MultiTick: Multiscale and multisensory modelling of the spatial distribution of tick-borne diseases
Description
The spatial distribution of vector-borne diseases is tied to environmental conditions in two primary ways. First, vectors and the pathogens they carry will only thrive under certain habitat conditions, relating to micro-habitat availability and also to climatic factors. While factors relating to microhabitats are best documented at the local scale, climatic ranges are often best described at the regional scale. Second, given a certain vector and pathogen distribution, for the disease to be transmitted to humans, there must be an overlap with the spatial distribution of human activities.
Tick-borne diseases are currently the most important vector-borne diseases in Europe. A dramatic upsurge in the incidence of tick-borne encephalitis was observed during the 1990s. Lyme borreliosis currently continues to increase in many European countries. Many explanations have been proposed to explain this, including climate change. However, climate alone is unlikely to account for the patterns observed.
Most applications of remote sensing to the study of vector-borne diseases so far have focused on one scale of analysis, using one type of remotely sensed data. Both low- and high-resolution data have been successful at mapping either vectors or disease cases. However, the knowledge that is currently available on environmental factors dictating the distribution of vectors or disease cases indicates that factors acting at more than one scale should be considered.
Using the example of tick-borne diseases in Europe, we propose to fill the gap existing in the use of remotely sensed data in spatial epidemiology consisting in the lack of combination of scales. In order to achieve this, we plan to use statistical techniques that allow the combination of nested, hierarchical data. Multilevel statistical methods, also called hierarchical models, have developed significantly over the past years, and have proved successful in a variety of fields that include sociology, epidemiology, and clinical studies. Their use in environmental studies, although promising, has remained sparse. These statistical methods can accommodate data nested in space or in time, and are therefore particularly suited for this question. Applying them to a remote sensing multi-scale study of disease risk is a particularly innovative component of the project.
MultiTick is funded by the Belgian Science Policy, Research Programme for Earth Observation Stereo II, Contract n°: SR/10/123
Contacts