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.

Minor effect of wind exposure and littoral slope on macrophyte characteristics in non-impacted lowland lakes of Poland.

Aquatic vegetation is a reliable indicator of the ecological condition of surface waters. Abundance, composition and spatial structure of aquatic communities are shaped by an array of factors, which include both natural abiotic features of an ecosystem and external influences. We investigated whether the physical features, i.e., wind exposure and slope of the lake basin, have a significant impact on the taxonomic composition and spatial structure of macrophyte communities from non-impacted, highly alkaline, lowland lakes of the European plains (Poland). We further examined whether these features can affect the classification of the ecological status of lakes assessed in accordance with the Water Framework Directive requirements. Morphological, botanical and physicochemical data from 260 transects in 16 non-disturbed lakes of Polish lowlands surveyed in the years 2011-2016 were analysed. For each transect, littoral slope and wind exposure were calculated. Additionally, the total phosphorus concentration was used as a proxy of water trophy. The relationships between environmental variables and macrophyte indices as well as the syntaxonomic composition of aquatic and rush vegetation (dependent variables) were analysed using multidimensional ordination techniques (redundancy analysis, variation partitioning and indicator values), correlation and regression analysis. Among the three analysed environmental factors (littoral slope, wind exposure and water trophy), in almost all cases the latter explained the highest variance in the macrophyte community, while the contribution of the first two was at most moderate, weak or usually statistically insignificant. However, lakes with steeper slopes were more frequently inhabited by stoneworts and had better ecological status than those with a gentle littoral shape. This may be attributed to the links between lake morphometry and rate of eutrophication, with deep lakes supporting more effective dilution of substances. Furthermore, lower light requirements of charophytes than of higher plants and the capacity to growth in unstable sediments facilitate charophyte establishment in deeper and steeper parts of the littoral over higher plants. Our findings suggest that in lowland lakes with relatively small areas, moderate depths and low wind exposure typical of European plains, slopes and weaving do not hamper vegetation development and do not negatively affect the macrophyte assessment of ecological status. In such ecosystems, eutrophication seems to be a more important factor determining aquatic vegetation than physical features.

Estimating nutrient thresholds for eutrophication management: Novel insights from understudied lake types.

Nutrient targets based on pressure-response models are essential for defining ambitions and managing eutrophication. However, the scale of biogeographical variation in these pressure-response relationships is poorly understood, which may hinder eutrophication management in regions where lake ecology is less intensively studied. In this study, we derive ecology-based nutrient targets for five major ecoregions of Europe: Northern, Central-Baltic, Alpine, Mediterranean and Eastern Continental. As a first step, we developed regressions between nutrient concentrations and ecological quality ratios (EQR) based on phytoplankton and macrophyte communities. Significant relationships were established for 13 major lake types; in most cases, these relationships were stronger for phosphorus than for nitrogen, and stronger for phytoplankton than for macrophytes. Using these regressions, we estimated the total phosphorus (TP) and total nitrogen (TN) concentrations at which lakes of different types are likely to achieve good ecological status. However, in the very shallow lakes of the Eastern Continental region, relations between nutrient and biological communities were weak or non-significant. This can be attributed to high nutrient concentrations (in the asymptotic zone of phosphorus-phytoplankton models) suggesting other factors (light, grazing) limit primary production. However, we also show that fish stocking is a major pressure on Eastern Continental lakes, negatively affecting ecological status: lakes with low fish stocking show low chlorophyll-a concentrations and good ecological status despite high nutrient levels, while the lakes with high fish stocking show high chlorophyll-a and low ecological status. This study highlights the need to better understand lakes in biogeographic regions that have been, for historical reasons, less studied. This, in turn, helps reveal factors that challenge the dominant paradigms of lake assessment and management.

Modelling of ecological status of Polish lakes using deep learning techniques.

Since 2000, after the Water Framework Directive came into force, aquatic ecosystems' bioassessment has acquired immense practical importance for water management. Currently, due to extensive scientific research and monitoring, we have gathered comprehensive hydrobiological databases. The amount of available data increases with each subsequent year of monitoring, and the efficient analysis of these data requires the use of proper mathematical tools. Our study challenges the comparison of the modelling potential between four indices for the ecological status assessment of lakes based on three groups of aquatic organisms, i.e. phytoplankton, phytobenthos and macrophytes. One of the deep learning techniques, artificial neural networks, has been used to predict values of four biological indices based on the limited set of the physicochemical parameters of water. All analyses were conducted separately for lakes with various stratification regimes as they function differently. The best modelling quality in terms of high values of coefficients of determination and low values of the normalised root mean square error was obtained for chlorophyll a followed by phytoplankton multimetric. A lower degree of fit was obtained in the networks for macrophyte index, and the poorest model quality was obtained for phytobenthos index. For all indices, modelling quality for non-stratified lakes was higher than this for stratified lakes, giving a higher percentage of variance explained by the networks and lower values of errors. Sensitivity analysis showed that among physicochemical parameters, water transparency (Secchi disk reading) exhibits the strongest relationship with the ecological status of lakes derived by phytoplankton and macrophytes. At the same time, all input variables indicated a negligible impact on phytobenthos index. In this way, different explanations of the relationship between biological and trophic variables were revealed.

Global patterns and determinants of lake macrophyte taxonomic, functional and phylogenetic beta diversity.

