Next Article in Journal
Assessing the Impacts of Chloride and Sulfate Ions on Macroinvertebrate Communities in Ohio Streams
Next Article in Special Issue
Ecological Status as the Basis for the Holistic Environmental Flow Assessment of a Tropical Highland River in Ethiopia
Previous Article in Journal
Kinetics and Mechanistic Studies of Photochemical and Oxidative Stability of Galaxolide
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Responses of Freshwater Diatoms and Macrophytes Rely on the Stressor Gradient Length across the River Systems

1
URBANZERO Institute for holistic environmental management, Ltd., Selo pri Mirni 17, 8233 Mirna, Slovenia
2
Institute for Water of the Republic of Slovenia, Einspielerjeva ulica 6, 1000 Ljubljana, Slovenia
3
Department of Biology, Biotechnical Faculty, University of Ljubljana, Jamnikarjeva 101, 1000 Ljubljana, Slovenia
*
Authors to whom correspondence should be addressed.
Water 2021, 13(13), 1814; https://doi.org/10.3390/w13131814
Submission received: 10 May 2021 / Revised: 26 June 2021 / Accepted: 27 June 2021 / Published: 30 June 2021
(This article belongs to the Special Issue Large Rivers: Ecology and Management in a Changing World)

Abstract

:
Phytobenthic diatoms and macrophyte communities respond differently to stressors in aquatic environments. For the assessment of the ecological status of rivers in Slovenia, we use several indices, including the River Macrophyte Index (RMI) and Trophic index (TI) based on macrophyte and phytobenthic diatoms communities, respectively. In the present study, we examined the relationships between nutrient variables and values of RMI and TI using varied stressor gradient lengths. We also aimed to explain the variability of macrophyte and diatom communities with different stressors, namely nutrients and land cover variables and their combinations. The relationships of RMI and TI with nutrient variables varied significantly and were affected by the length of the stressor gradient. We obtained a stronger relationship between the RMI and total phosphorous at an approximately <0.3-mg/L annual mean value, while, for the relationships with the TI, the values were significant at bigger gradient lengths. The greatest share of variability in the macrophyte and diatom community was explained by the combination of land use and nutrient variables and the lowest share by phosphorus and nitrogen variables. When we applied a composite stressor gradient, it explained a similar share of the variability of both macrophyte and diatom communities (up to 26%). A principal component analysis (PCA) based on land use and nutrient stressor gradient revealed that the relationship between RMI EQR and PCA1 that represents intensive agriculture depends on the length of the gradient. The relationship was stronger for shorter gradients at lower values and decreased as the gradient extended towards higher values. Both tested assessment methods showed that macrophyte communities are more sensitive to shorter stressor gradients of lower values, whereas diatom communities are more sensitive to longer stressor gradient and higher values of the stressor.

1. Introduction

Phytobenthos and macrophytes are primary producers and, thus, at the base of the food web in rivers. Phytobenthos is a crucial element of the periphytic community that colonises different submerged substrates in the photic zone of the majority of aquatic ecosystems, including rivers. The diversity of the phytobenthos community is directly related to the heterogeneity of the habitats in the water body [1,2]. Their occurrence and abundance also depend on other factors, such as the availability of light and nutrients, water temperature, changes in water levels, and water velocity [3,4]. Human activity can significantly alter these parameters, which can consequently affect the structure and function of phytobenthic communities [5]. After disturbance, the colonization of the substrate with diatoms, which are important elements of phytobenthos, occurs relatively quickly, while the complex community develops only over a longer period [6,7]. Phytobenthos has multiple functions. It stabilises the substrate and presents a habitat for other organisms. In many rivers, the contribution of organic carbon by phytobenthos photosynthesis is considerable. The short life cycle of phytobenthos enables a quick response to environmental changes; therefore, phytobenthos is a good indicator for assessing the ecological status of surface water bodies [8,9,10,11,12].
Macrophytes also present an important element of an aquatic ecosystem community. The presence and abundance of macrophytes in running waters depend primarily on the water quality, flow velocity, and substrate characteristics [13,14]. Moreover, they are involved in the energy flow, nutrient cycling, and sedimentation processes. When present in abundance, they affect ecosystems significantly [15]. They provide habitats, food, breeding locations, and refugia for aquatic invertebrates, fish, and a range of other organisms [16,17,18]. The degradation of macrophyte communities results in a degradation of the communities of dwelling organisms [19]. Macrophytes affect the water quality, and concurrently, they reflect the conditions in the water, in the riverbed, and in the interstitial water and sediment [20], which makes them valuable indicators of the water and sediment quality [21,22,23]. Macrophytes absorb nutrients through the root system from the sediment [24,25,26] and from the water column throughout their surface [27]. The proportion of uptake by the subterranean or aboveground parts of the plant depends on the specific adaptations of each species [28,29,30]. The knowledge about the structure of the macrophyte community is of great importance in assessing the ecological status of rivers [31].
Macrophytes and phytobenthos respond differently to nutrients and other stressors, since phytobenthos is more sensitive to nutrient enrichment while macrophytes better reflect the secondary effects of this enrichment [32]. However, the presence and abundance of representatives of the two groups reflect the trophic condition of the river, which is a measure of the autotrophic and heterotrophic bioproduction rates in the water body. The trophic status is one of the most important indicators of water quality [33,34,35]. Different trophic statuses are used as a stressor gradient in the development of indices for the assessment of the ecological status of waters in accordance with the Water Framework Directive [31,36,37]. Different studies have suggested that phosphorus and nitrogen are the most important nutrients regulating the trophic status in rivers [38]; however, the results are not consistent. Francoeur [39], who performed 237 nutrient enrichments on benthic algae in streams, found out that the algae in 16.5% of cases positively responded to N, 18.1% to P, 23.2% required N and P, 5% showed N or P inhibition, and 43% had no response to N or P.
For the evaluation of the ecological status of rivers in Slovenia on the basis of the biological element phytobenthos and macrophytes, we used the River Macrophytes Index (RMI), developed in 2007 [22,40] and supplemented in 2012, for intercalibration purposes at the European Union level [41] and the phytobenthic diatoms-based Trophic Index (TI) and Saprobic Index (SI), adapted for the ecological status assessment of Slovenian rivers from 2006 to 2008 [42,43] and supplemented in 2012 for intercalibration purposes at the European Union level [44]. Based on the RMI index, we primarily aimed to evaluate the trophic status and partly the general degradation of the river [22,40], as was the case in most of the methods evaluating the ecological status [32]. The phytobenthic diatoms-based TI index presumably reflects nutrients in the water (trophic state), whereas the SI index reflects the watercourse load with dissolved organic matter (saprobic state) [44,45,46,47]. Phytobenthos and macrophytes communities may respond differently to trophic conditions in the river. The response of macrophytes is relatively slow, while the phytobenthos respond quickly to changes of the trophic status. Therefore, both communities might reflect different aspects of the trophic status of the river ecosystem. However, rarely has it been tested whether the trophic stressor gradient length impacts the relationship between environmental variables and phytobenthos and macrophyte assemblage-based indices. The relations between trophic indices and nutrients are usually made on the basis of a few annual measurements of the nutrients, while the nutrient content in the water can vary greatly during the year [48]. The empirical research revealed relatively strong correlations between the nutrients and trophic indices based on phytobenthos; however, the correlations between nutrients and trophic indices based on macrophytes are rather weak [20,44,46,47]. This is possibly related to the length of the life cycle and growth form of both groups of primary producers. For macrophytes with a long life cycle and complex growth forms, the instantaneous chemical status of water could not be directly translated to the water quality during the macrophyte vegetation season. In addition, macrophytes are a very variable group regarding their growth forms, which also affects their response to various environmental factors [49,50,51], including water chemistry [52,53].
The aims of this study were (i) to examine the relationships between the nutrient variables and values of the River Macrophyte Index (RMI) and Trophic index (TI) using varied stressor gradient lengths and (ii) to explain the variability of the macrophyte and diatom communities with different stressors, namely nutrients and land cover variables and their combinations. We hypothesized (i) that TI will be related to nutrient concentrations in the water, while this will not be the case for RMI, (ii) that the relationships between stressor gradients and biotic indices will be affected by the length of the considered stressor gradients, and (iii) that a composite stressor gradient will explain a similar share of the variability of the macrophyte and phytobenthic communities.

