Next Article in Journal
Determination of Equilibrium Loading by Empirical Models for the Modeling of Breakthrough Curves in a Fixed-Bed Column: From Experience to Practice
Previous Article in Journal
Relationship Between Urbanization–Induced Land Use Changes and Flood Risk: Case Study in Chiang Mai, Thailand
Previous Article in Special Issue
Effects of War-Related Human Activities on Microalgae and Macrophytes in Freshwater Ecosystems: A Case Study of the Irpin River Basin, Ukraine
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Water
Volume 17
Issue 3
10.3390/w17030328
water-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Assessing Macrophyte Diversity in the Danube River: Comparing the Effectiveness of Different Sampling Procedures

by
Dragana Vukov
*,
Mirjana Ćuk
,
Nataša Nikolić
and
Miloš Ilić
Department of Biology and Ecology, Faculty of Sciences, University of Novi Sad, Trg Dositeja Obradovića 2, 21000 Novi Sad, Serbia
*
Author to whom correspondence should be addressed.
Water 2025, 17(3), 328; https://doi.org/10.3390/w17030328
Submission received: 20 November 2024 / Revised: 8 January 2025 / Accepted: 10 January 2025 / Published: 24 January 2025
(This article belongs to the Special Issue Biodiversity of Freshwater Ecosystems: Monitoring and Conservation)
Browse Figures
Versions Notes

Abstract

:
Accurate assessment of macrophyte diversity is essential for effective river ecosystem management. Discrepancies in sampling protocols can lead to variations in observed biodiversity, which may influence ecological monitoring and management strategies. This study evaluates and compares three sampling methods—the comprehensive “all” survey, Joint Danube Survey (JDS), and National Monitoring Program (NMP)—for assessing macrophyte diversity along the Serbian Danube. We applied Hill numbers and Chao estimators to evaluate sample completeness and diversity for species richness (q = 0), Shannon (q = 1), and Simpson (q = 2) diversity. Asymptotic diversity was estimated using rarefaction and extrapolation methods, and statistical analyses (ANOVA, permutation tests) assessed differences in diversity estimates across sampling methods. The “all” sampling method provided the highest sample completeness and reliable asymptotic diversity estimates, capturing 100% completeness for q = 1 and q = 2. The JDS and NMP datasets showed incomplete sampling for species richness (q = 0), with undetected species richness in both. However, both datasets captured all abundant species for higher diversity orders. Significant differences in diversity estimates between methods were found in some waterbodies, especially for species richness and Shannon diversity. The “all” sampling method most accurately captures true species richness and diversity. While JDS and NMP methods are valuable for assessing higher-order diversity, the two methods may underestimate species richness, particularly in heterogeneous river sections.

1. Introduction

Freshwater ecosystems, particularly rivers and their associated macrophyte communities, are vital components of biodiversity. Macrophytes are integral to the ecological functioning of rivers, providing habitat for aquatic fauna, stabilizing sediment, contributing to nutrient cycling, and influencing hydrological dynamics [1,2,3]. In rivers such as the Danube, which spans multiple countries and ecosystems, understanding the macrophyte community composition and diversity is crucial for effective ecological management and conservation strategies [4,5]. However, obtaining accurate and comparable diversity estimates across large and ecologically complex systems, such as the Danube River, remains challenging due to variations in sampling methodologies [6,7,8,9,10,11].
Sampling strategies designed to quantify species diversity vary widely in their scope, temporal coverage, and sensitivity to detect both common and rare species [12]. Standardized monitoring programs employ systematic sampling protocols, but their ability to fully capture the diversity of aquatic macrophytes across the entire river reach or in specific waterbodies is often unclear [9,13]. In contrast, more comprehensive survey approaches, which incorporate broader spatial and temporal scales, may better capture diversity but are more resource-intensive and difficult to implement over large areas [12].
While methods that include diversity metrics such as species richness, Shannon, and Simpson diversity are commonly used to assess the overall health and functioning of aquatic systems, these indices may provide misleading results if sampling completeness is insufficient [12,13,14,15,16,17]. Studies that incorporate rarefaction and extrapolation methods, including Hill numbers and Chao estimators, are increasingly used to assess the completeness of sampling efforts and estimate asymptotic diversity [9,14,16,17,18,19,20]. These approaches are particularly important when comparing multiple datasets derived from different sampling strategies, as they allow for more reliable inferences about true species richness and diversity.
Despite the growing body of work on sampling completeness and diversity estimation, there is still a significant gap in the comparative evaluation of common macrophyte sampling protocols in large river systems [20,21,22]. The lack of standardized methods for assessing diversity across spatially and temporally heterogeneous river systems complicates the ability to detect ecological changes or assess long-term trends in macrophyte populations [11,12,19,23,24]. Furthermore, the impact of sampling procedure on the detected diversity and the resulting ecological inferences remains understudied, particularly for large, complex systems like the Danube River.
This study aims to assess the effectiveness of three macrophyte sampling strategies—a comprehensive survey (“all”), the standardized Joint Danube Survey (“JDS”), and the national monitoring program (“NMP”)—in estimating macrophyte diversity along the Serbian stretch of the Danube River. The specific objectives of the study are as follows: 1. To quantify sampling completeness for each procedure and assess its impact on observed and inferred diversity, using various diversity orders across different waterbodies and reaches of the river. 2. To evaluate the performance of each sampling method in estimating true species richness and diversity, using rarefaction and extrapolation methods. 3. To analyze the influence of sampling procedure on ecological assessments, and to identify potential discrepancies between sampling methods. 4. To compare the spatial representativeness of each sampling strategy, and assess the sensitivity of each method to detect rare or highly abundant species across different river reaches.
We hypothesize the following: 1. The comprehensive “all” dataset will exhibit the highest sample completeness and asymptotic diversity estimates, as it encompasses a broad spatial scale, including both abundant and rare species [12,15,17,24]. Conversely, the “JDS” and “NMP” datasets will show incomplete sampling profiles, particularly for rare or infrequently detected species. 2. The “JDS” and “NMP” sampling strategies, due to their limited coverage or less frequent sampling intervals, will underestimate species richness and diversity, especially in spatially heterogeneous or ecologically complex areas [17,18,20]. 3. Significant differences in diversity estimates between sampling methods will result in variability in ecological assessments, which may influence management and conservation decisions, particularly for regions with high species turnover or seasonal variability [6,9,19,23]. 4. Discrepancies between empirical and asymptotic diversity estimates (e.g., q = 0 for species richness and q = 1, q = 2 for Shannon and Simpson diversities) will highlight the limitations of each sampling approach in capturing the full range of macrophyte diversity [14,15,17,25].
By comparing different macrophyte sampling strategies in terms of their completeness, diversity estimates, and ecological implications, this study provides critical insights into the methodology used in large river biodiversity assessments. The integration of advanced statistical frameworks such as rarefaction, extrapolation, and multivariate analysis enables a robust comparison of how different protocols influence ecological inferences. This research contributes to the broader scientific discourse on optimizing sampling methods for large river systems, offering guidance for future monitoring programs and ecological assessments in large rivers such as the Danube.
Moreover, the findings of this study will be invaluable for improving river management practices, as they emphasize the importance of methodological rigor in detecting true biodiversity and guiding conservation strategies. In particular, this work underscores the need for harmonizing sampling protocols to enhance the reliability and comparability of ecological assessments across large-scale aquatic ecosystems [13,20,22,23,26].

