Diet seasonality and resource partitioning by large catfishes in the Madeira River, Brazil

Sazonalidade da dieta e partilha de recursos por grandes bagres no Rio Madeira, Brasil

ABSTRACT

Large catfishes form a diverse group with complex life histories, different habitat requirements and home range sizes, which may be related to several ecological aspects. Being among the main top predators in Amazonian river channels, they play a critical role in trophic dynamics, making their conservation particularly significant. However, our ecological understanding of this group remains limited. We evaluated the trophic seasonality and interspecific relationships among large pimelodid catfishes with different life histories and habitat uses in the Madeira River, in the Brazilian Amazon. The dataset was compiled through collaboration with local communities and commercial fishermen at 15 fishing sites, with samplings conducted between April 2009 and September 2011. A total of 225 stomachs contained food items that were identified at the taxonomic level of family. The diet was described for eight species, of which diet seasonality was tested for four, while diet dissimilarity and overlap were estimated for seven regarding only the low-water season. The large pimelodids fed upon many fish families. Most Brachyplatystoma species did not exhibit seasonal shifts in prey consumption. Diet overlap was higher among congeneric and channel-restricted species, particularly Brachyplatystoma platynemum, B. rousseauxii, and B. vaillantii. Despite the diet overlap, some degree of trophic niche partitioning was observed, even among channel-restricted Brachyplatystoma species, mainly when comparing B. filamentosum to the other congeneric species.

KEYWORDS:
top predators; trophic ecology; feeding behavior; Amazon fish; Siluriformes

RESUMO

Grandes bagres constituem um grupo diversificado com histórias de vida complexas, diferentes requisitos de habitat e tamanhos de área de vida relacionados a vários aspectos ecológicos. Como figuram entre os principais predadores de topo em canais fluviais amazônicos, eles têm um papel crítico na dinâmica trófica, portanto sua conservação é particularmente relevante. Entretanto, nosso conhecimento ecológico deste grupo permanece restrito. Nós avaliamos a sazonalidade trófica e as relações interespecíficas de grandes bagres pimelodídeos com diferentes histórias de vida e uso de habitat que habitam o Rio Madeira, na Amazônia brasileira. O conjunto de dados foi construído em colaboração com a população local e pescadores comerciais em 15 localidades de pesca entre abril 2009 e setembro 2011. Um total de 225 estômagos contiveram itens alimentares, que foram identificados em nível taxonômico de família. A dieta foi descrita para oito espécies, das quais a sazonalidade da dieta foi testada para quatro e a similaridade e sobreposição da dieta para sete, apenas em relação à estação de águas baixas. Os grandes pimelodídeos alimentaram-se de muitas famílias de peixes. A maioria das espécies de Brachyplatystoma não apresentou mudanças sazonais no consumo de presas. A sobreposição da dieta foi maior entre espécies congenéricas e restritas ao canal do rio, sendo particularmente alta entre Brachyplatystoma platynemum, B. rousseauxii e B. vaillantii. Apesar da sobreposição nas dietas, observamos algum grau de partição de nicho trófico, mesmo entre espécies de Brachyplatystoma restritas ao canal, principalmente quando se compara B. filamentosum com suas congenéricas.

PALAVRAS-CHAVE:
predadores de topo; ecologia trófica; comportamento alimentar; peixes Amazônicos; Siluriformes

INTRODUCTION

Phylogenetically related species occupying similar habitats and with similar morphological attributes are more likely to compete for a shared resource (Webb et al. 2002). However, in nature, resource partitioning is more common due to slight differences in food consumption driven by ecological adaptations (Huang et al. 2021) and environmental heterogeneity, which provides a wider range of resource opportunities (van der Sleen and Rams 2023). By consuming slightly different forms of resources, using the same resource in different microhabitats or periods of the day, or employing diverse feeding tactics, species can reduce resource overlap, thereby making coexistence easier (Huang et al. 2021; van der Sleen and Rams 2023).

Home range size and habitat seasonality can also play a relevant role in diversifying resource use and mediating species interactions and trophic specialization (Huang et al. 2021; van der Sleen and Rams 2023). For example, specialization may be more likely in large-bodied organisms with large home range when habitat and resource availability fluctuate minimally over time (Huang et al. 2021). In the Neotropics, the seasonal oscillation of the level of large rivers and flooding of marginal habitats strongly regulates habitat availability and heterogeneity, as well as species phenology and movement, which, in turn, influences food availability for fish (Goulding et al. 1988). Flooding affects trophic guilds differently, increasing food for most (e.g., herbivores, omnivores and frugivores), but reducing it for piscivores due to prey dispersion (Goulding et al. 1988; McMeans et al. 2019). Conversely, during the low-water period, higher fish densities lead to increased food availability for piscivorous species (Flood et al. 2023). Habitat use and prey availability have been pointed out as key variables for species coexistence (Mérona and Rankin-de-Mérona 2004; Correa and Winemiller 2014; Mortillaro et al. 2015; van der Sleen and Rams 2023).

Large catfishes (> 1 meter in length) figure among the main natural fish predators in riverine environments (alongside river dolphins and crocodilians), inhabiting predominantly large river channels, but also occupying floodplain habitats (Barthem and Goulding 1997). Goliath catfishes of the genus Brachyplatystoma (Pimelodidae) primarily occupy the main channel of large rivers and undertake longitudinal movements along the rivers for growth and reproduction (Duponchelle et al. 2016; Barthem et al. 2017; Hauser et al. 2018; Hauser et al. 2019). Other pimelodid species belonging to the genera Pseudoplatystoma, Pinirampus, and Zungaro are known to use a diversity of connected habitats, moving seasonally from the river channels to floodplain habitats (Barthem and Goulding 1997; Loubens and Panfili 2000; Fabré and Barthem 2005; Barbarino-Duque 2005; Santos et al. 2006; Pereira et al. 2023). These species undertake both downstream and upstream short-medium migrations (sensuDuponchelle et al. 2021), as well as lateral migrations for reproduction and feeding (Loubens and Panfili 2000; Santos et al. 2006; Alves et al. 2007; Ziober et al. 2012; Avigliano et al. 2023). Movements for feeding purpose seem to be synchronized with fish schools that gather and change habitats during the same periods (Goulding 1980).

Over this complex spatial-temporal movements, habitat segregation has been suggested as one of the main factors responsible for differences in foraging strategies and prey consumption by this group of fish (Barbarino-Duque and Winemiller 2003). The most obvious habitat differences include species restricted to the main river channel and species which move seasonally from the main river channel to the floodplain and small tributaries (Barbarino-Duque and Winemiller 2003).

During the low-water season large catfish species are forced to retreat to the river channel, where they are more likely to co-occur (Goulding 1979). In that season, the density of prey increases while habitat heterogeneity decreases (Lowe-McConnell 1987) possibly leading to higher trophic overlap and competition (Barbarino-Duque and Winemiller 2003). As potential prey fish exhibit a high diversity of migratory behaviors and habitat use throughout the annual flooding cycle of river-floodplain systems, it is reasonable to expect seasonal variations in the diets of large pimelodid catfishes. However, species with large home range and habitat specialization, such as the river-channel catfishes, frequently feed on the most available prey in the habitat (Riotte-Lambert and Matthiopoulos 2020; Huang et al. 2021), which could suppress the seasonal effect of hydrological variation that affects most floodplain fish assemblages (Bogotá-Gregory et al. 2023).

In this context, in this study we examined the diet and feeding ecology of the most abundant large pimelodid catfishes captured in the Madeira River, in the southwestern Brazilian Amazon, between 2009 and 2011. We describe the diet and test hypotheses to address the following questions: (1) Do the diets ofBrachyplatystomaspecies, as habitat specialists with large home ranges, exhibit seasonal variation?; and (2) How do habitat use and congeneric traits relate to the trophic segregation of large catfish species, enabling their co-occurrence in the Madeira River? It was expected that catfish species restricted to the main river channel, such as Brachyplatystoma spp., exhibit low or no seasonal variation in their diet. During the low-water season, when large catfishes are forced to retreat into the river channel and are more likely to co-occur, we expected a higher degree of dietary overlap among congeneric species compared to non-congeneric ones. Regarding habitat use, main channel-restricted species were expected to show higher trophic overlap, segregating from other large river-floodplain catfish species.

MATERIAL AND METHODS

Study area

The Madeira River is formed by the confluence of the Beni and Mamoré rivers and is the largest tributary of the Amazon River in terms of drainage area, discharge, and sediment transport (Goulding 1981; Goulding et al. 2003; Junk et al. 2011; Latrubesse et al. 2017). Before dam construction, the Madeira River basin could be geographically divided into two distinct regions in relation to its main rapids and waterfalls (Jirau, Teotônio, and Santo Antônio waterfalls, which are now submerged by the reservoirs of the Jirau and Santo Antonio hydroelectric power plants; Cella-Ribeiro et al. 2017). Upstream of the waterfalls, the water is highly turbid for most of the year, with its tributaries flowing through deep (8-20 m), narrow, incised channels with very limited floodplain, reflecting its non-meandering morphology. Downstream of the waterfalls, the river widens, has a lower flow speed and larger floodplains with meandering tributaries towards the confluence with the Amazon River (Leite et al. 2011; Torrente-Vilara et al. 2011).

Data collection was carried out along a transect spanning approximately 600 km over the Mamoré - Madeira River basin, including three urban fish markets as well as 12 riverine communities with fisheries landings located on the banks of the river (Figure 1), as described in Doria et al. (2012) and Sant'Anna et al. (2020). The most upstream site was Guajará-Mirim, on the lower Mamoré River, and the most downriver site was in the municipality of Humaitá, in the middle Madeira River (Doria et al. 2012). Data collection occurred between April 2009 and September 2011, covering both the low-water (usually from May to October) and high-water periods (usually from November to April), before reservoir flooding by the dams, which started in September 2011.

