2013, influence of local and landscape characteristics on the distribution and diversity of fish...

8/11/2019 2013, Influence of Local and Landscape Characteristics on the Distribution and Diversity of Fish Assemblages of Str

1/12

, 2013, vol. 25, no. 4, p. 451-462 Acta Limnologica Brasiliensia http://dx.doi.org/10.1590/S2179-975X2013000400010

Inuence of local and landscape characteristics on thedistribution and diversity of sh assemblages of streams in the

Ivinhema River basin, Upper Paran RiverInuncia de caractersticas locais e da paisagem sobre a distribuio e diversidade

assembleias de peixes de riachos na bacia do Rio Ivinhema, Alto Rio Paran Ana Paula Lemke1,2and Yzel Rondon Sarez2

1Programa de Ps-graduao em Recursos Naturais, Universidade Estadual de Mato Grosso do Sul UECEP 79804-970, Dourados, MS, Brasile-mail: [email protected]

2Centro Integrado de Anlise e Monitoramento Ambiental, Laboratrio de Ecologia, Universidade Estadde Mato Grosso do Sul UEMS, Rod. Dourados, Itahum km 12, CEP 79804-970, Dourados, MS, Brasi

e-mail: [email protected]

Abstract: Objectives: Among the most relevant issues in community ecology is the

inuence of different ways of addressing the pattern of diversity and distribution of species,and based on this, the present study sought to evaluate the inuence of landscape andlocal characteristics on the richness and composition of sh assemblages in the IvinhemaRiver basin.Method: In the present study, we used data of distribution of sh speciesin 25 stretches of streams of the Ivinhema River basin, Upper Paran River, aiming toknow the role of local characteristics and land use on sh assemblages.Results: Weidentied 113 sh species, and those sampled in a greater number of sites were: AstyanaxaltiparanaeandSerrapinnus notomelas . Te richness varied between 4 and 65 species. Teregression tree allowed explaining 89.3% of richness variation, and permitted to identifythat altitude is the main predictor of richness. According to the canonical correspondenceanalysis, variables that most inuenced the species distribution were: altitude, depth,width, velocity, conductivity and percentage of built-up areas.Conclusions: Our resultssuggested that factors that most inuenced sh assemblages in the Ivinhema River basinwere physiographic and limnological characteristics, followed by land use.

Keywords: species distribution, species richness, sh assemblages, landscape ecology.

Resumo: Objetivos: Entre as questes mais importantes em ecologia de comunidadesest a inuncia das diferentes formas de abordagem sobre o padro de diversidade edistribuio das espcies, com base nisso o presente trabalho buscou avaliar a inunciadas caractersticas locais e de paisagem sobre a riqueza e composio das assembleiasde peixes na bacia do rio Ivinhema.Mtodo: no presente trabalho utilizamos dados dadistribuio das espcies de peixes em vinte e cinco trechos de riachos da bacia do rioIvinhema, Alto Rio Paran, visando compreender o papel de caractersticas locais e deuso do solo sobre as assembleias de peixes;Resultados: Identicamos 113 espcies depeixes, sendo que as amostradas em maior nmero de pontos foram Astyanax altiparanae e Serrapinnus notomelas . A riqueza variou entre 4 e 65 espcies. O resultado da rvorede regresso permitiu explicar 89.3% da variao da riqueza e permitiu identicar quea altitude o principal preditor da riqueza. Conforme a anlise de correspondnciacannica, as variveis que mais inuenciaram a distribuio das espcies foram: altitude,profundidade, largura, velocidade,condutividade e reas edicadas. Concluses: Nossosresultados sugerem que os fatores que mais inuenciam as assembleias de peixes na baciado rio Ivinhema so as caractersticas siogrcas e limnolgicas, seguidas pelo uso do solo.

Palavras-chave: distribuio de espcies, riqueza de espcies, assembleias de peixes,ecologia da paisagem.


2/12

452 Lemke, AP. and Sarez, YR. Acta Limnologica Brasiliensia

variation (Pavanelli and Caramaschi, 2003). Teinuence of physiographic characteristics of thewatershed was commented by Sarez et al. (2007)and Sarez and Lima Junior (2009), the inuenceof the riparian vegetation is mentioned by Cetra andPetrere Junior (2006), the effects from urbanization

on the diversity and distribution of species wasdetected by Felipe and Sarez (2010) in streams ofthe Ivinhema River basin.

Aquatic ecosystems are the most affected byhuman interferences( ejerina-Garro et al., 2005).Te overexploitation of the soil, pollution, changesin the ow, destruction of natural habitats arethe main threats to biodiversity(Dudgeon et al.,2006). Ward and Stanford (1989) affirm thatstream ecosystems should be studied and analyzed

in four dimensions: longitudinal, lateral, verticaland temporal, and laterally emphasizing mainly theland use. Tus, the present study aimed to evaluatethe inuence of landscape and local characteristicson the composition of fish community onthe Ivinhema River basin, by responding thefollowing questions: 1) How do landscape and localcharacteristics interfere with species richness in thesampled streams? 2) Which buffer range: 1, 5 or10km best explain the distribution of sh species

in the Ivinhema River basin? 3) What variablesbest explain the distribution of sh species in thesampled streams in the Ivinhema River basin?

2. Material and Methods

2.1. Study area

he Ivinhema River basin is located in thesouth-central portion of the Mato Grosso do SulState, in the Paran River basin, between latitudes

21 and 23 S and longitudes 53 30 and 56 W(Mato Grosso do Sul, 1990). It has an area of45.000 km and is formed by the rivers Dourados,Brilhante, Vacaria and Santa Maria. Its headwatersare located approximately 700 meters above sealevel and its mouth is on the Paran River, around218 meters altitude. 2.2. Sampling

wenty five sampling sites were established

across the Ivinhema River basin, includingheadwaters, middle course and mouth of the basin,as well as the major forms of land use (Figure 1).

In this study, we used a data base of sh speciesoccurrence between 2001 and 2013 throughoutthe Ivinhema River basin. At least six samplingwere taken along the year in all sampling sites, in

1. IntroductionTere are around 28.900 sh species in the world,

11.000 are exclusive to freshwater(Lvque et al.,2008).About 4,500 freshwater species weredescribed only in the Neotropics, a number thathas increased every year(Nelson, 2006).