2. Materials and Methods

2.1. Study Area

Original data were collected during a macrophyte survey of the Danube River in Serbia during the summers of 2014 and 2015. The survey covered the river’s main channel from river kilometer (rkm) 1433 to rkm 846 and was conducted separately on each riverside, except in transboundary sections of the river.
According to national legislation (details available at www.sepa.gov.rs accessed on 4 December 2024), the Danube in Serbia belongs to the Running Water Type 1—a large lowland river dominated by fine sediment. For ecological assessment and monitoring purposes, the Danube in Serbia is divided into ten contiguous waterbodies (Figure 1, Table 1).

2.2. Sampling Procedures

In this study, we employed three distinct sampling approaches used for macrophyte surveys on the Serbian Danube: the comprehensive “all” dataset, the “JDS” dataset, and the “NMP” dataset, each characterized by differing spatial coverage and sampling frequency. These methods were designed to test the hypotheses outlined in Section 1, where we proposed that the “all” dataset would exhibit superior sample completeness and diversity estimates due to its broader spatial scale, while the “JDS” and “NMP” datasets would reflect incomplete sampling profiles, particularly for rare species (see Section 1, Paragraph 6). The anticipated discrepancies in diversity estimates and sampling limitations were specifically assessed to evaluate their implications for ecological and conservation analyses. Original data were collected in summers of 2014 and 2015, using a standard survey procedure [27] (CEN-EN 14184, 2014), covering the entire stretch of the Danube in Serbia. The survey required recording of submerged, floating, and emergent aquatic plants growing in water during the mean water levels. The river was divided into contiguous longitudinal transects—survey units (SUs)—each one kilometer in length. In each SU, macrophyte species were recorded zigzagging along the SU, and their abundances were estimated using a five-level scale (1—rare; 2—occasional; 3—frequent; 4—abundant; 5—very abundant) [28]. Each sampling unit (SU) was surveyed once during the macrophyte survey campaign. Macrophytes were surveyed from a small boat, and in shallows on foot. The submerged species were sampled using rakes. The observed species were mainly identified in the field, while for some species it was necessary to identify the collected material in the laboratory. The collected plant material was deposited in the Aquatic Collection of the University of Novi Sad Herbarium (BUNS). In total, 51 plant species were recorded. The original dataset was used as the “all” sampling procedure (Table 1B).
From this original dataset, data corresponding to two additional sampling procedures were extracted:
  • JDS Sampling Procedure: Locations were based on the JDS3 (ICPDR Joint Danube Survey 3) sampling sites (www.danubesurvey.org/jds3 accessed on 4 December 2024). While the JDS methodology recommends surveying macrophytes in three contiguous river kilometers on each riverside, we expanded this to five SUs (Table 1B). This subset formed the “JDS” dataset.
  • NMP Sampling Procedure: Locations were based on the National Monitoring Program (NMP) conducted by the Serbian Environmental Agency (www.sepa.gov.rs accessed on 4 December 2024). We considered each sampling site as five contiguous SUs on each riverside (Table 1B). This subset formed the “NMP” dataset.

2.3. Data Analyses

2.3.1. Quantification of Sample Completeness and Comparison of Diversity Among Assemblages

As suggested by [29], plant abundance data were transformed into quantitative estimates (abundance3 = quantity), serving as proxies for the number of individuals recorded per species in each SU.
To evaluate the completeness of macrophyte surveys and the influence of sampling procedures on observed macrophyte diversity, we applied the framework proposed by [17], based on Hill numbers and Chao estimators. Hill numbers integrate species richness and abundance into a single class of diversity measures, defined as the effective number of equally abundant species. These measures are parameterized by diversity order q. Hill numbers for order q ≥ 0 include the three most widely used species diversity measures: q = 0—Species richness; q = 1—Shannon diversity; q = 2: Simpson diversity.
We assessed and compared sample completeness across the “all”, “JDS” and “NMP” datasets. The Chao framework [17] provides completeness measures parameterized by q ≥ 0, weighting species and individuals differently depending on q. For the q = 1, each species is proportionally weighted by its abundance. For q = 2, each species is proportionally weighted by its squared species abundance; therefore, this measure is disproportionally sensitive to highly abundant species. A continuous profile of completeness measures was used to visualize and compare the completeness of different assemblages. Confidence intervals (95%) were calculated using a bootstrap method.
To estimate asymptotic diversities, we applied the framework by [25], extended by [17]. Traditional methods rely on rarefaction to standardize sample sizes across assemblages, but this discards data. Instead, we used size-based rarefaction and extrapolation curves, as described by [15,16], to estimate richness and diversity profiles. The visual inspection of the size-based rarefaction and extrapolation sampling curves indicate whether the asymptotic estimates could be used to infer the true diversity of the assemblage. If the curve stays at a fixed level (often for q = 1, and q = 2), asymptotic estimates infer the true diversity of an assemblage, otherwise the asymptotic estimate represents only a lower bound. When the true diversity can be accurately inferred, the extent of undetected diversity within each dataset can be assessed by comparing the estimated asymptotic diversity profile with the observed profile. Confidence intervals (95%) were calculated using a bootstrap method.
Analyses were conducted using the “iNEXT.4steps” package [17] in R (version 4.4.2) for the entire Danube stretch and for individual waterbodies.

