Menidia menidia

ICES Journal of Marine Science publishes long-term fecundity study!


2 November 2021. We are happy report that the ICES Journal of Marine Science just published the last major experimental paper on Atlantic silverside CO2-sensitivity from our lab. Callie Concannon and co-authors report on two complementary, long-term rearing trials in 2015/16 and 2018/19, where silverside juveniles or newly fertilized embryos were reared under contrasting temperature and CO2 conditions to maturity. This revealed negative effects of high CO2 conditions on female fecundity, but only at the warm, not the cold temperature treatments (Fig. below). Our study and its data are novel, because they were generated by the first whole-life CO2 rearing experiment of a fish and are the first empirical fecundity effects shown for a broadcast-spawning fish species.

The paper is also special to us, because its publication marks the erstwhile conclusion of our yearlong, NSF-funded efforts (OCE#1536165) to understand the CO2 sensitivity and its mechanisms in this important forage fish and long-standing model in fish ecology and evolution. The project ran from 2015 - 2020, produced 15 publications, 2 book chapters, and over 40 presentations, while furthering the careers of a post-doc, a PhD student, 5 Master students and over 10 undergraduates.

[New Publication] MEPS publishes Julie’s Ms research on silverside otoliths

12 December 2019. We are happy to announce that Marine Ecology Progress Series just published our latest paper on Atlantic silversides, but this time not an experimental but a field study! During her time in our lab, Julie Pringle investigated the otolith microstructure of young-of-year silversides, finding intriguing patterns about differential growth in males and females that likely result in sex-selective survival during their growing season. Congratulations, Julie, well done!

Pringle, J.W. and Baumann, H. (2019) Otolith-based growth reconstructions in young-of-year Atlantic silversides (Menidia menidia) and their implications for sex-selective survival. Marine Ecology Progress Series 632:193-204

This graph shows reconstructed hatch distributions of male and female Atlantic silversides sampled in fall 2015. Counting daily otolith increments, young-of-year fish caught in October could be reliably aged, whereas those from November and December where likely underaged because water temperatures had already decreased below their growth threshold. This graph compbines previous knowledge, environmental monitoring and results of otolith microstructure analysis.

From the abstract:

“We examined the utility of otolith microstructure analysis in young-of-year (YoY) Atlantic silversides Menidia menidia, an important annual forage fish species along the North American Atlantic coast. We first compared the known hatch window of a local population (Long Island Sound, USA) to otolith-derived hatch distributions, finding that YoY collected in October were reliably aged whereas survivors from November and December were progressively under- aged, likely due to the onset of winter ring formation. In all collections, males outnumbered fe- males, and both sexes had bimodal size distributions. However, while small and large females were almost evenly represented (~60 and ~40%, respectively), over 94% of all males belonged to the small size group. We then examined increment widths as proxies for somatic growth, which suggested that bimodal size distributions resulted from 2 distinct slow- and fast-growing YoY phe- notypes. Length back-calculations of October YoY confirmed this, because fast- and slow-growing phenotypes arose within common bi-weekly hatch intervals. We concluded that the partial sexual size dimorphism in this population resulted largely from sex-specific growth differences and not primarily from earlier female than male hatch dates, as predicted by the well-studied phenome- non of temperature-dependent sex determination (TSD) in this species. Furthermore, observed sex ratios were considerably less male-biased than reconstructed thermal histories and published laboratory TSD values predicted. Assuming that selective mortality is generally biased against slower growing individuals, this process would predominantly remove male silversides from the population and explain the more balanced sex ratios at the end of the growing season.”

[New Publication] IUCN published ocean deoxygenation report!

9 December 2019. During the COP25 summit in Madrid, the International Union for Conservation of Nature (IUCN) released its latest comprehensive report titled “Ocean deoxygenation: everyone’s problem” that compiles the current evidence for the ongoing, man-made decline in the oceans oxygen levels. The 588 page, 11 chapter wake-up call to these detrimental changes was produced by leading experts in the field. We are happy to announce that Hannes is one of the many authors of this document, co-authoring chapter 6 “Multiple stressors – forces that combine to worsen deoxygenation and its effects“.

