Climate change projections for Portugal showed warming and drying trends, representing a substantial threat for the sustainability of forage production in perennial grassland. The objective of the present study was to assess climate change impacts on seasonal dry matter yield (DMY) in three locations (North-west-, Central-inner and South-Portugal) with different climatic conditions, for two grassland production systems deviating in growing season length, either early cuts in spring (ES) or late cuts in summer (LS). Impacts were estimated using the STICS (Simulateur mulTIdisciplinaire pour les Cultures Standard) crop model, by comparing a historical baseline period (1985–2006) with simulated projections over future periods (2021–2080). For this purpose, the STICS crop model was driven by high-resolution climate data from a coupled Global Climate Model/Regional Climate Model chain. As a result, we obtained that, during the baseline period, DMY of LS was consistently much higher than that of ES in all three locations. For LS, significant reductions in mean DMY were forecasted during 2061–2080, ranging from mild (–13%) in the north to severe (–31%) in the south of Portugal. In contrast, seasonal DMY was largely maintained for ES among sites until 2080, benefiting from low water deficits, the expected atmospheric CO2 rise and the forecasted temperature increase during cool season. Thus, the yield gap was projected to gradually decrease between the two regimes, in which mean DMY for ES was foreseen to exceed that of LS over 2061–2080 in the southern site. Moreover, ES was projected to have very low exposure to extreme heat and severe water stresses. Conversely, LS, subjected to high summer water deficit and irrigation needs, was projected to experience increased summertime water stress (9–11%) and drastically increased heat stress (33–57%) in 2061–2080, with more pronounced heat stress occurring in the south. Frequency of occurrence of extreme heat stress was projected to gradually increase in summer over successive study periods, with a concomitant increased intensity of DMY response to inter-annual variability of heat stress during 2061–2080. Heat stress tended to be more important than water stress under the prescribed irrigation strategy for LS, potentially being the main limiting factor for summertime DMY production under climate change scenario.
Implications of climate model biases and downscaling on crop model simulated climate change impacts
