468488v4 Full
468488v4 Full
468488v4 Full
1
7 Instituto Pró-Pampa (IPPampa), Laboratório de Ornitologia, Pelotas, Rio Grande do
8 Sul, Brazil
2
9 Instituto Cavanilles de Biodiversidad y Biología Evolutiva, Universidad de Valencia,
3
11 Theoretical and Experimental Ecological Station, French National Center for Scientific
4
13 Departamento de Ecología - IMEM Ramon Margalef, Universidad de Alicante,
14 Alicante, Spain
15
17
18 Abstract
25 five Eryngo wetlands patches occupied (n=3) and no occupied (n=2) by Straight-billed
32 and their populations were present during the entire study period. After fragmentation
33 events, local extinction in one of the wetland patches was observed, and individuals
34 were sporadically observed in two other initially unoccupied sites. The model in which
35 fragmentation affected only the extinction probability was the most plausible among the
36 set of candidate models. Fragmentation greatly increased the chance of local population
37 extinction within patches. Our results indicate that the conservation of populations of
39 wetland patches.
42
43 Introduction
44 Continental wetlands are among the most threatened ecological systems in the
45 World (Davidson 2014). In South America, habitat fragmentation of some regions has
46 reduced the surface of this ecosystem to less than 10% of its original area (Maltchik et
bioRxiv preprint doi: https://doi.org/10.1101/468488; this version posted November 15, 2018. The copyright holder for this preprint (which
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
available under aCC-BY 4.0 International license.
47 al. 2003; Guadagnin et al. 2005). One of the most interesting and little-known
50 These wetlands are distributed in the form of patches in the grassland landscape and are
52 intensive livestock, intentional fires and drainage for dam construction (Irgang 1999;
54 Collected for the first time in June 1833 by Charles Darwin (Steinheimer 2004),
58 movements and its geographical distribution is limited to southern Brazil (Fontana et al.
59 2008; Bencke et al. 2010), and to neighboring countries – Uruguay (Aldabe et al. 2009)
60 and Argentina (Chebez et al 2011). Globally, this species is included within the "near-
61 threatened" category and its populations have been decreasing across its distributional
62 range (BirdLife International, 2016). At national scales, the scarce knowledge about its
63 population size and life history, along with its reduced distribution and high habitat
64 specificity has led to its inclusion in threatened species lists in Uruguay and Argentina,
65 under "near-threatened" (Aldabe et al. 2009) and "threatened" (Chebez et al. 2011)
69 including the presence of many other threatened species, such as annual fishes,
70 amphibians, birds and mammals (Fontana et al. 2003, Teixeira de Mello et al. 2011;
71 Lanés et al. 2014). However, in southern Brazil, regions with high concentration of this
bioRxiv preprint doi: https://doi.org/10.1101/468488; this version posted November 15, 2018. The copyright holder for this preprint (which
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
available under aCC-BY 4.0 International license.
72 type of wetland are also characterized by low annual precipitation, and the constructions
73 of dams have been very important for the development of cities and local communities
74 (Boschi et al. 2011). This is of particular concern as dams significantly affect the
81 monitored for their presence and absence during the reservoir construction activities.
85 colonization probabilities.
86
87 Methods
88 Study area
90 31º 17’ 26,4’ and W 54º 09’ 32,7’). This region greatly represents the landscape of the
91 Pampa Biome (Rambo 1959; Overbeck et al. 2007). The study area is characterized by
93 (main economic activity in the region) and agricultural crops, such as corn, soy,
94 sorghum and exotic forest (IBGE 2013). Specifically, the place designated for the
95 reservoir construction covers an area of approximately 340 hectares (Figure 1). Inside
bioRxiv preprint doi: https://doi.org/10.1101/468488; this version posted November 15, 2018. The copyright holder for this preprint (which
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
available under aCC-BY 4.0 International license.