Figure 1
Location of the 15 commercial fishery landing localities (circles) and three urban fish markets (squares) that conform the study area along the Madeira and Mamoré rivers in Rondônia state (Brazil).

Study species and data collection

Eight large catfish species figured among the most landed species in the surveyed fish markets (Doria and Lima 2016) and were considered for stomach sampling: Brachyplatystoma platynemum (Boulenger 1898), B. filamentosum (Lichtenstein 1919), B. rousseauxii (Castelnau 1855), B. vaillantii (Valenciennes 1840), Pinirampus pirinampu (Spix and Agassiz 1829), Pseudoplatystoma punctifer (Castelnau1855), P. tigrinum (Valenciennes 1840), and Zungaro zungaro (Humboldt 1821). Data collection was conducted as part of the Fisheries Monitoring and the Fish Ecology and Biology Monitoring programs, associated with the environmental studies of the Santo Antonio and Jirau Hydroelectric power plants (Doria et al. 2010; Doria and Lima 2016; Hauser et al. 2018). Stomachs were obtained weekly from fishes captured by commercial fisheries and landed at the 15 fishing sites across the study area, preserved (see below) and sent monthly to the Laboratory of Ichthyology and Fisheries at the Federal University of Rondônia (LIP/UNIR) in Porto Velho, Rondônia, Brazil. At each site, a local trained collector or a technician from LIP/UNIR recorded the standard length (mm), total weight (g), fishing date and locality during fish landings. Stomachs containing food items were collected from a subsample of the landed target species, identified by species and preserved in a 4% formaldehyde solution, and later transferred to a 70% alcohol solution in the laboratory. The fishing techniques used by fishermen included longlines, bottom longlines, harpoons, gillnets, cast nets, and drift gillnets (Doria and Lima 2016). The main habitat targeted by fisheries and the source of the samples was the main channel of the Madeira River, primarily in the rapids’ sections (see details in Doria and Lima 2016; Sant'Anna et al. 2020).

Stomachs were examined under a stereomicroscope, and prey items were identified to the most detailed taxonomic level possible by taxonomy experts, using specialized literature and reference collections at UNIR (see Queiroz et al. 2013a,b,c). For each stomach containing food items, the relative volume of each food item was visually estimated relative to the total stomach content, considering the total volume of the items as 100% (Goulding et al. 1988). Because fishing with longlines and bottom longlines involves the use of bait that usually is commonly captured fishes (e.g., Curimatidae, Prochilodontidae, Anostomidae species), prey items were carefully validated. Fish prey showing minimal digestion and perforations caused by hooks (i.e., used as bait) were excluded from the relative volume estimate. Captures using other methods, such as nets or harpoons, typically did not involve bait, and are the most representative fisheries techniques except for Zungaro zungaro (Goulding 1979; Doria et al. 2012; Sant'Anna et al. 2020). Empty stomachs were not completely reported by our local collaborators, and it was not possible to assess the frequency of empty stomachs among the species and seasons of the year.

Data analysis

Of the 315 stomachs containing food items obtained during the sampling period, only those containing prey items identified at least to the level of order were included in the analysis. Analyses were performed by grouping the data at the level of order and at the level of family, except for the analysis of diet overlap, in which grouping at order level would overestimate niche overlap.

Diet description was based on all data available for each species, pooling samples from the high and low water seasons. However, due to variations in sample size among species and seasons (samples were about two times higher during low water, Table S1), we selected two distinct data sets to address the research questions posed in this study (Table 1). For Zungaro zungaro only the diet description is presented due to the low sample size. Details on the number of stomachs analyzed, seasonal distribution, fish size and weight, and habitat preferences per species are provided in the Supplementary Material (Tables S1 and S2).

Table 1
Analytical design to address the two main research questions on the trophic ecology of large catfishes from the Madeira River between 2009 and 2011. The numbers represent the total stomachs analyzed for each species and seasonal period.

Question	Analyses	Species	Period
			Low water	High water
1- Seasonal changes in diet	PERMANOVA by species with individual data	Brachyplatystoma filamentosum	20	11
		Brachyplatystoma platynemum	9	11
		Brachyplatystoma rousseauxii	27	25
		Brachyplatystoma vaillantii	72	8
2- Trophic segregation	1. Diet dissimilarity among species - PERMANOVA with individual data; 2. Trophic Niche breadth - PERMDISP with individual data; 3. Niche overlap - t-test for Pianka`s index comparing channel restricted species to others; t-test comparing congeneric and non-congeneric species		Low water
		Brachyplatystoma filamentosum*	20
		Brachyplatystoma platynemum*	8
		Brachyplatystoma rousseauxii*	27
		Brachyplatystoma vaillantii*	72
		Pseudoplatystoma punctifer**	13
		Pseudoplatystoma tigrinum**	6
		Pinirampus pirinampu**	6

• *River channel-restricted species; ** River-floodplain species

Diet seasonality - We tested diet change between seasons (high water x low water) using multivariate permutation analysis of variance (PERMANOVA; Anderson 2001) separately for each of the four Brachyplatystoma species. Diet composition was based on individual values of the relative volume of food items identified to the level of family. Month of capture was categorized as low water (May - October) or high water (November - April). Values were transformed by log (x+1) before constructing a Bray-Curtis dissimilarity matrix. To identify the food items contributing most to dietary differences, we employed the percentage similarity method (SIMPER, Clarke 1993) with 999 randomizations. A PERMANOVA test was also applied to verify sampling year effect, which indicated no significance for any species (results not shown).

Diet segregation during the low-water season - To assess diet dissimilarity, we transformed the individual relative volume values of food items by log (x+1). Next, we created a Bray-Curtis dissimilarity matrix of diet composition and applied a PERMANOVA combined with pairwise tests and Bonferroni correction (Armstrong 2014). To visualize the diet dissimilarity among the species a cluster analysis based on the UPGMA method was applied, based on the sum of relative volumes of each food item per species.

The dietary overlap was then calculated according to Pianka (1974), using the equation O_ij = (Σp_ij x p_ik/ √Σp_ij^2 Σp_ik^2), where O_jk is Pianka’s index of species j and k, p _ij is the relative volume of food item i in the overall food items of species j, p _ik is the proportion of food item i in the total food items of species k, and n is the total number of food items. We estimated the consumption of each food type by the sum of the relative volume of that item for each species, and the results were consistent when using the average of relative volume (results not shown). Dietary overlap values range from 0 (no overlap) to 1 (full overlap), with values greater than 0.6 considered high, values from 0.4 to 0.6 considered intermediate, and values below 0.4 considered low (modified from Grossman 1986). To test if congeneric species had higher trophic overlap compared to non-congeneric species a t-test with Welch’s correction was conducted using Pianka`s overlap values for each group. To test if channel-restricted species segregate from river-floodplain species, another t-test with Welsh correction was conducted by comparing Pianka`s overlap values among Brachyplatystoma species to the overlap values between Brachyplatystoma and other (non-congeneric) species.

The niche breadth was estimated using PERMDISP with a Bray-Curtis dissimilarity matrix of diet composition (Anderson 2006). Dietary niche breadth was measured as the average distance of each individual to the group’s centroid in a Principal Coordinate Analysis (PCoA) (Correa and Winemiller 2014).

All statistical analyses were performed in the R programming environment (R Core Team 2023), with a significance level of p < 0.05. Statistical analyses were conducted using the adonis2 (PERMANOVA), betadisper (PERMDISP), and simper (SIMPER) from the vegan package; hclust (Cluster) function from the stats package, and pairwise.perm.manova (post-hoc test) from the RVAideMemoire package. We used the niche.overlap function from the spaa package for niche overlap analyses.

RESULTS

Minimum and maximum standard length suggest that most fish were subadults or adults (Table S1). From 315 valid samples, 267 stomachs had prey identified at order level and 225 had items identification at family level. Since the overall results comparing family and order level of prey identification did not differ significantly, here we present only the results based on the family-level identification of prey. Gymnotiformes was the most affected group when family level identification is considered, underestimating the relevance of this group in the diet of large catfishes, mainly B. platynemum (Table S3 and S4). Results based on order-level identification are available in the Supplementary Material (Tables S5, S6, S7; Figure S1).

The diet based on the 225 stomachs containing food items identified to family level suggested that the large pimelodid catfishes predominantly consumed fish, with the occasional consumption of macroinvertebrates. Overall, 23 fish families were identified among the prey, with pimelodids representing the main prey consumed by the congeners B. platynemum, B. rousseauxii, and B. vaillantii. Pimelodidae and Doradidae were equally consumed by B. filamentosum (Figure 2a). Pinirampus pirinampu fed mainly on Triportheidae, P. punctifer consumed small Heptapteridae catfishes and Prochilodontidae equally, P. tigrinum strongly preyed upon Prochilodontidae, and Z. zungaro fed on Curimatidae and Prochilodontidae (Figure 2b; see Supplementary Material, Tables S3 and S4 for detailed prey identities).

Figure 2
Sum of relative volumes of prey taxonomically identified to the family level for each study species of large catfish sampled along the Madeira River (Rondônia state, Brazil) from 2009 to 2011. Invertebrates were grouped due to their low representativeness. Species were split into two graphics for better visualization. A - Brachyplatystoma species; B - Other large catfish species.

