See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/312593226 Hydrologic pulsing promotes spatial connectivity and food web subsidies in a subtropical coastal ecosystem Article  in  Marine Ecology Progress Series · March 2017 DOI: 10.3354/meps12060 CITATIONS 4 READS 382 11 authors, including: Some of the authors of this publication are also working on these related projects: Biosafety and Occupational Health in Aquaculture View project Food webs structure and dynamics in Southwestern Atlantic coastal systems View project Alexandre Garcia Universidade Federal do Rio Grande (FURG) 100 PUBLICATIONS   1,320 CITATIONS    SEE PROFILE Kirk O Winemiller Texas A&M University 398 PUBLICATIONS   15,643 CITATIONS    SEE PROFILE Rodrigo Ferreira Bastos Federal University of Pernambuco 22 PUBLICATIONS   96 CITATIONS    SEE PROFILE Fabiano Corrêa Universidade Federal de Pelotas 42 PUBLICATIONS   110 CITATIONS    SEE PROFILE All content following this page was uploaded by Alexandre Garcia on 08 February 2017. The user has requested enhancement of the downloaded file. https://www.researchgate.net/publication/312593226_Hydrologic_pulsing_promotes_spatial_connectivity_and_food_web_subsidies_in_a_subtropical_coastal_ecosystem?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_2&_esc=publicationCoverPdf https://www.researchgate.net/publication/312593226_Hydrologic_pulsing_promotes_spatial_connectivity_and_food_web_subsidies_in_a_subtropical_coastal_ecosystem?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_3&_esc=publicationCoverPdf https://www.researchgate.net/project/Biosafety-and-Occupational-Health-in-Aquaculture?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_9&_esc=publicationCoverPdf https://www.researchgate.net/project/Food-webs-structure-and-dynamics-in-Southwestern-Atlantic-coastal-systems?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_9&_esc=publicationCoverPdf https://www.researchgate.net/?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_1&_esc=publicationCoverPdf https://www.researchgate.net/profile/Alexandre_Garcia9?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_4&_esc=publicationCoverPdf https://www.researchgate.net/profile/Alexandre_Garcia9?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_5&_esc=publicationCoverPdf https://www.researchgate.net/institution/Universidade_Federal_do_Rio_Grande_FURG?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_6&_esc=publicationCoverPdf https://www.researchgate.net/profile/Alexandre_Garcia9?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_7&_esc=publicationCoverPdf https://www.researchgate.net/profile/Kirk_Winemiller?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_4&_esc=publicationCoverPdf https://www.researchgate.net/profile/Kirk_Winemiller?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_5&_esc=publicationCoverPdf https://www.researchgate.net/institution/Texas_A_M_University?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_6&_esc=publicationCoverPdf https://www.researchgate.net/profile/Kirk_Winemiller?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_7&_esc=publicationCoverPdf https://www.researchgate.net/profile/Rodrigo_Bastos3?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_4&_esc=publicationCoverPdf https://www.researchgate.net/profile/Rodrigo_Bastos3?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_5&_esc=publicationCoverPdf https://www.researchgate.net/institution/Federal_University_of_Pernambuco?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_6&_esc=publicationCoverPdf https://www.researchgate.net/profile/Rodrigo_Bastos3?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_7&_esc=publicationCoverPdf https://www.researchgate.net/profile/Fabiano_Correa?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_4&_esc=publicationCoverPdf https://www.researchgate.net/profile/Fabiano_Correa?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_5&_esc=publicationCoverPdf https://www.researchgate.net/institution/Universidade_Federal_de_Pelotas?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_6&_esc=publicationCoverPdf https://www.researchgate.net/profile/Fabiano_Correa?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_7&_esc=publicationCoverPdf https://www.researchgate.net/profile/Alexandre_Garcia9?enrichId=rgreq-514e9bef4fe8fff78f09c76a4f434063-XXX&enrichSource=Y292ZXJQYWdlOzMxMjU5MzIyNjtBUzo0NTk1NTE2MjAyNDM0NjJAMTQ4NjU3NzA1MjkyNA%3D%3D&el=1_x_10&_esc=publicationCoverPdf 1 Running head: 1 Pulse-driven subsidies in an estuarine ecosystem 2 3 Hydrologic pulsing promotes spatial connectivity and food web subsidies in a subtropical 4 coastal ecosystem. 5 6 Authors: Garcia, A.M*. 1 , K.O. Winemiller 2 , D.J. Hoeinghaus 3 , M.C. Claudino 1 , R. Bastos 1 , F. 7 Correa 1 , S. Huckembeck, 1 J. Vieira 1 , D. Loebmann 1 , P. Abreu 1 , and C. Ducatti 4 8 9 1 Oceanography Institute, Rio Grande Federal Universiy, Rio Grande, Rio Grande do Sul, 96203-10 900, BRAZIL 11 2 Department of Wildlife and Fisheries Sciences and Interdisciplinary Program of Ecology and 12 Evolutionary Biology, Texas A&M University, College Station, Texas 77843-2258, USA 13 3 Department of Biological Sciences and the Advanced Environmental Research Institute, 14 University of North Texas, Denton, Texas 76203-5017, USA 15 4 Paulista Júlio de Mesquita Filho State University, Stable Isotope Center for Environmental and 16 Life Sciences, Biosciences Institute, Botucatu, São Paulo 18608-000, BRAZIL. 17 18 # Corresponding author 19 amgarcia.ictiofurg@gmail.com 20 +55 53 32336539 21 22 23 2 Abstract 24 25 Resource pulsing is a widespread phenomenon, but its effects on ecosystem dynamics are often 26 difficult to predict. Hydrological pulsing, in particular, is known to influence the structure and 27 dynamics of fluvial and coastal ecosystems, but little information is available for its effects on 28 trophic connectivity between wetlands and estuaries. In this study, we investigated the 29 hypothesis that hydrologic pulsing drives one-way trophic subsidies (e.g. suspended organic 30 matter and freshwater fish) from wetland to estuary. Our study system is a coastal lagoon with an 31 ephemeral mouth that, when closed, stores freshwater as a sustained flood pulse that is 32 subsequently released when a connection with the sea is reestablished. We monitored isotopic 33 composition of consumers and food sources over the course of an entire flood pulse to infer 34 trophic linkages and spatial subsidies. Before the flood peak (April and May), freshwater and 35 estuarine zones were largely dependent on local primary production sources (seston and C3 36 plants vs. C4 plants and microphytobenthos, respectively), essentially functioning as 37 disconnected compartments. A sustained pulse of freshwater inflow (June to August) induced 38 greater habitat connectivity and a net flow of biomass and energy from the freshwater zone into 39 the estuarine zone. The opening of the lagoon outlet channel abruptly terminated the flood pulse 40 and reduced freshwater subsidies to estuarine consumers, and both zones returned to dependence 41 on autochthonous production. Our findings contribute to current concerns that artificial opening 42 of sandbars in coastal lagoons alters natural ecological dynamics with significant effects on 43 biodiversity and ecosystem processes. 44 45 3 Keywords: basal resource, Bayesian mixing model, biomass assimilation, estuary, hydrologic 46 connectivity, production source, salinity, trophic ecology 47 48 49 50 51 52 53 54 55 4 Introduction 56 Food web dynamics are influenced not only by consumer-resource interactions, but also 57 by environmental variation that directly affects ecosystem productivity, habitat suitability, and 58 other factors affecting organism fitness and population growth (Winemiller & Layman 2005). 