Integrated risk assessment for WFD ecological status classification applied to Llobregat river basin (Spain). Part I-Fuzzy approach to aggregate biological indicators

Artificial intelligence to explain the variables that favor the cyanobacteria steady-state in tropical ecosystems: A Bayeasian network approach

The steady-state is a situation of little variability of species dominance and total biomass over time. Maintenance of cyanobacteria are often observed in tropical and eutrophic ecosystems and can cause imbalances in aquatic ecosystem. Bayeasian networks allow the construction of simpls models that summarizes a large amount of variables and can predict the probability of occurrence of a given event. Studies considering Bayeasian networks built from environmental data to predict the occurrence of steady-state in aquatic ecosystems are scarce. This study aims to propose a Bayeasian network model to assess the occurrence, composition and duration of cyanobacteria steady-state in a tropical and eutrophic ecosystem. It was hypothesized long lasting steady-state events, composed by filamentous cyanobacteria species and directly influenced by eutrophication and drought. Our model showed steady-state lasting between 3 and 17 weeks with the monodominance or co-dominance of filamentous species, mainly Raphidiopsis raciborskii and Planktothrix agardhii. These evens occurred frequently under drought and high turbidity, however higher nutrients concentrations did not increase the probability steady-state occurrence or longer duration. The proposed model appears as a tool to assess the effects of future warming on steady-state occurrence and it can be a useful to more traditional process-based models for reservoirs.

Ecological status classification of the Taizi River Basin, China: a comparison of integrated risk assessment approaches

Integrated risk assessment approaches allow to achieve a sound evaluation of ecological status of river basins and to gain knowledge about the likely causes of impairment, useful for informing and supporting the decision-making process. In this paper, the integrated risk assessment (IRA) methodology developed in the EU MODELKEY project (and implemented in the MODELKEY Decision Support System) is applied to the Taizi River (China), in order to assess its Ecological and Chemical Status according to EU Water Framework Directive (WFD) requirements. The available dataset is derived by an extensive survey carried out in 2009 and 2010 across the Taizi River catchment, including the monitoring of physico-chemical (i.e. DO, EC, NH3-_N, chemical oxygen demand (COD), biological oxygen demand in 5 days (BOD5) and TP), chemical (i.e. polycyclic aromatic hydrocarbons (PAHs) and metals), biological (i.e. macroinvertebrates, fish, and algae), and hydromorphological parameters (i.e. water quantity, channel change and morphology diversity). The results show a negative trend in the ecological status from the highland to the lowland of the Taizi River Basin. Organic pollution from agriculture and domestic sources (i.e. COD and BOD5), unstable hydrological regime (i.e. water quantity shortage) and chemical pollutants from industry (i.e. PAHs and metals) are found to be the main stressors impacting the ecological status of the Taizi River Basin. The comparison between the results of the IRA methodology and those of a previous study (Leigh et al. 2012) indicates that the selection of indicators and integrating methodologies can have a relevant impact on the classification of the ecological status. The IRA methodology, which integrates information from five lines of evidence (i.e., biology, physico-chemistry, chemistry, ecotoxicology and hydromorphology) required by WFD, allows to better identify the biological communities that are potentially at risk and the stressors that are most likely responsible for the observed alterations. This knowledge can be beneficial for a more effective restoration and management of the river basin ecosystem.

Developing best-practice Bayesian Belief Networks in ecological risk assessments for freshwater and estuarine ecosystems: a quantitative review

Bayesian Belief Networks (BBNs) are being increasingly used to develop a range of predictive models and risk assessments for ecological systems. Ecological BBNs can be applied to complex catchment and water quality issues, integrating multiple spatial and temporal variables within social, economic and environmental decision making processes. This paper reviews the essential components required for ecologists to design a best-practice predictive BBN in an ecological risk assessment (ERA) framework for aquatic ecosystems, outlining: (1) how to create a BBN for an aquatic ERA?; (2) what are the challenges for aquatic ecologists in adopting the best-practice applications of BBNs to ERAs?; and (3) how can BBNs in ERAs influence the science/management interface into the future? The aims of this paper are achieved using three approaches. The first is to demonstrate the best-practice development of BBNs in aquatic sciences using a simple nutrient model. The second is to discuss the limitations and challenges aquatic ecologists encounter when applying BBNs to ERAs. The third is to provide a framework for integrating best-practice BBNs into ERAs and the management of aquatic ecosystems. A quantitative review of the application and development of BBNs in aquatic science from 2002 to 2014 was conducted to identify areas where continued best-practice development is required. We outline a best-practice framework for the integration of BBNs into ERAs and study of complex aquatic systems.

Biomarkers: are realism and control mutually exclusive in integrated pollution assessment?

