A Fishy Situation: How Hurricanes Shrank Fish in Puerto Rico

When Shayan Saniei was searching for an undergraduate research project at Pace University, his biostatistics professor, Dr. Erika Crispo, suggested he look into data available from the NEON program. He found a compelling topic to study: looking at fish body size data in Puerto Rico before and after hurricanes. The resulting study was published in BioOne Complete in December 2022: The effect of hurricanes on body size in Puerto Rico fish populations.

Fish, Hurricanes, and Eco-evolution

Two major hurricanes struck Puerto Rico in 2017: Irma and Maria. These hurricanes caused catastrophic damage across the island, with widespread impacts on humans, wildlife, and ecosystems. Other researchers had already studied factors such as soil health and marine populations in the aftermath of Maria, but Saniei found a dearth of data for freshwater aquatic vertebrates.

It's both an intriguing and important question: how did the recent hurricanes in Puerto Rico impact the average body size of freshwater fish found in Puerto Rican rivers? Saniei and Crispo used data on fish body size from the two aquatic sites in the Atlantic Neotropical Domain (D04): Río Cupeyes (CUPE) and Río Yahuecas (GUIL) (formerly Río Guilarte). They compared the data from before and after the hurricanes for four freshwater species: western mosquitofish (Gambusia affinis), green swordtail (Xiphophorus hellerii), guppy (Poecilia reticulata), and mountain mullet (Agonostomus monticola).

The study dives into aspects of eco-evolution. Eco-evolution (or eco-evolutionary dynamics) looks at how organisms and their environment change together over time. Changing conditions—such as rising temperatures, shifting precipitation patterns, the introduction of an invasive species, or resource availability—influence the direction of this change. This happens through natural selection; as the environment changes, different traits give advantages. Evolutionary changes in a species may, in turn, impact other species and the wider environment. Eco-evolutionary studies seek to tease out the drivers and impacts of evolution.

In fish, factors such as temperature and oxygen availability impact growth and average body sizes for a species in ways that can be measured in relatively short time periods. Saniei wanted to see if there were any measurable and significant changes in mean fish body size before and after the two recent hurricanes.

"Hurricanes have become stronger and more frequent [in this region], and that affects not only people but also plants, animals, and the whole ecosystem," Saniei explains. "We know that environmental conditions can impact body size for fish, as well as behavior and other traits. It's important to understand both the short- and long-term impacts of climate change, marine water temperature changes, and tropical storms and hurricanes on fish."

What Made These Fish Shrink?

The study uses fish body length data for the four species of fish sampled by the NEON program for a one-year period, including sampling data both before and after the 2017 hurricanes. Saniei and Crispo found that the mean body length was smaller after the hurricanes for all four species compared to pre-hurricane data. However, the change was only statistically significant for two species: green swordtail (X. helleri) and mountain mullet (A. monticola).

Although the study was not designed to identify drivers and mechanisms for the changes, Saniei and Crispo have some theories. Large hurricanes like Irma and Maria have a range of significant impacts on the environment, both during and after the event. There may be temporary changes in water temperature, salinity, turbidity, sedimentation, or nutrient availability due to torrential rains and flooding, high winds, and mixing of marine waters into freshwater environments due to storm surges. Large storms can also have both temporary and permanent impacts on stream morphology and flow rates.

These changes may impact fish populations in a variety of ways, potentially changing both physiology and behavior. For example, different conditions may favor smaller body size, directly impacting the survival rates and reproductive success for fish of different sizes. Or it may be that the storms impact fish behavior, perhaps driving larger fish into areas where they are less likely to be caught, or changing the timing of reproduction, resulting in a smaller and younger population at the sample time. Environmental stressors can increase blood cortisol levels in fish, which may impact swimming performance, metabolism, and survival rates.

Displacement and dispersion may also be factors. Crispo notes that larger fish may be more likely to be flushed out to sea from their river habitats than their smaller counterparts, which may be better able to hide in small crevices that provide protection. Larger fish could also be simply relocating to friendlier waters; a North Carolina study found that fish migration and movement increased up to 2550% in response to hurricane conditions.

Crispo says, "The results are interesting, but the question is, why did this change occur? To get to those answers, we would need to have a comparison group not impacted by the hurricane and a more extensive data set."

Whatever the reason, the study does suggest that hurricanes can impact the phenotypic composition of freshwater fish communities. It also provides a starting point for future studies into the eco-evolutionary effects of hurricanes. Saniei says, "Evolutionary biology takes a long time to study. Shorter studies like this provide preliminary data that can give us hints into how a species may be changing and provide a basis for longer-term studies."

Saniei hopes that other researchers will carry the work forward to dig into the drivers of evolutionary change in fish, both in the neotropical Atlantic and elsewhere. Fish body sizes have been shown to be shrinking globally, most likely in response to rising average water temperatures. These findings could have implications for conservation as well as direct impacts on fisheries and local economies.

Fish sampling at GUIL

Crispo says, "Populations are always fluctuating, both phenotypically and genetically. Species are always changing, and they are always going to change. But right now, we don't yet understand all the drivers of these changes. This kind of research is important to inform conservation managers about how they should be restoring populations, for example. People think that within a species, all members are interchangeable, but that is not the case. Within a species, some members will be better adapted to certain environmental conditions than others."

Enabling Remote Science and Student Inquiry

The study started as a summer research project funded through a competitive undergraduate student grant program at Pace University. Saniei became interested in biostatistics and data science while taking Crispo's class and wanted to continue working with ecological data. The NEON program proved to be an ideal resource for his project. Crispo served as his mentor for the study and co-authored the eventual paper.

Saniei began his research in the summer of 2020, as COVID-19 shut down field research opportunities across the country. The NEON data portal provided a free and open source of data that enabled him to conduct research from his living room. "I really appreciated having something big like this to work on that summer," he says. "I spent all my time exploring the data, learning to code, having remote meetings on data analysis, reading other papers and writing my paper." He appreciates the data science skills he developed that summer, including using GitHub.

He says, "NEON is a great resource. Most people think of ecology as field scientists collecting data on the ground, but there is an immense opportunity to use the data that is already collected and published. You can use data mining to conduct studies and publish papers on so many different topics—it's really infinite."

In part as a result of his study experience, Saniei changed his academic path from pre-med to data science and bioinformatics. He graduated from Pace in 2022 and is now applying his data science skills in the field of medicine. Thanks to the Pace grant program, Crispo, and NEON data, he graduated from his undergraduate program with his first authored paper. "It's rare for undergraduate students to have that opportunity, so I'm very grateful," he says. "Dr. Crispo has been a great mentor for me, and I learned so much through the experience." Saniei graduated with the highest GPA in the Pace Biology department for his year, earning the department's Scholastic Achievement Award.

Crispo anticipates that more of her students will find opportunities to use NEON data in the future. "NEON data are great for these short-term student projects," she says. "For a summer award project or capstone project, there isn't a lot of time—you don't have two or three years to collect your own data for a data analysis project. I will definitely recommend NEON as a potential option for my students." Saniei's abstract now hangs on her door as inspiration for future students.