http://orcid.org/0000-0003-0386-1618
Figures
Abstract
Run-of-the-river hydroelectric dams cause changes in seasonal inundation of the floodplains, and this may cause displacement of semi-aquatic vertebrates present before dam construction. This study evaluated the movement of crocodilians before and after the filling of the Santo Antônio hydroelectric reservoir on the Madeira River in the Brazilian Amazon, which occurred in November 2011. We radio-tracked four adult male Paleosuchus palpebrosus and four adult male Paleosuchus trigonatus before and after the formation of the reservoir between 2011 and 2013. The home ranges of the P. palpebrosus varied from < 1 km2 to 91 km2 and the home ranges of the P. trigonatus varied from < 1km2 to 5 km2. The species responded differently to time since filling and water level in weekly movement and home range. However, overall the dam appears to have had little effect on the use of space by the individuals that were present before dam construction.
Citation: Campos Z, Mourão G, Magnusson WE (2017) The effect of dam construction on the movement of dwarf caimans, Paleosuchus trigonatus and Paleosuchus palpebrosus, in Brazilian Amazonia. PLoS ONE 12(11): e0188508. https://doi.org/10.1371/journal.pone.0188508
Editor: Adam Stow, Macquarie University, AUSTRALIA
Received: June 20, 2017; Accepted: November 8, 2017; Published: November 27, 2017
Copyright: © 2017 Campos et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Data for all surveys are available from Dryad ("Movement of the Paleosuchus palpebrosus and Paleosuchus trigonatus,” 10.5061/dryad.1sj40). Laboratory Protocols are available from protocols.io (https://www.protocols.io/view/the-movement-of-the-paleosuchus-palpebrosus-and-pa-jzkcp4w).
Funding: The study was funded by Santo Antônio Energia. R. Zilca Campos and Guilherme Mourão are employees of Embrapa Pantanal which is a non-profit government company linked to the Ministry of Agriculture of Brazil. The funder provided support in the form of salaries for authors, but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.
Competing interests: The authors have declared that no competing interests exist. We declare that our affiliation with Embrapa Pantanal does not alter our adherence to PLOS ONE policies on sharing data and materials.
Introduction
Historically, traditional hydroelectric dams with large reservoirs flooded extensive areas of terrestrial habitat and changed the conditions from lotic to lentic, which presumably led to the exclusion of many aquatic species present in the area before dam construction [1, 2]. During recent decades, there has been a tendency for the construction of run-of-the-river dams [3, 4], which inundate much smaller areas and maintain the lotic nature of most of the upstream sections.
Presumably, run-of-the-river dams have less drastic effects on the aquatic and semi-aquatic fauna than traditional dams. Nonetheless, even if species assemblages similar to those present before dam construction eventually reestablish, it is not known what happens to the individuals present when the dam fills and inundates local wetlands. Crocodilians are long lived and many species are territorial [5], so dam construction could lead to severe disruption of home ranges, possibly resulting in individuals abandoning the area and dispersing into areas where they experience increased risk of mortality, such as areas near dam turbines [6] or areas with less nesting habitat [7, 8].
Species of Paleosuchus are small crocodilians, the adults of which normally occupy relatively small home ranges [9, 10]. In some places near dam walls, the reservoir may completely destroy all suitable habitat previously occupied by the species [11]. However, most of the river system is not as strongly affected and it is possible that individuals could reestablish in areas close to their former home range
We used radio telemetry to study the movement of individuals Paleosuchus palpebrosus and Paleosuchus trigonatus before the filling of the Santo Antônio Hydroelectric Dam on the Madeira River in Brazilian Amazonia, and in equivalent annual periods one and two years after the filling of the dam. The study was concentrated around the Jaci River, which is a tributary of the Madeira River in the central part of the area inundated by the reservoir. This allowed us to evaluate the effects of water level in the three periods and to determine to what extent dam filling resulted in changes in short-term rates of movement and displacement of home ranges.
Materials and methods
Ethics statement
The research project was approved by the Brazilian Environmental Agency (IBAMA permit N°. 017/02) and by the Chico Mendes Institute for Biodiversity Conservation (ICMBio permanent license N°. 13048–1) for capture and marking caimans (Normative regulation N°. 154/2007). All procedures followed ethical practices for animals approved by Committee on the Ethics of Animal Research of the Brazilian Agricultural Research Organization (Embrapa N° 009/2016). No biological material, such as blood or tissue, was collected in this study. These species are classified by the International Union for Conservation of Nature (IUCN) as Lower Risk, least concern for conservation.
