Climate Change Sensitivity of Threatened, and Largely Unprotected, Amazonian ﬁshes

Home , Aguarunichthys, Baryancistrus, Lebiasinidae, Ossubtus xinguense, Rhynchodoras, Sternarchorhynchus

AQUATIC CONSERVATION: MARINE AND FRESHWATER ECOSYSTEMS Aquatic Conserv: Mar. Freshw. Ecosyst. 26 (Suppl. 1): 91–102 (2016) Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/aqc.2658

Climate change sensitivity of threatened, and largely unprotected, Amazonian ﬁshes

RENATA G. FREDERICOa,c,*,, JULIAN D. OLDENa and JANSEN ZUANONb aSchool of Aquatic and Fishery Sciences, University of Washington, Washington, USA bCoordenação de Biodiversidade, Instituto Nacional de Pesquisas da Amazônia, Manaus, Amazonas, Brazil cInstituto de Ciências Biológicas, Universidade Federal de Goiás, Goiânia, Brazil

ABSTRACT 1. Climate change is poised to have fundamental impacts on the freshwater environments of the Amazon River Basin, and protected areas are routinely proposed as a possible management strategy to conserve freshwater fishes. However, there remains a paucity of information regarding the sensitivity of threatened fish species to climate- induced changes in water quantity and quality. 2. An expert-based survey was used to address the following questions: (1) Are currently threatened fish species in Brazil also sensitive to projected effects of climate change? (2) Does the current conservation status of fish species also reflect their degree of sensitivity to climate change? and (3) What are the specific aspects of climate change that are likely to contribute the most to species sensitivity? 3. Survey respondents evaluated 35 species (11 families) representing 50% of the threatened species in the Brazilian Amazon. The results suggest that the majority of threatened Brazilian fish species are considered highly sensitive to climate change impacts. 4. Climate-induced changes in water quality were, on average, considered a greater threat to species persistence than potential changes in water quantity. Survey results also suggest that fishes exhibit high sensitivity to changes in temperature and dissolved oxygen, and moderate to high sensitivity to changes in high-flow (i.e. flood) and low-flow (i.e. drought) regimes. 5. A considerable mismatch was found between species conservation status and sensitivity to climate change, suggesting that perceptions of present-day extinction risk do not necessarily provide insight into future risks associated with climate change. 6. Species sensitivity to climate change showed no relationship to dispersal ability, indicating that protected areas may serve as important refugia for those species unable to keep pace with climate change. Despite this, the number and size of protected areas in Brazil have decreased over the past decade, largely to support the exploitation of hydropower and mining. Strategic conservation planning that involves existing and new protected areas for those species most at risk to climate change is warranted. Copyright # 2016 John Wiley & Sons, Ltd.

Received 21 October 2015; Accepted 20 March 2016

KEY WORDS: tropical areas; impacts; hydrological regimes; freshwater ecosystems; protected areas; conservation

*Correspondence to: Renata G. Frederico, Instituto de Ciências Biológicas, Universidade Federal de Goiás, Goiânia, Brazil. E-mail: [email protected]

INTRODUCTION (Sousa and Freitas, 2008); thus climate-driven changes to the magnitude, duration and timing of Protected areas (PAs) represent one of many ecologically-critical flow events will have direct possible strategies to conserve freshwater fishes in consequences for the breeding, migration and the light of future climate change, albeit with a persistence of fish species (Bunn and Arthington, number of recognized challenges (Abell et al., 2002; Poff et al., 2002; Leigh et al., 2014). In 2007; Lawrence et al., 2011). Protected areas may addition, elevated temperature and depressed river be less affected by the impacts of climate change discharge will promote conditions of reduced and other human activities, thus providing habitat dissolved oxygen, which affects primary refugia for species in the future (Pittock et al., productivity (Poff et al., 2002; Leigh et al., 2014) 2008). For example, the network of PAs in and interacts with species-specific physiological tropical forests can buffer the effects of climate tolerances to determine fish performance and change by decreasing the velocity of future patterns of spatial occupancy (Brauner and Val, changes in temperature and precipitation (Loarie 2006; Barletta et al., 2010). et al., 2009). Brazil has the largest PA network in Recent decades have witnessed the rapid evolution the world (Bernard et al., 2014), together with of assessments on climate change vulnerability to some of the largest conservation units (with more help set priorities and guide conservation efforts, than 3 million ha), classified according to Peres including the identification of protected areas (2005) as mega-reserves. In total, approximately (Williams et al., 2008). Approaches include 43% of Brazilian Amazon territory are considered evaluating vulnerability according to intrinsic traits protected areas (Veríssimo et al., 2011). With the of species (Foden et al., 2013; Pearson et al., 2014), goal of reducing rates of deforestation, the past changes in geographic distribution (Bush et al., Brazilian Amazon witnessed rapid growth in the 2014), exposure to extrinsic conditions associated number of PAs established in the 2000s (Veríssimo with climatic changes (Foden et al., 2013), and the et al., 2011). As a result many PAs were located in combination of both intrinsic and extrinsic factors the south-eastern region of the Amazon where (Gardali et al., 2012; Bush et al., 2014; Case et al., numerous human threats already exist. These efforts 2015). For instance, Foden et al. (2013) combined were effective in slowing the rate of deforestation, information about species sensitivity and exposure and thus reduced carbon dioxide emissions (Soares- to projected climate change for the world’sbirds, Filho et al., 2010). However, the Amazon region is amphibians and corals to identify taxa and areas expected to face considerable threats in the future that require conservation attention. Similarly, Bush owing to the combined impacts of climate change et al. (2014) assessed the vulnerability of Australian and continued deforestation (Malhi et al., 2008). Odonata according to past changes in species It is predicted that the Amazon River Basin will ranges and projected exposure to future climate be especially prone to the ecological impacts of change. In data-limited situations, expert knowledge future climate change (Pacifici et al., 2015; Castello has emerged as a viable approach for assessing and Macedo, 2016). Models indicate that more species sensitivity to climate change (Moyle et al., than one-third of freshwater fish species in the 2013; Case et al., 2015). For instance, Moyle et al. Amazon, Tocantins and Araguaia river basins may (2011, 2013) used experts to evaluate the threatened be lost if the most pessimistic scenarios of climate status of Californian (USA) freshwater fishes change are realized (Xenopoulos et al., 2005). The according to future threats of climate change. Intergovernmental Panel on Climate Change In Brazil alone there are 310 species classified as (IPCC) recently forecast that the Amazon will face threatened according to the most recent national warmer and dryer conditions, with less predictable evaluation (normative instruction number 445, rainfall and more extreme events (e.g. droughts www.icmbio.gov.br). Of the many threats to and floods) in the future (Haylock et al., 2006; Brazilian fishes, habitat loss and degradation, river Brodie et al., 2012). Amazonian fish communities fragmentation by dams, invasive species, are largely structured by flood pulse dynamics overexploitation, pollution, and climate change,