59 Further complicating our ability to predict dynamics in response to environmental drivers is the 60 fact that perceived patterns of variation are influenced by temporal and spatial scales of analysis 61 (Petraitis & Latham 1999). Most, and probably all, ecosystems are subject to abiotic factors that 62 pulse. Resource pulses, in particular, are defined as discrete, significant increases in resource 63 availability over intervals of time and/or space (Yang et al. 2008). Environmental seasonality 64 affects local population dynamics and species interactions, but also can have regional effects 65 when pulsing factors induce mass migrations or synchrony among regional populations 66 (Leibhold et al. 2004). For example, Krenz et al. (2011) showed how coastal upwelling 67 influenced by the El Niño Southern Oscillation affects food web subsidies within intertidal 68 ecosystems over a large region. Resource subsidies associated with pulsing environmental 69 factors strongly influence bottom-up (e.g. a fertilizer effect) and top-down (consumer control) 70 effects in food webs of arid islands (Polis et al. 1997, Spiller et al. 2010). 71 Pulsing can induce responses that reverberate throughout the food web (Yang et al. 2008). 72 Mass spawning by corals and reef fishes in response to lunar and oceanographic cues produces 73 huge subsidies for planktivores (Pratchett et al. 2001), which can have long-term effects on 74 fitness of the latter (McCormick 2003). Regional insect outbreaks in response to favorable 75 climatic conditions can devastate plant biomass (Mattson & Addy 1975) while subsidizing 76 insectivores (Yang 2004).Seasonal migrations can create food web subsidies at the landscape 77 scale (Polis et al. 2004). For example, the decomposing carcasses of post-spawn anadromous 78 5 salmon provide a pulse of marine-derived nutrients that stimulates primary production not only 79 within oligotrophic streams, but also riparian terrestrial ecosystems (Naiman et al. 2002). 80 Theoretical models indicate diverse responses to pulsing that range from transient population 81 dynamics that may or may not influence species coexistence, to tipping the dynamics governing 82 alternative stable states of ecosystems (Ostfeld & Keesing 2000, Holt 2008; Petersen et al. 2008). 83 Hydrologic changes, in particular, can trigger ecosystem pulsing that affects the structure 84 and dynamics of fluvial and coastal ecosystems (Winemiller 1990, Poff et al. 1997, DeAngelis et 85 al. 2005, Childers 2006, Warfe et al. 2011). Fish dynamics in hydrologically pulsed wetland 86 ecosystems monitored across long time scales (e.g. Florida’s Everglades) are strongly associated 87 with seasonal cycles in precipitation and water depth that cascade to other components of the 88 food web (DeAngelis et al. 1997, Childers 2006). 89 In southeastern Brazil, inter-annual and seasonal variation in precipitation drives changes 90 in hydrology and salinity gradients that, in turn, affect the composition of local fish assemblages 91 within coastal lagoons (Garcia et al. 2003, Garcia et al. 2004). A suitable ecosystem model for 92 evaluating hydrologic pulsing effects on spatial food web subsidies between freshwater and 93 estuarine systems in this region is the Lagoa do Peixe National Park (LPNP). Ecological 94 conditions within LPNP are influenced not only by intra-annual variations in freshwater inflow, 95 but also by ephemeral connections with the sea. Freshwater draining from the coastal plain is 96 stored within the lagoon as a sustained flood pulse and then released over a relatively short 97 period when a connection with the sea is established either (i) naturally by inflows or (ii) 98 artificially by mechanical excavation (Lanés et al. 2015). Thus, hydrology induces a freshwater 99 flood pulse that causes aquatic ecosystem expansion with greater connectivity between the 100 6 freshwater wetland and the estuarine zone and conditions promoting input of freshwater basal 101 production sources and consumer taxa into the estuary. 102 We hypothesized that hydrologic pulsing influences food web dynamics along the 103 longitudinal fluvial gradient via one-way trophic subsidies (e.g. passive transport of suspended 104 organic matter and dispersal of freshwater fish) from wetland to estuary. We used stable isotope 105 methods to test this hypothesis based on extensive temporal and spatial sampling of production 106 sources and consumers at locations along the longitudinal fluvial gradient, and used a Bayesian 107 mixing model to estimate probability distributions for assimilation of alternative production 108 sources. Stable isotope analysis (SIA) is commonly used to reconstruct trophic links connecting 109 food web components across spatial (Hoeinghaus et al. 2011, Claudino et al. 2015) and temporal 110 (Claudino et al. 2013, Garcia et al. in press) scales, and several authors have used SIA to test 111 effects of hydrologic changes on food web organization and spatial subsidies among ecosystems 112 (Jardine et al. 2012, Abrantes et al. 2013, Kaymak et al. 2015, Ou & Winemiller 2016). In 113 contrast with most previous studies that used seasonal or before-and-after sampling designs, our 114 study analyzed a time-series of samples taken over frequent, short intervals, which enabled us to 115 view food web structure like a movie rather than a snapshot. This analysis provided a basis for 116 estimating the timing and magnitude of a spatial food web subsidy to estuarine consumers. 117 118 Material and methods 119 Study area 120 The Lagoa do Peixe National Park (LPNP) has an area of 344 km² and is situated along 121 the coastal plain of Rio Grande do Sul, the southernmost state in Brazil. This coastal plain is 122 characterized by a flat topography, low altitude (< 20 masl) and low tidal range (~ 0.5m) (Fig. 1). 123 7 Lagoa do Peixe is a shallow coastal lagoon (<50 cm, except its channel) surrounded by 124 freshwater wetlands (Maltchik et al. 2010), except for its eastern border where sand dunes are 125 prevalent (Fig. 1). The lagoon has a narrow outlet to the sea that is blocked from 126 February/March to August/September each year by sand dunes until freshwater inflows, usually 127 from winter rainfall, establish a connection. If the seasonal connection with the sea is not 128 established naturally, earth-moving machinery is used to construct an outlet channel (200 m long, 129 40 m wide, 1.5 m deep) during late winter (August-September). The overall dimensions of both 130 artificial and natural channels are similar. This periodic opening of the outlet is done to promote 131 the entrance of marine shrimp larvae into the lagoon to favor commercial fisheries and also to 132 drain water from the floodplain to increase pasture for livestock ranching (Lanés et al. 2015). 133 The LPNP has a humid subtropical climate, with mean temperatures ranging from 14.6°C 134 to 22.2°C, a mean annual temperature of 17.5°C, and annual precipitation in the study area 135 ranges from 1,150 to 1,450 mm yr −1 , with an annual mean of 1,250 mm yr −1 . Prior studies on 136 estuarine systems approximately 100 km south of the LPNP have shown that rainfall anomalies 137 cause hydrological changes that affect estuarine and freshwater fish assemblages (Garcia et al. 138 2004). To assess potential influence of climatic variation on ecological dynamics, temperature 139 and rainfall data were obtained from two meteorological stations (Rio Grande and Mostardas) 140 near the study area (Fig. 1), and the status of the lagoon connection with the sea was monitored 141 throughout the study. Rainfall varied seasonally, with two periods of high rainfall: May to 142 August 2008 and January to March 2009. The former wet period coincided with a closed lagoon 143 mouth, which occurred from April to August 2008. At the estuarine site in particular, water 144 flooded over the marginal areas during wet periods, increasing hydrological exchanges with 145 freshwater wetlands. Average air temperature varied seasonally, with lowest values during the 146 8 austral winter (14.1 ° C in June) and highest values in summer (23.7 ° C in February). Based on 147 this climatic variation, we defined flooded and non-flooded phases of the hydrologic pulse in 148 order to evaluate our initial hypothesis (Fig. 1). 