Evaluating the Forest Pest Invasion Potential of Trade-related and Recreational Transportation Pathways

The successful establishment of invasive alien species in North America is strongly associated with global trade. Invasive insects and pathogens that affect plants are regularly introduced through international trade and domestic freight shipments, but few studies have directly quantified the invasion risks caused by commercial transportation at regional or larger scales. Similarly, despite much recent attention to the potential spread of forest pests with firewood transported by cars and other private or commercial vehicles, there has been little quantitative assessment of the invasion risks posed by this dispersal vector. In each case, a lack of empirical studies regarding human-mediated, longdistance spread of invasive organisms represents a key knowledge gap impeding realistic prediction of invasion patterns. We present two studies in which we separately examined aspects of forest pest movement via commercial-traderelated and recreational transportation. Our first study estimated the likelihood of successful entry of alien forest insect species for >3,000 urban areas in the contiguous U.S. To develop these estimates, we first utilized data on historical merchandise imports and insect incursions to estimate an annual U.S. rate of alien insect species establishment with trade. Next, we used historical data on pest interceptions to calculate the proportion of all insects arriving at U.S. ports of entry that are associated with forest hosts. We then combined these results to estimate a nationwide establishment rate for alien forest insects. Finally, we used international and domestic commodity flow networks to allocate this nationwide rate to individual U.S. urban areas. For 2010, we estimated the nationwide rate as 1.89 new alien forest insect species per year. While the establishment rates observed at most urban areas are low (<0.005 new species/year), selected areas can expect the establishment of new alien forest insect species every 5-15 years. Our second study analyzed forest insect spread with firewood and developed related dispersal functions for application in spatially explicit models. The primary data source was the U.S. National Recreation Reservation Service database, which records camper reservations at >2500 locations nationwide. We used “origin-destination” data for >7 million individual reservations (including visits from Canada) to construct an empirical dispersal kernel that estimates the likelihood of spreading an invasive organism as a function of geographic distance. We then fitted the data with various theoretical distributions. The data appear to be strongly leptokurtic and fairly well fit by the unbounded Johnson (SU) distribution (the lognormal distribution fit similarly). Most campers (~53%) traveled <100 km, but ~10% traveled >500 km (and as far as 5500 km). Additionally, we examined the impact of geographic region and proximity to major national parks and urban centers on the shape of the dispersal kernel, and found that mixture distributions may fit better in such circumstances. These assessments represent initial steps toward improved depiction of human-assisted dispersal potential, and provide important functional inputs for quantitative models of invasion. Future work will focus on applications of our results in comprehensive, cross-border (U.S.-Canada) analyses of pest invasion risk and for specific probabilistic modeling efforts to trace the origins of established or anticipated infestations.