The Big Catch: Quantitative center is making every fish count
From April until July, Michigan State University (MSU) AgBioResearch evolutionary ecologist Kim Scribner is more likely to be found wading with students in the evergreen-lined Black River of northern Michigan than in his campus office.
December 29, 2014
From April until July, Michigan State University (MSU) AgBioResearch evolutionary ecologist Kim Scribner is more likely to be found wading with students in the evergreen-lined Black River of northern Michigan than in his campus office. The fisheries and wildlife professor studies lake sturgeon, a bottom-feeding Great Lakes fish species that can grow longer than 8 feet, weigh up to 300 pounds and live for nearly a century. Despite long lives, however, lake sturgeon populations have been in serious decline since the late 19th century. The U.S. Fish and Wildlife Service reports that in 1910, 80 percent of the species had disappeared from Lake Erie, and by 1929 the amount of sturgeon caught in Lake Michigan had dwindled from 3.8 million pounds to just 2,000. Numerous causes for the sturgeon decline are known. Scribner and his research team are trying to find a way to bring back the ancient fish.
“Sturgeon migrate from the Great Lakes to the rivers for spawning, which makes them an important part of that coupled ecosystem,” he said. “The species is a poster child for humans decreasing the population through overfishing and pollution, and we’re trying to help turn that around.”
Though female lake sturgeons lay thousands of eggs at a time, Scribner’s team has found that 95 to 98 percent never survive to hatching, and of those that do, 99 percent die in the larval stage. Those survivors must live 15 years before achieving sexual maturity and the ability to reproduce.
“Our primary focus is finding the factors that limit sturgeon survivability in the early stages of their life cycles,” Scribner said. “If we can identify the sources of mortality, we can create better management plans to help increase their numbers.”
Scribner and his team, in cooperation with the Michigan Department of Natural Resources (DNR), are in the 11th year of pursuing these answers. Though they employ a wide variety of experiments — from studying the organisms that prey on the young sturgeon to the microbial content of the stream and lake water they call home — accurate population estimates over time are absolutely essential to understanding the impact of those threats.
In the spring, when the sturgeon return from the lakes to their ancestral spawning grounds upstream, Scribner’s team is in the river daily, using nets to collect the fish and tagging them with an array of tracking devices. Some tags are simply colored markers, enabling the researchers to identify a specific fish on sight. Others are more complex, such as passive inducible transponders (PIT), which are implanted in the fish to detect their exact locations.
The ability to track and monitor Black River sturgeon has taken Scribner’s research into a new level of experimentation and analysis. His team has documented the genotype of the nearly 1,200 spawning lake sturgeons, allowing the researchers to identify the parents of newborn sturgeon and make predictions about whcih genetics are the most beneficial to the species’ survival. The technology has also allowed the team to discover which stream environments sturgeon spawn in, and how various environmental and water quality conditions affect the hatch time and the likelihood that a larval sturgeon will live to adulthood.
“Temperature has a definite effect on their growth rate and survival,” Scribner said. “Eggs laid in colder temperatures incubate longer, and the larva that emerges has a larger body size.”
Keeping tabs on the lake sturgeon population is crucial to understanding these fish and the threats they face.
“No two years are the same,” Scribner said. “You need long-term research when you’re studying a fish that can live over 100 years.”
Scribner’s lab has taken their work further, partnering with the MSU Quantitative Fisheries Center (QFC) to understand lake sturgeon populations in Lake Michigan where these same issues apply. Together, Scribner, a QFC associate researcher, and other QFC scientists are using genetic marker data from fish sampled in the lake, where different populations mix, to untangle the dynamics of populations spawning in different streams.
Without the ability to keep accurate records of lake sturgeon population levels, Scribner said this type of research would not be possible, and the future of lake sturgeon would continue to remain troubled.
Expertise for a challenging task
“There’s a famous phrase in fisheries management: counting fish is just like counting trees, only they move and you can’t see them,” saidMichael Jones, MSU professor of fisheries and wildlife.
Jones is co-director of the organization formed specifically to help fisheries management agencies tackle that exact problem. In 2004, MSU joined forces with the MDNR and the Great Lakes Fishery Commission (GLFC) to establish the QFC to provide research and training support to fisheries management agencies around the Great Lakes basin. One of the main objectives was to take advantage of the quantitative and statistical expertise of MSU faculty members. The techniques and technologies pioneered by QFC are very important to researchers such as Scribner and Jones, but they are even more crucial to management agencies, whose scientific needs do not always mesh with traditional academic research programs.
