Invasive smallmouth bass (Micropterus dolomieu): history, impacts, and control

In this review, we (i) describe smallmouth bass (Micropterus dolomieu Lacepède, 1802) invasions past, present, and future; (ii) summarize the impact that this species can have on native communities; and (iii) describe and discuss various options for control. M. dolomieu are invasive throughout much of the United States, southern portions of Canada, and in countries in Europe, Asia, and Africa. Historically, this species spread via stocking programs intended to improve sport fisheries. Currently, their spread is facilitated by anglers and global climate change. Models predict that M. dolomieu will continue to spread with consequences for native prey fish, sport fish, and food webs through predation, competition, and hybridization. Effective control methods are necessary to mitigate these impacts. Options for M. dolomieu control include biological control, chemical control, environmental manipulation, and physical removal. However, our review of the literature suggests that only a handful of the possible control options have been explored (usually in isolation and with limited success), and that there is a clear need for focused research and informed management. For example, our elasticity analysis of published M. dolomieu matrix population models suggests that M. dolomieu control will be most effective when it targets eggs, larvae, and juveniles. We recommend targeting these life stages by using nest failure as part of an adaptive and integrated pest management approaches that incorporate existing and emerging technologies. However, we also emphasize that M. dolomieu control, where necessary and possible, is more likely to take the form of suppression rather than permanent eradication. Therefore, we also recommend efforts to prevent M. dolomieu (re)introduction.


Introduction
The smallmouth bass (Micropterus dolomieu Lacepède, 1802) is a cool-warm water centrarchid (Brown et al. 2009;Shuter et al. 1980Shuter et al. , 1989 and a popular sport fish among anglers. M. dolomieu are littoral predators, generally consuming small prey fish and crayfish (Vander Zanden et al. 1999). During the spring, male M. dolomieu build and guard nests in the shallows of lakes and streams (Ridgway et al. 1991). M. dolomieu are native to freshwater systems in 23 states in the eastcentral United States (Rahel 2000;Fuller and Cannister 2011) and the southern portions of two Canadian provinces (Scott and Crossman 1973).
Outside of its native range, M. dolomieu is an invasive species for which there is currently no effective means of control. M. dolomieu are invasive across much of the United States, southern portions of Canada, and in 9 other countries throughout the world (Fuller and Cannister 2011;Lyons 2011;Iguchi et al. 2004a). As with many invasive fishes, the spread of M. dolomieu beyond its native range has been facilitated by intentional and accidental stocking and climate-mediated habitat expansion (Jackson 2002;Rahel and Olden 2008). Invasive M. dolomieu reduce native smallbodied fish abundance and diversity through predation, outcompete other piscivorous game fish, and indirectly change planktonic and benthic communities (Jackson 2002). To date, most attempts to control invasive M. dolomieu have either produced undesirable results (e.g., Zipkin et al. 2008) or proven prohibitively labor-intensive (e.g., Tyus and Saunders 2000).
Here, we review the literature on invasive M. dolomieu and consider various options for control or eradication. Where M. dolomieu literature is lacking, we draw from the literature on invasive largemouth bass (Micropterus salmoides Lacepède, 1802), a sister species with a similar life history. First, we summarize the history of M. dolomieu invasions and review research that predicts M. dolomieu spread with climate change. We then summarize the impacts that invasive M. dolomieu have on native species and food webs. Finally, we outline various options for controlling or eradicating invasive species and describe their relevance to M. dolomieu in light of life history and previous control attempts.

