The San Xusto SHP station is situated in the middle reaches of the Lérez River, in the southwest of Galicia. The climate in the area is oceanic: the average annual rainfall is between 1246 mm (Pontevedra) and 2000 mm (Campolameiro) and the average temperature is 14.8 °C [9,19].

River Lérez belongs to the Atlantic watershed of Galicia, and drains into the Ría de Pontevedra. It has a total length of 57 km [20] and its flow is 14.4 m³/s 3 km downstream the SHP station and 21.2 m³/s in the river mouth [21]. The flow regime of the water flows in Galicia is related with the rainfall distribution [10]. The periods of highest flow is between December and March, and minimum occurs in September and the bedrock of the watershed is mainly granitic [21] leading to low water conductivity and low pH [22]. Natural vegetation in the riversides is Atlantic deciduous forest with oak (Quercus robur), black alder (Alnus glutinosa), willow (Salix atrocinerea), hazel tree (Corylus avellana), elder (Sambucus nigra) and ash (Fraxinus sp.). This forest is narrow and restricted to the river banks due to the pressure of other productive uses of the land as eucalyptus (Eucalyptus globulus) plantations and pastures, as well as the spread of invasive species as Acacia melanoxylon and Acacia dealbata.

The Lérez stream has been classified as salmonid waters [23], i.e., waters which support or become capable of supporting fish belonging to species such as salmon (Salmo salar), trout (Salmo trutta), or other species in other areas of Europe (Directive 2006/44/EC [24]). Currently, there is a population of salmon inhabiting the Lérez after being close to extinction early in the 1990s [25]. The stretch of the river where the San Xusto SHP plant is situated has been classified as CEDEX type 21: Cantabric-Atlantic siliceous rivers [26].

The human population density in municipality where this reaches of the river belongs is 32.1–32.9 inhabitants/km², but the neighbouring municipality of Pontevedra has 693 inhabitats/km² [27], so a recreational use of the river downstream the SHP station as well as a conservationist concern exists. Some fishing grounds exist both downstream and upstream the dam [21].

The San Xusto SHP station is made up of a diversion dam, to block the river and divert the water, a pipeline which draws the water from a high level to the powerhouse, a penstock and a powerhouse building (Figure 1).

The dam has a fish ladder, which was completed in November 2006. The dam (UTM: 29T 541585; 4708041; 153 m above sea level (a.s.l.)) coronation has a length of 53 m and a height of 164 m a.s.l. and 11.05 m above the river. The surface of the water in the reservoir is 160 m high. The diversion pipe is 1960 m long and the penstock is 116 m long. The length of the river stretch between the dam and the tailrace is 3400 m. The power house (UTM: 29T 539412; 4708029; 83 m a.s.l.) has 2 turbines of 10 m³/s and 5.91 MW each. The gross head is 70 m. The discharge is 20 m³/s. and the station takes 59% of the water flow. The power is 11.81 MW and the total production is 37.81 GWh/year.

The Water Framework Directive 2000/60/EC (WFD) [17] was created to maintain and improve the aquatic environment in the EU countries. According to its article 2, the surface water status should be determined by its ecological and its chemical status. WFD establish that Member States shall ensure the establishment of programs to monitor the water status, which for surface waters shall cover: the volume and level of flow and the ecological and chemical status. WFD Annex V of [17] include the quality elements suggested for the classification of the ecological status of the surface water, including biological, hydromorphological and physicochemical elements, as well as criteria for definition of ecological status using that elements. Although these programs should be developed by the regional administrations, the indicators established by the WFD are useful to assess the water status that should be ideally achieved in reaches affected by small hydropower. Consequently, some of the water quality elements, established by WFD, have been taken in consideration in this work.

In order to describe the quality of the aquatic ecosystem in a river stretch including the San Xusto SHP station of the Lérez River, the physicochemical characteristics of the water, the biological quality of the water, and the ecological quality of riparian habitat were considered (Figure 2).