“The agencies’ interest in starting our center was motivated by a recognition that, at the time, they lacked the expertise in quantitative fisheries that was becoming increasingly important for good fisheries management in the Great Lakes,” Jones said. “Quantitative fisheries science takes a high amount of expertise.”
Jones and his colleagues had been developing that expertise, as well as a relationship with the fisheries management agencies of the Great Lakes, for more than a decade through the Partnership for Ecosystem Research and Management (PERM). PERM is a collaborative effort between MSU and the MDNR established in 1993. Its focus has been on traditional research, however, the need for an entity that could assist with the short-term research and training needs of agencies became apparent. That’s when QFC originated.
“PERM is something that QFC sort of grew out of,” said James Bence, QFC co-director along with Jones. “When we were first talking about the idea that would become the QFC, the discussion centered around the issue that a lot of times agencies need help on an array of projects that are not really suitable to being regularly sponsored research projects on their own.”
Having accurate population estimates is a critical tool in fisheries management because the estimates enable fisheries managers to determine the amount of fish that can be harvested by commercial and recreational fishers and to track the overall health of the ecosystems under their authority. Obtaining them is a challenging, complex task, one that the experts working for the QFC are especially qualified and equipped to perform.
A numbers game
One way that QFC helps management agencies is through stock assessments, in which all the data an agency has for a particular fishery is integrated to evaluate the status of an entire fish population. This means combining the numbers of fish with information on age, types of environment and harvest rate. What sounds like simple addition is actually a very complicated mathematical process requiring an in- depth understanding of both fisheries and statistical methods. QFC works with agencies across the Great Lakes to conduct and improve stock assessments.
“We’ve played a very important role in providing education on stock assessment methods to our agency partners, and we’ve done important work on species assessments around the basin,” Bence said.
QFC scientists have helped management agencies assess populations of walleye in Lake Erie, lake trout in Lake Superior and Chinook salmon across three of the five Great Lakes, among others. By applying highly developed statistical models to the various species, researchers can also develop a description of the underlying characteristics that will affect growth. This provides a window into the future as well as the present. QFC scientists are able to turn this data into forecasting models, which allow their management agency partners to plan for the long-term sustainability of the fisheries in their charge.
“What we do with stock assessments helps the agencies set reasonable harvest rates,” Bence said. “It helps them determine how best to use their resources, whether they should stock certain species, alter the fraction of a population that can be harvested or even how often they should conduct new stock assessments. How we go about figuring out an issue is often just as important as how we solve it.”
Developing better stock assessment methods is one of they key contributions made by QFC.
“Agencies need to know how many fish there are. That’s a simple question with a very complicated answer,” Jones said. “One of the main thrusts of our work is coming up with clever ways to count fish.”
QFC researchers collaborated to assist he Ontario Ministry of Natural Resources measure changes in the biodiversity of Lake Huron’s fish species using data collected over the past 30 years. Using a technique called dynamic factor analysis, the researchers were able to take data from five locations around the lake, adjust it to ensure that it was comparable and draw conclusions that have helped shape Ontario’s fishery management policies.
“The bottom line was that we found a major decline in the abundance of fish across the community,” said Brian Maurer, director of the MSU Center for Statistical Training and Consulting (CSTAT). “We also found that the proportion of species remained relatively constant, which impleid that all species had experienced this decline.”
CSTAT is dedicated to improving the quality of MSU research by providing expertise in various levels of statistical analysis, a mission that makes it a natural partner to the QFC.
“We’re still working on this project, but it’s already a very interesting story that we need to learn more about if we’re going to be wise about how we impact our natural resources,” said Maurer, a professor in the Department of Fisheries and Wildlife and MSU AgBioResearch community ecologist. “Using quantitative fisheries methods has revealed just how huge of an impact we’ve had on the Great Lakes ecosystem. You have these huge lakes that you’d think we couldn’t possibly have an impact on, but biologically they’ve changed dramatically even in the past three decades.”
The expertise of QFC is important not only for fisheries management agencies but for academic fisheries researchers as well. Better, more refined statistical models are essential for furthering knowledge of fisheries ecosystems.
Daniel Hayes, a MSU AgBioResearch fisheries scientist and a professor in the MSU Department of Fisheries and Wildlife, studies dams and dam removals, as well as habitat conditions such as water temperature, plant cover and food resources, and their effects on fish populations. Though his work is focused on the field, it remains quantitative in nature.