2. Materials and Methods

2.1. Study Area and Sampling

Slovenia covers a total area of 20,273 km2 and features 4573 km of rivers with catchments larger than 10 km2. The rivers extend over four European ecoregions: Po lowland (3), Alps (4), Dinaric Western Balkan (5), and Pannonian lowland (11) [54,55]. However, macrophytes in rivers are common and abundant especially in the ecoregions Dinaric Western Balkan, Pannonian lowland, and Po lowland. Our experimental set comprised macrophytes and phytobenthic diatoms data of 70 samples, obtained at 42 sampling sites (Figure 1). Twenty-five sample sites were sampled once, 12 sample sites two times, 2 sample sites three times, and 3 sample sites were sampled five times. Half of the data were from Pannonian lowland, 32 from Dinaric Western Balkan, and 3 from Po Lowland. The sampling sites covered (near) natural-to-highly distorted conditions reflecting the various levels of perturbance caused by nutrients (phosphorous and nitrogen) and catchment land use.
The biological and nutrient (nitrogen and phosphorous) data originated from a national surface water monitoring programme in Slovenia conducted by the Slovenian Environment Agency between 2007 and 2016. Macrophytes (vascular plants, bryophytes, and charophytes) were sampled during the main vegetation period (from June to September). The macrophyte sampling procedure followed the standardised Slovenian river bioassessment protocol [56]. Sampling was carried out on a 100-m-long river stretch during the low-to-medium water levels. Depending on the sampling site characteristics, sampling was performed by wading across the channel, from the bank of the stream, or from a boat. Submerged, floating, and emergent plants were recorded. Species abundance was estimated as the relative plant biomass using a five-degree scale: 1 = very rare, 2 = rare, 3 = common, 4 = frequent, and 5 = abundant and predominant [57,58]. Macrophyte species presence and abundance were used to calculate the River Macrophyte Index (RMI) [22], which is used for the assessment of the river ecological status and classification of river water bodies in Slovenia [37,59]. The RMI was calculated using the following equation [22]:
R M I = i = 1 n A Q A i + 1 2 i = 1 n A B Q A B i 1 2 i = 1 n B C Q B C i i = 1 n C Q C i i = 1 n S Q S i
where:
  • QAi = abundance of the taxon i from the group A;
  • QABi = abundance of the taxon i from the group AB;
  • QBCi = abundance of the taxon i from the group BC;
  • QCi = abundance of the taxon i from the group C;
  • QSi = abundance of the taxon i from the groups A, AB, B, BC, and C;
  • nA = total number of taxa in group A;
  • nAB = total number of taxa in group AB;
  • nBC = total number of taxa in group BC;
  • nC = total number of taxa in group C;
  • nS = total number of taxa in groups A, AB, B, BC, and C.
The phytobenthos sampling procedure followed the standardised Slovenian river bioassessment protocol [56]. The sampling procedure is based on the micro-habitat sampling approach in which one sample consists of subsamples from different habitats defined as a combination of the substrate, flow type, water depth, and riparian vegetation shading. In accordance with this methodology, phytobenthos subsamples were scraped with a sharp subject from various living and non-living solid underwater surfaces (pebbles, stones, rocks, macrophytes, submerged wood, etc.) that were at least 1-m distant from the bank. Collected subsamples were transferred into a bottle and preserved in 4% formaldehyde or alcohol. The determination of phytobenthos was out carried in the laboratory, where permanent slides were prepared, and then, at least 500 phytobenthic diatoms per sample were counted and identified. The presence and abundance of diatoms was used as a proxy for the phytobenthos community. These data were then used to calculate the Trophic index (TI) [46], which was used to assess the trophic status of rivers in Slovenia [59,60]. The TI was calculated using the following equation [46]:
T I = i = 1 n T W i G i H i i = 1 n G i H i
where:
  • TI = Trophic index;
  • TWi = trophic value of the taxon i;
  • Gi = indicative weight of the taxon i;
  • Hi = abundance of the taxon i.
Water samples for the nitrogen and phosphorous variables analyses were obtained monthly or at least four times a year (each season). In our analyses, only those variables were considered where data were available for all the selected sites: total nitrogen (TN), ammonium (NH4), nitrate (NO3), nitrite (NO2), total phosphorus (TP), and orthophosphate (PO43−). For each variable, three variables were calculated: annual mean, annual maximum, and summer mean (Table 1).
Land use characteristics were obtained at the catchment and subcatchment scales. The subcatchment areas were defined as parts of the catchment area upstream of a sampling site to the confluence of an important tributary (for details, see Pavlin [61]). Eight land use variables were calculated on the catchment and subcatchment scales using Corine Land Cover (CLC) data [62]. We distinguished four categories of land use: natural and semi-natural areas (CLC classes 3, 4, and 5); urban areas (CLC class 1); areas characterised by extensive agricultural practices (CLC categories 2.3.1, 2.4.3, and 2.4.4); and areas characterised by intensive agricultural practices (CLC categories 2.1, 2.2, 2.4.1, and 2.4.2).

2.2. Data Analyses

At the first step, we determined the relationship between the nutrient variables and biotic indices: the ecological quality ratio (EQR) values of the River Macrophyte Index (RMI EQR) and the Trophic Index (TI EQR) using the whole dataset (n = 70) (for details on the analytical procedure, see Figure 2). All relationships were defined using a Spearman rank correlation coefficient. Further on, we examined the influence of the stressor gradient length expressed as the mean annual total phosphorous range (Figure 3) on the relationships between selected nutrient variables (ammonium maximum annual value and orthophosphate maximum annual value) and the biotic indices (RMI EQR and TI EQR). We determined the relationships using six different gradient lengths (Table 2). Due to the asymmetric distribution of the data, we determined the relationships based on the Spearman rank correlation. We used SPSS Statistics version 21.0(Armonk, NY, USA) to perform the analyses [63].
We also related the environmental variables with macrophyte and phytobenthic diatoms communities. Five groups of environmental variable stressor gradients were defined: nitrogen (N variables); phosphorous (P variables); land use (CLC variables); nutrients (N and P variables); and nutrients and land use (N, P, and CLC variables). Prior to the analyses, the phytobenthic diatoms data were transformed using the ln(x + 1) function, whereas the macrophyte data were not transformed. For the nutrient variables, we used log(x + 1) transformation to correct the heteroscedasticity in the data. Canonical Correspondence Analyses (CCAs) direct ordination techniques were carried out to analyse the relations between different groups of stressor gradients and the macrophyte and phytobenthic diatoms communities. Altogether, 14 CCAs were performed, seven with macrophytes and seven with phytobenthic diatoms communities (Table 3). Besides the differences in the two aquatic organism groups, the CCAs differed also in selected environmental variable group(s) and the forward selection procedure. In ten CCAs, an automatic forward selection procedure was used and manual in four CCAs. When a manual forward selection procedure was used, in the nutrient group, the phosphorous variables were selected first and then the nitrogen variables, whereas, in the nutrient and land use stressor group, the land use variables were selected first and then the nutrient variables. In the first step, we examined the contribution of each environmental variable to the explained variance of the species data (marginal effect), while, in the second step, we examined the remaining contribution of each variable added to the model when the other variables were already taken into account (conditional effects). Significant variables (p < 0.05) were selected using the Monte Carlo permutation test with 999 unrestricted permutations. These analyses were performed using the Canoco 5 software package [64].
In the last step, we built the composite stressor gradient. This was developed when running the Principal Component Analysis (PCA) in the Canoco 5 software package using the nutrient variables and CLC land cover variables [64].

3. Results

3.1. Relationships between Nutrient Variables and Biotic Indices

The relationships of the River Macrophyte Index (RMI EQR) and the Trophic Index (TI EQR) with the nutrient variables varied significantly. The TI EQR was negatively weakly to moderately statistically significantly correlated (p < 0.05) with several nutrient variables (r = −0.33 to −0.59), while the RMI EQR was correlated (p < 0.05) with the summer mean total nitrogen values (Table 4).
The relationships between the stressor gradients and biotic indices were affected by the length of the gradient. The strength of the relationship varied with the length of the gradient and the selected biotic index. For the variable orthophosphate—maximum value (PO4 max), we got a stronger relationship with the RMI index at 0.3 and at the 0.2-mg/L total phosphorous—annual mean value (TP mean) (Figure 4). The whole gradient provided much lower and nonsignificant values. When we examined the relationships with the Trophic Index (TI), the values were significant only at gradient lengths > 0.3 mg/L of the TP mean and highest at the whole gradient. For the variable ammonium—maximum value (NH4 max), the strength of the relationship also varied with the length of the gradient (Figure 5). The RMI EQR revealed no statistically significant relationships, while the TI EQR showed statistically significant relationships at gradient lengths ≥ 0.3 mg/L of the TP mean.