Between several chemical and physicochemical elements described in Annex V of WFD, the following physicochemical parameters were considered: water temperature (°C), conductivity (mS/cm), dissolved oxygen (DO, mg/L), and pH. It was used a handheld multiparameter instrument YSI 556MPS (YSI, Yellow Springs, OH, USA), accompanied by the following sensors types: YSI Temperature Precision™ thermistor, 4-electrode cell with auto ranging, steady state polarographic, platinum button, glass combination electrode (see specifications in [28]). According to WFD, physicochemical conditions are considered good if the parameters do not reach levels outside the range established to ensure the functioning of the specific type of ecosystem. Then, the ranges established for salmonid waters in Directive 2006/44/EC [24] on the quality of fresh waters needing protection or improvement in order to support fish life were considered. The values registered for the parameters were compared with those registered in the closest station belonging to the Water Quality Control Network from Aguas de Galicia, as well as with the reference condition in Cantabric-Atlantic siliceous rivers (CEDEX type 21: Instrucciones de Planificación Hidrológica, henceforth IPH 2008 [29]).

According to WFD, biological elements that should be taken in consideration in order to define the ecological quality are: composition and abundance of aquatic flora, composition and abundance of benthic invertebrate fauna and composition, abundance and age structure of fish fauna. The biological quality of the water was assessed using the index Iberian Biological Monitoring Working Party IBMWP [30,31]. This index is widely accepted and extensively used in Spain. It has been adapted from the British Biological Monitoring Working Party BMWP to the Iberian macroinvertebrates [30]. It is also easily applicable: the taxa of machobenthonic fauna found in the stream is identified to family level, a predefined score is allocated for each family and the total IBMWP score for a sample is a summation of the scores of all the families found. The scores of the IBMWP (0 – >100) have been grouped in 5 quality classes [30]. Other index as NFAM (number of families) and IASPT (Iberian Average Score per Taxon) [26,32] were also obtained. The Equivalent Quality Ratio (EQR) was obtained [33,34]. IBMWP was scored according to [30] and was compared with the reference conditions in Cantabric-Atlantic siliceous rivers [35]. The WFD defines reference conditions as the values of biological, hydromorphological and physical–chemical quality elements at high ecological status.

Between the hydromorphological characteristics, high status is found weather the hydrological regime, river continuity, morphological conditions, structure and condition of the riparian zones correspond totally or nearly totally to undisturbed conditions. Nevertheless, when dealing with SHP stations, alterations occur. In this case good status is achieved if the hydro morphologic conditions are consistent with the achievement of the values specified above for the biological quality.

Although measures of the conservation status of riparian habitat are not often used to describe river health [36], the SHP building can affect the characteristics of the vegetation in the banks of the river. The Riparian Forest Quality or QBR index [36] has been used to assess the ecological quality of the riparian habitat in the reaches affected by the SHP plant construction. The index QBR varies between 0 and 100 and it is based on four components of riparian habitat: total riparian vegetation cover, cover structure, cover quality and channel alterations for the different geomorphology of the river from its headwaters to the lower reaches. After completing the analysis, the sum of scores for the four components gives the final QBR index.

In order to measure the physicochemical parameters of the water, 4 sampling points have been selected in places where the entry does not entail risk and trying to select areas accessible in the future in spite of changes in the flow, building works or vegetation changes (Figure 2, Table 1). The position of the sampling points was registered with a GPS GPSMAP 60CSx (Garmin, Olathe, KS, USA) and pictures were taken to describe them. They were situated in a zone with no excessive sunshine, excessively high or low stream velocity and medium depth.

On each sampling point, the handheld multiparameter instrument was immersed in the watercourse and, after a stabilization of 20 min, the instrument was keep registering during 20 min more in order to obtain representative data. Previous laboratory calibration was performed for all sensors following the procedures described by YSI Environmental [37].