Fish communities are inuenced by differentfactors such as limnological variables ( undisi and

undisi, 2008; Melo et al., 2009), biotic variables(Winemiller et al., 2008; Araujo and ejerina-Garro, 2009), hydrological factors (Poff, 1997),with the relative importance of each one of thesesources changing according to the environment andscale analyzed.

It can be highlighted the studies of Feyrer andHealey (2003), which determined the importanceof environmental variables on the sh fauna of theSouth Delta in California, and of Hanchet (2012)who examined the effect of four different land useson the distribution of native sh in the WaikatoRiver, New Zealand. Although Ward and Stanford(1989), Fitzgerald et al. (1998), Lammert and Allan(1999) have already sought to relate land use to thesh fauna in micro basins, studies on this subject inneotropical regions are scarce.

In the Neotropics, studies relating the inuenceof different forms of land use on sh are scarce,however there are studies presenting the inuencefrom urbanization (Paul and Meyer, 2001; Araujo and ejerina-Garro, 2009; Furlan et al.,2012),agriculture and livestock farming(Casatti et al., 2006; Ferreira and Casatti, 2006)and benets of riparian areas (Cetra and Petrere Junior, 2006; eresa and Casatti, 2010) on the shcommunity. Gerhard (2005) in the CorumbataRiver basin examined the relationship betweenland use: native forest, pasture, canebrake, and thediversity of sh species.

Te Paran River basin is the second largestbasin in South America and the fourth largest inthe world, and is an example of highly impactedriver, with great changes in its bed and in shfauna. Te Upper Paran River system includes allthe catchment of the Paran River upstream of theSete Quedas Waterfalls, currently dammed for theconstruction of the Itaipu hydroelectric power plant

(Agostinho and Jlio Junior, 1999; Langeani et al.,2007).Te sh community of the Upper Paran River

has received increased attention in recent years,with studies reporting the species composition(Castro et al., 2003; 2004; 2005; Langeani et al.,2007), and the influence of the longitudinal


3/12

2013, vol. 25, no. 4, p. 451-462 453Inuence of local and landscape characteristics

the Georeferenced Information Processing System(SPRING) (INPE, 2010) developed by the NationalInstitute for Space Research (INPE). It was createdbuffers with 1, 5 and 10 km radius within whichit was quantied the main forms of land use in thebasin: agriculture and livestock farming, built-uparea, forest fragments and wetlands.

A supervised classication of land use was donethrough the Bhattacharya algorithm (Moreira,2005), for developing the methodology are selectedareas with known types of land use and then thealgorithm recognizes similar areas based on spectralproperties of the image. 2.3. Data analysis

A possible spatial pattern in richness andcomposition of sh assemblages was analyzed bythe Morans I test for species richness, and forspecies occurrence we used a Mantel correlogram(Legendre and Fortin, 1989; Legendre, 1993). Te

Morans I autocorrelation index and its signicancewas calculated in the software SAM (Spatial Analysiin Macroecology) developed by Rangel et al.(2010), whereas the Mantel correlogram for speciescomposition was obtained by using the functionmgram in the software ecodist using the platformR (R Development Core eam, 2011).

order to include the inuence of seasonal variationon the composition of sh species. For samplings,it was used rectangular sieves (1.2 0.8m) with2mm mesh size, electroshing, seining nets withdifferent mesh sizes, and gill nets that were installedin the afternoon of one day and withdrawn at noonof the next day. At each location, samplings wereperformed in stretches of approximately 100mlength.

Collected sh were xed in 10% formaldehydeand after 72 hours they were transferred to 70%alcohol. Te identication was based on Graaand Pavanelli (2007) and consultations to expertswhenever necessary.

At each location the following environmentalvariables were measured: electrical conductivity(S/cm1), using a conductivimeter (Delta OHMConductivity Meter HD 8706-R2), depth (m)and width (m) were obtained at ve points in eachstretch sampled (100m), using a graduated wooden

rod and water velocity (m/s1

) was taken by a digitalow meter (Global Water FP101). At least veestimates were taken at approximately equidistantlocations, which were later used to describe theaverage of the site.

For the analysis of land use we used LANDSA5 imagery, from 2010; data were processed through

Figure 1. Sampling sites in the Ivinhema River basin, Upper Paran River from 2001 to 2013.


4/12


Te Mantel correlogram and Morans I spatialautocorrelation evidenced no signicant spatialautocorrelation for species richness and for speciescomposition only nal distance classes have spatialautocorrelation statistically different from zero(Figure 3).

Te rst partition of variance pointed that thebuffer with 1km radius is the most indicated toquantify the importance of the landscape on thesh distribution. Te 1km radius explained 21.84%variance of data obtained, while radii of 5 and 10km explained 19.26% and 15.42%, respectively(Figure 4a). Considering all buffers in the rangeof 10km, the main land use was agriculture andlivestock with 75.6% total area of buffers, followedby forest remnants (17.81%), wetlands (3.57%) and

built-up area (2.92%).Based on the second partition of variance, dataof sh species distribution were primarily inuencedby limnological and physiographic variables (r2=29.69%), followed by land use (r= 21.87%) andlocation of the sampling site in the basin (r2=15.83%) (Figure 4b).

Te regression tree identied the altitude as themain predictor of species richness, followed by theforest fragments, agriculture and livestock farming,

and depth of the sampled stretches. Te results alsoshowed that 89.3% of variation of species richnesswas explained by the selected variables. In this way,stretches with altitude higher than 338 m and area ofagriculture and livestock farming had lower speciesrichness (mean= 14.7). While places with altitudesbelow 338 m present mean richness 33.9 species.

wo partitions of variance were performed; therst to check which buffer range (1 km, 5 km or10 km) most inuenced the variation of the data,the second to check which data set 1) limnologicalparameters and physiographic characteristics of thebasin, 2) land use, or 3) the position of the sampling

site in the watershed that best explain the variationof species occurrence (Borcard et al., 1992).In order to analyze the inuence of environmental

variables on the diversity, represented in this studyby richness, a regression tree was carried out, usingrichness in response to environmental variablesand classes of land use as explanatory variables. According to Death and Fabricius (2000), theregression tree is a variance partitioning method,in which observations are sequentially divided into

more and more homogeneous groups according tothe explanatory variables provided.o verify the individual influence of

environmental variables on species occurring in morethan 10% samples (at least three sites) a CanonicalCorrespondence Analysis (CCA) was applied. Teindividual importance of environmental descriptorswas quantied through an envt routine, whichuses a randomization process (999 permutations)to dene the signicance of environmental variables

and landscape metrics for obtaining the value ofr2. All statistical analyses were run by means of thepackage vegan in the platform R (R DevelopmentCore eam, 2011).

