Integrating
fishermen knowledge and scientific analysis to assess
changes in
fish diversity and food web structure
Roberto Rosa
a, Adriana R. Carvalho
b, Ronaldo Angelini
c,*aPrograma de Pos Graduaç~ao em Desenvolvimento e Meio Ambiente, Universidade Federal do Rio Grande do Norte e UFRN, Brazil
bDepartamento de Ecologia, Universidade Federal do Rio Grande do Nortee UFRN, BR 101, Campus Universitario, 59078-970 Natal, RN, Brazil cDepartamento de Engenharia Civil, Universidade Federal do Rio Grande do Nortee UFRN, BR 101, Campus Universitario, 59078-970 Natal, RN, Brazil
a r t i c l e i n f o
Article history: Received 19 March 2014 Received in revised form 12 August 2014 Accepted 11 October 2014 Available online 18 October 2014 Keywords: Fishermen-based knowledge Alien species Shifting baseline Ecopath model Stream crossings Brazil
a b s t r a c t
In this paper the Ecosystem Approach to Fisheries (EAF) and the Fishermen Knowledge Approach (FKA) were applied to understand changes in a tropical coastal lagoon. Then, thefishermen experience-based knowledge, literature information and data sampling were combined aiming to understand biodiversity loss and food web changing in theirfishing environment exposed to the introduction of an invasive fish species. Fishermen indicated the changes infish species composition and simplification in trophic web, confirming biological data. However they do not attributed changes to the species introduction (our ecological hypothesis previously assumed), but to the sequential construction of bridges that interrupted waterflow between the lagoon and the ocean (their environmental hypothesis). This assertion fitted to Ecopath model simulations outputs in which exotic species removal from the ecosystem does not re-coveryfish composition or the trophic web complexity. Fishermen recalled as larger as smaller species, the matching between food web described byfishermen and the food web from literature and the justification for biodiversity loss presented by fishermen, broadened the confidence on their role as experts. This approach merged the historical views offishermen and provided evidence on the value of complement scientific data with experts consulting.
© 2014 Elsevier Ltd. All rights reserved.
1. Introduction
Biodiversity loss directly affects ecosystem function (Caliman et al., 2010) and ecosystem processes complexity. This complexity might be described as the number of pathways and mechanisms that determine in which way an individual species or species in-teractions affect ecosystems (Caliman et al., 2013).
Infisheries ecosystems, the loss of fish diversity may, in part, be attributed to the damming of the rivers (Gubiani et al., 2010), ur-banization (Peressin and Cetra, 2014), overfishing (Huang et al., 2013), introduction of exotic species (Rahel, 2007; Vitule et al., 2009) and inappropriately constructed stream crossings (Gibson et al., 2005). Thus, there is major interest in understanding which factors determine fisheries ecosystem dynamics and what infor-mation should be taken to better designfisheries management.
Therefore the ecosystem approach tofisheries (EAF) has reached major importance in developing countries (Le Fur et al., 2011; Heymans et al., 2011) intensifying the focus in ecosystem based
fisheries management e EBFM to understand fisheries and ecosystem dynamics (Pikitch et al., 2004; Christie et al., 2007; Marasco et al., 2007). Both approaches (EAF and EBFM) are devel-oped under the collection of large amount of ecological and social information (Rice, 2008), mainly when aiming ecosystem modeling to drivefishery management (Coll and Libralato, 2012).
In developing countries, the data requirement has been chal-lenged by lack of resources, large dimensions and unregulated development (Bene et al., 2010). However, fishermen-based knowledge has emerged in science as a useful source of data to marine and freshwater environments (Carvalho et al., 2009; Hill et al., 2010; Zappes et al., 2013). The assumption is that resources users are experts on the natural assets and on environment exploited. Hence the opinion from experts has been incorporate to science and data gathering by different means (Cooke, 1991).
Mainly after the seminal studies carried out byJohannes (1981), Sarda and Maynou (1998)andPoizat and Baran (1997)fishermen empirical and ecological knowledge have been accepted as a valuable data assessment under many viewpoints. Further proper assessments of the resource and resource users have a crucial role in successful conservation (Tesfamichael et al., 2014; Hallwass et al., 2013).
* Corresponding author.
E-mail addresses: r_rosa_prodema@hotmail.com (R. Rosa), acarvalho.ufrn@ gmail.com(A.R. Carvalho),ronangelini@gmail.com(R. Angelini).
Contents lists available atScienceDirect
Ocean & Coastal Management
j o u rn a l h o m e p a g e :w w w . e ls e v i e r . c o m / l o c a t e / o c e c o a m a n
http://dx.doi.org/10.1016/j.ocecoaman.2014.10.004
Underfisheries perspective, fishermen have a detailed knowl-edge about fish species and their environments (Morril, 1967; Silvano and Jorgensen, 2008; Silvano and Begossi, 2012) and they are able to provide information regarding to fisheries change, ecological aspects offish and fisheries as well as on the environ-ment. This Local Ecological Knowledgee LEK (Johannes, 1998) or Fisher's Ecological Knowledge (Leite and Gasalla, 2013) has long been the basis for the incorporation of fishermen as experts in fisheries studies. Recently scientific data and fishermen-based knowledge is applied to understand ecosystem process (Garcia and Cochrane, 2005; Le Fur et al., 2011) integrating Fishermen Knowledge Approache FKA into fisheries studies and enhancing fisheries science in many aspects.
Despite management decisions still lack the use of ecological knowledge fromfishermen under ecosystem approach to fisheries (Leite and Gasalla, 2013), some of the ineffective results of biology-orientated mainstreamfishery research methods (Holm, 2003) and the costly acquisition of data around large spatial and temporal scales (Johannes, 1998; Hilborn, 2011) has increased the interest on fishermen knowledge for fisheries management purposes (Lopes et al., 2013). Thus, the question is no longer if FKA should be used in fisheries researches but how biological data and information from fishermen complement each other aiming to understand ecosystem or species changes and to informfisheries management under EAF.
The present paper explores EAF integratingfishermen knowl-edge and biological data sampling to understand the effect of an invasivefish species introduction on biodiversity loss and on the food web changing in a coastal lagoon through time. The hypoth-eses verified were: (1) fishermen historical observation on fish diversity and food web structure will confirm the introduction of peacock bass (an invasive species) as the main cause for the alter-ation in the ecosystem as a whole (ecological hypothesis); (2) outcomes from Ecopath model simulations for the invasive fish species will be supported by fishermen experienced-based knowledge; and (3) as fishermen information as scientific data (from earlier literature and recently samples) will indicate similar past and current food web structure.
The predictions assumed both sources of data as complemen-tary since fishermen is expected to report effect of competitive exclusion from peacock bass introduction and modeling is expected to indicate the top down control by the species as well as the food
web structure recovery after the invasive species elimination. Rather, is expected that past and current food web structure indi-cate the matching between both sources of data.
2. Material and methods 2.1. Fishing community assessed
Currently not manyfishermen constitute the fishing community that exploit Extremoz Lagoon (Fig. 1) on both daily and weekly basis. In the past, around 60fishermen used to practice artisanal fishery in the lagoon. Nearby 35 fishermen are still settled in the area, but roughly half of them arefishing elsewhere, far away the urban area. The remainingfishermen are currently not fishing at all. Occasionally somefishermen exploit fishing resources in small groups or in pairs (with family members) for subsistence. Fishing sites are chosen according to the proximity of their homes, avoiding fishing places that were exploited in the last days. Fishing gears used in Extremoz are castnet, gillnets, dipnet, standing net and harpoons. The last one is applied for catch peacock bass (Cichla kelberi), an alien and top-predator species introduced in the lagoon. This lagoon is located in the Neotropical freshwater ecoregion in the urban area of Natal City. Local climate is tropical with dry season (Kottek et al., 2006) and the average temperature is 26.6C. The only intensive use currently taking place in Extremoz Lagoon relies in provide water intake to 70% of Natal citizens (Fig. 1).
