Spread Forecasting Model

The unprecedented economic, social, and environmental challenges caused by invasive species amplify the value of identifying and implementing optimal management options to prevent, detect, and control invasive species (Simberloff et al. 2013). Decision-makers face two major challenges when managing the spread of invasive species. First, pest risk management decisions frequently involve trade-offs between complex and often competing economic, social, and environmental objectives. Second, understanding of these risks is often marked by profound uncertainties (Liu et al. 2010). Spread forecasting maps illustrate the probability of invasion by an alien species vary temporally and spatially. Such maps are powerful tools to assist policymakers in

Since the seminal work of Fisher (1937) and Skellam (1951) ongoing attention has been devoted to the development of species spread models in ecology as a means of either understanding how organisms spread, developing new modeling techniques, or predicting their spread rates (reviewed in Hastings et al. 2005; Higgins and Richardson 1996). This form of modeling has also identified the role of different spread pathways (Robinet et al. 2009) and valued the adoption of a strategic control zone to slow the spread of invasive species (Buckley et al. 2005; Sharov 2004; Sharov and Liebhold 1998). Most recent modeling approaches incorporate the dynamics of population growth, stratified dispersal (local dispersal and long dispersal), propagule pressure, Allee effect, as well as realistic habitat landscape heterogeneity (GIS framework), and temporal variability (e.g. Keesing et al. 2006, Anderson et al. 2009; Keith et al. 2008; Pitt et al. 2009 and Meetemyer et al 2011). Shogren (2000) and Cook et al. (2007) also address the issue of incorporating economics into risk reduction strategies for invasive species using a model of endogenous risk. Methods that integrate pest spread and climate suitability, with economic and social factors have been developed to assess the degree to which economics policies or control strategies can be used to optimize management decisions of

Spatially explicit spread models are considered to require too many poorly known parameters for their projections to be as reliable in practice (Hartig et al. 2011). In addition, the diversity of uses and application oof these models pointed at inherent limitations to the predictability of the phenomenon (Caley et al. 2008). New analytical methods are being developed to provide formal quantitative measurement of uncertainty (Makowski, 2013) and to address the perceived risk aversion of some biosecurity decision-makers (Yemshanov et al. 2013). Initial investigations suggest that the incorporation of uncertainty analysis adds credibility to pest risk maps, and narrows the set of geographical locations that would need to be targeted by costly inspections and public outreach activities (Kriticos et al., 2013). However, Venette et al. (2010) called for substantial improvement in visual decision-support model documentation, communication of uncertainty, data accessibility, and improved training. It is important not to portray a false sense of accuracy to decision-makers by concealing what may or may not be captured by a species spread map, or the model behind this map. There is a large body of literature on understanding the propagation of error from model inputs,