Study site and sampling design
The study was undertaken in the area affected by the Santo Antônio hydroelectric dam (SAHD) on the Madeira River in western Amazonia. The general area and dam characteristics have been described in a previous study [11]. The climate in the region is hot and humid with rainy season from October to April [12]. The water level of the Jaci River before filling ranged from 0.90 m to 8.00 m throughout the year, and after filling ranged from 9.33 to 12.69 m.
The two species of Paleosuchus are small crocodilians that occupy small streams in continuous forest and inundated forests near large rivers in Amazonia [13], and many of these environments close to large rivers are permanently flooded by hydroelectric dams [11]. The two species engage in terrestrial movements in riparian areas and spend extended periods in terrestrial retreats, mainly in dry and cool periods [8, 14, 15].
Between January and August 2011, we captured dwarf caimans from a boat using a steel-cable noose attached to a pole and attached radio transmitters (MOD 400, Telonics) to the tails of four adult male P. trigonatus and four adult male P. palpebrosus. The movements of these individuals were monitored (8°47'00"S, 63°56'00"W) before and after the formation of the reservoir of the Santo Antônio hydroelectric dam. Snout-vent length (SVL cm) of the P. trigonatus ranged from 70.0 to 87.0 cm and body mass from 9.0 to 15.0 kg. For P. palpebrosus, SVL ranged from 69.0 to 82.5 cm and body mass from 9.0 to 18.0 kg (Table 1).
The code for the individual is the same as that used on the figures.
Surveys to locate animals with transmitters were undertaken at weekly intervals throughout the study period by boat and/or on foot. However, to maintain comparability, and take into account possible seasonal variation in space use, we analyzed only data for the calendar months sampled pre-inundation for each animal for both the pre- and post-inundation periods. Although this required discarding data in post-inundation periods, it did not affect the general location or size of home ranges. Data for all surveys are available from the Dryad Digital Repository titled “Data from: Movement of the Paleosuchus palpebrosus and Paleosuchus trigonatus” DOI:doi:10.5061/dryad.1sj40 and Laboratory Protocols ≤ https://www.protocols.io/view/the-movement-of-the-paleosuchus-palpebrosus-and-pa-jzkcp4w.
We divided the sampling period into three phases. Phase 1 was the period of monitoring before dam filling (Feb-Nov/2011); phases 2 and 3 included the same monitoring months in the following years (Phase 2 = Feb-Nov/2012; Phase 3 = Feb-Nov/2013).
Data analysis
All analyses were done in the R statistical language environment, version 2.1.11 [16]. We examined graphically the evolution of the distance from the origin, defined as distance from the capture location of the caiman, through time. We used linear mixed-effects models (lme) adjusted by maximum likelihood (ML) to test whether caiman movement, indexed by the distance moved between consecutive locations, differed among phases. Candidate model predictors were water level (fixed factor), phase nested within individuals (fixed), species (fixed) and individuals (random). We adjust three models with different degrees of complexity (“S1 and S2 Tables”) using the lme function of the nlme package [17]. The first model included as explanatory variables only the overall mean and the nested structure (phase within individuals). The second model included the phase, the interaction between phase and caiman species and the nested structure. The full model included phase, the interaction between phase, caiman species and water level, and the nested structure. We assessed the significance of our models by applying an ANOVA and selected the best model using the Akaike Information Criterion (AIC). Once the best model was selected, we re-ran it using restricted maximum likelihood (REML) to obtain the parameter estimates [18].
Home ranges were estimated for each caiman in each phase by the fixed-kernell 95% technique [19] with the package adehabitatHR [20]. We used the Friedman non-parametric analysis of variance to test for differences in the size of home ranges between phases for each species.
The map of the reservoir was produced in the Argis Desktop 10 program (www.esri.com/argis) by Santo Antônio Energia (SAE). The Spring 5.5.0 program (www.dpi.inpe/spring) was used by Z. Campos and collaborators from Embrapa to plot the locations of caimans. Permission to publish the maps was granted by SAE and Embrapa.
Results
Displacement from the origin through the phases did not appear to differ between caiman species (Fig 1). Some individuals displaced little from the capture local (origin) (Fig 1C and 1D), some moved continually away from the origin between the phases (Fig 1F, 1G and 1H). Others returned repeatedly to the original point of capture (Fig 1I and 1J). However, one P. palpebrosus (Fig 1I) was located one month after the end of the study period about twenty kilometers from the original location.
Distances moved from the point of origin by four Paleosuchus palpebrosus and four Paleosuchus trigonatus at different times during the study. The dam was filled in November 2011.
