Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Agricultural impacts on streams near Nitrate Vulnerable Zones: A case study in the Ebro basin, Northern Spain

  • Rubén Ladrera,

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Food and Agriculture Department, Science and Technology Complex, University of La Rioja, Logroño, La Rioja, Spain

  • Oscar Belmar,

    Roles Formal analysis, Methodology, Resources, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Marine and Continental Waters Program, IRTA, Sant Carles de la Ràpita, Catalonia, Spain

  • Rafael Tomás,

    Roles Formal analysis, Investigation, Resources

    Affiliation Food and Agriculture Department, Science and Technology Complex, University of La Rioja, Logroño, La Rioja, Spain

  • Narcís Prat,

    Roles Conceptualization, Methodology, Validation, Writing – review & editing

    Affiliation Freshwater Ecology, Hydrology and Management Group (FEHM), Departament de Biologia Evolutiva, Ecologia i Ciències Ambientals, Institut de Recerca de l'Aigua (IdRA), University of Barcelona, Catalonia, Spain

  • Miguel Cañedo-Argüelles

    Roles Conceptualization, Formal analysis, Methodology, Supervision, Visualization, Project administration, Writing – original draft, Writing – review & editing

    Affiliation Freshwater Ecology, Hydrology and Management Group (FEHM), Departament de Biologia Evolutiva, Ecologia i Ciències Ambientals, Institut de Recerca de l'Aigua (IdRA), University of Barcelona, Catalonia, Spain


Agricultural intensification during the last century has caused river degradation across Europe. From the wide range of stressors derived from agricultural activities that impact rivers, diffuse agricultural pollution has received most of the attention from managers and scientists. The aim of this study was to determine the main stressors exerted by intensive agriculture on streams around Nitrate Vulnerable Zones (NVZs), which are areas of land that drain into waters polluted by nitrates according to the European Nitrate Directive (91/676/EEC). The study area was located in the NW of La Rioja (Northern Spain), which has some of the highest nitrate concentrations within the Ebro basin. The relationships between 40 environmental variables and the taxonomic and functional characteristics of the macroinvertebrate assemblages (which are useful indicators of water quality) were analyzed in 11 stream reaches differentially affected by upstream agricultural activity. The streams affected by a greater percentage of agricultural land cover in the surrounding catchment had significantly higher nitrate concentrations than the remaining sites. However, hydromorphological alteration (i.e. channel simplification, riparian forest and habitat degradation), which is closely linked to agricultural practices, was the main factor affecting macroinvertebrate assemblages. We suggest that “good agricultural practices” should be implemented in streams affected by NVZs to reverse stream degradation, in concordance with the European Water Framework Directive (WFD).


Agriculture is one of the most important pressures affecting freshwater ecosystems around the world [1,2]. Intensive agriculture: i) strongly affects water quality through pesticides and fertilizers, mainly nitrates [3]; ii) requires large quantities of water for irrigation, and iii) degrades the fluvial habitat through riparian forest removal, channel incision and straightening, reduction of bank stability and sediment deposition [4].

To revert agricultural impacts on streams, different Directives [i.e. a legal act requiring member states to achieve a certain result) have been implemented in the European Union (EU) during the last decades. One example is the Nitrates Directive (91/676/EEC; ND), which was adopted in 1991. The ND aims to protect water quality across Europe reducing nitrates from agricultural sources and promoting the use of good farming practices. Within the ND, Nitrate Vulnerable Zones (NVZs) are defined as the areas of land that drain into waters polluted by nitrates, and farmers with land in these zones have to follow mandatory rules to restore water quality [5]. The ND forms an integral part of the European Water Framework Directive (2000/60/EC; WFD), which is aimed at establishing a framework for water protection, so all the water bodies in Europe reach a “good ecological status” by 2021 or 2027. To achieve this good ecological status, the WFD indicates that pressures and impacts at the basin scale must be identified to guide management measures. This identification is often inadequate in existing monitoring programs [6].

The European Water Framework Directive (WFD) also establishes the need to use biotic indices to assess the ecological status of rivers, and the most widely used indices are those based on macroinvertebrate assemblages [79]. Regarding agriculture, there is an extensive literature related to assessing its impacts on rivers through macroinvertebrate assemblages, using both the taxonomic composition and the functional structure [4,1015]. Compared with the taxonomic approach, the study of the functional structure (i.e. trait-based approaches) presents a higher temporal and spatial stability across regions because the variability attributed to species composition of different areas is reduced [16,17].

Various studies suggest that 10 ppm of nitrate can adversely affect, at least during long-term exposures, freshwater invertebrates [18,19]. However, beyond nitrate toxicity, nitrate pollution in rivers and streams has been associated with other environmental factors, for example increased autotrophic production due to reduced shading of the stream channel [20,21]. High periphyton biomass reduces oxygen concentrations in the water-column and the substrate interstices [22,23], with increased ecosystem respiration leading to impacts on pollution sensitive species such as some Ephemeroptera, Plecoptera and Trichoptera (EPT) [24]. Excessive algal growth can also smother the riverbed, reducing substrate availability for grazing mayflies, which are replaced by molluscs, oligochaetes and chironomids [22].

Additionally, stream macroinvertebrate assemblages located in agricultural areas are especially sensitive to habitat deterioration, which is mainly related with an increase in fine sediments [2527]. Fine sediments remove and homogenize habitats [27], leading to reduced taxa richness, especially some EPT species [26,27], and altered assemblage composition [28]. Sedimentation also alters the functional structure of macroinvertebrate assemblages [13,15,25,2931]. Higher amounts of sediment can cover vegetal and mineral coarse substrate leading to an increase in burrowers, and a decrease in crawlers and scrapers [13,15,3033].

