Impacts of Aquatic Invasive Species on Sport Fish and Recreational Fishing in the Great Lakes: Possible Future Scenarios March 2016 HDRU Series No 16-1 Prepared by: Richard C. Ready, T. Bruce Lauber, Gregory L. Poe, Lars G. Rudstam, Richard C. Stedman, and Nancy A. Connelly Human Dimensions Research Unit, Department of Natural Resources, Cornell University HUMAN DIMENSIONS RESEARCH UNIT PUBLICATION SERIES This publication is one of a series of reports resulting from investigations dealing with public issues in environmental and natural resources management. The Human Dimensions Research Unit (HDRU) in the Department of Natural Resources at Cornell University studies the social and economic aspects of natural resources and the environment and the application of social and economic insights in management planning and policy. A list of HDRU publications may be obtained by writing to the Human Dimensions Research Unit, Department of Natural Resources, Fernow Hall, Cornell University, Ithaca, NY 14853, or by accessing our World Wide Web site at: http://www.dnr.cornell.edu/hdru. TO CITE THIS REPORT Ready, R.C., T.B. Lauber, G.L. Poe, L.G. Rudstam, R.C. Stedman, and N.A. Connelly. 2016. Impacts of aquatic invasive species on sport fish and recreational fishing in the Great Lakes: Possible future scenarios. HDRU Publ. No. 16-1. Dept. of Nat. Resour., Coll. Agric. and Life Sci., Cornell Univ., Ithaca, N.Y. XX pp. This report is available electronically at http://www2.dnr.cornell.edu/hdru/pubs/HDRUReport16- 1.pdf. TABLE OF CONTENTS Table of Contents ............................................................................................................................. i Executive Summary ....................................................................................................................... iii Acknowledgments......................................................................................................................... vii Introduction ..................................................................................................................................... 1 Methods ....................................................................................................................................... 3 Scenario Development ............................................................................................................. 3 Economic Model...................................................................................................................... 6 Scenario Evaluation ................................................................................................................. 6 Future Effects of Aquatic Invasive Species on Sport Fish ............................................................. 8 Bighead and Silver Carp ............................................................................................................. 8 Scenario AC-1 ......................................................................................................................... 9 Scenario AC-2 ....................................................................................................................... 10 Northern Snakehead .................................................................................................................. 11 Grass Carp ................................................................................................................................. 13 Hydrilla...................................................................................................................................... 14 Quagga Mussel .......................................................................................................................... 16 Economic Model ........................................................................................................................... 17 Future Effects of Aquatic Invasive Species on Sport Fishing ...................................................... 41 Asian Carp ................................................................................................................................. 41 Scenarios AC-1a and AC-2a.................................................................................................. 41 Scenario AC-1b ..................................................................................................................... 46 Scenario AC-1c...................................................................................................................... 49 Scenario AC-2b ..................................................................................................................... 52 Scenario AC-2c...................................................................................................................... 55 Asian Carp Summary............................................................................................................. 58 Northern Snakehead .................................................................................................................. 59 Scenario NS-2 ........................................................................................................................ 59 Grass Carp ................................................................................................................................. 63 Scenarios GC-1 ...................................................................................................................... 