Diet seasonality

Among the Brachyplatystoma species, the diet significantly varied seasonally only for B. filamentosum, which fed predominantly on Pimelodidae (long-whiskered catfishes) and Serrasalmidae (pacus and piranhas) during the high-water season and shifted to Doradidae (thorny catfishes), Loricariidae (suckermouth catfishes), and several families of Characiformes during the low-water season (Table 2; Table S3).

Table 2
PERMANOVA results for seasonal variations (high vs. low water) in the diet composition of Brachyplatystoma species from the Madeira River, based on the relative volume of consumed prey items for the period of 2009 and 2011. Separate tests were conducted for each species, with statistically significant results highlighted in bold.

Species	DF	F	P
Brachyplatystoma filamentosum	30	2.555	0.03
Brachyplatystoma platynemun	19	1.787	0.13
Brachyplatystoma rousseauxii	51	0.959	0.38
Brachyplatystoma vaillantii	79	1.505	0.32

Trophic segregation during low water

The diets of seven large catfishes analyzed for trophic segregation differed significantly during the low water season (PERMANOVA, F = 2.563; df = 6; 146; p = 0.001; Table 3, above the diagonal; Figure 3). The diet of B. filamentosum differed significantly from that of B. rousseauxii by the consumption of other siluriforms, including Loricariidae (mostly the deep channel dwelling Planiloricaria cf. cryptodon) and a large proportion of Doradidae and Cetopsidae. Brachyplatystoma rousseauxii preyed mainly on Characiformes, especially Curimatidae and Triportheidae. Furthermore, the diets of B. filamentosum differed from Pinirampus pirinampu, which consumed a larger proportion of Triportheidae (Table 3, above the diagonal; Figure 3). Brachyplatystoma filamentosum and B. rousseauxii differed significantly from Pseudoplatystoma tigrinum, which mainly consumed Prochilodontidae and Macrobrachium shrimps (Supplementary Material, Table S8).

Table 3
Results of diet dissimilarity and trophic overlap among seven catfish species from the Madeira River for the period between 2009 and 2011. Pairwise comparisons (PERMANOVA) for diet dissimilarity among large catfish species (Bonferroni-adjusted p-values) are presented above the diagonal, while Pianka’s index of diet overlap is shown below the diagonal. Statistically significant p-values (PERMANOVA) and high trophic overlap values (Pianka’s index) are highlighted in bold.

	B. filamentosum	B. platynemum	B. rousseauxii	B. vaillantii	P. pirinampu	P. punctifer	P. tigrinum
Brachyplatystoma filamentosum	-	1.000	0.042	0.063	0.042	0.231	0.021
Brachyplatystoma platynemum	0.450	-	1.000	1.000	1.000	1.000	0.504
Brachyplatystoma rousseauxii	0.414	0.776	-	1.000	1.000	0.189	0.021
Brachyplatystoma vaillantii	0.531	0.852	0.933	-	0.399	0.357	0.084
Pinirampus pirinampu	0.064	0.488	0.433	0.451	-	1.000	1.000
Pseudoplatystoma punctifer	0.198	0.400	0.407	0.529	0.523	-	1.000
Pseudoplatystoma tigrinum	0.024	0.276	0.183	0.336	0.656	0.637	-

Figure 3
Cluster dendrogram of diet dissimilarity among seven large catfish species during the low-water season in the Madeira River (Rondônia state, Brazil) from 2009 to 2011. The prey items consumed by each species are represented by the sum of the relative volumes of items consumed by all individuals sampled for each species.

The trophic overlap (Table 3, below the diagonal) was significantly higher among the congeneric species pairs (Brachyplatystoma spp. and Pseudoplatystoma spp.), while most of the other pairwise comparisons showed low to moderate levels of overlap (t = 3.311, df = 11.125, p-value <0.01; mean Pianka`s overlap: congeneric = 0.656; non-congeneric = 0.331; Figure 3). Trophic overlap was also significantly higher among the channel-restricted species than between those and the river-floodplain species (t = 3.302, df = 8.466, p-value < 0.01; mean Pianka`s overlap: main channel-restricted = 0.66; main channel-restricted x river-floodplain= 0.313; Table 3; Figure 4). Despite the contrasting diet composition and trophic overlap, all species exhibited similar dietary niche breadths (PERMDISP, F= 0.816; df=6; 146; p > 0.1).

Figure 4
Pianka’s trophic overlap between groups of large catfishes from the Madeira River (Rondônia state, Brazil) based on congeneric relatedness and habitat use.A-congeneric vs. non-congeneric species;B- Pianka`s values among channel-restricted species vs. Pianka`s values among channel-restricted and river-floodplain species. The bold line indicates the mean, the box the standard deviation, the bar the range and the circles are individual values. The groups were compared with t-tests (with Welch’s correction).

DISCUSSION

Our study revealed three relevant aspects of the trophic ecology of large Pimelodidae catfishes in the Madeira River: 1 - most species with habitat use restricted to the main river channel did not exhibit seasonal shifts in prey consumption; 2 - diet overlap during the low water season was higher only among congeneric and channel-restricted species, particularly among B. platynemum, B. rousseauxii, and B. vaillantii; and 3 - there is evidence of trophic niche partitioning among the co-occurring species. This partitioning is particularly evident between B. filamentosum and its congeneric species.

The studied species did not exhibit trophic specialization in the form of predation on any specific prey family, suggesting that prey selection is driven by local and momentary availability (Barthem and Goulding 1997) and possibly mediated by differences in feeding tactics or foraging meso-habitat use. Seasonal variation in prey abundance, driven by fish movements between rivers and floodplains, is a key factor shaping predator-prey relationships in tropical rivers (Lowe-McConell 1987). Species movements, both longitudinal within the river channel and lateral to and from the floodplains, are strongly influenced by water level fluctuations, resulting in varying fish assemblages (Lowe-McConnell 1987; Röpke et al. 2016; Duarte et al. 2022; Bogotá-Gregory et al. 2023).

Many species of Characiformes, including Triportheus spp., Mylossoma spp., Prochilodus nigricans, Psectrogaster spp., and Potamorhina spp., which were the most consumed prey, abound in the river channel and form large schools that undertake annual migrations in both directions. These migrations are typically upstream for feeding and dispersal during flooding and return to the main river channel (or seek refuge in deep and permanent floodplain lakes) during low water periods (Lowe-McConnell 1987). Additionally, the evolution of long-distance migration in some of the studied catfish species exposes them to varying prey abundances across their home ranges, preventing fine-tuned specialization and favoring consumption of broadly and predictably available prey (Huang et al. 2021). As a result, main-channel species showed niche breadths similar to those using broader habitats, such as P. pirinampu, P. punctifer, and P. tigrinum, which was somewhat surprising.

Seasonal and spatial variability might explain why the same species are classified differently across regions, ranging from piscivorous to carnivorous, and why their diet composition differs in a non-specialized manner (see references in Supplementary Material, Table S9). For example, remarkably different diets have been described for Brachyplatystoma spp. across river basins. In the lower Amazon River, Barthem and Goulding (1997) found Characiformes as the main prey of these large catfishes. In the Apure-Arauca rivers, Barbarino-Duque and Winemiller (2003) recorded a high consumption of Gymnotiformes, while in the Madeira River, we observed predominant consumption of Siluriformes, mainly Pimelodidae and Doradidae, and several families of Characiformes.

The large dietary variation observed among all studied species in the Madeira River supports our findings on diet overlap, suggesting that resource partitioning likely occurs due to differences in foraging tactics and meso-habitat use. Nevertheless, higher values of dietary overlap were found among channel-restricted species. All Brachyplatystoma species are bottom feeders (Goulding 1980; Barthem and Goulding 1997; García et al. 2009), but B. filamentosum appears to exhibit some specialization in consuming Loricariidae, particularly Planiloricaria cf. cryptodon, a deep-water species usually captured only using bottom trawl-nets (Cella-Ribeiro 2010). In contrast, the other three Brachyplatystoma species predominantly consumed Pimelodidae (Pimelodina flavipinnis and Pimelodus blochii) and Triportheidae (Triportheus spp.), which are highly abundant along riverbank areas and in the shallow water column (Duarte et al. 2010).

Our results suggest that species co-occurrence likely influences the trophic ecology of Brachyplatystoma in the Madeira River basin. For instance, although the migratory behavior of B. filamentosum is poorly known, if individuals of this species have a smaller home range compared to their congeneric species, it could indicate a different ecological strategy. Goulding (1979) and Barthem et al. (1991) suggest that B. filamentosum may exhibit migratory patterns similar to B. platynemum, which is restricted to the Madeira River (Hauser et al. 2019). In this case, resource competition may have driven B. filamentosum to adapt its diet by exploiting more specific habitats.

Conversely, the broader home ranges of longer-distance migratory Brachyplatystoma species might have constrained dietary specialization, as these species are exposed to more variable prey availability across wide spatial scales. Contemporary niche differentiation would be expected if past competition shaped their trophic behavior (Cavender-Bares et al. 2009). Barbarino-Duque and Winemiller (2003) suggested that, despite high trophic overlap among adults and subadults of B. rousseauxii, B. vaillantii, and B. platynemum, competition among these species in the Apure-Arauca rivers might have been mitigated by their consumption of similar prey from different micro and meso-habitats. A similar pattern may occur in the Madeira River, as the proportion of shared prey families varies among the congeners. However, the higher diet overlap observed in this study, combined with the reduced growth rates of B. rousseauxii during the low water season in the Madeira River (Hauser et al. 2018) - despite high prey availability - supports the hypothesis of interspecific competition. It is worth mentioning that Barthem and Goulding (1997) reported moderate to low diet overlap for juvenile individuals of Brachyplatystoma species in the lower Amazon River, suggesting that asymmetric spatial competition may have driven the evolution of complex life cycles, particularly for B. rousseauxii and B. vaillantii. Our findings also suggest that subadults and adults of Brachyplatystoma species may experience a higher degree of interspecific competition. Further studies examining species co-occurrence at smaller spatial scales, alongside growth and body condition across life stages, could clarify these competitive dynamics.