ANOVA indicated that the full model explained more variability in the response variable than the other competing models, and the full model AIC value was 30 AIC units lower than the previous competing models (Table 2). The best model to explain the distance moved by the caimans indicated that the responses of the caimans to the phases were complex and largely depended of the species (Table 3). For both species, effects of phase 2 differed from the other phases (p < 0.001) and the P. trigonatus moved more than the P. palpebrosus during phase 3 (p = 0.022, “S1 Fig”). The caiman species responded differently to the interaction of phases and water level. P. palpebrosus (Fig 2A, 2B and 2C) tended to increase their distance moved between locations with the increase in water level during phase 1 (p = 0.008) and phase 2 (p > 0.001) and ceased to do so in phase 3 (p = 0.252). P. trigonatus (Fig 2D, 2E and 2F) did not respond to the increase in the water level during phases 1 and 2 (p = 0.530 and p = 0.923, respectively), but moved less between relocations in phase 3 (p = 0.017).
Distances moved between relocations (m) by phases in relation to water level for four male Paleosuchus palpebrosus (A, B, C) and four male Paleosuchus trigonatus (D, E, F).
Home ranges of P. palpebrosus (Fig 3) ranged from < 1 km2 to 90.94 km2 (median = 1.22 km2, IQR = 2.88) and differed between phases (χ2 = 6, df = 2, p = 0.049). The home ranges of the P. trigonatus (Fig 4) ranged from < 1 km2 to 4.89 km2 (median = 0.98 km2, IQR = 1.99) and did not differ significantly between phases (χ2 = 6, df = 2, p = 0.472).
Each horizontal panel represents one individual in different periods related to dam filling. The identifying numbers on each map indicate the individual (before the stroke) and the year of the observations (after the stroke). The map of the reservoir was produced in the Argis Desktop 10 program (www.esri.com/argis) by Santo Antônio Energia (SAE). The figure was created in the Spring 5.5.0 program (www.dpi.inpe/spring).
Each horizontal panel represents one individual in different periods in relation to dam filling. The identifying numbers on each map indicate the individual (before the stroke) and the year of the observations (after the stroke). The map of the reservoir was produced in the Argis Desktop 10 program (www.esri.com/argis) by Santo Antônio Energia (SAE). The figure was created in the Spring 5.5.0 program (www.dpi.inpe/spring).
Discussion
Movement of crocodilians is influenced by water level and seasonal variation in aquatic habitats [21, 22]. Hydroelectric dams modify the water-level regime [23] but there have been no studies on how caimans respond to such changes.
In this study, mean rates of movement differed among years for P. trigonatus, and among years, among individuals, and depending on water level for P. palpebrosus. However, all factors affected home ranges for both species, indicating that individuals are affected by, and respond idiosyncratically to, dam filling. The margins of major Amazonian rivers are subject to deep and irregular yearly flooding, so flooded-forest organisms have to be tolerant of the sorts of changes caused by run-of-the-river dams. The higher mean water levels could eliminate important resources locally, and we had expected some of the caimans to be displaced long distances, but, in general, this did not happen. The home range of one P. palpebrosus covered 91 km2 before the dam was constructed, and the mean home-range size of the P. trigonatus in this study before dam construction was 4.89 km2, which is similar to that found for the species in an area of rainforest in central Amazonia [7].
Despite the fact that dam filling resulted in displacement of most of the individuals of both species from their original home ranges, all individuals reestablished close to where they were originally captured and none engaged in long-range movements that might result in risk of mortality due to the dam turbines. At least in the short term, the dam appears to have had little effect on the use of space by the individuals that were present before dam construction. This is good news, both from the point of view of conservation and animal welfare. However, long-term studies are necessary to determine whether the caimans will be able to survive and breed in their new home ranges.
Supporting information
S1 Fig. Contrast plot of the best model (see text) representing the moved distance between relocations (m) by caiman species.
Pp = P. palpebrosus; pt = P. trigonatus. The plot also shows the 95% confidence band (grey) and the prediction linee (blue).
https://doi.org/10.1371/journal.pone.0188508.s001
(TIF)
S1 Table. The three competing linear mixed-effects models to exam whether the different phases explained the differences in caiman movement, indexed by the distance moved between consecutive locations. dist = distance moved between locations; id = id of each monitored caiman; wl = water level, and sp = species of monitored caiman.
https://doi.org/10.1371/journal.pone.0188508.s002
(DOCX)
S2 Table. ANOVA table comparing the three competing models for caiman movement.
https://doi.org/10.1371/journal.pone.0188508.s003
(DOCX)
Acknowledgments
We grateful to Embrapa Pantanal, INPA, and SETE Soluções e Tecnologia Ambiental. José Augusto Dias, Jefferson Almeida, Aurimar Sarmento, Pedro Oliveira, Tiago Rabelo, and Daniel Martins helped with radio telemetry. Leonardo Daligna (Santo Antônio Energia) and Luiz Pellegrin (Embrapa Pantanal) helped with GIS analyses and production of Figs 3 and 4. PPBio and CENBAM provided facilities for data storage.