The present study assesses the potential impact of agriculture on stream ecosystems around NVZs using the taxonomic and functional composition of macroinvertebrate assemblages as indicators. The Ebro basin was selected as a case study because of being subjected to heavy human pressures, especially agriculture [34,35]. The specific objectives of the study were to: i) determine the main agricultural stressors on streams located in areas affected by intensive agriculture around NVZs; and ii) assess their degree of compliance with the WFD according to physicochemical variables and macroinvertebrate indices. We predicted that these streams would be affected not only by nitrate pollution, but also by an increase in fine sediments and a reduction in riparian and aquatic habitat quality. As a consequence, we expected that agricultural intensification would lead to a shift in the taxonomic composition and functional structure of aquatic macroinvertebrate assemblages, from assemblages dominated by pollution-sensitive species that feed on coarse organic matter (e.g. leaf litter) to assemblages dominated by tolerant species feeding on fine organic matter (e.g. suspended particles) and algae.


Study area and sampling strategy

The study area is located in the NW of La Rioja (northern Spain, Ebro Basin; Fig 1), in the Oja-Tirón, Zamaca and Najerilla sub-basins. To reduce natural differences among sites, the study was carried out in six first-order streams (Fig 1) belonging to the typology "Calcareous Mediterranean mountain rivers" according to the classification developed in Spain to fulfil the WFD (ORDER ARM/2656/2008, Ministry for the Environment). Eleven sites (100 m length) were studied. All of them were free from significant urban or industrial effluents in the adjacent area and they mainly differed in the upstream agrarian activity (i.e. percentage of cultivated land). There are no large villages (i.e. more than 500 inhabitants) in the area and almost no industrial activity. The study sites were selected to fulfill the above-mentioned criteria after a visual inspection of the region. The information on the location of the sampling sites is provided in the supplementary material (Table A in S1 File).

Fig 1. Map of the study area.

Sampling sites in the study area (La Rioja, Spain). Nitrate Vulnerable Zones (NVZs) are coloured in grey (“Zamaca and Oja alluvial” and “Low Najerilla alluvial”).

Well-preserved forests exist in the upper course of Yalde and Reláchigo Streams (Fig 1), where agricultural activity is limited. The cultivated area increases as the streams flow into the Ebro Valley, where intensive farming dominates (mainly cereal, vineyard and horticultural crops). As a result of the intense agricultural activity, two NVZs exist in the study area, called “Zamaca and Oja alluvial” and “Low Najerilla alluvial”. The six streams selected in the present study are the only first-order around these NVZs. According to monitoring programs carried out by the Ebro Hydrographic Confederation (CHE;, nitrate concentrations in the study area had exceeded the highest value established by WFD to consider a river or stream under a good ecological status. However, some of these streams have been studied just sporadically and all of them just at one site.

Each site was sampled four times during 2017 (i.e., February, May, August, and November). Physicochemical variables were analyzed at each sampling occasion, whereas habitat quality and macroinvertebrates were sampled only in May.

Environmental variables

Forty environmental variables belonging to different categories (physicochemical, hydromorphological, geological, land cover and topographic; Table B in S1 File) were studied for each site.

Water temperature (°C), pH, dissolved oxygen (ppm) and electrical conductivity (mS/cm) were measured in situ using handheld probes (PCE-PHD1 multifunctional instrument). For ion analysis, 125 ml water samples were collected and transported in a cool-box to the laboratory, where they were frozen until being analyzed using chromatographic standard methods [36].

The riparian habitat of each site (100 m length) was inspected and characterized using the four components of the QBR Riparian Forest Quality Index by Munné et al. [37]: total riparian vegetation cover, cover structure, cover quality and channel alterations (Table B in S1 File). The fluvial instream habitat was also inspected and characterized using the IHF River Habitat Index [38], which measures habitat heterogeneity by analyzing seven features: hard substrate embeddedness in fine sediment, riffle frequency, substrate composition, velocity/depth conditions, shading of river bed, heterogeneity components (leaf litter, wood, roots or natural dams), and aquatic vegetation cover (Table B in S1 File). Both indexes range from 0 to 100 with high scores associated with better habitat conservation. Additionally, the percentage of macrophyte cover was visually determined.

The geological, land cover and topographic variables were respectively retrieved from the Spanish Geological Survey (IGME), the Spanish Land Cover and Use Information System (SIOSE) and the Spanish National Geographic Institute (IGN) using a Geographic Information System (GIS).

Macroinvertebrate sampling and biotic indices

Macroinvertebrates were collected according to the Standard protocol for benthic invertebrates of wadeable streams in Spain (ML-Rv-I-2013; [37]). This quantitative protocol considers the presence of five different habitat types: hard substrate, fine sediments, plant debris, submerged bank vegetation and submerged macrophytes. Twenty Surber samples (0.1 m2) were taken at each site, proportionally distributed among each type of habitat according to the surface that it occupied within the streambed. The permission to sample was issued by the Government of La Rioja. The collected material was stored in one sample and preserved in 70% ethanol and taken to the laboratory to be identified. The identification of macroinvertebrates was generally done to genus level, except for some Diptera families (subfamily level) and Oligochaeta. Sub-sampling was conducted to estimate the taxa abundances when 300 individuals per sample were counted, although sample exploration continued to check for new taxa.

The IBMWP (Iberian Biological Monitoring Working Party; [39]) and IMMi-T (Iberian Mediterranean Multimetric Index, using quantitative data [40]) biotic indices were calculated using the MAQBIR software [41]. The IBMWP is a qualitative index based on the tolerance of macroinvertebrate families to water quality and river alteration. The presence of each family provides a single score out of 10 (being 1 highly tolerant and 10 highly sensitive) with the cumulative scores providing the final IBMWP score. The IMMi-T is a quantitative multimetric index based on the number of families, EPT, IASPT (Iberian Average Score per Taxon, [42]) and log (selected EPTCD + 1), where EPTCD is the number of sensitive Ephemeroptera, Plecoptera, Trichoptera, Coleoptera and Diptera families. Following the WFD, this index was standardized by calculating the EQR (Ecological Quality Ratio) value dividing the value of each metric by the reference value for Calcareous Mediterranean mountain rivers.

