The quantity of wildlife extracted from the Amazon has increased in the past decades as a consequence of an increase in human population density and income growth. To evaluate the spatial distribution of studies on subsistence and/or commercial hunting conducted in the Brazilian Amazon, we selected eight mid-sized and large-bodied aquatic vertebrate species with a history of human exploitation in the region. We used a combination of searches in the gray and scientific literature from the past 24 years to provide an updated distributional map of studies on the target species. We calculated the distances between the study sites and the locations of the research institutes/universities that the first and last authors of the same study were affiliated to. For the period of 1990 to 2014, we found 105 studies on the subsistence and/or commercial hunting of aquatic vertebrates in the Brazilian Amazon in 271 locations that involved 43 institutions (37 Brazilian and 6 international). The spatial distribution of the studies across the Brazilian Amazon varied, but over 80% took place in the northeast and central Amazon, encompassing three States of the Legal Brazilian Amazon (Amazonas, 51.42%; Pará, 19.05%; and Amapá, 16.19%). Over half of the research study sites (52.91%) were within 500 km of the research institute/university of the first or last authors. Some research institutes/universities did not have any inter-institutional collaborations, while others collaborated with eight or more institutes. Some research institutes/universities conducted many studies, had an extensive collaboration network, and contributed greatly to the network of studies on Amazonian aquatic vertebrates. Our research contributes to the knowledge of studies on the subsistence and/or commercial hunting of the most exploited aquatic vertebrates of the Brazilian Amazon, illustrates the impact that collaboration networks have on research, and highlights potential areas for improvement and the generation of new collaborations.
Citation: Salinero MC, Michalski F (2016) Implications of Scientific Collaboration Networks on Studies of Aquatic Vertebrates in the Brazilian Amazon. PLoS ONE 11(6): e0158413. https://doi.org/10.1371/journal.pone.0158413
Editor: Peter van den Besselaar, VU University Amsterdam, NETHERLANDS
Received: December 8, 2015; Accepted: June 15, 2016; Published: June 28, 2016
Copyright: © 2016 Salinero, Michalski. 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: All relevant data are within the paper and its Supporting Information files.
Funding: MS received an MSc scholarship from the Federal Agency for Support and Evaluation of Graduate Education, Ministry of Education (“Coordenação de Aperfeiçoamento de Pessoal de Nível Superior” - CAPES). This work was partially funded by a research grant to FM (Project 88881.030414/2013-01) from the Federal Agency for Support and Evaluation of Graduate Education, Ministry of Education (“Coordenação de Aperfeiçoamento de Pessoal de Nível Superior” - CAPES). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
After habitat loss, direct human exploitation via hunting (for subsistence or commercial markets) is one of the biggest threats to sustainability in tropical regions [1–3]. Protein acquisition by local communities in tropical regions relies heavily on meat from wild vertebrates [4, 5], and some aquatic species, such as caimans [6, 7], freshwater turtles [8–10], manatees , and river dolphins , have a long history of human exploitation in the Amazon region. The use of these aquatic species is not only related to meat consumption but also to traditional medicine  and religious beliefs [7, 14], which in turn make the conservation of these species even more challenging.
Brazil has a territorial surface of 851 million hectares  and contains the largest protected area system in the world, which encompasses nearly 220 million ha . The Legal Brazilian Amazon has an area of 510 million ha  that contains approximately 40% of the world’s remaining tropical forests , and hosts a large fraction of the Earth’s biodiversity.
As of December 2010, the Legal Brazilian Amazon had 307 Conservation Units (CUs) covering over 117 million ha. Of these, the majority were sustainable-use reserves (n = 196) while 111 were strictly protected . Therefore, the majority of these reserves face the difficult challenge of balancing guaranteed access by local communities to natural resources whilst ensuring the persistence of biodiversity and ecological processes . This is a delicate balance that requires the support of legislation , and the effectiveness of CUs in protecting biodiversity can be compromised by weak or inappropriate management responses  to external threats that undermine the main goals of the reserves [23, 24].