References
- 1. Castello L, Macedo MN. Large-scale degradation of Amazonian freshwater ecosystems. Global Change Biology. 2015; pmid:26700407
- 2. Benchimos M, Peres CA. Widespread forest vertebrate extinctions induced by a Mega hydroelectric dam, in Lowland Amazon. Plos One 2015;
- 3. Finer M, Jenkis CN. Proliferations of hydroelectric dams in the Andean Amazon and implications for Andea-Amazon connectivity. Plos One 2012; 7(4), e35126. pmid:22529979
- 4. Tundisi JC. Goldemberg J, Matsumiri-Tundisi T, Saraiva ACF. How many more dams in the Amazon? Energy polity 2014; 74:703–708.
- 5.
Lang JW. Crocodilian behavior: implications for management. In Webb G. J. W., Manolis S. C., and Whitehead P. J. (eds.). Wildlife Management: Crocodiles and Alligators, pp. 273–293. Surrey Beatty Pty. Ltd., Chipping Norton, Australia.
- 6. Campos Z Size of caimans killed by humans at a hydroelectric dam in the Madeira River, Brazilian Amazon. Herpetozoa 2015; 28(1/2):101–104.
- 7. Magnusson WE, Lima AP The ecology of a cryptic predator, Paleosuchus trigonatus, in a tropical rainforest. J Herpet. 1991; 25:41–48.
- 8. Mourão G, Campos Z Survey of broad-snouted Caiman latirostris, marsh deer Blastocerus dichotomus, and capybara Hydrochaeris hydrochaeris in the area to be inundated by Porto Primavera dam, Brazil. Biolog. Conserv. 1995; 73:27–31.
- 9. Champion G, Downs CT. Status of the Nile crocodile population in Pongolapoart Dam after river impoundment. African Zoology 2017; 52(1): 55–63.
- 10. Campos Z, Sanaiotti T, Muniz F, Farias I, Magnusson WE. Parental care in the dwarf caiman, Paleosuchus palpebrosus Cuvier, 1807 (Reptilia: Crocodilia: Alligatoridae). J Natural History 2012; 46: 2979–2984.
- 11. Campos Z, Magnusson WE. Density and Biomass Estimates by Removal for an Amazonian Crocodilian, Paleosuchus palpebrosus. Plos One 2016; 11(5): e0156406. pmid:27224473
- 12. Fisch G, Marengo JA, Nobre CA. Uma revisão geral sobre o clima da Amazônia. Acta Amazonica 1998; 28(2): 101–126.
- 13. Magnusson WE. Paleosuchus palpebrosus. Catalogue of American amphibians and reptiles 1992; 554.1: 554.2.
- 14. Paolillo AO, Gorzula S Paleosuchus palpebrosus. Herpetol. Review. 1985; 16(1): 27.
- 15. Campos Z, Magnusson WE. Thermal relations of dwarf caiman, Paleosuchus palpebrosus, in a hillside stream: evidence for an unusual thermal niche among crocodilians. J Thermal Biology. 2013; 38: 20–23.
- 16.
R DEVELOPMENT CORE TEAM 2011. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria, www.R-project.org. Accessed 01 July 2016.
- 17.
Pinheiro J, Bates D, DebRoy S, Sarkar D. R Core Team _nlme:Linear and Nonlinear Mixed Effects Models_. R package version 3.1–131, 2017. <URL: https://CRAN.R-project.org/package=nlme>.
- 18.
Zuur A, Ieno EN, Walker N, Saveliev AA, Smith GM. Mixed effects models and extensions in ecology with R. Springer-Verlag, New York, New York, USA; 2009.
- 19. Seaman DE, Powell RA. An evaluation of the accuracy of kernel density estimators for home range analysis. Ecology, 1996; 77: 2075–2085.
- 20. Calenge C. The package adehabitat for the R software: a tool for the analysis of space and habitat use by animals. Ecological Modelling 2006; 197: 516–519.
- 21. Campos Z, Coutinho M, Mourão G, Bayliss P, Magnusson WE. Long distance movements by Caiman crocodilus yacare: implications for management of the species in the Brazilian Pantanal. Herpetol. J. 2006; 16: 123–132.
- 22. Campbell HA, Dwyer RG, Irwin TR, Franklin CE. Home range utilization and long-range movement of estuarine crocodiles during the breeding and nesting season. PLoS One 2013; 8(5): e62127. pmid:23650510
- 23. Magilligan RJ, Nislow KH. Changes in hydrology regime by dams. Geomorphology 2005: 71:61–78