Functional traits

Macroinvertebrate functional structure were characterized using five biological and two ecological traits, containing 35 and 12 categories respectively (Table C in S1 File; [43,44]). Biological traits describe life-cycle features (“reproductive cycles per year” trait), physiology (“respiration”), resilience or resistance potentials (“locomotion and substrate relation”) and feeding behavior (“food, and feeding habits”), which are the four groups of biological features described by Usseglio-Polatera et al. [43]. The ecological traits reflect the organism’s environmental preferences and the behaviours associated with these preferences. Here we used two ecological traits referring to the microscale characteristics of the habitat used by invertebrates (“microhabitat”) and to the trophic status of the freshwater that they inhabit (“trophic status”). The selected traits have been extensively used in previous studies related to agricultural and other land use impacts on rivers (e.g. [15,27,29,30,4446]), as well as fine sediment increase [31]. These ecological traits were useful to assess nutrient and hydromorphological impacts on the streams, whereas the traits “locomotion and substrate relation”, “food” and “feeding habits” have been previously used as effect traits (i.e. those that have a direct influence on a specific function of the ecosystem [4749]).

With this information, a dataset of relative abundance of trait categories per sample was built following a ‘fuzzy coding’ approach [50]. The trait categories [44] were assigned to an affinity score for each taxa ranging from 0 to 5, from null to high affinity, respectively [50].

Statistical analysis

A hierarchical cluster analysis of the macroinvertebrate assemblages (unweighted pair group method using arithmetic averages, UPGMA; PRIMER, [51]) was done to group sites by their similarity. Then, environmental variables were examined within these groups.

An Indicator Species Analysis (IndVal, [52]) was used to determine the invertebrate taxa significantly associated with each of the groups previously established. This method provides an indicator value (IV-value) for each taxon according to its presence and abundance in each group. A Monte Carlo permutation test with 9999 permutations was used to test the significance of each IV-value (P<0.05). For each trait category, a nonparametric pairwise Mann-Whitney U test was performed to detect differences between the groups defined by the cluster analysis. Asymmetric beanplots (software R3.4.2, package beanplot, [53]) were used to visualize the differences in trait abundance between groups.

In order to determine the main environmental variables related to differences in macroinvertebrate assemblages among sites, a DISTLM (distance-based linear model) analysis was performed (PERMANOVA + for PRIMER, [54]) based on genera abundances. DISTLM is a routine for analysing and modelling the relationship between a multivariate data cloud, as described by a resemblance matrix, and one or more predictor variables [54]. In our case, the resemblance matrix described dissimilarities among macroinvertebrate samples of studied sites, based on genus abundance data, and the predictor variables were the environmental variables (Table B in S1 File).

The macroinvertebrate distance matrix was created using the Bray Curtis distance after the data were log (x + 1) transformed. The environmental variables were log (x + 1) transformed and normalized. Variables that were highly correlated with others (Spearman correlation coefficient higher than 0.9) were removed from the analysis. The values of the physicochemical variables included in the analysis were the averages of the four sampling dates since they represent better the physicochemical state of the streams and their influence on the biological assemblages.

DISTLM provides quantitative measures and tests of the variation explained by one or more predictors, allowing the partitioning of variance according to a regression (or multiple regression) model [54]. The DISTLM routine was performed using a step-wise selection procedure based on the AIC selection criteria [55], to obtain a set of variables plausibly explaining changes in the composition of the macroinvertebrate assemblages. From every variable included in the final model just some of them significantly explain the variation in the aquatic macroinvertebrate assemblages between sites differentially affected by agriculture. The final model from DISTLM analysis was visualized by a distance-based Redundancy Analysis (dbRDA) plot. The dbRDA routine performs a constrained ordination of the macroinvertebrate assemblages data using the DISTLM final model.


Biotic, environmental and hydromorphological characteristics

Two groups of sampling sites were differentiated based on the hierarchical cluster analysis of the macroinvertebrate assemblages (Fig 2). The two groups presented a 60% of dissimilarity and separated sites with low (<50%) and high (>70%) agricultural -or cultivated- land cover upstream the river reach (Table 1).

Fig 2. Hierarchical cluster.

Hierarchical cluster analysis (UPGMA) based on the composition of the macroinvertebrate community of each site. The resultant groups are identified with black triangles (group 1) and grey inverted triangles (group 2).

Table 1. Environmental variables.

Values of the physicochemical variables and the biotic indices -IBMWP (Iberian Biological Monitoring Working Party) and IMMi-T (Iberian Mediterranean Multimetric Index)- used to assess the good ecological status of streams according to the Spanish legislation (RD 817/2015), based on the European Water Framework Directive (WFD). The limits established by the Spanish legislation to consider a water body at good ecological status or higher for “Calcareous Mediterranean mountain rivers” are shown in parentheses. For chemical variables, mean values and ranges (i.e. minimum and maximum) are shown. Values exceeding the threshold for the good ecological status according to the RD 817/2015 are highlighted in bold. Asterisk indicate that although the mean values were below the threshold, the threshold was exceeded at least once during the study period. The lowest detectable values for phosphate and ammonia were 0.010 and 0.020, respectively. IHF (River Habitat Index) and QBR (Riparian Forest Quality Index) values, which are not considered by RD 817/2015 to determine the good ecological status of a water body, are shown at the right part along with the percentage of area occupied by agricultural land upstream the study site (AGR). The ranges for IHF and QBR are shown in parentheses, with higher scores associated to better habitat conservation. The lowest values of IHF (<50) and QBR (<50) and the highest values of AGR (≥70) are highlighted in bold.

Regarding the environmental variables studied and included in the WFD (Table 1), nitrate showed particularly high concentrations at different dates and sites. Considering average values, 4 sites exceeded 25 ppm, the highest value established by RD 817/2015—Spanish legislation based on WFD—to consider a river or stream in a good ecological status (R3, Ea, Z1 and Z2; Table 1). However, there was a great variability in nitrate concentrations among dates, as a concentration of 25 ppm was exceeded at least once in every site of group 2 (Ea, Z1, Z2, T1 and T2) during the study (Table 1). The lowest values of the IBMWP and IMMi-T biotic indices were recorded at group 2, and they were below the good ecological status threshold for Z1, T1 and T2 (Table 1).

Sites of group 2 also showed the lowest values of the IHF and QBR indices (Table E in S1 File). The first component of the IHF, referred to the embeddedness of hard substrate in fine sediment (Table E in S1 File), had a score of 0 at every site of group 2, since more than 60% of boulders, cobbles and pebbles were embedded in fine sediment. Macrophytes occupied between 30 and 60% of the streambed at all study sites, except for R2 (20%) and for Y1, Y2 and R1, where macrophyte cover was less than 5% (Table E in S1 File).

