Predictive modelling is increasingly common in ecology, and statistical models created in one context are often used to predict the state of the system in other contexts. However, the fast development of predictive ecology calls for caution,as it is not always clear whether the current understanding of ecological processes is comprehensive enough to warrant predictions. The amplitude of rodent population cycles (i.e., peak-phase abundances) has been hypothesized to be determined by vegetation properties in tundra ecosystems. We assessed the spatial and temporal predictability of food and shelter plants effects on peak-phase small rodent abundance during two consecutive rodent population peaks.
During the rodent population peak in 2007 at Varanger peninsula, rodent abundance was related to the abundance of food and shelter plants. Still, plants alone were not enough to predict how high the rodent peak in 2011 was. The role of plants was then likely to be over driven by other food web components(like predators) and weather. This study highlights two challenges for predicting food web functioning to the future: i) gathering long enough time-series to include different situations in the predictive models and ii) including several key interactions instead of isolating one of them from the remaining food web.
You can access the full publication here.
Reference: Soininen, E.M., Henden, J.A., Ravolainen, V.T., Yoccoz, N.G., Brathen, K.A., Killengreen, S.T., Ims, R.A. (2018) Transferability of biotic interactions: temporal consistency of arctic plant-rodent relationships is poor. Ecology and Evolution 8(19):9697-9711.
Picture: Lemming grazing in a snowbed in Finnmark (photo: Eeva Soininen)