Copyright # 2016 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 26 (Suppl. 1): 91–102 (2016) CLIMATE CHANGE IN AMAZONIAN FISHES 93 are of major concern (Barletta et al., 2010; METHODS Nogueira et al., 2010; Castello et al.,2013).The The frameworks published by Galbraith and Price threatened species have at least 25% of their range (2009) and Moyle et al. (2013) were modified to already affected by at least one of these factors, thus assess sensitivity of threatened Brazilian fishes to putting into question whether existing protecting climate change; this ensured a repeatable and areas will be adequate to achieve conservation goals. transparent process for eliciting expert knowledge. Despite this growing awareness, management and A group of 44 experts were identified who had conservation efforts to protect Brazilian freshwater recently participated in the evaluation that resulted ecosystems and their constituent fish species remain in the Brazilian Red List of freshwater fishes. The scarce (Castello et al.,2013;Bernardet al., 2014). experts are professors and researchers, each with Managing Amazonian freshwater fishes into the more than 10 years experience in working with the future requires a robust understanding of which taxonomy and/or ecology of Amazonian fishes. species will be most sensitive to projected climatic They were contacted by e-mail and had 40 days to changes, and the factors that will make them more answer the survey. Half of them (22 experts) sensitive (Castello and Macedo, 2016). This study completed the survey. sought to evaluate the sensitivity of Amazonian freshwater fishes that are currently threatened (by Threatened species several factors) to potential climate change impacts. Restricted geographic ranges, small In December 2014, the Brazilian Ministry of the populations, and low tolerance to environmental Environment published a national Red List of all changes are intrinsic traits that typically make vertebrate and selected invertebrate species (http:// species more sensitive to a given level of exposure www.icmbio.gov.br/portal/biodiversidade/fauna- to climate change (Williams et al., 2008; Pearson brasileira/lista-de-especies.html). This list was the et al., 2014), and are thus considered vulnerable. outcome of numerous workshops where specialists An expert-based survey was used to assess evaluated almost 4000 freshwater species according sensitivity given the limited published information to the IUCN Red List criteria. This study evaluated fi on the physiological and life-history traits of 71 sh species occurring in the Amazon River Basin Amazonian freshwater fishes (Barletta et al., 2010; that are considered threatened according to the Baigún et al., 2012; Castello et al., 2013; Ota Brazilian Red List (Appendix S1, Supplementary et al., 2015). Here, the focus was on sensitivity material). The present-day threats to these species because all species are likely to be exposed to include habitat loss and fragmentation by land comparable climatic changes owing to similarities conversion, dams, and other human activities. It is in their geographic distributions. The following worth noting that the Brazilian Ministry of the questions were addressed: (1) are currently Environment (2014) (normative instruction threatened fish species also sensitive to projected number 445, www.icmbio.gov.br) did not include fi effects of climate change, specifically those effects large-bodied sh species that are exploited in fi fi associated with changing water quantity and sheries, such as the giant cat shes (Pimelodidae), quality? (2) Does the current conservation status owing to their wide distribution in the Amazon of fish species also reflect their degree of sensitivity basin. fi to climate change? and (3) What are the speci c Survey methods aspects of climate change that contribute the most to species sensitivity? By addressing these An online survey was developed and administered questions this paper aims to provide new insight using Qualtrics software, version 2.4 (# 2015, into which threatened Brazilian fish species are at Provo, UT). An online survey was chosen rather most risk to future climate change, thus identifying than in-person interviews because experts were the species requiring greater consideration for widely distributed across Brazil. All survey conservation and discussing the potential role of participants were involved with the species protected areas. evaluation conducted by the Brazilian Ministry of