2.2. Data sampling 2.2.1. Fishermen
The survey for gather current and past information on the status of fishery was carried out during 20 non-consecutive days of interviewing. During thefirst interviews every fisherman met was approached. Therefore thefirst four fishermen met had not more that 20 years offishing experience. Following the indication from these fishermen (snow ball method from Goodman, 1961), in-terviewers approached mainly only fishermen with at least 40 years of exclusive dedication to local fishery. The idea under assumption assumedfishing experience as a proxy to fishermen knowledge on the lake. Before interviews, each fisherman was informed on the purpose of the study.
Main questions regarded to 1) Whichfish species were present in Extremoz Lagoon 15 years ago; Which species are present nowadays; 2) When species disappeared and why; 3) If after spe-cies disappearance, new spespe-cies appeared in the lagoon and which ones; 4) When the peacock bass was introduced; 5) If following the fish species introduction, fishermen noticed changes on fish species composition. This question enabledfishermen to associate species elimination to the peacock bass; 6) Iffishery in Extremoz was more profitable in the past; and 7) If fishing activity is better or worse than 20 years ago and why. All questions referred to all species, regardless if commercially exploited or not.
Even thoughfishermen tend to remember mostly the larger or commercially appreciated species (Bender et al., 2013), each fish-ermen was asked to quote thefish species present in the lagoon in the past as far as he remembered. Further, a list with pictures of 35 fish species (previously recorded byVieira, 2002) was shown and eachfisherman pointed which species were present in the lagoon currently and in the past.
2.2.2. Fish sampling
Aiming to assess the current species composition in Extremoz Lagoon, twofish samples were taken in the dry season (March and September, 2012) and two fish samples were taken in the rainy season (April and June, 2012) atfive locations. Fishes were collected by a set of gillnets with different mesh sizes following those currently being used by localfishermen (3.5 cm, 4 cm and 5 cm, opposite knots). Gillnets were placed at 5 p.m. and verified at 12 p.m. and at 6 a.m., in order to capture species from nocturnal and diurnal behavior. Fishing effort was the same in all sampling sites and during all sample period. Localfishermen assisted the research group during all samplings.
During dry season one sample was also performed in two sites in River Doce, which connects Extremoz lagoon to the sea. Further, one sample was performed at three sites in Guarijú River and Mudos River. These samples were performed using the same fish-ing gears and were carried out durfish-ing periods of seasonal migration of saltwater fishes (e.g. genus Mugil,Lebreton et al., 2011) and aimed sample other species that could colonize the lagoon.
The stomach of all specimens collected was removed and pre-served in solution of ethanol 70% to perform the analysis of gut contents. Diet items were identified up to species level (if fish) and at order level (if other contents). Gut contents were quantified by the method of frequency of occurrence (Hynes, 1950; Bowen, 1992). 2.3. Data analysis
Analyses performed on past and currentfish biodiversity fol-lowed the main trends reported byfishermen. Accordingly, aiming to meet as much consensus as possible, only data cited by more than 40% of thefishermen were included (as suggestedSilvano and Jorgensen, 2008).
Using Mann Whitney U test the maximum length of species remembered byfishermen was compared to the maximum length of those species not recalled but recorded byVieira (2002)and/or
Starks (1913), which are the only reports onfish species composi-tion available to Extremoz Lagoon. Fishbase (Froese and Pauly, 2013) supplied information on maximum size for each species.
To verify the similarity in the presence/absence forfish species composition from different sources, Jaccard index was used to compare: a) the currentfish community (from samples); b) past fish species composition reported by fishermen; c) current fish species composition reported by fishermen; d) fish composition reported by Vieira (2002) and e) fish composition according to
Starks (1913).
Past and current food web resulting from biological data and fromfishermen knowledge were elaborated using Cytoscape soft-ware. Three energetic metrics related to energy paths were extracted to compare food webs: Shortest paths (i.e. the length of the shortest path between two nodes,Watts and Strogatz, 1998); Characteristic path length (i.e. the expected distance between two connected nodes) and Average number of neighbors that it in-dicates the average connectivity of a node into the network.
As not manyfishermen were willing to provide information on the diet offish, trophic connection to the less abundant species was assembled from the literature and from FishBase (Froese and Pauly, 2013). However, fishermen knowledge provided information on fish composition for current food web. Fish composition for the past food web followedVieira (2002).
An Ecopath model (see below) was elaborated with current sampled data in order to estimate some global properties (met-rics) of food web (Odum, 1969), which permit to understand the state of development of the Extremoz Lagoon: 1. Total Primary Production/Total Respiration (TPP/TR), that indicates higher maturity if close to one; 2. Overall overhead, which measures ecosystem resilience (Christensen, 1995), and may indicate the group responsible for the system stability; 3. Transfer Efficiency (%) among trophic levels; and 4. Ratio between production in detritus-based food chain and the production in grazing-based food chain. Ecopath model also aimed to assess the Key-stoneness species index (Libralato et al., 2006), which identified the key species of the Extremoz Lagoon.
Following the model balancing, an initialfishing pressure cor-responding to 10% of biomass for C. kelberi and Hoplias malabaricus was modeled aiming to represent fishery upon these two top predators. Subsequently, simulations were modeled increasing the exploitation on these species up to their elimination in order to assess the performance of the otherfish species under the scenario of C. kelberi and H. malabaricus exclusion, mainly the former, which is an alien Amazonian species introduced in the lagoon. Initial fishing mortality was multiplied by a factor of 10 during six years for all simulations. For details on modeling see below.
2.4. The Ecopath model
The Ecopath with Ecosim model (EwE, version 6.0) is based on the mass-balance assumption (Christensen and Pauly, 1992) as follows: Bi P=Bi EEi Sj Bj Q Bj DCji EXi¼ 0 (1)
Where Biis the biomass of group i; P/Biis the production/biomass rate of i and is equal to the total mortality Z or natural mortalitye M (Allen, 1971); EEiis the ecotropic efficiency of i, which varies from 0 to 1 and measures the fraction of the production by group that is transferred to higher TLs or exported from the system; Bjis the biomass of predator j; Q/Bjis the food consumption per unit of biomass for predator j; DCjiis the fraction of i in the diet of j; and EXi is the export of i and represents the biomass that is caught through fishing and/or that migrates to other ecosystems. The biomasses were expressed as tww km2 (metric tons of wet weight per square kilometer), and theflows in the food web were expressed in tww km2 year1.
For an ecosystem with n groups (compartments), the model will have a system of n linear equations. In the development of an EwE model, at least three of the four main input parameters (Bi, P/Bi, Q/ Bi, and EEi) has to be informed to allow the software to estimate the missing parameter under the assumption that the production of one group is consumed by another group within the system (Christensen and Pauly, 1992).
