The purpose of the present investigation was to examine the ability of the simple mechanistic model published by Jouven et al. (2006a) to simulate the effects of drought on the seasonal patterns of growth in managed grasslands. Data collected in Switzerland were used for reference. Different representations of the effects of water stress on transpiration and growth were tested, but otherwise all simulations were run using the standard model configuration. Results indicated that the model was able to correctly simulate the seasonality of growth across sites and years, as well as variations in soil water availability. Overall, the effects of drought on growth were realistically reproduced by the model. This was specifically the case with respect to the exceptional conditions characterizing the summer of 2003. On the other hand, there was a tendency for overestimating the impacts during conditions of moderate water stress. Replacing the original formulation of the water stress response with alternate schemes was not sufficient to bypass the problem. This suggests that it could be worth examining ways to represent drought avoidance or enhanced drought tolerance in the model. The failure to simulate a delayed start of growth in the spring of 2011 further indicate the necessity to address in a mechanistic way spring regrowth. Concerning future model developments, we argue that the inclusion of a root/soil compartment should be addressed with high priority to be able to account for the interactions between water and nutrient cycling. (C) 2015 Elsevier B.V. All rights reserved.