3.2. Relationships between Nutrient, Land Use Variables, and Biotic Communities

We tried to explain the variability of macrophytic and phytobenthic community structure by different simple and composite gradients, and we found out that the explained variabilities were very similar for the macrophytes and phytobenthic diatoms (Table 5 and Table 6 and Figure 6). When we examined the explanatory potential of a single variable, we found out that the most variability in the macrophyte community was explained by the variable NO2 mean. A somewhat lower share was explained by both the nitrogen (TN max, TN mean, NO3 max, NO2 max, and NO3 mean) and phosphorus (TP mean) variables and the CLC group variables (Intensive agriculture—catchment and Natural—catchment) (Table 2). For the phytobenthic diatoms community, the variability was best explained by the NO2 mean variable. The slightly lower share of variability was explained by the CLC variables (Intensive agriculture—catchment and Natural—catchment), nitrogen variables (TN mean, NO3 mean), and phosphorus variables (TP mean) (Table 6).
When we applied different stressor gradient groups, we explained the greatest share of variability in the macrophyte and phytobenthic communities by the LandNut stressor group (22–26%), the least by phosphorus variables (phytobenthic diatoms, 9.5%; macrophytes, 11.7%), or nitrogen variables (macrophytes, 11.7%) (Figure 6). Even though the gradient analyses revealed that land use and nutrients explain some of the variability of macrophyte communities, considering the data of the selected 70 samples, we did not find statistically significant correlations between the River Macrophyte Index (RMI) and nutrients. The gradient analysis revealed that the land use and nutrients groups explained different portions of the variability of macrophytes.

3.3. Composite Stressor Gradient

In the last step, we built the composite stressor gradient using a PCA. Along the first and second axes of the PCA plot, we obtained different land cover and nutrient gradients. However, the nutrient variables were poorly correlated with the first or second PCA axes (Spearman rank correlation < 0.5), which reflects the independent gradients in our data set. The first PCA axis represented mainly a gradient of intensive agriculture (mostly in the subcatchment), while the second axis represented a gradient of natural areas and urban areas (in opposite directions) in the subcatchment (Figure 7). The relationships between the RMI EQR values, TI EQR values, and the PCA1 and PCA2 varied and also depended on the length of the gradient. The RMI EQR values were statistically significantly correlated only with PCA1 up to 0.3 mg/L of the total phosphorous mean annual value, representing the intensive agriculture in the subcatchment. Stronger relationships were observed for shorter gradients of lower values and decreased as the gradients extended towards higher values (Figure 8). The TI EQR values did not show any statistically significant correlations with any PCA axes of any stressor gradient length.

4. Discussion

Nutrients are crucial for river biocenoses, supporting primary production, on one hand, while, on the other hand, their high concentrations compromise species relations within the community [65] and, thus, ecological status of the rivers. The response of macrophyte communities to nutrients differed significantly from that of diatom communities and so did the indices based on these two groups of organisms. This is possibly a consequence of the sampling time, since the macrophytes data used in our study were obtained at the peak growing season. The majority of studies relate to nutrients with macrophyte abundance only; however, we took into account species presence and abundance. For example, the study of O’Hare [66] revealed the increase of macrophyte biomass with P availability as a soluble reactive phosphorus and total phosphorus in water. The study of D’Aiuto [67] showed a significant correlation between stream nutrient enrichment (total phosphorus, soluble reactive phosphorus, and nitrate nitrogen) and macrophyte production. On the contrary, with epiphytic algae that obtains nutrients from water, an important aspect defining the response of the majority of macrophyte species to nutrients is also root uptake from the sediment. Carr [26] showed that, in streams, the macrophyte abundance increased linearly with increasing the sediment P concentration in water, while the study of Mebane [68] revealed that both the nitrogen contents in sediment and water were positively correlated to a macrophyte biomass. Thus, the response of the macrophyte community largely reflects the trophic conditions during their entire growth cycle and on the long-term accumulation of nutrients in the sediment, as shown by Clarke [28]. The absence of significant relationships of the RMI with the majority of the nutrient parameters is also a consequence of specific traits of species and communities, namely species life cycle, their growth dynamics, and their modes of nutrient uptake [69,70,71]. In the case of perennials with different underground storage organs, the impact of nutrients can be even more delayed and observed only over a long period [72]. An additional reason for the lack of correlation between the RMI with total phosphorus in our study could be the consequence of the relatively high total phosphorus concentrations in Slovenian watercourses, since, in our dataset, only 11% of the samples contained concentrations below 0.05 mg/L. The research of Fabris [73] applying the macrophyte Trophic Index (TIM) as a measure of the macrophyte community status revealed significant changes at relatively low concentrations of total phosphorus (up to 0.05 mg/L).
For the TI, we determined weak-to-moderate statistically significant correlations for the majority of the studied nutrient parameters. The significance of the correlations for the TI was expected due to the expansion mode of phytobenthos growth [11,74,75] and their quick response to nutrients [39].
The strength of the relationships between the RMI and TI and nutrients strongly depended on the length of the stressor gradient. The findings based on shorter stressor gradients with lower values (up to 0.3 mg/L of the TP mean) differed significantly from those based on longer ones. These inconsistent responses of the RMI relationships to nutrients may be a consequence of macrophytes that can function as a sink and a source of nutrients [76,77]. It was shown that the bulk of nutrients can be released from macrophyte beds during decay processes after disturbances and at the end of the vegetation period [78]. Besides that, macrophytes can also affect the nutrient balance by trapping fine organic and inorganic particles, enhancing the mineralization of organic matter through oxidation of the sediments, and altering the local environment, thus enabling P release due to the reducing conditions [79]. For the TI, statistically significant relationships with an annual mean total phosphorous value were obtained at gradient lengths ≥ 0.3 mg/L of the TP mean. For the variable maximum value of ammonium, the strength of the relationship also varied with the length of the gradient in the case of the RMI, but it showed an increase with the length in the case of the TI. Although some studies indicate a strong correlation between the stressor gradient and biotic indices/communities [80], the weak correlations obtained in our study were not against our expectations [61]. Weak correlations between the groups of stressor gradients (e.g., eutrophication) or individual stressor gradient (e.g., TP stressor gradient) and biotic indices confirmed that it is difficult to distinguish among the effects of individual stressor gradients or a group of stressor gradients on biotic communities in a multiple-stressor ecosystem. Moreover, different responses of biotic communities to the same stressor gradients are also possible [81]. Even in a case when we primarily assess the individual stressor gradient, there are always other events or influences that could change the conditions. For instance, removing riparian vegetation, which is a hydro-morphological element, affects the water temperature, nutrient cycling, and consequently, the nutrient uptake and trophic conditions for biotic communities [82,83]. In addition, different environmental factors may act at multiple scales [84].
The explained variability of macrophytic and phytobenthic assemblage structures by different stressor gradient groups was similar for macrophytes and phytobenthic diatoms. The greatest share of variability was explained by the land use and nutrients (LandNut) stressor group and the least by the phosphorus variables and nitrogen variables. Thus, the unexplained portion of variability was still prevalent. Different studies revealed that, even with indices that focus on individual stressors (e.g., trophic status), authors rarely explain >50% of the variability of an index [85]. However, in all these cases, one does not evaluate only the trophic status but also the general degradation of the riverine ecosystem. To comprise these complex effects, we built a composite gradient using a PCA. Along the first and second axes of the PCA plot, we obtained different land cover gradients. The relationship with the land cover variables was expected, since many studies showed that the land cover presented a key factor affecting the aquatic community [15,22,86]. The first axis represented mainly a gradient of intensive agriculture, while the second axis a gradient of natural areas and urban areas. The gradient vectors of natural areas and urban areas pointed in opposite directions. The relationships between the RMI EQR values, TI EQR values, and the PCA1 and PCA2 varied and were dependent on the gradient lengths. In the case of the RMI EQR values, stronger relationships were observed for shorter gradients of lower stressor values and decreased as the gradient extended towards higher values. We obtained a significant relationship with PCA1 at low mean annual values of the total phosphorus, which was related to intensive agriculture in the subcatchment. With the TI EQR values, no significant relationships were determined. This may be due to the long-term response of macrophytes to their habitat conditions, including those related to land use changes, which may result in continuous sediment accumulation and nutrient input [87], while phytobenthos mainly responded to short-term changes of the nutrients in the water [1,8].
The observed similarities in the explained variability of the macrophytic and phytobenthic assemblage structures by different stressor gradient groups (nutrients and land use) and differences in the indices performances might indicate the individual assessment index limitations. Each index was developed with a specific dataset and tested against selected stressor gradient data. Usually, the stressor gradient applied in studies is for necessity, a simplification of the true stressor gradient to which biota are exposed. Despite all the limitations, biological assemblages offer more stable information about the ecological status of a given ecosystem than abiotic variables, although we cannot always perfectly relate them. Macrophytes and phytobenthos assemblages reflect varied faces of the same river ecosystems.