2.3.2. Evaluation of Sampling Procedures’ Effectiveness in Capturing Community Structure

To evaluate the effectiveness of each sampling procedure in capturing macrophyte community structure, we analyzed multivariate homogeneity of group dispersions (variance) for the “all”, “JDS” and “NMP” datasets using the “betadisper” function from the R package “vegan” [30]. Hellinger-transformed Kohler abundances were used for the analyses. Group variance was tested using ANOVA and pairwise permutation tests (999 permutations), followed by Tukey’s Honest Significant Difference test. The applied significance levels were p < 0.001—***, p < 0.01—**, p < 0.05—*.

2.3.3. Evaluation of Sampling Procedures’ Impact on Ecological Status Assessments

The Serbian Environmental Agency monitors surface waterbodies and assesses their ecological status using the reference index (RI) proposed by [29]. While our datasets lacked the helophyte data needed for full ecological assessments, RI values were used as proxies.
RI values were calculated from transformed abundance data (abundance3 = quantity) and compared across “all”, “JDS” and “NMP” datasets using Kruskal–Wallis and pairwise Wilcoxon tests. In cases with significant differences, RI components (quantities of A, B, C, and submerged species) were further analyzed. A, B, and C species are type-specific indicators, and submerged species serve as RI correction factors. According to [29] species group A are taxa that indicate reference conditions, species group C indicate the lack of reference conditions, while species group B are taxa indifferent to the occurrence of reference conditions or pressures in waterbodies. Indicator species lists were adapted from [31] for the Serbian Danube.

3. Results

To evaluate the influence of sampling procedures on macrophyte taxa diversity observation, we applied framework based on Hill numbers and Chao estimators. First, we assessed sample completeness for a macrophyte survey across an entire river reach, using various sampling strategies (Figure 2a, Table 2). The “all” sampling method achieved maximum sample completeness (100%) for all diversity orders. However, sample completeness profiles for the “JDS” and “NMP” sampling procedures increased with diversity order, suggesting undetected diversity in both datasets. Due to data sparsity in the “NMP” and “JDS” samples, the profiles intersect, and the confidence bands are broad, rendering them statistically indistinguishable. The estimated sample completeness for “JDS” and “NMP” data at q = 0 was 84% and 93%, respectively, indicating that the “JDS” sample covered up to 84% and the “NMP” sample up to 93% of total species richness in the sampled assemblage. For diversity orders q = 1 and q = 2, both datasets reached 100% sample completeness, capturing all observed species abundances and highly abundant species.
Sample completeness was also estimated individually for each waterbody (Table 2A, Figure S1). Generally, sample completeness profiles for all datasets rose with diversity order, reflecting the pattern observed across the whole river reach. Exceptions included 100% completeness across all diversity orders in “JDS” and “NMP” datasets for waterbodies D9 and D10, and in the “NMP” dataset for waterbody D3. In waterbody D2, the “all” dataset reached 96% completeness for q = 0, while in waterbody D10 it covered only 76%. For q = 1, completeness in the “all” dataset was below 100% in waterbodies D1, D8, D9, and D10, with values covering up to 98%, 99%, 94%, and 99% of species abundances, respectively. The “NMP” datasets generally exhibited lower completeness than the “JDS” datasets, except in waterbodies D1 and D6.
Figure S2 shows that, for each dataset sampled across the entire river reach, the size-based rarefaction and extrapolation sampling curves for diversity orders q = 1 and q = 2 stabilize, indicating that the asymptotic diversity estimates for these measures are reliable for inferring true diversity (Figure 2b). However, only the species richness curve (q = 0) for the “all” dataset, extrapolated to double the observed (reference) sample size, tends to stabilize. This suggests that only the “all” dataset provides sufficient information to accurately estimate true (asymptotic) species richness. In contrast, the other two datasets lack sufficient information for accurate asymptotic species richness estimation and instead provide only a lower bound.
A similar analysis of individual waterbodies (Figure S3) suggests that asymptotic species richness values (q = 0; Table 2B) should be interpreted as minimum estimates for all datasets. The asymptotic values for q = 1 and q = 2 derived from the “all” dataset reliably infer true diversity in all waterbodies except D1. For the “JDS” and “NMP” datasets, the asymptotic Shannon and Simpson diversity values infer true diversity in waterbodies D3, D4, and D9. However, in the remaining waterbodies, these values represent only the lower bounds of true diversity.
Comparing asymptotic and empirical diversity profiles (solid and dashed lines in Figure 2b, Table 2B), we examined the undetected diversity within each dataset for q = 0, q = 1, and q = 2. The “all” dataset had no undetected diversity. For “JDS” and “NMP” datasets, undetected species richness (q = 0) was at least 8 (≥16%) and 2.67 (≥7%), respectively. Undetected Shannon diversity (q = 1) was minimal (0.03 for “JDS” and 0.06 for “NMP”), indicating that only a single abundant species remained undetected in each dataset. Differences between assemblages regarding abundant species were 0.9 and were statistically non-significant due to overlapping 95% confidence bands. The differences in Simpson diversity (q = 2) between datasets were also minor, with nearly all highly abundant species detected.
When analyzing individual waterbodies, patterns similar to those observed for the entire river reach appeared (Table 2B, Figure S4). Notable exceptions included higher asymptotic diversity values across all orders in waterbody D8 for the “NMP” dataset, as well as higher observed Shannon diversity (q = 1) and equal observed species richness (q = 0). In waterbody D2, the “NMP” dataset also had higher asymptotic species richness (q = 0) compared to “JDS”, while in waterbody D4, “NMP” outperformed “all” in asymptotic species richness (q = 0) and Shannon diversity (q = 1). In waterbody D3, the “JDS” dataset had a slightly higher asymptotic Simpson diversity (q = 2) than “all”. “NMP” significantly differed from both “all” and “JDS” in waterbody D3 across all diversity orders, with lower asymptotic and observed values. Shannon and Simpson diversity values in “NMP” and “JDS” datasets were significantly lower than in “all” for waterbodies D2, D6, and D7. In waterbodies D9 and D10, both “JDS” and “NMP” detected only one species each, while the “all” sampling indicated the presence of 9 and 7 species, respectively.
To evaluate the sampling procedures’ effectiveness in capturing the structure of macrophyte assemblages, we analyzed multivariate homogeneity of group dispersions (variances) for the entire dataset using ANOVA, pairwise permutation tests, and Tukey’s HSD, which found no significant differences between aquatic vegetation compositions sampled by “all”, “JDS”, and “NMP” methods. The Euclidean distances between group members, reduced to principal coordinate axes, are shown for the first two PCoA axes in Figure 3.
We applied the same analysis to each waterbody (Table S1). ANOVA showed significant differences in composition across sampling methods in waterbodies D2, D3, D4, and D5 (Figure 4). Pairwise permutation and Tukey’s HSD tests identified significant differences between “NMP” and both “all” and “JDS” in waterbodies D2, D3, and D5, and between “NMP” and “all” in D4.
Only in waterbody D5, the Kruskal–Wallis test revealed significant differences in Reference Indices across sampling methods (Figure 5), with the “NMP” dataset having higher indices than “all” and “JDS”. Pairwise Wilcoxon tests confirmed significant differences between “NMP” and both “all” and “JDS”. The Kruskal–Wallis test on indicator groups A, B, and C revealed significant differences in groups A and C (Figure 6 and Figure S7), with pairwise Wilcoxon tests showing significant differences in group A between “NMP” and both “all” and “JDS”, and in group C between “JDS” and “NMP”. Although no significant difference was found in submerged taxa, the pairwise test indicated a difference between “JDS” and “NMP” (Figure 6).

