Résumé
"A Who's Who of pesticides is therefore of concern to us all. If we are going to live so intimately with these chemicals eating and drinking them, taking them into the very marrow of our bones-we had better know something about their nature and their power."-Rachel Carson, Silent Spring.In her day, Rachel Carson was right: plant protection products (PPP) like all the other chemical substances that humans increasingly release into the environment without further precaution, are among our worst enemies today. All compartments of the biosphere, air, soil and water, are potential reservoirs within which all species that live there are impaired. Birds are particularly concerned: PPP are recognized as a factor in the decline of their abundance and diversity predominantly in agricultural landscapes. Due to the restrictions on vertebrates testing, \textit{in silico} based approaches are an ideal choice alternative given input data are available. This is where the problem lies as we will illustrate in this paper. We performed an extensive literature search covering a long period of time, a wide diversity of bird species, a large range of chemical substances and as many model types as possible to encompass all our future need to improve environmental risk assessment of chemicals for birds. In the end, we show that poultry species exposed to pesticides are the most studied at the individual level with physiologically based toxicokinetic models. To go beyond, with more species, more chemical types, over several levels of biological organization, we show that observed data are crucially missing. As a consequence, improving existing models or developing new ones could be like climbing Everest if no additional data can be gathered, especially on chemical effects and toxicodynamic aspects.