A comparative analysis revealing incongruence in the impacts of small and large hydroelectric power plants on fish communities

Luiz Guilherme dos Santos Ribas¹, Anderson Luís Maciel¹, Gilmar Baumgartner¹, Fernando Cesar Alves Ferreira², Juliano Tupan Coragem³ and Éder André Gubiani¹

PDF: EN    XML: EN | Supplementary: S1 S2 S3 S4 S5 | Cite this article

Associate Editor: Paulo Pompeu

Section Editor: Fernando Pelicice

Editor-in-chief: Carla Pavanelli

Abstract​

---

Introduction​

---

Habitat loss and alteration are the main drivers of the biodiversity crisis (Jaureguiberry et al., 2022). This is particularly evident in freshwater ecosystems, where hydroelectric power plants rank among the primary contributors to such changes (Johnson et al., 2008; Zarfl et al., 2015; Winemiller et al., 2016; Thieme et al., 2021). The environmental impacts of dams and reservoirs are wide-ranging and profound, especially given the global expansion of hydropower, a trend most pronounced in developing countries (Zarfl et al., 2015; Winemiller et al., 2016).

Hydroelectric power plants are commonly classified as either large (LHPs) or small (SHPs) based on size and energy production capacity, though this distinction remains somewhat arbitrary (Couto, Olden, 2018). Classification criteria typically include energy output potential, reservoir size, dam height, and the stream order of impounded rivers (Kelly-Richards et al., 2017; Couto, Olden, 2018). In addition, there is no global consensus on what qualifies a hydroelectric power plant as SHP, with ambiguity and inconsistencies across different countries. Specifically, we follow here the Brazilian classification by the National Electric Energy Agency (ANEEL for its Portuguese acronym, ANEEL normative resolution number 1070 from 29/08/2023; ANEEL, 2023) and Brazilian federal legislation (CONAMA 279/2001; Federal Law number 13.360/2016; Brasil, 2016), which define SHPs as hydroelectric power plants with an energy production capacity of less than 30 MW and a reservoir area of up to 13 km². This classification aligns with criteria used in other countries (Kelly-Richards et al., 2017; Couto, Olden, 2018).

The classification of hydroelectric power plant sizes is supported by a new perspective on hydropower development, in which SHPs are considered to entail low opportunity costs and minimal potential for negative environmental impacts (Santos et al., 2006; Couto, Olden, 2018). This assumption has led to multiple legislative measures facilitating protocols that exclude SHPs from decision-making processes in many countries (Winemiller et al., 2016). Even when environmental impact studies are required for SHP construction, they are typically simpler and less committed to monitoring, evaluating, and mitigating the impacts of these projects on ecosystems (Winemiller et al., 2016). As a result, there has been a global increase in the number of SHPs compared to LHPs (Couto, Olden, 2018).

However, limited evidence supports the assumption that SHPs are less threatening to biodiversity than LHPs (Kibler, Tullos, 2013; Couto, Olden, 2018; Lange et al., 2019). The evidence supporting this claim remains scarce, particularly given the paucity of studies evaluating the impacts of SHPs (e.g., Santos et al., 2006; Couto et al., 2023) compared to the relatively extensive body of evidence on the impacts of LHPs (e.g., Agostinho et al., 2016; Turgeon et al., 2016; Botelho et al., 2017; Keppeler et al., 2022). Moreover, most studies suggest that SHPs and LHPs have similar effects on biodiversity (e.g., Hirschmann et al., 2008; Carvalho, Araújo, 2024). Therefore, we still need to better understand how different types of hydroelectric power plants affect the environment to improve decision-making. This understanding is also crucial for supporting or challenging the arbitrariness of classifying a hydroelectric power plant as small or large. Along with related policies and regulations, particularly in light of the rapid expansion of hydropower based on SHPs facilities (Kibler, Tullos, 2013; Jumani et al., 2017; Kelly-Richards et al., 2017; Couto, Olden, 2018; Couto et al., 2023).

Fish are among the most studied biological groups impacted by hydroelectric power plants (Finer, Jenkins, 2012; Pelicice et al., 2015; Agostinho et al., 2016; Turgeon et al., 2016; Winemiller et al., 2016). The construction of dams and formation of reservoirs alters fish communities through multiple pathways, primarily by changing hydrologic conditions, organic matter, energy flow, and nutrient availability compared to natural environments (Poff et al., 1997; Poff, Zimmerman, 2010; Dias et al., 2020). These changes are particularly well understood for LHPs, especially during and immediately after reservoir filling (Agostinho et al., 1999). In this context, generalist, omnivorous, and anthropogenically-adapted specie, such as invasive and small-sized sedentary species, tend to benefit, while migratory, piscivorous, and large-sized species are often negatively affected (Agostinho et al., 1999, 2003, 2008; Johnson et al., 2008; Hoeinghaus et al., 2009; Delariva et al., 2013; Pereira et al., 2016; Dias et al., 2020). However, the overall impact on fish communities shows high variability (Turgeon et al., 2016), especially in megadiverse ecosystems such as tropical ones, and depends on each hydroelectric power plant’s specific characteristics (Agostinho et al., 1999; Poff, Hart, 2002; Abell et al., 2008; Couto et al., 2023).

Considering known patterns of reservoirs on fish assemblages, the effects of hydroelectric power plants in the years immediately before and after reservoir filling result in a phenomenon known as the “trophic upsurge period” (Kimmel, Groeger, 1986; Kimmel et al., 1990). This phenomenon has been well documented in multiple LHPs operating under different conditions and constructed in various environments (Petrere, 1996; Agostinho et al., 2007; Gubiani et al., 2011; Turgeon et al., 2016). In summary, the trophic upsurge period is characterized by increased nutrient input into the system, driven by physical and chemical changes as the environment shifts from a lotic (i.e., river) to a lentic (i.e., reservoir) state. Consequently, this phenomenon alters fish communities by increasing overall abundance and species richness while promoting the dominance of certain species, which typically leads to reduced diversity indices that account for relative abundance (i.e., low evenness; Agostinho et al., 2008, 2016).

In this study, we used standard environmental impact assessments (EIA) data from multiple SHPs and LHPs with similar ecological characteristics but different biological communities (Paraná, Dagosta et al., 2024; Piquiri, Cavalli et al., 2018; Tibagi, Raio, Bennemann, 2010; and Iguaçu, Baumgartner et al., 2012; Mezzaroba et al., 2021; River basins) to compare their impacts on fish assemblages across the two size categories. To accomplish this, we first applied impact evaluation methods to evaluate how these facilities affect three key metrics expected to change due to the trophic upsurge period: (i) fish abundance, (ii) species richness, and (iii) diversity. We specifically focused on the trophic upsurge period because the monitoring data used in this study corresponded to a short time frame immediately before and after dam construction and reservoir filling, precisely when the trophic upsurge is expected to occur. This temporal window allowed us to test hypotheses related to the short-term ecological responses typically observed during the early post-impoundment phase. We then compared impact estimates on the early described metrics between SHPs and LHPs, considering the two hypotheses outlined next. First, we hypothesized that the trophic upsurge period would increase fish abundance and species richness while decreasing diversity due to reduced evenness, with more pronounced impact estimates in LHPs compared to SHPs. This prediction stems from LHPs typically causing greater environmental alterations, which may consequently lead to more severe biodiversity impacts. Second, we hypothesized that both classes of hydroelectric power plants would have similar directional impacts across the three metrics. Specifically, in line with the hypothesis, we expected that both SHPs and LHPs to cause positive impacts on fish abundance and species richness, and negative impacts on diversity. Thus, while we anticipated stronger impacts from LHPs, we expected the trophic upsurge period to produce comparable trends in both systems, regardless of size, leading to similar trends in fish abundance, richness, and diversity.

