Source: CORNELL UNIVERSITY submitted to
SEED DISPERSAL BY FRUIT-EATING FISH IN A NEOTROPICAL FLOODPLAIN FOREST
Sponsoring Institution
State Agricultural Experiment Station
Project Status
TERMINATED
Funding Source
Reporting Frequency
Annual
Accession No.
0206874
Grant No.
(N/A)
Project No.
NYC-183322
Proposal No.
(N/A)
Multistate No.
(N/A)
Program Code
(N/A)
Project Start Date
Apr 1, 2006
Project End Date
Sep 30, 2007
Grant Year
(N/A)
Project Director
Flecker, A. S.
Recipient Organization
CORNELL UNIVERSITY
(N/A)
ITHACA,NY 14853
Performing Department
ECOLOGY & EVOLUTIONARY BIOLOGY
Non Technical Summary
This study will examine whether large-bodied, fruit-eating fishes known as pacu are key seed dispersal agents and critical to the regeneration biology of Amazonian floodplain forests. The purpose of this research is to understand whether the structure and function of flooded forests are dependent on the well-being of fishes that are heavily exploited throughout much of tropical South America. The purpose of this study is to test the hypothesis that seed dispersal by fruit-eating fishes is critical to the regeneration biology of floodplain forests of the Peruvian Amazon.
Animal Health Component
40%
Research Effort Categories
Basic
60%
Applied
40%
Developmental
(N/A)
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
1120330107050%
1350899107050%
Goals / Objectives
The broad objectives of this study are to test the hypothesis that seed dispersal by fruit-eating fishes is critical to the regeneration biology of floodplain forests of the Peruvian Amazon. Specifically, we will: (1) radio tag wild-caught, fruit-eating fish known as pacu (Characidae: Colossoma macropomum) to evaluate both the distance and the directionality of seed dispersal by fish, (2) investigate the abundance and diversity of seeds dispersed by fruit-eating fishes, and (3) test whether seeds germinate earlier and grow faster when they have passed through the digestive systems of living fish than when they are harvested directly from trees. Together this data will allow us to assess the capacity of fruit-eating fishes to be potent seed-dispersal agents in Amazonian flooded forests. Results from this study are important for understanding the functional significance of fruit-eating fishes in Amazonian floodplain forests, which is vital for the conservation and management of these unique species and ecosystems.
Project Methods
We will use three main approaches to examine the importance of fruit-eating fishes as seed dispersal agents in Amazonian floodplain forests: (I) Diet analysis of pacu (Colossoma macropomum). Throughout the flooded season, individual pacus will be captured every week using trotlines baited with fruits in the flooded forests. This fishing technique is used in some parts of South America and allows the selective capture of large, fruit-eating fishes. We will transport healthy fish to large tanks at the Peruvian field station where we will be based, collect defecated and regurgitated seeds for the germination experiment and record the species composition and abundance of seeds present in the diets of these fishes. (II) Feeding trials and movement patterns of fruit-eating fishes. To model the distance fish disperse seeds during the flooded season, data on movement patterns of fruit-eating fishes will be combined with data on gut-passage rates of seeds. After obtaining seeds from wild-caught fish placed in tanks, we will present these individuals with fruits present in their natural diets and the time fish need to process fruits prior to defecation or regurgitation (i.e., the gut-passage rates) will be recorded. To study fine-scale spatial movement patterns, we will select a sub-sample of individuals from the feeding trials and will insert radio transmitters. Radio-tagged fishes will be released where they were captured, and we will assess the distance and directionality of fish movement patterns in the floodplains (lateral movements) and in the river channel (longitudinal movements). (III) Germination experiment. Animal dispersers can improve the germination success of seeds by removing fruit pulp. Thus, we will compare the viability of seeds passed through the digestive system of these fishes with control seeds. At least 50 seeds will be planted in each treatment: (1) seeds passed through the digestive systems of C. macropomum individuals; (2) intact fruit pulp surrounding mature seeds, no gut passage; (3) seeds from which we have manually removed the pulp, no gut passage; and (4) seeds with intact fruit pulp that have been presoaked in water for one month. We will record time to germination, percent germination and stem elongation.