Smallmouth bass invasions: Past, present, and future
Movement of M. dolomieu beyond their native range began primarily through intentional stocking by fisheries managers during the 19 th century. The prevailing attitude in fisheries and wildlife management during this period was that nature should be controlled and improved upon. Additionally, recreational fishing was just beginning to come into its own during this era. Books aimed at outdoorsmen promoted bass fishing (e.g. Henshall 1889 and1903) which was becoming a popular sport. Consequently, managers believed that introducing M. dolomieu would prove beneficial. The building enthusiasm for bass fishing led to a wealth of research on bass spawning and rearing (Bower 1897;Cushman 1917;Lydell 1902;Ripple 1908) allowing hatcheries to produce M. dolomieu for stocking throughout North America. For example, M. dolomieu were introduced to California in 1874 to improve sport fisheries (Moyle 1976). Similarly, from 1868 to 1881 the Maine Commissioners of Fisheries not only authorized M. dolomieu introductions in 51 water bodies, but also encouraged indiscriminate introduction by the public (Warner 2005). In the early 20 th century, managers introduced M. dolomieu into lakes in Ontario, Alberta and Manitoba, and even into a national park in Saskatchewan (Rawson 1945). Unfortunately, fisheries managers in the 19 th and early 20 th century knew little about both the importance of native fishes and the threats that non-native fishes such as M. dolomieu could pose to biodiversity. Indeed, non-native fishes were often introduced to control certain "undesirable" native species and enrich biodiversity (Hey 1926;Moyle 1976). Some jurisdictions even enacted laws to protect these non-natives (Cambray 2003;Hey 1926). Enthusiasm for non-native stocking was eventually tempered by changing attitudes and a better understanding of the impacts of non-native fishes. Stocking non-native fish to provide angling opportunities lost momentum after a final surge in the 1950s (Crossman 1991) and by the 1980s most authorized introductions were into systems that had already been invaded (e.g. Rahel 2004;Carey et al. 2011).
Intentional and unintentional introductions by anglers have been and continue to be major drivers behind the spread of M. dolomieu (Jackson 2002). Unintentional introductions of non-native fishes commonly result from bait bucket transfers (Litvak and Madrak 1993). The most comprehensive analysis to date (Drake 2011) suggests that bait bucket transfers in Ontario are responsible for as many as 20 introduction events of M. dolomieu per system per year into waterbodies outside of their current range. Intentional introductions occur because M. dolomieu are a popular sport fish. Bass are responsible for millions of angler fishing days per year in the Pacific Northwest (Carey et al. 2011), and 77.8% of competitive fishing events in North America's inland waters (Schramm et al. 1991). Both casual and competitive angling are important sources of revenue and development for many communities (Chen et al. 2003). Bass fishing has such a positive image that the negative effects of bass introduction usually go ignored. As early as the late 19 th century, citizens recognized that introduced M. dolomieu alter fish assemblages (Warner 2005), but the general public has remained apathetic towards the spread of M. dolomieu (Jackson 2002). Anglers continue to intentionally introduce M. dolomieu to create more fishing opportunities for bass, without knowing or acknowledging the potential impacts.
Recently, the spread and establishment of M. dolomieu has also been facilitated by global climate change. The establishment of M. dolomieu is dependent on temperature because their range is limited by the severity of overwintering stress in coldwater lakes (Shuter et al. 1980;Shuter et al. 1989;Jackson et al. 2001). Suitable habitat for M. dolomieu is expanding because of warming of lakes and streams attributed to global climate change. Climate change can also facilitate the spread of M. dolomieu to uninvaded systems through flooding associated with an increase in extreme weather events (Rahel and Olden 2008).   (Vander Zanden et al. 2004); with climate change this number could increase to 20% by the year 2100 (Sharma et al. 2009b). A conservative estimate is that at least 50% of Canada will become thermally suitable M. dolomieu habitat, including some arctic locations (Sharma et al. 2007). Combining these predicted changes in habitat suitability with predictions of where introductions will occur suggests that Manitoba and Ontario are the provinces at greatest risk for M. dolomieu invasion (Chu et al. 2005). Similar increases in thermal habitat are expected at northern latitudes throughout the world. For example, although cold temperatures prevented the establishment of M. dolomieu introduced to Sweden in the 1960s (Curry-Lindahl 1966, Kullander et al. 2012, they may be able to establish there under a warmer climate. As a result of introductions and habitat expansion, M. dolomieu are currently invasive throughout much of the United States, southern portions of Canada, Mexico, and 8 countries on three other continents ( Figure 1, Table 1). Forty-two states (including Hawaii) and most Canadian provinces bordering the United States have M. dolomieu in areas where they are considered non-native (Fuller and Cannister 2011;Lyons 2011). M. dolomieu have also established viable populations in Europe, Africa, and most notably in Japan where habitat suitability models suggest that most waterbodies in the country are at risk of M. dolomieu invasion (Iguchi et al. 2004b).