Additionally, we applied two impact evaluation methods: one commonly used in environmental sciences, the Before-After-Control-Impact (BACI), and another considered more accurate according to impact evaluation theory (Cashmore et al., 2010; Morgan, 2012) for estimating impacts with EIA designs – the Regression Discontinuity in Time (RDiT). Through this comparative approach, we examine how EIA monitoring programs could better assess hydroelectric power plant impacts and identify necessary improvements for impact estimation using these methods. This discussion aims to strengthen the utility of EIA monitoring data for generating evidence relevant to decision-making during and after licensing procedures for both SHPs and LHPs, despite legislation in many countries increasingly weakening or eliminating these requirements (Erlewein, 2013; Couto, Olden, 2018; Lange et al., 2019; Couto et al., 2023).

Material and methods

---

Small and large hydroelectric power plants. We evaluated six hydroelectric power plants in Paraná State, Brazil. These included three SHPs formally named Bela Vista, Cantu 2, and São Francisco, and three LHPs formally named Tibagi Montante, Baixo Iguaçu, and Salto Caxias. The facilities are located in the Paraná, Piquiri, Tibagi, and Iguaçu River basins (Fig. 1; Tab. S1), which share similar fish communities while each maintaining some distinct ecological characteristics.

FIGURE 1| Location and general characteristics of small (SHP) and large (LHP) hydroelectric power plants in Paraná State, Brazil.

For our analysis, we considered these hydroelectric power plants comparable within their respective size classifications, as we aimed to identify general trends in overall fish community attributes that we expect to respond similarly across rivers within these basins. It is worth noting that the Paraná and Iguaçu River basins are among the most heavily modified by hydropower construction, with the number of SHPs increasing rapidly in recent decades (Winemiller et al., 2016; Santos et al., 2018; Pini et al., 2021).

The hydroelectric power plants evaluated were classified as either SHPs and LHPs. SHPs are characterized by a generation capacity below 30 MW and a reservoir area of up to 13 km², whereas LHPs surpass these thresholds (CONAMA, 2001; ANEEL, 2023). While primarily distinguished by size, these facilities also vary in other aspects including water residence time, reservoir area, operational systems, and other relevant characteristics (see Tab. 1).

TABLE 1 | Characteristics of hydroelectric power plants.

	Size class	Year of reservoir formation	Municipality	Basin	River	Generation capacity (MW)	Operation system	Reservoir area (km²)	Water residence time (days)
Bela Vista	SHP	2021	Verê and São João	Iguaçu	Chopim	29.8	Run-of-river	2.10	3.6
Cantu 2	SHP	2015	Nova Cantu	Piquiri	Cantu	19.8	Run-of-river	3.55	9.4
São Francisco	SHP	2010	Toledo and Ouro Verde do Oeste	Paraná	São Francisco Verdadeiro	14.0	Run-of-river	0.67	3.6
Tibagi Montante	LHP	2019	Tibagi	Tibagi	Tibagi	36.0	Run-of-river	6.83	3.6
Baixo Iguaçu	LHP	2019	Capanema and Capitão Leônidas Marques	Iguaçu	Iguaçu	350.2	Run-of-river	31.00	1.7
Salto Caxias	LHP	1998	Capitão Leônidas Marques and Nova Prata do Iguaçu	Iguaçu	Iguaçu	1240.0	Storage	141.00	60.0

Fish data and metrics. Fish data were collected using standardized sampling protocols across all hydroelectric power plants (see Tab. S2 for details). The data were gathered during EIA monitoring programs, which comply with federal, state, and municipal regulations, including those established by the Brazilian Institute of Environment and Renewable Natural Resources (IBAMA), the ANEEL, the Water and Land Institute (IAT) of Paraná State, and municipal environmental agencies. These EIA monitoring programs are mandatory for both the construction and operation of hydroelectric power plants in Paraná State, Brazil.

Fish were sampled over several months across consecutive years at each hydroelectric power plant (Tab. S3). A consistent approach was applied at three sampling sites: upstream, reservoir, and downstream. Upstream sites were located at the beginning of the reservoir’s up section, following the natural course of the dammed river. Reservoir sites were positioned near the dam, while downstream sites were situated below the dam, where the water flow returned to a lotic behavior. For more details, see Tab. S1.

Using the fish data, we calculated three metrics to evaluate the expected effects of the trophic upsurge period: (i) overall abundance, (ii) species richness, and (iii) a diversity metric accounting for relative species abundance. Overall abundance was measured as the total catch per unit of effort (CPUE; individuals per 1,000 m² of gillnets over a 16-hour period, Tab. S2), as proposed by King (2007). CPUE standardized abundance comparisons across hydroelectric power plants given differences in sampling methods (Tab. S2). Species richness was determined as the count of unique species per sample. Diversity was calculated using the Inverse Simpson Index (Simpson, 1949; Magurran, 2004), which highlights community dominance and evenness. The index ranges from one, indicating low-diversity communities dominated by a single species, to the total number of species in a sample, where all species are equally abundant, representing a highly diverse community. All metrics were calculated using the vegan package (Oksanen et al., 2025) in R environment (R Development Core Team, 2025).

Although we did not include direct measurements of environmental energy (e.g., primary productivity), the timing of the monitoring data, collected immediately before and after dam filling, coincides with the period when the trophic upsurge is expected to occur (Thornton et al., 1990; Agostinho et al., 2007). We therefore interpret changes in fish assemblages within this temporal window as indirect indicators of early post-impoundment dynamics.

Impact evaluation. We utilized the metrics calculated in the previous section to evaluate reservoir filling impacts by comparing pre- and post-filling periods. To assess how SHPs and LHPs affect fish communities, we employed two impact evaluation designs: BACI (Before-After-Control-Impact) and RDiT (Regression Discontinuity in Time).

The BACI method, adapted from Underwood (1992, 1993), is widely employed in environmental sciences to evaluate impacts, including assessments of SHPs and LHPs impacts on biodiversity (e.g., Dias et al., 2020; Santos et al., 2022; Scotti et al., 2022). This approach proves valuable for analyzing EIA monitoring data. Within the BACI framework, intervention impacts are assessed through a two-factor analysis examining both time, by comparing conditions before and after the intervention, and space, by contrasting impacted areas with control areas.

In the BACI design, we considered the upstream site as control for the other two sampling sites. We evaluated impacts on both reservoir and downstream sites by comparing their pre- and post-intervention periods relative to the upstream reference site. The impacts estimated using the BACI design were calculated with a linear mixed-effects model, specified by the following the equation:

Metric_t = β + Period_t + Site_t + (Period_t × Site_t) + u_t + ε_t,

Where βrepresents the intercept term, considered as the baseline represented by measurements taken before the reservoir filling at the upstream sampling site for each single evaluated metric. Period_t denotes the effect of time on the evaluated metric, distinguishing between pre- and post-filling periods. The term Site_t refers to the spatial effect of sampling sites, other than upstream, on the evaluated metric. The interaction term Period_t × Site_t represents the impact of the hydroelectric power plant on the evaluated metric at both the reservoir and downstream sampling sites compared to the upstream site, accounting for differences between pre- and post-filling conditions. Finally, u_t represents random effects associated with sampling periods, while ε_t is the residual term.