Extremoz Lagoon Ecopath model had six compartments comprised byfish species caught in our samplings (seeResults), plus: phytoplankton, zooplankton, macrophytes, insects and shrimps. To Extremoz Lagoon modeling, the values of B for mac-rophytes, insects and shrimps were estimated from values of EE, P/B and Q/B obtained from Brazilian freshwater ecosystems that shared species with Extremoz Lagoon (found inAngelini and Agostinho, 2005a, 2005b; Angelini et al., 2006a, 2006b and Angelini et al., 2013). Biomass values for Phytoplankton and Zooplankton com-partments were estimated by monthly samples (June 2012eJune 2013) performed in two sampling points at surface layer (between 0 and 0.5 m) of the lagoon. EwE estimated Biomass values forfish compartments that were sampled using gillnets, except for C. kelberi which was caught using trawl nets.
The Production/Biomass (P/B) was calculated using the Pauly (1980)equation.Palomares and Pauly (1998)equation was used to determine the Consumption/Biomass ratio (Q/B). Equations pa-rameters were obtained from studies on tropicalfish (Gubiani et al., 2012; Angelini and Agostinho, 2005a) or from FishBase (Froese and Pauly, 2013). In order to obtain the mass balance, diet composition matrix in EwE was modified using prey items sampled.
2.5. Keystoneness species index estimate and model simulations Keystoneness species index (KSi) (Libralato et al., 2006) is per-formed, using the Mixed Trophic Impact (MTI) matrix, which pro-vides the direct and indirect interactions between the species in a food web, as follows:
KSi¼ log½eið1 piÞ (2)
where eiis equal to√Pmij2(mijis calculated based on the MTI and represents the interaction between the impacting group i and the impacted group j); therefore, eiis a measure of the effects of each group on all other groups in the food web (including indirect ef-fects). The effect of the change on the group biomass itself (i.e., mii) is not included; piis equal to (Bi/total B), where Biis the biomass of the impacted group and total B is the total biomass (excluding detritus).
The tropho dynamic module of EwE (i.e., Ecosim) uses settings from the mass-balance module (i.e., Ecopath) as the initial condi-tions and parameter definitions (Christensen and Walters, 2004). Namely, the system equation (Eq.(1)) is transformed into a system of ordinary differential equation as follows (Walters et al., 1997):
dBi=dt ¼ gi SjQjiþ SjQijþ Ii ðMOiþ Fiþ eiÞ Bi (3)
where dBi/dt is the change in B of group I over time t, giis the net growth efficiency, Qjiis the consumption of group j by group i, n is the number of prey groups, Qijis the consumption of group i by group j, m is the number of predator groups, Iiis the extent of immigration of group i, MOiis the non-predation rate of natural mortality of group i, Fiis thefishing mortality of group i, and eiis the emigration of group i.
3. Results
3.1. Past and current species richness and food web structure Current observation indicated that 35fishermen from the latter community still remain living in the area. Some of them arefishing in Estivas beach at 40 km from the lagoon and others are exploiting others coastal areas (including lagoons). Nearby 20 fishermen remained gillnetting in the lagoon in an irregular basis (just one or twice a week) mostly for consumption or for recreation through
diving and harpooning. From those, seventeenfishermen at mean age of 50.6 years were interviewed.
Allfishermen consider that artisanal fishing is under collapse in Extremoz Lagoon and that localfishery is no longer profitable, since the weekly average catch reaches onlyfive kilograms. Fishermen claimed that the amount caught is enough to family consumption only. Most fishermen (80%) have more than five children, and monthly income is below US$340.00.
According tofishermen knowledge species richness was 22 fish species in the past (last 10 years before 2000) and only 13 of these are still being caught currently (Table 1). Fishermen knowledge endorsed the presence offive marine species in the past and the current absence of marine species in the lagoon. Earlier samplings indicated richness of 35 species, being four marine (Vieira, 2002) while historical samplings fromStarks (1913)reported the occur-rence of 13 species with the presence offive marine fish species (Table 1).
Current samples also indicated the presence of six species in the lagoon and 13 in the rivers, but justfive current riverine species were common to data reported in 1913. Marine species were not caught and just one estuarine fish species (Eleotri spisonis) was currently sampled in Doce River, which connects the lagoon to the sea (Fig. 1).
Mostlyfishermen (70%) quoted the mullet (Mugil curema) as an example of marinefish species that was usually caught in the past but not been currently caught or observed in the local. According to 60% offishermen, mullet used to reach the lagoon by migrating through Doce River. Besides, following most of the fishermen opinions, the peacock bass has been introduced before 2000.
Highest similarities were found between current fish species composition sampled and present fish species composition re-ported byfishermen (0.63) and between fish species composition reported by Vieira (2002)and the past fish species composition recalled byfishermen (0.6;Table 2).
Past food web (Fig. 2a) and current food web (Fig. 2b) elaborated following thefishermen information were much simpler than the food web elaborated with data from Vieira (2002; Fig. 2c). The simplest food web however resulted from current biological sam-plings (Fig. 2d). These differences are reflected in food web metrics revealing more paths and higher path lengths in older food webs (Table 3).
Additionally, allfishermen indicated that fish species diversity has decreased in the lagoon. Many fishermen (76%) pointed the construction of sequential bridges between Doce River and the lagoon (Fig. 1) as the main reason for the loss offish diversity in the lagoon. Accordingly to them, these bridges and the underneath culverts (Fig. 3) represented physical barriers hampering fish migration from the estuaries into the Extremoz Lagoon. However, fourfishermen, specifically those with almost 20 years of fishing experience, supposed that peacock bass introduction was the main reason for the loss of biodiversity in the lagoon.
The introduction of peacock bass was reported by 40% of fish-ermen as the second main reason for biodiversity loss in the lagoon. Furthermore, peacock bass was associated to the trophic web un-balance by 76% of the fishermen because it “eats what it sees ahead”.
Actually the diet of peacock bass was mainly piscivorous (45%) but also omnivorous and opportunistic (Table 4) likewise H. malabaricus that feeds on insects (30%) andfish (Table 4). Other piscivorous fed on shrimps and bivalve (like Crenicichla menezesi) or exclusively on shrimps (like Serrasalmus orientale;Table 4).
The maximum lengths of species reminded byfishermen were not different of maximum lengths of species do not quoted by them (ManneWhitney U ¼ 123.5, p ¼ 0.219). Around 20% of fishermen
did not recognize some species shown in the list and were unable to report if the species existed both in the past and in the present days. 3.2. Ecopath model (EwE)
The basic results of Ecopath model for Extremoz Lagoon are showed inTable 5. Diet composition matrix in EwE (Table 6) was modified usingTable 4, in order to obtain the mass balance. Ecopath model indicated that C. kelberi and H. malabaricus have the highest trophic level. Despite only C. kelberi had many path number among fish predators, shrimp was the compartment showing the largest path number (Table 5).
Ecopath metrics indicated that Extremoz Lagoon is a developed system, but with moderate resilience given that Total Primary Production/Total Respiration (PPT/RT) was 1.23. Transfer Efficiency among trophic levels was 8.3% and overall overhead was 62%. Additionally aggregation of TLs showed that detritus-based food chain is 5.84 times higher than the grazing food chain.
Keystoneness species index (KSi) pointed the main top preda-tors (H. malabaricus, KSi¼ 0.043 and C. kelberi, KSi ¼ 0.135) as the key species in the lagoon. Also, total overhead for the system (62%) is represented mainly by Shrimp (25.5%) and Detritus (24.6%), indicating that in addition to the top-down process (influenced by the top predators) the system also tend to be controlled by wasp-waist and bottom up mechanisms.
