The MODIS classified images when compared at three different spatial scales (at district level, taluk level and pixel level) shows that MLP based classification on minimum noise fraction components on MODIS 36 bands provide the most accurate LC mapping with 86% accuracy, while DT on MODIS 36 bands principal components leads to less accurate classification (69%).

Simulating the daily discharge of the Mandovi River, west coast of India

A hydrological modelling framework was assembled to simulate the daily discharge of the Mandovi River on the Indian west coast. Approximately 90% of the west-coast rainfall, and therefore discharge, occurs during the summer monsoon (June-September), with a peak during July-August. The modelling framework consisted of a digital elevation model (DEM) called GLOBE, a hydrological routing algorithm, the Terrestrial Hydrological Model with Biogeochemistry (THMB), an algorithm to map the rainfall recorded by sparse rain-gauges to the model grid, and a modified Soil Conservation Service Curve Number (SCS-CN) method. A series of discharge simulations (with and without the SCS method) was carried out. The best simulation was obtained after incorporating spatio-temporal variability in the SCS parameters, which was achieved by an objective division of the season into five regimes: the lean season, monsoon onset, peak monsoon, end-monsoon, and post-monsoon. A novel attempt was made to incorporate objectively the different regimes encountered before, during and after the Indian monsoon, into a hydrological modelling framework. The strength of our method lies in the low demand it makes on hydrological data. Apart from information on the average soil type in a region, the entire parameterization is built on the basis of the rainfall that is used to force the model. That the model does not need to be calibrated separately for each river is important, because most of the Indian west-coast basins are ungauged. Hence, even though the model has been validated only for the Mandovi basin, its potential region of application is considerable. In the context of the Prediction in Ungauged Basins (PUB) framework, the potential of the proposed approach is significant, because the discharge of these (ungauged) rivers into the eastern Arabian Sea is not small, making them an important element of the local climate system.

Effects of temperature on complexes I and II mediated respiration, ROS generation and oxidative stress status in isolated gill mitochondria of the mud crab Scylla serrata

Effects of fluctuations in habitat temperature (18-30 degrees) on mitochondrial respiratory behavior and oxidative metabolic responses in the euryhaline ectotherm Scylla serrate are not fully understood. In the present study, effects of different temperatures ranging from 12 to 40 degrees C on glutamate and succinate mediated mitochondrial respiration, respiratory control ratio (RCR), ATP generation rate, ratio for the utilization of phosphate molecules per atomic oxygen consumption (P/O), levels of lipid peroxidation and H2O2 in isolated gill mitochondria of S. serrata are reported. The pattern of variation in the studied parameters was similar for the two substrates at different temperatures. The values recorded for RCR ( >= 3) and P/O ratio (1.4-2.7) at the temperature range of 15-25 degrees C were within the normal range reported for other animals (3-10 for RCR and 1.5-3 for P/O). Values for P/O ratio, ATP generation rate and RCR were highest at 18 degrees C when compared to the other assay temperatures. However, at low and high extreme temperatures, i.e. at 12 and 40 degrees C, states III and IV respiration rates were not clearly distinguishable from each other indicating that mitochondria were completely uncoupled. Positive correlations were noticed between temperature and the levels of both lipid peroxidation and H2O2. It is inferred that fluctuations on either side of ambient habitat temperature may adversely influence mitochondria respiration and oxidative metabolism in S. serrata. The results provide baseline data to understand the impacts of acute changes in temperature on ectotherms inhabiting estuarine or marine environments. (C) 2014 Elsevier Ltd. All rights reserved.

Mixing process in estuaries

A theoretical expression for vertical profile-of horizontal velocity in terms of its depth-average is derived based on oscillatory boundary layer theory and estuarine flow characteristics. The derived theoretical profile is then incorporated into a vertical quasi-two-dimensional model, which is proved advantageous in more physical implications and less CPU time demand. To validate the proposed model, the calculated results are compared to the field data in the Yangtze River Estuary, exhibiting good agreement with observations. The proposed quasi-two-dimensional vertical model is used to study mixing process, especially dependence of salinity distribution and salt front strength on runoff and tides in estuaries.

PICES Press, Vol. 15, No. 1, January 2007

PICES science – 2006 (pdf, 0.1 Mb) 2006 Wooster Award (pdf, 0.1 Mb) Thank you note from the Past-Chairman of PICES (pdf, 0.1 Mb) A comparison of regional mechanisms for fish production: Ecosystem perspectives (pdf, 0.3 Mb) 2006 CREAMS/PICES international workshop and summer school (pdf, 0.2 Mb) PICES Calendar (pdf, 0.2 Mb) 2006 Harmful Algal Bloom Section annual workshop (pdf, 0.1 Mb) 2006 PICES Workshop on “Modeling iron biogeochemistry and ocean ecosystems” (pdf, 0.1 Mb) Strolling through the NEMURO ecosystem model (pdf, 0.1 Mb) Climate and marine birds and mammals in the North Pacific (pdf, 0.2 Mb) Photo highlights of the PICES Fifteenth Annual Meeting (pdf, 3.5 Mb) Recent trends in waters of the subarctic NE Pacific: Cooler and fresher in summer of 2006 (pdf, 0.2 Mb) The state of the western North Pacific in the first half of 2006 (pdf, 0.3 Mb) Latest and upcoming PICES publications (pdf, 0.3 Mb) A seven-year effort of the PICES CCCC MODEL Task Team culminates in a dedicated issue of Ecological Modelling (pdf, 0.1 Mb) Japan joins PICES Marine Metadata Federation (pdf, 0.3 Mb) Argo: A 2006 status report (pdf, 0.3 Mb) New Chairmen in PICES (pdf, 0.2 Mb) PICES Interns (pdf, 0.2 Mb)

PICES Press, Vol. 11, No. 1, January 2003

Editorial and Contents The state of PICES science - 2002 Second annual Wooster Award to Yutaka Nagata Qingdao Open Science Meeting: A major landmark for GLOBEC GLOBEC OSM Session Highhlights PICES/GLOBEC Data Management Workshop PICES and GLOBEC modelling Some personal impressions of the GLOBEC OSM Photo highlights of PICES XI and GLOBEC OSM PICES Climate Change and Carrying Capacity (CCCC) Integration Workshop PICES/CLIVAR Workshop on Climate Variability in the Pacific and its impact on the marine ecosystem IGBP/SCOR Open Science Meeting on Ocean Biogeochemistry and Ecosystem Analysis Subarctic Ecosystem Response to Iron Enrichment Study (SERIES): eastern subarctic Pacific in July 2002 Introducing the GLOBEC International Project Office / GLOBEC Calendar 2003 Introducing the PICES Secretariat PICES Calendar