In addition, due to a combination of factors such as overexploitation, water contamination, habitat modification, and the invasion of exotic species, aquatic environments are one of the most globally endangered habitats , with a higher biodiversity decline than terrestrial ecosystems . Aquatic environments in the Legal Brazilian Amazon are also home to important sources of protein that sustain most of the region’s local communities.
The quantity of wildlife extracted from the Amazon has increased over the past decades as a consequence of a rise in human population density, growth in income, and a greater participation of hunters in the market economy . As a result, studies on subsistence and commercial hunting in the Amazon region have also increased in the past decades [3, 4, 6–9, 12, 27–29].
However, to our knowledge, no evaluation has been conducted of the spatial distribution of studies conducted on the subsistence and commercial hunting of aquatic vertebrates in the Amazon. Therefore, there is a need to understand the spatial distribution of studies that have been conducted to evaluate the impacts of hunting on aquatic species in such an important and diversity-rich region as the Brazilian Amazon. In addition, co-authorship networks are considered an important class of academic social networking , and have been the subject of intense interest in past decades as they can influence the behavior, motivation, and performance of scientific collaborations [31, 32].
We had two research objectives in this study. Firstly, to produce a map that showed where studies of the most overexploited Amazonian aquatic vertebrates had been conducted. Secondly, to explore the distribution of these studies in relation to the location of the research institutes and universities of the affiliated authors, and to examine how researchers cooperate with each other in a scientific network of collaborations. The outcomes of these two research objectives can contribute significantly to knowledge of the current state of studies on heavily exploited aquatic species in the Brazilian Amazon. Research institutes and universities will be able to know that they are contributing to a network of collaborations as well as strengthening their scientific relationships with other research institutes that have studied aquatic vertebrates in past decades. Researchers will also be able to identify potential areas where more effort should be exerted in studying Brazilian Amazon aquatic vertebrates. Therefore, this study will identify locations that have not been studied in terms of the subsistence and/or commercial hunting of aquatic vertebrates, and improve scientific collaborations between research institutes.
Materials and Methods
We selected aquatic vertebrates that met the following criteria: 1) were distributed across Amazonia, 2) were used (historically or currently) for subsistence and/or commercial use, 3) were mid-sized or large-bodied and easily identifiable, and 4) were classified as Least Concern But Conservation Dependent, Data Deficient, or Vulnerable according to the International Union for Conservation of Nature Red List. We selected eight species that fully met these criteria. We chose two species of freshwater turtle (Podocnemis expansa, the South American river turtle and Podocnemis unifilis, the yellow-spotted river turtle), two species of caiman (Melanosuchus niger, the black caiman and Caiman crocodilus, the spectacled or common caiman), two species of manatee (Trichechus inunguis, the South American manatee and Trichechus manatus, the West Indian manatee), and two species of river dolphin (Sotalia fluviatilis, the Tucuxi and Inia geoffrensis, the Boto or pink river dolphin).
Compilation of studies
We reviewed the available literature by focusing on the subsistence hunting and commercial use of aquatic vertebrates in the Legal Brazilian Amazon. We first conducted searches of the ISI Web of Knowledge, SciVerse, Scopus, Google Scholar, and Scielo using combinations of the terms “Amazon”, “wildlife”, “conflict”, “hunting”, “subsistence hunting”, “consumption of animals”, “conservation unit”, “indigenous reserves”, “Chelonia”, “Crocodilia”, “Sirenia”, “Cetacea”, “Pará”, “Amapá”, “Amazonas”, “Acre”, “Roraima”, “Rondônia”, “Mato Grosso”, “Maranhão”, and “Tocantins”. Searches using equivalent terms in Portuguese and Spanish were also conducted to cover the period from 1990 to 2014. We conducted additional searches of the literature cited in each article obtained in our survey and from reports of the Instituto Chico Mendes de Conservação da Natureza (ICMBio) and the Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis (IBAMA) that are available on the internet. The gray literature (i.e., conference abstracts, reports, graduate theses, MSc theses, and PhD dissertations) was also searched using the same combination of terms in Google and Google Scholar.
