Report

Amazonian Ecology: Tributaries Enhance the Diversity of Electric Fishes

See allHide authors and affiliations

Science  24 Sep 2004:
Vol. 305, Issue 5692, pp. 1960-1962
DOI: 10.1126/science.1101240

Abstract

Neotropical rivers support a diverse array of endemic taxa, including electric fishes of the order Gymnotiformes. A comprehensive survey of the main channels of the Amazon River and its major tributaries (>2000-kilometer transect) yielded 43 electric fish species. Biogeographical analyses suggest that local mainstem electric fish diversity is enhanced by tributaries. Mainstem species richness tends to increase downstream of tributary confluences, and species composition is most similar between tributaries and adjacent downstream mainstem locations. These findings support a “nodal” or heterogeneous model of riverine community organization across a particularly extensive and diverse geographical region.

Biogeographers since Alfred Russel Wallace (1) have observed that the distribution of many terrestrial plant and animal species concords with the geography of major river systems. In the Amazon basin, for example, river and tributary channels appear to limit the ranges of taxa such as primates and lowland-forest birds [(2, 3) but see (4)]. Similarly, divides between river basins can circumscribe the distribution of aquatic taxa such as freshwater fishes (5). Less clear, especially for large river systems such as the lowland Amazon, is the relationship between the structure of rivers [including channel geometry, network configuration, and geomorphology (6, 7)] and the distribution of aquatic species.

Previous studies of fishes in temperate regions have suggested that local species diversity along main river channels is relatively high at tributary confluences (8, 9). Tributaries might enrich mainstem fish diversity by providing access to the mainstem for migrating fishes, offering refugia for early life stages of mainstem species, or enhancing local ecological heterogeneity and thus augmenting local niche diversity (6, 10). The potential impact of tributaries on fish distribution and diversity, however, has never been tested on as broad a spatial scale as that of the Amazon River basin.

Here, we report on the diversity and distribution of electric fishes (Teleostei, Gymnotiformes) along the Amazon mainstem and its major tributaries. Electric fishes are a distinctive and moderately diverse clade endemic to the freshwaters of South and Central America (1113). These fishes are best known for their electroreceptive sense and production of electric fields for near-field orientation and electrocommunication (14). Recent taxonomic studies of these fishes have revealed an impressive degree of diversity, with 46 new species described within the past quarter century (15). In 1992, two of us (C.C.F. and J.G.L.) initiated the “Calhamazon Project,” designed to document the fish fauna of the principal river channels of the Brazilian Amazon. Our field operations produced large samples of fishes trawl-netted in the deep main channels along >2,000 km of the Brazilian Solimões-Amazon mainstem, and in the lower reaches of major tributaries from the Içá River downstream to the Tocantins River (Fig. 1) (16). From these collections, we have recently described two new species of a new genus (17), identified 11 additional undescribed species, and resolved taxonomic errors caused by pronounced sexual dimorphism (18, 19). These efforts set the stage for the present analysis of species diversity and distribution.

Fig. 1.

Map of area sampled, used with permission from the National Aeronautics and Space Administration Jet Propulsion Laboratory. Red circles mark the 13 tributary sites. Samples were taken at three “stations” within each site: (i) within the lower reaches of the tributary, (ii) in the Amazon mainstem, upstream of the tributary confluence, and (iii) in the Amazon mainstem, downstream of the tributary confluence (16).

We focus here on three questions: (i) How many species of electric fishes are there in the mainstem channels of the Brazilian Amazon River and its major tributaries? (ii) What is the contribution, if any, of major tributaries to electric fish species diversity in the Amazon mainstem channels? (iii) How do patterns of electric fish diversity vary along the extent of the Amazon River?

Based on morphological criteria, we identify in our collections 43 electric fish species: 29 Apteronotidae, 8 Sternopygidae, 5 Rhamphichthyidae, and 1 Hypopomidae (table S1). The cumulative number of species collected, plotted as a function of the number of individuals sampled [in which sample order was randomized with the use of EstimateS (20)], yields a curve that is asymptotic (Fig. 2). Thus, our survey of channel species was arguably complete within the limits of our sampling method, and an accurate estimate of species richness was reached after about 16,000 individuals were captured. We do not imply that there are no additional electric fish species in the Amazon; other species are certainly present in microhabitats that were not sampled with our deep-water gear and possibly present in substrate depressions, among the branches of submerged trees, or in shallows near islands or the riverbank.