4. Discussion

The study’s findings underscore the critical influence of sampling methodologies on macrophyte diversity assessments, with implications for biodiversity conservation and ecosystem management. The use of Hill numbers and Chao estimators provides a robust framework for evaluating sample completeness and diversity across various orders, offering detailed insights into the strengths and limitations of different sampling procedures.

4.1. Sample Completeness and Asymptotic Diversity Estimates

The “all” dataset’s achievement of 100% sample completeness across all diversity orders (q = 0, 1, 2) highlights its effectiveness in capturing the full spectrum of macrophyte diversity. This is consistent with studies emphasizing the importance of exhaustive sampling to reduce biases in biodiversity assessments [12,15,17,24]. In contrast, the lower completeness of the “JDS” and “NMP” datasets for species richness (q = 0)—84% and 93%, respectively—demonstrates their limitations in detecting rare taxa, which are often under-sampled due to their patchy distributions [12,20,32].
The increase in completeness for higher diversity orders reflects the greater detectability of common and abundant species, corroborating findings by [16] that dominant taxa require less sampling effort for reliable detection. The stabilization of rarefaction and extrapolation curves for Shannon (q = 1) and Simpson (q = 2) diversity supports the reliability of these asymptotic estimates for inferring true diversity [17,33]. However, the inability of the species richness curves (q = 0) to stabilize in the “JDS” and “NMP” datasets highlights the challenges associated with rare species detection, consistent with [9,12,15,24].

4.2. Spatial Heterogeneity Across Waterbodies

The variability in sample completeness across individual waterbodies underscores the spatial heterogeneity in macrophyte distributions, which are shaped by local abiotic factors such as hydrology, substrate type, and nutrient availability [3,34,35]. Lower completeness values in waterbodies D1 and D8, even in the “all” dataset, suggest the presence of rare or highly localized taxa, while the higher completeness in D9 and D10 reflects homogenous communities with fewer species.
The finding that the “NMP” dataset generally exhibits lower completeness than the “JDS” dataset (except in D1 and D6) highlights the need to account for methodological differences in spatial coverage and sampling intensity. These results align with [12], who emphasized optimizing sampling designs to address spatial variability in community structure.

4.3. Undetected Diversity and Sampling Effectiveness

Undetected diversity estimates reveal key differences between sampling methods, particularly for rare species (q = 0). The “JDS” and “NMP” datasets under-detected at least 16% and 7% of species richness, respectively, highlighting the challenges of fully capturing rare taxa. Rare species contribute disproportionately to ecosystem resilience and functional diversity, buffering ecosystems against environmental fluctuations [36,37]. Their underrepresentation could thus bias ecological assessments and management strategies.
For Shannon (q = 1) and Simpson (q = 2) diversity, undetected diversity was minimal, suggesting that both “JDS” and “NMP” effectively captured common and abundant taxa. The lack of significant differences between the two datasets for these diversity orders, as indicated by overlapping confidence intervals, suggests that either method could be used to assess dominant assemblage components with comparable accuracy.

4.4. Community Composition and Beta Diversity

The overall lack of significant differences in community composition across sampling methods (ANOVA, pairwise permutation tests, and multivariate homogeneity analyses) indicates that all methods reliably captured the core assemblages, particularly at the river-reach scale. This finding aligns with [38], who noted that beta diversity estimates can remain robust when sampling is reasonably comprehensive.
However, significant differences in community composition in specific waterbodies (e.g., D2, D3, and D5) highlight the influence of localized environmental factors, such as habitat heterogeneity and disturbance regimes, on assemblage structure. Such spatial patterns are consistent with species sorting and environmental filtering processes [11,39].

4.5. Implications for Ecological Monitoring and Management

The reference index has proven to be a sufficiently robust measure for ecological status assessment in nine out of ten water bodies, considering that only in the D5 water body did the values of the reference index differ significantly depending on the sampling procedure applied. The higher reference index values for the “NMP” dataset in waterbody D5 suggests that it may better capture type-specific indicator species, which are critical for ecological status assessments [29]. However, the inconsistencies in detecting submerged taxa and indicator groups across methods highlight the need for methodological standardization to ensure comparability and reliability in monitoring programs [13,20,22,26].
The results highlight the trade-offs between sampling effort and data quality in biodiversity assessments. While the “all” dataset offers the most accurate and comprehensive estimates, its resource-intensive nature may limit its feasibility for routine monitoring [9,12,20]. Conversely, the “JDS” and “NMP” methods provide practical alternatives, though they require methodological refinements to improve rare species detection. This trade-off aligns with the recommendations of [12], who emphasized balancing efficiency and accuracy in biodiversity monitoring.
One promising approach is the adoption of stratified sampling designs that combine comprehensive and targeted strategies could enhance the efficiency and effectiveness of macrophyte surveys, as suggested by [12,20,24]. For instance, employing the “all” method in ecologically diverse areas while using “JDS” or “NMP” in homogenous regions could optimize resource allocation without compromising data quality.
The spatial heterogeneity observed in macrophyte assemblages further emphasizes the need for adaptive management strategies tailored to local environmental conditions. Integrating macrophyte diversity assessments with hydrological and water quality data could provide a more holistic understanding of ecosystem health and inform targeted conservation efforts [40].