Among the other environmental variables studied, the average conductivity resulted particularly high at Enc site (mean conductivity 2077 μS/cm). Accordingly, exceptionally high values of calcium and sulfate were recorded at this reach (Enc; sulfate = 896.79 ppm; calcium = 575.52 ppm; Table C in S1 File).

Macroinvertebrate assemblage and agriculture stressors

Group 1 was mainly characterized by EPT taxa such as Polycentropodidae (Plectrocnemia sp.), Heptageniidae (Ecdyonurus sp. and Rithrogena sp.), Leptophlebiidae and Leuctridae (Leuctra sp.), which are sensitive to pollution according to the IBMWP index (Table 2). Other taxa associated with this group were Gammaridae, Ceratopogonidae and the beetles Esolus and Elmis (Elmidae). Group 2 was significantly associated with taxa with a low IBMWP score (Table 2), like Oligochaeta and Chironomidae, together with the dipterans Simuliidae and Psychodidae and other taxa such as Hydroptilidae or Haliplidae.

Table 2. IndVal analysis.

IndVal analysis results showing taxa significantly associated with one of the previously established groups (Fig 2). For those taxa included in the IBMWP index, their score is shown. # means that the whole family was significantly associated with one of the groups, but no genus within the family was.

According to the Mann Whitney test, functional trait categories significantly differed between groups 1 and 2 (Fig 3). Group 1 was characterized by mono- or semivoltine taxa, gill and plastron respiration, crawlers living on large substrate, shredder and scrapers feeding on microphytes, and taxa adapted to oligotrophic environments. Group 2 was characterized by polyvoltine taxa, living and feeding on fine sediment and macrophytes and adapted to eutrophic conditions (Fig 3).

Fig 3. Functional traits analysis.

Asymmetric beanplots of the functional trait categories significantly different (Mann Whitney U test, P<0.05) between the previously established groups (Fig 2). The individual observations of each site are shown as small white lines and the average of each group is shown as a black line.

The final model of the DISTLM explained 99.2% of the total variance and it included the following significant (p<0.05) explanatory variables (ranked in order of importance): total value of the River Habitat Index (IHF; 43.9% of total variance explained), percentage of area occupied by agricultural land upstream the river reach (11.8%), nitrate concentration (9.4%), the ratio of valley width to channel width (8.6%) and the component of the riparian habitat index (QBR) referring to channel alterations (7.3%). Other variables included in the final model were not significant (p>0.05): the component of the QBR index referring to riparian vegetation cover (7.2%), percentage of land occupied by pasture upstream (4.3%), chloride concentration (4.2%) and macrophyte cover (2.2%) (Fig 4).

Fig 4. DISTLM analysis.

DISTLM analysis showing the dissimilarities between the macroinvertebrate assemblages of the studied sites, and predictor variables included in the final model -significant variables (p<0.05) are highlighted in bold-. Sites previously grouped by cluster analysis are shown with different symbols (black triangles: group 1; grey inverted triangles: group 2). IHF: Total score of the River Habitat Index; QBR-natur: component of the QBR referred to river channel alteration; NO3-: nitrate concentration; Cl: chloride concentration AGR: area occupied by agricultural land upstream the study site; VWI: Valley Width Index or ratio of valley width to channel width; PAS: area occupied by pasture upstream the study site.


Our results suggest that different stressors associated with agricultural activities could have a significant impact on the macroinvertebrate assemblages in the study region. As we hypothesized, and in agreement with previous studies [11,13,15,26], agricultural intensification led to a replacement of EPT taxa by tolerant taxa such as Oligochaeta or Chironomidae. Likewise, and according to our hypothesis, streams were affected not only by nitrate pollution, but also by hydromorphological stressors like the alteration of riparian and aquatic habitat, leading to functional and taxonomic changes in the macroinvertebrate assemblages. Overall, sites with a higher upstream area occupied by agricultural land were dominated by polyvoltine taxa, a biological trait that has been associated with resistance strategies in human impacted sites [29,56].

Concentrations of nitrate significantly contributed to alter macroinvertebrate assemblages in the studied streams, as it has been observed in other agricultural areas [13,21]. Nitrate is able to convert oxygen-carrying pigments of aquatic animals to forms that are incapable of carrying oxygen (e.g. methemoglobin; [18,57]). In this regard, the mean nitrate concentration registered at seven of our study sites could be enough to adversely affect, at least during long-term exposures, freshwater invertebrates [18, 21]. However, nitrate uptake in aquatic animals is limited [58] and nitrate impact in aquatic ecosystems can be related to other processes, like an increase in autotrophic production [4]. Accordingly, macrophyte cover was higher in the most agricultural impacted sites. This could be associated not only with a higher nitrate concentration, but also with the degradation of the riparian forest allowing more light to penetrate the streambed [59]. Overall, in these highly impacted streams, taxa living and feeding on macrophytes and adapted to eutrophic conditions were favored.

The variation in aquatic macroinvertebrate assemblages between sites was best explained by variables associated with the hydromorphological alteration of the stream habitat and with the riparian habitat. Specifically, the total IHF index, the QBR components that refer to riparian vegetation cover and river channel alteration and the ratio of valley width to channel width (VWI) were responsible for 67% of the variance in the macroinvertebrate assemblages. Different authors have pointed out that physical habitat degradation can have a greater influence on aquatic macroinvertebrates than the exposure to chemical pollution. For instance, Genito et al. [10] suggested that a high percentage of agricultural land cover caused a macroinvertebrate assemblage composition that reflected stream habitat degradation, whereas nitrate concentration had a weaker association with assemblage alteration. Shields et al. [60] pointed out the importance of stream habitat conservation in agricultural areas, especially related to channel degradation, large wood removal or increased deposition of fine sediments. Accordingly, fine sediment has been determined as a more pervasive stressor to macroinvertebrate assemblage than increased nutrient concentrations in streams around agricultural areas [27] and under controlled experimental conditions [15]. Beyond agricultural activities, several studies have confirmed the strong impacts of fine sediments on stream invertebrates (e.g. [6164]).

