Model-based extrapolation of ecological systems under future climate scenarios: The example of Ixodes ricinus ticks.

Models can be applied to extrapolate consequences of climate change for complex ecological systems in the future. The acknowledged systems' behaviour at present is projected into the future considering climate projection data. Such an approach can be used to addresses the future activity and density of the castor bean tick Ixodes ricinus, the most widespread tick species in Europe. It is an important vector of pathogens causing Lyme borreliosis and tick-borne encephalitis. The population dynamics depend on several biotic and abiotic factors. Such complexity makes it difficult to predict the future dynamics and density of I. ricinus and associated health risk for humans. The objective of this study is to force ecological models with high-resolution climate projection data to extrapolate I. ricinus tick density and activity patterns into the future. We used climate projection data of temperature, precipitation, and relative humidity for the period 1971-2099 from 15 different climate models. Tick activity was investigated using a climate-driven cohort-based population model. We simulated the seasonal population dynamics using climate data between 1971 and 2099 and observed weather data since 1949 at a specific location in southern Germany. We evaluated derived quantities of local tick ecology, e.g. the maximum questing activity of the nymphal stage. Furthermore, we predicted spatial density changes by extrapolating a German-wide tick density model. We compared the tick density of the reference period (1971-2000) with the counter-factual densities under the near-term scenario (2012-2041), mid-term scenario (2050-2079) and long-term scenario (2070-2099). We found that the nymphal questing peak would shift towards early seasons of the year. Also, we found high spatial heterogeneity across Germany, with predicted hotspots of up to 2,000 nymphs per 100 m2 and coldspots with constant density. As our results suggest extreme changes in tick behaviour and density, we discuss why caution is needed when extrapolating climate data-driven models into the distant future when data on future climate drive the model projection.