The RDiT design employed a different approach by evaluating each site individually, using the pre-filling period as a control for the post-filling period. This method allowed site-specific impact evaluations by leveraging the temporal discontinuity created by reservoir filling. While less common in environmental science, this approach is well-established in other disciplines (e.g., econometrics and social sciences) for accurate impact estimation (Hausman, Rapson, 2018; Cattaneo, Titiunik, 2022). The impacts estimated using the RDiT design were calculated with a generalized least squares model incorporating AR(1) autocorrelation, as specified by the equation:

Metric_t = β + Intervention_t + Sampling event_t + Slope_t + ε_t,

Where βdenotes the intercept term established during the initial sampling period. Intervention_t represents the impact of each SHP and LHP on the evaluated metric, accounting for the reservoir filling discontinuity period. The Sampling event_t refers to the linear trend in each sample relative to the operational commencement year of each SHP and LHP. Slope_t captures post-operational changes in the linear model’s slope for each facility. Finally, ε_t represents the residual term, which accounts for an autocorrelation structure over time, as depicted in the following equation:

ε_t = ρε_t−1 + η_t, η_t ∼ N(0,σ²)

The BACI design was conducted using the lmer function from the lme4 package (Bates et al., 2015), while the RDiT design was conducted using the gls function from the nlme package (Pinheiro, Bates, 2025). Both analyses were performed in R environment (R Development Core Team, 2025). Results are presented separately for each impact evaluation design.

Hypothesis tests. Next, we used the results of the impact evaluation methods (i.e., BACI and RDiT) as response variables to investigate how impact varies between the two classes of hydroelectric power plants (i.e., SHP and LHP) considering the expected effects of the trophic upsurge period. Accordingly, we tested the two hypotheses presented in the introduction. The first hypothesis proposed that the impact would be greater in LHPs compared to SHPs due to the broader and more significant effects of the trophic upsurge period on fish in LHPs. The second hypothesis suggested that, for both classes of hydroelectric power plants, the direction of their impacts on fish would be similar.

We employed linear mixed-effects models to test the first hypothesis. In these models, impact estimates were compared across hydroelectric power plant classes (SHP or LHP), incorporating a random effect to account for variation among individual hydroelectric power plants facilities. This approach enabled evaluation of each class’s effects relative to a zero-impact baseline, as well as comparing between the two size categories to assess our hypothesis.

We used binary logistic regression to test the second hypothesis. First, we categorized impacts for each metric as either positive or negative. We then applied the model to estimate the probability of obtaining positive versus negative impact estimates for each hydroelectric power plant size class across the metrics. As a result, the model estimated the odds ratio of finding a positive versus a negative impact when comparing SHPs and LHPs.

For the first hypothesis, we utilized the lme function from the nlme package (Pinheiro, Bates, 2025). For the second hypothesis, we employed the glm function from the stats package. Both packages are available in R environment (R Development Core Team, 2025). We set the p-level at 0.05 to determine the statistical significance of our findings.

To assess whether the observed species richness was robust to variation in sampling effort, we conducted additional analyses using presence–absence matrices derived from the CPUE data. First, we computed non-parametric richness estimators (Chao, Jackknife, and Bootstrap) using the specpool function from the vegan package (Oksanen et al., 2025). These estimators provided a benchmark for comparing observed richness against expected values under potential under-sampling scenarios (Tab. S4). Next, we constructed species accumulation curves for each hydroelectric facility to evaluate whether species richness approached asymptotic levels, indicating sampling completeness (Fig. S5). Finally, we examined the relationship between sampling effort and observed richness at each site (Fig. S5). Total CPUE was used as a proxy for effort, and its association with richness was tested using Spearman’s rank correlation and linear regression.

Results​

---

BACI design. The impact evaluation across all three metrics revealed no significant differences between SHPs and LHPs (Tab. 2; Fig. 2). However, when comparing each class against a zero-impact baseline, we found that SHPs exhibited increased species richness at reservoir sites (Tab. 2; Fig. 2C) and showed a marginal effect (with a p-value near 0.05) on fish abundance at downstream sites (Tab. 2; Fig. 2B), with LHPs generally exhibiting higher values.

TABLE 2 | Impact estimates for small (SHP) and large hydroelectric power plants (LHP) on fish abundance, species richness and Inverse Simpson index using Before-After-Control-Impact (BACI) and Regression Discontinuity in Time (RDiT). Asterisk indicate significant impact estimates, UP representing upstream sampling site, RES representing reservoir sampling site, and DOWN representing downstream sampling site.

			BACI			RDiT
			Abundance	Richness	Inverse Simpson	Abundance	Richness	Inverse Simpson
SHP	Bela vista	UP	control	control	control	-39.53	-1.29	0.98
		RES	151.44	7.11*	2.68	357.94	8.25	2.20
		DOWN	50.04	-0.78	-0.74	28.29	-3.59	-4.03
	Cantu 2	UP	control	control	control	-5.08	-1.68	-1.08
		RES	-56.32	3.42	-1.98	105.27*	4.86	-1.19
		DOWN	-158.67	-4.46	-4.64*	-12.36	2.48	1.91
	São Francisco	UP	control	control	control	-60.32	2.31	3.10*
		RES	-152.74	3.26*	-1.31	-50.15	4.33	3.05
		DOWN	228.35	3.97*	1.35	341.56	6.08*	2.69*
LHP	Tibagi Montante	UP	control	control	control	489.61	8.94	5.71
		RES	-182.95	-2.21	-2.26	264.21	2.03	-0.22
		DOWN	155.33	1.97	1.19	958.09*	8.06	2.62
	Baixo Iguaçu	UP	control	control	control	120.19	0.56	-0.36
		RES	238.04*	3.00*	2.18*	-65.99	1.58	2.08
		DOWN	472.70*	4.09*	2.84*	187.18	3.49	3.17
	Salto Caxias	UP	control	control	control	400.76	-0.85	-5.08*
		RES	1507.39*	1.29	0.88	521.14	3.16	4.80
		DOWN	1651.13*	0.76	-0.82	2773.95	0.16	-1.98
	*p < 0.05

FIGURE 2| Impact estimates for small (SHP) and large hydroelectric power plants (LHP) on fish abundance (A, B), species richness (C, D) and Inverse Simpson index (E, F) using Before-After-Control-Impact (BACI) at reservoir (A, C, E) and downstream (B, D, F) sampling sites. Bellow each graph are the results of linear mixed-effects models testing difference of SHP and LHP against a zero-impact intercept and between the two size classes, these encompass model estimate, z-value in parenthesis, and p-value. Dashed lines indicating zero-impact baseline.

Considering the direction of the impact estimates between SHPs and LHPs, the odds of a positive impact on fish abundance were approximately five times higher for LHPs than for SHPs (p-value = 0.239). The odds of a positive impact on species richness were approximately 2.5 times higher for LHPs compared to SHPs (p-value = 0.512). The odds of a positive impact on the Inverse Simpson Index were approximately four times higher for LHPs compared to SHPs (p-value = 0.258).

RDiT design. The impact on fish abundance was higher in upstream sites for LHPs compared to SHPs (Tab. 2; Fig. 3A). Additionally, we found that SHPs had a greater impact on species richness in reservoir sites than LHPs (Tab. 2; Fig. 3E). When comparing impacts of each class to a zero-impact baseline, our results indicated that SHPs resulted in higher species richness in reservoir sites (Tab. 2; Fig. 3E). Simultaneously, LHPs also exhibited high species richness in the reservoir sites (Tab. 2; Fig. 3E). Furthermore, we noted high fish abundance in both upstream and downstream sites for LHPs (Tab. 2; Figs. 2A, C).

