Masting – the synchronous, heavy production of seeds in a given year by individuals of a population, typically followed by several years with little or no seed production – is widely observed in tree species. There are many hypotheses to explain the evolution of this phenomenon, and many ecological consequences of the behavior. Over the past 20 years, we have examined the linkages between tree seed production and the abundance of generalist small mammals (primarily white-footed mice) in oak – northern hardwood forests at the Cary Institute in southeastern New York. Our primary interest in this relationship stems from its central role in the spatial and temporal dynamics of human risk of Lyme disease, the most common vector-borne human disease in the U.S.
Results/Conclusions
Our initial hypotheses about the dynamics of the mast-mouse-Lyme disease system were guided by the assumption that masting - particularly among the oak (Quercus) species – was synchronous both across sites and across species, and could be characterized by periodic resource “pulses” characterized by super-abundant acorn crops. We also initially assumed that mouse population density would most closely track the abundance of acorns of the red oak subgenus (Erythrobalanus), because they provided a critical over-winter food resource. A very different pattern, however, is emerging from our long-term data. Seed production by the oaks and other tree species with seeds large enough to constitute a significant fall food resource at our sites shows only weak synchrony across species or sites over time. Moreover, there is evidence that mouse densities the following summer respond to seed production of species other than the red oak group. The most striking pattern is the occurrence of periodic resource “crashes” in which there is little or no seed production by any of the tree species. There have been 2 of these events during the past 10 years at our study site. The seed crop failures are followed by crashes in the white-footed mouse population the following summer at all 6 of our study sites. During the subsequent years, mouse population densities among sites increase and diverge, as a function of spatial and temporal variation in the numbers of individuals and species of trees reproducing at a given site. Several of our metrics of human risk of Lyme disease show minimum thresholds for mouse population density, and densities below those thresholds have generally been associated with the resource crashes.