3.3. Simulations in EwE
Model simulations revealed that an increasing offishing pres-sure upon both key species (H. malabaricus or C. kelberi) by a factor of 10 would raise the biomass of all otherfish species (Fig. 4a). Similarly the elimination of only H. malabaricus will produce the increasing of biomass of all other species (Fig. 4b). However, if only C. kelberi species were removed from the system (Fig. 4c and d), increasing on biomass was observed just for otherfish predators like H. malabaricus and S. orientale. Shrimp, which was the most
Table 1
Fish species in Extremoz Lake (Brazil). max L (cm): maximum length (byFroese and Pauly, 2013); Our sampling: X: species captured in Extremoz Lake;C: species captured in the rivers; LEK: Local Ecological Knowledge (X: referred to by one thirdfishermen; XX referred by 60% of fishermen; XXX: referred by 70% or more fishermen).
No Species Original habitat Max L (cm) Our sampling (2012e2013)
Vieira (2002) Starks (1913) LEK (before 2000)
LEK (current) 1 Astronotus ocellatus Agassiz, 1831 Freshwater 40 X XX
2 Astyanax bimaculatus Linnaeus Freshwater 17.5 C X X XXX X 3 Awaous tajasica Lichtenstein, 1822 Marine 16.3 X XX
4 Characidium bimaculatum Fowler, 1941 Freshwater 3.2 C X XXX XX 5 Centropomus mexicanus Bocourt, 1868 Marine 47.5 X
6 Cichlasoma bimaculatum Linnaeus, 1758 Freshwater 12.3 C X X XX 7 Cichlasoma orientale Kullander, 1983 Freshwater 13.6
8 Crenicichla lepidota Heckel, 1840 Freshwater 18 X X X XXX 9 Crenicichla menezesi Ploeg, 1991 Freshwater 14.6 C
10 Cichla kelberi Kullander and Ferreira 2006 (Bloch and Schneider) cf.
Freshwater 27.6 X X XXX
11 Curimatella dorsalis Eigenmann and Eigenmann, 1889
Marine 11.4 X
12 Dormitator maculatus Bloch, 1972 Freshwater 70 X XX
13 Eleotris pisonis Gmelin, 1789 Freshwater 25 C X XXX XXX 14 Eucinostomus gula Quoy and Gaimard, 1824 Freshwater 23 XX 15 Erythrinus erythrinus Bloch and Schneider, 1801 Freshwater 20 X XXX 16 Gymnotus carapo Linnaeus, 1758 Freshwater 76 C X X XX 17 Gerres sp. Walbaum, 1792 Marine 30 X
18 Hemigrammus marginatus Ellis, 1911 Freshwater 2.6 C X
19 Hoplias malabaricus Bloch, 1794 Freshwater 55.2 X X X XXX 20 Hoplosternum littorale Hancock,1828 Marine 24 X
21 Hypostomus pusarumStarks, 1913 Freshwater 20.3 X XX 22 Leporinus maculatus Muller and Troschel, 1844 Freshwater 18 X X
23 Leporinus piau Fowler, 1941 Freshwater 33 XC X XXX 24 Limatulichthys griseus Fowler, 1941 Freshwater 18 X X
25 Megalops atlanticus Valenciennes, 1847 Marine 250 X XX
26 Metynnis maculatus Kner, 1860 Freshwater 18 X X X XX 27 Microphis brachyurus lineatus Kaup 1856 Freshwater 22 X X
28 Mugil liza Valenciennes, 1836 Marine 80 X XXX 29 Mugil curema Valenciennes, 1836 Marine 90 X XXX 30 Moenkhausia lepidura Kner, 1858 Freshwater 8.9 X X 31 Nannostomus beckfordi Gunther, 1872 Freshwater 6.5 X XX 32 Piaractus brachypomus Cuvier, 1818 Marine 88 XX 33 Pimelodella enochi Fowler, 1941 Freshwater 5.9 X XX 34 Plagioscion squamosissimus Heckel, 1840 Freshwater 80 X XX
35 Poecilia vivipara Bloch and Schneider 1801 Freshwater 4 C X XXX XX 36 Prochilodus brevis Steindachnner, 1874 Freshwater 27 X XX
37 Serrapinnus heterodon Eigenmann, 1915 Freshwater 4.1 X XX 38 Serrapinnus piaba Lütken, 1874 Freshwater 3.5 C X
39 Serrasalmus rhombeus Linnaeus, 1766 Freshwater 41.5
40 Serrasalmus orientale Kner, 1858 Freshwater 21 X X XXX 41 Steindachnerina notonota Miranda Ribeiro, 1937 Freshwater 9.8 C X
42 Synbranchus marmoratus Bloch, 1975 Freshwater 150 C X 43 Trachelyopterus galleatus Linnaeus, 1766 Freshwater 22 C X 44 Triportheus signatus Garman, 1890 Freshwater 15.8 X
45 Trinectes paulistanos Ribeiro, 1915 Freshwater X XX
frequent gut content for many species in the system (Table 4), would be slightly benefit by C. kelberi removal and slightly harmed by H. malabaricus elimination.
4. Discussion
This paper combined an ecosystem approach tofisheries e EAF (through modeling) with fishermen knowledge approach e FKA (through consulting by interviews) to present a quantitative method to understand changes in the environment and to test the
balance between biological data andfishermen information. In this case, resources users were assumed as experts on environment and resources exploited. Since World War II, expert opinion has com-plemented investigations and assisted decision makers on physics, math, psychology and anthropology assessments (Cooke, 1991). In the last years, experts consulting have spread to other areas and many papers have assumed natural resources users as reliable in-formants, mainly onfishery (e.g.Sarda and Maynou, 1998; Davis and Wagner, 2003; García-Allut et al., 2007; Johannes and Neis, 2007; Carvalho et al., 2009; Eddy et al., 2010; Tesfamichael et al., 2014). Accordingly, we focused on Fishermen Knowledge Approach (FKA) to assemble information aiming validate and
Table 2
Jaccard similarity index (presence/absence) for species richness in Extremoz Lake, based on our samplings (CS, Current Sampling), literature (Vieira, 2002; Starks, 1913), and Fishermen's Information for Past (before 2000) and currently.
Jaccard similarity index CS Vieira (2002) Starks (1913) Fishermen's information (past) Currently sampling (CS) 1 Vieira (2002) 0.40 1 Starks (1913) 0.20 0.20 1 Fishermen's information (Past) 0.14 0.60 0.10 1 Fishermen's information (Currently) 0.63 0.37 0.20 0.10
Fig. 2. Food webs of Extremoz Lagoon: a) Past food web elaborated according tofisher's information; b) Current food web elaborated according to fisher's information; c) Past food web elaborated on the basis ofVieira (2002)data; d) Current food web elaborated by Ecopath software using data currently sampled. Thefirst three food webs were developed in Cytoscape software.
Table 3
Food web's metrics for Extremoz Lake, based on our samplings, literature (Vieira, 2002), and Fishermen's information for Past (before 2000) and currently.
Food web's metrics Fishermen's information
Biological data
Past Currently Vieira (2002) Currently samples Shortest path 124 89 166 32 Path length 1.47 1.53 1.55 1.15 Av. no.of neighbor 4.26 3.66 3.27 4.1
complement biological data onfish community and fisheries in a tropical coastal lagoon. Experts indicated the loss of biodiversity, changes infish species composition and simplification in trophic web followed by the collapse offishery for commercial and sub-sistence purposes. Despite assumption from scientists on peacock bass causing those alterations (ecological hypothesis),fishermen indicated the sequential construction of bridges that interrupted waterflow between the lagoon and the ocean (environmental hy-pothesis) as the main cause for ecosystem changes. Environmental hypothesis envisaged by fishermen was supported by biological data and matched to the Ecopath with Ecosim (EwE) model simu-lationsfindings in which exotic species (C. kelberi) removal from the ecosystem did not recoveryfish composition or the trophic web complexity.