5. Conclusions

The responses of phytobenthic diatoms and macrophyte communities to stressors in an aquatic environment differed significantly and so did the responses of the biotic indices that were developed based on these two groups. The relationships between the nutrient variables on the RMI and TI revealed the effect of the length of the stressor gradient, namely a stronger relationship between the RMI and annual mean total phosphorous was found at smaller gradient lengths in comparison to the TI. With different stressor gradient groups, we explained similar but relatively low shares of macrophyte and phytobenthic diatom community variabilities. The greatest share of variability in the macrophyte and phytobenthic diatom communities was explained by the combination of land use variables and nutrients (LandNut stressor group) and the least by phosphorus variables and nitrogen variables. The relationships of the RMI and TI to the composite stressor gradient, which comprised multiple variables shaping the phytobenthic diatoms and macrophyte communities, varied and were dependent on the gradient lengths. The RMI EQR values showed a significant relationship with PCA1 at low mean annual values of the total phosphorus, while, with the TI EQR values, we obtained no significant relationships. The used assessment methods showed that macrophytes respond better to shorter stressor gradients of lower values, whereas phytobenthic diatoms respond better to longer stressor gradients and higher values.

Author Contributions

Conceptualization, G.U.; methodology, G.U., A.G., B.D., M.G. and U.K.; investigation: G.U., B.D., A.G., M.G. and U.K.; validation: G.U., B.D. and A.G.; writing—original draft preparation G.U., B.D., A.G., M.G. and U.K.; writing—review and editing, G.U., B.D., A.G., M.G. and U.K.; visualization, G.U. and B.D.; supervision, G.U. and A.G.; project administration, G.U. and M.G.; and funding acquisition, G.U. and M.G. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Ministry of Environment and Spatial Planning of the Republic of Slovenia and Ministry of Higher Education and Science and Technology-Research Programs Biology of Plants (P1-0212).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data were stored within the documentation of Dopolnitev in posodobitev metodologije vrednotenja ekološkega stanja vodotokov na podlagi fitobentosa in makrofitov (naloga 4). Data are available on special request.

Acknowledgments

This study was supported by the Ministry of the Environment and Spatial Planning of the Republic of Slovenia and Slovenian Research Agency. We would like to thank the anonymous reviewers for their useful comments to a previous version of the manuscript.

Conflicts of Interest

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of the data; in the writing of the manuscript; or in the decision to publish the results.

