All quiet on the western front? Using phenological inference to detect the presence of a latent gypsy moth invasion in Northern Minnesota

Following establishment in a new environment, invasive species expand their range through stratified diffusion, the coupling of local growth and spatial spread with long-distance movement of propagules that found new colonies. Consequently, the stages of arrival and establishment serially repeat until the entire habitat susceptible to invasion is occupied. An empirical example of such a phenomenon is the gypsy moth, Lymantria dispar (L.), invasion of northeastern Minnesota. Under a regional management program, a trapping grid is deployed along the L. dispar leading edge to detect male moth presence and inform management decisions. However, the presence of moths does not always indicate reproducing populations, and moths dispersing from populations behind the front can obscure the presence of latent invasions. We analyzed trapping data during the arrival and establishment of L. dispar in this region (2000-2012), supplemented with sentinel trap data to ascertain the timing of male moth flight (2005-2009), and derived yearly phenological predictions for male moth flight. We also used a male wing morphology metric (2007-2009) to further quantify spatial and temporal patterns associated with the L. dispar invasion. We provide evidence of an established L. dispar population when analyses suggest spatial randomness in trap catch data, and how the presence of phenologically distinct, immigrating adult moths from source populations outside of Minnesota recorded from sentinel traps could lead to the overestimation of the abundance of an established population. This work highlights the complexity of the initial invasion process even in a well-studied system for which a sensitive monitoring program exists.