Our results show that intensive agriculture can alter the streambed, simplify the stream habitats (i.e. high substrate embeddedness in fine sediment, low heterogeneity of substrates) and deteriorate the riparian forest (Fig 5). This can have important impacts on the aquatic ecosystem. For example, according to Genito et al. [10] and Piscart et al. [12], the reduction of riparian forests decreased the abundance of shredders in sites affected by intensified agricultural activity, leading to a lower litter breakdown rate. Riparian alteration also enhances, together with the intensive soil tillage, the amount of sediment entering the streams [4]. Although we did not measure sediment fluxes directly, the importance of the IHF index (both the overall score and the 1st component, referred to fine sediment accumulation) suggests that sedimentation was an important explanatory variable of the functional structure of the macroinvertebrate assemblages in the study sites. Increased stream sedimentation has often been associated with nonpoint sources arising from agricultural land uses and it is mainly related to the loss of riparian vegetation [27].

Fig 5.

Photographs of the studied sites with the highest (white acronyms) and the lowest (black acronyms) QBR and IHF total values. Y1 (QBR = 80; IHF = 71), R1 (75; 71), T1 (10; 29), T2 (35; 31) Z1 (10; 32), Z2 (10; 36).

In the present study, the greatest boulders, cobbles and pebbles embeddedness in fine sediment (i.e. component 1 of IHF index) were found at sites affected by a greater agricultural activity (Z1, Z2, T1, T2 and Ea). At these sites, there were less macroinvertebrates adapted to move on large substrate (i.e. crawlers), maybe because fine sediment decreased habitat heterogeneity and availability [27]. On the contrary, there were more invertebrates living in the mud and burrowers, as it has been previously reported for rivers experiencing sediment increases [15,30]. Agricultural intensification also favored organisms that fed on fine sediment, negatively affecting scrapers, as it was found by previous other studies [32,33]. This could be related to the sediment covering large substrates, thereby reducing microphytes’ availability for stream fauna [65]. The abundance of shredders was also reduced under agricultural intensification. As suggested by Burdon et al. [27], this is most likely due to degraded riparian habitats (Fig 5), since shredders do not seem to be directly affected by sediment deposition [33]. Finally, respiration by plastron and gills were significantly less frequent in sites surrounded by a greater agricultural area. For example, we observed a decrease in Elmis and Esolus, which have larvae with anal gills and aquatic adults with plastron respiration [44,45]. Concordantly, species of Coleoptera dependent on a bubble or plastron to breath have shown to be sensitive to high sediment concentration in water [46].

Several of the studied sites did not meet the legal requirements to be considered in good ecological status according to the WFD. Three sites did not reach the good ecological status based on the IBMWP and IMMi-T biotic indices. Nitrate exceeded the concentration for a good ecological status at seven sites, despite all of them (except for Z1) were outside of NVZs. Accordingly, the official reports point to nitrate as the main agricultural stressor in the studied streams, although they only suggest the implementation of “good agricultural practices” at a few sites (and these practices are not detailed [66]). However, the WFD indicators seem to underestimate hydromorphological alterations, since they were not specifically considered when determining the ecological status of the water bodies. According to Houlden [67], hydromorphology is a “supporting element” in the WFD, which is only used to confer the high ecological status (i.e. a site cannot have the highest ecological status if it is hydromorphologically altered). As a consequence, despite the potential effect of hydromorphological alteration on aquatic macroinvertebrates, it has been largely overlooked by water managers within the Ebro basin [66].

Finally, it should be noted that the increased salinization, associated with fertilizers and inefficient irrigation in agricultural areas [68], should also be considered as a potential source of stress for stream organisms. However, in the study area, conductivity was below 1400 μS/cm at every site, except at Encemero stream (2078 ± 158 μS/cm). The high conductivity in Encemero is naturally associated with sodium sulfate tertiary deposits and secondary minerals, mainly gypsum and calcite [69]. Thus, our results suggest that secondary salinization is not severe in the study area and it probably had a weak effect on aquatic macroinvertebrates given the conductivity values recorded in this study [70].

Conclusions and recommendations

Aquatic macroinvertebrates seem to be useful indicators of agricultural impacts, since both the taxonomic composition and mainly the functional structure responded to agricultural intensification. Intensive agriculture around NVZs in the study area significantly altered the stream habitats and its associated macroinvertebrate assemblages. High nitrate content could be partially responsible for such alteration, but hydromorphological impacts, especially the increase in the amount of sediment that enters the streams, seem to be the main driver behind stream degradation. The macroinvertebrate assemblage’s alteration caused by intensive agriculture includes significant changes in effects traits (locomotion, food and feeding habits), which can directly affect certain ecosystem services and properties as ecosystem engineering, nutrient cycling and resource processing [71,72].

The implementation of the ND and the WFD do not seem to be enough to prevent the degradation of streams in agricultural areas. As other authors did [5,73], we suggest to extend the NVZs in the study area since nitrate exceeded the limits established by the WFD to achieve the good ecological status in seven of the studied sites (despite 6 of them are located upstream of NVZs). Moreover, farming practices established by the regional administration in NVZs have not been effective to reduce nitrate pollution in these areas [74]. Thus, the compliance of these practices should be more rigorously controlled. Accordingly, Worral et al. [75] showed little improvements in nitrate content in surface water concentration of NVZs after 12–15 years in UK, although several limitations to assess the effectiveness of NVZs have been acknowledged [76,77]. It is necessary to improve water quality monitoring programs to identify and quantify the main pollution sources [78], which could be related to the increase of arable lands in the catchment [75]. Current water quality monitoring programs are not able to identify the hydromorphological stressors existing in the area. A proper monitoring of these stressors should be designed, and adequate practices should be implemented to achieve a good ecological status.

Supporting information

S1 File.