Impacts of invasive smallmouth bass
Invasive M. dolomieu can disrupt the native ecology of the systems to which they have been introduced. Bass are voracious predators that can decrease the abundance of, change the habitat used by, and even extirpate small prey fish such as brook stickleback (Culaea inconstans Kirtland, 1840), fathead minnow (Pimephales promelas Rafinesque, 1820), pearl dace (Margariscus margarita Cope, 1867), finescale dace (Phoxinus neogaeus Cope, 1867) and northern redbelly dace (Phoxinus eos Cope, 1861) (MacRae and Jackson 2001; Trumpickas et al. 2011). In streams, prey fish alter their behavior to avoid invasive M. dolomieu by moving from pools to riffles and areas with more structural complexity Invasive M. dolomieu can also impact top predators, many of which are also prized sport fish. These impacts occur primarily through competition for prey and predation on juveniles. Salmon and trout are particularly sensitive to M. dolomieu invasion (Sharma et al. 2009a). Stable isotope studies suggest that M. dolomieu predation alters food webs by forcing piscivorous lake trout (Salvelinus namaycush Walbaum in Artedi, 1792) to prey on zooplankton, a lowquality food source (Vander Zanden et al. 1999;Morbey et al. 2007). However, the presence of pelagic prey fish can buffer S. namaycush from the effects of competition with M. dolomieu by providing an alternative high -quality food source (Vander Zanden et al. 2004). In the event of a shift to sub-optimal prey, S. namaycush growth and reproduction could be limited. For example, in the 1960s M. dolomieu were introduced into Utah's Flaming Gorge Reservoir to control the native Utah chub (Gila atraria Girard, 1856),  Gomez and Wilkinson (2008) 3,4 conventional, uncontroversial, speciesand size-selective, applicable to all depths labor-intensive, ineffective, affects non-target species *Study contains proof of concept of the applicability of the control method to smallmouth bass, but does not attempt smallmouth bass control.
which was competing with salmonid sport fish (Teuscher and Luecke 1996). However, decades after introduction, competition with M. dolomieu for food appears to be inhibiting the growth of young S. namaycush in the reservoir (Yule and Luecke 1993). In Canada, climate change models predict that by 2100, 11% of S. namaycush populations will be negatively impacted by competition with M. dolomieu (Sharma et al. 2009b (Whitmore 1983;Whitmore and Hellier 1988). Hybridization with M. treculii is of particular concern because this species is endemic to the Edwards Plateau of south central Texas (Edwards 1980).
Fish are not the only taxa affected by invasive M. dolomieu; mammals, birds, amphibians, reptiles, and invertebrates can be impacted as well. M. dolomieu will consume almost any prey small enough to ingest including crayfish, rats, mice, young waterfowl, frogs, snakes, and salamanders (Sanderson et al. 2009). Frog species can be impacted by predation from M. dolomieu, although the severity could depend on the presence of other invasive species and the life stage of the frog (Kiesecker and Blaustein 1998). Any organism that depends on the prey of M. dolomieu can also be impacted. In an extreme case, competition with invasive M. dolomieu and M. salmoides for food contributed to the extinction of an endemic Guatemalan waterbird, the Atitlán grebe (Podilymbus gigas Griscom, 1929) (Hunter 1988). In ponds invaded by M. salmoides, the loss of prey fish and crayfish populations can lead to a reduction in top-down control of benthic invertebrates and macrophytes (Maezono and Miyashita 2003;Maezono et al. 2005).

Controlling invasive smallmouth bass
M. dolomieu control is essential for mitigating, minimizing and perhaps even eliminating the impacts of M. dolomieu on native species and food webs. However, there has been little documented work on M. dolomieu control to date, and many potential control options remain untested.
Here we describe and discuss various control options as they relate to M. dolomieu. This section draws from and builds upon a recent review by Halfyard (2010). We summarize control options in Table 2.

Removal
Removal refers to the physical capture and removal of fish from a system, typically via electrofishing, netting, explosives, or angling. These methods are labor-intensive and rarely result in successful control. However, in certain systems they may be an effective component of an integrated management plan.

