Laura Scott and Dr. Jerald B. Johnson, Integrative Biology
Tilapia is a commonly farmed fish throughout the world because of its nutritional value, reproductive efficiency, high resistance to disease, and the low-cost of production. However, despite such commercial benefits, there are some characteristics of this fish that are being overlooked. Previous field research in Costa Rica suggests that when tilapia escape from local farms into nearby streams, they negatively impact native fishes, both in terms of abundance and diversity (unpubl. data). The native fish most notably affected are small, livebearing poecealiids.
The purpose of this study was to better understand the mechanisms by which tilapia affect native Costa Rican fishes. In designing this study we had two objectives: 1) determine the effects of tilapia on diversity in regards to species evenness and richness, two measures of biodiversity and 2) identify the behavioral interactions between introduced tilapia and native livebearing fishes of Costa Rica.
For our first objective, evaluating tilapia’s effect on local habitat diversity, we identified a river in Costa Rica that had introduced tilapia (Fig. 1). Downstream (approximately 500 yards) was a local tilapia; upstream, the river forked and formed a small pool in one of the branches. Using seining nets, we sampled the stream repeatedly 30 times, each time identifying the different species of fish collected and counting the number of individuals from each species collected. Using this data, we were able to quantify the diversity using indices of species richness (the numbers of individuals in community) and species evenness (the relative abundance of a species in a given area). We also sampled a nearby stream that did not have any tilapia, using the same procedure to collect and analyze data.
Our results showed a significant decrease in the species richness and species evenness the stream with introduced tilapia. While more streams should be sampled to confirm the negative impact of tilapia, it is evident that the introduction of this commercial fish has potentially devastating effects to local habitats.
For the second objective, which was to better understand the behavioral interactions between tilapia and native livebearers, we designed a behavioral assay setup in the BYU Evolutionary Ecology Lab. Specifically we wanted to know if given a choice, can native livebearers distinguish the difference between tilapia (a non-carnivore predator) and a predator native to Costa Rica of similar size by demonstrating a preference for one of the other. For this experiment, we used Brachyrhaphis rhabdophora as our native livebearer test subject. Our hypothesis was that Brachyrhaphis rhabdophora would view the tilapia and guapote as equal predators and not show a preference towards either one.
We exposed a single B. rhabdophora to both an exotic tilapia and native predator (guapote) in the same 20 gallon tank. The tank was divided into three chambers by clear plexiglass, allowing visual contact only between all compartments (Fig. 2). In the middle chamber, we put a single B. rhabdophora and in each of the end compartments we placed a single native predator (guapote) and tilapia. Markings on the outer glass wall divide B. rhabdophora chamber into three regions: Left, Center, and Right. After placing all the fish in their respective compartments, we recorded the frequency and duration of time the B. rhabdophora spends in each of the left, center or right regions in 15 minutes. We used a computer program to keep track of fish position in the tank and the amount of time spent in each section and ran 8 trials, each time with a different B. rhabdophora. This behavioral assay allowed us to measure for preference the B. rhabdophora may have shown for either tilapia or guapote by noting how frequently it moved towards the tilapia compared to the guapote.
The number of visits the subject fish spent in each section is shown in Fig. 3. We tested for statistical difference in B. rhabdophora preference for tilapia versus guapote. Results showed no significant difference between the amounts of time B. rhabdophora spent between either the tilapia or guapote side of the tank (p-value 0.9173) (Fig. 3). This could indicate that livebearers view tilapia the same as guapote- a predator. If so, their activity levels would likely decrease and in turn negatively affecting growth, feeding, and reproduction which may account for the loss of diversity found in Costa Rican streams. Further experiments are needed to explore the activity levels of B. rhabdophora in the presence of tilapia.
While the overall project/data gave us a better understanding of the affects tilapia have on native livebearers of Costa Rica, we did encounter some unexpected problems/difficulties in the implementation of this project.
For one, we were only able to identify one stream in Costa Rica that had introduced tilapia. This was due to time constraints, difficulties in finding tilapia farms near streams, and the season we were in Costa Rica. Second, in the behavioral assay the handling and transportation of fish from their home tank to the experimental tank could have caused the fish to become scared and less active because they were in a new environment and not because they were in the presence of a predator fish. Thus, our results could have been confounded by this.
Further experiments should include the sampling of more streams in Costa Rica to confirm the problem is consistent throughout multiple locations. Also, a behavioral assay that eliminates any nervousness the fish may experience when being moved from one tank to another would produce more accurate results.