63 Scenarios GC-2 ...................................................................................................................... 67 Scenarios GC-3 ...................................................................................................................... 70 Grass Carp Summary ............................................................................................................. 73 Hydrilla...................................................................................................................................... 74 i Scenarios H-2 ........................................................................................................................ 74 Scenarios H-3 ........................................................................................................................ 78 Hydrilla Summary ................................................................................................................. 81 Quagga Mussel .......................................................................................................................... 82 Conclusions ............................................................................................................................... 86 Literature Cited ......................................................................................................................... 89 ii EXECUTIVE SUMMARY The rate of invasions of non-native aquatic invasive species (AIS) has accelerated in the Great Lakes in recent years. To date, several hundred non-native species have entered the lakes. Some AIS have been shown or are predicted to cause considerable harm, and the effects of AIS on recreational fisheries has been a point of particular concern. Recreational angling in the Great Lakes is popular and highly valued. Poe et al. (2012) estimated that Great Lakes anglers enjoy 18 million fishing days per year with a net economic value of $0.4 to 1.3 billion annually. Management decisions aimed at limiting AIS introduction or spread should be informed by estimates of the potential consequences of those AIS on this highly valued recreational fishery. Ready et al. (2012) developed a model to address this need. Based on a survey of over 3,500 recreational anglers in 12 states in the Great Lakes, Upper Mississippi and Ohio River basins, they developed a model of recreational angler behavior that explains: (a) how often anglers go fishing, (b) where they go fishing, and (c) what types of fish (coldwater or warmwater) they target. Using these data, the model estimates the net economic value that anglers receive from the sportfishing resource. The model can therefore be used to project the impact of AIS-induced changes in sportfish catch rates on recreational fishing effort and on the net economic value that anglers place on the fishery. The model estimated by Ready et al. is the most comprehensive model available for evaluating and predicting recreational angling in the Great Lakes region. However, the model has some limitations. First, the model was estimated using data on trips from a very large study region that extends far beyond the boundaries of the Great Lakes basin. A model estimated using data on angler behavior from only the Great Lakes region would better reflect angler behavior in that region. Second, the model is relatively simplistic in how it models angler behavior, particularly with regards to how anglers choose what type of fishing to do, and how that choice would change in response to changes in fishing quality. A refined model is needed that better captures how anglers make these choices. Projecting of the impacts of AIS on recreational angling and anglers requires first estimating the ecological effects of AIS on recreational fish populations.Anticipating the impacts of AIS on recreational fish populations in a large, complex system, such as the Great Lakes, is challenging. Even under the best of circumstances, there is often considerable uncertainty about the likely impacts of AIS. Trying to reach single-estimate ecological forecasts of how AIS will affect fish may not be wise. Scenario-building may be a more appropriate approach to assess the possible future effects of AIS in the Great Lakes. This approach recognizes that uncertainty is fundamental in planning for the future. Scenario development involves preparing multiple internally consistent descriptions that represent a range of plausible futures and outcomes. While any individual scenario may have a very low probability, as a set, scenarios can set boundaries around a range of possible futures. Our project, therefore, had three objectives: • Refine the model estimated by Ready et al. (2012) to better reflect and better model the behavior of recreational anglers in the Great Lakes states. iii • Develop a set of plausible, science-based, internally consistent ecological scenarios about the possible effects of AIS on recreationally important fish populations in the Great Lakes. • Use the refined recreational angling model to project the impact that these AIS scenarios would have on angler behavior and on the net economic value of recreational fishing in the Great Lakes. Methods Using angler survey data collected by Ready et al. (2012), a new model of recreational angling behavior was estimated. Compared to the model estimated by Ready et al., the new model 1) is based on a smaller geographic area, to better reflect angler behavior in the Great Lakes region, 2) better models the decision of what type of fishing to engage in and how that