The conventional view of pollution monitoring is that any choice is a trade-off between realism and precision, as the control over confounding variables decreases with the increasing degree of organization of the test system. Dublin Bay is subject to considerable anthropogenic pressures and there have been many attempts to quantify the status of the system at organizational levels from DNA strand breaks (Comet) to the system itself (Ecological Network analysis, ENA). Using Dublin Bay as an example, the data show there was considerable variability at all levels of organization. At intracellular level, Lysosome Membrane Stability (LMS, assessed by Neutral Red Retention, NRR) varied almost 4-fold with season and individual condition, while the community level AZTI Marine biotic Index (AMBI) had a similar range within a single, supposedly homogeneous, site. Overall, there was no evidence that biomarkers of the lower levels of organisation reduced the variability of the measure, despite the extra control over influencing variables, nor was there any evidence that variability was additive at higher levels of organisation. This poses problems for management, especially given the fixed limits of Ecological Quality Standards (EQSs). Clearly while the integrated approach to pollution monitoring does offer the potential to link effects across the organizational range, it should also be possible to improve their capability by widening the database for reference values, particularly at the higher level of organization, and by process models, including the confounding variables found in the field, for those at lower level.

Ecological screening indicators of stress and risk for the Llobregat river water

The objective of this article is to develop and apply several simple and rough indicators for river aquatic ecosystems assessment in order to screen potential chemical stressors. Several indicators, based on toxicity (PNEC) and on legislation levels (EQS) have been developed. All these indicators are ratios that were calculated by using public and private data of concentrations of a large list of compounds during a period of five years, including metals and organic compounds in the lower part of the Llobregat river basin at the intake of the drinking water treatment plant. Additionally, new campaigns were executed for increasing the information available on the presence of compounds not routinely analyzed, such as some other pesticides and pharmaceuticals. In the case of inorganic pollutants, the indicators obtained in this river section showed significant risk especially for zinc, but also for copper, nickel and barium. For organic pollutants, the pesticides terbuthylazine, diazinon, 2-methyl-4-chlorophenoxyacetic (MCPA), and in a few cases, chlorpyrifos and lindane, also showed indexes above the threshold. Among the pharmaceuticals, the antibiotics clarithromycin and ciprofloxacin were the only ones with risk indicators adverse to ecosystems. The specific values of the indexes obtained rely on the quantity and quality of the data available, so their interpretation should take into account that some values can be high due to the use of too conservative toxicological information.

Water quality analysis in rivers with non-parametric probability distributions and fuzzy inference systems: application to the Cauca River, Colombia

The integration of water quality monitoring variables is essential in environmental decision making. Nowadays, advanced techniques to manage subjectivity, imprecision, uncertainty, vagueness, and variability are required in such complex evaluation process. We here propose a probabilistic fuzzy hybrid model to assess river water quality. Fuzzy logic reasoning has been used to compute a water quality integrative index. By applying a Monte Carlo technique, based on non-parametric probability distributions, the randomness of model inputs was estimated. Annual histograms of nine water quality variables were built with monitoring data systematically collected in the Colombian Cauca River, and probability density estimations using the kernel smoothing method were applied to fit data. Several years were assessed, and river sectors upstream and downstream the city of Santiago de Cali, a big city with basic wastewater treatment and high industrial activity, were analyzed. The probabilistic fuzzy water quality index was able to explain the reduction in water quality, as the river receives a larger number of agriculture, domestic, and industrial effluents. The results of the hybrid model were compared to traditional water quality indexes. The main advantage of the proposed method is that it considers flexible boundaries between the linguistic qualifiers used to define the water status, being the belongingness of water quality to the diverse output fuzzy sets or classes provided with percentiles and histograms, which allows classify better the real water condition. The results of this study show that fuzzy inference systems integrated to stochastic non-parametric techniques may be used as complementary tools in water quality indexing methodologies.

Integrated Risk Assessment for WFD Ecological Status classification applied to Llobregat river basin (Spain). Part II - Evaluation process applied to five environmental Lines of Evidence

Many indicators and indices related to a variety of biological, physico-chemical, chemical, and hydromorphological water conditions have been recently developed or adapted by scientists in order to support water managers in the Water Framework Directive (WFD) implementation. In this context, the achievement of a comprehensive and reliable Ecological Status classification of water bodies across Europe is hampered by the lack of harmonised procedures for selecting an appropriate set of indicators and integrating heterogeneous information in a flexible way. To this purpose, an Integrated Risk Assessment (IRA)(2) methodology was developed based on the Weight of Evidence approach. This method analyses and combines a set of environmental indicators grouped into five Lines of Evidence (LoE), i.e. Biology, Chemistry, Ecotoxicology, Physico-chemistry and Hydromorphology. The whole IRA methodology has been implemented as a specific module into a freeware GIS (Geographic Information System)-based Decision Support System, named MODELKEY DSS. This paper focuses on the evaluation of the four supporting LoE (i.e. Chemistry, Ecotoxicology, Physico-chemistry and Hydromorphology), and includes a procedure for a comparison of each indicator with proper thresholds and a subsequent integration process to combine the obtained output with the LoE Biology results in order to provide a single score expressing the Ecological Status classification. The approach supports the identification of the most prominent stressors, which are responsible for the observed alterations in the river basin under investigation. The results provided by the preliminary testing of the IRA methodology through application of the MODELKEY DSS to the Llobregat case study are finally reported and discussed.