From the executive summary:

“The equilibrium state of the ocean-atmosphere system has been perturbed these last few decades with the ocean becoming a source of oxygen for the atmosphere even though its oxygen inventory is only ~0.6% of that of the atmosphere. Different analyses conclude that the global ocean oxygen content has decreased by 1-2% since the middle of the 20th century. Global warming is expected to have contributed to this decrease, directly because the solubility of oxygen in warmer waters decreases, and indirectly through changes in the physical and biogeochemical dynamics.”

From the summary of chapter 6:

  • Human activities have altered not only the oxygen content of the coastal and open ocean, but also a variety of other physical, chemical and biological conditions that can have negative effects on physiological and ecological processes. As a result, marine systems are under intense and increasing pressure from multiple stressors.
  • The combined effects of ‘stressors’ can be greater than, less than, or different from the sum of each stressor alone, and there are large uncertainties surrounding their combined effects.
  • Warming, acidification, disease, and fisheries mortality are important common stressors that can have negative effects in combination with low oxygen.
  • Warming, deoxygenation, and acidification commonly co-occur because they share common causes. Increasing carbon dioxide (CO2) emissions simultaneously warm, deoxygenate, and acidify marine systems, and nutrient pollution increases the severity of deoxygenation and acidification.
  • A better understanding of the effects of multiple stressors on ocean ecosystems should improve the development of effective strategies to reduce the problem of deoxygenation and aid in identifying adaptive strategies to protect species and processes threatened by oxygen decline.


Access the full report from

[Presentation] Callie presents research at the Graduate Climate Conference

Callie presenting her poster to other graduate students
November 8, 2019. Callie Concannon joined other graduate students of the Department of Marine Sciences to present her thesis research at the Graduate Climate Conference in Woods Hole, MA. She presented a poster entitled “Long-term CO2 and temperature effects on fecundity and oocyte recruitment in the Atlantic silverside
Her preliminary findings can be summarized as:

Warmer, more acidic environments impact reproductive output in the Atlantic silverside

The participants of the Graduate Climate Change conference in November 2019


[Lab news] Deanna Elliott completes her NSF-REU project

10 August 2019. Deanna Elliott from Arizona State University has just successfully completed her summer research project as our third NSF-REU student. For her REU-project she reared Atlantic silverside larvae under different feeding regimes to create fish of different body sizes and then analyzed them for trace levels of mercury in their tissue. She tested the hypothesis that mercury concentrations in fish can be used as a proxy for ingestion rates, which are important to trophic ecosystem models to perform better.

Here’s what Deanna had to say about her REU research experience:

This summer, I spent 10 weeks in the Baumann Evolutionary Fish Ecology lab and had a blast! The entire lab was incredibly welcoming, and made me feel at home immediately. We jumped right into my project and I had so many new experiences, it was almost overwhelming. We went seining for silversides in Mumford Cove, fertilized fish eggs… I became a Fish Mommy for the first time, rearing approximately 500 juvenile silversides for five weeks—I had never even had a fish tank before! I also got valuable experience in the chemistry lab, analyzing the mercury content of my Fish Babies. I felt very welcomed and received a lot of encouragement on my project and the presentation I had to give at the end of the program. Hannes and Zosia especially made me feel appreciated and supported, and that made all the difference in my experience with UCONN’s marine biology REU.

Check out some of the impressions from Deanna’s time at UConn. Great job, Deanna!


[New publication] Science publishes our silverside genetic study!

Fishing changes silverside genes
1 August 2019. We are overjoyed that our paper on genetic changes in experimental silverside populations subjected to strong size-selective fishing has just been published by Science!

Therkildsen, N.O., Wilder, A.P., Conover, D.O., Munch, S.B., Baumann, H., and Palumbi, S.R. (2019)
Contrasting genomic shifts underlie parallel phenotypic evolution in response to fishing
Science 365:487-490
Related perspective: Fishing for answers Science 365: 443-444 | Cornell Press release | UConn Press release

Over recent decades, many commercially harvested fish have grown slower and matured earlier, which can translate into lower yields. Scientists have long suspected that rapid evolutionary change in fish caused by intense harvest pressure is the culprit.

Now, for the first time, researchers have unraveled genome-wide changes that prompted by fisheries – changes that previously had been invisible, according to a study published in Science by a team of researchers including Hannes Baumann, UConn assistant professor of Marine Sciences, who collaborated with researchers at Cornell University, the University of Oregon, the National Marine Fisheries Service, and Stanford University.