To guarantee that only studies related to our research subject were included in our study, we first tracked all studies obtained from the combinations of keywords. Subsequently, we excluded studies that were not conducted in the Legal Brazilian Amazon and that did not involve the eight target species. The number of studies excluded and those retained were recorded for each of the screening stages according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (Fig 1) .
We examined, filtered, and collated the results returned by the searches to ensure that we considered all relevant studies that reported on the impact of subsistence hunting and the commercial use of the selected aquatic species. We then reviewed the relevant studies to extract the following data: 1) order of authorship, 2) authorship affiliation (including institution and country), 3) geographical location and coordinates of the study area, and 4) the species studied. If a document failed to provide geographical coordinates, we used Google Earth (GE) to obtain the approximate coordinates, supported by maps of the study area and key landmarks such as roads, rivers, islands, and other visual features that could be clearly distinguished in the GE images. All of the sites in our analyses were considered, even when studies reported more than one study site coordinate/location. We used ArcGIS 9.3  to produce a final distribution map of all of the study sites that focused on the subsistence and commercial hunting of aquatic vertebrates within the Legal Brazilian Amazon.
Compilation of research institutes/universities
From the results of the searches in the scientific and gray literature, we extracted the affiliations of the first and last authors in all of the documents. We chose the first and last authors as they traditionally represent those that most contributed intellectually to the paper. Usually the last author is also the senior and/or the head of the research group, and represents the senior investigator of the study. To better reflect collaboration networks in graduate theses, MSc theses, and PhD dissertations, we included the supervisor and co-supervisor (when clearly stated in the document) as co-authors. If the city for a university or institute was not cited in a document, then it was obtained from a Brazilian Ministry of Education database . Coordinates of cities that contained Brazilian universities and research institutes were obtained from Instituto Brasileiro de Geografia e Estatística (IBGE, available at ftp://geoftp.ibge.gov.br/organizacao_territorial/localidades/), and coordinates of cities that contained international universities and research institutes were obtained using the information available from the website of the institute/university or GE. We used ArcGIS 9.3  to produce a final distribution map of all the research institutions and universities of the first and last authors of the documents obtained by our literature survey. These points were overlaid with international and national borders obtained from Natural Earth (public domain, http://www.naturalearthdata.com/) and IBGE (public domain, http://downloads.ibge.gov.br/downloads_geociencias.htm). We considered the distance between research institutions and universities to be close when the linear distance between them was 500 km or less.
Collaborations between research institutes/universities that published studies together were obtained and graphically represented with a spider diagram using the extension ET Geowizard 11.2  for ArcGIS 9.3 . We also used the package Igraph  in R  to graphically display collaboration networks among institutions.
We calculated the distances between the locations of the study sites and the locations of the research institutes/universities affiliated with the first and last authors of the same study in radians using the haversine formula  in R . We then transformed the radian distances to kilometers. Finally, with the package Igraph  in R , we determined binary collaboration networks between national and international research institutes/universities and the aquatic vertebrate species. Binary network models are simple, and have been widely used in studies that have evaluated scientific collaboration networks [31, 40, 41]. We used the degree of node centrality  to characterize the network; the degree is the simplest of the node centrality measures, and only uses the local structure around nodes. In a binary network, the degree is the number of links possessed by a node .
Spatial and temporal distribution of studies
Our searches returned 105 studies in 271 study locations in the Legal Brazilian Amazon in which the subsistence and/or commercial hunting of aquatic vertebrates occurred (S1 Table). Most studies were conducted after 2007 (80.6%). The highest number of studies was published in 2013 (n = 24), followed by 2014 (n = 18) (Fig 2).
The color gradient is proportional to the number of studies in each year.
