Ecological Modelling 247 (2012) 262–272 Contents lists available at SciVerse ScienceDirect Ecological Modelling jo u r n al hom ep age : www.elsevier.com/locate/ecolmodel Modeling the spread of the Argentine ant into natural areas: Habitat suitability and spread from neighboring sites Katherine Fitzgerald a,∗ , Nicole Heller b , Deborah M. Gordon a a Department of Biology, Stanford University, 371 Serra Mall, Stanford, CA 94305-5020, United States b Climate Central, 895 Emerson Street, Palo Alto, CA 94301, United States a r t i c l e i n f o Article history: Received 18 February 2012 Received in revised form 26 July 2012 Accepted 30 July 2012 Available online 2 October 2012 Keywords: Invasion model Grid-based model Argentine ants Linepithema humile Prenolepis imparis a b s t r a c t To predict a fine-scale invasion of Argentine ants (Linepithema humile) into a natural area from the sur- rounding suburban matrix, we introduce a grid-based invasion model, similar to a cellular automaton model. Our model was based on observations of ant presence and absence but, unlike other models based on presence–absence data, it incorporated the process of invasion by spread from neighboring areas. Simulations were parameterized from a statistical analysis of a 17-year survey of ant distributions in the Jasper Ridge Biological Preserve in northern California. We simulated the effects of Argentine ant pres- ence at neighboring grid squares, distance to development, presence of the native winter ant Prenolepis imparis, and other habitat and climate variables, and used these models to simulate invasion over many decades. The best predictions of the extent of Argentine ant invasion were based on the distance of each site to developed areas. Adding the effect of neighbors improved the predictions of the time at which sites would be invaded. Winter ants responded mainly to vegetation cover. Our results suggest that Argentine ants may reach their potential distribution in insular urban reserves rapidly, perhaps within 10 years, and that reserve size determines whether the reserve is likely to become fully invaded. © 2012 Published by Elsevier B.V. 1. Introduction 1.1. Invasion models Biological invasions can damage both natural ecosystems and human economic activities. It is important for land managers to predict where and how quickly invasions will occur, and researchers have developed a wide variety of models to accomplish these goals. Many invasion models fall into one of two categories: habitat suitability and mechanistic, which differ in application and in the data required to parameterize them (Jeschke and Strayer, 2008). Habitat suitability models can be parameterized using sim- ple presence/absence data, but can predict only the outcome of an invasion, in eventual spatial extent, not the dynamic process leading to that outcome. By contrast, mechanistic models require detailed information to find values for the parameters, but can pre- dict the course of invasions through space and time (Carrasco et al., 2010; Kot et al., 1996), and can be used to investigate the con- sequences of management interventions (Miller and Tenhumberg, 2010; Shea et al., 2010). ∗ Corresponding author. Present address: United States Fish and Wildlife Service, Yreka Fish and Wildlife Office, 1829 South Oregon Street, Yreka, CA 96097, United States. Tel.: +1 415 515 3350. E-mail addresses: kfitzger@alumni.stanford.edu (K. Fitzgerald), heller.nicole@gmail.com (N. Heller), dmgordon@stanford.edu (D.M. Gordon). Habitat suitability models are used, when little information is available about an invasive species’ population growth and dis- persal, to identify regions similar to the invader’s known range, where the invader would probably become established if it were introduced there (Jeschke and Strayer, 2008; Loo et al., 2007; Peterson et al., 2004). However, such species distribution modeling is predicated on the assumption that a species is in equilibrium with its environment. This may lead to an underestimate the extent of eventual invasion, because early in the invasion, when the invader may not yet have been introduced to all types of suitable habi- tat, equilibrium may not be reached (Jones et al., 2010; Robinson et al., 2010; Welk, 2004). Moreover, an invasive species may not prefer the same habitat in all parts of its range, due either to differences in biotic interactions or to physiological differences between populations (Dullinger et al., 2009; Rödder and Lötters, 2010; Sutherst and Maywald, 2005). Some models of habitat suit- ability avoid some of these pitfalls by basing their predictions on detailed measurements of physiological reactions to climate (e.g., temperature-dependent mortality or reproduction rates as in Abril et al., 2009; Hartley et al., 2006; Hartley and Lester, 2003), but this type of model requires detailed knowledge about the invader’s physiology. Mechanistic models, such as integro-difference models and individual-based simulations, predict the course of an invasion (Carrasco et al., 2010; Kot et al., 1996). They can be also be useful in identifying the best management interventions, e.g., by identifying which life stages have the greatest effect on population growth or 0304-3800/$ – see front matter © 2012 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.ecolmodel.2012.07.036