5. Conclusions

This study highlights the critical role of sampling methodologies in shaping our understanding of macrophyte diversity in river habitats, with significant implications for biodiversity conservation, ecological monitoring, and resource management. Our findings demonstrate that the comprehensive “all” dataset provides the most accurate and complete estimates of species richness and diversity across all orders, owing to its broad spatial coverage and exhaustive sampling effort. In contrast, the “JDS” and “NMP” datasets, while practical and resource-efficient, exhibit limitations in detecting rare species, which are crucial for ecosystem resilience and functional diversity.
The study underscores the importance of adopting a stratified and context-sensitive sampling design to address spatial heterogeneity and optimize resource allocation. For instance, combining comprehensive methods in ecologically diverse areas with targeted approaches in more homogeneous regions can balance the trade-offs between data quality and feasibility. Moreover, the observed inconsistencies in detecting rare and indicator species across methods emphasize the need for methodological standardization to ensure comparability and reliability in biodiversity assessments and ecological monitoring programs.
A fundamental challenge of this research lies in understanding the composition and distribution of aquatic species assemblages in river habitats and determining effective methods for their detection. River habitats are characterized by high spatial and temporal variability, making it difficult to achieve comprehensive sampling that captures the full range of species present, particularly rare and cryptic taxa. Addressing this challenge requires not only improvements in traditional sampling methods but also the integration of emerging technologies such as environmental DNA (eDNA) analysis, automated image recognition, and remote sensing. These tools can complement field-based surveys by enhancing the detection of elusive species and providing a more complete picture of aquatic biodiversity.
Looking forward, future research should focus on integrating macrophyte diversity assessments with hydrological and water quality data to develop a more holistic understanding of ecosystem health. Advanced technologies, such as those mentioned above, hold promise for addressing the limitations of existing methods and fully capturing aquatic species assemblages in river habitats. Such innovations, coupled with adaptive management strategies tailored to local environmental conditions, can significantly enhance conservation efforts and inform sustainable management of freshwater ecosystems.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/w17030328/s1, Figure S1: The plots of estimated sample completeness curves as a function of order q between 0 and 2 in “all”, “JDS”, and “NMP” datasets obtained for waterbodies, Figure S2: Sample size-based rarefaction (solid lines) and extrapolation (dashed lines) analysis of the sampling curves for diversity orders q = 0, q = 1, and q = 2 of the whole river reach, Figure S3: Sample size-based rarefaction (solid lines) and extrapolation (dashed lines) analysis of the sampling curves for diversity orders q = 0, q = 1, and q = 2 of the waterbodies, Figure S4: The asymptotic estimates of diversity profiles (solid lines) and empirical diversity profiles (dashed lines) in the “all”, “JDS”, and “NMD” datasets for the waterbodies, Table S1: Results of the ANOVA of the distances to group means, pairwise permutation test of group mean dispersions (with 999 permutations), and Tukey’s Honest Significant Differences between groups, Figure S5: Multivariate dispersions with plotted distances to medians obtained for macrophyte communities applying “all”, “JDS”, and “NMP” sampling procedures obtained for the waterbodies D1, D6-D10, Figure S6: Results of Kruskal–Wallis test and Wilcoxon pairwise tests of Reference Indices calculated for macrophyte assemblages in waterbodies D1-D4, D6-D10, sampled using “all”, “JDS”, and “NMP” procedure, Figure S7: Results of Kruskal–Wallis test and Wilcoxon pairwise tests of type specific indicator group B quantities in macrophyte assemblages in the D5 waterbody, sampled using “all”, “JDS”, and “NMP” procedure.

Author Contributions

Conceptualization, D.V., M.Ć., N.N. and M.I.; Data Preparation, M.Ć. and N.N.; Methodology and Analysis, D.V. and M.I.; Writing, D.V., M.Ć., N.N. and M.I. All authors have read and agreed to the published version of the manuscript.

Funding

This research is funded by Ministry of Science, Technological Development and Innovation of the Republic of Serbia (Grants No. ‪451-03-66/2024-03/200125 & 451-03-65/2024-03/200125) and Provincial Secretariat for Higher Education and Scientific Research of Autonomous Province of Vojvodina, Serbia (Project title: Habitats in Decline: Stressors in the Ecosystems of Floodplain Forests in Vojvodina; Grant No. 000882993 2024 09418 003 000 000 001 04 002.