Pseudoplatystoma punctifer, P. tigrinum, and P. pirinampu also seem to exploit the water column and shallow riverbank areas, mainly preying on Prochilodontidae and Curimatidae. These large catfishes likely have greater ability to exploit the water column and use structurally complex shallow habitats compared to other large Pimelodidae species (Goulding 1980; Barthem and Goulding 1997). Different studies (Supplementary Material, Table S9) found that small Characiformes were the main prey of Pseudoplatystoma spp. and P. pirinampu. These types of prey are highly abundant across diverse habitats of the Amazon river-floodplain systems (Lowe-McConnell 1987), where prey migrations may reduce competitive interactions among those species (van der Sleen and Rams 2023).

Some studies have highlighted that effective conservation of large catfishes requires not only fisheries management but also protection of prey species. Based on Barthem and Goulding’s (1997) findings, Angelini et al. (2006) proposed that the conservation of stocks of B. rousseauxii, B. vaillantii, B. filamentosum, P. tigrinum, and P. fasciatum (= P. punctifer) depends on the protection and management of key Characiformes prey species, such as those in the families Triportheidae, Curimatidae, and Prochilodontidae. Our results support this hypothesis, as these prey groups were predominantly consumed during the high-water period, a time when fish prey density is relatively low for piscivorous predators, especially in large river channels. However, in the Madeira River, adults and subadults of Brachyplatystoma species consumed a large amount of Siluriformes year-round. Thus, protecting main prey groups is essential but challenging at the scale of the Amazon basin.

Studies on the ecology of large Amazonian catfishes remain relatively scarce and geographically limited, especially given the vast home ranges of some species. The huge scale of the Amazon basin, coupled with the high diversity of catfishes and their prey, as well as the variety of habitats they occupy, presents challenges in collecting sufficient data to test hypotheses about prey preferences, dietary specialization, and trophic competition. Commercial fisheries remain the primary source of specimens and data for large catfishes across the Amazon basin. However, reliance on fisheries-derived samples can introduce inherent biases due to the seasonal and selective nature of fishing practices. For example, fisheries in the Madeira River exploit predominantly the main river channel, which strongly limited the number of samples for river-floodplain species during the high-water period in the study. Also, fisheries practices often result in specimens with viscera removed or in advanced stages of digestion or decomposition, making prey identification difficult, along with uncertainties about the precise location and habitat of capture of the large catfishes. These difficulties likely influenced our study by limiting the sample size for some species.

Understanding the relationship of top predators with their prey and how it may shape habitat use, home-range, and movements over the life cycle of large migratory catfishes allows inferring possible evolutionary impacts of anthropic disturbances (Riotte-Lambert and Matthiopoulos 2020). A decade after the construction of two large hydropower plants on the Madeira River, changes in migratory behavior and reductions in the home ranges ofB. rousseauxiihave already been detected (Hauser et al. 2024), though the full impact of disrupted migration routes remains unclear. Documented post-damming declines in catfish and prey abundance (Sant’Anna et al. 2020; Santos et al. 2018; Arantes et al. 2022) further underscore the urgency of basin-wide conservation.

Despite limitations, some consistent patterns of the trophic ecology of the studied species across different regions suggest a robust ecological understanding for these species. The relevance of bottom-dwelling preys and short- to medium-sized migratory prey species highlights the importance of habitat diversity and connectivity in sustaining prey-predator interaction. Seasonal movements of prey between floodplains and river channels are essential to support top predator populations in the Madeira River and throughout the Amazon basin. Therefore, conserving floodplain habitats linked to the home range of those large migratory catfishes and their connectivity to the main river is essential for maintaining these ecological processes and the integrity of aquatic food webs.

CONCLUSIONS

Overall, the catfish species examined did not show predation on a specific prey family. Except for B. filamentosum, channel-restricted species did not change diet seasonally. The trophic partitioning is strong among species, even during the low-water season, and diet overlap was high only among congeneric species that were restricted to the main river channel. Among the Brachyplatystoma species, the lowest overlap was observed when comparing B. filamentosum to its congeneric species, suggesting the more specialized foraging strategy.

ACKNOWLEDGMENTS

The authors thank Santo Antônio Energia, Energia Sustentável do Brasil, Instituto de Estudos e Pesquisas Agroambientais e Organizações Sustentáveis, and Universidade Federal de Rondônia - UNIR for their financial and logistic assistance. We also thank the entire team of the Laboratório de Ictiologia e Pesca - LIP, the fishermen and the local collecting teams that contributed to the data collection. CPR thanks Fundação de Amparo à Pesquisa do Estado de São Paulo- FAPESP for a pos-doctoral fellowship (# 2022/05832-3); JZ thanks Conselho Nacional de Desenvolvimento Científico e Tecnológico - CNPq (# 313184-2014-7) for a former productivity grant.