96 the construction area, an extensive riparian forest surrounds the most important
97 watercourse of the locality, called "Arvorezinha". Grasslands and drainage lines with
98 grass and low shrubs are the dominant cover inside the area, as well as patches of
101
103 Firstly, from satellite imagery, we identified all potential wetlands within the
104 study area. During a pilot inspection in December 2011, we excluded all sites without
105 Eryngo habitat. Five Eryngo wetlands patches were selected for research (Table 1). The
106 size of wetlands varied between 5444 m2 and 34150 m2. The presence of two
108 during the pilot study. Posteriorly, between September 2012 and April 2013 (except for
109 November), we conducted seven monthly visits to the five selected wetlands. In this
110 sense, including the first observation performed during the pilot inspection, eight
111 sampling visits were performed for each wetland. Nests were recorded in three patches
112 initially occupied by the Straight-billed Reedhaunter – see Gonçalves et al. (2017) for
116 be any change in the natural structure of a wetland that implied a new split of the
117 continuous patch. All samplings were made when weather conditions were favorable
118 (without rain and little wind). To facilitate the detection of Straight-billed Reedhaunte,
119 observations were always made by two researchers. "Playback" techniques were also
bioRxiv preprint doi: https://doi.org/10.1101/468488; this version posted November 15, 2018. The copyright holder for this preprint (which
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
available under aCC-BY 4.0 International license.
120 run on the edge of patches and the search time in each site varied between 1 and 2
121 hours.
122
126 destruction of the wetlands). Four parameters were evaluated: probability of occupation
127 (ψ); probability of colonization (γ); probability of extinction (ε) and probability of
128 detection (p) (this last was kept constant in all models as the species can be easily
129 detected). We tested the fit of four models: 1) ψ(.)γ(.)ε(.)p(.) – model in which the
130 probability of occupation, colonization, extinction and detection are constants (.); 2)
135 probability. We evaluated model fit using a multimodel inference approach within an
137 estimate model plausibility we used Akaike Information Criterion (AIC) and AIC
138 weight (w), which measures the relative likelihood of the model given the data,
139 normalized across the set of candidate models (Burnham & Anderson 2003, Anderson
140 2008). Occupancy models were fitted using the Unmarked Package (Fisk and Chandler
142
143
bioRxiv preprint doi: https://doi.org/10.1101/468488; this version posted November 15, 2018. The copyright holder for this preprint (which
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
available under aCC-BY 4.0 International license.
144 Results
146 Figure 2. Of these three wetlands, two were not altered by the reservoir construction and
147 the presence of individuals was constant throughout the study period. One of the
148 occupied patches (number 3, Figure 2) was partially destroyed during the breeding
149 period (Figure 3). During the next sampling periods, the species was absent in this
150 patch, and individuals were concomitantly observed in two other initially unoccupied
152 The model in which fragmentation affected only the probability extinction was
153 the most plausible (Tab. 1). Model in which fragmentation affects both the colonization
154 and extinction was also highly plausible (ΔAIC < 2), although its probability was much
155 lower than the model in which the probability of extinction was a function of habitat
158
159 Discussion
160 Our results highlight that the effect of habitat fragmentation on Straight-billed
164 consequences to their patterns of regional and global distribution (Henle et al. 2004).
165 Although species are able to occupy fragmented landscapes when their life cycles
166 include multiple fragments (Redpath 1995), the effects of fragmentation is stronger in
167 those species with specific ecological requirements (Swihart et al. 2003; DeVictor et al.
bioRxiv preprint doi: https://doi.org/10.1101/468488; this version posted November 15, 2018. The copyright holder for this preprint (which
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
available under aCC-BY 4.0 International license.
168 2008). This seems to be the case of the Straight-billed Reedhaunter. After local
169 extinction in one of the wetlands, individuals were sporadically observed in two other
170 initially unoccupied patches, including a fragmented site. Although we cannot confirm
171 that these individuals are the same as those in the previously destroyed wetlands, the
172 temporary occurrence of the Straight-billed Reedhaunter in these sites could indicate an
173 immediate attempt to extend its territory, as a result of the habitat loss and displacement
175 The size and stability of the populations in natural wetlands depend strongly on a
176 range of habitat and landscape factors, as well as on body size, morphology, behavior,
177 effects of niche breadth and the effect of geographic range boundaries (Redpath, 1995;
178 Marsh and Trenham 2000; Swihart et al. 2003). Stable subpopulations were observed
179 only in unaltered wetlands, which indicate that the species’ ecological plasticity is
180 apparently low. Paradoxically, the species has been known to occupy wetlands in
181 widely altered landscapes, e.g., on the edge of roads and dams (Ricci and Ricci 1984;
182 Barbarskas and Fraga 1998). We recommend taking these observations cautiously as the
183 populations may have colonized these wetlands after the alteration of the landscape,
184 which favors the colonization of Eryngo due to artificial water concentrations by
185 construction of roads and dams. Hence, further research is needed to elucidate the
186 effects of the area and habitat structure on the size of the subpopulations located in
188 Our results show the Straight-billed Reedhaunter populations have greater
189 stability in unaltered patches, which reduces the chances of permanence and
191 demonstrate a more accurate view of this species’ ecological plasticity and their
192 tolerance to habitat fragmentation, and contribute to lower the uncertainties of its degree
bioRxiv preprint doi: https://doi.org/10.1101/468488; this version posted November 15, 2018. The copyright holder for this preprint (which
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
available under aCC-BY 4.0 International license.
