In this study we developed and applied a multimetric index to assess the biological water quality of rivers in Vietnam as a complementary tool to the traditional physical–chemical analysis. ...Macroinvertebrate samples were collected at 15 monitoring sites in the Cau river basin, Northern of Vietnam. Eighteen candidate metrics were tested for their range, stability, sensitivity and responsiveness to anthropogenic impacts. The MMI was calculated as the arithmetic mean of five metrics that were retained being the Biological Monitoring Working Party (BMWP) – Viet, total number of taxa, Margalef index, number of ephemeroptera, plecoptera and trichoptera (EPT) and percent of insects. The MMI is split up in five water quality classes, ranging from class one (high biological status) to class five (bad biological status). The study demonstrated that the multimetric approach is suitable for application in the Vietnamese national monitoring and assessment program.
In the scope of the European Water Framework Directive (WFD) river restoration has received increased attention. By restoring the physical habitat it is expected that the natural dynamics of the ...aquatic system improve and thus the good ecological status can be achieved or maintained by 2015. To this end, several restoration actions, such as the construction of ecologically sound banks (ESBs) have been implemented. ESBs are sites where the riparian zone is restored to enhance the natural aquatic functions and related physical, chemical and biological variables. However, the impact of typical restoration measures, such as the construction of ESBs, on the ecological water quality is not yet quantified. Currently, few restoration projects rely on scientific evidence.
In this study, the effect of ESB construction on the ecological status of water bodies is analysed and the key elements important for ESB design and maintenance are investigated. In cooperation with six Dutch water boards a large dataset was collected consisting of 926 records comprising data on water quality, ecological status and ESB variables. After data pre-processing, 82 and 112 records were retained for the macroinvertebrate and macrophyte community, respectively. Data-driven classification trees were developed for both biological communities with sufficient reliabilities; the correctly classified instances amounted to 81±3.5% and 81±3.6% and the kappa statistic to 0.62±0.06 and 0.61±0.08 for the macrophyte and macroinvertebrate community, respectively. Stakeholders participated in the modelling process and evaluated all statistically reliable modes for their ecological relevance and applicability for decision support in water management.
Our results suggest that ESB construction is beneficial in the scope of the WFD. We found that ESBs contribute to a diverse macrophyte and macroinvertebrate community. The key variables for proper ESB site selection and design are: water type, bank type, water level management, sediment type, ESB type, water function. Also, maintaining the ESB that are constructed is crucial for their effectiveness. Models consisting of rules concerning the design and maintenance conditions were developed and communicated to the river managers by means of the easily interpretable classification trees.
Macroinvertebrates were sampled at 15 locations in the Iskar river basin in Bulgaria for the purpose of water quality assessment. Based on the chemical as well as the biological parameters, it was ...concluded that the water quality was still good upstream of Sofia, however, despite a huge waste water treatment plant, a strong decrease was observed when the river passed Sofia. Due to self-purification and dilution, a gradual amelioration of the water quality was observed 40 and 80km downstream of Sofia, however, water quality was still insufficient. The Irish Biotic Index (IBI), which is currently used in Bulgaria for the national monitoring of macroinvertebrates for water quality assessment, does not fulfil the requirements of the European Union Water Framework Directive (WFD). The Multimetric Macroinvertebrate Index Flanders (MMIF), on the contrary, is a WFD compliant method developed for the northern part of Belgium, which is based on (1) the total number of taxa, (2) the number of Ephemeroptera, Plecoptera and Trichoptera taxa, (3) the number of other sensitive taxa, (4) the Shannon–Wiener index and (5) the mean tolerance score. The outcome of this MMIF was strongly correlated with the outcome of the Irish Biotic Index. Therefore, it should be possible to develop a similar multimetric index for macroinvertebrates to evaluate the biological water quality in Bulgaria without much effort.
To meet the requirements of the EU Water Framework Directive, models are useful to predict communities in watercourses based on the abiotic characteristics of their aquatic environment. For that ...purpose back-propagation Artificial Neural Network (ANN) algorithms were used to induce predictive models on a dataset of the Zwalm river basin (Flanders, Belgium). This dataset consisted of 120 samples, collected over a 2-year period. Fifteen environmental variables were measured at each site, as well as the abundance of the aquatic macroinvertebrate taxa. Different neural networks were developed and optimized to obtain the best model configuration for the prediction of the habitat suitability of macroinvertebrate taxa. The best performing number of hidden layers and neurons and training algorithms have been searched for. The different options were theoretically and practically validated and assessed. The theoretical validation was based on cross-validation. For the practical validation, potential applications of the neural network models were analyzed, and the predictive performance of the models was assessed using ecological expert knowledge. The results indicate that the number of times a taxon was found in the whole river basin influences the performance measures and the architecture of the network. Based on the Cohen’s kappa, it could be concluded that ANN models predicting the presence/absence of very rare taxa (e.g.
