In southeastern France, vineyard irrigation has been increasing and now concerns about 8% of the total vineyard area in the region. Irrigation management aims to ensure both fruit quality and yield by maintaining the vineyard above a time-dependent threshold of soil water. Extension services disseminate information and advice on vineyard irrigation, based in large part on the WaLIS (Water Balance for Intercropped Systems) model described here, which computes the fraction of transpirable soil water of a vineyard with or without an intercrop. Extension bulletins currently report model results, but the extension services have indicated that adding uncertainty information would be important for informing users as to the reliability of those results. We show how uncertainty information could be calculated and presented. The major focus here is on the uncertainty due to using estimated rather than measured input variables, taking into account the way users estimate the input variables in practice. This is the source of uncertainty most relevant for users because it is an uncertainty that they can affect. However, it is also important to know how this uncertainty compares with the overall uncertainty in the predictions. We therefore also show how total uncertainty can be estimated.