Setting nutrient boundaries to protect aquatic communities: The importance of comparing observed and predicted classifications using measures derived from a confusion matrix.

Defining nutrient thresholds that protect and support the ecological integrity of aquatic ecosystems is a fundamental step in maintaining their natural biodiversity and preserving their resilience. With increasing catchment pressures and climate change, it is more important than ever to develop clear methods to establish thresholds for status classification and management of waters. This must often be achieved using complex data and should be robust to interference from additional pressures as well as ameliorating or confounding conditions. We use both artificial and real data to examine challenges in setting nutrient thresholds in unbalanced and skewed data. We found significant advantages to using binary logistic regression over other techniques. However, one of the key challenges is objectively selecting a probability from which to derive the nutrient threshold. For this purpose, the examination of the proportions of matching and mismatching status classifications of nutrients and a biological quality element using a confusion matrix is a key step that should be more widely adopted in threshold selection. We examined a large array of statistical measures of classification accuracy and their performance over combinations of skewness and imbalance in the data. The most appropriate threshold probability is a compromise between maximising overall classification accuracy and reducing mismatches expressed as commission (false positives) without excessive omission (false negatives). An application to a lake type indicated total phosphorus thresholds that would be around 50 µg l-1 lower than the threshold achieved by an 'unguided' approach, indicating that this approach is a very significant development meriting attention from national authorities responsible for water management.

Establishing ecologically-relevant nutrient thresholds: A tool-kit with guidance on its use.

One key component of any eutrophication management strategy is establishment of realistic thresholds above which negative impacts become significant and provision of ecosystem services is threatened. This paper introduces a toolkit of statistical approaches with which such thresholds can be set, explaining their rationale and situations under which each is effective. All methods assume a causal relationship between nutrients and biota, but we also recognise that nutrients rarely act in isolation. Many of the simpler methods have limited applicability when other stressors are present. Where relationships between nutrients and biota are strong, regression is recommended. Regression relationships can be extended to include additional stressors or variables responsible for variation between water bodies. However, when the relationship between nutrients and biota is weaker, categorical approaches are recommended. Of these, binomial regression and an approach based on classification mismatch are most effective although both will underestimate threshold concentrations if a second stressor is present. Whilst approaches such as changepoint analysis are not particularly useful for meeting the specific needs of EU legislation, other multivariate approaches (e.g. decision trees) may have a role to play. When other stressors are present quantile regression allows thresholds to be established which set limits above which nutrients are likely to influence the biota, irrespective of other pressures. The statistical methods in the toolkit may be useful as part of a management strategy, but more sophisticated approaches, often generating thresholds appropriate to individual water bodies rather than to broadly defined "types", are likely to be necessary too. The importance of understanding underlying ecological processes as well as correct selection and application of methods is emphasised, along with the need to consider local regulatory and decision-making systems, and the ease with which outcomes can be communicated to non-technical audiences.

European aquatic ecological assessment methods: A critical review of their sensitivity to key pressures.

The European Union has embarked on a policy which aims to achieve good ecological status in all surface waters (i.e. rivers, lakes, transitional and coastal waters). In theory, ecological status assessment methods should address the effects of all relevant human pressures. In this study, we analyze the degree to which methods European countries use to assess ecological status tackle various pressures affecting European waters. Nutrient pollution is by far the best-covered pressure for all four water categories. Out of total of 423 assessment methods, 370 assess eutrophication and pressure-specific relationships have been demonstrated for 212 of these. "General degradation" is addressed by 238 methods, mostly validated by relationships to combined pressure indices. Other major pressures have received significantly less effort: hydromorphological degradation is assessed by 160 methods and pressure-specific relationships have been demonstrated for just 40 of these. Hydromorphological pressures are addressed (at least by one BQE) only by 25% countries for coastal waters and 70-80% for lakes and transitional waters. Specific diagnostic tools (i.e. single-pressure relationships) for hydromorphology have only been developed by a few countries: only 20% countries have such methods for lakes, coastal and transitional waters and less than half for rivers. Toxic contamination is addressed by 90 methods; however, pressure-specific relationships have been demonstrated for just eight of these. Only two countries have demonstrated pressure-specific acidification methods for rivers, and three for lakes. In summary, methods currently in use mostly address eutrophication and/or general degradation, but there is not much evidence that they reliably pick up the effects of other significant pressures such as hydromorphology or toxic contamination. Therefore, we recommend that countries re-examine: (1) those pressures which affect different water categories in the country; (2) relevant assessment methods to tackle those pressures; (3) whether pressure-response relationships have been developed for each of these.

Establishing nutrient thresholds in the face of uncertainty and multiple stressors: A comparison of approaches using simulated datasets.

Various methods have been proposed to identify threshold concentrations of nutrients that would support good ecological status, but the performance of these methods and the influence of other stressors on the underlying models have not been fully evaluated. We used synthetic datasets to compare the performance of ordinary least squares, logistic and quantile regression, as well as, categorical methods based on the distribution of nutrient concentrations categorised by biological status. The synthetic datasets used differed in their levels of variation between explanatory and response variables, and were centered at different positions along the stressor (nutrient) gradient. In order to evaluate the performance of methods in "multiple stressor" situations, another set of datasets with two stressors was used. Ordinary least squares and logistic regression methods were the most reliable when predicting the threshold concentration when nutrients were the sole stressor; however, both had a tendency to underestimate the threshold when a second stressor was present. In contrast, threshold concentrations produced by categorical methods were strongly influenced by the level of the stressor (nutrient enrichment, in this case) relative to the threshold they were trying to predict (good/moderate in this instance). Although all the methods tested had limitations in the presence of a second stressor, upper quantiles seemed generally appropriate to establish non-precautionary thresholds. For example, upper quantiles may be appropriate when establishing targets for restoration, but not when seeking to minimise deterioration. Selection of an appropriate threshold concentration should also attend to the regulatory regime (i.e. policy requirements and environmental management context) within which it will be used, and the ease of communicating the principles to managers and stakeholders.

Nutrient criteria for surface waters under the European Water Framework Directive: Current state-of-the-art, challenges and future outlook.

The aim of European water policy is to achieve good ecological status in all rivers, lakes, coastal and transitional waters by 2027. Currently, more than half of water bodies are in a degraded condition and nutrient enrichment is one of the main culprits. Therefore, there is a pressing need to establish reliable and comparable nutrient criteria that are consistent with good ecological status. This paper highlights the wide range of nutrient criteria currently in use by Member States of the European Union to support good ecological status and goes on to suggest that inappropriate criteria may be hindering the achievement of good status. Along with a comprehensive overview of nutrient criteria, we provide a critical analysis of the threshold concentrations and approaches by which these are set. We identify four essential issues: (1) Different nutrients (nitrogen and/or phosphorus) are used for different water categories in different countries. (2) The use of different nutrient fractions (total, dissolved inorganic) and statistical summary metrics (e.g., mean, percentiles, seasonal, annual) currently hampers comparability between countries, particularly for rivers, transitional and coastal waters. (3) Wide ranges in nutrient threshold values within shared water body types, in some cases showing more than a 10-fold difference in concentrations. (4) Different approaches used to set threshold nutrient concentrations to define the boundary between "good" and "moderate" ecological status. Expert judgement-based methods resulted in significantly higher (less stringent) good-moderate threshold values compared with data-driven approaches, highlighting the importance of consistent and rigorous approaches to criteria setting. We suggest that further development of nutrient criteria should be based on relationships between ecological status and nutrient concentrations, taking into account the need for comparability between different water categories, water body types within these categories, and countries.