References

  1. Schaumburg, J.; Schranz, C.; Foerster, J.; Gutowski, A.; Hofmann, G.; Meilinger, P.; Schneider, S.; Schmedtje, U. Ecological classification of macrophytes and phytobenthos for rivers in Germany according to the water framework directive. Limnologica 2004, 34, 283–301. [Google Scholar] [CrossRef] [Green Version]
  2. Spitale, D.; Scalfi, A.; Cantonati, M. Niche partitioning, shape of species response, and diversity in the phytobenthos across the rocky shoreline of a large peri-Alpine lake. J. Great Lakes Res. 2012, 38, 620–627. [Google Scholar] [CrossRef]
  3. Horner, R.R.; Welch, E.B.; Seeley, M.R.; Jacoby, J.M. Responses of periphyton to changes in current velocity, suspended sediment and phosphorus concentration. Freshw. Biol. 1990, 24, 215–232. [Google Scholar] [CrossRef]
  4. Tanabe, Y.; Ohtani, S.; Kasamatsu, N.; Fukuchi, M.; Kudoh, S. Photophysiological responses of phytobenthic communities to the strong light and UV in Antarctic shallow lakes. Polar Biol. 2010, 33, 85–100. [Google Scholar] [CrossRef]
  5. Spitale, D.; Scalfi, A.; Cantonati, M. Urbanization effects on shoreline phytobenthos: A multiscale approach at lake extent. Aquat. Sci. 2014, 76, 17–28. [Google Scholar] [CrossRef]
  6. Allan, J.D.; Castillo, M.M. Stream Ecology; Springer: Dordrecht, The Netherlands, 2007; p. 372. [Google Scholar]
  7. Graham, L.E.; Graham, J.M.; Graham, J.E.; Wilcox, L.W. Algae; Benjamin Cummings: San Francisco, CA, USA; University of Wisconsin: Madison, WI, USA, 2009; p. 616. [Google Scholar]
  8. Danilov, R.A.; Ekelund, N.G. The use of epiphyton and epilithon data as a base for calculating ecological indices in monitoring of eutrophication in lakes in central Sweden. Sci. Total Environ. 2000, 248, 63–70. [Google Scholar] [CrossRef]
  9. Lavoie, I.; Vincent, W.F.; Pienitz, R.; Painchaud, J. Benthic algae as bioindicators of agricultural pollution in the streams and rivers of southern Québec (Canada). Aquat. Ecosyst. Health Manag. 2004, 7, 43–58. [Google Scholar] [CrossRef]
  10. Hering, D.; Feld, C.K.; Moog, O.; Ofenböck, T. Cook book for the development of a Multimetric Index for biological condition of aquatic ecosystems: Experiences from the European AQEM and STAR projects and related initiatives. In The Ecological Status of European Rivers: Evaluation and Intercalibration of Assessment Methods; Springer: Dordrecht, The Netherlands, 2006; pp. 311–324. [Google Scholar]
  11. Townsend, S.A.; Garcia, E.A.; Douglas, M.M. The response of benthic algal biomass to nutrient addition over a range of current speeds in an oligotrophic river. Freshw. Sci. 2012, 31, 1233–1243. [Google Scholar] [CrossRef]
  12. Kelly, M.G.; Phillips, G.; Juggins, S.; Willby, N.J. Re-evaluating expectations for river phytobenthos assessment and understanding the relationship with macrophytes. Ecol. Indic. 2020, 117, 106582. [Google Scholar] [CrossRef]
  13. Franklin, P.; Dunbar, M.; Whitehead, P. Flow controls on lowland river macrophytes: A review. Sci. Total Environ. 2008, 400, 369–378. [Google Scholar] [CrossRef]
  14. Kuhar, U.; Germ, M.; Gaberščik, A. Macrophytes of watercourses in the Slovenian Danube Basin. In Macrophytes of Watercourses in the Slovenian Danube Basin; Academia: Praha, Czech Republic, 2018; pp. 234–251. [Google Scholar]
  15. Baattrup-Pedersen, A.; Riis, T. Macrophyte diversity and composition in relation to substratum characteristics in regulated and unregulated Danish streams. Freshw. Biol. 1999, 42, 375–385. [Google Scholar] [CrossRef]
  16. Madsen, J.D.; Chambers, P.A.; James, W.F.; Koch, E.W.; Westlake, D.F. The interaction between water movement, sediment dynamics and submersed macrophytes. Hydrobiologia 2001, 444, 71–84. [Google Scholar] [CrossRef]
  17. Thomaz, S.M.; Cunha, E.R.D. The role of macrophytes in habitat structuring in aquatic ecosystems: Methods of measurement, causes and consequences on animal assemblages’ composition and biodiversity. Acta Limnol. Bras. 2010, 22, 218–236. [Google Scholar] [CrossRef]
  18. Ambrožič, Š.; Gaberščik, A.; Vrezec, A.; Germ, M. Hydrophyte community structure affects the presence and abundance of the water beetle family Dytiscidae in water bodies along the Drava River. Ecol. Eng. 2018, 120, 397–404. [Google Scholar] [CrossRef]
  19. Hansen, J.P.; Wikström, S.A.; Axemar, H.; Kautsky, L. Distribution differences and active habitat choices of invertebrates between macrophytes of different morphological complexity. Aquat. Ecol. 2011, 45, 11–22. [Google Scholar] [CrossRef]
  20. Schneider, S.; Melzer, A. The Trophic Index of Macrophytes (TIM)—A New Tool for Indicating the Trophic State of Running Waters. Int. Rev. Hydrobiol. 2003, 88, 49–67. [Google Scholar] [CrossRef]
  21. Haslam, S.M. River Plants of Western Europe: The Macrophytic Vegetation of Watercourses of the European Economic Community; Cambridge University Press: Cambridge, UK, 1987; p. 512. [Google Scholar]
  22. Kuhar, U.; Germ, M.; Gaberščik, A.; Urbanič, G. Development of a River Macrophyte Index (RMI) for assessing river ecological status. Limnologica 2011, 41, 235–243. [Google Scholar] [CrossRef]
  23. Szoszkiewicz, K.; Budka, A.; Pietruczuk, K.; Kayzer, D.; Gebler, D. Is the macrophyte diversification along the trophic gradient distinct enough for river monitoring? Environ. Monit. Assess. 2017, 189, 4. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  24. Chambers, P.A.; Prepas, E.E.; Bothwell, M.L.; Hamilton, H.R. Roots versus Shoots in Nutrient Uptake by Aquatic Macrophytes in Flowing Waters. Can. J. Fish. Aquat. Sci. 1989, 46, 435–439. [Google Scholar] [CrossRef]
  25. Barko, J.W.; Gunnison, D.; Carpenter, S.R. Sediment interactions with submersed macrophyte growth and community dynamics. Aquat. Bot. 1991, 41, 41–65. [Google Scholar] [CrossRef]
  26. Carr, G.M. Macrophyte growth and sediment phosphorus and nitrogen in a Canadian prairie river. Freshw. Biol. 1998, 39, 525–536. [Google Scholar] [CrossRef]
  27. Robach, F.; Hajnsek, I.; Eglin, I.; Trémolières, M. Phosphorus sources for aquatic macrophytes in running waters: Water or sediment? Acta Bot. Gall. 1995, 142, 719–731. [Google Scholar] [CrossRef]
  28. Clarke, S.J. Vegetation growth in rivers: Influences upon sediment and nutrient dynamics. Prog. Phys. Geogr. Earth Environ. 2002, 26, 159–172. [Google Scholar] [CrossRef]
  29. Madsen, T.V.; Cedergreen, N. Sources of nutrients to rooted submerged macrophytes growing in a nutrient-rich stream. Freshw. Biol. 2002, 47, 283–291. [Google Scholar] [CrossRef]
  30. Janauer, G.A.; Gaberščik, A.; Kvet, J.; Germ, M.; Exler, N. Macrophytes of the River Danube Basin; Academia: Praha, Czech Republic, 2018; p. 408. [Google Scholar]
  31. EEA. European Waters—Assessment of Status and Pressures; European Environment Agency: Copenhagen, Denmark, 2012; p. 96. [Google Scholar]
  32. Poikane, S.; Salas Herrero, F.; Kelly, M.G.; Borja, A.; Birk, S.; van de Bund, W. European aquatic ecological assessment methods: A critical review of their sensitivity to key pressures. Sci. Total Environ. 2020, 740, 140075. [Google Scholar] [CrossRef]
  33. Rott, E.; Pipp, E.; Pfister, P. Diatom methods developed for river quality assessment in Austria and a cross-check against numerical trophic indication methods used in Europe. Algol. Stud. Für Hydrobiol. Suppl. Vol. 2003, 110, 91–115. [Google Scholar] [CrossRef]
  34. Lebkuecher, J.G.; Tuttle, E.N.; Johnson, J.L.; Willis, N.K.S. Use of algae to assess the trophic state of a stream in Middle Tennessee. J. Freshw. Ecol. 2015, 30, 349–376. [Google Scholar] [CrossRef] [Green Version]
  35. Turner, R.E.; Rabalais, N.N.; Justic, D.; Dortch, Q. Global patterns of dissolved N, P and Si in large rivers. Biogeochemistry 2003, 64, 297–317. [Google Scholar] [CrossRef]
  36. Dodkins, I.; Aguiar, F.; Rivaes, R.; Albuquerque, A.; Rodríguez-González, P.; Ferreira, M.T. Measuring ecological change of aquatic macrophytes in Mediterranean rivers. Limnologica 2012, 42, 95–107. [Google Scholar] [CrossRef]
  37. Aguiar, F.C.; Segurado, P.; Urbanič, G.; Cambra, J.; Chauvin, C.; Ciadamidaro, S.; Dörflinger, G.; Ferreira, J.; Germ, M.; Manolaki, P.; et al. Comparability of river quality assessment using macrophytes: A multi-step procedure to overcome biogeographical differences. Sci. Total Environ. 2014, 476–477, 757–767. [Google Scholar] [CrossRef] [PubMed]