Location of the sampling sites (Table A); Environmental variables studied at each site, and categories and codes used in the present study (Table B); Functional (biological and ecological) traits and categories studied according to Tachet et al. (2006) (Table C); Mean values (for four sampling dates) and ranges (i.e. minimum and maximum) of the physicochemical variables not included in Table 1 (Table D); Values of hydromorphological variables determined in the May sampling campaign and not included in Table 1 (components of IHF -River Habitat Index-, QBR -Riparian Forest Quality Index- and percentage of macrophytes cover) (Table E); Values of topographic, geologic and land cover variables not included in Table 1 (Table F); Density (ind/m2) of macroinvertebrate taxa at each site (Table G); Relative abundance of each macroinvertebrate functional trait at study sites (Table H).



  1. 1. Flávio HM, Ferreira P, Formigo N, Svendsen JC. Reconciling agriculture and stream restoration in Europe: a review relating to the EU Water Framework Directive. Sci Total Environ. 2017;596:378–95. pmid:28448914
  2. 2. IPBES. Summary for policymakers of the global assessment report on biodiversity and ecosystem services of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services. Advance Unedited Versión. 2019 May 6.
  3. 3. Haas MB, Guse B, Fohrer N. Assessing the impacts of Best Management Practices on nitrate pollution in an agricultural dominated lowland catchment considering environmental protection versus economic development. J Environ Manage. 2017;196:347–64. pmid:28324851
  4. 4. Allan JD. Landscaps and Riverscapes: The Influence of Land Use on Stream Ecosystems. Annu Rev Ecol Evol Syst. 2004;35:257–84.
  5. 5. Arauzo M, Valladolid M. Drainage and N-leaching in alluvial soils under agricultural land uses: Implications for the implementation of the EU Nitrates Directive. Agric Ecosyst Environ. 2013;179:94–107.
  6. 6. Voulvoulis N, Arpon KD, Giakoumis T. The EU Water Framework Directive: from great expectations to problems with implementation. Sci Total Environ. 2017;575:358–66. pmid:27744201
  7. 7. Ector L, Rimet F. Using bioindicators to assess rivers in Europe: An overview. In: Lek S, Scardi M, Verdonschot PFM, Descy JP, Park YS, editors. Modelling community structure in freshwater ecosystems. Berlin: Springer; 2005. p. 7–19.
  8. 8. Bonada N, Prat N, Resh VH, Statzner B. Developments in aquatic insect biomonitoring: a comparative analysis of recent approaches. Annu Rev Entomol. 2006;51:495–523. pmid:16332221
  9. 9. Birk S, Bonne W, Borja A, Brucet S, Courrat A, Poikane S, et al. 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.
  10. 10. Genito D, Gburek WJ, Sharpley AN. Response of stream macroinvertebrates to agricultural land cover in a small watershed. J Freshw Ecol. 2002;17(1):109–19.
  11. 11. Song MY, Leprieur F, Thomas A, Lek-Ang S, Chon TS, Lek S. Impact of agricultural land use on aquatic insect assemblages in the Garonne river catchment (SW France). Aquat Ecol. 2009;43(4):999–1009.
  12. 12. Piscart C, Genoel R, Doledec S, Chauvet E, Marmonier P. Effects of intense agricultural practices on heterotrophic processes in streams. Environ Pollut. 2009;157(3):1011–18. pmid:19028003
  13. 13. Magbanua FS, Townsend CR, Blackwell GL, Phillips N, Matthaei CD. Responses of stream macroinvertebrates and ecosystem function to conventional, integrated and organic farming. J Appl Ecol. 2010;47(5):1014–25.
  14. 14. Moore AA, Palmer MA. Invertebrate Biodiversity in Agricultural and Urban Headwater Streams: Implications for Conservation and Management. Ecol Appl. 2012;15(4):1169–77.
  15. 15. Wagenhoff A, Townsend CR, Matthaei CD. Macroinvertebrate responses along broad stressor gradients of deposited fine sediment and dissolved nutrients: A stream mesocosm experiment. J Appl Ecol. 2012;49(4):892–902.
  16. 16. Statzner B, Dolédec S, Hugueny B. Biological trait composition of European stream invertebrate communities: assessing the effects of various trait filter types. Ecography. 2004;27(4):470–88.
  17. 17. Statzner B, Bonada N, Dolédec S. Conservation of taxonomic and biological trait diversity of European stream macroinvertebrate communities: a case for a collective public database. Biodivers Conserv. 2007;16(12):3609–32.
  18. 18. Camargo JA, Alonso A, Salamanca A. Nitrate toxicity to aquatic animals: A review with new data for freshwater invertebrates. Chemosphere. 2005;58(9):1255–67. pmid:15667845
  19. 19. Camargo JA, Alonso Á. Ecological and toxicological effects of inorganic nitrogen pollution in aquatic ecosystems: A global assessment. Environ Int. 2006;32(6):831–49. pmid:16781774
  20. 20. Liess A, Le Gros A, Wagenhoff A, Townsend CR, Matthaei CD. Landuse intensity in stream catchments affects the benthic food web: consequences for nutrient supply, periphyton C:nutrient ratios, and invertebrate richness and abundance. Freshw Sci. 2012;31(3):813–24.
  21. 21. Pearson CE, Ormerod SJ, Symondson WOC, Vaughan IP. Resolving large-scale pressures on species and ecosystems: propensity modelling identifies agricultural effects on streams. J Appl Ecol. 2016;53(2):408–17. pmid:27867215
  22. 22. Harding JS, Young RG, Hayes JW, Shearer KA, Stark JD. Changes in Agricultural Intensity and River Health Along a River Continuum. Freshw Biol. 1999;(42):345–57.
  23. 23. Rixen T, Baum A, Sepryani H, Pohlmann T, Jose C, Samiaji J. Dissolved oxygen and its response to eutrophication in a tropical black water river. J Environ Manage. 2010;91(8):1730–7. pmid:20435403