Copyright # 2016 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 26 (Suppl. 1): 91–102 (2016) 94 R. G. FREDERICO ET AL. the Environment, so they were aware of species Physiological and/or behavioural tolerances to threats and gaps in knowledge. Experts were asked climate change impact to evaluate as many species as they considered Fish species demonstrate different degrees of having sufficient knowledge. The survey comprised tolerance to climate-associated changes to the seven multi-part questions on each species, environment (Eaton and Scheller, 1996; Ficke addressing basic ecology, current threats and the et al., 2007). The experts were asked to classify the degree of sensitivity to projected climate change. tolerance of species as: (1) very low; (2) low; (3) The survey instrument is presented in the moderate; or (4) high, to changes in timing, Appendix S2 and summarized below. magnitude and duration of the high-flow season fl fl Species range ( ood) and the low- ow season (hydrological drought), and to dissolved oxygen concentration Fish species with small geographic ranges are more and water temperature. In addition, the experts likely to be sensitive to environmental change were asked to state their degree of certainty for because they exhibit narrower habitat requirements each response as high, moderate or low. (Eaton and Scheller, 1996; Nogueira et al., 2010). To estimate range size, experts were asked to Statistical analysis classify the species’ current distribution according to: (1) species limited to a portion of one Amazon A composite measure of climate change sensitivity sub-basin; (2) species that occur in just one sub- for each species was calculated using the following basin; (3) species that occur in two sub-basins; and protocol. First, each question (with the exception (4) species that occur in three or more sub-basins. of species range and current threats) was arranged on an ordinal scale from the lowest to the highest Species dispersal ability perceived sensitivity and then standardized from 0 to 1, where 0 represented low sensitivity and 1 Dispersal ability is considered paramount to represented high sensitivity to climate-related threat fi whether sh species will shift their distribution in (Appendix S3). This ensured that all the questions response to climate change (Poff et al., 2002; had equal weighting. For species with multiple Freitas et al., 2013). The experts were asked to entries, the average score obtained before fi classify species dispersal ability (of adult sh only) standardization was used. High inter-respondent fi into one of ve categories that included: (1) very agreement supported this decision (described below). < À1 low dispersal rate ( 2 km year ); (2) low dispersal Second, the overall sensitivity score of each species to – À1 rate (2 5 km year ); (3) moderate dispersal rate climate change was represented by the sum of the – À1 (5 10 km year ); (4) high dispersal rate standardized scores for dispersal ability (DA), flood – À1 (10 50 km year ) and; (5) very high dispersal rate (F), drought (D), temperature (T) and dissolved > À1 ( 50 km year ). oxygen (DO) (DA + F + D + T + DO = overall sensitivity). This resulted in a composite measure Present impacts other than climate change ranging from 0 to 5, where 0 represents lowest overall Fish species threatened by other environmental sensitivity and 5 represents highest overall sensitivity stressors may be more vulnerable to the effects of to climate change impacts (Appendix S3). Range size climate change (Olden et al., 2010; Moyle et al., was not included because all the threatened species 2013). The experts were asked to classify present evaluated have small range sizes. environmental stressors to the focal fish species The level of inter-respondent agreement for (following the recent review by Castello et al., species with multiple entries was quantified using 2013) that included: (1) dams and water Fleiss’ Kappa index (Conger, 1980). Fleiss’ kappa diversions; (2) timber logging; (3) pollution; (4) ranges from 0 (indicating no agreement) to 1 mining; (5) urbanization; (6) invasive species; (7) (indicating complete agreement), and was calculated overharvesting; and (8) no stressor other than to estimate the concordance among the entries. climate change. Pearson correlation coefficients were used to assess

Copyright # 2016 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 26 (Suppl. 1): 91–102 (2016) CLIMATE CHANGE IN AMAZONIAN FISHES 95 the relationship between overall sensitivity and the ability and sensitivity (tolerance to water quality degree of expert certainty. During the species and quantity parameters). All the analyses were evaluation, the Brazilian government did not take performed in the R programming language (R Core into account the potential climate change impacts. Team, 2012). Thus, a Kruskall–Wallis non-parametric analysis of variance was performed to determine whether there RESULTS was a relationship between the current threatened status assigned by the Brazilian government and the Survey respondents evaluated 35 species (11 species’ sensitivity to climate change. The frequency families) representing 50% of the threatened distribution of sensitivity scores to different water species in the Brazilian Amazon (Figure 1, quantity and quality parameters were examined to Appendix S1). The majority of threatened identify the primary threats associated with climate Brazilian Amazon ﬁsh species are considered change. Ordinary least squares (OLS) regression was highly sensitive to climate change impacts applied to quantify the relationship between dispersal according to the expert survey (Figure 2). Almost

Figure 1. Map of occurrence points of the 35 threatened freshwater ﬁsh species in the Brazilian Amazon that were covered by the survey.

Copyright # 2016 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 26 (Suppl. 1): 91–102 (2016) 96 R. G. FREDERICO ET AL.

Figure 3. Overall sensitivity to climate change for freshwater fish species Figure 2. Overall sensitivity to climate change for 35 threatened fi according to present-day threatened status given by the Brazilian freshwater sh species in the Brazilian Amazon. Government. Boxplots show the 25th, 50th and 75th percentiles; whiskers show the 10th and 90th percentiles, with circles representing half of these species had multiple entries with outliers. statistically significant concordance among the climate change were not incorporated into the experts’ answers (Fleiss’ Kappa: mean = 0.42, government evaluation. Survey respondents P < 0.05, n = 17, Table 1). Little correlation was identified dams and diversions, pollution and found between species’ overall sensitivity scores urban development as the leading non-climate and the degree of certainty reported by the threats to fishes (Figure 4). respondents (Pearson’s correlation R = À0.113, Climate-induced changes in water quality were, P = 0.521, n = 35). on average, considered a greater threat to species Species overall sensitivity to climate change persistence compared with potential changes in impacts showed little concordance with current water quantity. Survey results suggest that fishes threatened status assigned by the Brazilian exhibit high sensitivity to changes in temperature government (Kruskal–Wallis, K = 0.64, P = 0.72, and dissolved oxygen, and moderate to high df = 2; Figure 3). Notably, the potential effects of sensitivity to changes in high-flow (i.e. flood) and low-flow (i.e. drought) regimes (Figure 5). Table 1. Concordance in survey responses as measured by inter-rate reliability analysis (Fleiss’ kappa) for each fish species receiving multiple entities. Question number corresponds to survey presented in Appendix S2, Supplementary material