  38. Dodds, W.K. Eutrophication and trophic state in rivers and streams. Limnol. Oceanogr. 2006, 51, 671–680. [Google Scholar] [CrossRef] [Green Version]
  39. Francoeur, S.N. Meta-analysis of lotic nutrient amendment experiments: Detecting and quantifying subtle responses. J. North Am. Benthol. Soc. 2001, 20, 358–368. [Google Scholar] [CrossRef]
  40. Germ, M.; Urbanič, G.; Gaberščik, A.; Kuhar, U.; Šiško, M. Prilagoditev Trofičnega Indeksa Zahtevam Vodne Direktive (Direktiva 2000/60/ES) Za Vrednotenje Ekološkega Stanja Rek V Sloveniji Na Podlagi Makrofitov: Končno Poročilo; Nacionalni inštitut za biologijo: Ljubljana, Slovenia, 2007. [Google Scholar]
  41. Urbanič, G.; Germ, M. River Macrophyte Index (RMI)—An Update on Index Application with New Boundary Values for Assessment of Rivers in Slovenia; Institute for Water of the Republic of Slovenia: Ljubljana, Slovenia, 2012; p. 18. [Google Scholar]
  42. Kosi, G.; Šiško, M.; Bricelj, M.; Urbanič, G.; Grbovič, J.; Rotar, B.; Stanič, K. Prilagoditev Trofičnega Indeksa Zahtevam Vodne Direktive (Direktiva 2000/60/ES) Za Vrednotenje Ekološkega Stanja Rek V Sloveniji Na Podlagi Fitobentosa; Nacionalni inštitut za biologijo: Ljubljana, Slovenia, 2006; p. 81. [Google Scholar]
  43. Kosi, G.; Šiško, M.; Bricelj, M.; Urbanič, G.; Grbovič, J.; Rotar, B.; Stanič, K. Prilagoditev Saprobnega Indeksa Zahtevam Vodne Direktive (Direktiva 2000/60/ES) Za Vrednotenje Ekološkega Stanja Rek V Sloveniji Na Podlagi Fitobentosa; Univerza v Ljubljani: Ljubljana, Slovenia, 2006; p. 83. [Google Scholar]
  44. Urbanič, G.; Kosi, G. Interkalibracija Metod Vrednotenja Ekološkega Stanja—Fitobentos. Poročilo O Delu Za Leto 2012; Inštitut za vode Republike Slovenije: Ljubljana, Slovenia, 2012; p. 32. [Google Scholar]
  45. Rott, E.; Hofmann, G.; Pall, K.; Pfister, P.; Pipp, E. Indikationslisten Für Aufwuchsalgen. Teil 1: Saprobielle Indikation; Bundesministerium für Land-und Forstwirtschaft: Wien, Austria, 1997; p. 73. [Google Scholar]
  46. Rott, E.; Pipp, E.; Pfister, P.; van Dam, H.; Ortler, K.; Binder, N.; Pall, K. Indikationslisten Für Aufwuchsalgen. Teil 2: Trophieindikation; Bundesministerium für Land-und Forstwirtschaft: Wien, Austria, 1999. [Google Scholar]
  47. Štupnikar, N.; Urbanič, G. Metodologija Vrednotenja Ekološkega Stanja S Podpornimi Splošnimi Fizikalno-Kemijskimi Elementi, Za Vrednotenje Stanja Hranil (Celotni Fosfor); Institute for Water of the Republic of Slovenia: Ljubljana, Slovenia, 2012; p. 46.
  48. Urbanič, G. Reka Ščavnica—Pomembna dobrina. In Sveta Ana Skozi Čas: Zbornik Občine Sveta Ana; Toš, M., Zemljič, I., Eds.; Ostroga: Maribor, Slovenia, 2009; pp. 29–46. [Google Scholar]
  49. Šraj-Kržič, N.; Germ, M.; Urbanc-Berčič, O.; Kuhar, U.; Janauer, G.A.; Gaberščik, A. The quality of the aquatic environment and macrophytes of karstic watercourses. Plant Ecol. 2007, 192, 107–118. [Google Scholar] [CrossRef]
  50. Klančnik, K.; Mlinar, M.; Gaberščik, A. Heterophylly results in a variety of “spectral signatures” in aquatic plant species. Aquat. Bot. 2012, 98, 20–26. [Google Scholar] [CrossRef]
  51. Klančnik, K.; Vogel-Mikuš, K.; Kelemen, M.; Vavpetič, P.; Pelicon, P.; Kump, P.; Jezeršek, D.; Gianoncelli, A.; Gaberščik, A. Leaf optical properties are affected by the location and type of deposited biominerals. J. Photochem. Photobiol. B Biol. 2014, 140, 276–285. [Google Scholar] [CrossRef]
  52. Granéli, W.; Solander, D. Influence of aquatic macrophytes on phosphorus cycling in lakes. Hydrobiologia 1988, 170, 245–266. [Google Scholar] [CrossRef]
  53. Lukács, B.A.; Dévai, G.; Tóthmérész, B. Aquatic macrophytes as bioindicators of water chemistry in nutrient rich backwaters along the Upper-Tisza river (in Hungary). Phytocoenologia 2009, 39, 287–293. [Google Scholar] [CrossRef]
  54. Illies, J. Limnofauna Europaea: Eine Zusammenstellung Aller Die Europäischen Binnengewässer Bewohnenden Mehrzelligen Tierarten mit Angaben Über Ihre Verbreitung Und Ökologie; Fischer: Stuttgart, Germany, 1978. [Google Scholar]
  55. Urbanic, G. Redelineation of European inland water ecoregions in Slovenia. Rev. Hydrobiol. 2008, 1, 17–25. [Google Scholar]
  56. Ministrstvo Za Okolje in Prostor. Metodologija Vrednotenja Ekološkega Stanja Vodotokov Na Podlagi Fitobentosa in Makrofitov; Ministrstvo za okolje in prostor: Ljubljana, Slovenia, 2016; p. 83. [Google Scholar]
  57. Kohler, A. Methoden der Kartierung von Flora und Vegetation von Süβwasserbiotopen. Landsch. Stadt 1978, 10, 73–85. [Google Scholar]
  58. Kohler, A.; Janauer, G.A. Zur Methodik der Untersuchung von aquatischen Makrophyten in Fließgewässern. In Handbuch Angewandte Limnologie; Steinberg, C., Bernhardt, H., Klapper, H., Eds.; Ecomed: Landsberg am Lech, Germany, 1995; Volume III-1.1.3, pp. 3–22. [Google Scholar]
  59. OGRS. Rules on Monitoring of Surface Water. In Pravilnik o Monitoringu Stanja Površinskih Voda; 2009; Volume 10, pp. 832–839. Available online: http://pisrs.si/Pis.web/pregledPredpisa?id=PRAV9315 (accessed on 15 April 2018).
  60. Almeida, S.F.P.; Elias, C.; Ferreira, J.; Tornés, E.; Puccinelli, C.; Delmas, F.; Dörflinger, G.; Urbanič, G.; Marcheggiani, S.; Rosebery, J.; et al. Water quality assessment of rivers using diatom metrics across Mediterranean Europe: A methods intercalibration exercise. Sci. Total Environ. 2014, 476, 768–776. [Google Scholar] [CrossRef]
  61. Pavlin, M.; Birk, S.; Hering, D.; Urbanič, G. The role of land use, nutrients, and other stressors in shaping benthic invertebrate assemblages in Slovenian rivers. Hydrobiologia 2011, 678, 137–153. [Google Scholar] [CrossRef]
  62. Corine Land Cover. 2012. Available online: https://land.copernicus.eu/pan-european/corine-land-cover/clc-2012 (accessed on 12 April 2018).
  63. IBM. IBM SPSS Statistics 21 Core System User’s Guide; International Business Machines Corporation: Armonk, NY, USA, 2012. [Google Scholar]
  64. ter Braak, C.J.F.; Šmilauer, P. Canoco Reference Manual and User’s Guide: Software for Ordination; Version 5.0; Microcomputer Power: Houston, TX, USA, 2012; p. 496. [Google Scholar]
  65. Miltner, R.J.; Rankin, A.E.T. Primary nutrients and the biotic integrity of rivers and streams. Freshw. Biol. 1998, 40, 145–158. [Google Scholar] [CrossRef] [Green Version]
  66. O’Hare, M.T.; Clarke, R.T.; Bowes, M.J.; Cailes, C.; Henville, P.; Bissett, N.; McGahey, C.; Neal, M. Eutrophication impacts on a river macrophyte. Aquat. Bot. 2010, 92, 173–178. [Google Scholar] [CrossRef]
  67. D’Aiuto, P.E.; Makarewicz, J.C.; Bosch, I. The impact of stream nutrient loading on macrophytes and metaphyton in Conesus Lake, USA. SIL Proc. 1922–2010 2006, 29, 1373–1376. [Google Scholar] [CrossRef]
  68. Mebane, C.A.; Simon, N.S.; Maret, T.R. Linking nutrient enrichment and streamflow to macrophytes in agricultural streams. Hydrobiologia 2014, 722, 143–158. [Google Scholar] [CrossRef]
  69. Hill, B.H. Uptake and release of nutrients by aquatic macrophytes. Aquat. Bot. 1979, 7, 87–93. [Google Scholar] [CrossRef]
  70. Bornette, G.; Puijalon, S. Response of aquatic plants to abiotic factors: A review. Aquat. Sci. 2011, 73, 1–14. [Google Scholar] [CrossRef]
  71. O’Hare, M.T.; Baattrup-Pedersen, A.; Baumgarte, I.; Freeman, A.; Gunn, I.D.M.; Lázár, A.N.; Sinclair, R.; Wade, A.J.; Bowes, M.J. Responses of Aquatic Plants to Eutrophication in Rivers: A Revised Conceptual Model. Front. Plant Sci. 2018, 9, 451. [Google Scholar] [CrossRef]
  72. Larcher, W. Physiological Plant Ecology, 4th ed.; Springer: Berlin, Germany, 2003; p. 513. [Google Scholar]
  73. Fabris, M.; Schneider, S.; Melzer, A. Macrophyte-based bioindication in rivers—A comparative evaluation of the reference index (RI) and the trophic index of macrophytes (TIM). Limnologica 2009, 39, 40–55. [Google Scholar] [CrossRef] [Green Version]
  74. Squires, M.M.; Lesack, L.F.W. Benthic algal response to pulsed versus distributed inputs of sediments and nutrients in a Mackenzie Delta lake. J. N. Am. Benthol. Soc. 2001, 20, 369–384. [Google Scholar] [CrossRef]
  75. Veraart, A.J.; Romaní, A.M.; Tornés, E.; Sabater, S. Algal response to nutrient enrichment in forested oligotrophic stream. J. Phycol. 2008, 44, 564–572. [Google Scholar] [CrossRef] [Green Version]
  76. Hutchinson, G.E. A Treatise on Limnology. Volume III. Limnological Botany; John Wiley & Sons: New York, NY, USA, 1975; p. 660. [Google Scholar]