FIGURE 3| Impact estimates for small (SHP) and large hydroelectric power plants (LHP) on fish abundance (A, B, C), species richness (D, E, F) and Inverse Simpson index (G, H, I) using Regression Discontinuity in Time (RDiT) at upstream (A, D, G), reservoir (B, E, H), and downstream (C, F, I) sampling sites. Bellow each graph are the results of linear mixed-effects models testing difference of SHP and LHP against a zero-impact intercept and between the two classes, these encompass model estimate, z-value in parenthesis, and p-value. Dashed lines indicating zero-impact baseline.

Considering the impact estimates for SHPs and LHPs, the odds of finding a positive impact on fish abundance were approximately ten times higher for LHPs compared to SHPs (p-value = 0.066). Similarly, the odds of a positive impact on species richness were approximately four times greater for LHPs than for SHPs (p-value = 0.277). Additionally, the odds of a negative impact on the Inverse Simpson Index were approximately 0.6 times lower for LHPs than for SHPs (p-value = 0.63).

Discussion​

---

Our findings demonstrate that the impacts of hydroelectric power plants on fish assemblages were predominantly site-specific, with no consistent directional patterns clearly distinguishing SHPs from LHPs facilities. This conclusion is supported by the substantial variability observed for all three ecological metrics, fish abundance, species richness, and diversity, across different river sections. Contrary to our initial hypothesis that LHPs would exert stronger impacts, several SHPs exhibited higher impact estimates. Considering second hypothesis, LHPs generally exhibited more positive impact estimates compared to SHPs, contrasting with our expectation that both size classes would exhibit similar directional effects on fish communities. These patterns remain consistent across both impact evaluation methods. Overall, the evidence only partially supports our hypotheses, with particularly notable divergences observed for the species richness component in the reservoir sites.

Regarding expectations associated with the trophic upsurge period, we found evidence that LHPs increased fish abundance after reservoir filling, particularly when estimated using the RDiT method. However, impact estimates showed high variability, consistent with previous findings from other studies of trophic upsurge period effects in hydroelectric power plants (e.g., Petrere, 1996; Agostinho et al., 2007; Gubiani et al., 2011; Turgeon et al., 2016). This variability underscores how local and regional factors, such as reservoir morphometry, discharge patterns, habitat structure, and species composition, influence post-impoundment fish community dynamics, as previously documented in the literature (Agostinho et al., 2016). Nonetheless, we acknowledge that our study relies on indirect evidence of the trophic upsurge period based solely on fish assemblage data, without direct measurements of environmental energy or nutrient availability. Future studies incorporating primary productivity, water quality, and other ecological covariates will be essential for explicitly testing the mechanisms and dynamics underlying the trophic upsurge phenomenon, especially when comparing the impacts of SHPs and LHPs.

Despite the absence of consistent overall trends for both hydroelectric power plants classes, we observed some expected patterns associated with the trophic upsurge period in both SHPs and LHPs. While most documented trends reported in the literature focus on the impacts of LHPs on biodiversity, particularly reductions in species richness and diversity (e.g., Pusey et al., 1995; Gehrke et al., 2002), our results indicate this phenomenon appears less pronounced or absent in SHPs. Instead, SHP impacts showed greater variability, with some effects contradicting theoretical expectations. This aligns with studies suggesting SHPs and LHPs differentially alter ecosystem processes, yielding distinct biodiversity outcomes (e.g., Stanley et al., 2002; Wu et al., 2010).

Notably, inconsistent trends persisted even within individual facilities. For instance, species richness impacts varied substantially across sampled regions for most facilities, contradicting researches suggesting consistent within-site patterns (e.g., Kubecka et al., 1997; Jumani et al., 2017; Couto et al., 2023). Unlike previous studies, we detected no significant shifts in fish assemblages at downstream sites, at least for the basins that we evaluated, contrasting with reported dewatering impacts like reduced fish size and migratory species declines, particularly for SHPs (Kubecka et al., 1997; Jumani et al., 2017). These findings may reflect sampling characteristics and the lack of migratory species, particularly for hydroelectric power plants present in the Iguaçu basin.

The composition of fish assemblages in different basins may also influence the detectability and magnitude of dam impacts. For example, the Iguaçu River basin is known to host predominantly small-bodied, non-migratory fish species, with few long-distance migrators (Baumgartner et al., 2012; Mezzaroba et al., 2021). This composition may buffer the fish community against some of the more disruptive downstream effects typically associated with hydropower development, such as the fragmentation of migratory routes. Although our data are not sufficient to statistically test basin-level differences, we acknowledge this as a potentially important factor influencing our results. Future studies with expanded spatial replication and basin-level sampling will be critical to clarify how regional faunal traits modulate the ecological impacts of dams.

The diversity in size and operational modes of hydroelectric power plants leads to a wide range of unpredictable ecological consequences. Given the arbitrary nature of size classifications, LHPs generally demonstrate broader impacts than SHPs, as they include facilities with varying characteristics that affect biodiversity responses. For example, among the evaluated LHPs, Tibagi Montante resembles most SHPs in size, more than other LHPs, yet it sometimes showed greater biodiversity impacts than larger LHPs. This suggests that hydropower impacts extend beyond simple size-based expectations. These findings carry significant implications for hydropower policy and regulation, particularly in distinguishing “small” and “large” facilities during environmental licensing (Kelly-Richards et al., 2017; Couto, Olden, 2018). Current classification schemes primarily rely on energy generation capacity, which poorly correlates with ecological consequences (Kibler, Tullos, 2013; Couto, Olden, 2018). Notably, Tibagi Montante’s generation capacity is also more similar to SHPs (see Tab. 1), further emphasizing the need for rigorous, site-specific impact assessments regardless of size-based classifications.

Although differences between SHPs and LHPs were statistically unclear, we highlight that most high-impact estimates, whether positive or negative, were associated with LHPs. This suggests that large facilities generally induce greater changes in fish communities. However, this does not mean that SHPs do not cause impacts. On the contrary, SHPs appear to produce more variable impacts, likely due to differences in operational procedures, catchment morphometry, native ichthyofauna, and habitat modifications (Agostinho et al., 2004, 2008). As SHPs continue to expand globally, outnumbering LHPs 11 to 1 (Couto, Olden, 2018), their impacts require more rigorous study rather than being presumed minimal. In some cases, as stated in our results, SHPs can sometimes generate greater impacts than LHPs. This variability underscores the need for regular biodiversity monitoring to identify trends and inform decision-making. Such an approach is crucial for reconciling hydropower development with biodiversity conservation, a challenge that is becoming increasingly critical.

Beyond impact estimates, both impact evaluation designs and EIA monitoring approaches merit closer examination. Our BACI evaluation detected no significant differences between SHPs and LHPs across the three metrics. However, comparisons against a zero-impact baseline suggest some trends. SHPs were associated to higher species richness at reservoir sites, indicating these smaller facilities may create localized conditions that promote species accumulation. Conversely, SHPs showed marginal effects on downstream fish abundance, indicating potential ecological relevance despite lacking statistical significance. Meanwhile, LHPs generally exhibited greater fish abundance, supporting the premise that larger reservoirs enhance habitat availability, though this effect varied substantially among sites.