Species kappa Entries Questions P

Baryancistrus niveatus 0.447 3 5 0.000 Hypancistrus zebra 0.346 4 20 0.000 Lebiasina marilynae 0.255 2 20 0.006 Lebiasina melanoguttata 0.451 2 23 0.000 Lebiasina minuta 0.444 2 22 0.000 Melanocharacidium nigrum 0.513 6 21 0.000 Mylesinus paucisquamatus 0.220 4 12 0.000 Ossubtus xinguense 0.244 5 20 0.000 Pimelodus halisodous 0.318 3 20 0.000 Pimelodus joannis 0.387 2 22 0.000 Pimelodus stewarti 0.363 3 20 0.000 Prochilodus britskii 0.302 2 19 0.010 Rhinopetitia potamorhachia 0.359 2 21 0.000 Rhynchodoras xingui 0.491 3 4 0.000 Sartor tucuruiense 0.601 5 4 0.000 Teleocichla cinderella 0.609 3 20 0.000 ﬁ Microglanis robustus 0.804 3 25 0.000 Figure 4. Number of freshwater sh species considered under threat from non-climate related factors by one and more than one impact.

Copyright # 2016 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 26 (Suppl. 1): 91–102 (2016) CLIMATE CHANGE IN AMAZONIAN FISHES 97

Fish species demonstrated considerable variability in their tolerances to climate change impacts (Figure 5). The overall sensitivity scores varied from 2.9 (Scobinancistrus pariolispos) to 4.8 (Sternarchorhynchus jaimei and Sternarchorhynchus kokraimoro) (Table 2). Sternarchorhynchus jaimei and S. kokraimoro are considered highly sensitive to changes in water quality and quantity parameters, whereas Sternarchorhynchus pariolispos showed the lowest overall sensitivity to climate change impacts, owing to moderate tolerance to changes in water quality and quantity (Table 2). Among the species with high overall sensitivity to climate change, ﬁve are categorized as critically endangered: Apteronotus lindalvae, Baryancistrus niveatus, Hypancistrus zebra, Microglanis robustus and Sternarchorhynchus higuchii. Another two species Leporinus pitingai and Crenicichla cyclostoma are highly sensitive to projected climate- induced changes in both water quality and quantity parameters, but demonstrate moderate to high dispersal ability thus reducing their overall sensitivity (Table 1). Survey experts reported considerable variability in ﬁsh dispersal ability, which was not associated with sensitivity to projected climate-induced changes in water quantity and quality (R =0.04,F=0.95,P = 0.33).

DISCUSSION

Approximately half the threatened fish species evaluated in the Brazilian Amazon are considered sensitive to climate change impacts according to the expert survey. Altered hydrological regimes have direct consequences for Amazonian fish biodiversity given that feeding and reproductive strategies of many species are dependent on flooding dynamics (Alho et al., 2015), and thus are a major determinant of fish community composition (Sousa and Freitas, 2008). In the past decade alone, the Amazon River basin has been subjected to two severe supra-seasonal droughts Figure 5. Overall sensitivity to climate change for freshwater fish (2005 and 2010) and one record flood in 2009 species according to factors describing water quantity and quality: (A) high flow, flood À duration, magnitude and timing (duration (Marengo et al., 2012). Increasing frequency of these and timing overlap); (B) low flow, drought À duration, magnitude extreme events will have important implications for and timing; (C) temperature and dissolved oxygen. Kernel density is the estimation of the statistical distribution based on the freshwater biodiversity in the Amazon. According to score values. Freitas et al. (2013) the environmental impacts

Copyright # 2016 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 26 (Suppl. 1): 91–102 (2016) 98 R. G. FREDERICO ET AL.

Table 2. Rank of sensitivity to climate change of 35 threatened ﬁsh species in the Brazilian Amazon. Sensitivity scores indicate only the tolerance to changes in water quantity and quality and overall sensitivity (water quality and quantity, and dispersal ability). The sensitivity rank is based on the overall sensitivity to climate change impacts. Threat of extinction categories ofﬁcially adopted by the Brazilian Government in December 2014

Family Species Threat category Sensitivity score Overall sensitivity Rank of sensitivity