  77. Haslam, S.M. River Plants: The Macrophytic Vegetation of Watercourses; Cambridge University Press: Cambridge, UK, 2006; p. 450. [Google Scholar]
  78. Barko, J.W.; James, W.F. Effects of Submerged Aquatic Macrophytes on Nutrient Dynamics, Sedimentation, and Resuspension. In The Structuring Role of Submerged Macrophytes in Lakes; Springer: New York, NY, USA, 1998; pp. 197–214. [Google Scholar]
  79. Thiébaut, G. Phosphorus and aquatic plants. In The Ecophysiology of Plant-Phosphorus Interactions; White, P.J., Hammond, J.P., Eds.; Springer: Dordrecht, The Netherlands, 2008; pp. 31–49. [Google Scholar]
  80. Sponseller, R.A.; Benfield, E.F.; Valett, H.M. Relationships between land use, spatial scale and stream macroinvertebrate communities. Freshw. Biol. 2001, 46, 1409–1424. [Google Scholar] [CrossRef] [Green Version]
  81. Kovalenko, K.E.; Brady, V.J.; Brown, T.N.; Ciborowski, J.J.H.; Danz, N.P.; Gathman, J.P.; Host, G.E.; Howe, R.W.; Johnson, L.B.; Niemi, G.J.; et al. Congruence of community thresholds in response to anthropogenic stress in Great Lakes coastal wetlands. Freshw. Sci. 2014, 33, 958–971. [Google Scholar] [CrossRef] [Green Version]
  82. Mackay, S.; Marsh, N. The Response of Aquatic Macrophytes to Riparian Shading in a Stream Rehabilitation Site; Monash University: Melbourne, Australia, 2005; p. 41. [Google Scholar]
  83. Giling, D.P.; Reich, P.; Thompson, R.M. Riparian vegetation removal alters consumer—Resource stoichiometry in an Australian lowland stream. Mar. Freshw. Res. 2012, 63, 1–8. [Google Scholar] [CrossRef]
  84. Brazner, J.C.; Danz, N.P.; Trebitz, A.S.; Niemi, G.J.; Regal, R.R.; Hollenhorst, T.; Host, G.E.; Reavie, E.D.; Brown, T.N.; Hanowski, J.A.M.; et al. Responsiveness of Great Lakes wetland indicators to human disturbances at multiple spatial scales: A multi-assemblage assessment. J. Great Lakes Res. 2007, 33, 42–66. [Google Scholar] [CrossRef]
  85. Birk, S.; Bonne, W.; Borja, A.; Brucet, S.; Courrat, A.; Poikane, S.; Solimini, A.; van de Bund, W.; Zampoukas, N.; Hering, D. Three hundred ways to assess Europe’s surface waters: An almost complete overview of biological methods to implement the Water Framework Directive. Ecol. Indic. 2012, 18, 31–41. [Google Scholar] [CrossRef]
  86. Strayer, D.L.; Lutz, C.; Malcom, H.M.; Munger, K.; Shaw, W.H. Invertebrate communities associated with a native (Vallisneria americana) and an alien (Trapa natans) macrophyte in a large river. Freshw. Biol. 2003, 48, 1938–1949. [Google Scholar] [CrossRef]
  87. Kuhar, U.; Gregorc, T.; Renčelj, M.; Šraj-Kržič, N.; Gaberščik, A. Distribution of macrophytes and condition of the physical environment of streams flowing through agricultural landscape in north-eastern Slovenia. Limnologica 2007, 37, 146–154. [Google Scholar] [CrossRef] [Green Version]
Figure 1. Location of the sampling sites (Ecoregions: 3—Po lowland, 4—Alps, 5—Dinaric Western Balkan, and 11—Pannonian lowland).
Figure 1. Location of the sampling sites (Ecoregions: 3—Po lowland, 4—Alps, 5—Dinaric Western Balkan, and 11—Pannonian lowland).
Water 13 01814 g001
Figure 2. Flowchart of the analytical procedure. CCA: Canonical Correspondence Analyses, FS: Forward selection, and PCA: Principal Component Analysis.
Figure 2. Flowchart of the analytical procedure. CCA: Canonical Correspondence Analyses, FS: Forward selection, and PCA: Principal Component Analysis.
Water 13 01814 g002
Figure 3. The distribution of the data within our dataset for the variable total phosphorous—annual mean (TP mean).
Figure 3. The distribution of the data within our dataset for the variable total phosphorous—annual mean (TP mean).
Water 13 01814 g003
Figure 4. Changes in the correlation coefficient (R) between the orthophosphate—maximum value (PO4 max) and River Macrophyte Index (RMI EQR) and Trophic Index (TI EQR), respectively, as a function of the stressor gradient length expressed as the total phosphorous—annual mean value (TP mean).
Figure 4. Changes in the correlation coefficient (R) between the orthophosphate—maximum value (PO4 max) and River Macrophyte Index (RMI EQR) and Trophic Index (TI EQR), respectively, as a function of the stressor gradient length expressed as the total phosphorous—annual mean value (TP mean).
Water 13 01814 g004
Figure 5. Changes in the correlation coefficient (R) between the ammonium—maximum values (NH4 max) and River Macrophyte Index (RMI EQR) and Trophic Index (TI EQR), respectively, as a function of the stressor gradient length expressed as the total phosphorous—annual mean value (TP mean).
Figure 5. Changes in the correlation coefficient (R) between the ammonium—maximum values (NH4 max) and River Macrophyte Index (RMI EQR) and Trophic Index (TI EQR), respectively, as a function of the stressor gradient length expressed as the total phosphorous—annual mean value (TP mean).
Water 13 01814 g005
Figure 6. Percentage of explained variance in the macrophytes and phytobenthic assemblages by the stressor gradient groups (LandNut, NutLand; P&N, Nut N and P Land; LandNut—land use and nutrients, NutLand—nutrients and land use, P&N—phosphorous and nitrogen, Nut—phosphorous and nitrogen, N—nitrogen, P—phosphorous, and Land—land use).
Figure 6. Percentage of explained variance in the macrophytes and phytobenthic assemblages by the stressor gradient groups (LandNut, NutLand; P&N, Nut N and P Land; LandNut—land use and nutrients, NutLand—nutrients and land use, P&N—phosphorous and nitrogen, Nut—phosphorous and nitrogen, N—nitrogen, P—phosphorous, and Land—land use).
Water 13 01814 g006
Figure 7. PCA ordination diagram of the first two ordination axes with the land use and nutrient variables (n = 70). Environmental variables correlated <0.5 with PCA1 or PCA2 were omitted. Urb—urban, Nat—natural, IAg—intensive agriculture, Cat—catchment, and SCat—subcatchment. See Table 1 for the environmental variable codes.
Figure 7. PCA ordination diagram of the first two ordination axes with the land use and nutrient variables (n = 70). Environmental variables correlated <0.5 with PCA1 or PCA2 were omitted. Urb—urban, Nat—natural, IAg—intensive agriculture, Cat—catchment, and SCat—subcatchment. See Table 1 for the environmental variable codes.
Water 13 01814 g007
Figure 8. Changes in the correlation coefficient between the first and second axes of the Principal Component Analysis (PCA) and the River Macrophyte Index (RMI EQR) and Trophic index (TI EQR), respectively, as a function of the stressor gradient length expressed as the total phosphorous—annual mean (TP mean).
Figure 8. Changes in the correlation coefficient between the first and second axes of the Principal Component Analysis (PCA) and the River Macrophyte Index (RMI EQR) and Trophic index (TI EQR), respectively, as a function of the stressor gradient length expressed as the total phosphorous—annual mean (TP mean).
Water 13 01814 g008
Table 1. Mean, median, minimum, and maximum values of the selected nutrient and Corine Land Cover (CLC) variables.
Table 1. Mean, median, minimum, and maximum values of the selected nutrient and Corine Land Cover (CLC) variables.
VariableCodeMeanMedianMinimumMaximum
Total nitrogen—maximumTN max2.41.80.77.7
Total nitrogen—annual meanTN mean1.71.40.56.5
Total nitrogen—summer meanTN Smean1.61.20.47.5
Ammonium—maximumNH4 max0.30.10.03.2
Ammonium—annual meanNH4 mean0.10.10.01.5
Ammonium—summer meanNH4 Smean0.10.00.03.2
Nitrite—maximumNO2 max0.10.10.00.8
Nitrite—annual meanNO2 mean0.10.00.00.3
Nitrite—summer meanNO2 Smean0.10.00.00.8
Nitrate—maximumNO3 max8.66.72.232.0
Nitrate—annual meanNO3 mean6.15.01.226.7
Nitrate—summer meanNO3 Smean4.84.00.528.6
Total phosphorous—maximumTP max0.50.30.13.1
Total phosphorous—annual meanTP mean0.30.20.01.1
Total phosphorous—summer meanTP Smean0.30.20.01.8
Orthophosphate—maximumPO4 max0.20.10.02.0
Orthophosphate—annual meanPO4 mean0.10.10.00.8
Orthophosphate—summer meanPO4 Smean0.10.10.01.1
CLC Natural—subcatchmentNat SCat34.135.10.095.4
CLC Natural—catchmentNat Cat57.964.126.293.2
CLC Intensive agriculture—subcatchmentIAg SCat40.739.40.0100.0
CLC Intensive agriculture—catchmentIAg Cat26.218.73.358.8
CLC Extensive agriculture—subcatchmentEAg SCat11.99.20.039.0
CLC Extensive agriculture—catchmentEAg Cat13.511.92.730.0
CLC Urban—subcatchmentUrb SCat13.20.20.098.2