Progress 04/01/06 to 09/30/07

Outputs
OUTPUTS: During the course of this study, we have tested the importance of seed dispersal by fruit-eating fish. To this end, we have: 1) assessed the abundance and diversity of seeds dispersed by fish; 2) compared the viability of seeds that have passed through the digestive system of fish; 3) established botanical transects to determine how many seeds are available to fish during lengthy flooded seasons; and 4) tracked the movement patterns of fruit-eating fish using radio telemetry to evaluate the distance and directionality of seed dispersal by fish. We conducted this research both in the field in Pacaya-Samiria National Reserve (Peru), and in controlled aquaculture settings at the Institute for the Study of the Peruvian Amazon (Iquitos, Peru). Our results indicate the fish consume and disperse massive quantities of seeds. We have found more than 800,000 intact seeds in the diets of 238 individual pacus (Colossoma macropomum) and 55 individual red pacus (Piaractus brachypomus). These seeds represented 38 species, which is nearly half of the 72 species of trees and lianas that produced fruit during the 2006 flooded season. Furthermore, fish can carry seeds up to 5.5 km. Thus, these results indicate that fish are very effective seed dispersers. To date, we have presented the results of this study at national and international scientific conferences, including the annual meeting of the Ecological Society of America (Portland, 2005), and the Latin American Conference of Botany (Dominican Republic, 2006). Additionally, we have given seminars about this research at Cornell University, the University of Pittsburgh, Utah State University, and Sam Houston State University. This research has also been the foundation of the undergraduate theses of four Peruvian students. Additionally, we have trained five Peruvian biologists in experimental design, and research techniques such as radio telemetry. PARTICIPANTS: Partner Organization: Instituto para Investigaciones de la Amazonia Peruana (Institute for the Study of the Peruvian Amazon, Iquitos, Peru) Collaborators: Joe Saldana Rojas, Institute for the Study of the Peruvian Amazon Dr. Timothy Nuttle, Assistant Professor, Indiana University of Pennsylvania Training or Professional Development: We have guided four Peruvian students through their honors theses. These students are Cahuide del Busto Rojas, Elmer Vela Morales, Percy Mejia Poquioma, and Roy Rosales Fatama. We have trained an additional five Peruvians in research methodologies.

Impacts
This study illuminates the importance of fruit-eating fish as vectors of seed dispersal; very little was known about the seed dispersal activities of fruit-eating fish prior to this research. These fish are overexploited throughout their range, and this overexploitation could influence the dynamics of regeneration in floodplain forests. The conservation implications are clear: unless we begin to conserve these fish, the structure of floodplain forests may change dramatically.

Publications

  • No publications reported this period


Progress 01/01/06 to 12/31/06

Outputs
This project examines the seed dispersal activities of two species of fruit-eating fishes in Pacaya-Samiria, a two million hectare floodplain forest reserve in northeastern Peru. In 2006, research was organized around three objectives. First, the diets of Colossoma macropomum and Piaractus brachypomus (both large bodied fish in the family Characidae) were analyzed for three months during the flooded season within the reserve. Second, seeds were collected for germination trials, and these trials were begun in the reserve. Third, 11 Colossoma macropomum were selected for a radiotelemetry study to assess their movement patterns in relation to their activities as seed dispersers. Fourth, botanical transects were established to quantify the number of tree and vine species that fruit during the flooded season. Finally, gut retention time trials were conducted in aquaria during the dry season to determine the time necessary for seeds to travel through the digestive tracts of Colossoma macropomum. Results from this year, and the two previous years of the study, indicate that fish disperse large quantities of seeds long distances. The results from this project indicate that these fishes play a crucial role in the seed dispersal ecology of floodplain forests.

Impacts
Fish constitute the most important source of animal protein for human consumption in South America; historically frugivorous fishes have been a major component of the fisheries. For example, Colossoma macropomum comprised over 40 percent of the fish sold in markets in Manaus, Brazil, in the late 1970s. Overexploitation accounts for the current threatened status of this species. Despite the importance of fruit-eating fishes in the commercial market, we do not yet understand the depth of the interaction between these fishes and floodplain trees. The results from this study suggest that the primary role of fish is to disperse seeds long distances against the prevailing water current. This work is the first to elucidate the importance of fruit-eating fish in the reproductive biology of floodplain forest trees. A thorough understanding of this interaction will compel interest in the conservation of this fruit-eating fishes and floodplain forests.

Publications

  • No publications reported this period