REFERENCES

• Alves, C.B.M.; Silva, L.G.M.D.; Godinho, A.L. 2007. Radiotelemetry of a female jaú, Zungaro jahu (Ihering, 1898) (Siluriformes: Pimelodidae), passed upstream of Funil Dam, Rio Grande, Brazil. Neotropical Ichthyology 5: 229-232.
• Anderson, M.J. 2001. A new method for non-parametric multivariate analysis of variance.Austral Ecology26:32-46.
• Anderson, M.J. 2006. Distance-based tests for homogeneity of multivariate dispersions. Biometrics 62: 245-253.
• Angelini, R.; Fabré, N.N.; Silva-JR, U.L. 2006. Trophic analysis and fishing simulation of the biggest Amazonian catfish. African Journal of Agricultural Research 1: 151-158.
• Arantes, C.C.; Laufer, J.; Pinto, M.D.D.S.; Moran, E.F.; Lopez, M.C.; Dutka‐Gianelli, J.; et al 2022. Functional responses of fisheries to hydropower dams in the Amazonian Floodplain of the Madeira River. Journal of Applied Ecology 59: 680-692.
• Armstrong, R.A. 2014. When to use the Bonferroni correction. Ophthalmic and Physiological Optics 34: 502-508.
• Avigliano, E.; Pouilly, M.; Vargas, F.; Penha, J.; Silva, N.; Bouchez, J.; et al 2023. Environmental history of one of the largest freshwater neotropical fish (Zungaro jahu) from otolith isotopic analysis. River Research and Applications 39: 887-896.
• Barbarino Duque, A. 2005. Aspectos biológicos y pesqueros de los bagres rayados Pseudoplatystoma fasciatum (Linnaeus 1766) y P. tigrinum (Valenciennes 1840) (Siluriformes: Pimelodidae) en la parte baja de los ríos Apure y Arauca, Venezuela. Memoria de la Fundación La Salle de Ciencias Naturales 163: 71-91.
• Barbarino-Duque, A.; Winemiller, K.O. 2003. Dietary segregation among large catfishes of the Apure and Arauca Rivers, Venezuela. Journal of Fish Biology 63: 410-427.
• Barthem, R.B.; Goulding, M. 1997. Os bagres balizadores: ecologia, migração e conservação de peixes amazônicos. Sociedade Civil Mamirauá, MCT-CNPq, IPAAM, Brasília, 130p.
• Barthem, R.; Goulding, M.; Leite, R.; Cañas, C.; Forsberg, B.; Venticinque, E.; Petry, P.; Ribeiro, M.L.B.; Chuctaya, J.; Mercado, A. 2017. Goliath catfish spawning in the far western Amazon confirmed by the distribution of mature adults, drifting larvae and migrating juveniles. Scientific Reports 7: 41784.
• Barthem, R.B.; Ribeiro, M.C.L.B.; Petrere, M. 1991. Life strategies of some long-distance migratory catfish in relation to hydroelectric dams in the Amazon Basin. Biological Conservation 55: 339-345.
• Bogotá-Gregory, J.D.; Jenkins, D.G.; Lima, F.C.T.; Magurran, A.E.; Crampton, W.G.R. 2023. Geomorphological habitat type drives variation in temporal species turnover but not temporal nestedness in Amazonian fish assemblages. Oikos 2023: e09967. https://doi.org/10.1111/oik.09967
  » https://doi.org/10.1111/oik.09967
• Cavender‐Bares, J.; Kozak, K.H.; Fine, P.V.; Kembel, S.W. 2009. The merging of community ecology and phylogenetic biology. Ecology Letters 12: 693-715.
• Cella-Ribeiro, A. 2010. Variação espacial e temporal na estrutura e composição de assembleias de peixes bentônicos em um trecho do Alto rio Madeira, Rondônia. Master´s dissertation, Instituto Nacional de Pesquisas da Amazônia (INPA), Brazil, 94p. (https://repositorio.inpa.gov.br/bitstream/1/11244/1/Dissertação_%20Ariana%20Cella%20Ribeiro.pdf).
  » https://repositorio.inpa.gov.br/bitstream/1/11244/1/Dissertação_%20Ariana%20Cella%20Ribeiro.pdf
• Cella-Ribeiro, A.; Doria, C.R.C.; Dutka- Gianelli, J.; Alves, H.; Torrente-Vilara, G. 2017. Temporal fish community responses to two cascade run-of-river dams in the Madeira River, Amazon basin. Ecohydrology 10: e1889. https://doi.org/10.1002/eco.1889
  » https://doi.org/10.1002/eco.1889
• Clarke, K.R. 1993. Non-parametric multivariate analyses of changes in community structure. Austral Ecology 18: 117-143.
• Correa, S.B.; Winemiller, K.O. 2014. Niche partitioning among frugivorous fishes in response to fluctuating resources in the Amazonian floodplain forest. Ecology 95: 210-224.
• Doria, C.R.C.; Lima, M.A.L.; Lima, H.M.; Sant’Anna, I.R.A.; Monteiro-Neto, J.M. 2010. Monitoramento da atividade pesqueira no Rio Madeira. In: Doria, C.R.C.; Leite, R.G.; Röpke, C.P.; Queiroz, L.J. (Eds.). Relatório Técnico Anual do Programa de Monitoramento e Conservação da Ictiofauna do Rio Madeira Unpublished technical report, Convênio SAE/UNIR/RIOMAR, Porto Velho, 215p.
• Doria, C.R.C.; Ruffino, M.L.; Hijazi, N.C.; Cruz, R.L. 2012. A pesca comercial na bacia do Rio Madeira, estado de Rondônia, Brasil. Acta Amazonica 42: 29-40.
• Doria, C.R.C.; Lima, M.A.L. 2016. Rio Madeira, Seus Peixes e Sua Pesca EDUFRO, Porto Velho, 163p.
• Duarte, C.; Antão, L.H.; Magurran, A.E.; Deus, C.P. 2022. Shifts in fish community composition and structure linked to seasonality in a tropical river. Freshwater Biology 67: 1789-1800.
• Duarte, C.; Py-Daniel, L.H.R.; Deus, C.P.D. 2010. Fish assemblages in two sandy beaches in lower Purus River, Amazonas, Brazil. Iheringia. Série Zoologia 100: 319-328.
• Duponchelle, F.; Isaac, V.J.; Doria, C.R.C.; Van Damme, P.A.; Herrera‐R, G.A.; Anderson, E.P.; et al 2021. Conservation of migratory fishes in the Amazon basin.Aquatic Conservation: Marine and Freshwater Ecosystems31: 1087-1105.
• Duponchelle, F.; Pouilly, M.; Pécheyran, C.; Hauser, M.; Renno, J-F.; Panfili, J.; et al 2016. Trans-Amazonian natal homing in giant catfish. Journal of Applied Ecology 53: 1511-1520.
• Fabré, N.N.; Barthem, R.B. 2005. O Manejo da Pesca dos Grandes Bagres Migradores: Piramutaba e Dourada no Eixo Solimões-Amazonas IBAMA/PróVárzea, Manaus, 114p.
• Flood, P.J.; Loftus, W.F.; Trexler, J.C. 2023. Fishes in a seasonally pulsed wetland show spatiotemporal shifts in diet and trophic niche but not shifts in trophic position. Food Webs 34: e00265.
• García, A.; Sánchez, H.; Rodríguez, R.; Montreuil, V.; Vargas, G.; Tello, S.; Dunpochelle, F. 2009. Hábitos alimenticios del dorado Brachyplatystoma rousseauxii (Castelnau, 1855) en la Amazonia peruana. Folía Amazónica 18: 7-13.
• Goulding, M. 1979. Ecologia da pesca do Rio Madeira Editora INPA, Manaus, 172p.
• Goulding, M. 1980. Catfishes. In: Goulding, M. (Ed.). The Fishes and the Forest: Explorations in Amazonian Natural History, University of California Press, Berkeley, p187-208.
• Goulding, M. 1981. Man and fisheries on an Amazon frontier. In: Dumont, H.J. (Ed.). Physical and Biological Portrait of the Rio Madeira Basin, W. Junk Publishers, The Hague, p.1-16.
• Goulding, M.; Barthem, R.; Ferreira, E.J. 2003. The Smithsonian Atlas of the Amazon Smithsonian Institution, Washington, 253p.
• Goulding, M.; Carvalho, M.L.; Ferreira, E.G. 1988. Rio Negro, Rich Life in Poor Water: Amazonian Diversity and Food Chain Ecology as Seen Through Fish Communities SPB Academic Publishing, The Hague, 200p.
• Grossman, G.D. 1986. Food resource partitioning in a rocky intertidal fish assemblage. Journal of Zoology 1: 317-355.
• Hauser, M.; Doria, C.R.; Melo, L.R.; Santos, A.R.; Ayala, D.M.; Nogueira, L.D.; et al 2018. Age and growth of the Amazonian migratory catfish Brachyplatystoma rousseauxii in the Madeira River basin before the construction of dams. Neotropical Ichthyology 16: e170130.
• Hauser, M.; Doria, C.R.C.; Santos, R.V.; García-Vasquez, A.; Pouilly, M.; Pécheyran, C.; et al 2019. Shedding light on the migratory patterns of the Amazonian goliath catfish, Brachyplatystoma platynemum, using otolith 87Sr/86Sr analyses. Aquatic Conservation: Marine and Freshwater Ecosystems 29: 397-408.
• Hauser, M.; Doria, C.R.C.; Pécheyran, C.; Ponzevera, E.; Panfili, J.; Torrente-Vilara, G.; et al 2024. Quantitative impacts of hydroelectric dams on the trans-Amazonian migrations of goliath catfish. Conservation Letters 17: e13046.
• Huang, S.; Tucker, M.A.; Hertel, A.G.; Eyres, A.; Albrecht, J. 2021. Scale-dependent effects of niche specialization: The disconnect between individual and species ranges. Ecological Letters 24: 1408-1419.
• Junk, W.J.; Piedade, M.T.F.; Schöngart, J.; Cohn-Haft, M.; Adeney, J.M.; Wittmann, F. 2011. A classification of major naturally occurring Amazonian lowland wetlands. Wetlands 31: 623-640.
• Latrubesse, E.M.; Arima, E.Y.; Dunne, T.; Park, E.; Baker, V.R.; d’Horta, F.M.; et al 2017. Damming the rivers of the Amazon basin. Nature 546: 363-369.
• Leite, N.K.; Krusche, A.V.; Ballester, M.V.R.; Victoria, R.L.; Richey, J.E.; Gomes, B.M. 2011. Intra and interannual variability in the Madeira River water chemistry and sediment load. Biogeochemistry 105: 37-51.
• Loubens, G.; Panfili, J. 2000. Biologie de Pseudoplatystoma fasciatum et P. tigrinum (Teleostei: Pimelodidae) dans le Bassin du Mamoré (Amazonie Bolivienne). Ichthyological Exploration of Freshwaters 11: 13-34.
• Lowe-McConnell, R.H. 1987. Ecological Studies in Tropical Fish Communities Cambridge Tropical Biology Series, Cambridge University Press, Cambridge, 258p.
• McMeans, B.C.; Kadoya, T.; Pool, T.K.; Holtgrieve, G.W.; Lek, S.; Kong, H.; et al 2019. Consumer trophic positions respond variably to seasonally fluctuating environments. Ecology 100: e02570.
• Mérona, B.; Rankin-de-Mérona, J. 2004. Food resource partitioning in a fish community of the central Amazon floodplain. Neotropical Ichthyology 2: 75-84.
• Mortillaro, J.M.; Pouilly, M.; Wach, M.; Freitas, C.E.C.; Abril, G.; Meziane, T. 2015. Trophic opportunism of central Amazon floodplain fish. Freshwater Biology 60: 1659-1670.
• Pereira, L.A.; Castello, L.; Orth, D.J.; Duponchelle, F.; Hallerman, E.M. 2023. A Synthesis of the ecology and conservation of Pseudoplatystoma catfishes in the neotropics. Fishes 8: 306. doi.org/10.3390/fishes8060306
  » https://doi.org/10.3390/fishes8060306
• Pianka, E.R. 1974. Niche overlap and diffuse competition. Proceedings of the National Academy of Sciences 71: 2141-2145.
• Queiroz, L.J.; Torrente-Vilara, G.; Ohara, W.M.; Pires, T.H.S.; Zuanon, J.; Doria, C.R.C. 2013a. Peixes do Rio Madeira, v.1. Dialeto, São Paulo, 400p.
• Queiroz, L.J.; Torrente-Vilara, G.; Ohara, W.M.; Pires, T.H.S.; Zuanon, J.; Doria, C.R.C. 2013b. Peixes do Rio Madeira, v.2. Dialeto, São Paulo , 352p.
• Queiroz, L.J.; Torrente-Vilara, G.; Ohara, W.M.; Pires, T.H.S.; Zuanon, J.; Doria, C.R.C. 2013c. Peixes do Rio Madeira , v.3. Dialeto, São Paulo, 414p.
• R Core Team. 2023. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna. https://www.R-project.org/
  » https://www.R-project.org/
• Riotte-Lambert, L.; Matthiopoulos, J. 2020. Environmental predictability as a cause and consequence of animal movement. Trends in Ecology & Evolution 35: 163-174.
• Röpke, C.P.; Amadio, S.A.; Winemiller, K.O.; Zuanon, J. 2016. Seasonal dynamics of the fish assemblage in a floodplain lake at the confluence of the Negro and Amazon Rivers. Journal of Fish Biology 89: 194-212.
• Sant’Anna, I.R.A.; de Carvalho Freitas, C.E.; Sousa, R.G.C.; dos Anjos, H.D.B.; da Costa Doria, C.R.C. 2020. Fishing production of Pinirampus pirinampu and Brachyplatystoma platynemum catfish has been affected by large dams of the Madeira River (Brazilian Amazon). Boletim do Instituto de Pesca 46: e581. https://doi.org/10.20950/1678-2305.2020.46.2.581
  » https://doi.org/10.20950/1678-2305.2020.46.2.581
• Santos, G.M.; Ferreira, E.J.G.; Zuanon, J.A.S. 2006. Peixes Comerciais de Manaus INPA, Manaus, 144p.
• Santos, R.E.; Pinto-Coelho, R.M.; Fonseca, R.; Simões, N.R.; Zanchi, F.B. 2018. The decline of fisheries on the Madeira River, Brazil: the high cost of the hydroelectric dams in the Amazon Basin. Fisheries Management and Ecology 25: 380-391
• Torrente-Vilara, G.; Zuanon, J.; Leprieur, F.; Oberdorff, T.; Tedesco, P.A. 2011. Effects of natural rapids and waterfalls on fish assemblage structure in the Madeira River (Amazon Basin). Ecology of Freshwater Fish, 20: 588-597.
• van der Sleen, P.; Rams, M. 2023. Flood pulses and fish species coexistence in tropical rivers-a theoretical food web model. Environmental Biology of Fishes 106: 1785-1796.
• Webb, C.O.; Ackerly, D.D.; McPeek, M.A.; Donoghue, M.J. 2002. Phylogenies and community ecology. Annual Review of Ecology and Systematics 33: 475-505.
• Ziober, S.R.; Bialetzki, A.; de Mateus, L.A.F. 2012. Effect of abiotic variables on fish eggs and larvae distribution in headwaters of Cuiabá River, Mato Grosso State, Brazil.Neotropical Ichthyology10: 123-132.