The spatial distribution of the studies within the Legal Brazilian Amazon varied, but over 80% took place in the northeast and central Amazon and encompassed three States (Amazonas, 51.42%; Pará, 19.05%; and Amapá, 16.19%). The other regions of the Legal Brazilian Amazon had lower proportions of studies and involved another five States (Maranhão, 3.81%; Tocantins, 3.81%; Acre, 1.91%; Mato Grosso, 1.91%; Rondônia, 0.95%; and Roraima, 0.95%) (Fig 3).
(A) Geographical distribution of studies with aquatic vertebrate species (black dots) and locations of research institutes (red circles with black dots) in South America; (B) Locations of national and international research institutes that performed the studies (red circles with black dots) and network collaborations among them (black lines).
Collaboration networks of research institutes/universities on studies of aquatic vertebrates
The studies involved 43 institutions (37 Brazilian and 6 non-Brazilian) (Fig 3, S2 Table). Nearly half of the studies analyzed had up to two authors (n = 52, 49.52%), giving us confidence that our analysis using the first and last authors was representative of the dataset. Only 44.76% (n = 47) of the documents were scientific articles or book chapters; the majority were conference abstracts (n = 26) and graduate and MSc theses (n = 22), followed by PhD dissertations (n = 6), reports (n = 2), and magazine articles (n = 2). Some (n = 10, 23.26%) research institutes/universities did not collaborate with any other institute/university (Table 1). Considering only the 33 institutes/universities with collaborations, there were on average 2.06 collaborations per institution (range, 1–10, SD = 2.34). Three institutions had substantially more collaborations than the others, and represented the top 5% of the value range (IDSM, UFPA, and INPA, with 10, 9, and 8 collaborations, respectively; Table 1, Fig 4).
Each circle (node) represents an institution, and the lines connecting the nodes indicate collaborative relationships among institutions. Node sizes correspond to the degree of centrality.
Overall, studies on the subsistence and/or commercial hunting of Amazonian aquatic vertebrates occurred more frequently in areas close to research institutes/universities: 52.9% of the research study sites were located within 500 km of the research institute/university of the first or last authors (Fig 5). There was a decrease in the number of studies when distances from the affiliations of the authors increased. International institutions (4000 km or more away) that conducted research on aquatic vertebrates were exceptions to this trend.
The long dashed line shows the locations of studies conducted at 500-km distances from the authors’ affiliations. Red bars represent a high frequency of counts of distances from the institutions.
Some research institutes/universities, such as IDSM, INPA, IPI, UFPA, UFRA, and UNESP, conducted many studies and had large collaboration networks, and contributed most to the network of studies on Amazonian aquatic vertebrates (Table 1, Fig 6). Among the species selected, the most studied was the yellow-spotted river turtle, followed by the South American river turtle, the common caiman, the black caiman, the Boto, the South American manatee, the Tucuxi, and the West Indian manatee (Fig 6). The same trend was observed in contributions to the network structure in degrees: a large number of institutions were involved in studies of the two species of freshwater turtle.
Sizes and colors of the circles represent the contribution of each institute to the network. Aquatic species (A–H) are shown in descending order of number of studies in the Legal Brazilian Amazon. A, Podocnemis unifilis; B, Podocnemis expansa; C, Caiman crocodilus; D, Melanosuchus niger; E, Inia geoffrensis; F, Trichechus inunguis; G, Sotalia fluviatilis; H, Trichechus manatus.
Only one institution (IDSM) in the collaboration network performed studies on all of the aquatic species evaluated, while other institutions (INPA, UFPA, UFRA, UNESP, and UNIFAP) performed studies on seven species (Table 2). The majority of the institutions (IBAMA, ICMBio, IFPA, OMACHA, UEMA, UFLA, UFPE, UNB, UNEMAT, UFL, and ULISBOA) studied only one aquatic species (Table 2).
