Réseau national de surveillance aérobiologique (RNSA), Chemin des Gardes, BP 8, 69610 Saint-Genis-l’Argentière, Association pour le contrôle et la préservation de l’air dans la région grenobloise (ASCOPARG), 44, avenue Marcelin Berthelot, 38100 Grenoble, SUP’AIR, 22, rue Nicolas Avit, BP 345, 38150 Salaise-sur-Sanne
- Key words: cluster analysis, environnemental monitoring, pollen, rhinitis, allergic, seasonal
- DOI : 10.1684/ers.2008.0143
- Page(s) : 103-11
- Published in: 2008
The French network for monitoring airborne pollen (RNSA) is currently composed of 70 Hirst pollen traps distributed throughout France. The aim of this article is to present a simple statistical methodology that can be used to characterize the spatial variation of pollen. The results are discussed primarily in terms of improving network management. This pilot study is restricted to a limited portion of the RNSA network (18 monitoring stations), eight taxons of allergenic relevance, and a three-year period (2003-2005). In a first step, we quantified the similarity in pollen measurements between pairs of monitors according to an original index, called the mean Pollinic Distance (mPD), based on the comparison of concentration time series. Unexpectedly, there was a significant positive linear relation between the separation distance and the mPD. To characterize the spatial variation further, a cluster analysis was performed with mPD as a distance ’similarity) measure. When plotted on a map, the groups of site identified according to the similarity of their pollen profiles, correspond to distinct geographic areas that might be interpreted as homogeneous “air basins”. The results have major implications for monitoring network management since they provide an objective basis for choosing the most relevant spatial scale for producing and supplying pollen-related information, and for optimizing network architecture.