Similarly, RDiT results indicate that LHPs generally exert greater influence on fish abundance through their more extensive hydrological alterations and habitat modifications. Interestingly, the increased species richness observed at reservoir sites for both SHPs and LHPs reinforces the concept that impoundments initially enhance biodiversity by creating more heterogeneous habitats and greater resource availability. However, this pattern typically reverses over time as environmental filters progressively eliminate fluvial-adapted species (Mol et al., 2007; Orsi, Britton, 2014; Agostinho et al., 2016). For example, older Neotropical reservoirs often exhibit stabilized species richness at reduced levels, with assemblages composed by species pre-adapted to lentic conditions with limited reliance on fluvial environments (Agostinho et al., 2016). Additionally, in some cases, the post-impoundment increase in species richness may be spurious, due to the introduction of non-native species, as reservoirs act as stepping-stones for invasions (Johnson et al., 2008; Daga et al., 2020). The persistently high fish abundance observed at both upstream and downstream LHPs sites suggests that larger reservoirs may mitigate certain flow regulation impacts, a phenomenon that warrants further investigation, especially in downstream sites.

Interpreting BACI results requires caution, particularly given common EIA data collection practices. Using upstream sites as controls is problematic, as it often fails to meet two critical assumptions: (i) that upstream sites maintain ecological characteristics similar to the rest of the river, and (ii) that upstream sites are unaffected by the dam being evaluated as well as by any upstream dams (Morrison et al., 2008; Smokorowski, Randall, 2017; statistic independence). The first assumption is inherently flawed, as ecological differences naturally exist along river gradients (Vannote et al., 1980; Ward, Stanford, 1995; Jackson et al., 2001). The second assumption is also unrealistic, as upstream areas may still experience impacts (see our results; Agostinho et al., 2007, 2008, 2016). When these conditions are unmet, impact evaluations risk biased and inaccurate estimates (Ferraro, 2009; Smokorowski, Randall, 2017). To improve assessment reliability, control sites should be sufficiently distant from the facilities, both upstream and downstream, to minimize influence while maintaining ecological similarity, as demonstrated in Agostinho et al. (2007, 2016). However, appropriate distances are highly context-dependent, and further research is needed to evaluate dam impacts across spatial and temporal scales within river systems in order to establish clear guidelines. Additionally, Brazilian EIAs face persistent logistical, political, and financial constraints that limit proper sampling coverage, making the BACI design particularly challenging for impact evaluation using EIA data.

In contrast, RDiT and similar approaches provide stronger evidence by avoiding these assumptions (Hausman, Rapson, 2018). In RDiT models, the same site is compared before and after intervention, preventing inappropriate comparisons between ecologically distinct areas. However, from an EIA perspective, RDiT requires extensive time-series data, posing similar logistical challenges. To ensure accuracy, increased temporal sampling is necessary, particularly before hydropower development. This would improve model precision and impact estimates (Gertler, 2016; Hausman, Rapson, 2018), especially for fish communities which are naturally influenced by climatic, environmental, and regional factors (Jackson et al., 2001). Therefore, effective EIA sampling should prioritize capturing seasonal variations, an aspect often overlooked in assessments limited to short pre-implementation periods. Considering this, it would be valuable to distribute sampling over time rather than concentrating numerous collections in a brief period.

More importantly, we highlight the critical need to revise EIA protocols used to assess impacts on ichthyofauna. We suggest that extending the temporal scope of pre-dam sampling is more effective than simply increasing the number of samples over a short period. This recommendation is based on the rationale that capturing the natural temporal variability in fish communities (Winemiller, 1990; Silveira et al., 2018; Dallas, Kramer, 2022), particularly for metrics such as abundance, richness, and diversity, requires baseline monitoring that span at least two to three years prior to impoundment and includes sampling that reflects the environmental seasonality at each site. This extended sampling window improves the reliability of impact detection, particularly for regression-based designs such as RDiT, which rely on temporal trends rather than spatial comparisons. Considering BACI designs, we recommend adopting integrated basin-scale assessments with ecologically similar, yet spatially independent, control sites. This approach reduces bias and enhances the ecological validity of impact estimates. These improvements also support a broader recommendation: the need to move beyond size-based dam classifications by incorporating factors such as hydrological alterations, ecological connectivity, cumulative impacts, and river basin characteristics.

We also advocate incorporating sampling sufficiency diagnostics into standard monitoring practices. Species accumulation curves, abundance accumulation plots, and diversity profiles (e.g., Hill numbers) offer valuable tools to assess whether the collected data adequately represent the local fish community. These diagnostics are especially relevant in EIA contexts, where decision-making must be grounded in robust, representative biodiversity baselines.

Altogether, these improvements can help refine Brazil’s EIA framework by aligning sampling protocols with ecological theory and practical limitations. Terms of reference should explicitly encourage site-specific and evidence-based monitoring strategies, tailored to river order, basin characteristics, expected community variability, and socio-political context. By balancing scientific rigor with stakeholder engagement and realistic constraints, it becomes possible to enhance the transparency, reliability, the potential for multiple water resource uses, and conservation effectiveness of hydropower impact assessments. Therefore, terms of reference better adapted to local conditions, along with more effectively conducted environmental impact studies, can help ensure more appropriate impact mitigation measures.

In conclusion, our findings highlight the complexity of evaluating hydroelectric power plants impacts on fish assemblages. The lack of consistent patterns across size classes and sites emphasizes the importance of considering local ecological and operational characteristics. Both SHPs and LHPs showed highly variable, context-dependent effects on fish communities. We further recommend enhancing EIA monitoring practices, as current data collection methods remain insufficient for accurate impact assessment (Alshuwaikhat, 2005; Cashmore et al., 2010; Morgan, 2012). Therefore, developing site-specific terms of reference, assessing sampling sufficiency, and extending the pre-dam sampling period with seasonally distributed samples should make environmental impact studies more robust. Consequently, developing more sophisticated hydropower classification systems and implementing precise impact evaluation approaches becomes crucial for informing sustainable energy policies and conserving freshwater biodiversity.

Acknowledgments​

---

The authors would like to thank the Instituto Neotropical de Pesquisa Ambiental (Ineo), and the Grupo de Pesquisa em Recursos Pesqueiros e Limnologia (Gerpel). We also would like to thank Companhia Paranaense de Energia (Copel) and Consórcio Empreendedor Baixo Iguaçu (CEBI) for providing data and support this research.

References​

---

Abell R, Thieme ML, Revenga C, Bryer M, Kottelat M, Bogutskaya N et al. Freshwater ecoregions of the world: a new map of biogeographic units for freshwater biodiversity conservation. BioScience. 2008; 58(5):403–14. https://doi.org/10.1641/B580507

Agência Nacional de Energia Elétrica (ANEEL). Resolução Normativa ANEEL nº 1.070, de 29 de agosto de 2023. [Internet]. 2023. Available from: https://www2.aneel.gov.br/cedoc/ren20231070.pdf

Agostinho AA, Gomes LC, Pelicice FM. Ecologia e manejo de recursos pesqueiros em reservatórios do Brasil. Maringá: EDUEM; 2007.

Agostinho AA, Gomes LC, Santos NCL, Ortega JCG, Pelicice FM. Fish assemblages in Neotropical reservoirs: colonization patterns, impacts and management. Fish Res. 2016; 173:26–36. https://doi.org/10.1016/j.fishres.2015.04.006

Agostinho AA, Gomes LC, Suzuki HI, Júlio HF. Migratory fish from the upper Paraná River basin, Brazil. In: Carolsfield J, Harvey B, Ross C, Baer A, editors. Migratory fishes of South America: biology, social importance and conservation status. Victoria: World Fisheries Trust; 2003. p.19–99.