3. ResultsIn total, 113 species were sampled along the

stretches sampled in the Ivinhema River basin.Species registered in a greater number of stretcheswere As tyanax al tiparanae (84% of sampled

stretches) andSerrapinnus notomelas (80% ofsampled stretches). welve species occurred in morethan 50% stretches of streams sampled, while 24species occurred in only one stretch. In this way,most species occurred at low frequency in thestudied stretches (Figure 2).

Te species richness in the stream stretchesvaried from 4 to 65 species (mean=22.8; sd=16.09).Sites with higher richness (65 and 54 species) arelocated in the lower portion of the basin. Te site

with the lowest richness (four species) is located inthe upper portion of the basin, near the headwaters.Te electrical conductivity ranged from 12.3

to 401.6S/cm1, the width of streams varied from0.4 to 52 m, the depth from 0.3 to 3,5 m, thewater velocity between 0.16 and 1.38m/s1 and thealtitude between 244 and 680 m.

Figure 2. Frequency of occurrence of sh species sampledin streams of the Ivinhema River basin, Upper ParanRiver, from 2001 to 2013.


5/12


Te results of CCA also suggested that the groupmade up ofParodon nasus , Hypostomus strigaticeps , Astyanax fasciatus mainly occurs in locations withhigher altitude, degree of urbanization and electricalconductivity, complementarilyP. reticulata occurisolated of others species. InverselyS. marginatus and L. lacustris occur mainly in lower altitudesites and with small values of conductivity andurbanization rate. On the other hand, a set ofspecies includingRoeboides paranensis ,Pselogrammuskennedyi , Salminus brasiliensis, Prochilodus lineatus ,

Places with forest fragmentation highest 9% of thebuffer area present a mean richness of 25.5 species(Figure 5).

Te canonical correspondence analysis suggestedthat 41.88% of data variation was explained by therst two CCA axes, the rst axis explained 26.46%,and the second15.42%. he most importantenvironmental descriptors to determine the speciesoccurrence in the Ivinhema River basin were:altitude, depth and width, followed by conductivityand % of built-up areas ( able 1 and Figure 6).

Figure 3. Spatial autocorrelation for species richness (Mantel correlogram) and composition (Morans I) Ivinhema River basin, Upper Paran River, from 2001 to 2013.

Figure 4. A) Venn diagram with the variation percentage explaining the different buffer radii (1, 5 and 1in the partition of variance.B) Venn diagram with the variation percentage explaining the composition of

community based on the partition of variance based on: limnological and physiographic variables (L), geoposition (C), land use (U), fraction explained by environmental and land use characteristics (L:U), fraction eby environmental variables and location of the site (L:C), fraction explained by land use metrics and geoposition (U:C), fraction explained by environmental factors, land use, and geographical position (L:U:C).


6/12


Te most complete survey of the sh fauna ofthe Upper Paran River basin listed 310 sh species(Langeani et al., 2007), in addition to other 50

species under description totaling 360 species, inthis way, the present study found 28.9% of thespecies richness of the Upper Paran River.

Te richness registered in the present study washigher than that recorded by Sarez et al. (2011),who recorded 111 species along 200 samplingstretches in the Ivinhema River basin. Te streams

Serrasalmus maculatus among others, occurs indeeper and wider sites.

4. DiscussionBrazil is home to about 43% of the known

freshwater sh fauna (Buckup et al., 2007), but theknowledge on the composition of sh assemblagesin different Brazilian river basins is still very weakand irregular(Silva Filho et al., 2011).

Figure 5. Regression tree of sh species richness in streams of the Ivinhema River basin, Upper Paran Rive2001 to 2013.

Table 1. Results of the Canonical Correspondence Analysis (CCA) for sh communities of the Ivinhema RUpper Paran River. *= Signicant at = 0.05; ***=Signicant at = 0.001.

Environmental Characteristics Axis1 Axis2 r

Agriculture and livestock farming (%) 0.10 0.99 0.10Forest Fragments (%) 0.34 0.94 0.05Wetlands (%) 0.99 0.13 0.01Built-up areas (%) 0.81 0.57 0.20Stream average width 0.91 0.40 0.73***Stream average depth 0.95 0.30 0.66***

Average water velocity 0.74 0.66 0.16 Altitude 0.68 0.72 0.50***

Electrical Conductivity 0.22 0.97 0.41***Summary statistics of axes 1 and 2 Explanation Percentage (%) 26.46% 15.42%


7/12


the most common species for the Ivinhema Riverbasin are also registered as common in other sub-basins of the Upper Paran River basin, we believethat the widespread distribution of these species isa result of their dispersal ability. Cottenie (2005)analyzed the relative inuence of environmentalcharacteristics vs. spatial processes and observedthat neutral processes were considered the onlyresponsible for structuring only 8% of the naturalcommunities analyzed, but by disregarding thedispersal process according to the neutral theory,37% of communities would not have an importantpattern of organization. Tereby, probably the greatdispersal ability of sh species, associated withother ecological characteristics of them and/or ofenvironments can interact and dene the structureof local communities.

Despite the low number of sampling siteswhen compared to Sarez et al., (2011) also in

Ivinhema River Basin, we observed the variationof the longitudinal gradient in the species richness.Nevertheless, these results were partially maskedby other environmental variables. Wiens (1984)argues that communities are inuenced by a spatialgradient, among few studies on large scale of shdistributions Bistoni and Hued (2002) was detected

sampled in this study are mostly the same sampledin the study of Sarez et al. (2011), with only sixexceptions, besides the inclusion of the most currentdata, which allowed complementing the species listthrough the record of large sized species, typicalof rivers with larger volumes (eg. Auchenipterusucayalensis , Clarias gariepinnus , Hemiodus orthonops andSorubim lima ).