Data Availability Statement

Data are available from the authors by reasonable request.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Carpenter, S.R.; Lodge, D.M. Effects of Submersed Macrophytes on Ecosystem Processes. Aquat. Bot. 1986, 26, 341–370. [Google Scholar] [CrossRef]
  2. Lacoul, P.; Freedman, B. Environmental influences on aquatic plants in freshwater ecosystems. Environ. Rev. 2006, 14, 89–136. [Google Scholar] [CrossRef]
  3. Bornette, G.; Puijalon, S. Macrophytes: Ecology of Aquatic Plants. In Encyclopedia of Life Sciences (ELS); John Wiley & Sons, Ltd.: Chichester, UK, 2009; pp. 1–9. [Google Scholar]
  4. Bennett, A.F.; Haslem, A.; Cheal, D.C.; Clarke, M.F.; Jones, R.N.; Koehn, J.D.; Lake, P.S.; Lumsden, L.F.; Lunt, I.D.; Mackey, B.G.; et al. Ecological processes: A key element in strategies for nature conservation. Ecol. Manag. Restor. 2009, 10, 192–199. [Google Scholar] [CrossRef]
  5. Bennett, E.M.; Peterson, G.D.; Gordon, L.J. Understanding relationships among multiple ecosystem services. Ecol. Lett. 2009, 12, 1394–1404. [Google Scholar] [CrossRef]
  6. Kennard, M.; Pusey, B.; Harch, B.; Dore, E.; Arthington, A. Estimating local stream fish assemblage attributes: Sampling effort and efficiency at two spatial scales. Mar. Freshw. Res. 2006, 57, 635–653. [Google Scholar] [CrossRef]
  7. Hughes, R.M.; Herlihy, A.T.; Gerth, W.J.; Pan, Y. Estimating vertebrate, benthic macroinvertebrate and diatom taxa richness in raftable Pacific Northwest rivers for bioassessment purposes. Environ. Monit. Assess. 2012, 184, 3185–3198. [Google Scholar] [CrossRef] [PubMed]
  8. Li, L.; Liu, L.; Hughes, R.M.; Cao, Y.; Wang, X. Towards a protocol for stream macroinvertebrate sampling in China. Environ. Monit. Assess. 2014, 186, 469–479. [Google Scholar] [CrossRef]
  9. Budka, A.; Lacka, A.; Szoszkiewicz, K. Estimation of river ecosystem biodiversity based on Chao estimator. Biodivers. Conserv. 2018, 27, 205–216. [Google Scholar] [CrossRef]
  10. Paller, M. Estimating fish species richness across multiple watersheds. Diversity 2018, 10, 42. [Google Scholar] [CrossRef]
  11. Tonkin, J.D.; Altermatt, F.; Finn, D.S.; Heino, J.; Olden, J.D.; Pauls, S.U.; Lytle, D.A. The role of dispersal in river network metacommunities: Patterns, processes, and pathways. Freshw. Biol. 2018, 63, 141–163. [Google Scholar] [CrossRef]
  12. Sgarbi, L.F.; Bini, L.M.; Heino, J.; Jyrkankallio-Mikkola, J.; Landeiro, V.L.; Santos, E.P.; Schneck, F.; Siqueira, T.; Soinien, J.; Tolonen, K.T.; et al. Sampling effort and information quality provided by rare and common species in estimating assemblage structure. Ecol. Indic. 2020, 110, 105937. [Google Scholar] [CrossRef]
  13. 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]
  14. Chao, A.; Jost, L. Coverage-based rarefaction and extrapolation: Standardizing samples by completeness rather than size. Ecology 2012, 93, 2533–2547. [Google Scholar] [CrossRef]
  15. Colwell, R.K.; Chao, A.; Gotelli, N.J.; Lin, S.Y.; Mao, C.X.; Chazdon, R.L.; Longino, J.T. Models and estimators linking individual-based and sample-based rarefaction, extrapolation, and comparison of assemblages. J. Plant Ecol. 2012, 5, 3–21. [Google Scholar] [CrossRef]
  16. Chao, A.; Gotelli, N.J.; Hsieh, T.C.; Sander, E.L.; Ma, K.H.; Colwell, R.K.; Ellison, A.M. Rarefaction and extrapolation with Hill numbers: A framework for sampling and estimation in species diversity studies. Ecol. Monogr. 2014, 84, 45–67. [Google Scholar] [CrossRef]
  17. Chao, A.; Kubota, Y.; Zeleny, D.; Chiu, C.-H.; Li, C.-F.; Kusumoto, B.; Yasuhara, M.; Thorn, S.; Wei, C.-L.; Costello, M.J.; et al. Quantifying sample completeness and comparing diversities among assemblages. Ecol. Res. 2020, 35, 292–314. [Google Scholar] [CrossRef]
  18. Gotelli, N.J.; Chao, A. Measuring and estimating species richness, species diversity, and biotic similarity from sampling data. In Encyclopedia of Biodiversity, 2nd ed.; Levin, S.A., Ed.; Academic Press: Waltham, MA, USA, 2013; Volume 5, pp. 195–211. [Google Scholar]
  19. Budka, A.; Lacka, A.; Szoszkiewicz, K. The use of rarefaction and extrapolation as methods of estimating the effects of river eutrophication on macrophyte diversity. Biodivers. Conserv. 2019, 28, 385–400. [Google Scholar] [CrossRef]
  20. Szoszkiewicz, K.; Budka, A.; Lacka, A.; Pietruczuk, K. Determining macrophyte species richness and dark diversity sources—A novel approach to improve the biodiversity estimation based on species traits. Sci. Total Environ. 2022, 816, 151496. [Google Scholar] [CrossRef] [PubMed]
  21. Roni, P.; Liermann, M.C.; Jordan, C.; Steel, E.A. Steps for designing a monitoring and evaluation program for aquatic restoration. In Monitoring Stream and Watershed Restoration; Roni, P., Ed.; American Fisheries Society: Bethesda, MA, USA, 2005; pp. 13–34. [Google Scholar]
  22. Birk, S.; Willby, N. Towards harmonization of ecological quality classification: Establishing common grounds in European macrophyte assessment for rivers. Hydrobiologia 2010, 652, 149–163. [Google Scholar] [CrossRef]
  23. Szoszkiewicz, F.; Ferreira, T.; Korte, T.; Baattrup-Pedersen, A.; Davy-Bowker, J.; O’Hare, M. European River plant communities: The importance of organic pollution and the usefulness of existing macrophyte metrics. Hydrobiologia 2006, 566, 211–234. [Google Scholar] [CrossRef]