decision is affected by angler characteristics, and 3) accounts for response patterns in the survey data that may have biased the previous model to overstate the degree to which anglers will change their behavior in response to changes in fishing quality. To generate AIS scenarios, we recruited 10 aquatic ecologists and fisheries managers from the Great Lakes region to participate in a scenario-building process. This process took place in three stages: (a) an initial Delphi survey to identify AIS of concern for the Great Lakes; (b) a two-day workshop in which scenarios describing the possible effects of 5 different AIS on recreational fish stocks and catch rates were developed; and (c) an iterative process—following the workshop--of reviewing and refining these scenarios and assessing their likelihood. Each scenario included quantitative estimates of the possible effects of AIS on different types of recreational fish populations and recreational fishing. Although we did not expect any single one of these individual estimates to represent a likely outcome, collectively the estimates for the set of scenarios for each AIS portrayed a range of plausible outcomes from the perspective of the expert panel. Using the refined angler behavior model, we projected the impacts that each scenario would have on angler behavior and the economic value of the Great Lakes recreational fishery. Results Workshop participants developed scenarios for five species considered among the most likely to affect recreational fish stocks: bighead and silver carp, northern snakehead, grass carp, hydrilla, and quagga mussel. Fifteen ecological scenarios were developed representing a range of possible ecological outcomes if one of these five AIS became established or spread in the Great Lakes. The different scenarios for each AIS were based on different assumptions about the ecological processes that would be most important. • Bighead and silver carp. The best case ecological scenario considered included: (a) a 5% decrease in salmonids throughout the Great Lakes; and (b) 10-25% increases in largemouth and smallmouth bass, yellow perch, and walleye in high productivity areas (Green Bay, Saginaw Bay, Bay of Quinte, Lake St. Clair, and the Western and Central Basins of Lake Erie). The worst case scenarios included: (a) a 10% decrease in iv largemouth and smallmouth bass and a 40% decrease in yellow perch and walleye in high productivity areas; or (b) an 80% decrease in coho and chinook salmon in Lakes Michigan and Ontario. • Northern snakehead. The best case ecological scenario considered was no net effect on the populations of fish species that are important to recreational anglers. The worst case scenario included: (a) a 15% decrease in walleye throughout the Great Lakes and a 15% decrease in largemouth and smallmouth bass and yellow perch in high productivity areas; and (b) a 5% decrease in salmonids throughout the Great Lakes. • Grass carp. The best case ecological scenario considered included projections of a 5% decrease in yellow perch and a 15% decrease in centrarchids (members of the freshwater black bass family, including bluegills and other sunfish) besides largemouth and smallmouth bass throughout the Great Lakes. The worst case considered was a 10% decrease in yellow perch and a 50% decrease in largemouth bass, northern pike, and other centrarchids besides smallmouth bass. • Hydrilla. The best case scenario was a 15% increase in yellow perch, largemouth bass, northern pike, and muskellunge throughout the Great Lakes. The worst case scenario included decreases of yellow perch, largemouth bass, northern pike, and muskellunge of 15% throughout the Great Lakes, 25% in Green Bay, Saginaw Bay, and the Bay of Quinte, and 30% in Lake St. Clair. • Quagga Mussel. The best case scenario was no further effect on the recreational fishery. The worst case scenario included an 80% decrease in coho and chinook salmon in Lake Michigan. Based on these scenarios, we projected that the five AIS considered could have a range of possible effects on recreational fishing participation and value. The worst case scenarios for Asian carp and quagga mussel could involve losses of $129,000,000 to $139,000,000 in angler consumer surplus and 375,000 to 400,000 fewer fishing trips annually in the Great Lakes states. But aquatic invasive species could also lead to improvements in recreational fishing. Certain scenarios for hydrilla and Asian carp led to projected gains of up to $30,000,000 in value and 85,000 fishing trips annually. Scenarios projecting improvements were much less common than those involving losses, however. The pattern of states affected would vary depending on the particular scenario, but generally those in the central Great Lakes region were expected to bear the greatest impacts. Illinois and Michigan had the potential to be most negatively affected. Wisconsin, Indiana, Ohio, and New York also bore substantial negative effects under some scenarios while Pennsylvania and Minnesota tending to be less affected. In those scenarios involving improvements to recreational fishing, Michigan and Ohio would be most likely to experience the greatest benefits. It is important to recognize that the impacts on recreational fishing participation are often less severe than the ecological effects of AIS with which they are associated. There are several reasons for this. To begin with, anglers target some species much more heavily than others. If an AIS affects species that receive less attention from anglers