149 150 Field collections and sample processing 151 Samples were obtained monthly from April 2008 to May 2009 (with the exception of 152 September 2008 and March 2009) in three regions along the main longitudinal axis of the LPNP: 153 1) a freshwater wetland located near the northern limit of the park, 2) the upper–middle reach of 154 the lagoon that encompasses an ecotone between estuarine and freshwater zones, and 3) the 155 middle reach of the lagoon near the ephemeral connection with the sea (Fig. 1). These sites are 156 subsequently referred to as freshwater wetland, estuarine zone and lagoon mouth, respectively. 157 Fish were caught using four fishing gears (gillnets, beach seine, beam trawl, dipnet) 158 during each monthly survey. To catch larger fishes (>200 mm total length), two gillnets (4 x 2 m, 159 comprised by panels with different mesh sizes of 15, 20, 30 and 35 mm) were deployed in the 160 channel (2.0 – 2.5 m deep). One to two beach seine (9 m long, 2.4 m high, mesh size 13 mm 161 wings and 5 mm center) hauls were made in unvegetated nearshore areas. Densely vegetated 162 marginal habitats were sampled with three hauls of the beam trawl (mouth = 1 x 1 m, mesh = 5 163 mm) and dip netting for approximately 15 min. Beach seine and beam trawl hauls and dip 164 netting were conducted in shallower waters (<1.5 m) and were effective in capturing smaller 165 species (< 50 mm), including those that take refuge in structurally complex habitats. Sampling 166 with multiple gear types provides a more representative sample of the fish assemblage, especially 167 when the ecosystem is comprised by multiple habitat types with differing complexities and 168 species of various sizes and behaviors (Rozas & Minello 1997, Chick et al., 2004, Garcia et al. 169 9 2006). Representative macroinvertebrates, such as adult and immature aquatic insects, 170 gastropods, polychaete worms, shrimps and crabs, were collected at each study site. 171 Samples of microphytobenthos, seston (phytoplankton and suspended fine particulate 172 organic matter), and leaves from floating, emergent and submerged macrophytes were collected 173 at each site during each survey. This material was used for determination of the isotopic 174 composition of major basal production sources. Macrophytes were collected by hand, samples 175 of microphytobenthos were obtained by carefully removing the thin upper layer of flocculent or 176 consolidated biofilm from substrates, and seston samples were obtained by filtering water 177 through a pre-combusted (450° C, 4 h) Whatman glass fiber filter (GF/F) using a manual pump. 178 Immediately after collection, all specimens were placed on ice for transport to the 179 laboratory where they were stored frozen. After thawing, fish were weighed (g), measured (mm 180 total length, TL) and dissected to extract approximately 5 g of dorsal muscle tissue for isotopic 181 analysis. For fish <50 mm TL, a composite sample was obtained by combining muscle tissue 182 from 5–15 conspecifics from the same site. Gastropods, shrimps and crabs were dissected, and a 183 sample of muscle tissue was extracted for stable isotope analysis. Due to their small sizes, 184 polychaetes and aquatic insects were processed as whole specimens. Tissue samples from fish 185 and macroinvertebrates were inspected and any significant non-muscle material (e.g. bone, scales, 186 exoskeleton) was removed before the samples were rinsed with distilled water, placed in sterile 187 Petri dishes, and dried in an oven at 60 ° C until attainment of a constant weight (minimum of 48 188 h). Dried samples were ground to a fine powder with a mortar and pestle and stored in clean 189 Eppendorf tubes. Sub-samples were weighed to 6-10 mg, pressed into Ultra-Pure tin capsules 190 (Costech, Valencia, CA), and sent to the Centro de Isótopos Estáveis, Universidade Estadual 191 Paulista (UNESP) for measurement of stable isotope ratios ( 13 C/ 12 C and 15 N/ 14 N). Results are 192 10 reported as parts per thousand (‰) differences from a corresponding standard:  H X = [(Rsample / 193 Rstandard) - 1] x 1000, where X is a particular element (C, carbon or N, nitrogen), the superscript H 194 denotes the heavy isotope mass of the given element ( 13 C, 15 N) and R is the ratio of the heavy 195 isotope to the light isotope for the element ( 13 C/ 12 C, 15 N/ 14 N). Standards were carbon in the 196 PeeDee Belemnite and molecular nitrogen in air. Standard deviations of  13 C and  15 N replicate 197 analyses were 0.14‰ and 0.13‰, respectively. 198 199 Data analysis 200 For some analyses, fishes were pooled into five functional guilds based on prior 201 classification proposed by Garcia et al. (2001) and also considering Myers’ (1938) model of 202 primary- and secondary-division freshwater fishes, as follows: 1) estuarine resident species that 203 typically occur and breed within the estuary; 2) estuarine dependents that are marine or 204 freshwater spawning species found in large numbers within the estuary during certain periods of 205 their life cycle; 3) marine vagrants typically inhabiting marine habitats and rarely occurring 206 within the estuary; 4) primary freshwater fishes with no tolerance to salinity that are confined to 207 freshwater; and 5) secondary freshwater fishes with some salinity tolerance that may enter 208 brackish water occasionally. Also, due to consistent differences in retention of 13 C by plants 209 using the Calvin cycle (C3) and Hatch-Slack cycle (C4) photosynthetic pathways (Marshall et al. 210 2007), sampled macrophytes were pooled and analyzed as two groups, C3 vs. C4, according to 211 the literature (Marshall et al. 2007), the range of  13 C values observed in this study and previous 212 research in nearby coastal lagoon ecosystems (Garcia et al. 2007, Hoeinghaus et al. 2011). C3 213 plants tend to have significantly lower average carbon stable isotope ratios (~ -27‰) than C4 214 plants (~ -4‰) mainly due to differences in the enzymes involved in carbon fixation during 215 11 photosynthesis (Marshall et al. 2007). Abundance patterns of these groups shift along the salinity 216 gradient of the study site, with C3 being the dominant aquatic macrophytes in the freshwater 217 wetland and C4 plants dominant in the estuarine/marine sites (Knak 2004, Rolon et al. 2011). 218 Bi-plots of  15 N and  13 C values of primary producers and organic sources, invertebrates 219 and fishes were used to evaluate patterns of isotopic variation across spatial and temporal scales. 220 Sources of organic carbon assimilated by consumers are indicated by proximity between carbon 221 isotope ratios ( 13 C) of consumers and their food sources, whereas trophic position of each 222 consumer is indicated by the relative position of their nitrogen isotope ratios ( 15 N) in relation to 223 an isotopic baseline (Peterson & Fry, 1987). Analysis of variance (ANOVA) was used to 224 compare average  13 C and  15 N values of sources and consumers among study sites and, for 225 fishes only, among months. When a significant main effect was observed, pairwise comparisons 226 of site means were performed using Newman-Keuls post-hoc procedure. Normality and 227 homogeneity of variances were evaluated by the Kolmogorov-Smirnov and Cochran tests, 228 respectively (Zar 1984). 