In many places of the world the construction of culverts has turned into serious obstacle for fish species migration (Gibson et al., 2005; Kemp and O'Hanley, 2010; Bourne et al., 2011; Rolls et al., 2011). These structures hamper the accumulation of enough water to allow fish passing, especially the larger ones (Gibson et al., 2005; Karling et al., 2013). Besides, multiple bar-riers (like bridges building) may have cumulative effects and may prevent migration (Padgham and Webb, 2010). In the case of the
Extremoz Lagoon, the four bridges in the Doce River were con-structed after filling the area underneath with ground and canalizing the water flowing from the lagoon to the river downstream. Then, the construction of the bridges created an elevation (ranging from 50 to 80 cm) that works as a dam. As a consequence the bridges changed the environment when were built and still have until the present the potential to alter the circulation patterns in the water column, to reduce water depth and interrupt the ecological connectivity between these environments.
This evidence, however, was contrary to thefirst expectation, given that piscivorous fish species, especially exotic, can deeply alter the aquatic trophic web (Ogutu-Ohwayo, 1993; Pelicice and Agostinho, 2009; Vitule et al., 2009; Richardson, 2011). Thus, the loss of connectivity between lagoon and sea seems to be the main reason for the marine species exclusion. C. kelberi however, surely have the potential to regulate growth and abundance of potential preys like as Astyanax bimaculatus and Leporinus piau especially in coastal lagoons (Menezes et al., 2012). Beyond the present lack of marine species in the lagoon, there was a reduction in the number of species within some trophic guilds (e.g. piscivores and omni-vores) and the exclusion of detritivores and herbivores.
Fig. 3. Underneath culverts from bridges built in Doce River (for bridges' location seeFig. 1).
Table 4
Prey Items in the diet offish species in Extremoz Lake (Sum of columns is equal to 1). n: number of analyzed stomachs.
Fish speciese item Macrophyt Insects Shrimps Gastropod Bivalves C. kelberi C. menezesi L. piau M. maculatus H. malabaricus S. orientale Detritus n Cichla kelberi 0.15 0.15 0.05 0.15 0.15 0.2 0.1 0.05 27 Crenicichla menezesi 0.85 0.15 5 Leporinus piau 0.45 0.55 1 Metynnis maculatus 0.45 0.55 1 Hoplias malabaricus 0.3 0.1 0.05 0.05 0.1 0.15 0.15 0.1 11 Serrasalmus orientale 1 7
These results underlined that experts opinion may, under certain circumstances, be a valuable source of data (Le Fur et al., 2011). Indeed querying fishermen as experts to assemble infor-mation is inexpensive (Turvey et al., 2009; Daw et al., 2012), plentiful and virtually inexhaustible survey. Further, fishermen consulting offer us the opportunity of hindsight important histor-ical facts do not recorded even in grey literature. Many studies that have receded in pastfisheries through memories from fishermen have documented noteworthy information for managers and sci-entists (e.g.Saenz-Arroyo et al., 2005; Bunce et al., 2008; Daw,
2010; Eddy et al., 2010; Tesfamichael et al., 2014).
However data assembled from experts likefishermen are not a source of rational consensus mainly because an expert's opinion is not the same of an expert's knowledge (Cooke, 1991). Specifically in thefishery activity the knowledge on the ecosystem is as local as resource dependent. Such expertize is gradually accumulated through time due to thefishing experience or by oral communi-cation through stories, proverbs or songs (Kurien, 1998). This makes fishermen knowledge highly dependent on fishing experience (in years) and on age. At the other side, the opinion results from general observation and may be driven by political interest or by remarkable events informed by popular means of communication, such as television or radio. As the use of empirical experts in fish-eries has spread throughfisheries assessments (Davis and Wagner, 2003; García-Allut et al., 2007; Gerhardinger et al., 2009) this distinction should be taken into account. Accordingly, the collection
of information from older and more experiencedfishermen (more than 40 years of fishing activity), despite the limited set of in-terviewees, allowed us to register historical and relevant informa-tion regarding to the effects of bridges upon the environment and in the localfishery. Conversely the indication of peacock bass as the main cause for species depletion provided by fourfishermen that havefished for less than 20 years was likely driven by the common sense on the effect of peacock bass in the environment or by the shifting baseline syndrome, which determines that younger fish-ermen fail in understanding the extent of human changes in the environment in the long term (Bunce et al., 2008). Thus the simultaneous use of FKA and biological data as complementary source of data enlarged results and as complemented as enhanced the information obtained by both sources.
Actually our main goal was not strictly uphold that FKA could be proven by scientific data. Indeed, the enhancing nature of fisher-men knowledge to science and fisheries management has been reported worldwide through several reports (seeAswani and Lauer, 2006; Berkes et al., 2007; Johannes and Neis, 2007; Tesfamichael et al., 2014). Ourfirst intent was by using a factual case, underline the increase in achievements tofishery assessments by using both source of data. The complementary focus between scientific data and FKA has been less evaluated heretofore with few exceptions (see for instancePoizat and Baran, 1997; García-Allut et al., 2007; Williams and Bax, 2007). In the present paper this focus provided the development of the environmental hypothesis byfishermen in substitution to the ecological hypothesis previously stated.
The peacock bass introduction (ecological hypothesis) was ex-pected to trigger changes infish diversity and food web structure due to this species regulating effect upon the growth of other species by top-down effect (Menezes et al., 2012). Fishermen con-fidence in bridges construction instead of peacock bass introduc-tion as the main reason for local fishery collapse and food web simplification was a surprising statement. The information on the effects of bridges on the ecosystem would not be envisaged orfind by any other informant mainly because there is not official report of those builds construction.
Further,fishermen reasoning on ecosystem vulnerability to the bridges construction, despite their awareness on voracity and invasive features of C. kelberi provided reasonable justification to outcomes from modeling and simulation and to apply both ap-proaches in a complementary basis. Afterfishermen consulting and their views of secondary role of peacock bass in the system, the evidence that trophic web is not top down controlled, but wasp-waist controlled (by shrimps) seemed reasonable. Even though biomass estimate by EwE model might be a limitation to this analysis, is important to emphasize that all information supporting the estimates was provided by current study on gut content. Be-sides, biomass of peacock bass was reliable, since biomass was estimated using data sampled by trawl net.
Even though no conceptual framework is available to test on validity offishermen information, the reliability on fishermen as informant onfishery and fishing ecosystem has been proved by different means (Ticheler et al., 1998; Maurstad et al., 2007; Carvalho et al., 2009). In this paper we presented an ecosystem approach tofisheries e EAF through modeling and the fishermen knowledge approache FKA (through consulting by interviews) as a quantitative method to test previous hypothesis and to test the balance between biological data andfishermen information. This approach used both sources of data as complementary bases and enhanced the usual assessment based in the association between literature and information fromfishermen. To the purpose of this study, the fact thatfishermen did not mention preferably larger species than smaller provided evidence on reliability of their memories. Further the coincidence between food web described by
Table 5
Basic inputs parameters and outputs (in brackets) for Ecopath of the Extremoz Lake. TL: Trophic Level (TL), B: Biomass (t km2); P/B: Production/Biomass (t km2year1); Q/B: Consumption/Biomass (t km2year1); EE: Ecotrophic Effi-ciency; Paths: Number of paths inputting on compartment.