For the biological characteristics of the water, a 50 m radius-river stretches around the described points 2–4 were sampled (Table 1, Figure 2). The position of the borders of these stretches was registered with a GPS and photos of the sampled stretches were taken. The protocol [38] was followed: in the sampling stretches selected, different types of habitats were identified according to depth, flow velocity, substratum, and vegetation; each habitat was sampled; captures were identified over a white tray on the field using an atlas and key [39] and putting back the identified individuals into the river. Benthos which could not be identified in the field were kept in bottles with alcohol 70% to be identified in the laboratory.

The QBR index was calculated following the protocol [36] with some modifications. A river stretch of 400 m around the dam, and another river stretch of 650 m around the SHP station and tail race were selected (Figure 2). These reaches were divided in sections with similar characteristics and we calculated QBR index for each one. Points between two sections with different QBR were located with GPS. The creators of the index [36] established that both river banks should be considered together. Nevertheless, since some SHP plant elements only affect one bank, we calculate one QBR index for each bank.

During the 36 months period between December 2007 and November 2010, all the points were sampled every 3 months for the physicochemical and biological water quality: in March, June, September/October and November/December. QBR index was calculated twice a year: once in March and once in September. For the physicochemical parameters we used data registered by the SHP station owners previously with the same methodology on points 1–4: in March and September 2005, March, August, September and November 2006, and September 2007.

The software Ecowatch was used to download and process physicochemical data taken by the multiparameter instrument. When data was registered for more than 20 min, the series was trimmed to 200 samples or 20 min. We calculated descriptive for each 20 min sampling series: mean, standard deviation, maximum and minimum.

In the laboratory, we identified the captured macroinvertebrates taxa to family level [38] using a glass Motic ST-37 20–80× (Motic, Xiamen, China).

The values of the mean physicochemical variables of the series registered in each sampling point, as well as the IBMWP index were plotted over time to assess if their values stabilized themselves across time, as well as to observe differences between points. Differences between years and points were analyzed using Kruskal–Wallis tests. pH, temperature and DO were compared with extreme values allowed in salmonid waters (Directive 2006/44/EC [24]). Values of pH, temperature, DO and conductivity were compared with those registered during the same time of the year in the closest river Lérez Station 14RW03070 [40] from the Water Quality Control Network belonging to Augas de Galicia, which is situated 2.4 km downstream the Power Station. IBMWP scores were classified according to [30]. Physicochemical variables and biological index were compared with the Reference Condition in Cantabric-Atlantic siliceous rivers from IPH 2008 and [35].

The physicochemical parameters pH, temperature, dissolved oxygen, and conductivity did not show statistically significant differences between sampling points (Table 2). Previous studies neither observed significant influence of the presence of SHP plants on the physical and chemical quality of the water [13,41].

Lower scores for the IBMWP index were observed in point 3, upstream the power station (mean IBMWP = 110; SD = 26) compared with point 2 (mean IBMWP = 115; SD = 15) and point 4 (mean IBMWP = 114; SD = 30), although differences between points were not statistically significant (Kruskal–Wallis test = 1.756; df = 2; p = 0.416; n = 37).

Some studies did not observed severe adverse consequences on benthic macroinvertebrate community due to the SHP schemes [45,46], and even some observed a slight increase in the benthic fauna [46]. Others observed a lower density and specific richness of invertebrates in the points situated downstream the SHP plant compared to the points situated upstream [14]. Mean index during the study period were scored inside biological quality class I: unpolluted or not considerably altered water [30]. In a river stretch around a small hydroelectric power plan in the Ardena river (Portugal), the benthic macroinvertebrate communities have presented high richness, diversity and equitability [15]), although some decrease of the water quality and an alteration of the structure and composition of the benthic macroinvertebrate community due to the operation of a SHP plant was observed by these authors.