Agostinho AA, Gomes LC, Veríssimo S, Okada EK. Flood regime, dam regulation and fish in the Upper Paraná River: effects on assemblage attributes, reproduction and recruitment. Rev Fish Biol Fish. 2004; 14(1):11–19. https://doi.org/10.1007/s11160-004-3551-y

Agostinho AA, Miranda LE, Bini LM, Gomes LC, Thomaz SM, Suzuki HI. Patterns of colonization in Neotropical reservoirs, and prognoses on aging. In: Tundisi JG, Straskraba M, editors. Theoretical reservoir ecology and its applications. São Carlos: Backhuys Publishers; 1999. p.227–65.

Agostinho AA, Pelicice FM, Gomes LC. Dams and the fish fauna of the Neotropical region: impacts and management related to diversity and fisheries. Braz J Biol. 2008; 68(4):1119–32. https://doi.org/10.1590/S1519-69842008000500019

Alshuwaikhat HM. Strategic environmental assessment can help solve environmental impact assessment failures in developing countries. Environ Impact Assess Rev. 2005; 25(4):307–17. https://doi.org/10.1016/j.eiar.2004.09.003

Baumgartner G, Pavanelli CS, Baumgartner D, Bifi AG, Debona T, Frana VA. Peixes do baixo rio Iguaçu. Maringá: EDUEM; 2012.

Bates D, Mäechler M, Bolker B, Walker S. Fitting linear mixed-effects models using lme4. J Stat Softw. 2015; 67(1):1–48. https://doi.org/10.18637/jss.v067.i01

Botelho A, Ferreira P, Lima F, Pinto LMC, Sousa S. Assessment of the environmental impacts associated with hydropower. Renew Sustain Energy Rev. 2017; 70:896–904. https://doi.org/10.1016/j.rser.2016.11.271

Brasil. Lei nº 13.360, de 17 de novembro de 2016. [Internet]. 2016. Available from: https://www.planalto.gov.br/ccivil_03/_ato2015-2018/2016/lei/L13360.htm

Carvalho DR, Araújo FG. Heterogenisation of riverine ichthyofauna diversity by small hydropower dams. Ecol Freshw Fish. 2024; 33(3):e12775. https://doi.org/10.1111/eff.12775

Cashmore M, Richardson T, Hilding-Ryedvik T, Emmelin L. Evaluating the effectiveness of impact assessment instruments: theorising the nature and implications of their political constitution. Environ Impact Assess Rev. 2010; 30(6):371–79. https://doi.org/10.1016/j.eiar.2010.01.004

Cattaneo MD, Titiunik R. Regression discontinuity designs. Annu Rev Econom. 2022; 14(1):821–51. https://doi.org/10.1146/annurev-economics-051520-021409

Cavalli D, Frota A, Lira AD, Gubiani EA, Margarido VP, Graça WJ. Update on the ichthyofauna of the Piquiri River basin, Paraná, Brazil: a conservation priority area. Biota Neotrop. 2018; 18(2):e20170350. http://dx.doi.org/10.1590/1676-0611-BN-2017-0350

Conselho Nacional do Meio Ambiente (CONAMA). Resolução CONAMA nº 279, de 27 de junho de 2001. [Internet]. 2001. Available from: https://www.icmbio.gov.br/cepsul/images/stories/legislacao/Resolucao/2001/res_conama_279_2001_licenciamentoambientalsimplificadoparaempreendimentoseletricos.pdf

Couto TBA, Olden JD. Global proliferation of small hydropower plants – Science and policy. Front Ecol Environ. 2018; 16(2):91–100. https://doi.org/10.1002/fee.1746

Couto TBA, Rezende RS, Aquino PPU, Costa-Pereira R, Campos GL, Occhi TVT et al. Effects of small hydropower dams on macroinvertebrate and fish assemblages in southern Brazil. Freshw Biol. 2023; 68(6):956–71. https://doi.org/10.1111/fwb.14078

Dallas TA, Kramer AM. Temporal variability in population and community dynamics. Ecology. 2022; 103(2):e03577. https://doi.org/10.1002/ecy.3577

Daga VS, Olden JD, Gubiani EA, Piana PA, Padial AA, Vitule JR. Scale-dependent patterns of fish faunal homogenization in Neotropical reservoirs. Hydrobiologia. 2020; 847(18):3759–72. https://doi.org/10.1007/s10750-019-04145-5

Dagosta FCP, Monção MS, Nagamatsu BA, Pavanelli CS, Carvalho FR, Lima FCT et al.Fishes of the upper rio Paraná basin: diversity, biogeography and conservation. Neotrop Ichthyol. 2024; 22(1):e230066. https://doi.org/10.1590/1982-0224-2023-0066

Delariva RL, Hahn NS, Kashiwaqui EAL. Diet and trophic structure of the fish fauna in a subtropical ecosystem: impoundment effects. Neotrop Ichthyol. 2013; 11(4):891–904. https://doi.org/10.1590/S1679-62252013000400017

Dias RM, Ortega JCG, Strictar L, Santos NCL, Gomes LC, Luz-Agostinho KDG et al. Fish trophic guild responses to damming: variations in abundance and biomass. River Res Appl. 2020; 36(3):430–40. https://doi.org/10.1002/rra.3591

Erlewein A. Disappearing rivers – The limits of environmental assessment for hydropower in India. Environ Impact Assess Rev. 2013; 43:135–43. https://doi.org/10.1016/j.eiar.2013.07.002

Ferraro PJ. Counterfactual thinking and impact evaluation in environmental policy. New Dir Eval. 2009; 2009(122):75–84. https://doi.org/10.1002/ev.297

Finer M, Jenkins CN. Proliferation of hydroelectric dams in the Andean Amazon and implications for Andes-Amazon connectivity. PLoS ONE. 2012; 7(4):e35126. https://doi.org/10.1371/journal.pone.0035126

Gehrke PC, Gilligan DM, Barwick M. Changes in fish communities of the Shoalhaven River 20 years after construction of Tallowa Dam, Australia. River Res Appl. 2002; 18:265–86. https://doi.org/10.1002/rra.669

Gertler PJ, Martinez S, Premand P, Rawlings LB, Vermeersch CM. Impact evaluation in practice. Washington: World Bank Publications; 2016.