in the Great Lakes (e.g., centrarchids), the effects on recreational fishing participation and value will not be as substantial. In addition, some types of fishing are much less affected by the opportunity to catch fish than others. Those anglers who fish anadromous runs in particular are less likely to reduce their fishing as fish v populations decrease. Finally, many anglers switch from one type of fishing to another as the quality of their preferred type of fishing declines. In these scenarios, if one type of Great Lakes fishing declined, many anglers would take more trips for other types of Great Lakes fishing and, even more importantly, would take more trips to inland waters. Our research is not able to generate precise estimates of the future effects of AIS on recreational fishing participation and value. Indeed, our approach was premised on the assumption that precise estimates are impossible given the uncertainty associated with large ecological systems. Nevertheless, our work considerably narrows the range of possible AIS impacts that must be considered. Accepting the best and worst case scenarios from the set would involve projections from a $30 million improvement in the Great Lakes recreational fishery to a $139 million loss. Although that range is quite wide, it provides reasonable endpoints that policy makers can consider when evaluating options to control the AIS considered in this report and perhaps AIS more generally. vi ACKNOWLEDGMENTS This study was funded by the Great Lakes Fishery Commission. We thank David Bunnell, Ed Rutherford, Robert Hecky, Stephanie Guildford, Tom Stewart, Marion Wittmann, Stuart Ludsin, Marten Koops, Tomas Höök, and Andy Todd for their participation in the surveys and workshop that were used to generate assessments of the possible effects of aquatic invasive species on sport fish populations. Without their help and cooperation this project would not be possible. vii INTRODUCTION The rate of invasions of non-native aquatic invasive species (AIS) has accelerated in the Great Lakes in recent years (Mills et al. 1993, 2003). To date, several hundred non-native species have entered the lakes, many through ballast waters (Holeck et al. 2004). In the near future, we may expect to see several new species in the Great Lakes from the area around the Black and Caspian Seas (Kolar and Lodge 2001) as well as several species of Asian carp (Kolar et al 2007). Some AIS have been shown or are predicted to cause considerable economic costs (Pimentel 2001). For example, zebra mussels (Dreissena polymorpha) and quagga mussels (Dreissena rostriformis bugensis) are reducing water flows in pipes and increasing costs to private and municipal plants that use water from the lakes (Naleepa and Schloesser 2014). They also increase water clarity thereby promoting Cladophora growth in the nearshore (Hecky et al. 2004, Higgins and Vander Zanden 2010) and other aquatic macrophytes in embayments and lakes (Zhu et al. 2006, Mayer et al. 2014), which may or may not be seen as a positive outcome. However, mussels can also increase property values as a result of the increased water clarity associated with mussel filtering (Limburg et al. 2010). Other species may be replacing a native species without causing large scale ecosystem perturbations (such as the amphipod Echinogammarus echinus, a replacement that is not likely to be noticed by anglers or affect most people’s enjoyment of the lakes (Limen et al. 2005). Other species may improve the quality of angling and recreational activities – the round goby (Neogobius melanostromus) consumes zebra mussels and provide a forage species for warmwater fish species such as black bass and some coldwater species such as brown trout (Salmo trutta). While much research has been conducted on the ecological implications of AIS in the Great Lakes, less research has addressed the implications for recreational angling. Although exotic species certainly alter the original food web structure of the Great Lakes, their impact on recreational fishing and on anglers can be varied. For example, many cold-water anglers in the Great Lakes are seeking chinook salmon (Oncorhynchus tshawytscha) that rely on alewife (Alosa pseudoharengus) and rainbow smelt (Osmerus mordax) as the main prey species. None of these three species are native to the system. It is important to understand the impact of AIS on recreational anglers and to be able to quantitatively predict how angler behavior will change in response to those impacts. This is important for two reasons. First, recreational angling in the Great Lakes is popular and highly valued. Poe et al. (2012) estimated that Great Lakes anglers enjoy 18 million fishing days per year with a net economic value of $0.4 to 1.3 billion annually (measured in 2012 dollars).1 Second, management decisions aimed at limiting AIS introduction or spread should be informed by estimates of the potential consequences of those AIS on the highly-valued recreational fishery. 1 Net economic value is defined as the benefit that recreational anglers receive from being able to use the fishery in its current condition. This is the appropriate measure of benefit for use in cost-benefit analysis of management actions. This benefit measure is conceptually different from measures of regional economic impact due to expenditures by anglers. The contribution of the Great Lakes fishery (commercial and recreational) to the regional economy has been estimated as high as $8 billion (American Sportfishing Association 2013). 1