Time series analysis has found that small rodent population cycles may be driven by an interaction between food shortage during winters and delayed density dependence caused by predators. The region where small rodents exhibit cyclic dynamics have long la sting thick snow cover that probably facilitates the cycles. The layer between the snow and the ground will under some circumstances be transferred to a layer of fragile pyramidal ice crystals making the subnivean space. Small rodents sustain life through out the winter in the subnivean space. The subnivean space is fragmented by ice frozen to the ground that effectively reduce carrying capacity and may promote overgrazing. Overgrazing might be most prevalent during snowme lt when the small rodents are for ced by water from the melting snow to higher ground. The aim of this study is to experimentally test how much small rodent survival is dependent on access to food patches under the snow and to develop and parameterize an explicit functional response model to test whether the small rodent population cycle ca n be driven by a plant-herbivore interaction. Access to food will be manipulated by creating a network of artificialtunnels under the snow permitting access to else un-accessible food patches.