BUSHTICK - Changing farming, bush encroachment, and tick-borne disease risk in southern Norway

Context and objectives

Marginal areas of Europe are dynamic, and some of the contemporary land use and land cover changes occurring in these areas may have implications for human health, in particular with regard to the risk of tick-borne diseases and the abundance of ticks, currently the most important disease vector in Europe. In many areas, such as southern Norway, the distribution and abundance of ticks, and the incidence of tick-borne diseases both in animals and human populations have been shown to increase. This situation might result from the interactions between humans and the environment through land use and land management. However, standard land cover datasets do not always permit to investigate in a detailed and meaningful way the complex web of interactions between land cover, land use, and tick and tick-borne diseases distribution.

In this context, we will answer the following questions:
What are the spatio-temporal dynamics and extent of bush encroachment in southern Norway? ;
What are the relations between bush encroachment and ticks and tick-borne disease risks?

For that, we will produce land cover and land cover change maps based on spatially and temporally detailed remotely sensed data. This will provide a better understanding of the phenomenon of bush encroachment in Norway and other marginal areas of Europe. Then, these maps will be combined with other potentially important factors to model the distribution of ticks.

BUSHTICK will also allow:
A comparison with models calibrated using standard datasets such as CORINE. Beyond the immediate evaluation, this may inform the use of standard datasets in contexts where the production of detailed, custom-made data is not possible;
The evaluation of the constraints and potential solutions for mapping land cover change phenomena that are highly dynamic temporally and spatially.

Project outcome

  • A detailed map of land cover and land cover changes based on a time series of Landsat data over the period 1984 to 2010 and focusing on tick-relevant features were produced. Trends and breakpoints in NDVI;
  • Tick presence was assessed using two sets of environmental factors: one using the habitual static land cover classes , and one dynamic, using continuous variables describing the fraction of forest cover around collection points. Preliminary results indicate an association between high incidence of Lyme borreliosis and agricultural land abandonment;
  • Our results open interesting new perspectives on how to represent and monitor tickinfested landscapes. A patch-based monitoring of landscapes relying on discrete land cover classes may be insufficient to get a holistic perspective on tick habitat; while the amount of deciduous forest, as mapped using discrete landscape classes, was a significant predictor of tick presence in south-western Norway, the best model and predictive performance is obtained by adding the standard deviation of the tree cover fraction per pixel within a buffer surrounding the collection point. This indicates that discrete representations of the landscape miss complexities of landscape composition and configuration that have an impact on tick habitat;
  • Environmental changes during the last decade may explain, at least in part, actual pattern of human cases of Lyme borealis: Changes associated with the vegetation were related to actual variation of rates of infection at municipality level.