Gubiani EA, Angelini R, Vieira LCG, Gomes LC, Agostinho AA. Trophic models in Neotropical reservoirs: testing hypotheses on the relationship between aging and maturity. Ecol Modell. 2011; 222(23-24):3838–48. https://doi.org/10.1016/j.ecolmodel.2011.10.007

Hausman C, Rapson DS. Regression discontinuity in time: considerations for empirical applications. Annu Rev Resour Economics. 2018; 10(1):533–52. https://doi.org/10.1146/annurev-resource-121517-033306

Hirschmann A, Majolo MA, Grillo HCZ. Alterações na ictiocenose do rio Forqueta em função da instalação da Pequena Central Hidrelétrica Salto Forqueta, Putinga, Rio Grande do Sul. Iheringia Sér Zool. 2008; 98(4):481–88. https://doi.org/10.1590/S0073-47212008000400011

Hoeinghaus DJ, Agostinho AA, Gomes LC, Pelicice FM, Okada EK, Latini JD et al. Effects of river impoundment on ecosystem services of large tropical rivers: Embodied energy and market value of artisanal fisheries. Conserv Biol. 2009; 23(5):1222–31. https://doi.org/10.1111/j.1523-1739.2009.01248.x

Jackson DA, Peres-Neto PR, Olden JD. What controls who is where in freshwater fish communities the roles of biotic, abiotic, and spatial factors. Can J Fish Aquat Sci. 2001; 58(1):157–70. https://doi.org/10.1139/f00-239

Jaureguiberry P, Titeux N, Wiemers M, Bowler DE, Coscieme L, Golden AS et al. The direct drivers of recent global anthropogenic biodiversity loss. Sci Adv. 2022; 8(45):eabm9982. https://doi.org/10.1126/sciadv.abm9982

Johnson PTJ, Olden JD, Vander Zanden MJ. Dam invaders: impoundments facilitate biological invasions into freshwaters. Front Ecol Environ. 2008; 6(7):357–63. https://doi.org/10.1890/070156

Jumani S, Rao S, Machado S, Prakash A. Big concerns with small projects: evaluating the socio-ecological impacts of small hydropower projects in India. Ambio. 2017; 46(4):50011. https://doi.org/10.1007/s13280-016-0855-9

Kelly-Richards S, Silber-Coats N, Crootof A, Tecklin D, Bauer C. Governing the transition to renewable energy: a review of impacts and policy issues in the small hydropower boom. Energy Policy. 2017; 101(1):251264. https://doi.org/10.1016/j.enpol.2016.11.035

Keppeler FW, Andrade MC, Trindade PAA, Sousa LM, Arantes CC, Winemiller KO et al. Early impacts of the largest Amazonian hydropower project on fish communities. Sci Total Environ. 2022; 838:155951. https://doi.org/10.1016/j.scitotenv.2022.155951

Kibler KM, Tullos DD. Cumulative biophysical impact of small and large hydropower development in Nu River, China. Water Resour Res. 2013; 49(6):3104–18. https://doi.org/10.1002/wrcr.20243

Kimmel BL, Groeger AW. Limnological and ecological changes associated with reservoir aging. In: Hall GE, Van Den Avyle MJ, editors. Reservoir fisheries management: strategies for the 80’s. Bethesda/Maryland: American Fisheries Society; 1986. p.103–09.

Kimmel BL, Lind OT, Paulson LJ. Reservoir primary production. In: Thornton KW, Kimmel LB, Payne FE, editors. Reservoir limnology: ecological perspectives. New York: John Wiley & Sons; 1990. p.133–93.

King M. Fisheries biology, assessment and management. Oxford: Blackwell Publishing; 2007.

Kubecka J, Matena J, Hartvich P. Adverse ecological effects of small hydropower. River Res Appl. 1997; 13:101–13. https://doi.org/10.1002/(SICI)1099-1646(199703)13:2<101::AID-RRR439>3.0.CO;2-U

Lange K, Wehrli B, Åberg U, Bätz N, Brodersen J, Fischer M et al. Small hydropower goes unchecked. Front Ecol Environ. 2019; 17(5):256–58. https://doi.org/10.1002/fee.2049

Magurran AE. Measuring biological diversity. Hoboken: Blackwell Publishing; 2004.

Mezzaroba L, Debona T, Frota A, Graça WJ, Gubiani EA. From the headwaters to the Iguassu Falls: inventory of the ichthyofauna in the Iguassu River basin shows increasing percentages of nonnative species. Biota Neotrop. 2021; 21(2):e20201083. https://doi.org/10.1590/1676-0611-BN-2020-1083

Mol JH, Mérona B, Ouboter PE, Sahdew S. The fish fauna of Brokopondo Reservoir, Suriname, during 40 years of impoundment. Neotrop Ichthyol. 2007; 5(3):351–68. http://dx.doi.org/10.1590/S1679-62252007000300015

Morgan RK. Environmental impact assessment: the state of the art. Impact Assess Proj Apprais. 2012; 30(1):5–14. https://doi.org/10.1080/14615517.2012.661557

Morrison ML, Block WM, Strickland MD, Collier BA, Peterson MJ. New York: Springer New York; 2008.

Oksanen J, Simpson G, Blanchet FG, Kindt R, Legendre P, Minchin PR et al. vegan: community ecology package [Internet]. Oksanen J; 2025. Available from: https://vegandevs.github.io/vegan/

Orsi ML, Britton JR. Long-term changes in the fish assemblage of a Neotropical hydroelectric reservoir. J Fish Biol. 2014; 84(6):1964–70. https://doi.org/10.1111/jfb.12392

Pelicice FM, Pompeu PS, Agostinho AA. Large reservoirs as ecological barriers to downstream movements of Neotropical migratory fish. Fish Fish. 2015; 16(4):697–715. https://doi.org/10.1111/faf.12089

Pereira LS, Agostinho AA, Delariva RL. Effects of river damming in Neotropical piscivorous and omnivorous fish: feeding, body condition and abundances. Neotrop Ichthyol. 2016; 14(2):e150044. https://doi.org/10.1590/1982-0224-20150044

Petrere Jr. M. Fisheries in large tropical reservoirs in South America. Lakes Reserv. 1996; 2(1–2):111–33. https://doi.org/10.1111/j.1440-1770.1996.tb00054.x

Pinheiro J, Bates D. nlme: Linear and nonlinear mixed effects models. R package version 3. [Internet]. Pinheiro; 2025. Available from: https://CRAN.R-project.org/package=nlme

Pini SFR, Makrakis MC, Neves MP, Makrakis S, Shibatta OA, Kashiwaqui EAL. Ichthyofauna in the last free-flowing river of the Lower Iguaçu basin: the importance of tributaries for conservation of endemic species. Zookeys. 2021; 1041:183–203. https://doi.org/10.3897/zookeys.1041.63884

Poff NL, Allan JD, Bain MB, Karr JR, Prestgaard KL, Richter BD et al. The natural flow regime: a paradigm for river conservation and restoration. BioScience. 1997; 47(11):769–84. https://doi.org/10.2307/1313099

Poff NL, Hart DD. How dams vary and why it matters for the emerging science of dam removal: An ecological classification of dams is needed to characterize how the tremendous variation in the size, operational mode, age, and number of dams in a river basin influences the potential for restoring regulated rivers via dam removal. BioScience. 2002; 52(8):659–68. https://doi.org/10.1641/0006-3568(2002)052[0659:HDVAWI]2.0.CO;2

Poff NL, Zimmerman JKH. Ecological responses to altered flow regimes: a literature review to inform the science and management of environmental flows. Freshw Biol. 2010; 55(1):194–205. https://doi.org/10.1111/j.1365-2427.2009.02272.x

Pusey BJ, Arthington AH, Read MG. Species richness and spatial variation in fish assemblage structure in two rivers of the Wet Tropics of northern Queensland, Australia. Environ Biol Fishes. 1995; 42:181–99. https://doi.org/10.1007/BF00001996

Raio CB, Bennemann ST. A ictiofauna da bacia do rio Tibagi e o projeto de construção da UHE Mauá, Paraná, Brasil. Semina. 2010; 31(1):15–20.