  24. Croft, M.V.; Chow-Fraser, P. Non-random sampling and its role in habitat conservation: A comparison of three wetland macrophyte sampling protocols. Biodivers. Conserv. 2009, 18, 2283–2306. [Google Scholar] [CrossRef]
  25. Chao, A.; Jost, L. Estimating diversity and entropy profiles via discovery rates of new species. Methods Ecol. Evol. 2015, 6, 873–882. [Google Scholar] [CrossRef]
  26. Szoszkiewicz, K.; Jusik, S.; Pietruczuk, K.; Gebler, D. The Macrophyte Index for Rivers (MIR) as an Advantageous Approach to Running Water Assessment in Local Geographical Conditions. Water 2020, 12, 108. [Google Scholar] [CrossRef]
  27. CEN EN 14184; Water Quality Guidance Standard for the Surveying of Aquatic Macrophytes in Running Waters. BSI: London, UK, 2014.
  28. Kohler, A. Methoden der Kartierung von Flora und Vegetation von Süβwasserbiotopen. Landsch. Stadt 1978, 10, 73–85. [Google Scholar]
  29. 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]
  30. Oksanen, J.; Simpson, G.; Blanchet, F.; Kindt, R.; Legendre, P.; Minchin, P.; O’Hara, R.; Solymos, P.; Stevens, M.; Szoecs, E.; et al. Vegan: Community Ecology Package, R Package, Version 2.7-0; Bavarian Environment Agency: Augsburg, Germany, 2024. [Google Scholar]
  31. Schaumburg, J.; Schranz, C.; Stelzer, D.; Vogel, A.; Gutowski, A. Instruction Manual for the Assessment of Running Water Ecological Status in Accordance with the Requirements of the EC-Water Framework Directive: Macrophytes and Phytobenthos; Bavarian Environment Agency: Augsburg, Germany, 2012. [Google Scholar]
  32. Magurran, A.E. Measuring Biological Diversity; Blackwell Science: Oxford, UK, 2013. [Google Scholar]
  33. Hsieh, T.C.; Ma, K.H.; Chao, A. iNEXT: An R package for rarefaction and extrapolation of species diversity (Hill numbers). Methods Ecol. Evol. 2016, 7, 1451–1456. [Google Scholar] [CrossRef]
  34. Poff, N.L.; Allan, J.D.; Bain, M.B.; Karr, J.R.; Prestegaard, K.L.; Richter, B.D.; Sparks, R.E.; Stromberg, J.C. The natural flow regime. BioScience 1997, 47, 769–784. [Google Scholar] [CrossRef]
  35. Riis, T. Plant distribution and abundance in relation to physical conditions and location within Danish stream systems. Hydrobiologia 2001, 448, 217–228. [Google Scholar] [CrossRef]
  36. Mouillot, D.; Bellwood, D.R.; Baraloto, C.; Chave, J.; Galzin, R.; Harmelin-Vivien, M.; Kulbicki, M.; Lavergne, S.; Lavorel, S.; Mouquet, N.; et al. Rare Species Support Vulnerable Functions in High-Diversity Ecosystems. PLoS Biol. 2013, 11, e1001569. [Google Scholar] [CrossRef]
  37. Díaz, S.; Purvis, A.; Cornelissen, J.H.C.; Mace, G.M.; Donoghue, M.J.; Ewers, R.M.; Jordano, P.; Pearse, W.D. Functional traits, the phylogeny of function, and ecosystem service vulnerability. Ecol. Evol. 2013, 3, 2958–2975. [Google Scholar] [CrossRef]
  38. Anderson, M.J.; Ellingsen, K.E.; McArdle, B.H. Multivariate dispersion as a measure of beta diversity. Ecol. Lett. 2006, 9, 683–693. [Google Scholar] [CrossRef] [PubMed]
  39. Baattrup-Pedersen, A.; Larsen, S.E.; Riis, T. Long-term effects of stream management on plant communities in two Danish lowland streams. Hydrobiologia 2002, 481, 33–45. [Google Scholar] [CrossRef]
  40. Feld, C.K.; da Silva, P.M.; Sousa, J.P.; de Bello, F.; Bugter, R.; Grandin, U.; Hering, D.; Lavorel, S.; Mountford, O.; Pardo, I.; et al. Indicators of biodiversity and ecosystem services: A synthesis across ecosystems and spatial scales. Oikos 2014, 123, 523–530. [Google Scholar] [CrossRef]
Water 17 00328 g001
Figure 1. Study area with JDS and NMP sampling locations and delineated waterbodies D1–D10.
Figure 1. Study area with JDS and NMP sampling locations and delineated waterbodies D1–D10.
Water 17 00328 g001
Water 17 00328 g002
Figure 2. (a) Sample completeness curves across diversity orders (q = 0 to q = 2) for the whole Serbian Danube reach; (b) asymptotic (solid lines) and empirical (dashed lines) diversity profiles. Shaded areas indicate 95% confidence intervals derived from bootstrap replicates (n = 100), with narrow confidence bands in some instances. Numerical values for the three special cases of q = 0, 1, 2 are shown in Table 2.
Figure 2. (a) Sample completeness curves across diversity orders (q = 0 to q = 2) for the whole Serbian Danube reach; (b) asymptotic (solid lines) and empirical (dashed lines) diversity profiles. Shaded areas indicate 95% confidence intervals derived from bootstrap replicates (n = 100), with narrow confidence bands in some instances. Numerical values for the three special cases of q = 0, 1, 2 are shown in Table 2.
Water 17 00328 g002
Water 17 00328 g003
Figure 3. Multivariate dispersions with plotted distances to medians obtained for macrophyte communities applying “all”, “JDS”, and “NMP” sampling procedures obtained for the whole river reach.
Figure 3. Multivariate dispersions with plotted distances to medians obtained for macrophyte communities applying “all”, “JDS”, and “NMP” sampling procedures obtained for the whole river reach.
Water 17 00328 g003
Water 17 00328 g004
Figure 4. Multivariate dispersions with plotted distances to medians obtained for macrophyte communities in waterbodies D2, D3, D4, and D5, applying “all”, “JDS”, and “NMP” sampling procedures. Significance codes for ANOVA (at the upper left corner of plots) and Tukey’s HSD (at the upper right corner of plots): p < 0.001—***, p < 0.01—**, p < 0.05—*.