  24. 24. Jacobsen D, Rostgaard S, Vásconez JJ. Are macroinvertebrates in high altitude streams affected by oxygen deficiency? Freshw Biol. 2003;48(11):2025–32.
  25. 25. Magbanua FS, Townsend CR, Hageman KJ, Matthaei CD. Individual and combined effects of fine sediment and the herbicide glyphosate on benthic macroinvertebrates and stream ecosystem function. Freshw Biol. 2013;58(8):1729–44.
  26. 26. Piggott JJ, Lange K, Townsend CR, Matthaei CD. Multiple stressors in agricultural streams: a mesocosm study of interactions among raised water temperature, sediment addition and nutrient enrichment. PLoS One. 2012;7(11):e49873. pmid:23185471
  27. 27. Burdon FJ, McIntosh AR, Harding JS. Habitat loss drives threshold response of benthic invertebrate communities to deposited sediment in agricultural streams. Ecol Appl. 2013;23(5):1036–47. pmid:23967573
  28. 28. Larsen S, Vaughan IP, Ormerod SJ. Scale-dependent effects of fine sediments on temperate headwater invertebrates. Freshw Biol. 2009;54(1):203–19.
  29. 29. Dolédec S, Phillips N, Scarsbrook M, Riley RH, Townsend CR. Comparison of structural and functional approaches to determining landuse effects on grassland stream invertebrate communities. J North Am Benthol Soc. 2006;25(1):44–60.
  30. 30. Lange K, Townsend CR, Matthaei CD. Can biological traits of stream invertebrates help disentangle the effects of multiple stressors in an agricultural catchment? Freshw Biol. 2014;59(12):2431–46.
  31. 31. Murphy JF, Jones JI, Arnold A, Duerdoth CP, Pretty JL, Naden PS, et al. Can macroinvertebrate biological traits indicate fine‐grained sediment conditions in streams? River Res Appl. 2017;33(10):1606–17.
  32. 32. Burdon FJ, McIntosh AR, Harding JS. Mechanisms of trophic niche compression: evidence from landscape disturbance. bioRxiv. 2018;329623.
  33. 33. Rabení CF, Doisy KE, Zweig LD. Stream invertebrate community functional responses to deposited sediment. Aquat Sci. 2005;67(4):395–402.
  34. 34. Rodríguez-Martín JA, Arias ML, Grau-Corbí JM. Heavy metals contents in agricultural topsoils in the Ebro basin (Spain). Application of the multivariate geoestatistical methods to study spatial variations. Environ Pollut. 2006;144(3):1001–12. pmid:16580763
  35. 35. Mondy CP, Muñoz I, Dolédec S. Life-history strategies constrain invertebrate community tolerance to multiple stressors: A case study in the Ebro basin. Sci Total Environ. 2016;572:196–206. pmid:27498381
  36. 36. ASTM. Standard Methods for the Examination of Water and Wastewater 19th ed. Washington, DC.: American Public Health Association, American Water Works Association, and Water Environment Federation (APHA, AWWA, and WEF); 1995.
  37. 37. Munné A, Prat N, Solá C, Bonada N, Rieradevall M. A simple field method for assessing the ecological quality of riparian habitat in rivers and streams: QBR index. Aquat Conserv Mar Freshw Ecosyst. 13(2):147–63.
  38. 38. Pardo I, Álvarez M, Casas J, Moreno JL, Vivas S, Bonada N, et al. El hábitat de los ríos mediterráneos. Diseño de un índice de diversidad de hábitat. Limnetica. 2002;21(3–4):115–33.
  39. 39. Alba-Tercedor J, Jáimez-Cuéllar P, Álvarez M, Avilés J, Bonada N, Casas J, et al. Caracterización del estado ecológico de ríos mediterráneos ibéricos mediante el índice IBMWP (antes BMWP’). Limnetica. 2002;21(3–4):175–85.
  40. 40. Munné A, Prat N. Use of macroinvertebrate-based multimetric indices for water quality evaluation in Spanish Mediterranean rivers: an intercalibration approach with the IBMWP index. Hydrobiologia. 2009;628(1):203–25.
  41. 41. Munné A. Análisis de métriques d´avaluació de la qualitat de l´aigua mitjançant l´us de macroinvertebrats cm a bioindicadors y ls seva resposta a les alteracions antrópiques i a la tipologia fluvial en ecosistemes mediterranis. Tesis Doctoral. Departament d`Ecologia. Universitat de Barcelona; 2009.
  42. 42. Bonada N, Prat N, Munné A, Plans M, Solà C, Álvarez M, et al. Intercalibración de la metodología GUADALMED. Selección de un protocolo de muestreo para la determinación del estado ecológico de los ríos mediterráneos. Limnetica. 2002;21(3–4):13–33.
  43. 43. Usseglio-Polatera P, Bournaud M, Richoux P, Tachet H. Biological and ecological traits of benthic freshwater macroinvertebrates: Relationships and definition of groups with similar traits. Freshw Biol. 2000;43(2):175–205.
  44. 44. Tachet H, Richoux P, Bournaud M, Usseglio-Polaterra P. Invertébrés d’eau douce. Systématique, biologie, écologie. Paris: CNRS Editions; 2006.
  45. 45. Elliott JM. The Ecology of Riffle Beetles (Coleoptera: Elmidae). Freshw Rev. 2008;1(2):189–203.
  46. 46. Hauer FR, Resh VH. Benthic Macroinvertebrates. In: Hauer FR, Lamberti GA, editors. Methods in Stream Ecology. San Diego: Academic Press; 1996. p. 339–65.
  47. 47. Belmar O, Bruno D, Guareschi S, Mellado-Díaz A, Millán A, Velasco J. Functional responses of aquatic macroinvertebrates to flow regulation are shaped by natural flow intermittence in Mediterranean streams. Freshw Biol. 2019;64(5):1064–77.
  48. 48. Schmera D, Heino J, Podani J, Erős T, Dolédec S. Functional diversity: a review of methodology and current knowledge in freshwater macroinvertebrate research. Hydrobiologia. 2017;787(1):27–44.
  49. 49. Hevia V, Martín-López B, Palomo S, García-Llorente M, de Bello F, González JA. Trait-based approaches to analyze links between the drivers of change and ecosystem services: Synthesizing existing evidence and future challenges. Ecol Evol. 2017;7(3):831–44. pmid:28168020