• CITE AS:
  Röpke, C.; Lonardoni, A.; Zuanon, J.; Doria, C.R.C.; Lima, A.L.; Sant’Anna, I.R.A.; Gunther, H.M.L.; Silva, T.M.; Souza, S.T.B.; Torrente-Vilara, G. 2025. Diet seasonality and resource partitioning by large catfishes in the Madeira River, Brazil. Acta Amazonica 55: e55bc25003.

Data availability

The data that support the findings of this study are available in the Mendeley repository and can be accessed at DOI: 10.17632/b7kwsyzmyf.1.

SUPPLEMENTARY MATERIAL

Röpke et al. Diet seasonality and resource partitioning by large catfishes in the Madeira River, Brazil

Figure S1
Cluster dendrogram of diet dissimilarity among eight large catfish species in the Madeira River (Rondônia, Brazil) between 2009 and 2011. Cluster based on order level of items identification during low water season. The cluster was generated based on the average method from the Bray-Curtis metric. The items consumed by each species were considered as the sum of the items consumed by all individuals sampled for each species.

Table S1
Information on habitat use, the number of stomachs containing food items identified at least to family level (N), and the minimum and maximum values of standard length (SL, in cm) and total weight (TW, in g) for eight large catfish species per season (high water and low water period). The fish were sampled in the Madeira River (Rondônia, Brazil) between 2009 and 2011.

Species	Habitat	High water			Low water			Total
		N	SL (Min-Max)	TW (Min-Max)	N	SL (Min-Max)	TW (Min-Max)	N	SL (Min-Max)	TW (Min-Max)
Brachyplatystoma filamentosum	Channel-restricted	11	63-123	4500-30000	20	64-119	3220-24000	31	63-123	3220-30000
Brachyplatystoma platynemum	Channel-restricted	11	54-84	1500-8000	9	51-64	1000-3000	20	51-84	1000-8000
Brachyplatystoma rousseauxii	Channel-restricted	25	67.5-105	880-18000	27	50-101.5	1540-28000	52	50-105	880-28000
Brachyplatystoma vaillantii	Channel-restricted	8	33-59	500-3500	72	25-47	250-4000	80	25-59	250-4000
Pinirampus pirinampu	River-floodplain	-	-	-	6	39-44.2	630-1309	6	39-44.2	630-1309
Pseudoplatystoma punctifer	River-floodplain	3	-	-	13	44-85	1000-6000	16	44-85	1000-6000
Pseudoplatystoma tigrinum	River-floodplain	-	-	-	6	77-88	6000-8000	6	77-88	6000-8000
Zungaro zungaro	Channel-restricted	11	-	-	3	137	-	14	137	-

Table S2
Number of stomachs per species and season with content identified at least to the order level. The data refer to eight large catfish species sampled in the Madeira River (Rondônia, Brazil) between 2009 and 2011.

Species	High water	Low water	Total
Brachyplatystoma filamentosum	16	23	39
Brachyplatystoma platynemum	14	15	29
Brachyplatystoma rousseauxii	25	29	54
Brachyplatystoma vaillantii	9	87	96
Pinirampus pirinampu		8	8
Pseudoplatystoma punctifer	3	16	19
Pseudoplatystoma tigrinum		7	7
Zungaro zungaro	11	4	15
Total	78	189	267

Table S3
Sum of relative volumes of prey taxonomically identified at family level in the stomachs of eight large catfish species sampled in the Madeira River (Rondônia, Brazil) between 2009 and 2011 in the high and low water season. Insect orders and families were pooled due to low representativeness. Values are based on 225 stomachs.

Prey family	Brachyplatystoma filamentosum		Brachyplatystoma platynemum		Brachyplatystoma rousseauxii		Brachyplatystoma vaillantii		Pinirampus pirinampu		Pseudoplatystoma punctifer		Pseudoplatystoma tigrinum		Zungaro zungaro
	High	Low	High	Low	High	Low	High	Low	High	Low	High	Low	High	Low	High	Low
Synbranchidae	0	0	0	0	0	0	0	0	-	0	0	100	-	0	0	0
Acestrorhynchidae	0	0	0	0	0	0	0	0	-	0	0	100	-	0	0	0
Doradidae	100	660	0	100	0	80	100	810	-	0	0	0	-	0	0	0
Loricariidae	0	400	0	0	0	0	0	0	-	0	0	0	-	0	0	0
Callichthyidae	0	0	0	0	0	0	0	0	-	0	0	100	-	0	0	0
Sciaenidae	0	0	0	0	0	0	0	100	-	0	0	0	-	0	0	0
Anostomidae	0	100	0	0	300	0	0	0	-	0	0	100	-	0	0	0
Pristigasteridae	100	0	0	0	180	0	0	0	-	0	0	0	-	0	0	0
Bryconidae	0	0	0	0	0	0	0	100	-	0	0	0	-	0	0	0
Cichlidae	0	0	0	0	0	0	0	0	-	0	100	50	-	0	0	0
Serrasalmidae	300	0	120	0	100	0	100	0	-	100	0	60	-	0	50	0
Characidae	0	100	0	0	0	30	0	365	-	0	0	150	-	0	100	0
Chilodontidae	0	0	0	0	0	60	0	0	-	0	0	0	-	0	90	0
Curimatidae	0	100	100	230	357	450	0	1380	-	40	0	100	-	70	535	0
Cynodontidae	0	0	0	0	180	0	0	0	-	0	0	0	-	0	0	0
Hemiodontidae	0	100	0	30	30	0	0	0	-	0	0	0	-	0	0	0
Iguanodectidae	0	0	0	0	0	100	0	0	-	0	0	0	-	0	0	0
Prochilodontidae	0	0	0	0	0	0	0	485	-	110	0	200	-	300	325	0
Triportheidae	0	0	580	150	293	440	210	820	-	300	0	100	-	130	0	0
Apteronotidae	0	0	0	150	0	0	0	60	-	0	0	0	-	0	0	0
Cetopsidae	100	200	100	0	0	200	0	0	-	0	0	0	-	0	0	0
Heptapteridae	0	0	0	0	0	0	0	0	-	0	200	0	-	0	0	0
Pimelodidae	450	300	100	200	900	1280	390	2450	-	50	0	100	-	0	0	200
Macrobrachium spp.	0	0	0	0	45	0	0	5	-	0	0	50	-	0	0	0
Trichodactylus sp.	0	0	0	0	0	0	0	3	-	0	0	0	-	0	0	100
Megaloptera	0	0	0	0	0	0	0	100	-	0	0	0	-	0	0	0

Table S4
Sum of relative volumes of food items (prey) identified to the lowest taxonomic level in 315 stomachs of eight large catfish species sampled in the Madeira River (Rondônia, Brazil) during the high and low water season between April 2009 and September 2011.