In recent years, there has been an increase in the number of studies on Amazonian aquatic vertebrates and most of them have been multi-authored, indicating that scientific collaboration plays an important role in the study of aquatic vertebrates. Although international and interdisciplinary scientific collaboration has expanded rapidly in all fields of research  and increases citation rates of ecological articles , we found that there have been few international collaborations on Amazonian aquatic vertebrates over the past 24 years. Consequently, there is a huge opportunity for national and international research institutes to collaborate in the study of aquatic vertebrates in the largest remaining expanse of tropical forest in the world.
Of the species evaluated, the two species of freshwater turtle (P. unifilis and P. expansa) were the most studied. Freshwater turtles have a long history of consumption by Brazilian Amazon riverine communities [9, 45], and this has generated great attention from the scientific community. As protein acquisition for local communities in tropical regions relies heavily on meat from wild vertebrates [4, 5], it is expected that this subject will attract research interest. Studies that have investigated human exploitation effects on aquatic vertebrates have increased in the past 10 years.
However, some aquatic species, such as manatees and Botos, are still poorly studied. This could be related to the difficulty of studying these animals in the wild, because of their biology and natural history [11, 12]. The semi-aquatic nature of freshwater turtles, including their habits of basking and beach-nesting , makes them easier subjects to study than the other species included here. Caimans also have the habit of basking on fallen trees and in open areas close to riverbanks, and care for their eggs and hatchlings on riverbanks . Therefore, the fact that caimans are the most studied species after freshwater turtles was to be expected. There is a myriad of uses of aquatic species in the Brazilian Amazon that enhances their potential interest for scientific research. These uses include traditional medicine  and religious beliefs [14, 47], which interact additively or synergistically with meat consumption and make the study of these species challenging.
Scientific collaborations have been developing interdisciplinary approaches in recent years . Overall, 43 research institutes were included in our 105 studies, 33 of which had some collaboration with other institutes. Institutes such as IDSM, UFPA, and INPA had the highest number of collaborations. In Amazonia, these institutes are known to have high international connectivity due to the number of international researchers hired in each institute, and to their long history of studies on ecology. INPA and UFPA were created in 1952 and 1957, respectively, and are amongst the most traditional research institutes in the Brazilian Amazon. In contrast, some institutes did not have any collaborations, and worked alone in their studies on aquatic vertebrates (IFPA, PUC Goiás, UEPA, UEPB, UFL, UFSC, UFT, UNB, UNEMAT, and USP). A potential limitation of our study was that we analyzed documents only using the first and last authors; datasets including all authors from all research groups would provide a better overview of the scientific network of collaborations.
Our study demonstrates that the locations of aquatic vertebrate study sites were close to the research institutes of the first and last authors. Over 50% of the studies were conducted within 500 km of the institute to which the researchers were affiliated to, so the spatial distribution of studies on Amazonian aquatic vertebrates has a clear bias that needs to be better explored and analyzed.
Faculty effort and time commitments for developing and delivering graduate courses have been evaluated previously ; however, to our knowledge, no studies have been conducted that have evaluated the effort and time researchers allocate in order to conduct field research in ecology. There are obvious constraints in terms of costs and time if researchers need to travel long distances to conduct fieldwork. However, in studies on subsistence and commercial hunting, there are increasing levels of hunting in areas close to human settlements . Therefore, our study provides clear evidence that the development of new strategies and improvements in scientific collaboration [44, 48] will benefit studies on aquatic vertebrates.
An enhanced understanding of the dynamics of scientific collaborations will help to direct efforts and improve communication effectiveness, enhance interactions between institutes, and reduce costs from a teamwork perspective .
Studies on Amazonian aquatic vertebrates will become increasingly important for tropical wildlife conservation because of the growing number of built and planned hydroelectric dams in the Amazon region. In the present study, we have shown that most of the studies that have been conducted were within 500 km of the lead authors’ affiliated institutes. This indicates a strong potential bias in studies of the subsistence and commercial hunting of aquatic vertebrates. We also show that improvements in national, and particularly international, collaborations would benefit both the studies and the research institutes by increasing article citation rates and reducing the costs associated with fieldwork. To enhance scientific collaborations in studies on Amazonian aquatic vertebrates, we consider it essential to improve communication between research institutes and researchers and increase the distance between survey sites and the institutes to which the researchers are affiliated.