R Development Core Team. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2025. Available from: https://www.r-project.org/

Santos JM, Ferreira MT, Pinheiro AN, Bochechas JH. Effects of small hydropower plants on fish assemblages In medium-sized streams in central and northern Portugal. Aquat Conserv. 2006; 16(4):373–88. https://doi.org/10.1002/aqc.735

Santos NCL, García-Berthou E, Dias JD, Lopes TM, Affonso IP, Severi W et al. Cumulative ecological effects of a Neotropical reservoir cascade across multiple assemblages. Hydrobiologia. 2018; 819(1):77–91. https://doi.org/10.1007/s10750-018-3630-z

Santos JA, Silva CB, Santana HS, Cano-Barbacil C, Agostinho AA, Normando FT et al. Assessing the short-term response of fish assemblages to damming of an Amazonian river. J Environ Manage. 2022; 307:114571. https://doi.org/10.1016/j.jenvman.2022.114571

Scotti A, Jacobsen D, Ștefan V, Tappeiner U, Bottarin R. Small hydropower – small ecological footprint? A multi-annual environmental impact analysis using aquatic macroinvertebrates as bioindicators. Part 1: effects on community structure. Front Environ Sci. 2022; 10:902603. https://doi.org/10.3389/fenvs.2022.902603

Silveira EL, Ballester EL, Costa KA, Scheffer EW, Vaz‐dos‐Santos AM. Fish community response to environmental variations in an impacted Neotropical basin. Ecol Freshw Fish. 2018; 27(4):1126–39. https://doi.org/10.1111/eff.12420

Simpson EH. Measurement of diversity. Nature. 1949; 163(4148):688. https://doi.org/10.1038/163688a0

Smokorowski KE, Randall RG. Cautions on using the before-after-control-impact design in environmental effects monitoring programs. Facets. 2017; 2(1):212–32. https://doi.org/10.1139/facets-2016-0058

Stanley EH, Luebke MA, Doyle MW, Marshall DW. Short-term changes in channel form and macroinvertebrate communities following low-head dam removal. J North Am Benthol Soc. 2002; 21:172–87. https://doi.org/10.2307/1468307

Thieme ML, Tickner D, Grill G, Carvallo JP, Goichot M, Hartmann J et al. Navigating trade-offs between dams and river conservation. Glob Sustain. 2021; 4(e17):1–07. https://doi.org/10.1017/sus.2021.15

Thornton KW, Kimmel BL, Payne FE, editors. Reservoir limnology: ecological perspectives. New York: J. Wiley; 1990.

Turgeon K, Solomon CT, Nozais C, Gregory-Eaves I. Do novel ecosystems follow predictable trajectories? Testing the trophic surge hypothesis in reservoirs using fish. Ecosphere. 2016; 7(12):e01617. https://doi.org/10.1002/ecs2.1617

Underwood AJ. Beyond BACI: the detection of environmental impacts on populations in the real, but variable, world. J Exp Mar Biol Ecol. 1992; 161(2):145–78. https://doi.org/10.1016/0022-0981(92)90094-Q

Underwood AJ. The mechanics of spatially replicated sampling programs to detect environmental impacts in a variable world. Austral Ecol. 1993; 18(1):99–116. https://doi.org/10.1111/j.1442-9993.1993.tb00437.x

Vannote RL, Minshall GW, Cummins KW, Sedell JR, Cushin CE. The river continuum concept. Can J Fish Aquat Sci. 1980; 37:130–37. https://doi.org/10.1139/f80-017

Ward JV, Stanford JA. Ecological connectivity in alluvial river ecosystems and its disruption by flow regulation. River Res Appl. 1995; 11(1):105–19. https://doi.org/10.1002/rrr.3450110109

Winemiller KO. Spatial and temporal variation in tropical fish trophic networks. Ecol Monogr. 1990; 60(3):331–67. https://doi.org/10.2307/1943061

Winemiller KO, McIntyre PB, Castello L, Fluet-Chouinard E, Giarizzo T, Nam S et al. Balancing hydropower and biodiversity in the Amazon, Congo, and Mekong. Science. 2016; 351:128–29. https://doi.org/10.1126/science.aac7082

Wu NC, Jiang WX, Fu XC, Zhou SC, Li FQ, Cai QH et al. Temporal impacts of a small hydropower plant on benthic algal community. Fundam Appl Limnol. 2010; 177:257–66. https://doi.org/10.1127/1863-9135/2010/0177-0257

Zarfl C, Lumsdon AE, Berlekamp J, Tydecks L, Tockner K. A global boom in hydropower dam construction. Aquat Sci. 2015; 77:161–70. https://doi.org/10.1007/s00027-014-0377-0

Authors

---

Luiz Guilherme dos Santos Ribas¹, Anderson Luís Maciel¹, Gilmar Baumgartner¹, Fernando Cesar Alves Ferreira², Juliano Tupan Coragem³ and Éder André Gubiani¹

^[1]    Grupo de Pesquisa em Recursos Pesqueiros e Limnologia (GERPEL), Laboratório de Ictiologia e Estatística Pesqueira (ICTES), Universidade Estadual do Oeste do Paraná, Campus Toledo, Rua Guaíra, 3141, 85903-220 Toledo, PR, Brazil. (LGSR) lg_ribas@hotmail.com, (ALM) maciel_ander@yahoo.com.br, (GB) gilmarbaum03@gmail.com, (EAG) eder.gubiani@unioeste.br (corresponding author).

^[2]    Companhia Paranaense de Energia (COPEL), Rua José Izidoro Biazetto, 158, 81200-240 Curitiba, PR, Brazil. (FCAF) fernando.cesar@copel.com.

^[3]    Consórcio Empreendedor Baixo Iguaçu (CEBI), Rua Tupinambas, 1187, 85760-000 Capanema, PR, Brazil. (JTC) juliano.tupan@baixoiguacu.com.br.

Authors’ Contribution

---

Luiz Guilherme dos Santos Ribas: Conceptualization, Data curation, Formal analysis, Writing-original draft.

Anderson Luís Maciel: Data curation, Investigation, Methodology, Writing-original draft.

Gilmar Baumgartner: Data curation, Funding acquisition, Investigation, Methodology, Writing-review and editing.

Fernando Cesar Alves Ferreira: Data curation, Funding acquisition, Investigation, Project administration, Writing-review and editing.

Juliano Tupan Coragem: Data curation, Funding acquisition, Investigation, Project administration, Writing-review and editing.

Éder André Gubiani: Conceptualization, Data curation, Supervision, Validation, Visualization, Writing-review and editing.

Ethical Statement​

---

Not applicable.

Competing Interests

---

The author declares no competing interests.

Data availability statement

---

Research data is only available upon request. The data supporting this study are available from Companhia Paranaense de Energia (COPEL). Restrictions apply to the availability of these data, which were used under license for this study. Data are available from the authors upon reasonable request and with permission from Companhia Paranaense de Energia (COPEL).

AI statement

---

During the preparation of this paper the authors used ChatGPT (Version 5) by OpenAI in order to review language. After using this tool/service, the authors reviewed and edited the content as needed and takes full responsibility for the content of the publication.

Funding

---

This study was partially funded by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Finance Code 001.

Supplementary Material

---

Supplementary material S1

Supplementary material S2

Supplementary material S3

Supplementary material S4

Supplementary material S5

How to cite this article

---

Ribas LGS, Maciel AL, Baumgartner G, Ferreira FCA, Coragem JT, Gubiani EA. A comparative analysis revealing incongruence in the impacts of small and large hydroelectric power plants on fish communities. Neotrop Ichthyol. 2025; 23(4):e250053. https://doi.org/10.1590/1982-0224-2025-0053

Copyright​

---

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Distributed under

Creative Commons CC-BY 4.0

© 2025 The Authors.

Diversity and Distributions Published by SBI

Accepted March 28, 2025

Submitted August 14, 2025

Epub February 2, 2026