Te predominance of A. altiparanae and S.notomelas is consistent with previous studies

conducted in the Ivinhema River basin (Pavanelliand Caramaschi, 2003, Costa-Pereira et al., 2012)and is the result of the great dispersal ability andgeneralist habit of these species (Orsi et al., 2004;Gomiero and Braga, 2008) which have a widedistribution in the Upper Paran River basin. Onthe other hand, the low number of species withwidespread occurrence in the basin, with 10.1% ofspecies occurring in more than 50% of the sampledsites suggests that few species have ecological

characteristics that allow an extensive occupationof the Ivinhema River basin. According to the neutral theory of biodiversity

(Hubbell, 2001) the species composition at regionallevel signicantly inuences local communities,since species with wider distribution should bethose with higher dispersal ability. In this sense, as

Figure 6. Scatter plot of distribution of sh species in the Ivinhema River basin, Upper Paran River, fromto 2013.


8/12


the importance of different spatial scales on theorganization of sh communities. Although ourstudy did not present a distribution of samplingsites that allow a biogeographical approach, theobservation that local characteristics are moreimportant than landscape characteristics and the

mere geographic location allows directing futurestudies on sh communities in neotropical streams.On the other hand, the inclusion of the altitude asa local characteristic interfere with our conclusions,since it is known that the longitudinal variation isextremely important for sh distribution in thewatershed scale (Vannote et al., 1980).

In the Ivinhema River basin are present totallyor partially 25 municipalities and about 26% ofthe state population, in this way the basin is under

anthropogenic pressures that cause changes in waterquality, which in turn impact sh assemblages in thewater body. Casatti et al., (2010) studied changesin sh communities along an integrity gradient instreams of the Serra da Bodoquena, and observeddifferences between sh communities in urbanareas, and veried that species such asPoeciliareticulataandCorydoras aeneus , in this studied area,can be considered indicators of urban areas.

In a study developed in urban streams of the

city of Dourados, Ivinhema River basin, Felipe andSarez (2010) observed that sites with higher levelof urbanization presented predominance of Astyanaxaltiparanae , Poecilia reticulataandCorydoras aeneus ,species with higher tolerance to environmentaldegradation, but the scale considered in this studywas smaller and totally inside the urban area, wherelocal characteristics (eg. sewage disposal) still havea great importance. Another research performed byCunico et al. (2012) examined sh communities

under different levels of human inuence in the cityof Maring (Paran State) and concluded that theurbanization had interfered with species richnessand composition, with dominance ofPoeciliareticulata in the streams.

One of the main problems of the growingurbanization surrounding rivers is the increaseddischarge of effluents into rivers, leading toan increase in the electrical conductivity ofwater(Metcalf and Eddy, 1991). Studies undertaken

in streams of the Ivinhema River basin under lowhuman inuence have shown that the electricalconductivity varies between 4 and 50 S.cm1 inheadwaters stretches and up to 100 S.cm1 in lowerstretches of the basin (Valrio et al., 2007; Sarezand Lima Junior, 2009), considering relativelyintact streams, however streams situated in urban

altitudinal variation richness on the gradient inrivers of Argentina.

Te relationship of richness with volume ofstreams was expected, drawing an analogy withthe island biogeography theory, once the volumeof streams is a measure equivalent to the area of

islands (MacArthur and Wilson, 1967; Gorman andKarr, 1978) and has been detected in other studieson stream sh. Sarez et al. (2007) detected thathabitat characteristics along with the separationof microbasins and the colonization dynamics areinvolved in the determination of sh assemblages.Couto and Aquino (2011) characterized thespatial and seasonal distribution of streams insidea conservation unit in the Federal District, Brazil,and concluded that the sh assemblage was strongly

affected by the habitat complexity. In this context,habitats in the lower portion of the basin, withhigher volume, present a higher complexity ofhabitats than those located in the headwaters,where the high velocity and low volume simplifythe habitats.

In a complementary way, the river continuumconcept (Vannote et al., 1980) suggests thatcharacteristics as diversity and productivity changesalong the water course, causing the distribution of

organisms to be determined according to physicalcharacteristics of the stream, which is expectedaccording to its position in the drainage network.Sites close to the headwaters are shallow and havea higher amount of riparian vegetation, favoringthus the presence of individuals whose existencerequires allochthonous foods. As the volume of thewater body increases, also increases the incidenceof solar radiation favoring the primary production,increasing the diversity of foods, and thus a higher

diversity of sh.Values found for the spatial autocorrelationindicate no spatial autocorrelation for speciesrichness and signicant autocorrelation for speciescomposition only in final distance gradient,which was also observed by Cunico et al. (2012)who studied the effects of local and regionalenvironmental factors on the structure of fishassemblage in the Pirap River basin, Brazil, andfound no correlation between the matrix of distance

of sampled sites and collection periods, but, thisconclusion was inuenced by the smaller spatial andtemporal scale analyzed by these authors.

Although the concept of the local compositionof species considers it as the result of factors at thelocal, regional and biogeographically scales (Gotelliand Graves, 1996), few studies have examined


9/12


anthropogenic perturbations on stream fishassemblages, Upper Paran River, Central Brazil.Neotropical Ichthyology , vol. 7, no. 1, p. 31-38.

BIS ONI, MA. and HUED, AC. 2002. Patterns of shspecies in rivers of the central region of Argentina.Brazilian Journal of Biology , vol. 62, no. 4b,p. 753-764.http://dx.doi.org/10.1590/S1519-69842002000500004

BORCARD, D., LEGEDRE, P. and DRAPEAU, P.1992. Partialling out the spatial component ofecological variation.Ecology , vol. 71, p. 1045-1055.http://dx.doi.org/10.2307/1940179

BUCKUP, PA., MENEZES, NA. and GHAZZI, MS.2007.Catlogo das espcies de peixes de gua doce doBrasil . Rio de Janeiro: MuseuNacional. 195 p.

CASA I, L., LANGEANI, F., SILVA, AM. andCAS RO, RMC. 2006. Stream sh, water and habitat

quality in a pasture dominated basin, southeasternBrazil.Brazilian Journal of Biology , vol. 66, no. 2B,p. 681-696.http://dx.doi.org/10.1590/S1519-69842006000400012

CASA I, L., ROMERO, RM., ERESA, FB.,SABINO, J. and LANGEANI, F. 2010. Fishcommunity structure along a conservation gradientin Bodoquena Plateau streams, central West of Brazil. Acta Limnologica Brasiliensia , vol. 22, no. 1, p. 50-59.http://dx.doi.org/10.4322/actalb.02201007

CAS RO, RMC.,CASA I, L, SAN OS, HF.,FERREIRA, KM., RIBEIRO, AC., BENINE, RC.,DARDIS, GZP., MELO, ALA., ABREU, X.,BOCKMANN, FA., CARVALHO, M., GIBRAN,FZ. and LIMA, FC . 2003. Estrutura e composioda ictiofauna de riachos do Rio Paranapanema,sudeste e sul do Brasil.Biota Neotropica , vol. 3, no. 1,p. 01-31.