Electrofishing
Electrofishing is a common, uncontroversial removal method in fisheries management that has been applied to invasive M. dolomieu with limited success. Electrofishing programs in small reaches of the Colorado and Yampa Rivers decreased M. dolomieu abundance, but only temporarily due to immigration (Burdick 2008; Hawkins et al. 2008). Other attempts to control invasive M. dolomieu via electrofishing have ultimately failed because of increased recruitment following treatment (Boucher 2005 andWeidel et al. 2007;Hawkins et al. 2008). In one striking example, the mass removal of 47,474 M. dolomieu over a 6-year period from an Adirondack lake in New York initially reduced M. dolomieu abundance by 90%, but ultimately resulted in increased abundance (Weidel et al. 2007;Zipkin et al. 2008). This unexpected increase was attributed to decreased intraspecific competition that led to accelerated maturation of juveniles and, ultimately, improved recruitment (Ridgway et al. 2002;Zipkin et al. 2008). This phenomenon is known as the hydra effect or overcompensation (Abrams 2009;Strevens and Bonsall 2011;Zipkin et al. 2008). Because electrofishing gear tends to remove more adults than juveniles (Moore et al. 1986;Kulp and Moore 2000;Earle and Lajeunesse 2007), this method can lead to overcompensation. Therefore, a control plan that involves electrofishing should include one or more methods that reduce the abundance of young M. dolomieu. Electrofishing is only likely to be effective in shallow, isolated streams and ponds absent of complex habitat. This method is labor-intensive, inefficient, non-species-specific, and requires repeated, longterm application.

Netting
Netting is another common fisheries management tool but is generally ineffective at controlling invasive M. dolomieu. For example, a springtime attempt to net invasive M. dolomieu in a 70 ha pond captured only 7 individuals in 2,103 trap net hours (0.003 fish/hour), as compared to 200 M. dolomieu captured in 8.62 hours of electrofishing effort (23.2 fish/hour; Boucher 2006). In another attempt, less than 1% of the 3083 fish captured in gill nets were M. dolomieu (Gomez and Wilkinson 2008). These results might reflect the relative abundance of M. dolomieu in the system, the low vulnerability of centrarchids to passive netting (Hayes et al. 1996), or seasonal variation in catchability (i.e., M. dolomieu are most trappable in mid-summer, Wright 2000). In general, nets are much less effective at catching M. dolomieu than electrofishing (Bacula et al. 2011 and references therein). Therefore, although netting is a familiar, available, and uncontroversial control option that can be both size-and species-selective and deployed at most depths, the inability of nets to catch large numbers of M. dolomieu, even when effort is high, is a significant shortcoming that precludes the use of nets for control, even in combination with other methods. However, netting may be the only option in systems that are not conducive to electrofishing. For example, invasive young-of-the-year M. dolomieu in Lake Opeongo, a low-conductivity lake in Algonquin Park, Canada, are routinely (and effectively) sampled using minnow traps (e.g. Dunlop et al. 2005aDunlop et al. , 2005b.

Explosives
Explosives have been proposed for invasive fish control (Lee 2001). Although there are no known applications to M. dolomieu control, the use of detonation cord to sample M. dolomieu in deep reaches of the Middle Snake River, Idaho (Munther 1970) suggests that at least some degree of removal is possible. Detonation cord tends to kill most fish within 9 meters of the blast (Metzger and Shafland 1986). In general, explosives are effective at killing adult and larval fish with swim bladders, but not fish eggs (Baxter II et al. 1982;Metzger and Shafland 1986;Bayley and Austen 1988;Keevin et al. 2002;Settle et al. 2002;Faulkner et al. 2008).
Explosives are an effective and relatively cheap method for killing fish in almost any habitat, but there are a number of issues that are likely to limit their use in M. dolomieu control. Explosives are difficult to obtain, highly controversial, and dangerous to both human and environmental health. They are not selective for fish species, and can destroy habitat and leave behind toxic chemical residues (Hayes et al. 1996;Lotufo and Lydy 2005). Explosives may also be inappropriate for large waterbodies because of scale. Several authors recommend explosives only when there are no other options for sampling or control (e.g., Bayley and Austen 1988;Hayes et al. 1996).