Fig. 2.

Estimated species richness, randomized over 50 iterations with EstimateS (20), as a function of the number of individuals captured. The asymptote value of 43 species is achieved after about 16,000 individuals.

Our estimate of 43 electric fish species in Amazon channels exceeds that from another major Neotropical river, the Orinoco, for which the same deep-water sampling methods were used and in which 28 species were identified (11). Twenty-two electric fish species were identified from multiple gear (including trawl) samples from the Amazonian Napo and Aguarico rivers of Eastern Ecuador (21); collections in the Negro River also made with multiple gears identified 35 species of electric fishes (22); and 31 electric fish species were collected with seine nets in the marginal habitats of Catalão island, at the confluence of the Amazon and Negro rivers (23). Twenty-two of the Catalão species were also found in our deep-water collections. We believe that the comparatively high species diversity we identified in the channels of the Amazon and its tributaries reflects both our intensive sampling effort and the unusually high richness of the region. It will be instructive to apply a comparable sampling effort to additional rivers to assess the extent to which the Amazon may be a “hot spot” for electric fish diversity (24).

To test whether tributaries enrich local mainstem electric fish diversity, we calculated Coleman rarefaction estimates of species richness on a site-by-site basis (16, 20). This rarefaction procedure allowed us to control for differences among stations in sampling effort and abundance (Table 1). In our calculations, when comparing stations within each site (e.g., upstream of Içá, Içá, and downstream of Içá) we used the lowest observed abundance among the three stations as the basis for estimating and comparing species richness between stations.

Table 1.

Sample of electric fishes collected across the 39 collection stations.

Station n trawls n individuals n species Coleman estimate
Upstream of Içá 49 316 14 14
Içá 49 479 26 23
Downstream of Içá 41 462 21 19
Upstream of Jutaí 34 305 15 11
Jutaí 45 1737 28 15
Downstream of Jutaí 32 103 18 18
Upstream of Juruá 22 17 8 8
Juruá 32 1938 31 13
Downstream of Juruá 25 62 15 8
Upstream of Japurá 9 18 9 9
Japurá 23 1884 25 16
Downstream of Japurá 42 2203 30 11
Upstream of Coari 4 27 5 5
Coari 5 1232 11 7
Downstream of Coari 5 21 7 7
Upstream of Purus 19 231 10 10
Purus 27 2532 31 23
Downstream of Purus 13 808 19 16
Upstream of Manacapuru 5 32 5 5
Manacapuru 5 147 14 8
Downstream of Manacapuru 5 93 9 6
Upstream of Negro 17 746 26 23
Negro 58 3957 34 26
Downstream of Negro 46 497 24 24
Upstream of Madeira 60 1225 29 29
Madeira 63 4806 31 24
Downstream of Madeira 40 1457 31 30
Upstream of Trombetas 34 2761 29 19
Trombetas 67 3937 32 20
Downstream of Trombetas 27 188 16 16
Upstream of Tapajós 37 1087 24 16
Tapajós 38 166 5 5
Downstream of Tapajós 68 919 28 20
Upstream of Xingu 40 669 27 25
Xingu 51 553 25 24
Downstream of Xingu 19 392 21 21
Upstream of Tocantins 14 184 14 10
Tocantins 29 253 13 12
Downstream of Tocantins 3 41 12 12

We identified a significant positive effect of tributaries on Amazon mainstem species richness, in two respects. First, we found that sample stations downstream of each tributary contained more species than did their respective upstream stations (10 of 12 pairwise comparisons, tie at Juruá excluded, one-tail binomial test, P = 0.019; Fig. 3). This pattern occurred in spite of a similar sampling effort at upstream and downstream stations, both in the number of trawl tows (344 at upstream stations compared with 366 at downstream stations) and in the number of individuals captured (7618 at upstream stations compared with 7246 at downstream stations) (Table 1).

Fig. 3.

Pairwise comparison of mainstem electric fish species richness upstream and downstream of 13 tributary confluences. Species richness estimates were calculated with the Coleman rarefaction method (16, 20) (Table 1). Stations downstream of tributaries tended to contain more species than did their respective upstream stations.