CAS RO, RMC., CASA I, L., SAN OS, HF.,MELO, ALA., MAR INS, LSF., FERREIRA, KM.,GIBRAN, FZ., BENINE, RC., CARVALHO, M.,RIBEIRO, AC., ABREU, X., BOCKMANN, FA.,DARDIS, GZP., S OPIGLIA, R. and LANGEANI,F. 2004. Estrutura e composio da ictiofauna deriachos da bacia do Rio Grande, no Estado de SoPaulo, Sudeste do Brasil.Biota Neotropica , vol. 4,no. 1, p. 01-39.

CAS RO, RMC., CASA I, L., SAN OS, HF., VARI,RP., MELO, ALA., MAR INS, LDF., ABREU, X.,BENINE, RC., GIBRAN, FZ., RIBEIRO, AC.,BOCKMANN, FA., CARVALHO, M., PELIO,GZP., FERREIRA, KM., S OPIGLIA, R. and AKAMA, A. 2005. Structure and composition ofthe stream ichthyofauna of four tributary rivers ofthe upper Rio Paran basin, Brazil.IchthyologicalExploration of Freshwaters , vol. 16, no. 3, p. 193-214.

CE RA, M. and PE RERE JUNIOR, M. 2006. Fish- Assemblage Structure of the Corumbatai Riverbasin, So Paulo State, Brazil: characterization andanthropogenic disturbances.Brazilian Journal of

areas present conductivity of approximately 700S.cm1and lower values of species richness, as wellas the predominance ofC. aeneusandP. reticulata ,conrming previous studies.

With the increase in urban areas there isan increase in the runoff due to the increase in

impervious surfaces ( ucci, 2003). Given the needfor large areas for agriculture and livestock farming,the deforestation of areas with native vegetationbecomes indispensable. In this way, Roth et al.(1996) detected that streams with little agricultureand greater amount of wetlands in their vicinity,also present a higher diversity of habitats. Likewise, Wang et al. (2001) states that the urbanization has anegative effect on the sh fauna due to soil sealing.

Te altitude can be determinant by summarizing

some limnological characteristics such as: watertemperature, ow and reproductive rate ( ondatoand Sarez, 2010). Many species have limitationas for high altitudes (Matthews, 1998),or bynding factors limiting its dispersal to higher sites(Valrio et al., 2007).

5. ConclusionsTen, our conclusions are that the buffer with

1km radius is the most indicated to quantify the

importance of the landscape on the stream shdistribution. Te sh species diversity are mainlyexplained by a combination of limnological/physiographic and landscape characteristics(altitude, % by the forest fragments, % agricultureand livestock farming, and stream depth). Andfinally, the most important environmentaldescriptors to determine the species occurrence inthe Ivinhema River basin were: altitude, depth andwidth, followed by conductivity and % of built-up

areas. Acknowledgements

o CNPq and Fundect for nancial support. oCAPES for the scholarship to the rst author. oDleth Fernanda Silva Santos, Gabriel Nakamurade Souza and Marcelo Maldonado de Souza forhelp in the eld work. Y. R. Sarez is supported byproductivity grants from CNPq.

References AGOS INHO, AA. and JLIO JUNIOR, HF. 1999.

Peixes da Bacia do Alto Rio Paran. In Lowe-McConnel, R. H.Estudos de comunidades de peixestropicais . So Paulo: EdUSP. p. 374-399.

ARAUJO, NB. and EJERINA-GARRO, FL.2009. Influence of environmental variables and
http://dx.doi.org/10.1590/S1519-69842002000500004http://dx.doi.org/10.1590/S1519-69842002000500004http://dx.doi.org/10.2307/1940179http://dx.doi.org/10.1590/S1519-69842006000400012http://dx.doi.org/10.1590/S1519-69842006000400012http://dx.doi.org/10.4322/actalb.02201007http://dx.doi.org/10.4322/actalb.02201007http://dx.doi.org/10.1590/S1519-69842006000400012http://dx.doi.org/10.1590/S1519-69842006000400012http://dx.doi.org/10.2307/1940179http://dx.doi.org/10.1590/S1519-69842002000500004http://dx.doi.org/10.1590/S1519-69842002000500004


10/12


FURLAN, N., ES EVES, KE. and QUINGLIA, GA.2012. Environmental factors associated with shdistribution in an urban neotropical river (Upper

iet River Basin, So Paulo, Brazil).EnvironmentalBiology of Fishes , vol. 96, no. 1, p. 77-92.http://dx.doi.org/10.1007/s10641-012-0024-3

GERHARD, P. 2005.Comunidades de peixes de riachosem funo da paisagem da bacia do rio Corumbata,estado de So Paulo. Piracicaba: Escola Superior de Agricultura Luiz de Queiroz, Universidade de SoPaulo. 241 p. [ ese de Doutorado].

GOMIERO, LM. and BRAGA, FMS. 2008. Feedinghabits of the ichthyofauna in a protected area inthe state of So Paulo, southeastern Brazil.BiotaNeotropica . vol. 8, no. 1, p. 41-47.http://dx.doi.org/10.1590/S1676-06032008000100004

GORMAN, O . and KARR, JR. 1978. Habitat structureand stream sh communities.Ecology , vol. 59, no. 3,p. 507-515.http://dx.doi.org/10.2307/1936581

GO ELLI, NJ. and GRAVES, GR. 1996. Null Modelsin Ecology. Washington: Smithsonian InstitutionPress. 386 p.

GRAA, WJ. and PAVANELLI, CS. 2007.Peixesda plancie de inundao do alto rio Paran e reasadjacentes . Maring: EdUEM. 241 p.

HANCHE , SM. 2012. Effect of land use on thedistribution and abundance of native fish intributaries of the Waikato River in the HakarimataRange, North Island, New Zealand.New Zealand Journal of Marine and Freshwater Research, vol. 2p. 159-171.

