Publications des agents du Cirad


Modelling coffee rust (Hemileia vastatrix)

Avelino J.. 2009. In : Rapidel Bruno (ed.), Roupsard Olivier (ed.), Navarro Muriel (ed.). Modelling agroforestry systems : Workshop Proceedings,Turrialba, Costa Rica, 25-29 February 2008. Turrialba : CATIE, p. 81-90. (Technical Meetings, 14).

Plant disease epidemiologists have built models in order to describe, explain or predict development of epidemics, taking into account the multiple factors that can affect a disease. Indeed, the existence and severity of a disease are determined by the effects of a host, a pathogen, an environment, a cropping system and their interactions. The host component concerns types of resistance, complete or partial. It also involves physiological, morphological or even architectural aspects. The parasite component concerns the biology of the organism and genetic aspects like virulence and aggressiveness. The environmental component concerns the climate (primarily wetness, temperature, radiation, wind) as well as soil, topography, landscape structure and the biological environment. The crop management component brings into play all the agricultural practices that can affect the disease, directly or indirectly. For coffee diseases, only statistical models have been developed. Some of them are very simple. They explain a characteristic of the disease (for example, the latent period3) or of the epidemic4 (the percentage of diseased leaves for example) by only a few climatic or host characteristics predictors. Others are much more complicated. They integrate a great number of predictors and sometimes make a hierarchy between them by order of influence on the response. All of these models have been very useful for the understanding of the conditions that are propitious to the development of an epidemic. However, further understanding of the mechanisms that govern coffee-disease epidemics requires different approach, including the development of mechanistic models, which permit analysing in a quantitative manner the way a system is functioning. It has been widely used for modelling plants diseases, as it allows quantitative integration of the individual effects, and interactions, of various factors on complex biological processes.

Mots-clés : hemileia vastatrix; coffea; costa rica

Communication de congrès

Agents Cirad, auteurs de cette publication :