Group name TL B P/B Q/B EE Paths 1 Phytoplankton 1.00 2.97 250.00 (0.127) 2 Cladocera 2.00 0.072 25.00 150.00 (0.001) (2) 3 Copepodo 2.00 0.827 35.00 150.00 (0.001) (2) 4 Macrophytes 1.00 (10.079) 4.00 0.7 (2) 5 Insectos 2.00 (0.418) 25.00 250.00 0.7 (1) 6 Shrimp 2.00 (7.347) 8.00 80.00 0.91 (17) 7 Cichla kelberi 3.37 1.860 1.50 8.54 (0.92) (12) 8 Hoplias malabaricus 3.14 (1.973) 0.90 5.67 0.99 (2) 9 Crenicichla lepidota 2.95 (0.906) 1.30 7.75 0.95 (3) 10 Leporinus piau 2.50 (1.111) 1.59 5.74 0.95 (3) 11 Metynnis maculatus 2.50 (1.193) 1.48 37.45 0.95 (1) 12 Serrasalmus orientale 3.00 (1.286) 1.30 10.48 0.95 13 Detritus 1.00 0.81 Table 6
Diet composition in Ecopath model for Extremoz Lake (Numbers from 2e12 refers to prey/predator numbered in thefirst column).
Prey/predator 2 3 5 6 7 8 9 10 11 12 1 Phytoplankton 0.7 0.7 2 Cladocera 3 Copepodo 4 Macrophytes 0.05 0.45 0.45 5 Insects 0.25 0.3 6 Shrimp 0.35 0.2 0.95 0.5 0.5 1 7 Cichla kelberi 0.15 8 Hoplias malabaricus 0.1 9 Crenicichla lepidota 0.1 10 Leporinus piau 0.15 11 Metynnis maculatus 0.15 12 Serrasalmus orientale 0.1 13 Detritus 0.3 0.3 0.8 1 0.05 0.1 0.05 0.05 0.05 14 Import 0.15 Total 1 1 1 1 1 1 1 1 1 1
fishermen and that reported in the literature [in this case,Vieira (2002)andStarks (1913)] and thefishermen justification for fish-ery collapse, broadened the confidence on their evaluation as experts.
Such differences endorse thatfishermen knowledge actually is subject to bias due to the influence of local facts, personal skills, time offishing practice, cultural behavior, fishing gears applied, individual preferences and ability to observe learn and remember (Johannes, 1998; Henry et al., 1994). But these features also un-derscore their ability in observing and report local events and may, under certain situations, be the only source of information. Besides, the traditional aspect offishing activity provides us the opportunity to gather memory report from experts in regards on ecosystem and fisheries change, and to understand the extent of ecological process in the course of the historical modifications.
5. Conclusion
The combination between ecosystem approach to fisheries e EAF andfishermen knowledge approach e FKA enhanced the usual method to assessfishermen knowledge, which is often based on the match between broad literature data and information from fishermen.
Fishermen assessed provide good historical information enabling us to assume the environmental hypothesis instead of the ecological hypothesis to explain local los offish diversity and the food web changes. Their understanding on the environment also provided reasonable explanation to modeling outcomes showing the wasp-waist control on the food web (by shrimps) instead of top down control (by peacock bass or other piscivorous).
Fishermen memories were highly valued evidence to change local scientists view on the ecosystem changes and were valuable information to combine with our usual quantitative tools. Acknowledgments
This paper is based on the Master's Degree dissertation offirst author under the supervision of R. Angelini in the Prodema Post-Graduate Program at Universidade Federal do Rio Grande do Norte (Brazil). The authors thank to CNPq (Edital Universal, proc. 476347/ 2010-6) and Capes (scholarship to thefirst author).
References
Allen, R.R., 1971. Relation between production and biomass. J. Fish. Res. Board Can. 28, 1573e1581.
Angelini, R., Agostinho, A.A., 2005a. Food web model of the Upper River Foodplain: description and aggregation effects. Ecol. Model. 181, 109e121.
Angelini, R., Agostinho, A.A., 2005b. Parameter Estimates for Fishes of the Upper Parana River Floodplain and Itaipu Reservoir (Brazil), vol. 28. NAGA, The World Fish Center Quarterly, Manila, pp. 53e57.
Angelini, R., Fabre, N.N., Silva Jr., U.L., 2006a. Trophic analysis and fishing simulation of the biggest Amazonian catfish. Afr. J. Agric. Res. 1151e1158.
Angelini, R., Gomes, L.C., Agostinho, A.A., 2006b. Modeling energyflow in a large Neotropical reservoir: a tool to evaluatefishing and stability. Neotrop. Ichthyol. 4 (2), 253e260.
Angelini, R., Morais, R.J., Catella, A.C., Resende, E.K., Libralato, S., 2013. Aquatic food webs of the oxbow lakes in the Pantanal: a new site forfisheries guaranteed by alternated control? Ecol. Model. 253, 82e96.
Aswani, S., Lauer, M., 2006. Benthic mapping using local aerial photo interpretation and resident taxa inventories for design marine protected areas. Environ. Cons. 33 (3), 263e273.
Bender, M.G., Floeter, S.R., Mayer, F.P., Vila-Nova, D.A., Longo, G.O., Hanazaki, N., Carvalho-Filho, A., Ferreira, C.E.L., 2013. Biological attributes and major threats as predictors of the vulnerability of species: a case study with Brazilian reef fishes. Fauna Flora Int. Oryx 47 (2), 259e265. http://dx.doi.org/10.1017/ S003060531100144X.
Bene, C., Hersoug, B., Allison, E.H., 2010. Not by rent alone: analysing the pro-poor func-tions of small-scalefisheries in developing countries. Dev. Policy Rev. 28, 325e358.
Berkes, F., Berkes, M.K., Fast, H., 2007. Collaborative integrated management in Canada's North: the role of local and traditional knowledge and community-based monitoring. Coast. Manag. 35 (1), 143e162.
Bourne, C.M., Kehler, D.G., Wiersma, D.F., Cote, D., 2011. Barriers of thefish passage and barriers tofish passage assessments: the impact of assessment methods and assumptions on barrier identification and quantification of watershed connectivity. Aquat. Ecol. 45, 389e403.
Bowen, S.H., 1992. Quantitative description of the diet. In: Nielsen, L.A., Johnson, D.L. (Eds.), Fisheries Techniques. American Fisheries Society, Bethesda, pp. 325e336.
Bunce, M., Rodwell, L.D., Gibb, R., Mee, L., 2008. Shifting baselines infishers' per-ceptions of island reeffishery degradation. Ocean Coast. Manag. 51, 285e302.
Caliman, A., Pires, A.F., Esteves, F.A., Bozelli, R.L., Farjalla, V.F., 2010. The prominence of and biases in biodiversity and ecosystem functioning research. Biodivers. Conserv. 19, 651e664.
Caliman, A., Carneiro, L.S., Leal, J.J.L., Farjalla, V.F., Bozelli, R.L., Esteves, F.A., 2013. Biodiversity effects of ecosystem engineers are stronger on more complex ecosystem processes. Ecology 94 (9), 1977e1985.
Carvalho, A.R., Williams, S., January, M., Sowman, M., 2009. Reliability of community-based data monitoring in the Olifants River estuary (South Africa). Fish. Res. 96, 119e128.
Christensen, V., 1995. Ecosystem maturityetowards quantification. Ecol. Model. 77, 3e32.
Christensen, V., Pauly, D., 1992. ECOPATH IIe a software for balancing steady-state ecosystem models and calculating network characteristics. Ecol. Model. 61, 169e185.