In unprecedented detail, the study shows sweeping genetic changes and how quickly those changes occur in fish populations extensively harvested by humans, says Baumann.

“Most people think of evolution as a very slow process that unfolds over millennial time scales, but evolution can, in fact, happen very quickly,” said lead author Nina Overgaard Therkildsen, Cornell assistant professor of conservation genomics in the Department of Natural Resources.

Observed shifts in adult size. Trends across generations in mean length at harvest (standardized as the difference from the mean of the control populations in each generation) ± the standard deviations in up-selected (blue shades), down-selected (yellow and orange shades), and control populations (green shades).

The all-pervasive human meddling in our planet’s affairs undeniably reached the genetic make-up of its organisms.
— Hannes Baumann, UConn.

In heavily exploited fish stocks, fishing almost always targets the largest individuals. “Slower-growing fish will be smaller and escape the nets better, thereby having a higher chance of passing their genes on to the next generations. This way, fishing can cause rapid evolutionary change in growth rates and other traits,” said Therkildsen. “We see many indications of this effect in wild fish stocks, but no one has known what the underlying genetic changes were.”

Therkildsen and her colleagues took advantage of an influential experiment published back in 2002. Six populations of Atlantic silversides, a fish that grows no bigger than 6 inches in length, had been subjected to intense harvesting in the lab. In two populations, the largest individuals were removed; in another two populations, the smallest individuals were removed; and in the final two populations, the fishing was random with respect to size.

After only four generations, these different harvest regimes had led to evolution of an almost two-fold difference in adult size between the groups. Therkildsen and her team sequenced the full genome of almost 900 of these fish to examine the DNA-level changes responsible for these striking shifts.

The team identified hundreds of different genes across the genome that changed consistently between populations selected for fast and slow growth. They also observed large linked-blocks of genes that changed in concert, dramatically shifting the frequencies of hundreds of genes all at the same time.

Surprisingly, these large shifts only happened in some of the populations, according to the new paper. This means that there were multiple genomic solutions for the fish in this experiment to get either larger or smaller.

“Some of these changes are easier to reverse than others, so to predict the impacts of fisheries-induced evolution, it is not enough to track growth rates alone, we need to monitor changes at the genomic level,” said Therkildsen.

When the experiment was originally conducted nearly two decades ago by co-authors David Conover, professor of biology at the University of Oregon, and Stephan Munch of the National Marine Fisheries Service, the tools to study the genomic basis of the rapid fisheries-induced evolution they observed were not available. Fortunately, Conover and Munch had the foresight to store the samples in a freezer, making it possible to now return – armed with modern DNA sequencing tools – and reveal the underlying genomic shifts.

Research like this can assess human impacts, and improve humanity’s understanding of “the speed, consequences and reversibility of complex adaptations as we continue to sculpt the evolutionary trajectories of the species around us,” Therkildsen said.

“What’s most fascinating about this is that life can find different genetic ways to achieve the same result. In this study, two experimental populations evolved smaller body size in response to the selective removal of the largest fish, which is what most trawl fisheries do. However, only by looking at the genetic level we demonstrated that these two experimental populations evolved via two completely different genetic paths,” says Baumann.

The good news for the Atlantic silversides is that the fisheries selection was able to tap into the large reservoir of genetic variation that exists across the natural range of this species from Florida into Canada, said Therkildsen: “That genetic bank fueled rapid adaptation in the face of strong fishing pressure. Similar responses may occur in response to climate-induced shifts in other species with large genetic variability.”

“Scientists have coined the term Anthropocene in recognition of the all-pervasive human alteration of the earth’s climate, oceans, and land. No matter how ‘pristine’ a piece of nature may look to us at first glance, examine it thoroughly enough and you will find a trace of human in it. Take a cup of water from the middle of Pacific Ocean and a handful of sand from a ‘pristine’ beach – and you will find little plastic particles under the microscope,” says Baumann. “The parallel to this study is that the all-pervasive human meddling in our planet’s affairs now undeniably reached the genetic make-up of its organisms. Today’s fishes may superficially look the same as always, but their genes are not. They bear witness to human alteration.”

In addition to Baumann, Therkildsen, Conover, and Munch, co-authors included former Cornell postdoctoral researcher Aryn P. Wilder, now a researcher at San Diego Zoo Institute for Conservation Research; and Stephen R. Palumbi, Stanford University.