Angling
Removal by angling is another control option that is unlikely to be effective. In the only documented attempt, anglers removed M. dolomieu from a 70 ha pond in Maine at a rate of 0.31 fish/hour as compared to 23.2 fish/hour for electrofishing (Boucher 2006). The author did not believe that angling was an appropriate control measure for that system. In another program, nearly 300 angler hours over two years resulted in the removal of just 150 M. dolomieu (Gomez and Wilkinson 2008). Even though angling for M. dolomieu control has not been successful, increased fishing pressure either from liberalized regulations or intensive effort has been advocated for the control of other invasive fishes (Wydoski and Wiley 1999;Beamesderfer et al. 1996;Moore et al. 2005). Because angling is an inefficient removal method, it is probably most effective in small systems and/or when fishing pressure is high.

Options for enhancing removal
The efficacy of a particular removal method can be enhanced through techniques that improve catchability. Here we describe two such techniques: pheromone-baited traps and "Judas fish". Pheromone-baited traps use species-specific chemical attractants to improve trap efficiency (Sorensen and Stacey 2004 (Taylor and Katahira 1988). A Judas fish is a conspecific that has been implanted with a radio tag and then tracked to an aggregation of fish. These fish can then be more effectively targeted for removal. The Judas fish technique should be evaluated for its potential to locate aggregations of M. dolomieu (e.g., during spawning or winter shoaling), particularly in large systems.

Chemical
The addition of chemical piscicides to a system is perhaps the most common method of fish control (Wydoski and Wiley 1999;Meronek et al. 1996). Piscicides such as rotenone and antimycin-A are effective at killing a large proportion of fish in the system with minimal effort (Lennon et al. 1971, Baker et al. 2008. Both rotenone and a newly-developed piscicide, Supaverm®, can effectively kill invasive M. dolomieu (Smith 1941;Ward 2005). Rotenone eliminated M. dolomieu from Potter's Lake, New Brunswick (46 ha), and Supaverm® holds promise as a more selective option that tends to spare native minnows. Overall, and given adequate funding, chemical control can be a quick and effective option for M. dolomieu control in many systems. However, most piscicides are lethal to all fish in the system (Finlayson 2001;Dawson and Kolar 2003) and can also affect amphibians, aquatic invertebrates, and zooplankton (Smith 1940;Brown and Ball 1943;Morrison 1979;Finlayson et al. 2000;Arnekleiv et al. 2001;Ling 2001;Dinger and Marks 2007). Additionally, stakeholders may be opposed to this option on ethical grounds or due to its non-target effects and high cost. Managers should also be aware that piscicides are not always 100% effective at extirpating invasive fishes. Chemical eradication is, on average, only 35% effective 10 years after treatment (Wydoski and Wiley 1999). For chemical treatment to be effective, managers must acknowledge and manage for public opposition, non-target effects, and the potential need for additional control measures.

Biological control
Biological control (biocontrol) refers to the introduction or enhancement of an invasive species' predators or pathogens, or the sterilization of the invasive species. Although these methods tend to be controversial and are largely untested for M. dolomieu, there is evidence to suggest that they can be effective. To this end, we encourage research into the efficacy of biocontrol methods as they apply to invasive M. dolomieu.

Predation
Predation is the most common and effective method of biological control, through either the introduction of new predators or the enhancement of existing ones (Wydoski and Wiley 1999). This type of control would be most effective for small non-game fish or young sport fish (e.g., M. dolomieu eggs, fry, and juveniles) given their vulnerability to predation. The impact of this kind of biocontrol on M. dolomieu may be limited by the tendency of nesting males to aggressively guard eggs and larvae from predators during the spawning season (Ridgway et al. 1991). However, nest guarding may not be a limiting factor in all cases. For example, native Japanese dace (Tribolodon hakonensis Günther, 1877) are extremely effective nest predators, consuming on average 92.4% of invasive M. dolomieu eggs, even while males guard their nests (Iguchi and Yodo 2004). In this case and others, removing the guarding male would enhance nest predation. Nest predators such as crayfish, yellow perch (Perca flavescens Mitchill, 1814), sunfish (Lepomis spp. Rafinesque, 1819), and the introduced round goby (Neogobius melanostomus Pallas, 1814) can consume an entire nest of unprotected eggs in as little as 17 minutes (Kieffer et al. 1995;Steinhart et al. 2004).
Although invasive species are unlikely to be appropriate predators to introduce for biocontrol, management plans that enhance native predators or introduce otherwise benign predators could be useful. Once an appropriate predator is identified, this type of control is usually inexpensive, requires minimal effort, and can be effective in a variety of habitats. Predatory biocontrol is nonetheless risky and controversial (Hoddle 2004). Before introducing a predator species (or enhancing natural predators), it is important to consider the vulnerability of the target species, the potential for non-target effects, and the likelihood that introduced predators will survive, establish, and spread (Wydoski and Wiley 1999). Managers should carefully weigh the pros and cons of this untested method and researchers should consider its study, as it appears to be a promising option for M. dolomieu control.