S1 Table. List of 105 studies on the subsistence and/or commercial hunting of Amazonian aquatic vertebrates.
S2 Table. List of research institutes/universities that conducted studies on Amazonian aquatic vertebrates.
The Federal University of Amapá (UNIFAP) provided logistical support. We are deeply indebted to Marie-Josée Fortin and Darren Norris for providing insightful comments and suggestions on earlier versions of the manuscript. This article resulted from a Programa de Pós-Graduação em Biodiversidade Tropical (PPGBIO) workshop that was held in October 2015.
Conceived and designed the experiments: MS FM. Performed the experiments: MS. Analyzed the data: MS FM. Contributed reagents/materials/analysis tools: FM. Wrote the paper: MS FM.
- 1. Redford KH. The empty forest. Bioscience. 1992;42(6):412–22. WOS:A1992HW24500018.
- 2. Wright SJ. The myriad consequences of hunting for vertebrates and plants in tropical forests. Perspect Plant Ecol Evol Syst. 2003;6(1–2):73–86. WOS:000184344400006.
- 3. Peres CA, Palacios E. Basin-wide effects of game harvest on vertebrate population densities in Amazonian forests: Implications for animal-mediated seed dispersal. Biotropica. 2007;39(3):304–15. WOS:000245939300003.
- 4. Peres CA. Effects of subsistence hunting on vertebrate community structure in Amazonian forests. Conserv Biol. 2000;14(1):240–53. WOS:000085366700029.
- 5. Redford KH, Robinson JG. The game of choice—patterns of Indian and Colonist hunting in the Neotropics. Am Anthropol. 1987;89(3):650–67. WOS:A1987K111800007.
- 6. Da Silveira R, Thorbjarnarson JB. Conservation implications of commercial hunting of black and spectacled caiman in the Mamiraua Sustainable Development Reserve, Brazil. Biol Conserv. 1999;88(1):103–9. WOS:000078180000010.
- 7. Baia PC Junior, Guimaraes DA, Le Pendu Y. Non-legalized commerce in game meat in the Brazilian Amazon: a case study. Rev Biol Trop. 2010;58(3):1079–88. WOS:000281071900019. pmid:20737856
- 8. Coway-Gomez K. Effects of human settlements on abundance of Podocnemis unifilis and P. expansa turtles in Nortbeastern Bolivia. Chelonian Conserv Biol. 2007;6(2):199–205. WOS:000252138600005.
- 9. Schneider L, Ferrara CR, Vogt RC, Burger J. History of Turtle Exploitation and Management Techniques to Conserve Turtles in the Rio Negro Basin of the Brazilian Amazon. Chelonian Conserv Biol. 2011;10(1):149–57. WOS:000292635400022.
- 10. Norris D, Michalski F. Socio-economic and spatial determinants of anthropogenic predation on Yellow-spotted River Tuttle, Podocnemis unifilis (Testudines: Pelomedusidae), nests in the Brazilian Amazon: Implications for sustainable conservation and management. Zoologia. 2013;30(5):482–90. WOS:000326771500003.
- 11. Rosas FCW. Biology, conservation and status of the Amazonian Manatee Trichechus-inunguis. Mammal Rev. 1994;24(2):49–59. WOS:A1994NV13900001.
- 12. Martin AR, Da Silva VMF, Salmon DL. Riverine habitat preferences of botos (Inia geoffrensis) and tucuxis (Sotalia fluviatilis) in the central Amazon. Mar Mamm Sci. 2004;20(2):189–200. WOS:000220806900001.