229 To estimate relative contributions of basal production sources to fish assemblages of the 230 freshwater wetland, estuarine zone and lagoon mouth, we employed Bayesian mixing model with 231 four basal production sources as end-members (seston, microphytobenthos, C3 and C4 plants) 232 using data specific to each site. Mixing models were run using the Stable Isotope Analysis in R 233 (SIAR) package in R statistical software (R Core Team 2012), which employs a Bayesian 234 approach that incorporates uncertainties associated with sample variability and trophic 235 enrichment on the assumption they are normally distributed (Parnell et al. 2010). Each mixing 236 model was fit via a Markov chain Monte Carlo (MCMC) method that generate simulations of 237 plausible values of dietary proportions consistent with the data using a Dirichlet prior 238 12 distribution (Parnell et al. 2010). Each model was run based on 500,000 iterations, discarding the 239 first 50,000, and considering a non-informative prior to guide the dietary proportion simulations. 240 Upper and lower credibility intervals (95, 75 and 50%) were used to describe the range of 241 feasible contributions for each food source to the consumer (Parnell et al. 2010). 242 In order to quantify food web subsidies occurring due to movement of carbon sources 243 from the freshwater wetland into the estuarine zone, we ran five-end-member SIAR mixing 244 models to estimate, on a monthly basis, the relative contribution of freshwater- and estuarine-245 derived sources for estuarine resident fishes (silversides Atherinella brasiliensis and Odonthestes 246 argentinensis) and individuals of primary freshwater fish species that were collected within the 247 estuarine zone during the hydrologic pulse. Importantly, silversides are not piscivorous, and thus 248 assimilation of freshwater-derived sources by these estuarine-resident taxa is not by consumption 249 of freshwater fishes that moved into the estuarine zone. We used average values of carbon and 250 nitrogen isotope ratios of seston and C3 plants collected in the wetland, and microphytobenthos, 251 C4 plants and seston collected in the estuary as representative freshwater- and estuarine-derived 252 basal production sources. We chose freshwater seston and C3 plants (mostly floating 253 macrophytes) because these basal sources are transported from the freshwater wetland into the 254 estuarine zone during floods, which create hydrologic connectivity between regions. Estuarine 255 sources included in this model were based on the four-end-member mixing model that identified 256 the primary sources assimilated by the estuarine fish assemblage. Mixing models were computed 257 assuming the silversides are two trophic levels above the basal production sources, and we used 258 values of 0.54‰ ± 1.23 and 2.54‰ ± 0.11 to correct for trophic fractionation of isotopic ratios of 259 carbon and nitrogen, respectively (Vanderklift & Ponsard 2003). 260 261 13 Results 262 A total of 684 samples representing 52 fish species was obtained in the freshwater 263 wetland (n= 352), estuarine zone (n= 187) and lagoon mouth (n= 145) between April 2008 and 264 May 2009 for analysis of carbon and nitrogen stable isotope ratios (Table S1). Local fish 265 assemblages showed a gradual zonation in species composition across the freshwater-estuarine 266 gradient. Primary- and secondary-freshwater fishes were restricted to freshwater wetland and 267 estuarine zones, with the only exception Jenynsia multidentata (secondary freshwater species), 268 which occurred at all three sites. In contrast, estuarine residents, estuarine dependents and marine 269 vagrants were captured only within the estuarine zone and the lagoon mouth (Table S1). 270 At the assemblage scale, there was a significant trend in  13 C and  15 N along the 271 freshwater to estuarine gradient, with lower values in freshwater ( 13 C: -26.3‰ ±2.86,  15 N: 272 7.5‰ ±1.5) and much higher values in the estuarine ( 13 C: -15.7‰ ±3.4,  15 N: 10.4‰ ±1.8) and 273 lagoon mouth zones ( 13 C: -13.8‰ ±2.4,  15 N: 13.3‰ ±2.1) (Fig, 2, Table S1). Overall, 274 invertebrates (e.g. crabs, shrimps, polychaete worms), macrophytes (C3 and C4) and other 275 potential food resources (seston, microphytobenthos) showed similar trends. Specifically, both 276  13 C and  15 N values significantly increased from the freshwater wetland to the lagoon mouth 277 (Fig. 2, Table S1). 278 For fish, quantitative estimates from Bayesian isotopic mixing-models revealed that 279 seston and C3 plants (0.53 to 0.69 and 0.23 to 0.44, respectively) were the basal sources 280 assimilated in greatest proportions by fish inhabiting the freshwater wetland. Conversely, within 281 the two mixohaline sites (estuarine zone and lagoon mouth), C4 plants and microphytobenthos 282 were the basal carbon sources assimilated in greatest proportions. For the estuarine zone, in 283 particular, C4 plants had a much greater contribution to fish biomass (0.70 to 0.83) compared 284 14 with C3 plants (0.08 to 0.21), and C4 plants and microphytobenthos were the basal carbon 285 sources with highest contributions near the lagoon mouth (0.53 to 0.71 and 0.07 to 0.42, 286 respectively) (Fig. 2, right panel). 287 Monthly analysis of the isotopic composition of fish assemblages revealed important 288 changes in the spatial trend in average  13 C across study sites (Fig. 3), which coincided with the 289 effect that the closure of the lagoon’s connection with the sea has on the flood dynamics of this 290 system. During the flooding episode recorded between April and October 2008, mean fish  13 C 291 values across all sites were significantly lower (-23.5‰) compared with those recorded during 292 the non-flooded period (-20.0‰) (F1, 687= 32.28, p< 0.00). When functional guilds were taken 293 into account, we observed that this difference was primarily due to the occurrence of salinity-294 intolerant, primary freshwater fishes, especially small characids (Astyanax eigenmanniorum, 295 59.7 ±11.8 mm TL; A. fasciatus, 35.7 ±8.8 mm TL; Cheirodon interruptus, 32.4 ±6.0 mm TL) 296 and poeciliids (Phalloceros caudimaculatus, 28.6 ±8.7 mm TL), that occurred in the estuarine 297 zone from July to November (Fig. 3). Fish collected from the estuary in July had lower  13 C 298 values (-21.6‰), but the isotopic composition of these primarily freshwater species increased 299 (F4,38= 7.51, p< 0.00) nearer to the estuarine site until approaching values (-17.7‰) similar to 300 those observed for estuarine-dependent fishes. Estuarine-resident fishes also showed significant 301 differences in their average  13 C values during the flood period (F1,106= 108.37, p< 0.00), and 302 had lower values (-18.7‰) when compared with the non-flooded period (-13.5‰) (Fig. 3). 303 Temporal variation in  13 C during flooded vs. non-flooded conditions (Appendix B) also 304 occurred for some basal sources, such as microphytobenthos (F1,18= 6.47, p< 0.02) and seston 305 (F1,17= 11.44, p< 0.00), which had lower  13 C during flooded (-21.8‰ and -22.6‰, respectively) 306 vs. non-flooded (-17.2‰ and -18.4‰, respectively) conditions. Average  13 C values of other 307 15 sources (C3 and C4 plants) did not change significantly between flooding conditions (F1,16= 308 0.0656, p< 0.80 and F1,5 = 0.0057, p< 0.94, respectively) (Appendix B). 309 Mixing models revealed that freshwater-derived basal sources (C3 plants and seston) 310 contributed to estuarine residents (silversides A. brasiliensis and O. argentinensis) during flood 311 conditions (April through July) (Fig. 4, left panel). Contributions of these sources were nearly 312 zero in April, but gradually increased to up to 40% in June and July. Over the same period, 313 contribution of estuarine-derived basal sources to estuarine fish biomass declined. However, two 314 months after the opening of the lagoon mouth and cessation of flood conditions and hydrologic 315 connectivity with the freshwater wetland, the contribution of freshwater-derived basal sources to 316 estuarine resident fish decreased, and C4 plants and microphytobenthos again were the main 317 sources supporting estuarine resident fishes (Fig. 4, left panel). Primary freshwater fishes, such 318 as characids and poeciliids, assimilated carbon derived mostly from seston and C3 plants in the 319 freshwater wetland. Conspecifics caught within the estuarine zone during the late flooding 320 period (July- August) assimilated freshwater- and estuarine-derived basal sources in similar 321 amounts. Similar to estuarine-resident fishes, the percent contribution of estuarine-derived 322 sources (microphytobenthos and C4 plants) to these freshwater species inhabiting the estuarine 323 zone increased to up to 69% following cessation of flood conditions (Fig. 4, right panel). 