Apteronotidae Sternarchorhynchus jaimei CR 4 4.75 1 Apteronotidae Sternarchorhynchus kokraimoro VU 4 4.75 1 Characidae Mylesinus paucisquamatus EN 4 4.67 2 Apteronotidae Apteronotus lindalvae CR 3.66 4.66 3 Loricariidae Baryancistrus niveatus CR 4 4.62 4 Characidae Brycon nattereri VU 4 4.50 5 Loricariidae Ancistrus cryptophthalmus EN 3.33 4.33 6 Loricariidae Hypancistrus zebra CR 3.33 4.33 6 Lebiasinidae Lebiasina melanoguttata VU 3.33 4.33 6 Lebiasinidae Lebiasina minuta VU 3.33 4.27 7 Prochilodontidae Prochilodus britskii EN 3.50 4.25 8 Crenuchidae Melanocharacidium nigrum EN 3.33 4.25 8 Characidae Brycon gouldingi EN 4 4.25 8 Characidae Rhinopetitia potamorhachia EN 3.48 4.23 9 Pimelodidae Pimelodus joannis VU 3.52 4.21 10 Lebiasinidae Lebiasina marilynae VU 3.33 4.20 11 Loricariidae Lamontichthys parakana CR 3.38 4.13 12 Pimelodidae Pimelodus stewarti VU 3.42 4.12 13 Pimelodidae Pimelodus halisodous VU 3.33 4.04 14 Pimelodidae Aguarunichthys tocantinsensis EN 4 4.00 15 Loricariidae Lamontichthys avacanoeiro EN 3.44 3.94 16 Anostomidae Sartor tucuruiense EN 3.21 3.93 17 Pseudopimelodidae Microglanis robustus CR 3.12 3.90 18 Characidae Ossubtus xinguense VU 3.22 3.90 19 Anostomidae Leporinus guttatus VU 3.36 3.86 20 Apteronotidae Sternarchorhynchus severii VU 3.08 3.83 21 Apteronotidae Sternarchorhynchus higuchii CR 2.91 3.66 22 Apteronotidae Sternarchogiton zuanoni VU 2.66 3.41 23 Apteronotidae Sternarchorhynchus villasboasi VU 2.62 3.37 24 Cichlidae Teleocichla cinderella EN 2.55 3.30 25 Doradidae Rhynchodoras xingui EN 2.44 3.17 26 Anostomidae Leporinus pitingai CR 2.66 3.16 27 Cichlidae Crenicichla cyclostoma CR 2.66 2.91 28 Apteronotidae Sternarchorhynchus mareikeae VU 2.16 2.91 28 Loricariidae Scobinancistrus pariolispos VU 2.33 2.83 29

CR – critically endangered, EN – endangered, VU - vulnerable associated with climate change could lead to the loss unexpected given that the primary reason for the of at least 7% of fish species. original expansion of PAs in the Amazon was to A major constraint to using PAs to protect slow the rate of deforestation, and by some estimates freshwater fishes is that riverine ecological integrity it has been successful in accomplishing this goal is subject to human disturbances that occur outside (Soares-Filho et al., 2010; Veríssimo et al., 2011). the boundaries. This is one of the main criticisms Land-cover changes in the Amazon River Basin of applying the terrestrial protected area approach have direct consequences for freshwater to freshwater ecosystems (Abell et al., 2007, 2011). ecosystems. The expansion of protected areas in Evidence from the USA suggests that the National the Brazilian Amazon between 2004 and 2006 led Park System, a network of reserves established for to a 37% reduction in deforestation rates (Soares- the conservation of terrestrial features, adequately Filho et al. 2010). These are favourable trends represents freshwater fish diversity (Lawrence given that deforestation has been linked to reduced et al., 2011). However, this is unlikely to be the precipitation, and in some catchments such as the case in the Amazon. In fact, the majority of Tocantins and Araguaia Rivers, deforestation has Amazonian threatened fish species are not well already altered the regional water balance (Coe represented (in terms of their range) in the present et al., 2009). In catchments where less than 25% of PA network (Figure 1, www.icmbio.br). This is not the land surface has been deforested, changes in

Copyright # 2016 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 26 (Suppl. 1): 91–102 (2016) CLIMATE CHANGE IN AMAZONIAN FISHES 99 regional hydrological cycles may be too small to be attributed these decreases to the redistribution of detected (Coe et al., 2009). some species to deeper waters (refugia) to avoid The results of this study indicate that nearly 80% the shallow, warmer and hypoxic waters and of the (evaluated) Amazonian threatened fish elevated rates of predation (Thomé-Souza and species of Brazil are considered highly sensitive to Chao, 2004). climate-induced changes in flood timing, Floods and droughts have direct influence on magnitude and duration. Many larger-bodied fish water quality, and the results described here show species demonstrate a periodic reproductive that the majority of species demonstrate high strategy where recruitment occurs during the early sensitivity to climate-induced changes to flood season when wetland and off-channel temperature and dissolved oxygen. The water habitats enlarge, become enriched with nutrients, temperature of Amazonian streams and rivers and primary and secondary productivity spike ranges between 24 °C and 32 °C (Barletta et al., (Beesley et al., 2012; Alho et al., 2015). Enhanced 2010; pers. obs.), and the regional fish fauna is lateral connectivity during these times facilitates well adapted to these high temperatures. Tropical the dispersal of native species, but at the same species, however, are stenotherms, having a time may promote the spread of non-native species smaller required temperature range compared with (Leigh et al., 2014; Reich and Lake, 2015). most temperate fishes; thus they have low Hydrological changes during the wet season may tolerance to extremes in temperature. Low especially affect species with seasonal reproductive concentrations of dissolved oxygen occur in the strategies by causing a mismatch between flood Amazon basin (Barletta et al., 2010), and many dynamics and reproductive activity and cues species have evolutionary and behavioural (Olden, 2016). adaptations to cope with these conditions (Brauner The survey showed that most species (nearly and Val, 2006). However, strong and persistent 90%) are considered highly sensitive to climate- oxygen depletion may be more common in the induced changes in hydrological drought regimes. future, and could lead to chronic stress for many Extreme drought has a number of well-recognized fish species. According to Loarie et al. (2009), due consequences for freshwater fishes, including the to the size of protected areas in tropical and fragmentation of floodplain lakes from rivers, subtropical moist broadleaf forests, the rate of decreased lake surface area, and the concentration climate change impacts could be lower when of fish in confined refugia leading to increased compared with other regions of the world. Thus, competition and predation (Magoulick and Amazonian fish species could be sheltered, at least Kobza, 2003; Alho et al., 2015). For example, the in part, if the Brazilian government maintains the 2005 extreme drought in the Brazilian Amazon led current network of PAs in the Amazon. However, to early and severe disconnection among lakes and it is still unclear whether protected areas afford rivers, and many species were trapped in shrinking protection to freshwater fishes, particularly in the lakes, later dying due to hypoxia, extreme water future when fishes will be shifting their temperatures or predation (Tomasella et al., 2013). distributions in response to climate change. Freitas et al. (2013) observed that the beta Unfortunately, the Brazilian government diversity in six floodplain lakes along the Solimões continues to fall short in the protection of River decreased after the extreme drought in the biodiversity (Ferreira et al., 2014; Loyola, 2014). Amazon in 2005. They concluded that these Since 2008, around 3.4% of Brazilian conservation extreme events could lead to both local and units were reduced in size or degazetted, and in the regional loss of fish, especially those species in low Amazon nearly 42% of PAs were downgraded, abundance and demonstrating niche specialization. downsized or degazetted (Bernard et al., 2014). During an intense drought in 1997, Thomé-Souza The loss of protected areas often paves the way for and Chao (2004) reported a significant decrease in additional hydropower and mining developments the biomass and richness of benthic fish fauna in (Bernard et al., 2014; Ferreira et al., 2014). the Negro and Branco rivers. The authors Moreover, developing Amazonian countries are