Second, we found that species richness within tributaries exceeded that within their adjacent upstream mainstem stations (10 of 13 pairwise comparisons, one-tail binomial test, P = 0.046) but was comparable between tributaries and their adjacent downstream mainstem stations (P > 0.05). An analysis of species composition across sampling stations further illustrates the contribution of tributaries to the mainstem fish fauna. We found that downstream mainstem stations tend to share more species with their adjacent tributaries than with their corresponding upstream mainstem stations (table S2, compare first and second columns, 10 of 13 pairwise comparisons, one-tail binomial test, P = 0.046).

These results agree with our observation that tributaries contained comparatively high abundances of electric fishes. We collected, on average, 48 individuals per trawl in the tributaries, as compared with only 22.1 and 19.8 fishes per trawl in the mainstem (upstream and downstream) stations. The high abundance of fish in tributaries enhances the likelihood that more species move from tributaries into the mainstem, even if relative richness (number of species divided by number of individuals captured) is lower in tributaries than in mainstem stations, as in our samples (calculated from Table 1). We do not imply that electric fishes move only from tributaries to the mainstem downstream of confluences; movements in other directions seem likely. In a study of lateral migration, a small number of electric fishes were captured within large groups of other fishes moving between floodplains and the Amazon mainstem (25).

Our sample also allows a large-scale test of the hypothesis that species diversity accumulates in a downstream direction. This pattern has been shown for fishes in numerous rivers, and has been attributed to increases downriver in food availability, habitat heterogeneity, and habitat stability (7). For this test, we first combined samples from each pair of upstream and downstream mainstem stations to provide a comprehensive estimate of species richness at each site. We then estimated mainstem species richness at each site, again standardized with the Coleman rarefaction method. Across the Amazon, we did not find support for the species addition hypothesis (n = 8 sites; r2 = 0.245; P = 0.212).

Our main finding, that tributaries tend to enhance mainstem electric fish diversity, is perhaps surprising given that our sample covered such a large and diverse geographical region (table S3) (26). If the patterns exhibited by gymnotiforms prove to be general to fishes or other taxa, they would provide a notable exception to the “river continuum” model of lotic community organization, which posits that rivers, particularly the main channels, are continuous entities that express gradients of biotic and physicochemical change. Yet the organization of Amazon River channel fish diversity, at least for electric fishes, shows a large local or “nodal” effect of large tributaries. This pattern is consistent with recent data sets that emphasize river habitat heterogeneity and its influence on species distribution (6, 10, 27).

One possible explanation for the “tributary effect” we describe here is that tributaries carry nutrients and organic material from terrestrial sources and floodplain lakes into the mainstem (28), and may thus provide food and habitat resources for electric fishes. Three major categories of Amazonian tributaries are represented in our sample (29): “black water” (Jutaí and Negro), “clear water” (Trombetas, Tapajós, Xingu, and Tocantins), and “white water” (all other tributaries). White waters are comparatively rich in nutrients and invertebrates, such as zooplankton and aquatic insect larvae, on which electric fishes feed (11, 28). The nine westernmost sites (Fig. 1) were uniform in the positive contribution of tributaries to species diversity and composition. All but two of these tributaries, Jutaí and Negro, are white water. In contrast, exceptions to the reported patterns occurred in the four easternmost tributaries, all of which are clear water. Another possible explanation for the tributary effect centers on differences in flow volume and water levels between the Amazon mainstem and its tributaries, which are caused by basinwide seasonal variation in rainfall and flood stages (30). Resultant “backwater effects” tend to reduce current speed in the lower reaches of tributaries above their confluences, and might directly increase the density and distribution of food resources.

We found that tributaries enrich the species diversity and composition of Amazon mainstem electric fish communities, particularly in the westernmost white water tributaries. However, at a regional scale along the Amazon, these local increases do not result in an overall accumulation of species from upriver to downriver. Together, these results support the idea that rivers are not homogeneous or graded physical and biological entities (6). The generality of the patterns found here for the Amazon remains to be tested with other taxa, including the numerous other fishes collected along with the gymnotiforms.

Supporting Online Material

www.sciencemag.org/cgi/content/full/305/5692/1960/DC1

Materials and Methods

Tables S1 to S3

References

References and Notes

View Abstract

Navigate This Article