Prey	Brachyplatystoma filamentosum			Brachyplatystoma platynemum			Brachyplatystoma rousseauxii			Brachyplatystoma vaillantii			Pinirampus pirinampu			Pseuplatystoma punctifer			Pseuplatystoma tigrinum			Zungaro zungaro
	Total	High	Low	Total	High	Low	Total	High	Low	Total	High	Low	Total	High	Low	Total	High	Low	Total	High	Low	Total	High	Low
	41	16	25	32	14	18	60	29	31	125	9	116	10	-	10	21	4	17	7	-	7	15	11	4
Acestrorhynchus spp.																100		100
Adontosternarchus balaenops				100		100
Amblydoras spp.										100		100
Anadoras weddellii	100		100							430		430
Ancistrus sp. sideral	100		100
Anodus spp.	100		100				30	30
Anostomidae							200	200
Aphyocharax sp.																50		50
Apteronotus bonaparti				50		50
Astrodoras asterifrons										20		20
Astyanax aff. bimaculatus										10		10
Brycon spp.										100		100
Bryconops alburnoides							100		100
Caenotropus sp.							60		60													90	90
Calophysus macropterus				90		90	100		100	200		200										100		100
Cetopsis candiru	100		100
Cetopsis coecutiens	200	100	100				200		200
Cetopsis spp.				100	100
Characidae																						100	100
Characiformes	250	250		180	180		100		100	437		437	100		100	340		340	180		180	100		100
Clupeiformes													100		100
Coleoptera										23		23
Crenicichla spp.																100	100
Curimatella meyeri	100		100							100		100
Curimatidae				60		60	90	90		170		170	40		40	100		100				150	150
Doradidae	200		200	100		100				40	20	20
Duopalatinum peruano	50	50
Escales n.i.										200		200
Exallodontus aguanai	100		100
Gymnotyformes	20		20	400		400	140	100	40	300		300
Hemiodontidae				30		30
Hoplosternum littorale																100		100
Hypophthalmus marginatus							680		680
Hypophthalmus spp.							400	100	300	20		20
Leporinus spp.	100		100
Leptodoras juruensis	200	100	100
Macrobrachium spp.							45	45		5		5				50		50
Megalonema spp.										135		135
Megaloptera										100		100
Moenkhausia spp.										55		55
Myleinae	200	200											100		100
Mylossoma duriventre	100	100		120	120																	50	50
Mylossoma spp.							100	100		100	100					60		60
Nemadoras humeralis	100		100
Bones n.i.										130	30	100
Fish n.i.	240		240	300	100	200	635	415	220	2870	100	2770	200		200	400	100	300	20		20
Pellona castelnaeana							180	180
Pellona spp.	100	100
Pimelodella spp.																200	200
Pimelodidae	300	200	100				300	300		580		580
Pimelodina flavipinnis	200	200		100	100		600	400	200				50		50
Pimelodus blochii	100		100	110		110	100	100		1685	190	1495										100		100
Pimelodus spp.										120	100	20				100		100
Planiloricaria cryptodon	300		300
Plasgioscion spp.										100		100
Potamorhina spp.										100		100
Potamorhina altamazonica										220		220
Potamorhina latior							200	200											70		70	100	100
Prochilodus nigricans										485		485	110		110				300		300	325	325
Psectrogaster amazonica				70		70	100		100	200		200
Psectrogaster rutiloides				100		100	320		320	390		390										285	285
Psectrogaster spp.				100	100		30		30	200		200
Psectrogastrer essequibensis							66.66	66.7
Pseudanos trimaculatus																100		100
Pterodoras granulosus	100		100							220	80	140
Insects										10		10
Rhaphiodon vulpinus							180	180
Roeboides myersi							30		30
Schizodon fasciatus							100	100
Siluriformes	580	300	280	460	120	340	100		100	1145	70	1075				100		100
Sternarchogiton sp.										60		60
Symbranchus spp.																100		100
Tetragonopterinae										100		100				100		100
Tetragonopterus argenteus	100		100
Tetragonopterus spp.										200		200
Trachydoras spp.	60		60				80		80	100		100
Trichodactylus sp.										3		3										100		100
Trichoptera										7		7
Triportheus spp.				680	580	100	273.3	33.3	240	640	110	530	100		100				130		130
Triportheus angulatus				50		50	460	260	200	390	100	290	200		200
Triportheus auritus																100		100

Table S5
Results of PERMANOVA analysis for seasonal change in diet (high water x low water) based on prey identified at least to the order level for four species of large catfish in the Madeira River (Rondônia, Brazil) between 2009 and 2011. The statistic presented is derived from a separate test for each species. Statistically significant results are presented in bold.

Species	DF	F	P
Brachyplatystoma filamentosum	38	1.977	0.15
Brachyplatystoma platynemun	28	3.679	0.05
Brachyplatystoma rousseauxii	54	1.629	0.215
Brachyplatystoma vaillantii	95	5.212	0.705

Table S6
Results of pairwise comparison through PERMANOVA (DF = 188; R2 = 0.16; F = 4.977; p = 0.001) for diet dissimilarity (Bonferroni adjusted p-value) among eight species of large catfish sampled in the Madeira River (Rondônia, Brazil) from 2009 to 2011. Analysis based on prey identified at least to the order level. Significant p values are in bold.

	Brachyplatystoma filamentosum	B. platynemum	B. rousseauxii	B. vaillantii	P. pirinampu	P. punctifer
Brachyplatystoma platynemum	0.504	-	-	-	-	-
Brachyplatystoma rousseauxii	1.000	1.000	-	-	-	-
Brachyplatystoma vaillantii	0.392	1.000	1.00	-	-	-
Pinirampus pirinampu	0.028	0.252	0.364	0.168	-	-
Pseudoplatystoma punctifer	0.028	0.448	0.560	0.308	1.000	-
Pseudoplatystoma tigrinum	0.028	0.056	0.224	0.056	1.000	1.000
Zungaro zungaro	1.000	1.000	1.000	1.000	1.000	0.392

Table S7
Similarity Percentage - Simper pairwise contrast analysis for differences in diet composition during the low water period for species of large catfish sampled in the Madeira River (Rondônia, Brazil) between 2009 and 2011. Contrast analyses are presented only when the PERMANOVA indicated significant difference between species based on food items identified at order level. Ratio = ratio of the average dissimilarity to the standard deviation of the dissimilarity; Sd = standard deviation; Av.sp1 = average of relative volume for the first species; Av.sp2 = average of relative volume for the second species; cumsum = cumulative sum in explicability to difference in diet; p = significance level of difference for each item. Significant p-values are shown in bold.

Brachyplatystoma filamentosum X Pseudoplatystoma punctifer
Order	average	sd	ratio	Av.sp1	Av.sp2	cumsum	p
Siluriformes	0.34220	0.22436	152.510	378.100	0.86500	0.474	0.001
Characiformes	0.31860	0.22908	139.090	0.80300	337.600	0.915	0.001
Synbranchiformes	0.03100	0.12043	0.25750	0.00000	0.28800	0.958	0.024
Cichliformes	0.01960	0.07596	0.25760	0.00000	0.24600	0.985	0.025
Gymnotiformes	0.01090	0.05117	0.21260	0.13200	0.00000	1.000	0.907
Brachyplatystoma filamentosum X Pinirampus pirinampu
Order	average	sd	ratio	Av.sp1	Av.sp2	cumsum	p
Siluriformes	0.36730	0.20722	177.260	378.100	0.49100	0.464	0.001
Characiformes	0.35130	0.21206	165.640	0.80300	395.300	0.908	0.001
Clupeiformes	0.06200	0.16477	0.37640	0.00000	0.57700	0.987	0.012
Gymnotiformes	0.01070	0.05041	0.21200	0.13200	0.00000	1.000	0.853
Brachyplatystoma filamentosum X Pseudoplatystoma tigrinum
Order	average	sd	ratio	Av.sp1	Av.sp2	cumsum	p
Characiformes	0.40840	0.18807	217.130	0.80300	458.400	0.494	0.001
Siluriformes	0.40730	0.18959	214.830	378.100	0.00000	0.987	0.001
Gymnotiformes	0.01100	0.05181	0.21250	0.13200	0.00000	1.000	0.839

Table S8
Similarity Percentage-Simper pairwise contrast analysis for differences in diet composition based on family level of prey identification during the low water period between species of large catfish sampled in the Madeira River from 2009 to 2011. Contrast analyses are presented only when the PERMANOVA indicated significant difference between species based on food items identified at family level. Ratio -refers to the ratio of the average dissimilarity to the standard deviation of the dissimilarity for a particular táxon; Sd - Standard deviation; Av.sp1 - average of relative volume for the first species; Av.sp2 - average of relative volume for the second species; cumsum - cumulative sum in explicability to difference in diet; p - significance level of difference for each item. Significant p values are shown in bold.