Pathogens and parasites
The introduction or enhancement of novel or existing pathogens or parasites is another biocontrol option for invasive fish. We find one promising example in Australia, where research is underway to use koi herpes virus to control common carp (McColl et al. 2007). Although we are unaware of attempts to use pathogens or parasites to control invasive M. dolomieu, there are at least two M. dolomieu-specific parasites that hold promise. The first parasite is a protozoan that attaches itself to M. dolomieu gills and can cause mortality (Davis 1937(Davis , 1942. The second parasite is a tapeworm (Proteocephalus ambloplitis Leidy) that limits M. dolomieu fecundity by preferentially infesting its oocytes (McCormick and Stokes 1982). With proper testing and development, these parasites could be used as biocontrol agents for invasive M. dolomieu. Similarly, the pathogen known as the largemouth bass virus (Family Iridoviridae; genus unknown) causes a disease that is specific to M. salmoides (Grizzle et al. 2003) but could potentially be genetically engineered to target M. dolomieu. Testing and developing M. dolomieu-specific pathogens and parasites is likely to be both expensive and timeconsuming. Once developed, however, these agents could be used simply and cheaply in almost any system. Nonetheless, development and application should be undertaken with caution to avoid effects on non-target species, the development of resistance, and public controversy (Wydoski and Wiley 1999).

Sterilization
Sterilization is a form of biocontrol that involves the release of sterile conspecifics or the alteration of individual physiology to limit reproductive success. Sterilization is not currently available as a control technique for invasive M. dolomieu, but the idea merits further research. One promising option is to use pheromones to effectively sterilize M. dolomieu by altering their behavior. Treatment with pheromones can reduce a male bass's ability to guard his nest, causing increased nest failure (Dey et al. 2010). Another option is the release of sterile males, which has been used successfully for sea lamprey (Twohey et al. 2003). Recombinant gene therapy could be used to develop sterile M. dolomieu, but this technology is largely untested (Thresher 2008). Other sterilization techniques such as irradiation and chemically-induced sterilization should be researched for M. dolomieu. Sterilization is species-specific and applicable to all types of habitats, but it can be expensive to develop, controversial, and labor-intensive to implement.

Environmental manipulation
Environmental manipulation seeks to control invasive fishes via changes to their physical environment. Here we focus on manipulation of water levels and dissolved oxygen concentration.

Water level manipulation
Draining, or dewatering, a system is considered to be the most effective way to guarantee complete removal of fish (Finlayson et al. 2002;Ling 2001;McClay 2000). However, this approach affects non-target species and organisms and is impractical in many places due to logistics or public opposition. Partial dewatering may be a more feasible option, especially in managed reservoirs, rivers, and streams in which it is easy to manipulate water levels. Partial dewatering can cause behavioral changes in M. dolomieu and increase predation on juveniles (Rogers and Bergersen 1995;Heman et al. 1969). Because M. dolomieu spawn in shallow water (Ridgway et al. 1991), partial dewatering can also be used to limit the availability of spawning habitat or kill incubating eggs. The only example of water level manipulation specifically targeted at controlling invasive M. dolomieu induced discharge pulses in a stream during the spawning season (Kleinschmidt 2008). These pulses evacuated all fry from 43% of the study nests and removed some of the fry from another 21%. While the author believes that this reduction in M. dolomieu young could improve trout habitat, the long-term impacts of this treatment have not been determined as of 2013. Water level management is probably an inexpensive and effective option for M. dolomieu control in certain systems. However, this method cannot be applied everywhere, has the potential for non-target effects, and could be controversial given conflicting water uses (agriculture, industry, residential, recreational, hydroelectric power, navigation, etc.).