- 13. Nobrega Alves RR, Rosa IL. Trade of Animals Used in Brazilian Traditional Medicine: Trends and Implications for Conservation. Human Ecology. 2010;38(5):691–704. WOS:000283364800009.
- 14. Alves RRN, Rosa IL, Leo Neto NA, Voeks R. Animals for the Gods: Magical and Religious Faunal Use and Trade in Brazil. Human Ecology. 2012;40(5):751–80. WOS:000309863900008.
- 15. IBGE. Área Territorial Oficial. Available from: http://www.ibge.gov.br/home/geociencias/cartografia/default_territ_area.shtm. Acessed 2015 Nov 08. 2015.
- 16. Bernard E, Penna LAO, Araujo E. Downgrading, downsizing, degazettement and reclassification of protected areas in Brazil. Conserv Biol. 2014;28(4):939–50. WOS:000339501100006. pmid:24724978
- 17. FAO. Computerized Data Gathering and Networking as a Control and Monitoring System for the improvement of and reporting on forest management in the Amazon: the case of Brazil. Forest Management Working Papers, Working Paper 27. Forest Resources Development Service, Forest Resources Division. Rome. Available at: ftp://ftp.fao.org/docrep/fao/009/j5416e/j5416e00.pdf Acessed 2015 Nov 08. 2005.
- 18. Kirby KR, Laurance WF, Albernaz AK, Schroth G, Fearnside PM, Bergen S, et al. The future of deforestation in the Brazilian Amazon. Futures. 2006;38(4):432–53. WOS:000236337100004.
- 19. Veríssimo A, Rolla A, Vedoveto M, Futada SM. Áreas Protegidas na Amazônia Brasileira: avanços e desafios. Belém/São Paulo: Imazon e ISA; 2011. 87 p.
- 20. Schwartzman S, Moreira A, Nepstad D. Rethinking tropical forest conservation: Perils in parks. Conserv Biol. 2000;14(5):1351–7. WOS:000089816800021.
- 21. Peres CA. Conservation in Sustainable-Use Tropical Forest Reserves. Conserv Biol. 2011;25(6):1124–9. WOS:000296962700018. pmid:22070267
- 22. Leverington F, Costa KL, Pavese H, Lisle A, Hockings M. A Global Analysis of Protected Area Management Effectiveness. Environ Manage. 2010;46(5):685–98. WOS:000283360300003. pmid:20859627
- 23. Laurance WF, Useche DC, Rendeiro J, Kalka M, Bradshaw CJA, Sloan SP, et al. Averting biodiversity collapse in tropical forest protected areas. Nature. 2012;489(7415):290–4. WOS:000308635900041. pmid:22832582
- 24. Pfaff A, Robalino J, Herrera D, Sandoval C. Protected Areas' Impacts on Brazilian Amazon Deforestation: Examining Conservation—Development Interactions to Inform Planning. PLoS One. 2015;10(7):e0129460. WOS:000358837700002. pmid:26225922
- 25. Laurance WF, Sayer J, Cassman KG. Agricultural expansion and its impacts on tropical nature. Trends Ecol Evol. 2014;29(2):107–16. WOS:000332749300011. pmid:24388286
- 26. Sala OE, Chapin FS, Armesto JJ, Berlow E, Bloomfield J, Dirzo R, et al. Biodiversity—Global biodiversity scenarios for the year 2100. Science. 2000;287(5459):1770–4. WOS:000085775300030. pmid:10710299
- 27. Godoy R, Undurraga EA, Wilkie D, Reyes-Garcia V, Huanca T, Leonard WR, et al. The effect of wealth and real income on wildlife consumption among native Amazonians in Bolivia: estimates of annual trends with longitudinal household data (2002–2006). Anim Conserv. 2010;13(3):265–74. WOS:000277122500006.
- 28. Peres CA, Lake IR. Extent of nontimber resource extraction in tropical forests: Accessibility to game vertebrates by hunters in the Amazon basin. Conserv Biol. 2003;17(2):521–35. WOS:000181736400021.
