Reconstructing past population processes with general equilibrium models: House mice in Kern County, California, 1926–1927

Reconstructing past ecosystem conditions is helpful in addressing a range of basic and applied questions, but difficult because of a lack of inferential or modeling approaches. Here, we demonstrate the use of a general equilibrium ecosystem model (GEEM) to investigate ecological conditions surrounding a population eruption of the house mouse (Mus musculus) in Kern County, California in the 1920s, arguably the most dramatic mammalian population eruption reported for North America. Our application of GEEM to this ecosystem allowed us to evaluate the plausibility of reported population densities of house mice that were extreme, even in comparison with values for the same species introduced in Australia, and to examine alternative hypothetical community interactions and their effects on populations. The modeled ecosystem included vegetation, native small rodents and introduced house mice as herbivores, and mammalian predators. Individual plants and animals behaved as net energy optimizers, with net energy directed to reproduction. Optimizing predators determined the biomass taken from prey, and prey, through feeding choices, determined their vulnerability to predators. Model inputs incorporated the competitive superiority of house mice to native small rodents, the time of arrival of house mice in the area, and the timing and intensity of predator removal. Without an invasion of house mice, human removal of predators was predicted to minimally affect native small rodents. Without human removal of predators, invading house mouse populations were predicted to attain high, oscillating densities by the 1920s that stabilized at lower densities by the 1930s, with native small rodents persisting in the area. With near-complete human removal of predators and with house mice present, native small rodents were predicted to go nearly extinct, and house mice to attain densities similar to those reported. The predicted lag between the introduction and eruption of house mice approximated reported values. Predation on native and introduced rodents was predicted to allow both rodent groups to coexist, whereas near-complete removal of predators caused native small rodents to approach extinction during the house mouse eruption, consistent with empirical studies. GEEM appears to be a useful tool for reconstructing past ecosystem conditions and trophic interactions.