“My approach is statistical — that’s the quantitative part,” said Hayes, an associate QFC researcher. “In science, sometimes you find things and nobody uses them, but we help discover how things work through statistical modeling that has real impacts in the world. It’s very applied.”
Hayes and his colleagues have increasingly found that variance among individual fish in a population is an important element in the success of that species, and they have developed a new statistical model to incorporate that. The individual-based model allows researchers to examine how the differences between individuals in a population affect the whole. For example, though most fish in a population spawn within the same time period, not all fish spawn at exactly the same time. Scribner found through his research that the time a lake sturgeon spawns can determine much about the survivability of its offspring.
“If the water is warm when they spawn, the young develop too fast and hatch in the winter; if it’s cold, they might develop slower and hatch in more favorable conditions,” Hayes said. “We use the individual model to understand this better.”
In Scribner’s lake sturgeon research, only a small percentage of the thousands of eggs live on to become reproducing fish. The individual model used by Hayes looks to answer why that is.
“If you look at young fish, the average fish dies,” Hayes said. “With this model, we’re trying to determine what makes that small percentage of survivors more fit. The average doesn’t always tell us what’s going to happen in the long run, so we need to look more closely.”
An eye for the future
In addition to pioneering the cutting-edge assessment methods employed in fisheries management and research, QFC is also working to train the next generation of quantitative fisheries scientists. Many graduates of the QFC’s Ph.D. program now work for the federal government, primarily the National Marine Fisheries Service, tasked with managing on a national level the fisheries of the United States.
“There is a recognized shortage of appropriately trained stock assessment scientists coming into the field,” said Travis Brenden, QFC assistant director. “There is going to be a significant wave of retirements, and a lot of students don’t have the training to fill that gap. Our center is one of the few major programs that provide that training.”
One of the driving forces behind management agency support for the center was the need to improve agencies’ capacity for performing activities such as stock assessments, and to that end the QFC provides training for both fisheries students and agency staff members. Through workshops and online training courses, QFC is helping to expand the capabilities of management agencies, allowing them to improve their use of data. The center has also begun new secondment and visiting scholars programs, in which agency staff and researchers would come work at the center for several months in order to gain a more in-depth knowledge of the methods employed by QFC scientists. The center is also considering expanding its training opportunities with a professional master’s degree program.
“The data they collect is expensive, so it makes sense to keep looking for better ways to use it,” Bence said. “Better training helps them make better management choices, so fish stocks don’t collapse and bigger harvests can be obtained with less risk. The training we provide has helped fisheries management become more rigorous than it was even five years ago.”
Fellow fisheries scientist Christopher Vandergroot sought out the center to improve his quantitative fisheries skills. He brought back the knowledge he gained at QFC to his work on Lake Erie through the Ohio Department of Natural Resources.
“As a fisheries biologist, the training I received has given me a new perspective to problem-solving and analysis in my current position,” Vandergroot said. “QFC’s commitment to quantitative and scientific rigor is exemplified in their desire to train not only the next generation of fisheries professionals, but those currently working in the field, as well.”
By bringing the right people together in the right environment, QFC has been able to improve the abilities of both scientists and agency employees to sustainably monitor and manage the fisheries of the Great Lakes.
“QFC plays a critical role in training people, in transferring knowledge and in linking students, scientists and professionals,” Hayes said.
Building a better fishery
QFC scientists and students continue to have a significant impact on the way fisheries are managed. Across the Great Lakes and the United States, they have changed and refined the ways we understand one of the region’s most significant natural resources.
“We’re helping to improve the way we manage our natural world,” Jones said. “I’ve always been somebody who’s more interested in that than in pushing back the frontiers of knowledge in a more fundamental sense. I’d rather solve messy practical problems than elegant abstract ones, and that’s exactly what we do here.”
Many partners of the center testify to its positive impacts.
“Ever since its inception, the Quantitative Fisheries Center has provided fisheries managers with the expertise needed to ensure the long-term stability of the Lake Erie fishery resource,” Vandergroot said. “Because of my experience with QFC, I look forward to interacting with them into the future.”
Equipped to handle a diverse range of challenges, from streams to lakes, food supply to fish mortality and walleye to lake sturgeon, QFC continues to stand as one of the preeminent organizations in its field.
“You never know what request will come in,” Brenden said. “It could be something as simple as developing a program for sorting data or it could be helping to evaluate a water quality management program or a stock assessment model. A lot of what we work on deals with complex species in very large systems, and that can be difficult. It can be challenging, but it’s also very rewarding.”