Documenting the patterns of biological diversity on Earth has always been a central challenge in macroecology and biogeography. However, for the diverse group of freshwater plants, such research program is still in its infancy. Here, we examined global variation in taxonomic, functional and phylogenetic beta diversity patterns of lake macrophytes using regional data from six continents. A data set of ca. 480 lake macrophyte community observations, together with climatic, geographical and environmental variables, was compiled across 16 regions worldwide. We (a) built the very first phylogeny comprising most freshwater plant lineages; (b) exploited a wide array of functional traits that are important to macrophyte autoecology or that relate to lake ecosystem functioning; (c) assessed if different large-scale beta diversity patterns show a clear latitudinal gradient from the equator to the poles using null models; and (d) employed evolutionary and regression models to first identify the degree to which the studied functional traits show a phylogenetic signal, and then to estimate community-environment relationships at multiple spatial scales. Our results supported the notion that ecological niches evolved independently of phylogeny in macrophyte lineages worldwide. We also showed that taxonomic and phylogenetic beta diversity followed the typical global trend with higher diversity in the tropics. In addition, we were able to confirm that species, multi-trait and lineage compositions were first and foremost structured by climatic conditions at relatively broad spatial scales. Perhaps more importantly, we showed that large-scale processes along latitudinal and elevational gradients have left a strong footprint in the current diversity patterns and community-environment relationships in lake macrophytes. Overall, our results stress the need for an integrative approach to macroecology, biogeography and conservation biology, combining multiple diversity facets at different spatial scales.

Macrophyte assessment in European lakes: Diverse approaches but convergent views of 'good' ecological status.

The European Water Framework Directive has been adopted by Member States to assess and manage the ecological integrity of surface waters. Specific challenges include harmonizing diverse assessment systems across Europe, linking ecological assessment to restoration measures and reaching a common view on 'good' ecological status. In this study, nine national macrophyte-based approaches for assessing ecological status were compared and harmonized, using a large dataset of 539 European lakes. A macrophyte common metric, representing the average standardized view of each lake by all countries, was used to compare national methods. This was also shown to reflect the total phosphorus (r2 = 0.32), total nitrogen (r2 = 0.22) as well as chlorophyll-a (r2 = 0.35-0.38) gradients, providing a link between ecological data, stressors and management decisions. Despite differing assessment approaches and initial differences in classification, a consensus was reached on how type-specific macrophyte assemblages change across the ecological status gradient and where ecological status boundaries should lie. A marked decline in submerged vegetation, especially Charophyta (characterizing 'good' status), and an increase in abundance of free-floating plants (characterizing 'less than good' status) were the most significant changes along the ecological status gradient. Macrophyte communities of 'good' status lakes were diverse with many charophytes and several Potamogeton species. A large number of taxa occurred across the entire gradient, but only a minority dominated at 'less than good' status, including filamentous algae, lemnids, nymphaeids, and several elodeids (e.g., Zannichellia palustris and Elodea nuttallii). Our findings establish a 'guiding image' of the macrophyte community at 'good' ecological status in hard-water lakes of the Central-Baltic region of Europe.

Global patterns in the metacommunity structuring of lake macrophytes: regional variations and driving factors.

We studied community-environment relationships of lake macrophytes at two metacommunity scales using data from 16 regions across the world. More specifically, we examined (a) whether the lake macrophyte communities respond similar to key local environmental factors, major climate variables and lake spatial locations in each of the regions (i.e., within-region approach) and (b) how well can explained variability in the community-environment relationships across multiple lake macrophyte metacommunities be accounted for by elevation range, spatial extent, latitude, longitude, and age of the oldest lake within each metacommunity (i.e., across-region approach). In the within-region approach, we employed partial redundancy analyses together with variation partitioning to investigate the relative importance of local variables, climate variables, and spatial location on lake macrophytes among the study regions. In the across-region approach, we used adjusted R2 values of the variation partitioning to model the community-environment relationships across multiple metacommunities using linear regression and commonality analysis. We found that niche filtering related to local lake-level environmental conditions was the dominant force structuring macrophytes within metacommunities. However, our results also revealed that elevation range associated with climate (increasing temperature amplitude affecting macrophytes) and spatial location (likely due to dispersal limitation) was important for macrophytes based on the findings of the across-metacommunities analysis. These findings suggest that different determinants influence macrophyte metacommunities within different regions, thus showing context dependency. Moreover, our study emphasized that the use of a single metacommunity scale gives incomplete information on the environmental features explaining variation in macrophyte communities.

ESMI: a macrophyte index for assessing the ecological status of lakes.

The paper introduces the Ecological State Macrophyte Index (ESMI), a method compliant with the Water Framework Directive (2000/60/EC) for assessing the ecological status of lakes based on macrophytes. A description of the elaboration of macrophyte metrics, relevant reference conditions and a classification system for two types of high-alkalinity lowland lakes (stratified and polymictic), and a customised field survey procedure based on belt transects are presented. The ESMI evaluates two aspects of macrophyte community: taxonomic composition (index of evenness J) and abundance (colonisation index Z), which are combined into one multimetric. ESMI values range from 0 to 1, where 1 denotes pristine conditions and 0, highly degraded habitats. The high/good class boundary (H/G) was set at the first quartile of ESMI values of reference lakes. For the other classes, boundaries were set by dividing the range of ESMI values between the H/G boundary and the minimum value recorded in the dataset in a logarithmic scale into four. The ESMI correlated best with water transparency (Pearson's R = 0.62 in stratified lakes and 0.79 in polymictic ones), whereas the correlations with phosphorus and nitrogen concentrations were somewhat weaker (R = -0.48 to -0.57). Based on the results of international intercalibration, the original class boundaries were modified (merged for stratified and polymictic lakes, the good/moderate boundary tightened to approximately 20%), to make ESMI-based assessment results comparable with the outcomes of other European methods.