324 325 Discussion 326 Pulsing hydrology in Lagoa do Peixe affected both habitat productivity and habitat 327 connectivity which together resulted in trophic subsidies from the freshwater wetland to 328 estuarine consumers. This subsidy occurred via passive transport of basal sources (e.g. terrestrial 329 detritus in the form of fine particulate organic matter) as well as dispersal by freshwater fishes 330 16 and probably other freshwater taxa along the fluvial gradient. Seston (phytoplankton plus 331 suspended particulate organic matter of undetermined origin) from the freshwater wetland was 332 apparently transported in significant quantities into the estuary during the flood pulse. The 333 assimilation of this isotopically distinct seston (approximately 4‰ and 6‰ lower  13 C than 334 seston from the estuarine zone and lagoon mouth, respectively) by estuarine primary consumers 335 (e.g. zooplankton) and their subsequent consumption by fish could explain why we observed an 336 average reduction of approximately 5‰ in carbon isotope ratios of estuarine fishes (Odonthestes 337 argentinesis and Atherinella brasiliensis) during the flood period. Carbon isotope ratios of 338 estuarine fish gradually shifted back to typical estuarine isotopic values observed during pre-339 flooding conditions, suggesting their prey transitioned back to assimilating locally produced 340 estuarine seston. 341 In addition to passive transport of seston during flooding conditions, food web subsidies 342 also occurred via animal dispersal from the wetland into the estuary. Animal dispersal in fluvial 343 systems is often directional, either by downstream drift or upstream swimming against prevailing 344 current (Flecker et al. 2010, Oliveira et al. 2014). For example, anadromous fish that ascend 345 rivers to spawn have been shown to deliver large amounts of nutrients from excretion and 346 carcass decomposition that subsidize freshwater and riparian ecosystems by enhancing primary 347 production (Naiman et al. 2002, Koshino et al. 2013). Juvenile marine fish that migrate into 348 small coastal streams in Brazil excrete marine-derived nutrients and are consumed by resident 349 freshwater piscivores (Oliveira et al. 2014), thereby subsidizing the base and top of the food web 350 simultaneously. During the Lagoa do Peixe flood pulse, salinity-intolerant primary freshwater 351 fishes (e.g. small characins such as A. eigenmanniorum, A. fasciatus, and C. interruptus) moved 352 17 into the estuarine region and began to assimilate material derived from local basal production 353 sources. 354 In this system, these species generally consume cladocerans, other microcrustacea, algae 355 and vascular plant fragments (FC, unpublished stomach contents data). Consumption of locally-356 produced microcrustacea and vegetation would explain the spatial and temporal dynamics in 357 isotopic values for these species, especially for those specimens sampled after the opening of the 358 lagoon mouth. In addition to assimilating significant amounts of estuarine resources during their 359 brief periods of residence in the estuary, these freshwater fishes contribute to nutrient dynamics 360 via excretion and also delivered a pulse of biomass and energy that can be exploited by avian 361 (Bugoni et al. 2005) and mammalian (Colares & Waldemarin 2000) predators. Studies of other 362 coastal lagoons have demonstrated that a great diversity of freshwater fishes from different 363 trophic guilds (e.g. detritivores, zoobenthivores, piscivores) can colonize estuarine zones during 364 high freshwater discharges (Garcia et al. 2003). However, their potential effects on fitness, 365 feeding habits and survival of estuarine competitors and predators as well as overall effects to 366 estuarine food web organization remain largely unknown. 367 Estimates of spatial food web subsidies generally have involved asymmetric productivity 368 across aquatic-terrestrial or marine-terrestrial ecotones, with the more productive habitat being a 369 net donor (e.g. coastal marine systems, riparian forests) and less productive habitat being a net 370 recipient (e.g. arid islands, forest streams) (Polis et al. 2004). We are unaware of prior evidence 371 for reciprocal subsidies between aquatic habitats with similar productivity, such as freshwater 372 wetlands and estuaries. Whether or not inputs of wetland-derived allochthonous organic matter 373 and freshwater fish into estuaries induce trophic cascades probably depends on the relative 374 productivity of the two systems and the degree to which there are reciprocal subsidies. Potential 375 18 bottom-up and top-down effects could be further elucidated by field experiments. For example, 376 Nakano and colleagues conducted field experiments and dietary analyses to estimate reciprocal 377 subsidies across aquatic-terrestrial ecotones and their influence on trophic cascades (Nakano et al. 378 1999, Nakano & Murakami 2001). Baxter et al. (2004) demonstrated how habitat alteration and 379 introduction of alien predators suppressed flows of resources across ecotones and associated 380 reciprocal subsidies. 381 Fish guilds responded differently to the pulse of freshwater inflow and degree of 382 connectivity with the sea. Estuarine-dependent (e.g. Brevoortia pectinata, Micropogonias 383 furnieri, Paralichthys orbignyanus) and marine-vagrant (e.g. Diapterus rhombeus, Eucinostomus 384 melanopterus, Menticirrhus littoralis, Trachinotus marginatus, Ulaema lefroyi) fishes were 385 absent in mixohaline zones (estuary, lagoon mouth) during the latter stages of flooding when 386 characids, poeciliids and other freshwater fishes were prevalent. Marine fishes responded to the 387 freshwater pulse by moving southward toward the sea, but they could not exit the lagoon because 388 sand dunes blocked the outlet. Several species from the marine guild (e.g. croakers, mullets, 389 silversides) were observed in high densities within shallow pools along the margins of the outlet 390 zone. Mass mortality of marine species sometimes occurs during periods of drought as well as 391 periods of freshwater expansion into mixohaline zones (AMG, personal observation) and would 392 explain the absence of estuarine-dependent and marine-vagrant fish when the lagoon’s mouth 393 was closed. Similar to other coastal lagoons in the region (Garcia et al. 2012), marine fishes 394 returned to the mixohaline zones soon after the connection to the sea was opened with machinery. 395 Because we did not sample marine habitat outside the lagoon outlet, we can only speculate about 396 effects of estuarine discharge of resources on the nearshore marine food web. Using stable 397 isotope analysis, Savage et al. (2012) showed that freshwater primary production subsidizes 398 19 suspension-feeding bivalves in coastal waters of New Zealand when high flow pulses transport 399 material from the estuary to the adjacent coast. 400 High resolution temporal sampling of food web components was crucial to reveal how 401 pulsed freshwater inflows affect trophic subsidies between zones of a coastal lagoon ecosystem. 