HUBBELL, SP. 2001.Te Unied Neutral Teory ofBiodiversity and Biogeography . Princenton UniversityPress: Princeton. 396 p.

Instituto Nacional de Pesquisas Espaciais. 2010.SPRING - Sistema de Processamento de InformaesGeorreferenciadas . Available from: . Access in:15 Feb. 2013.

LAMMER , M. and ALLAN, JD. 1999. Assessing bioticintegrity of streams: Effects of scale in measuringthe inuence of land use/cover on habitat structureon fish and macroinvertebrates.Environmental Management , vol. 23, no. 2, p. 257-270.http://dx.doi.org/10.1007/s002679900184

LANGEANI, F., CAS RO, RMC., OYAKAWA, O .,SHIBA A, OA., PAVANELLI, CS. and CASA I,L. 2007. Diversidade da ictiofauna do Alto RioParan: composio atual e perspectivas futuras.Biota Neotropica , vol. 7, no. 3, p. 1-17.http://dx.doi.org/10.1590/S1676-06032007000300020

LEGENDRE, P. 1993. Spatial autocorrelation: troubleor a new paradigm?Ecology . vol. 74, no. 6, p. 1659-1673.http://dx.doi.org/10.2307/1939924

Biology , vol. 66, no. 2A, p. 431-439.http://dx.doi.org/10.1590/S1519-69842006000300007

COS A-PEREIRA, R., ROSA, FR. and RESENDE, EK.2012.Estrutura rca da comunidade de peixes deriachos da poro Oeste da Bacia do Alto Paran . Dadoseletrnicos. Corumb: EMBRAPA Pantanal. 25 p.

CO ENIE, K. 2005. Integrating environmental andspatial processes in ecological community dynamics.Ecology Letters , vol. 8, p. 1175-1182.http://dx.doi.org/10.1111/j.1461-0248.2005.00820.x

COU O, BA. and AQUINO, PPU. 2011. Structureand integrity of sh assemblages in streams associatedto conservation units in Central Brazil.NeotropicalIchthyology , vol. 9, no. 2, p. 445-454.http://dx.doi.org/10.1590/S1679-62252011000200023

CUNICO, AM., FERREIRA, EA., AGOS INHO, AA., BEAUMORD, AC. and FERNANDES, R.2012. Te effects of local and regional environmentalfactors on the structure of sh assemblages in thePirap Basin, Southern Brazil.Landscape and UrbanPlanning , vol. 105, p. 336-344.http://dx.doi.org/10.1016/j.landurbplan.2012.01.002

DEA H, G. and FABRICIUS, KE. 2000. Classicationand regression trees: a powerful yet simple techniquefor ecological data analysis.Ecology , vol. 81, no. 11,p. 3178-3192.http://dx.doi.org/10.1890/0012-9658(2000)081[3178:CAR AP]2.0.CO;2

DUDGEON, D., AR HING ON, AH., GESSNER,MO., KAWABA A, ZI., KNOWLER, DJ.,LVQUE , C., NAIMAN, RJ., PRIEUR-RICHARD, A-H., SO O, D., S IASSNY, MLJ.and SULLIVAN, CA. 2006. Freshwater biodiversity:importance, threats, status and conservationchallenges.Biological Reviews , vol. 81, p. 163-182.http://dx.doi.org/10.1017/S1464793105006950

FELIPE, RA. and SAREZ, YR. 2010. Caracterizaoe inuncia dos fatores ambientais nas assembleiasde peixes de riachos em duas microbacias urbanas,

Alto Rio Paran.Biota Neotropica . vol. 10, no. 2,p. 143-151.FERREIRA, CP. and CASA I, L. 2006. Inuncia

da estrutura do hbitat sobre a ictiofauna de umriacho em uma micro-bacia de pastagem, So Paulo,Brasil.Revista Brasileira de Zoologia , vol. 23, no. 3,p. 642-651.http://dx.doi.org/10.1590/S0101-81752006000300006

FI ZGERALD, DG., KO , E., LANNO, RP. andDIXON, DG. 1998. A quarter century of changes inthe sh assemblages of three small streams modiedby anthropogenic activities. Journal of AquaticEcosystem Stressand Recovery, vol. 6, p. 111-127.

FEYRER, F. and HEALEY, MP. 2003. Fish communitystructure and environmental correlates in the highlyaltered southern Sacramento-San Joaquin Delta.Environmental Biology of Fishes , vol. 66, p. 123-132.http://dx.doi.org/10.1023/A:1023670404997
http://dx.doi.org/10.1007/s10641-012-0024-3http://dx.doi.org/10.1007/s10641-012-0024-3http://dx.doi.org/10.1590/S1676-06032008000100004http://dx.doi.org/10.1590/S1676-06032008000100004http://dx.doi.org/10.2307/1936581http://www.dpi.inpe.br/spring/portugues/index.htmlhttp://www.dpi.inpe.br/spring/portugues/index.htmlhttp://dx.doi.org/10.1007/s002679900184http://dx.doi.org/10.1007/s002679900184http://dx.doi.org/10.1590/S1676-06032007000300020http://dx.doi.org/10.1590/S1676-06032007000300020http://dx.doi.org/10.2307/1939924http://dx.doi.org/10.1590/S1519-69842006000300007http://dx.doi.org/10.1590/S1519-69842006000300007http://dx.doi.org/10.1111/j.1461-0248.2005.00820.xhttp://dx.doi.org/10.1111/j.1461-0248.2005.00820.xhttp://dx.doi.org/10.1590/S1679-62252011000200023http://dx.doi.org/10.1590/S1679-62252011000200023http://dx.doi.org/10.1016/j.landurbplan.2012.01.002http://dx.doi.org/10.1016/j.landurbplan.2012.01.002http://dx.doi.org/10.1890/0012-9658(2000)081[3178:CARTAP]2.0.CO;2http://dx.doi.org/10.1890/0012-9658(2000)081[3178:CARTAP]2.0.CO;2http://dx.doi.org/10.1017/S1464793105006950http://dx.doi.org/10.1590/S0101-81752006000300006http://dx.doi.org/10.1590/S0101-81752006000300006http://dx.doi.org/10.1023/A:1023670404997http://dx.doi.org/10.1023/A:1023670404997http://dx.doi.org/10.1590/S0101-81752006000300006http://dx.doi.org/10.1590/S0101-81752006000300006http://dx.doi.org/10.1017/S1464793105006950http://dx.doi.org/10.1890/0012-9658(2000)081[3178:CARTAP]2.0.CO;2http://dx.doi.org/10.1890/0012-9658(2000)081[3178:CARTAP]2.0.CO;2http://dx.doi.org/10.1016/j.landurbplan.2012.01.002http://dx.doi.org/10.1016/j.landurbplan.2012.01.002http://dx.doi.org/10.1590/S1679-62252011000200023http://dx.doi.org/10.1590/S1679-62252011000200023http://dx.doi.org/10.1111/j.1461-0248.2005.00820.xhttp://dx.doi.org/10.1111/j.1461-0248.2005.00820.xhttp://dx.doi.org/10.1590/S1519-69842006000300007http://dx.doi.org/10.1590/S1519-69842006000300007http://dx.doi.org/10.2307/1939924http://dx.doi.org/10.1590/S1676-06032007000300020http://dx.doi.org/10.1590/S1676-06032007000300020http://dx.doi.org/10.1007/s002679900184http://dx.doi.org/10.1007/s002679900184http://www.dpi.inpe.br/spring/portugues/index.htmlhttp://www.dpi.inpe.br/spring/portugues/index.htmlhttp://dx.doi.org/10.2307/1936581http://dx.doi.org/10.1590/S1676-06032008000100004http://dx.doi.org/10.1590/S1676-06032008000100004http://dx.doi.org/10.1007/s10641-012-0024-3http://dx.doi.org/10.1007/s10641-012-0024-3