Dissolved oxygen concentration
It may be possible to control M. dolomieu by reducing the wintertime concentration of dissolved oxygen (DO) below the lethal threshold of ~1.2 mg/L (Smale and Rabeni 1995). One can induce lethal concentrations of DO in winter (producing a winterkill event) by partially drawing down water or causing disturbances (e.g., artificial mixing of the water column) that stir up anoxic water and sediments containing oxygen-consuming bacteria (Verrill and Berry 1995;Shroyer 2007). Winterkill has been used successfully for other invasive fishes (Verrill and Berry 1995;Shroyer 2007), but not for M. dolomieu. The applicability of this option is likely limited to shallow, eutrophic lakes and ponds, where it could be effective and inexpensive. However, inducing low DO is likely to kill non-target organisms and could be controversial. Elasticity is a measure of the sensitivity of population growth rate to systematic and proportional changes to age-specific parameters for survival and fertility. Our results suggest that population growth rate is most sensitive to survival in the first 1-4 years of life ( Figure 2). Therefore, from a biological perspective, targeting M. dolomieu eggs, larvae, and juveniles is the most efficient approach to controlling invasive M. dolomieu. Additionally, targeting these early life stages alone or in concert with management of adult M. dolomieu could potentially prevent overcompensation.

Targeting specific life stages
Options for targeting M. dolomieu eggs and larvae outnumber options for targeting M. dolomieu juveniles, and would be far more effective due to the relatively narrow habitat requirements for nesting and incubation. Whereas juveniles can be removed via electrofishing, angling, and perhaps minnow traps, options for targeting eggs and larvae include the physical removal of eggs from nests, the chemical treatment of nests, nest predator management, sterilization, explosives (larvae only), water level management during nest guarding, and the removal of nest-guarding males. Although angling is itself an inefficient removal method (Boucher 2006;Gomez and Wilkinson 2008), angling may be an effective means of inducing nest failure. Generally, the aggression of nest-guarding males makes them easy to angle during spawning season (Ridgway et al. 1991). Both field and modeling studies involving M. dolomieu suggest that even short handling times during catch-and-release angling can increase the risk of nest predation (Ridgway and Shuter 1997;Suski et al. 2003;Steinhart et al. 2005), and that released bass can be too exhausted to adequately defend their nests (Kieffer et al. 1995;Hinch and Collins 1991;Hanson et al. 2008).

Conclusion
Human introductions and global climate change are facilitating the spread of M. dolomieu beyond their native range in Canada and the United States, and into other countries around the world. Invasive M. dolomieu can impact native food webs through predation, competition, and hybridization, and can even extirpate native fishes. These impacts are well documented and fairly well understood, but our ability to control invasive M. dolomieu is severely limited. Although numerous options for M. dolomieu control exist, few have been tested or developed and even fewer have been successful.
To improve M. dolomieu control, we recommend integrated pest management plans that include several nest failure strategies, perhaps in combination with other options (e.g., adult removal). For example, using catch-and-keep angling to remove nest-guarding males while simultaneously enhancing native nest predators is probably more effective at inducing nest failure than either of these options alone. Additionally, because a small subset of spawning males (~5%) can produce over half of a population's young of year (Gross and Kapuscinski 1997), we recommend using genetic techniques to identify and subsequently target the most productive males. We also recommend research to develop control methods that are not yet available for M. dolomieu (e.g., pathogens, parasites, pheromone traps). Of course, the combination of control options to use in an invaded system also depends on environmental constraints (e.g., lentic vs. lotic, depth, structural complexity, substrate type, ecology), logistic constraints (e.g., budget, timeline, available infrastructure/equipment), and social factors (e.g., acceptability, political climate). M. dolomieu control may be unnecessary or impossible in some systems. For those systems in which control is an option, it is important to learn from previous attempts (M. dolomieu or other species) and practice adaptive management (Pine et al. 2009;Zipkin et al. 2009).
It is also important to maintain realistic expectations: M. dolomieu extirpation is unlikely except for in small and/or isolated systems, and M. dolomieu suppression will require multiple applications, probably in perpetuity (even rotenone can be <100% effective; Wydoski and Wiley 1999). For these reasons, in addition to sound research and management, we stress the importance of preventing M. dolomieu (re)introductions, for example through bait regulations and public awareness campaigns. Creative, comprehensive solutions need to be developed and implemented to minimize further spread of invasive M. dolomieu and mitigate their impacts on native ecosystems. Bacula TD, Blackwell BG, Willis DW (2011)