  50. 50. Chevenet F, Dolédec S, Chessel D. A fuzzy coding approach for the analysis of long-term ecological data. Freshw Biol. 1994;31:295–309.
  51. 51. Clarke KR, Gorley RN. PRIMER v6: User manual/tutorial. Plymouth. United Kingdom: PRIMER-E Ltd.; 2006.
  52. 52. Dufrêne M, Legendre P. Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecol Monogr. 1997;67:345–66.
  53. 53. Kampstra P. Beanplot: A Boxplot Alternative for Visual Comparison of Distributions. J Stat Softw. 2008;28(1):1–9.
  54. 54. Anderson MJ, Gorley RN, Clarke KR. PRIMER + for PERMANOVA: Guide to Software and Statistical Methods. Plymouth, United Kingdom: PRIMER-E. Ltd; 2008.
  55. 55. Akaike H. Information theory and an extension of the maximum likelihood principle. In: Petrov BN, Csake F, editors. Second International Symposium on Information Theory. Budapest: Akademiai Kiado; 1973. p. 267–281.
  56. 56. Bogan MT, Boersma KS, Lytle DA. Resistance and resilience of invertebrate communities to seasonal and supraseasonal drought in arid-land headwater streams. Freshw Biol. 2014;60(12):2547–58.
  57. 57. Cheng SY, Chen JC. Study on the oxyhemocyanin, deoxyhemocyanin, oxygen affinity and acid-base balance of Marsupenaeus japonicus following exposure to combined elevated nitrite and nitrate. Aquat Toxicol. 2002;61(3–4):181–93. pmid:12359389
  58. 58. Jensen FB. Uptake, elimination and effects of nitrite and nitrate in freshwater crayfish (Astacus astacus). Aquat Toxicol. 1996;34:95–104.
  59. 59. Finlay JC. Stream size and human influences on ecosystem production in river networks. Ecosphere. 2011;2(8):art 87.
  60. 60. Shields FD, Langendoen EJ, Doyle MW. Adapting existing models to examinate effects of agricultural conservation programs on stream habitat quality. J Am Water Resour Assoc. 2006;42:25–33.
  61. 61. Beermann AJ, Elbrecht V, Karnatz S, Ma L, Matthaei CD, Piggott JJ, et al. Multiple-stressor effects on stream macroinvertebrate communities: A mesocosm experiment manipulating salinity, fine sediment and flow velocity. Sci Total Environ. 2018;610:961–71. pmid:28830056
  62. 62. Gieswein A, Hering D, Lorenz AW. Development and validation of a macroinvertebrate-based biomonitoring tool to assess fine sediment impact in small mountain streams. Sci Total Environ. 2019;652:1290–301. pmid:30586815
  63. 63. Brown LE, Aspray KL, Ledger ME, Mainstone C, Palmer SM, Wilkes M, et al. Sediment deposition from eroding peatlands alters headwater invertebrate biodiversity. Glob Chang Biol. 2019;25(2):602–19. pmid:30414307
  64. 64. Sánchez-Morales M, Sabater F, Muñoz I. Effects of urban wastewater on hyporheic habitat and invertebrates in Mediterranean streams. Sci Total Environ. 2018;642:937–45. pmid:29929145
  65. 65. Jones JI, Murphy JF, Collins AL, Sear DA, Naden PS, Armitage PD. The impact of fine sediment on macro‐invertebrates. River Res Appl. 2012;28(8):1055–71.
  66. 66. Ebro Hydrographic Confederation. Plan Hidrológico del Ebro. Anexo 4.1. Estado y objetivos medioambientales de las masas de agua y fichas justificativas. Zaragoza, 2015.
  67. 67. Houlden V. Hydromorphology, the forgotten facet of the Water Framework Directive. HR Wallinford. 2018 Sep 13.
  68. 68. Estévez E, Rodríguez-Castillo T, González-Ferreras AM, Cañedo-Argüelles M, Barquín J. Drivers of spatio-temporal patterns of salinity in Spanish rivers: a nationwide assessment. Philos Trans R Soc B Biol Sci. 2018;374(1764):20180022.
  69. 69. Menduiña J, Ordóñez S, del Cura MG. Geología del yacimiento de glauberita de Cerezo de Rio Tirón (Burgos). Bol Geol Min. 1984;95(1):35–51.
  70. 70. Cañedo-Argüelles M, Kefford BJ, Piscart C, Prat N, Schäfer RB, Schulz CJ. Salinisation of rivers: An urgent ecological issue. Environ Pollut. 2013;173:157–67. pmid:23202646
  71. 71. Covich AP, Palmer MA, Crowl TA. The role of benthic invertebrate species in freshwater ecosystems: Zoobenthic species influence energy flows and nutrient cycling. Bioscience. 1999;49(2):119–27.
  72. 72. Statzner B. Geomorphological implications of engineering bed sediments by lotic animals. Geomorphology. 2012;157–158:49–65.
  73. 73. Arauzo M, Valladolid M, Martínez-Bastida JJ. Spatio-temporal dynamics of nitrogen in river-alluvial aquifer systems affected by diffuse pollution from agricultural sources: Implications for the implementation of the Nitrates Directive. J Hydrol. 2011;411(1–2):155–68.
  74. 74. Gobierno de La Rioja. Informe sobre la aplicacion de la Directiva 91/676/CEE, relativa a la protección de las aguas contra la contaminación producida por nitratos procedentes de fuentes agrarias, correspondiente al cuatrienio 2012–2015 en la Comunidad Autónoma de La Rioja. Logroño, 2015.
  75. 75. Worrall F, Spencer E, Burt TP. The effectiveness of nitrate vulnerable zones for limiting surface water nitrate concentrations. J Hydrol. 2009;370(1–4):21–8.
  76. 76. Howden NJK, Burt TP, Worrall F, Whelan MJ. Monitoring fluvial water chemistry for trend detection: Hydrological variability masks trends in datasets covering fewer than 12 years. J Environ Monit. 2011;13(3):514–21. pmid:21347486
  77. 77. Weatherhead EK, Howden NJK. The relationship between land use and surface water resources in the UK. Land use policy. 2009;26:243–50.
  78. 78. Kay P, Grayson R, Phillips M, Stanley K, Dodsworth A, Hanson A, et al. The effectiveness of agricultural stewardship for improving water quality at the catchment scale: Experiences from an NVZ and ECSFDI watershed. J Hydrol. 2012;422:10–6.