194 (BirdLife International, 2016), but has been recently excluded from the list of
195 endangered species in Brazil (MMA, 2014) and Rio Grande do Sul State (Rio Grande
196 do Sul, 2014). Indeed, the number of records of this species has significantly increased
197 in recent years, especially in southernmost Brazil (Develey et al. 2008; ICMBIO, 2014;
198 Wikiaves, 2016). On the other hand, it is important to note that the knowledge about
199 many aspects of its biology and ecology is completely lacking. In this work, we discuss
200 the effect of fragmentation promoted by an activity that tends to drastically alter
201 landscapes. Additionally, these wetlands are commonly channeled for water subtraction
202 – a practice that also fragments wetlands and may have similar consequences to the
204 relationship of both the habitat structure and degree of connectivity of patches with the
206 spatial stochastic processes on different landscape scales influence the species
207 demographic dynamics. Finally, the natural distribution of the gravatazais may imply
208 that the Straight-billed Reedhaunter may be distributed as a metapopulation, and we end
209 this article by encouraging further research about not only the Straight-billed
210 Reedhaunter, but also about the vast biodiversity associated with wetlands dominated
211 by Eryngo.
212
213 Acknowledgements
214 We are grateful to the Ecossis Soluções Ambientais for support while undertaking
215 fieldwork.
216
bioRxiv preprint doi: https://doi.org/10.1101/468488; this version posted November 15, 2018. The copyright holder for this preprint (which
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
available under aCC-BY 4.0 International license.
217
218
219 References
220 Aldabe, J., Rocca, P., Claramunt, S. 2009. Uruguay. Pág. 383 – 392 in Devenish, C.;
221 Díaz Fernández, D. F.; Clay, R. P.; Davidson, I. & Zabala, Y. I. (eds.). Important Bird
222 Areas Americas - Priority sites for biodiversity conservation. Quito, Ecuador: BirdLife
223 International.
226 Bencke, G. A., Fontana, C. S., Dias, R. A., Maurício, G. N., Mahler Jr., J. K. (2003).
227 Aves. In Fontana CS, Bencke GA and Reis RE (ed) Livro vermelho da fauna ameaçada
228 de extinção no Rio Grande do Sul. Porto Alegre, Brasil (in Portuguese).
229 Bencke, G. A.; Dias, R. A.; Bugoni, L.; Agne, C. E.; Fontana, C. S.; Maurício, G. N. &
230 Machado, D. M. 2010. Revisão e atualização da lista das aves do Rio Grande do Sul,
231 Brasil. Iheringia, Sér. Zool. 100(4):519-556.
232 BirdLife International. 2016. Limnoctites rectirostris. The IUCN Red List of Threatened
233 Species 2016: e.T22702652A93885012. http://dx.doi.org/10.2305/IUCN.UK.2016-
234 3.RLTS.T22702652A93885012.en. Downloaded on 25 February 2018.
235 Boschi, R. A.; Oliveira, S. R. M. & Assad, E. D. 2011. Técnicas de mineração de dados
236 para análise da precipitação pluvial decenal do Rio Grande do Sul. Eng. Agríc.
237 Jaboticabal 31(6):1189-1201
238 Chebez, J. C.; Gasparri, B., Hansen Cier, M., Nigro, N. A., Rodríguez, L. 2011. Estado
239 de conservación de los tetrápodos de la Argentina. In: Porini, G. & Ramadori, D. (eds.).
240 Manejo de Fauna Silvestre en Argentina. Conservación de especies amenazadas.
241 Fundación de Historia Natural “Félix de Azara”.
242 Davidson, N. C. (2014). How much wetlands has the world lost? Long-term and recent
243 trends in global wetland area. Marine and Freshwater Research, 65, 934–941.