Aplexa) or very common taxa (e.g. Tubificidae) were rather irrelevant, although their correctly classified instances (CCI) was high. Predicting the presence/absence of Asellidae (a moderately present taxon), the highest performances (CCI and Cohen’s kappa) were found for the network model with two hidden layers each having 10 neurons. When calculation time was also taken into account, the network model with one hidden layer having 10 neurons could be preferred. Applying this network architecture, performances were only slightly worse, while calculation time was a lot shorter. One may also conclude that not all network models resulted in a relevant relation between a variable and a specific taxon. For Gammaridae for example, a rather small ANN structure gave a better idea of the impact of dissolved oxygen on its presence than a larger one. More reliable predictions and ecological interpretations for river ecosystem management would thus be possible provided the best configuration could be found.
Harbours, which are often characterised by anthropogenic stress in combination with intensive international ship traffic, tend to be very susceptible to aquatic invasions. Since alien ...macrocrustaceans are known to be very successful across many European waters, a study was made on their distribution and impact in the four Belgian coastal harbours (Nieuwpoort, Ostend, Blankenberge and Zeebrugge). Biological and physical–chemical data were gathered at 43 sampling sites distributed along a salinity gradient in the four harbours. One-fourth of all crustacean species recorded were alien and represented on average 30% of the total macrocrustacean abundance and 65% of the total macrocrustacean biomass. The large share of alien crustaceans in the total macrocrustacean biomass was mainly due to several large alien crab species. Most alien species were found in the oligohaline zone, whereas the number of indigenous species slightly increased with increasing salinity. The low number of indigenous species present at low salinities was probably not only caused by salinity, but also by the lower water quality in this salinity range. Based on the site-specific biocontamination index (SBCI), which was used to assess the ecological water quality, the harbour of Nieuwpoort and Ostend scored best and were classified as good, indicating the limited abundance and the low number of alien macrocrustaceans. Sampling locations situated more inland generally had a higher SBCI and a lower ecological water quality. Zeebrugge and Blankenberge were characterised by a severe biocontamination. For Zeebrugge, this is probably related to the intensive transcontinental commercial ship traffic, whereas for Blankenberge, this could be due to introduction of alien species via recreational crafts or due to its geographical location in the proximity of Zeebrugge. Consistent monitoring of estuarine regions and harbours, which are seen as hotspots for introductions, could help in understanding and predicting the impact of alien species on native biota.
Predicting freshwater organisms based on machine learning is becoming more and more reliable due to the availability of appropriate datasets, advanced modelling techniques and the continuously ...increasing capacity of computers. A database consisting of measurements collected at 360 sampling sites in non-navigable watercourses in Flanders was applied to predict the absence/presence of benthic macroinvertebrate taxa by means of decision trees. The measured variables were a combination of physical–chemical (temperature, pH, dissolved oxygen concentration, conductivity, total organic carbon, Kjeldahl nitrogen and total phosphorus), structural (granulometric analysis of the sediment, width, depth and flow velocity of the river) and two ecotoxicological variables. The predictive power of decision trees was assessed on the basis of the number of Correctly Classified Instances (CCI). A genetic algorithm was introduced to compare the predictive power of different sets of input variables for the decision trees. The number of input variables was reduced from 15 to 2–8 variables without affecting the predictive power of the decision trees significantly. Furthermore, reducing the number of input variables allowed to ease the identification of general data trends.
Teaching students to develop data-driven models is a challenging task as a good balance has to be found between the theoretical background of the models, the ecological relevance of the knowledge ...rules inferred and their socio-economic applicability. In this context it is unclear which aspects of the modeling process are easily understood by students, and in particular, how theoretical issues interfere with practical boundary conditions and socio-economic relevance (ecosystem protection, water management, policy development, ecological engineering). In order to fill this knowledge gap, students developed static data-driven models and tutors assessed students' performances. Criteria such as the theoretical, ecological and socio-economic relevance of the derived knowledge rules were used to select the most optimal models.
We noticed an inverse relationship between the complexity of the subtasks and the number of students that succeeded. Students evaluated their models with respect to the theoretical reliability, but were not likely to consider the other two criteria. Half of the students succeeded in assessing the models based on their ecological relevance and only 17% of the students checked the socio-economic relevance of the knowledge rules. Four groups out of seven assessed their models merely based on the predictive power of the models. Only one group integrated the theoretical, ecological and socio-economic relevance to assess the models.