Fig. 4. Simulation Ecopath with Ecosim model: a) Fishing effort on Cichla kelberi and on Hoplias malabaricus increased 10 times; b) Fishing effort increased by 10 times on H. malabaricus; c) Fishing effort on C. kelberi increased by 10 times; d) Fishing effort on C. kelberi increased by 20 times.
Christensen, V., Walters, C.J., 2004. Ecopath with Ecosim: methods, capabilities and limitations. Ecol. Model. 172, 109e139.
Christie, P., Fluharty, D.L., White, A.T., Eisma-Osorio, L., Jatulan, W., 2007. Assessing the feasibility of ecosystem-basedfisheries management in tropical contexts. Mar. Policy 31 (3), 239e250.
Coll, M., Libralato, S., 2012. Contributions of food web modelling to the ecosystem approach to marine resource management in the Mediterranean Sea. Fish Fish. 13, 60e88.
Cooke, R.M., 1991. Experts in Uncertainty: Opinion and Subjective Probability in Science. Oxford University Press, New York, 321p.
Davis, A., Wagner, J.R., 2003. Who knows? On the importance of identifying experts when researching local ecological knowledge. Hum. Ecol. 31 (3), 463e489.
Daw, T.M., 2010. Shifting baselines and memory illusions: what should we worry about when inferring trends from resource user interviews? Anim. Cons. 13, 534e535.
Daw, T.M., Cinner, J.E., McClanahan, T.R., Brown, K., Stead, S.M., 2012. Tofish or not tofish: factors at multiple scales affecting Artisanal Fishers' readiness to exit a declining fishery. PLoS One 7 (2), 1e10. http://dx.doi.org/10.1371/ journal.pone.0031460.
Eddy, T.D., Gardner, J.P.A., Perez-Matus, A., 2010. Applying Fishers' ecological knowledge to construct past and future lobster stocks in the Juan Fernandez Archipelago, Chile. PLoS One 5 (11), 1e12. http://dx.doi.org/10.1371/ journal.pone.0013670.
Froese, R., Pauly, D. (Eds.), 2013. FishBase. W. W. W. Electronic Publication.www. fishbase.org. version (10/2013).
Garcia, S.M., Cochrane, K.L., 2005. Ecosystem approach tofisheries: a review of implementation guidelines. ICES J. Mar. Sci. 62 (3), 311e318.
García-Allut, A., Freire, J., Barreiro, A., Losada, D.E., 2007. Methodology for integra-tion offisher's ecological knowledge in fisheries biology and management using knowledge representation (artificial intelligence). In: Haggan, N., Neis, B., Baird, I.G. (Eds.), The Value of Anecdote. Fishers' Knowledge in Fisheries Science and Management, Coastal Management Sourcebooks, vol. 4. UNESCO Publish-ing, Paris, pp. 227e237.
Gerhardinger, L.C., Godoy, E.A.S., Jones, P.J.S., 2009. Local ecological knowledge and the management of marine protected areas in Brazil. Ocean Coast. Manag. 52 (3e4), 154e165.
Gibson, R.J., Haedrich, R.L., Wernerheim, C.M., 2005. Loss of habitat as a conse-quence of inappropriately constructed stream crossings. Fisheries 30, 10e17.
Goodman, L., 1961. Snowball sampling. Ann. Math. Stat. 32, 148e170.
Gubiani, E.A., Gomes, L.C., Agostinho, A.A., Baumgartner, G., 2010. Variations infish assemblages in a tributary of the upper Parana River, Brazil: a comparison between pre and post-closure phases of dams. Rivers Res. Appl. 26, 848e865.
Gubiani, E.A., Gomes, L.C., Agostinho, A.A., 2012. Estimates of population parame-ters and consumption/biomass ratio forfishes in reservoirs, Parana State, Brazil. Neotrop. Ichthyol. 10, 177e188.
Hallwass, G., Lopes, P.F., Juras, A.A., Silvano, R.A., 2013. Fishers' knowledge identifies
environmental changes and fish abundance trends in impounded tropical
rivers. Ecol. Appl. 23 (2), 392e407.
Henry, B., Moffitt, T.E., Caspi, A., Langley, J., Silva, P.A., 1994. On the “remembrance of things past”: a longitudinal evaluation of the retrospective method. Psychol. Assess. 6, 92e101.
Heymans, J.J., Howell, K.L., Ayers, M., Burrows, M.T., Gordon, J.D.M., Jones, E.G., Neat, F., 2011. Do we have enough information to apply the ecosystem approach to management of deep-seafisheries? An example from the West of Scotland. ICES J. Mar. Sci. 68, 265e280.
Hilborn, R., 2011. Future directions in ecosystem basedfisheries management: a personal perspective. Fish. Res. 108, 235e239.
Hill, N.A.O., Michael, K.P., Frazer, A., Leslie, S., 2010. The utility and risk of local ecological knowledge in developing stakeholder drivenfisheries management: the Foveaux Strait dredge oysterfishery, New Zealand. Ocean Coast. Manag. 53, 659e668.
Holm, P., 2003. Crossing the border: on the relationship between science and fishermen's knowledge in a resource management context. Marit. Stud. 2, 5e33.
Huang, L.L., Wu, Z.Q., Li, J.H., 2013. Fish fauna, biogeography and conservation of freshwater fish in Poyang Lake Basin, China. Environ. Biol. Fishes 96, 1229e1243.
Hynes, H.B.N., 1950. The food of fresh-water sticklebacks (Gasterosteus aculeatus and Pygosteus pungitius), with a review of methods used in studies of the food fishes. Anim. Ecol. 19, 36e57.
Johannes, R.E., 1981. Working withfishermen to improve coastal tropical fisheries and resource management. Bull. Mar. Sci. 13 (3), 673e680.
Johannes, R.E., 1998. The case for data-less marine resource management: examples from tropical near shorefin fisheries. Trends Ecol. Evol. 13, 243e246.
Johannes, R.E., Neis, B., 2007. The value of anecdote.In Fishers' knowledge in fish-eries science and management. In: Haggan, N., Neis, B., Baird, I.G. (Eds.), Coastal Management Sourcebooks, vol. 4. UNESCO Publishing, Paris, pp. 41e58.
Karling, L.C., Isaac, A., Affonso, I.P., Takemoto, R.M., Pavanelli, G.C., 2013. The impact of a dam on the helminth fauna and health of a neotropicalfish species Sal-minus brasiliensis (Cuvier 1816) from the upper Parana River, Brazil. J. Helminthol. 87, 245e251.
Kemp, P.S., O'Hanley, J.R., 2010. Evaluation of barriers tofish migration and prior-ization of removal and mitigation projects. Fish. Manag. Ecol. 17, 297e322.
Kottek, M., Grieser, J., Beck, C., Rudolf, B., Rubel, F., 2006. World map of the k€oppen-geiger climate classification updated. Meteorol. Sch. 15, 259e263.
Kurien, J., 1998. Traditional ecological knowledge and ecosystem sustainability: new meaning to Asian coastal proverbs. Ecol. Appl. 8 (1), S2eS5. Supplement.
Le Fur, J., Guilavogui, A., Teitelbaum, A., 2011. Contribution of localfishermen to improving knowledge of marine ecosystem and resources in the Republic of Guinea, West Africa. Can. J. Fish. Aquat. Sci. 68, 1454e1469.
Lebreton, B., Richard, P., Parlier, E.P., Guillou, G., Blanchard, G.F., 2011. Trophic ecology of mullets during their spring migration in a European saltmarsh: a stable isotope study. Estuar. Coast. Shelf Sci. 91, 502e510.