- 29. Redford KH, Stearman AM. Forest-dwelling native Amazonians and the conservation of biodiversity—interests in common or in collision. Conserv Biol. 1993;7(2):248–55. WOS:A1993LE12500014.
- 30. Uddin S, Hossain L, Rasmussen K. Network Effects on Scientific Collaborations. PLoS One. 2013;8(2):e57546. WOS:000315524900120. pmid:23469021
- 31. Newman MEJ. The structure of scientific collaboration networks. Proc Natl Acad Sci U S A. 2001;98(2):404–9. WOS:000166485300010. pmid:11149952
- 32. Uddin S, Hossain L, Abbasi A, Rasmussen K. Trend and efficiency analysis of co-authorship network. Scientometrics. 2012;90(2):687–99. WOS:000299088900022.
- 33. Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Plos Medicine. 2009;6(7):e1000097. WOS:000268452400005.
- 34. ESRI. ArcGIS Desktop: Release 9.3. Redlands, CA: Environmental Systems Research Institute; 2009.
- 35. MEC. Instituições de Educação Superior e Cursos Cadastrados. Available: http://emec.mec.gov.br. Acessed 2015 May 08. 2015.
- 36. ET Spatial Techniques. ArcGIS and ArcView extensions. Available: http://www.ian-ko.com. 2014.
- 37. Csardi G, Nepusz T. The igraph software package for complex network research. InterJournal. 2006;Complex Systems:1695.
- 38. R Development Core Team. R 3.1.1: A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing; 2014.
- 39. Sinnott RW. Virtues of the Haversine. Sky and Telescope. 1984;68(2):159. WOS:A1984TA75300024.
- 40. Bian J, Xie M, Hudson TJ, Eswaran H, Brochhausen M, Hanna J, et al. CollaborationViz: Interactive Visual Exploration of Biomedical Research Collaboration Networks. PLoS One. 2014;9(11):e111928. WOS:000347121300018. pmid:25405477
- 41. Liu XM, Bollen J, Nelson ML, Van de Sompel H. Co-authorship networks in the digital library research community. Information Processing & Management. 2005;41(6):1462–80. WOS:000231332600011.
- 42. Opsahl T, Agneessens F, Skvoretz J. Node centrality in weighted networks: Generalizing degree and shortest paths. Social Networks. 2010;32(3):245–51. WOS:000279409000009.
- 43. Glanzel W. Coauthorship patterns and trends in the sciences (1980–1998): A bibliometric study with implications for database indexing and search strategies. Library Trends. 2002;50(3):461–73. WOS:000176374100010.
- 44. Leimu R, Koricheva J. Does scientific collaboration increase the impact of ecological articles? Bioscience. 2005;55(5):438–43. WOS:000229009300011.
- 45. Alves RRN, Vieira KS, Santana GG, Vieira WLS, Almeida WO, Souto WMS, et al. A review on human attitudes towards reptiles in Brazil. Environmental Monitoring and Assessment. 2012;184(11):6877–901. WOS:000309476100033. pmid:22134858
- 46. Somaweera R, Brien M, Shine R. The role of predation in shaping Crocodilian natural history. Herpetological Monographs. 2013;27(1):23–51. WOS:000332670700002.
- 47. Alves RRN, Rosa IL. Use of tucuxi dolphin Sotalia fluviatilis for medicinal and magic/religious purposes in north of Brazil. Human Ecology. 2008;36(3):443–7. WOS:000256823800011.
- 48. Katz JS, Martin BR. What is research collaboration? Research Policy. 1997;(26):1–18. WOS:A1997XC57700001.
- 49. Visser JA. Faculty work in developing and teaching web-based distance courses: A case study of time and effort. American Journal of Distance Education. 2000;14(3):21–32.
- 50. Melin G, Persson O. Studying research collaboration using co-authorships. Scientometrics. 1996;36(3):362–77. WOS:A1996VG50500006.