11/12


RANGEL, F., DINIZ-FILHO, JAF. and BINI,LM. 2010. SAM: a comprehensive applicationfor spatial analysis in macroecology.Ecography ,vol. 33, p. 46-50.http://dx.doi.org/10.1111/j.1600-0587.2009.06299.x

RO H, NE., ALLAN, JD. and ERICKSON, DE. 1996.Landscape influences on stream biotic integrityassessed at multiple spatial scales.Landscape Ecology ,vol. 11, p. 141-156.http://dx.doi.org/10.1007/BF02447513

SILVA FILHO, EG., SAN ANA, FMS. and SEREVI, W.2011. Ictiofauna doreservatrio de Duas Unas, baciado rio Jaboato, Pernambuco: resultados preliminaresda composio e estrutura da assembleia.RevistaBrasileira de Cincias Agrrias , vol. 6, no. 2, p. 351-361.http://dx.doi.org/10.5039/agraria.v6i2a1168

SAREZ, YR., VALRIO, SB., ONDA O, KK.,

XIMENES, LQL. and FELIPE, RA. 2007.Determinantes ambientais da ocorrncia de espciesde peixes em riachos de cabeceira da bacia do rioIvinhema, alto rio Paran. Acta Scientiarum. BiologicalSciences , vol. 29, no. 2, p. 145-150.

SAREZ, YR. and LIMA JUNIOR, SE. 2009. Variaoespacial e temporal nas assemblias de peixes deriachos na bacia do rio Guira, Alto Rio Paran.BiotaNeotropica , vol. 9, no. 1, p. 101-111.

SAREZ, YR., SOUZA, MM., FERREIRA, FS.,PEREIRA, MJ., SILVA, EA., XIMENES, LQL., AZEVEDO, LG., MAR INS, OC. and LIMA- JNIOR, SE. 2011. Patterns of species richnessand composition of fish assemblages in streamsof the Ivinhema River basin, Upper Paran River. Acta Limnologica Brasi liensia , vol. 23, no. 2,p. 177-188.http://dx.doi.org/10.1590/S2179-975X2011000200008

EJERINA-GARRO, FL., MALDONADO, M.,IBEZ, C., PON , D., ROSE , N. andOBERDOFF, . 2005. Effects of natural and

anthropogenic environmental changes on riverine shassemblages: a framework for ecological assessmentof rivers.Brazilian Archives of Biology and echnology ,vol. 48, no. 1, p. 91-108.http://dx.doi.org/10.1590/S1516-89132005000100013

ERESA, FB. and CASA I, L. 2010. Importncia davegetao ripria em regio intensamente desmatadano sudeste do Brasil: um estudo com peixes de riacho.Pan-American Journal of Aquatic Sciences , vol. 5,no. 3, p. 444-453.

ONDA O, KK and SAREZ, YR. 2010. emporalchanges in sh species composition of headwaterstreams of the upper Paraguay and Paran basins,Brazil. Acta Scientiarum. Biological Sciences , vol. 32,no. 3, p. 279-284.

UCCI, CEM. 2003.Impactos da variabilidade climticae do uso do solo nos recursos hdricos . Braslia: AgnciaNacional de guas. p. 150.

LEGENDRE, P. and FOR IN, MJ. 1989. Spatial patternand ecological analysis.Vegetatio, vol. 80, p. 107-138.http://dx.doi.org/10.1007/BF00048036

LVQUE, C., OBERDORFF, ., PAUGY, D.,S IASSNY, MLJ. and EDESCO, PA. 2008.Global diversity of fish (Pisces) in freshwater.Hydrobiologia , vol. 595, no. 1, p. 545-567.http://dx.doi.org/10.1007/s10750-007-9034-0

MacAR HUR, RH. and WILSON, EO. 1967.Tetheory of island Biogeography . Princeton UniversityPress: Princenton. 203 p.

Mato Grosso do Sul. 1990. Atlas multireferencial: Mato Grosso do Sul . Campo Grande: Secretaria dePlanejamento e Coordenao Geral. 27 p.

MA HEWS, WJ. 1998.Patterns in freshwater shecology . New York: Chapman & Hall. 756 p.

MELO, L., EJERINA-GARRO, FL. and MELO,CE. 2009. Inuence of environmental parameterson sh assemblage of a neotropical river with a oodpulse regime, Central Brazil.Neotropical ichthyology ,vol. 7, no. 3, p. 421-428.http://dx.doi.org/10.1590/S1679-62252009000300009

ME CALF & EDDY. 1991.Wastewater Engineering :reatment, Disposal, Reuse. New York: McGraw-

Hill. 1334 p.MOREIRA, MA. 2005.Fundamentos do sensoriamento

remoto e metodologias de aplicao. Viosa: UFV.307 p.