IBMWP index showed a positive trend from 2 years after the construction of the SHP plant and its values stabilized themselves on biological quality I. In December 2007, one year after the SHP station construction, only points 2 and 4 were sampled for biological status of the water. IBMWP in point 2 showed a score of 75 (quality II: some effects of pollution are evident) and in point 4 a score of 120 (quality I: unpolluted or not considerably altered water) (Figure 7). In March 2008, 16 months after the construction, the IBMWP scores observed in point 2 (downstream the dam) showed a water-biological quality II (moderate effect of pollution are evident), and scores observed in points 3 and 4 showed quality IV (very polluted water). Since this moment, when the worse biological quality of the waters were observed, biological conditions were improving until to the point, in December 2008, when the IBMWP scores on the three sampling points level off in status I (very clean waters and unpolluted or not considerably altered water). Nevertheless, our results suggest, that river stretches around SHP plants operational conserve the conditions for the development of well-structured biotic communities as well as observed in other works [13]. In this sense it was previously observed that in the aestival period, with the suspension of the operations of the small hydroelectric plant, there was an almost total recovery of the communities [13].

For the rivers classified as type 21, reference conditions related with IBMWP has not been defined in the IPH 2008. So, in order to classify the biological quality of the water, the ecological index were compared with the reference condition given by [35] (Table 4). The IBMWP EQR was calculated and compared with the IPH 2008 limits of the river type which was considered more similar and for which IPH 2008 provide reference conditions for (rivers type 25: IPH 2008). The river Lérez stretch studied showed an unfavourable ecological status according with [35], or a moderate ecological quality according with IPH 2008, except for the year 2008 when it was deficient (EQR < 0.42, Table 4). The NFAM show a unfavourable status every year [35]. Nevertheless, IASPT showed higher values than the reference condition and EQR >1. These results suggest that the diversity of macrobenthos in this stretch of the river around the San Xusto SHP plant is lower than the reference condition, but the present families have high ecological value. This effects had been described previously [45]) in Wales, where these authors also obtained a low score of the index BMWP but the taxa present indicated good quality, thus giving a reasonable ASPT.

Biological status relies also on the composition, abundance and age structure of fish fauna. Moderate status occurs when the composition and abundance of fish species differ moderately from the type-specific communities attributable to anthropogenic impacts on physico-chemical or hydromorphological quality elements (Directive 2000/60/EC, Annex V. 1.2.1 [17]). The San Xusto SHP plant incorporates a fish ladder, managed by the regional government (Xunta de Galicia). This ladder, if used, would reduce the negative impact of the dam on the fish population [11], but we registered no use of this ladder and consequently a probable effect of the dam on the fish population. Diversion dams of SHP plant hinder the passage of eels [47,48]), and also alter the populations of other fish species [41,46,49]. Some authors [41,46] found differences in population size–structure among sites upstream and downstream from SHPs with suitable fish passes for Barbus bocagei and Squalius carolitertii, as well as for B.bocagei, S. carolitertii and Salmo trutta in SHPs with unsuitable fish passes. Specifically there have been observed a change in the structure of fish assemblages when comparing before and after SHP plant started to operate [49]. Elements of SHP plants, specially dams, are not always complete barriers for fish [50] even if they are fitted with no effective fish passes [41]. Mitigation measures, such as environmental flows and fish passage facilities made it possible to limit to some extent the negative impacts of the SHP plants, although every obstruction in the river, even fitted with effective fish passage facilities creates a delay in migration [12].

The evaluation of the riparian habitat showed two river bank sections with extreme degradation (QBR ≤ 25, Figure 8): (1) a right bank stretch of 200 m situated around the dam, the diversion pipe and fish ladder; (2) a right bank stretch of 50 m where the power house is situated. Around the power house we observed a river stretch of 110 m, on both river banks, with alterations from extreme degradation, on the power house, to important disturbance. In the remaining sections the riparian habitat was classified as good quality (QBR = 75–90), or even natural condition (QBR ≥ 95). SHP plant effect on riparian habitat was only observed in the sections affected by the constructions, totalling a stretch of 300 m.