Copyright # 2016 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 26 (Suppl. 1): 91–102 (2016) 100 R. G. FREDERICO ET AL. experiencing high rates of dam construction (Finer respond (Jaeger et al.,2014).Therefore,better and Jenkins, 2012), constituting one of the most knowledge is needed of which areas are the most serious threats to the region (Castello and important to protect to ensure that catchment Macedo, 2016). functionality is maintained (Pittock et al. 2008). This The present and future impacts from climate is especially true in areas, such as the Brazilian change cannot be understated. A considerable Amazon, where the majority of the PAs were created mismatch was found between the conservation for the purposes of conserving terrestrial assets. status of a species and its sensitivity to climate change – a phenomenon also reported for fishes in California by Moyle et al. (2013). This suggests ACKNOWLEDGEMENTS that perceptions of present-day extinction risk do not necessarily provide insight into future risks We extend our sincere appreciation to all the associated with climate change. Taking into experts who participated in the survey, and Lauren account the current and future (climate change) Kuehne for help with formatting the survey in impacts in the Amazon basin and the ecological Qualtrics. This manuscript benefited from the characteristics such as small range size and low detailed comments from three anonymous dispersal ability, it is possible to predict those reviewers. Funding was provided by the Brazilian species most likely to face elevated risk of Council of Science and Technology (Conselho extinction (Table 2). Nacional de Desenvolvimento Científico e Evidence suggests that dispersal ability is a Tecnológico - CNPq) - Science Without Borders fi primary predictor of how the ranges of sh species to RGF, CNPq and FAPEAM (Fundação de have shifted in response to past climate change Amparo à Pesquisa do Estado do Amazonas) to (Heino et al., 2009; Comte et al., 2014). For JZ, and the H. Mason Keeler Endowed example, Comte et al. (2014) found that high Professorship (School of Aquatic and Fishery fi dispersal was a critical attribute enabling sh to Sciences, University of Washington) to JDO. JZ colonize higher altitudes in order to track suitable received a productivity grant from CNPq (process thermal conditions. The analysis here revealed that 313183/2014-7). fish species identified as highly sensitive to climate change (specifically, alterations in water quantity and quality) demonstrated variable dispersal REFERENCES abilities. This suggests that regardless of sensitivity, species may vary considerably in their ability to Abell R, Allan JD, Lehner B. 2007. Unlocking the potential of protected areas for freshwaters. Biological Conservation 134: track changes in the climatic conditions. Indeed, 48–63. species dispersal ability is considered paramount in Abell R, Thieme M, Ricketts TH, Olwero N, Ng R, Petry P, the ultimate effectiveness of PA networks to Dinerstein E, Revenga C, Hoekstra J. 2011. Concordance conserve species (Hannah et al., 2007; Hannah, of freshwater and terrestrial biodiversity. Conservation Letters 4: 127–136. 2008). This is particularly true in freshwater Alho CJR, Reis RE, Aquino PPU. 2015. Amazonian ecosystems where fish are dispersing through freshwater habitats experiencing environmental and dendritic channel networks that drain upstream socioeconomic threats affecting subsistence fisheries. Ambio 44: 412–425. catchments and connect critical habitats downstream Baigún CRM, Colautti D, López HL, Van Damme PA, Reis (Abell et al., 2007). Strategies that incorporate RE. 2012. Application of extinction risk and conservation longitudinal connectivity in conservation, including criteria for assessing fish species in the lower La Plata River basin, South America. Aquatic Conservation: Marine and the management of current PAs and the Freshwater Ecosystems 22: 181–197. identification of future ones, are critical (Hermoso Barletta M, Jaureguizar AJ, Baigun C, Fontoura NF. 2010. et al., 2011). This challenge is especially pressing Fish and aquatic habitat conservation in South America: a continental overview with emphasis on neotropical systems. given that climate change is likely to reduce 76 – fi Journal of Fish Biology : 2118 2176. hydrological connectivity within rivers and sh Beesley L, King AJ, Amtstaetter F, Koehn JD, Gawne B, Price dispersal may ultimately determine their ability to A, Nielsen DL, Vilizzi L, Meredith S. 2012. Does flooding