CLC Urban—catchmentUrb Cat2.52.80.07.0
Table 2. Total phosphorous gradient length given as the total phosphorous—annual mean (TP mean) and number of samples (N). The minimum value was, in all cases, 0.03 mg/L.
Table 2. Total phosphorous gradient length given as the total phosphorous—annual mean (TP mean) and number of samples (N). The minimum value was, in all cases, 0.03 mg/L.
TP Mean (mg/L)N
0.129
0.236
0.340
0.450
0.560
1.170
Table 3. Canonical Correspondence Analyses (CCA) option codes performed with macrophytes and phytobenthic diatoms assemblages and used with the environmental variable stressor groups and forward selection options.
Table 3. Canonical Correspondence Analyses (CCA) option codes performed with macrophytes and phytobenthic diatoms assemblages and used with the environmental variable stressor groups and forward selection options.
CCA Option CodeStressor GroupForward Selection Option
NNitrogenautomatic
PPhosphorousautomatic
NutNitrogen and Phosphorousautomatic
P&NNitrogen and Phosphorousmanual
LandLand Useautomatic
NutLandNutrients and Land useautomatic
LandNutNutrients and Land usemanual
Table 4. The relationship (Spearman rank correlation coefficients) between the nutrient variables and the ecological quality ratio of the River Macrophyte Index (RMI EQR) and the Trophic Index (TI EQR); n = 70. Statistically significant relationships are marked **—p < 0.01 and *—p < 0.05.
Table 4. The relationship (Spearman rank correlation coefficients) between the nutrient variables and the ecological quality ratio of the River Macrophyte Index (RMI EQR) and the Trophic Index (TI EQR); n = 70. Statistically significant relationships are marked **—p < 0.01 and *—p < 0.05.
VariableRMI EQRTI EQR
Total nitrogen—maximum0.21−0.51 **
Total nitrogen—annual mean0.19−0.49 **
Total nitrogen—summer mean0.30 *−0.40 **
Ammonium—maximum0.16−0.50 **
Ammonium—annual mean0.15−0.57 **
Ammonium—summer mean0.09−0.51 **
Nitrite—maximum−0.07−0.49 **
Nitrite—annual mean−0.03−0.58 **
Nitrite—summer mean−0.06−0.59 **
Nitrate—maximum0.09−0.33 **
Nitrate—annual mean0.07−0.37 **
Nitrate—summer mean0.06−0.22
Total phosphorous—maximum−0.03−0.40 **
Total phosphorous—annual mean0.05−0.40 **
Total phosphorous—summer mean0.13−0.40 **
Orthophosphate—maximum−0.07−0.34 **
Orthophosphate—annual mean−0.01−0.37 **
Orthophosphate—summer mean0.06−0.37 **
Table 5. Percentage of macrophyte assemblage variance explained by each environmental variable before forward selection (λ1) and after forward selection (λa) within each stressor gradient group. Values of λa are given only for statistically significant explanatory variables (p < 0,05). LandNut—land use and nutrients, NutLand—nutrients and land use, P&N—phosphorus and nitrogen, Nut—phosphorous and nitrogen, N—nitrogen, P—phosphorous, and Land—land use.
Table 5. Percentage of macrophyte assemblage variance explained by each environmental variable before forward selection (λ1) and after forward selection (λa) within each stressor gradient group. Values of λa are given only for statistically significant explanatory variables (p < 0,05). LandNut—land use and nutrients, NutLand—nutrients and land use, P&N—phosphorus and nitrogen, Nut—phosphorous and nitrogen, N—nitrogen, P—phosphorous, and Land—land use.
Stressor Gradient
Environmental VariableVariable Groupλ1 (%)p-ValueLandNut NutLandP&NNutNPLand
λa (%)p-Valueλa (%)p-Valueλa (%)p-Valueλa (%)p-Valueλa (%)p-Valueλa (%)p-Valueλa (%)p-Value
Nitrite—annual meanNutrients5.30.00111.80.00111.80.0017.30.00115.20.00120.30.001
Total nitrogen—maximumNutrients4.90.001 7.80.0016.80.00110.10.00113.50.001
Total nitrogen—annual meanNutrients4.90.001
Nitrate—maximumNutrients4.80.001
Nitrite—maximumNutrients4.70.0014.10.0364.20.0335.30.0345.40.037.30.026
Nitrate—annual meanNutrients4.70.0016.60.0016.50.0016.80.0078.30.00111.10.001
CLC Intensive agriculture—catchmentLand Use4.60.00180.0014.30.009 30.30.001
Total phosphorous—annual meanNutrients4.60.001 130.001 32.40.001
CLC Natural—catchmentLand Use4.50.001
Orthophosphate—maximumNutrients4.20.0014.30.025 8.30.001 20.70.001
Total phosphorous—summer meanNutrients4.20.0014.10.031 5.80.012 14.60.018
Orthophosphate—annual meanNutrients40.001
Ammonium—annual meanNutrients40.0014.70.0115.30.00160.0076.80.0019.20.001
Total phosphorous—maximumNutrients3.80.001 6.30.004 15.70.007
CLC Urban—catchmentLand Use3.70.0015.10.0024.50.013 21.60.001
Orthophosphate—summer meanNutrients3.50.001
Nitrite—summer meanNutrients3.20.001
Total nitrogen—summer meanNutrients3.20.001
Ammonium—maximumNutrients30.001
CLC Intensive agriculture—subcatchmentLand Use2.80.00140.042 12.60.037
CLC Natural—subcatchmentLand Use2.60.00550.0054.70.011 17.60.001
Nitrate—summer meanNutrients2.40.017
Ammonium—summer meanNutrients2.30.011
CLC Urban—subcatchmentLand Use2.10.065
CLC Extensive agriculture—catchmentLand Use1.80.102
CLC Extensive agriculture—subcatchmentLand Use1.80.129
Table 6. Percentage of the phytobenthic assemblage variance explained by each environmental variable before forward selection (λ1) and after forward selection (λa) within each stressor gradient group. Values of λa are given only for variables with statistically significant explanatory variables (p < 0,05). LandNut—land use and nutrients, NutLand—nutrients and land use, P&N—phosphorous and nitrogen, Nut—phosphorous and nitrogen, N—nitrogen, P—phosphorous, and Land—land use.
Table 6. Percentage of the phytobenthic assemblage variance explained by each environmental variable before forward selection (λ1) and after forward selection (λa) within each stressor gradient group. Values of λa are given only for variables with statistically significant explanatory variables (p < 0,05). LandNut—land use and nutrients, NutLand—nutrients and land use, P&N—phosphorous and nitrogen, Nut—phosphorous and nitrogen, N—nitrogen, P—phosphorous, and Land—land use.
Environmental VariableVariable GroupStressor Gradient
λ1 (%)p-ValueLandNut NutLandP&NNutNP Land
λa (%)p-Valueλa (%)p-Valueλa (%)p-Valueλa (%)p-Valueλa (%)p-Valueλa (%)p-Valueλa (%)p-Value
Nitrite—annual meanNutrients6.50.00114.80.00114.80.00110.00.00118.90.001250.001
CLC Intensive agriculture—catchmentLand Use5.90.00110.60.0016.00.001 38.20.001
CLC Natural—catchmentLand Use5.90.0014.40.0104.50.002
Total nitrogen—annual meanNutrients5.80.001
Total phosphorous—annual meanNutrients5.80.0014.40.0097.10.00116.90.0019.00.001 43.70.001
Nitrate—annual meanNutrients5.60.0015.20.0014.30.0164.90.0485.50.01311.70.001
Total nitrogen—maximumNutrients5.40.001 9.00.001
Nitrate—maximumNutrients5.20.0013.90.0185.80.0017.30.0017.40.0018.30.001
Nitrite—maximumNutrients5.20.001
Total phosphorous—summer meanNutrients5.10.001
Ammonium—annual meanNutrients5.00.001
Orthophosphate—annual meanNutrients4.90.001 5.40.028 14.10.030
Orthophosphate—summer meanNutrients4.70.001
Total phosphorous—maximumNutrients4.60.0013.70.0363.80.0385.30.0364.90.043 13.70.033
Orthophosphate—maximumNutrients4.40.001
Total nitrogen—summer meanNutrients4.10.001
Nitrite—summer meanNutrients4.10.001
Ammonium—maximumNutrients4.00.001
CLC Urban—catchmentLand Use3.70.001 21.10.001
Nitrate—summer meanNutrients3.50.0014.30.0154.90.0036.30.0026.30.0037.50.01
Ammonium—summer meanNutrients2.30.007
CLC Extensive agriculture—catchmentLand Use2.10.027
CLC Intensive agriculture—subcatchmentLand Use1.90.053.90.0283.90.017 13.10.010
CLC Natural—subcatchmentLand Use1.80.089
CLC Extensive agriculture—subcatchmentLand Use1.50.319
CLC Urban—catchmentLand Use1.40.453
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Urbanič, G.; Debeljak, B.; Kuhar, U.; Germ, M.; Gaberščik, A. Responses of Freshwater Diatoms and Macrophytes Rely on the Stressor Gradient Length across the River Systems. Water 2021, 13, 1814. https://doi.org/10.3390/w13131814

AMA Style

Urbanič G, Debeljak B, Kuhar U, Germ M, Gaberščik A. Responses of Freshwater Diatoms and Macrophytes Rely on the Stressor Gradient Length across the River Systems. Water. 2021; 13(13):1814. https://doi.org/10.3390/w13131814

Chicago/Turabian Style

Urbanič, Gorazd, Barbara Debeljak, Urška Kuhar, Mateja Germ, and Alenka Gaberščik. 2021. "Responses of Freshwater Diatoms and Macrophytes Rely on the Stressor Gradient Length across the River Systems" Water 13, no. 13: 1814. https://doi.org/10.3390/w13131814

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop