FORUM Combining Tactics to Exploit Allee Effects for Eradication of Alien Insect Populations 1 2 3 DAVID MAXWELL SUCKLING, PATRICK C. TOBIN, DEBORAH G. MCCULLOUGH, 4 AND DANIEL A. HERMS J. Econ. Entomol. 105(1): 1Ð13 (2012); DOI: http://dx.doi.org/10.1603/EC11293 ABSTRACT Invasive species increasingly threaten ecosystems, food production, and human welfare worldwide. Hundreds of eradication programs have targeted a wide range of nonnative insect species to mitigate the economic and ecological impacts of biological invasions. Many such programs used multiple tactics to achieve this goal, but interactions between tactics have received little formal consideration, speciÞcally as they interact with Allee dynamics. If a population can be driven below an Allee threshold, extinction becomes more probable because of factors such as the failure to Þnd mates, satiate natural enemies, or successfully exploit food resources, as well as demographic and environmental stochasticity. A key implication of an Allee threshold is that the population can be eradicated without the need and expense of killing the last individuals. Some combinations of control tactics could interact with Allee dynamics to increase the probability of successful eradication. Combinations of tactics can be considered to have synergistic (greater efÞciency in achieving ex- tinction from the combination), additive (no improvement over single tactics alone), or antagonistic (reduced efÞciency from the combination) effects on Allee dynamics. We highlight examples of combinations of tactics likely to act synergistically, additively, or antagonistically on pest populations. By exploiting the interacting effects of multiple tactics on Allee dynamics, the success and cost- effectiveness of eradication programs can be enhanced. KEY WORDS Allee effect, biological invasion, density dependence, eradication, invasive species There has been a steady accumulation of an increas- (Wiedemann) (Diptera: Tephritidae), Cochliomyia ingly wide range of plant-feeding insects in forests, hominivorax Coquerel (Diptera: Calliphorida), Adel- agro-ecosystems, and urban environments postborder ges tsugae Annand (Hemiptera: Adelgidae), and Ly- and beyond their native range (Levine and DÕAntonio mantria dispar (L.) (Lepidoptera: Lymantriidae) cost 2003, Brockerhoff et al. 2006, Hulme et al. 2008, property owners, local and national government agen- Aukema et al. 2010). Most species that arrive in a new cies, and private industries billions of dollars annually habitat fail to establish (Williamson and Fitter 1996, (Aukema et al. 2010, 2011; Holmes et al. 2010; Kovacs Ludsin and Wolfe 2001, Simberloff and Gibbons 2004, et al. 2011). The distribution of costs from incursions Lockwood et al. 2005) or have relatively minor effects is often contentious. in their expanded range (Mack et al. 2000, Aukema et Preventing the introduction of species has long al. 2010). A portion of alien species, however, become been recognized as the most effective means to reduce invasive with substantial economic and ecological im- impacts of invaders (Sakai et al. 2001, Hulme et al. pacts, often increasing the energy footprint of food 2008, Liebhold and Tobin 2008). Numerous interna- and Þber production systems because of an increased tional phytosanitary regulations and agreements, be- need for pest management, or irreversibly altering the ginning in the United States with the 1912 U.S. Plant invaded ecosystem and its biodiversity (Pimentel Pest Act, have been implemented to reduce risks of 2002, Gandhi and Herms 2010, Aukema et al. 2011). inadvertent transport of insects and other organisms High proÞle invaders such as Agrilus planipennis Fair- through the movement of infested materials (e.g., U.S. maire (Coleoptera: Buprestidae), Ceratitis capitata Code of Federal Regulations, Title 7, Chapter III, Part 301). Nevertheless, nonnative insects continue to be 1 Corresponding author: The New Zealand Institute for Plant and introduced and newly established species are de- Food Research Ltd., PB 4704, Christchurch, New Zealand (e-mail: tected postborder every year (Work et al. 2005, Brock- [email protected]). 2 Forest Service, U.S. Department of Agriculture, Northern Re- erhoff et al. 2006, Liebhold et al. 2006, McCullough et search Station, Morgantown, WV 26505. al. 2006). Given current and projected rates of global 3 Departments of Entomology and Forestry, Michigan State Uni- trade and travel, it is inevitable that unwanted, non- versity, 243 Natural Science Building, East Lansing, MI 48824. native insects will continue to be introduced, and 4 Department of Entomology, Ohio Agricultural Research and Development Center, The Ohio State University, 1680 Madison Ave., some will establish and ultimately become invasive Wooster, OH 44691. pests. 