Brachyplatystoma filamentosum X Brachyplatystoma rousseauxii
Family	average	sd	ratio	Av.sp1	Av.sp2	cumsum	p
Pimelodidae	0.23917	0.24704	0.96810	0.69230	221.390	0.266	0.056
Doradidae	0.17553	0.23447	0.74860	159.010	0.16280	0.461	0.007
Curimatidae	0.10736	0.19455	0.55180	0.23080	0.92370	0.581	0.840
Loricariidae	0.09798	0.19710	0.49710	0.92300	0.00000	0.690	0.001
Triportheidae	0.08335	0.18052	0.46170	0.00000	0.82130	0.782	0.837
Cetopsidae	0.07873	0.18159	0.43350	0.46150	0.34190	0.870	0.004
Characidae	0.03204	0.11285	0.28390	0.23080	0.12720	0.905	0.718
Anostomidae	0.02450	0.10731	0.22830	0.23080	0.00000	0.933	0.192
Hemiodontidae	0.02450	0.10731	0.22830	0.23080	0.00000	0.960	0.145
Iguanodectidae	0.01857	0.09479	0.19590	0.00000	0.17090	0.981	0.143
Chilodontidae	0.01750	0.08933	0.19590	0.00000	0.15230	1.000	0.138
Brachyplatystoma filamentosum x Pinirampus pirinampu
Family	average	sd	ratio	Av.sp1	Av.sp2	cumsum	p
Triportheidae	0.25072	0.25181	0.99570	0.00000	230.760	0.256	0.005
Doradidae	0.15848	0.21995	0.72050	159.010	0.00000	0.418	0.181
Prochilodontidae	0.10782	0.15319	0.70380	0.00000	134.040	0.528	0.217
Pimelodidae	0.10539	0.17835	0.59090	0.69230	0.65530	0.636	0.992
Loricariidae	0.09136	0.18553	0.49240	0.92300	0.00000	0.729	0.048
Serrasalmidae	0.08357	0.18768	0.44530	0.00000	0.76920	0.814	0.017
Curimatidae	0.06772	0.13880	0.48790	0.23080	0.61890	0.883	0.903
Cetopsidae	0.04568	0.13898	0.32870	0.46150	0.00000	0.930	0.222
Anostomidae	0.02284	0.10091	0.22630	0.23080	0.00000	0.953	0.292
Characidae	0.02284	0.10091	0.22630	0.23080	0.00000	0.977	0.735
Hemiodontidae	0.02284	0.10091	0.22630	0.23080	0.00000	1.000	0.189
Brachyplatystoma filamentosum X Pseudoplatystoma tigrinum
Family	average	sd	ratio	Av.sp1	Av.sp2	cumsum	p
Prochilodontidae	0.29985	0.21878	137.060	0.00000	286.730	0.302	0.001
Doradidae	0.17048	0.23554	0.72380	159.010	0.00000	0.474	0.118
Triportheidae	0.13016	0.19574	0.66500	0.00000	134.150	0.605	0.321
Loricariidae	0.09823	0.19876	0.49420	0.92300	0.00000	0.704	0.036
Curimatidae	0.09682	0.19403	0.49900	0.23080	0.71040	0.802	0.785
Pimelodidae	0.07367	0.17735	0.41540	0.69230	0.00000	0.876	0.995
Cetopsidae	0.04912	0.14895	0.32980	0.46150	0.00000	0.926	0.212
Anostomidae	0.02456	0.10817	0.22700	0.23080	0.00000	0.951	0.199
Characidae	0.02456	0.10817	0.22700	0.23080	0.00000	0.975	0.634
Hemiodontidae	0.02456	0.10817	0.22700	0.23080	0.00000	1.000	0.152
Brachyplatystoma rousseauxii X Pseudoplatystoma tigrinum
Family	average	sd	ratio	Av.sp1	Av.sp2	cumsum	p
Prochilodontidae	0.29307	0.21579	135.810	0.00000	286.730	0.316	0.001
Pimelodidae	0.23158	0.24504	0.94510	221.390	0.00000	0.566	0.240
Triportheidae	0.16525	0.21288	0.77620	0.82130	134.150	0.745	0.163
Curimatidae	0.13882	0.21015	0.66060	0.92370	0.71040	0.894	0.496
Cetopsidae	0.03638	0.12986	0.28020	0.34190	0.00000	0.934	0.225
Iguanodectidae	0.01819	0.09362	0.19430	0.17090	0.00000	0.953	0.116
Doradidae	0.01774	0.09134	0.19430	0.16280	0.00000	0.973	0.974
Chilodontidae	0.01713	0.08823	0.19420	0.15230	0.00000	0.991	0.117
Characidae	0.00825	0.04233	0.19500	0.12720	0.00000	1.000	0.868

Table S9
Most important food items consumed by seven species of large catfish according to the published literature.

Species	Rivers	Main items consumed	Trophic category	Reference
Brachyplatystoma filamentosum	Amazon, tributaries and estuary	Semaprochilodus sp. and Pimelodidae	Not determined	Barthem and Goulding 1997
Brachyplatystoma filamentosum	Amazon	Fish	Carnivorous	Santos et al. 2006
Brachyplatystoma filamentosum	Tocantins	Fish	Carnivorous	Santos et al.1984
Brachyplatystoma filamentosum	Machado	Curimatidae, Pimelodidae and Loricariidae	Not determined	Goulding 1980
Brachyplatystoma flavicans (=Brachyplatystoma rousseauxii)	Machado	Characiformes	Not determined	Goulding 1980
Brachyplatystoma flavicans (=Brachyplatystoma rousseauxii)	Amazon, tributaries and estuary	Hemiodus sp. and Triportheus sp.	Not determined	Barthem and Goulding 1997
Brachyplatystoma flavicans (=Brachyplatystoma rousseauxii)	Apure and Arauca	Gymnotiformes and fish remains	Piscivorous	Barbarino and Winemiller 2003
Brachyplatystoma flavicans (=Brachyplatystoma rousseauxii)	Madeira	Fish	Piscivorous	Goulding 1981
Brachyplatystoma flavicans (=Brachyplatystoma rousseauxii)	Tocantins	Fish	Carnivorous	Santos et al.1984
Brachyplatystoma rousseauxii	Ucayali, Marañón and Amazon	Curimatidae and Characidae	Piscivorous	García et al. 2009
Brachyplatystoma rousseauxii	Amazon	Fish	Piscivorous	Santos et al. 2006
Brachyplatystoma rousseauxii	Amazon estuary	Amuré (Gobioides sp.), sardinha (Triportheus sp.) and shrimp	Not determined	Fabré and Barthem 2005
Brachyplatystoma rousseauxii	Amazon	Sardinha (Triportheus sp.) and shrimp	Not determined	Fabré and Barthem 2005
Brachyplatystoma rousseauxii	Solimões	Sardinha (Triportheus sp.) and branquinha (Curimatidae)	Not determined	Fabré and Barthem 2005
Brachyplatystoma vaillantii	Amazon, tributaries and estuary	Semaprochilodus sp. and Doradidae	Not determined	Barthem and Goulding 1997
Brachyplatystoma vaillantii	Apure and Arauca	Fish remains	Piscivorous	Barbarino and Winemiller 2003
Brachyplatystoma vaillantii	Amazon	Fish and occasional invertebrates	Piscivorous	Santos et al. 2006
Brachyplatystoma vaillantii	Tocantins	Fish	Carnivorous	Santos et al. 1984
Brachyplatystoma vaillantii	Amazon estuarye	Amuré (Gobioides sp.), sardinha (Triportheus sp.) and shrimp	Not determined	Fabré and Barthem 2005
Brachyplatystoma vaillantii	Amazon	Sardinha (Triportheus sp.) and shrimp	Not determined	Fabré and Barthem 2005
Brachyplatystoma vaillantii	Solimões	Sardinha (Triportheus sp.) and branquinha (Curimatidae)	Not determined	Fabré and Barthem 2005
Goslinea platynema (=Brachyplatystoma platynemum)	Apure and Arauca	Gymnotiformes	Piscivorous	Barbarino and Winemiller 2003
Goslinea platynema (=Brachyplatystoma. platynemum)	Amazon	Fish and invertebrates	Carnivorous	Santos et al. 2006
Goslinea platynema (Brachyplatystoma platynemum)	Tocantins	Fish	Carnivorous	Santos et al. 1984
Pseudoplatystoma fasciatum (=Pseudoplatystoma punctifer)	Machado	Anostomidae, Loricariidae, crustaceans and Cichlidae	Not determined	Goulding 1980
Pseudoplatystoma fasciatum (=Pseudoplatystoma punctifer)	Amazon, tributaries and estuary	Curimatidae	Not determined	Barthem and Goulding, 1997
Pseudoplatystoma fasciatum (=Pseudoplatystoma punctifer)	Apure and Arauca	Fish remains, M. duriventre and P. mariae	Piscivorous	Barbarino and Winemiller 2003
Pseudoplatystoma fasciatum (=Pseudoplatystoma punctifer)	Amazon	Fish and csrustaceas	Carnivorous	Santos et al. 2006
Pseudoplatystoma fasciatum (=Pseudoplatystoma punctifer)	Tocantins	Fish	Carnivorous	Santos et al. 1984
Pinirampus pirinampu	Apure and Arauca	Fish remains	Piscivorous	Barbarino and Winemiller 2003
Pinirampus pirinampu	Mamoré	Fish	Piscivorous	Pouilly et al. 2004
Pinirampus pirinampu	Tocantins	Fish and Decapoda	Carnivorous non specialist	Mérona et al. 2001
Pinirampus pirinampu	Amazon	Fish and occasional invertebrates	Piscivorous	Santos et al. 2006
Pinirampus pirinampu	Tocantins	Fish	Carnivorous non specialist	Santos et al. 1984
Paulicea lutkeni (=Zungaro zungaro)	Madeira	Characiformes	Not determined	Goulding 1981
Paulinea lutkeni (=Zungaro zungaro)	Tocantins	Fish	Carnivorous	Santos et al. 1984
Zungaro zungaro	Amazon	Fish and Crustacean	Carnivorous	Santos et al. 2006

References citted only in the supplementary material

Muñoz, H.; Van Damme P.A. 1998. Parametros de Reproduccion de 4 Especies de Peces Comerciales (Pseudoplatystoma fasciatum, P. tigrinum, Colossoma macropomum y Piaractus brachypomum) en la Cuenca del Rio Ichilo (Bolivia). Revista Boliviana de Ecologia y Conservacion Ambiental 4:39-54.

Mérona, B.; Santos, G.M.; Almeida, R.G. 2001. Short term effects of Tucuruí Dam (Amazonia, Brazil) on the trophic organization of fish communities.Environmental Biology of Fishes60, 375-392.

Pouilly, M.; Yunoki, T.; Rosales, C.; Torres, L. 2004. Trophic structure of fish assemblages from Mamore ́ River floodplain lakes (Bolivia). Ecology of Freshwater Fish 2004: 13: 245-257

Santos, G.M.; Jegu; M.; Mérona B. 1984. Catálogo de peixes comerciais do baixo rio Tocantins. Eletronorte/INPA, Brasilia. 83 pp.

Edited by

• ASSOCIATE EDITOR:
  Naraiana Benone https://orcid.org/0000-0003-4155-9938

Publication Dates

• Publication in this collection
  29 Sept 2025
• Date of issue
  2025

History

• Received
  14 Jan 2025
• Accepted
  13 June 2025