Figure 4. Multivariate dispersions with plotted distances to medians obtained for macrophyte communities in waterbodies D2, D3, D4, and D5, applying “all”, “JDS”, and “NMP” sampling procedures. Significance codes for ANOVA (at the upper left corner of plots) and Tukey’s HSD (at the upper right corner of plots): p < 0.001—***, p < 0.01—**, p < 0.05—*.
Water 17 00328 g004
Water 17 00328 g005
Figure 5. Results of Kruskal–Wallis test and Wilcoxon pairwise tests of reference index values calculated for macrophyte assemblages in the D5 waterbody, sampled using “all”, “JDS”, and “NMP” procedure.
Figure 5. Results of Kruskal–Wallis test and Wilcoxon pairwise tests of reference index values calculated for macrophyte assemblages in the D5 waterbody, sampled using “all”, “JDS”, and “NMP” procedure.
Water 17 00328 g005
Water 17 00328 g006
Figure 6. Results of Kruskal–Wallis test and Wilcoxon pairwise tests of type specific indicator group A (left), C (middle) and submerged species (right) quantities in macrophyte assemblages in the D5 waterbody, sampled using “all”, “JDS”, and “NMP” procedure.
Figure 6. Results of Kruskal–Wallis test and Wilcoxon pairwise tests of type specific indicator group A (left), C (middle) and submerged species (right) quantities in macrophyte assemblages in the D5 waterbody, sampled using “all”, “JDS”, and “NMP” procedure.
Water 17 00328 g006
Table 1. Danube waterbodies in Serbia. (A). Names and starting/ending river km; (B). nSU—number of survey units sampled by “all”, “JDS”, and “NMP” procedures; overlap—number of overlapping survey units when “JDS” and “NMP” procedures are applied.
Table 1. Danube waterbodies in Serbia. (A). Names and starting/ending river km; (B). nSU—number of survey units sampled by “all”, “JDS”, and “NMP” procedures; overlap—number of overlapping survey units when “JDS” and “NMP” procedures are applied.
(A). Water Body(B). nSU
NameFrom [rkm]To [rkm]AllJDSNMPOverlap
D101433138153553
D913801296861054
D8129512548410104
D7125312157820106
D6121411718810100
D51170110513230100
D41104107658101010
D310759441321553
D294386376550
D186284617552
Table 2. The q = 0, 1, and 2 diversity values for “all”, “JDS”, and “NMP” datasets across the river reach and specific waterbodies. (A) Sample completeness; (B) asymptotic, observed and undetected diversity profiles.
Table 2. The q = 0, 1, and 2 diversity values for “all”, “JDS”, and “NMP” datasets across the river reach and specific waterbodies. (A) Sample completeness; (B) asymptotic, observed and undetected diversity profiles.
(A). Completeness(B). Asymptotic, Observed, and Undetected Diversity Profiles
AllJDSNMPAll JDS NMP
Asym.Obs.Und.Asym.Obs.Und.Asym.Obs.Und.
Dq = 0100%84%93%46.1746.000.1749.0041.008.0037.6735.002.67
q = 1100%100%100%13.9613.950.0114.6014.570.0313.1713.110.06
q = 2100%100%100%9.619.600.0110.2510.240.019.049.030.01
D1q = 086%73%94%13.9812.001.9815.0211.004.028.488.000.48
q = 198%83%96%10.579.740.8210.527.952.579.167.591.56
q = 2100%95%96%9.728.830.897.386.051.3310.507.333.17
D2q = 096%98%73%25.9025.000.9014.2514.000.2514.9911.003.99
q = 1100%100%99%8.118.090.016.736.670.054.764.660.10
q = 2100%100%100%6.286.280.005.205.180.024.264.220.04
D3q = 080%74%100%41.0033.008.0038.0028.0010.0014.0014.000.00
q = 1100%100%100%12.2212.210.0213.2313.140.088.148.040.11
q = 2100%100%100%9.049.030.0110.1310.100.036.546.470.07
D4q = 097%93%79%36.0035.001.0035.6733.002.6739.1631.008.16
q = 1100%100%100%12.9512.930.0212.7512.690.0613.0312.900.13
q = 2100%100%100%9.619.600.019.669.640.029.439.390.04
D5q = 092%88%79%29.2527.002.2524.0021.003.0021.5017.004.50
q = 1100%100%100%9.049.030.018.158.120.047.707.640.06
q = 2100%100%100%6.746.740.006.005.990.016.026.000.02
D6q = 087%83%90%34.5030.004.5020.5517.003.5515.4914.001.49
q = 1100%92%99%14.3914.250.1411.299.731.575.825.600.23
q = 2100%99%100%10.1710.110.057.406.820.583.553.510.04
D7q = 089%88%79%28.0025.003.0023.9921.002.9920.1516.004.15
q = 1100%99%98%10.5510.420.137.887.640.244.123.970.16
q = 2100%100%100%7.477.430.044.554.510.042.622.600.02
D8q = 090%85%56%18.9917.001.995.915.000.919.005.004.00
q = 199%85%69%8.548.280.265.374.111.256.864.172.69
q = 2100%88%88%6.106.000.104.583.461.135.143.521.62
D9q = 077%100%100%11.639.002.631.001.000.001.001.000.00
q = 194%100%100%2.852.590.261.001.000.001.001.000.00
q = 2100%100%100%1.631.620.021.001.000.001.001.000.00
D10q = 076%100%100%9.247.002.241.001.000.001.001.000.00
q = 199%100%100%1.251.230.021.001.000.001.001.000.00
q = 2100%100%100%1.071.070.001.001.000.001.001.000.00
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Vukov, D.; Ćuk, M.; Nikolić, N.; Ilić, M. Assessing Macrophyte Diversity in the Danube River: Comparing the Effectiveness of Different Sampling Procedures. Water 2025, 17, 328. https://doi.org/10.3390/w17030328

AMA Style

Vukov D, Ćuk M, Nikolić N, Ilić M. Assessing Macrophyte Diversity in the Danube River: Comparing the Effectiveness of Different Sampling Procedures. Water. 2025; 17(3):328. https://doi.org/10.3390/w17030328

Chicago/Turabian Style

Vukov, Dragana, Mirjana Ćuk, Nataša Nikolić, and Miloš Ilić. 2025. "Assessing Macrophyte Diversity in the Danube River: Comparing the Effectiveness of Different Sampling Procedures" Water 17, no. 3: 328. https://doi.org/10.3390/w17030328

APA Style

Vukov, D., Ćuk, M., Nikolić, N., & Ilić, M. (2025). Assessing Macrophyte Diversity in the Danube River: Comparing the Effectiveness of Different Sampling Procedures. Water, 17(3), 328. https://doi.org/10.3390/w17030328

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