A simple range expansion model of multiple pathways: the case of nonindigenous green crab Carcinus aestuarii in Japanese waters

Forecasting the range expansion of nonindigenous organisms enables effective quarantine and the development of pre-arrival countermeasures, as well as raises public and scientific concerns among the general public. Here, we present an approach to forecasting the range expansion of the nonindigenous green crab Carcinus in Japanese waters, with consideration of the human-mediated shipping and natural dispersal. Two types of shipping, namely, primary transport to Japan via long-distance ocean-going shipping and secondary transport within Japan via short-distance coastal shipping, were considered. The presence and absence of the crab in 50-km shoreline segments were obtained at 5 year intervals from 1985 to 2005, based on observation records by professional and amateur naturalists. Two types of dispersal kernel, namely, conventional logistic regression and a multiplicative immigration model considering multiple pathways explicitly, were compared. The multiplicative immigration model gave significantly better results. Natural dispersal was the most significant factor, in spite of its low expansion rate: secondary transport by coastal shipping was the second. Transport by international ocean-going shipping was not statistically significant, suggesting quite a small immigration probability of the crab by long voyages. Stochastic simulations forecasted that the crab will invade most of the coasts of western Japan along the Pacific Ocean and Seto Inland Sea by 2055, and will become widespread all over the country by 2205. Quarantine scenarios to eliminate transport by vessels revealed that preventing crab transport by domestic coastal vessels may delay the arrival of crabs in Hokkaido and the northern Japan area by 700 years at maximum.