In order to evaluate costs associated with invasive species and cope with them, recognizing spatiotemporal distribution of the species and predicting area at high risk of invasion in the future are one of the fundamental principles of action plan against invasive species. In case of red-necked longicorn (Aromia bungii), which have already been spread in 21 cities in Osaka Prefecture 2023, the prediction could start from the relatively simple assumption. Namely, host species of the longicorn is mostly limited to the arboreous Rosaceae and their spread should be restricted to the host species, particularly cherry trees, which are popular as ornament tree. We therefore presumed that spatial distribution of host trees helps predicting area at high risk of invasion and future distribution of the longicorn.
For the elucidation of the spatial distribution of host trees, we conceived to use remote-sensing images, based on the fact substantial proportion of ornament cherry trees were a clonal breed (Prunus x yedoensis) and their blooming period are synchronized in a region, which indicates they are distinguished by the color tone of the bloom.
We selected several areas in Osaka Prefecture and manually recognized cherry trees based on the Google Earth satellite images to test whether the distribution of cherry trees elucidated by remote-sensing image help predicting range expansion of the longicorn. We used distribution of the longicorn infested trees (trees with signs of the frass excreted by the red-necked longicorn) since 2020 (hearing by Osaka Prefecture) and analyzed with the cherry tree distribution from the following perspectives. Firstly, at the edge of the longicorn distribution, the correlation between the canopy area of cherry trees and probabilities to find the longicorn infested trees were analyzed, assuming the accumulation of their host trees should encourage colonization and range expansion of the longicorn. Secondly, relationship between the annual increase of the infested trees and the canopy area of cherry trees were analyzed to test if area with abundant host trees provoked population growth of the longicorn in the spread phase. Our preliminary analysis based on a survey area at the edge of the longicorn invasion, the presence of the longicorn infested tree were positively correlated with canopy area of cherry tree surrounding the tree, which supports the notion that elucidating spatial distribution of host trees for the invasive species using satellite imagery data improve the area at high risk of invasion.