402 All estuarine systems experience hydrologic pulsing, with patterns varying from seasonal (e.g. 403 wet/dry tropics, temperate regions with spring snowmelt) to highly variable and relatively 404 unpredictable regimes (dryland rivers). Appropriate sampling designs are required to reveal 405 spatiotemporal patterns of food web dynamics caused by hydrologic pulsing (Yang et al. 2008). 406 Resource pulses occur at multiple scales, and a local subsidy for one species (e.g. short-lived, 407 sedentary organism) may be insignificant for another (e.g. long-lived, migratory organism) 408 (Zackrisson et al. 1999, Yang et al. 2008). Therefore, in order to quantify resource pulses, it is 409 crucial to define relevant spatial and temporal scales matching resource-consumer interactions. 410 Because we sampled food web components with greater frequency (monthly) compared with 411 most prior studies (e.g. pre- vs. post-flood or seasonal sampling), our analysis was able to reveal 412 the timing, magnitude and duration of trophic subsidies along the longitudinal fluvial gradient. 413 Analysis of fish guilds also allowed us to identify pulses at ecologically relevant scales in space 414 and time. 415 Our findings provide evidences that human interference with the connectivity of coastal 416 wetlands, lagoons and the sea can greatly affect food web dynamics, including trophic subsidies 417 among their compartments. Mechanical opening the lagoon outlet terminated the flood pulse and 418 reduced freshwater subsidies delivery to the estuarine zone, but it also allowed estuarine-419 dependent species to access critical habitat for early life stages. Excavation of the Lagoa do 420 Peixe outlet channel has been done multiple times over the past 150 years to allow entrance of 421 20 shrimp (Farfantepenaeus paulensis) from the sea into the lagoon and to drain water from the 422 floodplain to increase pasture for cattle (Lanés et al. 2015). Ecological responses to artificial 423 connections between coastal lagoons and marine systems are difficult to predict, and fish species 424 diversity may decline (Saad et al. 2002), increase (Griffiths 1999), or change little (Lanés et al. 425 2015). There is general concern that artificial opening of sandbars in coastal lagoons alters 426 natural ecological dynamics with potential negative effects on biodiversity and ecosystem 427 services (Griffiths 1999, Saad et al. 2002). Our study contributes to this discourse by 428 demonstrating how the timing of freshwater inflows and the outlet opening affect food web 429 dynamics in different regions of the lagoon system. 430 Future studies of coastal ecotones could examine the influence of pulsing allochthonous 431 inputs on the distribution of trophic interactions of varying strength within the regional food web. 432 Prior theoretical and empirical work suggests that greater proportions of weak trophic 433 interactions enhance stability in species-rich food webs, and strong trophic interactions promote 434 destabilizing oscillations (Kokkoris et al. 2002, Cross et al. 2013). Landscape mosaics, including 435 fluvial ecosystems and other ecotones, might have greater proportions of weak links within the 436 regional food web when there is periodic exchange of organisms among habitat compartments 437 (Bellmore et al. 2015). Research that combines comparative and experimental approaches to 438 estimate interaction strength could test this hypothesis and further elucidate ecological responses 439 to pulsing hydrology and food web subsidies. 440 441 Acknowledgements 442 This study received financial support from Brazil’s Conselho Nacional de Desenvolvimento 443 Científico e Tecnológico, CNPq (Grant No. 482920/2007-6) and International Foundation of 444 21 Science, IFS (Grant No.A/4419-1). AMG acknowledges fellowship support from CNPq 445 (305888/2012-9) and the ICMBIO for providing permit (14523-2 and 14523-4) for sample 446 collections. KOW acknowledges support from the US National Science Foundation (DEB 447 1257813 and IGERT 0654377). This is a contribution of the Grupo de Análises de Isótopos 448 Estáveis em Ambientes Aquáticos (GAIA/CNPq). 449 450 451 Literature Cited 452 Abrantes KG, Barnett A, Bouillon S (2013) Stable isotope-based community metrics as a tool to 453 identify patterns in food web structure in east African estuaries. Funct Ecol 28:270-282 454 Baxter CV, Fausch KD, Murakami M, Chapman PL (2004) Fish invasion restructures stream and 455 forest food webs by interrupting reciprocal prey subsidies. Ecology 85:2656–2663 456 Bellmore JR, Baxter CV, Connolly PJ (2015) Spatial complexity reduces interaction strengths in 457 the meta-food web of a river floodplain mosaic. Ecology 96:274-283 458 Bugoni L, Cormons TD, Boyne AW, Hays H (2005) Feeding Grounds, Daily Foraging Activities, 459 and Movements of Common Terns in Southern Brazil, Determined by Radio-telemetry. 460 Waterbirds 28(4): 468-477 461 Chick JH, Ruetz III CR, Trexler JC (2004) Spatial scale and abundance patterns of large fish 462 communities in freshwater marshes of the Florida Everglades. Wetlands 24(3): 652–664 463 Childers DL (2006) A synthesis of long-term research by the Florida Coastal Everglades LTER 464 Program. Hydrobiologia 569:531–544 465 22 Claudino MC, Abreu PC, Garcia AM (2013) Stable isotopes reveal temporal and between-habitat 466 changes in trophic pathways in a southwestern Atlantic estuary. Mar Ecol Progr Ser 489:29-467 42 468 Claudino MC, Pessanha ALM, Araújo FG, Garcia AM (2015) Trophic connectivity and basal 469 food sources sustaining tropical aquatic consumers along a mangrove to ocean gradient. 470 Estuar Coast Shelf Sci 167:45-55 471 Colares EP, Waldemarin HF (2000) Feeding of the Neotropical River Otter (Lontra longicaudis) 472 in the Coastal Region of the Rio Grande do Sul State, Southern Brazil IUCN Otter Spec. 473 Group Bull. 17 (1): 6 - 13 474 Cross WF, Baxter CV, Rosi-Marshall EJ, Hall RO, Kennedy TA, Donner KC, Wellard Kelly HA, 475 Seegert SEZ, Behn KE, Yard MD (2013) Food-web dynamics in a large river discontinuum. 476 Ecol Monogr 83:311–337 477 DeAngelis DL, Loftus WF, Trexler JC, Ulanowicz RE (1997) Modeling fish dynamics and 478 effects of stress in a hydrologically pulsed ecosystem. J Aquat Ecosyst Stress Recovery 6: 1–479 13 480 DeAngelis DL, Trexler JC, Loftus WF (2005) Life history trade-offs and community dynamics 481 of small fishes in a seasonally pulsed wetland. Canadian J Fish Aquat Sci 62: 781–790 482 Flecker AS, Mcintyre PB, Moore JW, Anderson JT, Taylor BW, Hall RO. (2010) Migratory 483 fishes as material and process subsidies in riverine ecosystems. American Fisheries Society 484 Symposium 73:559–592 485 Garcia AM, Vieira JP, Winemiller KO (2001) Dynamics of the shallow-water fish assemblage of 486 the Patos Lagoon estuary (Brazil) during cold and warm ENSO episodes. J Fish Biol 487 59:1218–1238 488 23 Garcia AM, Raseira MB, Vieira JP, Winemiller KO, Grimm AM (2003) Spatiotemporal 489 variation in shallow-water freshwater fish distribution and abundance in a large subtropical 490 coastal lagoon. Environ Biol Fish 68:215–228 491 Garcia AM, Vieira JP, Winemiller KO, Grimm AM (2004) Comparison of the 1982-1983 and 492 1997-1998 El Niño effects on the shallow-water fish assemblage of the Patos Lagoon estuary 493 (Brazil). Estuaries 27:905-914 494 Garcia AM, Bemvenuti MA, Vieira JP, Marques DMLM, Burns MDM, Moresco A, Condini 495 MVL (2006) Checklist comparison and dominance patterns of the fish fauna at Taim 496 Wetland, South Brazil. Neotrop Ichthyol 4: 261–268 497 Garcia AM, Hoeinghaus DJ, Vieira JP, Winemiller KO (2007) Isotopic variation of fishes in 498 freshwater and estuarine zones of a large subtropical coastal lagoon. Estuar Coast Shelf S 499 73:399-408 500 Garcia AM, Vieira JP, Winemiller KO, Moraes LE, Paes ET (2012) Factoring scales of spatial 501 and temporal variation in fish abundance in a subtropical estuary. Mar Ecol Prog Ser 502 461:121-135 503 Garcia AM, Claudino MC, Mont’alverne R, Pereyra PER, Copertino