NELSON, JS. 2006.Fishes of the world . New York: John Wiley & Sons. p. 523.

ORSI, ML., CARVALHO, ED. and FORES I, FO.2004. Biologia populacional de AstyanaxaltiparanaeGarutti & Britski ( eleostei, Characidae) do mdioRio Paranapanema, Paran, Brasil.Revista Brasileirade Zoologia , vol. 21, no. 2, p. 207-218.http://dx.doi.org/10.1590/S0101-81752004000200008

PAUL, MJ. and MEYER, JL. 2001. Streams in the urbanlandscape. Annual Review of Ecology and Systematics ,vol. 32, p. 333-65.http://dx.doi.org/10.1146/annurev.ecolsys.32.081501.114040

PAVANELLI, CS. and CARAMASCHI, EP. 2003.emporal and Spatial Distribution of the Ichthyofauna

in wo Streams of the Upper Rio Paran Basin.Brazilian Archives of Biology and echnology , vol. 46,no. 2, p. 271-280.http://dx.doi.org/10.1590/S1516-89132003000200019

POFF, NL. 1997. Landcape lters and species traits:towards mechanistic understanding and predictionin stream ecology. Journal of the North AmericanBenthological Society , vol. 16, no. 2, p. 391-409.http://dx.doi.org/10.2307/1468026

R Development Core eam. 2011.R : A language andenvironment for statistical computing. [2. 12. 2]Vienna: R Foundation for Statistical Computing.
http://dx.doi.org/10.1111/j.1600-0587.2009.06299.xhttp://dx.doi.org/10.1111/j.1600-0587.2009.06299.xhttp://dx.doi.org/10.1007/BF02447513http://dx.doi.org/10.1007/BF02447513http://dx.doi.org/10.5039/agraria.v6i2a1168http://dx.doi.org/10.1590/S2179-975X2011000200008http://dx.doi.org/10.1590/S2179-975X2011000200008http://dx.doi.org/10.1590/S1516-89132005000100013http://dx.doi.org/10.1590/S1516-89132005000100013http://dx.doi.org/10.1007/BF00048036http://dx.doi.org/10.1007/s10750-007-9034-0http://dx.doi.org/10.1007/s10750-007-9034-0http://dx.doi.org/10.1590/S1679-62252009000300009http://dx.doi.org/10.1590/S1679-62252009000300009http://dx.doi.org/10.1590/S0101-81752004000200008http://dx.doi.org/10.1590/S0101-81752004000200008http://dx.doi.org/10.1146/annurev.ecolsys.32.081501.114040http://dx.doi.org/10.1146/annurev.ecolsys.32.081501.114040http://dx.doi.org/10.1590/S1516-89132003000200019http://dx.doi.org/10.1590/S1516-89132003000200019http://dx.doi.org/10.2307/1468026http://dx.doi.org/10.2307/1468026http://dx.doi.org/10.1590/S1516-89132003000200019http://dx.doi.org/10.1590/S1516-89132003000200019http://dx.doi.org/10.1146/annurev.ecolsys.32.081501.114040http://dx.doi.org/10.1146/annurev.ecolsys.32.081501.114040http://dx.doi.org/10.1590/S0101-81752004000200008http://dx.doi.org/10.1590/S0101-81752004000200008http://dx.doi.org/10.1590/S1679-62252009000300009http://dx.doi.org/10.1590/S1679-62252009000300009http://dx.doi.org/10.1007/s10750-007-9034-0http://dx.doi.org/10.1007/s10750-007-9034-0http://dx.doi.org/10.1007/BF00048036http://dx.doi.org/10.1590/S1516-89132005000100013http://dx.doi.org/10.1590/S1516-89132005000100013http://dx.doi.org/10.1590/S2179-975X2011000200008http://dx.doi.org/10.1590/S2179-975X2011000200008http://dx.doi.org/10.5039/agraria.v6i2a1168http://dx.doi.org/10.1007/BF02447513http://dx.doi.org/10.1007/BF02447513http://dx.doi.org/10.1111/j.1600-0587.2009.06299.xhttp://dx.doi.org/10.1111/j.1600-0587.2009.06299.x


12/12


multiple spatial scales.Environmental Management ,vol. 28, p. 255-266.http://dx.doi.org/10.1007/s0026702409

WARD, JV. and S ANFORD, JA. 1989. Riverineecosystems: the influence of man on catchmentdynamics and fish ecology.Canadian SpecialPublication of Fisheries and Aquatic Sciences , vol. 106,p. 56-64.

WIENS, JA. 1984. Spatial scaling in ecology.FunctionalEcology , vol. 3, p. 386-397.

WINEMILLER, KO., AGOS INHO, AA. andCARAMASCHI, EP. 2008. Fish ecology in tropicalstreams. In DUDGEON, D., ed.ropical StreamEcology . Elsevier, Academic Press. 370 p.

Received: 28 May 2013 Accepted: 12 February 2014

UNDISI, JG. and UNDISI, M. 2008.Limnologia .So Paulo: Ocina de textos. 632 p.

VANNO E, RL., MINSHALL, GW., CUMMINS,KW., SEDEL, JR. and CUSHING, CE. 1980.Te river continuum concept.Canadian Journal ofFisheries and Aquatic Sciences , vol. 37, p. 130-137.

VALRIO, SB., SAREZ, YR., FELIPE, RA.,ONDA O, KK. and XIMENES, LQL. 2007.

Organization patterns of headwater-stream fishcommunities in the Upper Paraguay-Paran basins.Hydrobiologia , vol. 583, no. 1, p. 241-250.http://dx.doi.org/10.1007/s10750-006-0533-1

WANG, L., LYONS, J. and KANEHL, P. 2001. Impactsof urbanization on stream habitat and sh across
http://dx.doi.org/10.1007/s0026702409http://dx.doi.org/10.1007/s0026702409http://dx.doi.org/10.1007/s10750-006-0533-1http://dx.doi.org/10.1007/s10750-006-0533-1http://dx.doi.org/10.1007/s10750-006-0533-1http://dx.doi.org/10.1007/s10750-006-0533-1http://dx.doi.org/10.1007/s0026702409http://dx.doi.org/10.1007/s0026702409

2013, influence of local and landscape characteristics on the distribution and diversity of fish...

Documents