Multivariate evaluation of cold tolerance in domestic and foreign populations for addressing climate mismatch in biological control of Alternanthera philoxeroides in the USA

Lack of successful biological control of alligatorweed, Alternanthera philoxeroides (Mart.) Griseb. (Amaranthaceae), by Agasicles hygrophila Selman & Vogt (Coleoptera: Chrysomelidae) in the USA is an emerging problem as A. philoxeroides expands into temperate climates due to poor cold tolerance of the beetles. Sourcing climatically suited biotypes of biological control agents is traditionally done through foreign exploration into their native range; however, surveying introduced populations may provide a cost-effective and regulatory-friendly alternative. To compare foreign biotypes to those established in the USA, two populations of A. hygrophila were collected from the coolest extent of their range in South America, and two populations were collected from stable populations in the southeastern USA. For comparing cold tolerance among introduced populations, eight populations were collected from across a climatic gradient. For all populations, cold tolerance was measured as a function of supercooling point, critical thermal minimum (CTmin), chill coma, and temperature-dependent survival and development of eggs and larvae. Results of cold-tolerance measures for North American populations were subjected to principal component analysis (PCA) to reduce dimensionality and detect trends in cold tolerance among populations. For foreign and locally collected A. hygrophila, South American populations were significantly more cold tolerant as measured by CTmin (?1.98 ?), chill coma (?1.95 ?), larval survival (?35.0%), and development time of eggs (24.7% faster) and larvae (17.4% faster) at 15 ?. Comparisons among North American populations found significant differences in cold tolerance across populations; however, results were less consistent than those observed for foreign populations. The first principal component (PC) explained 43.0% of variation in cold-tolerance metrics and was strongly related to mean temperature of the coldest quarter. The use of a multivariate approach to evaluate and compare cold-tolerance phenotypes provides strong evidence of local adaptation in introduced populations of A. hygrophila. However, South American populations may still provide the best chance for control of A. philoxeroides in northern infestations.