M, Vieira JP (in press) 504 Temporal variability in food assimilation of basal food sources by an omnivorous fish at 505 Patos Lagoon Estuary revealed by stable isotopes (2010-2014). Mar Biol Res 506 10.1080/17451000.2016.1206939 507 Griffiths SP (1999) Consequences of artificially opening coastal lagoons on their fish 508 assemblages. Inter J Salt Lake Research 8:307-327 509 Hoeinghaus DJ, Vieira JP, Costa CS, Bemvenuti CE, Winemiller KO, Garcia AM (2011) Estuary 510 hydrogeomorphology affects carbon sources supporting aquatic consumers within and among 511 24 ecological guilds. Hydrobiologia 673:79-92 512 Holt RD (2008) Theoretical perspectives on resource pulses. Ecology 89:671-681 513 Jardine TD, Pettit NE, Warfe DM, Pusey BJ, Ward DP, Douglas MM, Davies PM. Bunn SE 514 (2012) Consumer–resource coupling in wet–dry tropical rivers. J Anim Ecol 81: 310–322 515 Kaymak N, Winemiller KO, Akin S, Altuner Z, Polat F, Dal T (2015) Stable isotope analysis 516 reveals relative influences of seasonal hydrologic variation and impoundment on assimilation 517 of primary production sources by fish in the Upper Yesilırmak River, Turkey. Hydrobiologia 518 753:131–147 519 Knak RB (2004) Management plan of Lagoa do Peixe National Park (in Portuguese). 520 IBAMA/FNMA/FURG/ NEMA/UFPel, Brasília. 465 p. 521 Kokkoris GD, Jansen VAA, Loreau M, Troumbis AY (2002) Variability in interaction strength 522 and implications for biodiversity. J Anim Ecol 71:362–371 523 Koshino Y, Kudo H, Kaeriyama M (2013) Stable isotope evidence indicates the incorporation 524 into Japanese catchments of marine-derived nutrients transported by spawning Pacific 525 Salmon. Freshwater Biol 58:1864–1877 526 Krenz C, Menge BA, Freidenburg TL, Lubchenco J, Chan F, Foley MM, Nielsen KJ (2011) 527 Ecological subsidies to rocky intertidal communities: linear or non-linear changes along a 528 consistent geographic upwelling transition? J Exp Mar Biol Ecol 409:361–370 529 Lanés LEK, Rolon AS, Stenert C, Maltchik L (2015) The effects of annual artificial opening of a 530 natural sandbar of a lagoon system on the fish community. J Appl Ichthyol 31:321-327 531 Leibhold A, Koenig WD, Bjørnstad ON (2004) Spatial synchrony in population dynamics. 532 Annual Review of Ecology and Systematics 35:467-490 533 Maltchik L, Lanés LEK, Stenert C, Medeiros ESF (2010) Species-area relationship and 534 25 environmental predictors of fish communities in coastal freshwater wetlands of southern 535 Brazil. Environ Biol Fish 88:25–35 536 Marshall JD, Brooks JR, Lajtha K (2007) Sources of variation in the stable isotopic composition 537 of plants. In: Mitchener R, Lajtha K (eds) Stable isotopes in ecology and environmental 538 science. Blackwell Scientific, Oxford, p 22-60 539 Mattson WJ, Addy ND (1975) Phytophagous insects as regulators of forest primary production. 540 Science 190:515-522 541 McCormick MI (2003) Consumption of coral propagules after mass spawning enhances larval 542 quality of damselfish through maternal effects. Oecologia, 136:37-45 543 Myers GS (1938) Freshwater fishes and West Indian zoogeography. Annual Report of the Board 544 of Regents of the Smithsonian Institution 92: 339–364 545 Naiman RJ, Bilby RE, Schindler DE, Helfied JM (2002) Pacific salmon, nutrients, and the 546 dynamics of freshwater and riparian ecosystems. Ecosystems 5:399-417 547 Nakano S, Murakami M (2001) Reciprocal subsidies: dynamic interdependence between 548 terrestrial and aquatic food webs. P Natl Acad Sci 98:166–170 549 Nakano S, Miyasaka H, Kuhara N (1999) Terrestrial-aquatic linkages: riparian arthropod inputs 550 alter trophic cascades in a stream food web. Ecology 80:2435–2441 551 Oliveira MCLM, Bastos RF, Claudino MC, Assumpção CM, Garcia AM (2014) Transport of 552 marine-derived nutrients to subtropical freshwater food webs by juvenile mullets: a case 553 study in southern Brazil. Aquat Biol 20:91-100 554 Ostfeld RS, Keesing F (2000) Pulsed resources and community dynamics of consumers in 555 terrestrial ecosystems. Trends Ecol Evol 15:232-237 556 Ou C, Winemiller KO (2016) Seasonal hydrology shifts production sources supporting fishes in 557 26 rivers of the Lower Mekong Basin. Canadian Journal of Fisheries and Aquatic Sciences, 558 10.1139/cjfas-2015-0214 559 Parnell AC, Inger R, Bearhop S, Jackson AL (2010) Source partitioning using stable isotopes: 560 coping with too much variation. PLoS ONE 5:e9672 561 Petersen JK, Hansen JW, Laursen MB, Clausen P, Carstensen J, Conley DJ (2008) Regime shift 562 in a coastal marine ecosystem. Ecol Appl 18(2): 497–510 563 564 Peterson BJ, Fry B (1987) Stable isotopes in ecosystem studies. Annu Rev Ecol Evol Syst 565 18:293-320 566 Petraitis PS, Latham RE (1999) The importance of scale in testing the origins of alternative 567 community states. Ecology 80:429-442 568 Poff NL, Allan JD, Bain MB, Karr JR, Prestegaard KL, Richter BD, Sparks RE, Stromberg JC 569 (1997) The natural flow regime: a paradigm for river conservation and restoration. 570 Bioscience 47:769-784 571 Polis GA, Power ME, Huxel GR eds. (2004) Food webs at the landscape level. University of 572 Chicago Press, Chicago, Illinois, USA. 573 Polis GA, Anderson WB, Holt RD (1997) Toward an integration of landscape and food web 574 ecology: the dynamics of spatially subsidized food webs. Annu Rev Ecol Syst 28:289-316 575 Pratchett MS, Gust N, Goby G, Klanten SO (2001) Consumption of coral propagules represents 576 a significant trophic link between corals and reef fish. Coral Reefs 20:13-17 577 Rolon AS, Rocha O, Maltchick L (2011) Diversity of aquatic macrophytes in the Lagoa do Peixe 578 National Park. Neotrop Biol Conserv 6(1):5-12 579 Rozas LP, Minello TJ (1997) Estimating densities of small fishes and decapod crustaceans in 580 27 shalow estuarine habitats: a review of sampling design with focus on gear selection. 581 Estuaries, 20: 199-213 582 Saad AM, Beaumond AC, Caramaschi EP (2002) Effects of artificial canal openings on fish 583 community structure of Imboassica coastal lagoon, Rio de Janeiro, Brazil. J Coastal Res 584 36:634–639 585 Savage C, Thrush SF, Lohrer AM, Hewitt JE (2012) Ecosystem services transcend 586 boundaries: estuaries provide resource subsidies and influence functional diversity in 587 coastal benthic communities. PLoS ONE: 7:e42708. 588 Spiller DA, Piovia-Scott J, Wright AN, Yang LH, Takimoto G, Schoener TW, Iwata T 589 (2010) Marine subsidies have multiple effects on coastal food webs. Ecology 91:1424-590 1434 591 Vanderklift MA, Ponsard S (2003) Sources of variation in consumer-diet d15N enrichment: a 592 meta-analysis. Oecologia 136:169-182 593 Warfe DM, Pettit NE, Davies PM, Pusey BJ, Hamilton SK, Kennard MJ, Townsend SA, Bayliss 594 P, Ward DP, Douglas MM, Burford MA, Finn M, Bunn SE, Halliday IA (2011) The ‘wet-595 dry’ in the wet-dry tropics drives ecosystem structure and function of northern Australian 596 rivers. Freshw Biol 56: 2169–2195 597 Winemiller KO (1990) Spatial and temporal variation in tropical fish trophic networks. Ecol 598 Monogr 60:331-367Winemiller KO, Layman CA (2005) Food web science: moving on the 599 path from abstraction to prediction. In: Ruiter PC, Wolters V, Moore JC (eds) Dynamic Food 600 Webs: Multispecies Assemblages, Ecosystem Development and Environmental Change. 601 Elsevier, Amsterdam, Netherlands, p 10-23 602 28 Yang LH (2004) Periodical cicadas as resource pulses in North American forests. Science 603 306:1565-1567 604 Yang LH, Bastow JL, Spence KO, Wright AN (2008) What can we learn from resource pulses? 605 Ecology 89:621-634 606 Zackrisson O, Nilsson MC, Jaderlund A, Wardle DA (1999) Nutritional effects of seed fall 607 during mast years in boreal forest. Oikos 84:17-26 608 Zar JH (1984) Biostatistical analysis. 