Copyright # 2016 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 26 (Suppl. 1): 91–102 (2016) CLIMATE CHANGE IN AMAZONIAN FISHES 101

affect spatiotemporal variation of fish assemblages in temperate Gardali T, Seavy NE, Digaudio RT, Comrack LA. 2012. A floodplain wetlands? Freshwater Biology 57:2230–2246. climate change vulnerability assessment of California’s at- Bernard E, Penna LAO, Araújo E. 2014. Downgrading, risk birds. PloS One 7: e29507 downsizing, degazettement, and reclassification of protected Hannah L. 2008. Protected areas and climate change. Annals of areas in Brazil. Conservation Biology 28: 939–950. the New York Academy of Sciences 1134: 201–212. Brauner CJ, Val AL. 2006. Oxygen transfer. In The Physiology Hannah L, Midgley G, Andelman S, Araújo M, Hughes G, of Tropical Fishes, Val AL, Almeida-Val VMF, de Randall , Martinez-Meyer E, Pearson R, Williams P. 2007. Protected DJ (eds). Academic Press: London; 277–306. areas needs in changing climate. Frontiers in Ecology and Brodie J, Post E, Laurance WF. 2012. Climate change and the Environment 5: 131–138. tropical biodiversity: a new focus. Trends in Ecology & Haylock MR, Peterson TC, Alves LM, Ambrizzi T, Anunciação Evolution 27: 145–150. YMT, Baez J, Barros VR, Berlato MA, Bidegain M, Coronel G, Bunn SE, Arthington AH. 2002. Basic principles and ecological et al. 2006. Trends in total and extreme South American consequences of altered flow regimes for aquatic biodiversity. rainfall in 1960–2000 and links with sea surface temperature. Environmental Management 30: 492–507. Journal of Climate 19: 1490–1512. Bush AA, Nipperess DA, Duursma DE, Theischinger G, Heino J, Virkkala R, Toivonen H. 2009. Climate change and Turak E, Hughes L. 2014. Continental-scale assessment of freshwater biodiversity: detected patterns, future trends and risk to the Australian Odonata from climate change. PloS adaptations in northern regions. Biological Reviews 84: One 9: e88958 39–54. Case MJ, Lawler JJ, Tomasevic JA. 2015. Relative sensitivity Hermoso V, Linke S, Prenda J, Possingham HP. 2011. Addressing to climate change of species in northwestern North longitudinal connectivity in the systematic conservation America. Biological Conservation 187: 127–133. planning of fresh waters. Freshwater Biology 56:57–70. Castello L, Macedo MN. 2016. Large-scale degradation of Jaeger KL, Olden JD, Pelland NA. 2014. Climate change Amazonian freshwater ecosystems. Global Change Biology poised to threaten hydrologic connectivity and endemic 22: 990–1007. fishes in dryland streams. Proceedings of the National Castello L, McGrath DG, Hess LL, Coe MT, Lefebvre PA, Academy of Sciences 111: 13894–13899. Petry P, Macedo MN, Renó VF, Arantes CC. 2013. The Lawrence DJ, Larson ER, Reidy Liermann CA, Mims MC, vulnerability of Amazon freshwater ecosystems. Pool TK, Olden JD. 2011. National parks as protected Conservation Letters 6:1–13. areas for US freshwater fish diversity. Conservation Letters Coe MT, Costa MH, Soares-Filho BS. 2009. The influence of 4: 364–371. historical and potential future deforestation on the stream Leigh C, Bush A, Harrison ET, Ho SS, Luke L, Rolls RJ, flow of the Amazon River: land surface processes and Ledger ME. 2014. Ecological effects of extreme climatic atmospheric feedbacks. Journal of Hydrology 369: 165–74. events on riverine ecosystems: insights from Australia. Comte L, Murienne J, Grenouillet G. 2014. Species traits and Freshwater Biology 60: 2620–2638. phylogenetic conservatism of climate-induced range shifts in Loarie RL, Duffy PB, Hamilton H, Asner GP, Field CB, stream fishes. Nature Communications 5:1–9. Ackerly DD. 2009. The velocity of climate change. Nature Conger AJ. 1980. Integration and generalization of Kappas for 462:24–31. multiple raters. Psychological Bulletin 88: 322–328. Loyola R. 2014. Brazil cannot risk its environmental Eaton JG, Scheller RM. 1996. Effects of climate warming on leadership. Diversity and Distributions 20: 1365–1367. fish thermal habitat in streams of the United States. Magoulick DD, Kobza RM. 2003. The role of refugia for fishes Limnology and Oceanography 41: 1109–1115. during drought: a review and synthesis. Freshwater Biology Ferreira J, Aragão LEOC, Barlow J, Barreto P, Berenguer E, 48: 1186–1198. Bustamante M, Gardner TA, Lees AC, Lima A, Louzada Malhi Y, Roberts JT, Betts AR, Killeen TJ, Li W, Nobre CA. J, et al. 2014. Brazil’s environmental leadership at risk. 2008. Climate change, deforestation, and fate of the Amazon. Science 346: 706–707. Science 319: 169–172. Ficke AD, Myrick CA, Hansen LJ. 2007. Potential impacts of Marengo JA, Tomasella J, Soares WR, Alves LM, Nobre CA. global climate change on freshwater fisheries. Reviews in Fish 2012. Extreme climatic events in the Amazon basin. Biology and Fisheries 17: 581–613. Theoretical and Applied Climatology 107:73–85. Finer M, Jenkins CN. 2012. Proliferation of hydroelectric dams Moyle PB, Katz JVE, Quiñones RM. 2011. Rapid decline of in the Andean Amazon and implications for Andes–Amazon California’s native inland fishes: a status assessment. connectivity. PloS One 7: e35126 Biological Conservation 144: 2414–2423. Foden WB, Butchart SHM, Stuart SN, Vié J-C, Akçakaya HR, Moyle PB, Kiernan JD, Crain PK, Quiñones RM. 2013. Angulo A, DeVantier LM, Gutsche A, Turak E, Cao L, et al. Climate change vulnerability of native and alien freshwater 2013. Identifying the world’s most climate change vulnerable fishes of California: a systematic assessment approach. PloS species: a systematic trait-based assessment of all birds, One 8: e63883 amphibians and corals. PloS One 8: e65427 Nogueira C, Buckup PA, Menezes NA, Oyakawa OT, Silva Freitas CEC, Siqueira-Souza FK, Humston R, Hurd LE. 2013. MC, Kasecker TP, Ramos MB. 2010. Restricted-range An initial assessment of drought sensitivity in Amazonian fish fishes and the conservation of Brazilian freshwaters. PloS communities. Hydrobiologia 705: 159–171. One 5: e11390 Galbraith H, Price J. 2009. A framework for categorizing the Olden JD. 2016. Challenges and opportunities for fish relative vulnerability of threatened and endangered species to conservation in dam-impacted waters. In Conservation of climate change, US Environmental Protection Agency: Freshwater Fishes, Closs GP, Krkosek M, Olden JD (eds). Washington, DC. Cambridge University Press: Cambridge; 107–148.