0022-0493/12/0001Ð0013$04.00/0 ᭧ 2012 Entomological Society of America 2JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 105, no. 1 Options available for eradication or, in the event of failure or lack of feasibility, long-term management of invasive insects, depend on biological attributes, hosts and projected impacts of the pest. When an alien species is detected but expected to have little impact, regulatory agencies typically elect to take no action. In New Zealand, for example, an average of one new organism is discovered postborder every week (Kriti- cos et al. 2005), but eradication programs are rare and mounted only when there is a high probability of expected economic and environmental cost. Deci- sions to initiate an eradication program are not un- dertaken lightly, given the signiÞcant expense and potential controversy that often accompany such ef- forts (Kean et al. 2012), particularly so for multi-year projects (Knipling 1979). Moreover, if a nonnative species appears to be established across a large geo- graphic area, eradication is unlikely to be practical (Brockerhoff et al. 2010). Similarly, erroneously de- claring success is embarrassing and costly, and under- mines future conÞdence. Although there are hundreds of cases of successful Fig. 1. Allee dynamics resulting from combining treat- eradication programs (Kean et al. 2012), there remains ment tactics (dashed arrows). (A) Representation of Allee considerable pessimism about the feasibility of erad- dynamics in which the change in population density (Ntϩ1/ N ) is plotted against the initial density (N ). Initial densities ication (Dahlsten and Garcia 1989, Myers et al. 2000). t t above an Allee threshold (solid square) will lead to a positive Some failed eradication attempts can be attributed to rate of population increase until governed by overcrowding biological, tactical, resource or political limitations dynamics (e.g., a carrying capacity). Initial densities below (Myers et al. 1998, Government Accountability OfÞce an Allee threshold will lead to a declining population density 2006). The outcomes of some efforts remain ambigu- and extinction. (B) Synergistically combining a density-in- ous for reasons ranging from political aversion to an dependent tactic to reduce population density with one that admission of defeat, the difÞculty determining when a increases an Allee threshold. (C) Combining two tactics that population is truly eradicated, as well as the poten- do not affect density but jointly increase the Allee threshold. tially embarrassing and costly consequences of erro- (D) Combining two tactics that do not alter the Allee thresh- neously declaring success (Dreistadt 1983, Dreistadt old but jointly decrease population density below an Allee threshold. (E) Antagonistic combination of tactics in which and Weber 1989). To circumvent some of these chal- one decreases density while another negates an Allee effect. lenges, previous studies have described probabilistic models used to estimate the conÞdence that an erad- ication program was successful given a continual lack can be integrated into the design and implementation of detection in monitoring efforts (Barclay and Har- of operational eradication programs (Tobin et al. grove 2005, Kean and Suckling 2005). Ultimately, the 2011). A key concept is that of pest density, which damage or potential damage associated with a non- involves knowledge of the number of individuals per native species must warrant the investment required unit area. Pest control tactics can be broadly classiÞed to detect and eradicate the population, and viable as density-independent, such as insecticide applica- methods to do so must be available. Given the long- tions where a certain proportion of the population is term costs of damage and pest management averted by killed, or density-dependent, such as mating disrup- eradication (Popham and Hall 1958, Klassen 1989, tion, where efÞcacy is inversely dependent on the Brockerhoff et al. 2010), eradication should not be absolute density and scarcity plays a role. Tactics discounted as an option, especially if novel approaches could also be used to subdivide or fragment popula- can facilitate success. Advances in understanding and tions, which can then be progressively tackled using technology can generate new tactics or strategies that the rolling carpet principle (Dyck et al. 2005). can be used in programs to eradicate insect pests. For In many populations, there is a critical population example, identiÞcation and synthesis of long-range size or density, known as the Allee threshold, below pheromones and other attractants have provided which the per capita population growth rate is nega- highly effective detection tools, and facilitated