244 Develey, P. F., Setubal, R. B., Dias, R. A., & Bencke, G. A. (2008). Conservação das
245 aves e da biodiversidade no bioma Pampa aliada a sistemas de produção animal. Revista
246 Brasileira de Ornitologia, 16(4), 308-315.
bioRxiv preprint doi: https://doi.org/10.1101/468488; this version posted November 15, 2018. The copyright holder for this preprint (which
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
available under aCC-BY 4.0 International license.
247 Devictor, V., Julliard, R., & Jiguet, F. (2008). Distribution of specialist and generalist
248 species along spatial gradients of habitat disturbance and fragmentation. Oikos, 117(4),
249 507-514.
250 Fontana, C. S., Rovedder, C. E., Repenning, M., Gonçalves, M. L. 2008. Estado atual
251 do conhecimento e conservação da avifauna dos Campos de Cima da Serra do sul do
252 Brasil, Rio Grande do Sul e Santa Catarina. Revista Brasileira de Ornitologia,
253 16(4):281-307.
254 Guadagnin, D. L., Peter, A. S., Perello, L. F. C., & Maltchik, L. (2005). Spatial and
255 temporal patterns of waterbird assemblages in fragmented wetlands of Southern Brazil.
256 Waterbirds, 28, 261–272.
257 IBGE, Instituto Brasileiro de Geografia e Estatística. (2013). Bagé – RS. Disponível
258 em: <http://www.ibge.gov.br> Acesso em: 05.02.2015.
259 Kingsford, R. T. (2000). Ecological impacts of dams, water diversions and river
260 management on floodplain wetlands in Australia. Austral Ecology, 25(2), 109-127.
261 ICMBIO – Instituto Chico Mendes de Conservação da Biodiversidade. 2014. Plano de
262 Ação Nacional para a Conservação dos Passeriformes Ameaçados dos Campos Sulinos
263 e Espinilho.
317
318
319 Tables
320 Table 1. Model-selection table with candidate models ranked according to their AIC weights(ω) , from highest values to lowest values.
321
322
323
324
325
326
327
328
329 Table 2. Probabilities estimates (±se) for the Straight-billed Reedhaunter for the best model after the fragmentation event. Subscript for
330 extinction probabilities denote non-fragmented (εnf) and fragmented patches (εf).
ψ(.)γ()ε(Fragmentation)p(.) 0.60 (0.22) 0.12 (0.08) 0.06 (0.06) 0.67 (0.27) 1 (0)
331 E21
bioRxiv preprint doi: https://doi.org/10.1101/468488; this version posted November 15, 2018. The copyright holder for this preprint (which
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
available under aCC-BY 4.0 International license.
332
333 Fig. 1. Study area located in southernmost Brazil. The continuous line demarcates the
334 approximated flooding area of the reservoir. Selected patches are numbered and
335 marked. Image dates to April 2007 taken from Google Earth Pro (accessed on 12 June
336 2016).
337
bioRxiv preprint doi: https://doi.org/10.1101/468488; this version posted November 15, 2018. The copyright holder for this preprint (which
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
available under aCC-BY 4.0 International license.
338
339
341 dynamics. A dotted line in Figure 2a marks the area affected by the construction
342 activities. The areas with the presence and absence of species are filled in gray and
343 white, respectively. The patches with gridlines represent the occurrence of a
344 fragmentation event. The dates of the samples were: 2a) December 2011; 2b) September
345 2012; 2c) October 2012; 2d) December 2012; 2e) January 2013; 2f) February 2013; 2g)
347
bioRxiv preprint doi: https://doi.org/10.1101/468488; this version posted November 15, 2018. The copyright holder for this preprint (which
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
available under aCC-BY 4.0 International license.
348
349
350 Fig. 3a-3c. 3a, Straight-billed Reedhaunter perched on Eryngium pandanifolium; 3b,
351 Nest built on the stems of "Eryngo"; 3c, wetland partially destroyed by the reservoir
352 construction activities. Photos: 3a, Christian Andretti; 3b and 3c, Priscila Pons.
353
bioRxiv preprint doi: https://doi.org/10.1101/468488; this version posted November 15, 2018. The copyright holder for this preprint (which
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
available under aCC-BY 4.0 International license.
354
355
356
357 Fig. 4. Extinction probability (95% Confidence Interval) of Straight-billed Reedhaunter
358 estimated for each sampling month in areas with and without fragmentation. Orange:
359 fragmented patches; Grey: non-fragmented patches.
360
361