Copyright # 2016 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 26 (Suppl. 1): 91–102 (2016) 102 R. G. FREDERICO ET AL.

Olden JD, Kennard MJ, Leprieur F, Tedesco PA, Winemiller National Academy of Sciences of the United States of KO, García-Berthou E. 2010. Conservation biogeography America 107: 10821–10826. of freshwater fishes: recent progress and future challenges. Sousa RGC, Freitas CEDC. 2008. The influence of flood pulse Diversity and Distributions 16: 496–513. on fish communities of floodplain canals in the Middle Ota RR, Message HJ, da Graça WJ, Pavanelli CS. 2015. Solimões River, Brazil. Neotropical Ichthyology 6: 249–255. Neotropical Siluriformes as a model for insights on Thomé-Souza MJF, Chao NL. 2004. Spatial and temporal determining biodiversity of animal groups. PloS One 10: variation of benthic fish assemblages during the extreme e0132913 drought of 1997-98 (El Niño) in the middle rio Negro, Pacifici M, Foden WB, Visconti P, Watson JEM, Butchart Amazonia, Brazil. Neotropical Ichthyology 2: 127–136. SHM, Kovacs KM, Scheffers BR, Hole DG, Martin TG, Tomasella J, Pinho PF, Borma LS, Marengo JA, Nobre Akçakaya HR, et al. 2015. Assessing species vulnerability CA, Bittencourt ORFO, Prado MCR, Rodriguez DA, to climate change. Nature Climate Change 5: 215–225. Cuartas LA. 2013. The droughts of 1997 and 2005 in Pearson RG, Stanton JC, Shoemaker KT, Aiello-Lammens Amazonia: floodplain hydrology and its potential ME, Ersts PJ, Horning N, Fordham DA, Raxworthy CJ, ecological and human impacts. Climatic Change 116: Ryu HY, McNees J, et al. 2014. Life history and spatial 723–746. traits predict extinction risk due to climate change. Nature Veríssimo A, Rolla A, Souto Maior APC, Monteiro A, Climate Change 4: 217–221. Brito B, Souza C, Augusto CC, Cardoso D, Conrado D, Peres CA. 2005. Why we need megareserves in Amazonia. Araújo E, et al. 2011. Áreas protegidas na Amazônia Conservation Biology 19: 728–733. Brasileira: Avanços e Desafios, Belém e São Paulo: Pittock J, Hansen LJ, Abell R. 2008. Running dry: freshwater Imazon/ISA. biodiversity, protected areas and climate change. Williams SE, Shoo LP, Isaac JL, Hoffmann AA, Langham G. Bioanalysis 9:30–38. 2008. Towards an integrated framework for assessing the Poff NL, Brinson MM, Day JW. 2002. Aquatic Ecosystems and vulnerability of species to climate change. PLoS Biology 6: Global Climate Change: Potential Impacts on Inland e325 Freshwater and Coastal Wetland Ecosystems in the United Xenopoulos MA, Lodge DM, Alcamo J, Marker M, Schulze States. Pew Center on Global Climate, Change: Arlington, VA. K, Van Vuuren DP. 2005. Scenarios of freshwater fish R Core Team (2012). R: a language and environment for extinctions from climate change and water withdrawal. statistical computing. R Foundation for Statistical Global Change Biology 11: 1557–1564. Computing, Vienna, Austria. http://www.R-project.org/. Reich P, Lake PS. 2015. Extreme hydrological events and the ecological restoration of flowing waters. Freshwater Biology 60:1–14. Soares-Filho B, Moutinho P, Nepstad D, Andreson A, SUPPORTING INFORMATION Rodrigues H, Garcia R, Dietzsch L, Merry F, Bowman M, Hissa L, et al. 2010. Role of Brazilian Amazon protected Additional supporting information may be found in the areas in climate change mitigation. Proceedings of the online version of this article at the publisher’sweb-site.