Leite, M.C.F., Gasalla, M.A., 2013. A method for assessingfishers' ecological knowl-edge as a practical tool for ecosystem-basedfisheries management: seeking consensus in Southeastern Brazil. Fish. Res. 145, 43e53.
Libralato, S., Christensen, V., Pauly, D., 2006. A method for identifying keystone species in food web models. Ecol. Model. 195, 153e171.
Lopes, P.F., Rosa, E.M., Salyvonchyk, S., Nora, V., Begossi, A., 2013. Suggestions for fixing top-down coastal fisheries management through participatory ap-proaches. Mar. Pol. 40, 100e110.
Marasco, R.J., Goodman, D., Grimes, C.B., Lawson, P., Punt, A.E., Quinn, T.J., 2007. Ecosystem-basedfisheries management: some practical suggestions. Can. J. Fish. Aquat. Sci. 64 (6), 928e939.
Maurstad, A., Dale, T.T., Bjørn, P.A., 2007. You wouldn't spawnin a septic tank, would you? Hum. Ecol. 35, 601e610.
Menezes, R.F., Attayde, J.L., Lacerot, G., Kosten, S., Coimbra e Souza, L., Costa, L.S., Nes, E.H., Jeppesen, E., 2012. Lower biodiversity of native fish but only marginally altered plankton biomass in tropical lakes hosting introduced piscivorous Cichla cf. ocellaris. Biol. Invasions 14, 1353e1363.
Morril, W.T., 1967. Ethnoichthyology of the cha-cha. Ethnology 6, 405e417.
Odum, E.P., 1969. The strategy of ecosystem development. Science 164, 262e270.
Ogutu-Ohwayo, R., 1993. The effects of predation by Nile Perch, Lates niloticus L., on thefish of Lake Nabugabo, with suggestions for conservation of endangered endemic cichlids. Conserv. Biol. 7, 701e711.
Padgham, M., Webb, J.A., 2010. Multiple structural modifications to dendritic ecological networks produce simple responses. Ecol. Model 221, 2537e2545.
Palomares, M.L.D., Pauly, D., 1998. Predicting food consumption offish populations as functions of mortality, food type, morphometrics, temperature and salinity. Mar. Freshw. Res. 49 (5), 447e453.
Pauly, D., 1980. On the interrelationships between natural mortality, growth parametersand mean environmental temperatures in itsfish stocks. J. Cons. Int. Explor. Mer. 39 (3), 175e192.
Pelicice, F.M., Agostinho, A.A., 2009. Fish fauna destruction after the introduction of a non-native predator (Cichla kelberi) in a Neotropical reservoir. Biol. Invasions 11, 1789e1801.
Peressin, A., Cetra, M., 2014. Responses of the ichthyofauna to urbanization in two urban areas in Southeast Brazil. Urb. Ecosyst. 17 (2), 1e12.http://dx.doi.org/ 10.1007/011252-014-0352-5.
Pikitch, E.K., Santora, C., Babcock, E.A., Bakun, A., Bonfil, R., Conover, D.O., Dayton, P., Doukakis, P., Fluharty, D., Heneman, B., Houde, E.D., Link, J., Livingston, P.A., Mangel, M., McAllister, M.K., Pope, J., Sainsbury, K.J., 2004. Ecosystem-based fishery management. Science 305, 346e347.
Poizat, G., Baran, E., 1997. Fishermen's knowledge as background information in tropicalfish ecology: a quantitative comparison with fish sample results. En-viron. Biol. Fishes 50, 435e449.
Rahel, F.J., 2007. Biogeographic barriers, connectivity and homogenization of freshwater faunas: it's a small world after all. Freshw. Biol. 52, 696e710.
Rice, J., 2008. Can we manage ecosystems in a sustainable way? J. Sea Resour. 60 (1e2), 8e20.
Richardson, D., 2011. Fifty Years of Invasion Ecologye the Legacy of Charles Elton. Blackwell Publishing, Oxford, UK.
Rolls, R.J., Boulton, A.J., Growns, I.O., Maxwell, S.E., 2011. Response byfish assem-blages to an environmentalflow release in a temperate coastal Australian river: a paired catchment analysis. Rivers Res. Appl. 27, 867e880.
Saenz-Arroyo, A., Roberts, C.M., Torre, J., Cari~no-Olvera, M., 2005. Using fishers' anecdotes, naturalists' observations and grey literature to reassess marine species at risk: the case of the Gulf grouper in the Gulf of California, Mexico. Fish Fish. 6, 121e133.
Sarda, F., Maynou, F., 1998. Assessing perceptions: do Catalanfishermen catch more shrimp on Fridays? Fish. Res. 36, 149e157.
Silvano, R.A.M., Begossi, A., 2012. Fishermen's local ecological knowledge on southeastern Brazilian coastalfishes: contributions to research, conservation, and management. Neotrop. Ichthyol. 10, 133e147.
Silvano, R.A.M., Jorgensen, J.V., 2008. Beyondfishermen's tales: contributions of fishers' local ecological knowledge to fish ecology and fisheries management. Environ. Dev. Sustain. 10, 657e675.
Starks, E.C., 1913. The Fishes of the Stanford Expedition to Brazil. Leland Stanford Publications University Series, 77 pp.
Tesfamichael, D., Pitcher, T.J., Pauly, D., 2014. Assessing changes infisheries using Fishers' knowledge to generate long time series of catch rates: a case study from the Red Sea. Ecol. Soc. 19 (1), 18e30.
Ticheler, H.J., Kolding, J., Chanda, B., 1998. Participation of localfishermen in sci-entific fisheries data collection: a case study from the Bangweulu swamps, Zambia. Fish. Manag. Ecol. 5, 81e92.
Turvey, S.T., Barrett, L.A., Yujiang, H., Lei, Z., Xinqiao, Z., Xianyan, W., Yadong, H., Kaiya, Z., Hart, T., Ding, W., 2009. Rapidly shifting baselines in Yangtzefishing communities and local memory of extinct species. Cons. Biol. 24 (3), 778e787.
Vieira, D.B., 2002. Icthyofauna Freshwater Survey in Rio Grande do Norte State (Brazil). Final Undergraduate Report. Londrina, Parana, 104p. (in Portuguese).
Vitule, J.R.S., Freire, C.A., Simberloff, D., 2009. Introduction of non-native freshwater fish can certainly be bad. Fish Fish. 10, 98e108.
Walters, C., Christensen, V., Pauly, D., 1997. Structuring dynamic models of exploited ecosystems from trophic mass-balance assessments. Rev. Fish Biol. Fish. 7, 139e172.
Watts, D.J., Strogatz, S.H., 1998. Collective dynamics of ‘small-world’ networks. Nature 393, 440e442.
Williams, A., Bax, N., 2007. Integratingfisher's knowledge with survey data to un-derstand the structure, ecology and use of a seascape off south-eastern
Australia. In: Haggan, N., Neis, B., Baird, I.G. (Eds.), Fishers' Knowledge in Fisheries Science and Management, Coastal Management Sourcebooks, vol. 4. UNESCO Publishing, Paris, pp. 41e58.
Zappes, C.A., Sa Alves, L.C.P., Silva, C.V., Azevedo, A.F., Di Beneditto, A.P.M., Andriolo, A., 2013. Accidents between artesanalfisheries and cetaceans on the Brazilian coast and Central Amazon: proposals for integrated management. Ocean Coast. Manag. 85, 46e57.