This work was funded by the National Science Foundation.

[Lab news] NSF-REU student Deanna Elliott joins the Baumann lab

Deanna Elliott is a junior at Arizona State University who has joined the Baumann lab in summer 2019 as our third NSF-REU student. Deanna has experimented with locusts before, but now strives to become an expert fish rearer. Her project will rear Atlantic silverside larvae under different feeding regimes to create fish of different body sizes and then analyze the these fish for trace levels of mercury in their tissue. She will test the hypothesis that mercury concentrations in fish can be used as a proxy for ingestion rates, which are important to improve trophic ecosystem models. Welcome, Deanna!

[Atlantic silverside, Menidia menidia, mercury, ingestion rates]

An early brainstorming sketch on the whiteboard, outlining Deanna’s REU experiment
Deanna starts her REU experiment by fertilizing strip-spawned silverside eggs

[Lab news] Emma turns 30 and starts a new silverside experiment!

3 May 2019. It is Emma’s 30th birthday today, so naturally she celebrates it by starting a new, large experiment with Atlantic silversides, thus sharing her special day with more than 5,000 little embryos that are now developing in our system.

Like in our previous experiments, we are mimicking current and future coastal environments that fluctuate daily in CO2 and oxygen levels – thanks to our computer-controlled system that manipulates these levels in up to nine tanks simultaneously.

But this time, our additional goal is to keep track of sib-ship. We produced full sibs (same mother, same father), half-sibs (same mother or father, different father or mother) and unrelated individuals, and by keeping them separate we will later be able to calculate additive genetic variances in the various traits under different conditions (i.e., heritability) and examine trait correlations.

Breeding design

As usual, this could not be done by one person, so the entire lab helped preparing, seining, and fertilizing embryos on this frantic day. Great job all!


[Publication] Meta-analysis of silverside CO2 experiments published!

28 November 2018. Hannes, Emma, and Chris are happy to announce that Biology Letters just published our latest study, a meta-analysis of 20 standard CO2 exposure experiments conducted on Atlantic silverside offspring between 2012-2017. All these years of sustained experimental work resulted in the most robustly constrained estimates of overall CO2 effect sizes for a marine organism to date.
The study demonstrated:

  • A general tolerance of Atlantic silverside early life stages to pCO2 levels of ~2,000 µatm
  • A significant overall CO2 induced reduction of embryo and overall survival by -9% and -13%, respectively
  • The seasonal change in early life CO2 sensitivity in this species
  • The value of serial experimentation to detect and robustly estimate CO2 effects in marine organisms

Baumann, H., Cross, E.L., and Murray, C.S. Robust quantification of fish early life CO2 sensitivities via serial experimentation. Biology Letters 14:20180408

This figure shows the summary of early life responses to high CO2 conditions in Atlantic silversides across all experiments conducted between 2012-2017. Effect size was estimated using the log-transformed response ratio (A-D). Error bars are 95% confidence intervals. The responses are considered significant if the confidence interval does not include zero. Panels E-F: seasonal decomposition of response ratios, showing that silverside early life stages are most sensitive to high CO2 at the beginning and end of their spawning season.

[New publication] Complex CO2 x temperature effects in Menidia offspring

20 July 2018. We are happy to announce that Diversity just published Chris Murray’s paper on complex CO2 x temperature effects in Atlantic silverside offspring. The paper synthesizes 5 large multistressor experiments conducted since 2014, finding evidence for the large CO2 tolerance in this species across a large temperature range.

Congrats, Chris, to the second chapter published!

  • Murray, C.S. and Baumann, H. You better repeat it: complex temperature × CO2 effects in Atlantic silverside offspring revealed by serial experimentation. Diversity 10:69

  • MurrayBaumann-Fig1
    M. menidia. Offspring responses to control (blue), high (red), and extreme (green) CO2 conditions at four temperatures across five CO2 × temperature factorial experiments. Traits include embryo survival (A–E), hatch length (F–I), larval survival (J–N) and larval growth rate (O–R). Individual replicates are represented by small faded circles. Treatment means (±SD) are depicted by large, bold circles and connected by dotted lines. Note: different scales used for hatch length measurements due to differences in sample timing; panels F and G use 1dph length Y axis (left) while panels H and I use hatch length Y axis (right).