2nd ed. Prentice-Hall, Englewood Cliffs, New Jersey, USA 609 610 29 Figure and legends 611 Figure 1. Upper. Panel. Map showing the Lagoa do Peixe National Park (LPNP) in the coastal 612 plain of southern Brazil with the location of the sampled areas (squares) at the freshwater 613 wetland (1), at the estuarine zone (2) and at the lagoon’s mouth (3). Red dashed line denotes the 614 boundaries of the national park. Pictures taken at the mouth of the lagoon on April 2 th and 615 November 11 th 2008 showing the status of its connection with the sea (C and D) and in the same 616 location of the estuarine zone showing the flooded (A) and non-flooded conditions (B). Photos 617 by A.M.Garcia. Lower panel. Monthly variation in rainfall (mm, bars) and water temperature ( o C, 618 line) during the study period (April 2008 – May 2009). 619 620 Figure 2. Mean (+SD) of  15 N and  13 C values for primary producers and organic sources (open 621 circles numbered 1 to 4), prey and fishes (left panel) and relative contributions of basal food 622 sources (C3 and C4 plants, seston and microphytobentos-MiPhBe) to fishes at the freshwater 623 wetland, estuarine zone and lagoon mouth sites (right panel). 624 625 Figure 3. Monthly variation in the  15 N and  13 C values for fish guilds (primary freshwater– 626 open circles; secondary freshwater– open triangles; estuarine dependent– grey circles; estuarine 627 resident– grey triangles; marine vagrants– dark squares) from the freshwater wetland, estuarine 628 zone and lagoon mouth. Vertical bar on the right denotes flood status within the estuarine zone. 629 630 Figure 4. Monthly relative contributions of basal food sources (C3 and C4 plants, seston and 631 microphytobentos) to biomass of estuarine-resident fish (Atherinella brasiliensis, Odonthestes 632 argentinensis) (left panel) and primary freshwater fish (Astyanax eigenmanniorum, A. fasciatus, 633 30 Cheirodon interruptus, Phalloceros caudimaculatus) from the estuarine zone during flooding 634 and non-flooding conditions. Each plot shows the 50 (black), 75 (gray) and 95% (lighter gray) 635 Bayesian credibility intervals of the feasible contributions of each basal production source to fish 636 in each month. 637 638 639 640 641 642 31 643 644 Figure 1 645 646 647 32 648 Figure 2 649 650 33 651 652 Figure 3 653 654 34 655 Figure 4 656 657 35 658 Supplemental material 659 660 Appendix A. δ 13 C and δ 15 N values (X±SD) and total length (TL, mm) of fishes, invertebrates and 661 plants and organic sources at the freshwater wetland, estuarine zone and lagoon mouth of the 662 Lagoa do Peixe National Park. n, sample size. Functional guilds (GUI): primary freshwater (PF), 663 secondary freshwater (SF), estuarine resident (ER), estuarine dependent (ED), marine vagrants 664 (MV). 665 666 Appendix B. Monthly variation in the  15 N and  13 C values for basal food sources 667 (microphytobenthos, open circles; seston, open triangles; C3 plants, grey circles; C4 plants) at the 668 freshwater wetland, estuarine zone and lagoon mouth sites. Vertical bar on the right denotes 669 status of the flooding conditions in the estuarine zone. 670 671 36 SUPPLEMENTAL MATERIAL 672 Appendix A 673 GUI n TL 13C 15N n TL 13C 15N n TL 13C 15N Fishes Astyanax eigenmanniorum PF 34 56.8 ± 16.5 -26.0 ± 2.6 8.2 ± 1.5 8 59.8 ± 11.8 -21.1 ± 2.6 10.1 ± 1.9 1 76.0 -21.1 11.3 Astyanax fasciatus PF 5 56.5 ± 24.4 -25.4 ± 2.8 7.4 ± 1.3 4 35.7 ± 8.8 -17.8 ± 1.8 10.0 ± 0.3 Astyanax jacuhiensis PF 12 57.1 ± 25.6 -29.1 ± 3.2 7.4 ± 0.8 1 45.0 -21.5 8.9 Astyanax sp PF 1 47.0 -28.4 7.8 Callichthys callichthys PF 12 106.1 ± 39.0 -26.0 ± 3.6 7.7 ± 0.9 Characidium rachovii PF 2 40.5 ± 10.7 -28.7 ± 5.7 6.3 ± 1.1 Cheirodon ibicuhiensis PF 2 33.5 ± 13.4 -25.1 ± 3.4 9.4 ± 0.6 1 47.0 -19.2 8.3 Cheirodon interruptus PF 20 35.0 ± 6.8 -24.9 ± 4.0 8.7 ± 1.4 11 32.4 ± 6.0 -19.3 ± 2.6 9.5 ± 0.7 Cnesterodon decemmaculatus PF 1 35.0 -21.1 9.0 Corydoras paleatus PF 12 47.9 ± 12.7 -27.5 ± 2.2 7.1 ± 1.0 Cynopoecilus melanotaenia PF 3 43.2 ± 12.9 -24.8 ± 1.0 7.0 ± 0.8 1 29.0 -21.4 7.3 Cyphocharax saladensis PF 7 50.6 ± 10.2 -30.8 ± 3.4 5.1 ± 0.9 Cyphocharax voga PF 10 79.9 ± 47.6 -27.9 ± 3.9 6.1 ± 2.1 Gymnotus carapo PF 1 273.0 -27.0 8.0 Hoplias malabaricus PF 45 202.8 ± 91.9 -25.0 ± 1.8 8.8 ± 1.4 Hoplosternum littorale PF 1 212.0 -27.7 8.2 Hyphessobrycon bifasciatus PF 24 36.7 ± 10.6 -27.6 ± 2.1 6.8 ± 1.0 Hyphessobrycon boulengeri PF 18 44.0 ± 8.8 -26.4 ± 1.6 7.0 ± 1.2 1 43.0 -17.0 8.1 Hyphessobrycon luetkenii PF 4 28.0 ± 7.7 -28.1 ± 2.6 7.3 ± 1.3 Mimagoniates inequalis PF 11 27.3 ± 5.5 -27.7 ± 1.8 8.1 ± 1.4 Oligosarcus jenynsii PF 2 109.5 ± 10.6 -25.4 ± 2.3 11.2 ± 1.5 7 81.4 ± 23.6 -19.0 ± 2.2 11.1 ± 1.8 Oligosarcus robustus PF 3 38.7 ± 10.6 -25.0 ± 1.5 6.1 ± 0.7 Phalloceros caudimaculatus PF 18 31.3 ± 6.7 -25.4 ± 2.3 7.4 ± 0.9 7 28.7 ± 8.7 -22.0 ± 2.0 9.6 ± 0.7 Pimelodella australis PF 1 53.0 -20.0 9.4 Pseudocorynopoma doriae PF 5 48.8 ± 11.3 -25.7 ± 0.4 8.0 ± 1.3 Rhamdia quelen PF 7 180.5 ± 134.4 -23.5 ± 2.5 7.2 ± 0.9 2 48.3 ± 11.0 -16.7 ± 1.1 7.3 ± 1.0 Synbranchus marmoratus PF 6 216.0 ± 120.7 -23.7 ± 2.1 7.1 ± 0.9 Australoheros facetum SF 35 83.1 ± 29.2 -26.9 ± 2.4 6.9 ± 0.8 Cichlasoma portalegrense SF 7 84.1 ± 34.6 -23.8 ± 2.4 6.4 ± 0.9 Crenicichla lepidota SF 19 127.2 ± 36.5 -27.4 ± 1.3 7.0 ± 1.2 Geophagus brasiliensis SF 15 95.7 ± 37.5 -27.9 ± 1.3 6.6 ± 0.6 Jenynsia multidentata SF 9 41.5 ± 18.3 -22.1 ± 1.7 6.8 ± 1.0 33 39.9 ± 13.2 -14.4 ± 2.5 10.0 ± 1.1 8 44.6 ± 12.2 -11.8 ± 1.6 11.8 ± 0.9 Platanichthys platana SF 1 47.0 -16.0 12.7 Atherinella brasiliensis ER 17 76.3 ± 36.4 -13.8 ± 1.4 12.0 ± 1.4 15 69.7 ± 33.4 -12.8 ± 2.6 13.4 ± 0,7 Odontesthes argentinensis ER 34 104.6 ± 84.7 -15.5 ± 3.5 12.1 ± 1.1 41 140.3 ± 76.2 -14.5 ± 2.6 13.9 ± 1.2 Brevoortia pectinata ED 11 138.1 ± 111.7 -17.0 ± 1.3 10.5 ± 2.1 9 83.6 ± 20.9 -13.5 ± 1.8 11.5 ± 0.5 Lycengraulis grossidens ED 3 117.7 ± 9.2 -14.0 ± 0.6 15.1 ± 0.3 Micropogonias furnieri ED 3 134.3 ± 76.0 -14.9 ± 2.1 12.0 ± 0.8 19 180.1 ± 97.4 -14.1 ± 1.4 15.1 ± 2.4 Mugil curema ED 8 68.0 ± 38.8 -13.3 ± 2.5 7.9 ± 1.1 Mugil platanus ED 31 135.3 ± 88.7 -13.2 ± 1.6 9.1 ± 1.2 22 133.5 ± 101.7 -12.9 ± 3.2 10.6 ± 1.7 Paralichthys orbignyanus ED 7 170.1 ± 99.3 -14.0 ± 1.0 14.0 ± 1.3 Anchoa marinii MV 1 74.0 -15.3 12.8 Diapterus rhombeus MV 2 70.0 -15.4 ± 0.1 11.3 ± 0.1 Eucinostomus gula MV 1 68.0 -15.2 11.9 Eucinostomus melanopterus MV 1 57.0 -12.2 14.4 3 90.7 ± 6.4 -14.0 ± 0.4 13.9 ± 0.6 Gereidae MV 1 152.0 -14.4 15.1 Harengula clupeola MV 1 65.0 -14.5 14.2 Hemiramphus MV 1 232.0 -14.1 14.1 Menticirrhus littoralis MV 6 278.5 ± 15.2 -14.0 ± 0.3 17.0 ± 1.3 Stellifer brasiliensis MV 4 37.3 ± 11.1 -14.4 ± 0.9 12.3 ± 0.7 Trachinotus marginatus MV 4 44.1 ± 28.1 -13.7 ± 1.2 14.1 ± 1.3 Ulaema lefroyi MV 2 59.0 ± 1.4 -12.6 ± 0.2 14.5 ± 0.1 Invertebrates Crab 4 -29.3 ± 3.2 5.6 ± 0.5 35 -13.4 ± 1.8 10.7 ± 1.5 36 -12.3 ± 1.6 11.7 ± 1.4 Gastropode 31 -27.4 ± 5.3 2.8 ± 1.6 Insect (adult) 22 -26.3 ± 5.0 3.6 ± 1.7 Insect (larvae) 17 -27.2 ± 3.6 4.0 ± 1.6 Polychaeta 20 -15.9 ± 1.2 8.5 ± 1.6 5 -14.4 ± 1.1 10.2 ± 0.6 Shrimp 1 -25.9 8.4 12 -14.1 ± 3.6 9.4 ± 1.4 12 -11.6 ± 1.5 11.5 ± 0.7 Plants and organic sources C3 plants 20 -30.5 ± 2.6 1.6 ± 1.7 18 -26.6 ± 3.2 6.0 ± 2.9 7 -26.7 ± 2.5 2.3 ± 2.6 C4 plants 4 -17.3 ± 6.3 1.7 ± 2.3 7 -15.0 ± 4.4 4.1 ± 2.0 3 -12.8 ± 0.3 8.9 ± 3.3 Seston 14 -24.8 ± 3.7 2.2 ± 2.1 19 -20.7 ± 3.4 4.3 ± 1.8 10 -19.2 ± 5.4 7.6 ± 2.3 Microphytobenthos 8 -33.6 ± 3.7 0.8 ± 1.3 20 -19.7 ± 4.6 4.3 ± 3.0 3 -16.9 ± 3.0 8.2 ± 0.6 Table S1 - 13C and  15N values (X±SD) and total length (TL, mm) of fishes, invertebrates and plants and organic sources at the freshwater wetland, estuarine zone and lagoon mouth of teh Lagoa do Peixe National Park. n, sample size. Functional guilds (GUI): primary freshwater (PF), secondary freshwater (SF), estuarine resident (ER), estuarine dependent (ED), marine vagrants (MV). Freshwater wetland Estuarine zone Lagoon mouth 674 675 37 SUPPLEMENTAL MATERIAL 676 Appendix B 677 678 679 View publication statsView publication stats https://www.researchgate.net/publication/312593226