The key findings of our research can be used to optimize the efficiency of data mining courses. We reveal which aspects of the modeling process students seem to overemphasize and give recommendations about the topics trainers should emphasize in the future to ensure that students develop advanced skills. Based on our results the theory–practice dichotomy in higher education can be further reduced. Our learning-by-doing approach showed students how to solve common problems in ecological data sets (e.g. missing data, outliers, collinearity, non-normal distribution, parameterization, uncertainty, etc.), which are often only briefly discussed in basic statistical courses.
When forecasting invasions, models built on a dataset from a certain region often have to be used for simulations in another geographic region. Results on the reliability and usefulness of such ...models are missing in literature. The present study compares habitat suitability models for the invasive amphipod species Dikerogammarus villosus developed based on data gathered in recently invaded rivers and channels in Flanders (Belgium), with similar models developed on the basis of long-term colonised systems in Croatia. The models were tested on their reliability in both regions. Two techniques, logistic regressions (LR) and classification trees (CT) were used to analyse the habitat preference of this species based on physical–chemical and morphological habitat characteristics. It was found that in Flanders, D. villosus prefers rivers with a non-natural bank structure, high oxygen saturation, low conductivity and good chemical water quality, which could be related to its distribution in large rivers and canals. In Croatian rivers, high oxygen saturation was the most important prerequisite for the species to be present. Despite the longer history of invasion in Croatia, the species seemed to have similar habitat preferences in both invaded regions. Both data-driven approaches yielded similar results, but CT performed somewhat better based on the used performance criteria (% Correctly Classified Instances, Kappa and Area Under Curve) and were easier to interpret compared to the LR. The CT models developed based on the data of Flanders performed moderately when applying on the data of Croatia, but had a lower performance when applied vice versa. The LR models did not perform well when applying on a dataset of another geographic area. Extrapolation of the logistic regression model seemed to be more difficult compared to classification tree models. Our results indicate that it is possible to determine the habitat preference of an invasive species and that these models could be applied to other regions in Europe in order to take preventive measures to control the further spread of invasive species. However, a major concern is that the models are developed based on a representative range of all relevant variables reflecting the stream conditions and that accurate data are important.
A literature survey and the identification of all available collection material resulted in a checklist and distribution maps for the mayflies occurring in Flanders. In addition, the relationship ...between the occurrence of mayflies and water characteristics was analysed. Of the 32 species that have been recorded, six are now extinct in Flanders (three of which are potamal species), while the majority of the remaining species are rare and their populations are often strongly isolated and therefore extremely vulnerable. Waters with relatively low oxygen levels and high conductivities were characterized by the most tolerant mayfly species Cloeon dipterum and Caenis robusta. However, most other species only occurred at higher oxygen concentrations and lower conductivities and could be separated into two groups. The first group mainly occurred in waters with a high pH and were often restricted to the loamy region or to stagnant waters, while the second group occurred in waters with a lower pH and mainly occurred in the Campine region. For most mayfly species, sustainable populations can only be achieved when their current habitats are adequately protected and, in addition, measures should be taken to connect and enlarge the remaining populations.
Antropogenic activities have severely deteriorated the river systems in Flanders as a result of which many functions such as drinking water supply, fishing, … are being threatened. Because ...restoration of these river systems entails drastic social and economical consequences, actions should be considered in advance. In this context, migration dynamics of predicted organisms and migration barriers along the river can deliver important additional information for habitat suitability models on the effectiveness of restoration plans. To this purpose, migration models for
Baetis (Insecta, Ephemeroptera),
Ephemera (Insecta, Ephemeroptera) and Limnephilidae (Insecta, Trichoptera) have been developed for the Zwalm river basin, Flanders, Belgium. The migration models consisted of three resistance layers: one for migration through the air/over land and two for migration through the river in upstream and downstream direction. Based on the Cost Weighted Distance function, source populations and migration potential could be calculated. In combination with habitat suitability calculations based on Artificial Neural Networks (ANNs), the migration models were used to simulate the effect of removing a weir used for water quantity control. According to the ANN habitat suitability models, this removal did not affect the habitat suitability for
Baetis,
Ephemera and Limnephilidae. The ANN models predicted that after restoration the habitat was still not suitable for the taxa considered. In spite of this, the migration model for
Baetis could be applied to simulate the possible recolonization of the restored river section in case of further habitat improvement. As calculated by the model, the shortest path with the least accumulative cost for migration would be through the air. Based on the migration model of
Baetis, it would take approximately 275 days to recolonize the restored river.