The deposition of active nitrogen on the special areas of conservation (SAC) is too high for several ecosystems in Flanders, Belgium. Therefore, in response to the 92/43/EEC (Habitats Directive) article 6 obligations of the member states, a policy to reduce nitrogen deposition is currently implemented in Flanders. Part of this policy are adjustments to the regulations for permitting new nitrogen (i.e. NH3, NOx) emitting installations. According to §3 of article 6, any plan or project that likely has a significant effect on the SAC, either individually or in combination with other plans or projects, is subject to an appropriate assessment of its implications for the site. It can only be approved if there is certainty that it will not adversely affect the integrity of the SAC. As is regularly done for permitting, this test is performed by using a model. The model chain in this case is the VLOPS-IFDM model. The model is well validated and based on these results, the uncertainty on the model results has been estimated. However, this uncertainty is quite large, in part due to the limited knowledge on the dry deposition process of reactive nitrogen. This study will first present the validation results for several validation cases. Thereafter, a discussion about the implications for policy related cases is initiated. At the moment the model uncertainty is ignored in the permitting procedure. However, one could state that this is not in agreement with the necessity of proving that the permit will not adversely affect the integrity of the SAC. Nevertheless, if the uncertainty was to be taken into account too rigorously, permitting new installations in Flanders would be impossible. The goal of this part is not to present the definite answer to this question but to initiate a discussion on how to take into account uncertainties in policy supporting cases.