Climate projections point to modifications in the magnitude, frequency and timing of floods in the future. However, robust methodologies to quantify how climate change will modify the catchment response in flood events are required. Continuous hydrological modelling usually smooth magnitudes of extreme events. This paper proposes a methodology to improve the assessment of flood changes in the future driven by climate change. Climate change projections of the EURO-CORDEX programme obtained under the Fifth Assessment Report (AR5) of the Intergovernmental Panel on Climate Change (IPCC) supplied are used. Four catchments located on the Douro River Basin have been considered as case studies. Precipitation and temperature projections have been bias corrected to reduce errors with observations in the control period (1971–2004). The HBV continuous hydrological simulation model has been used to simulate the soil moisture content on the day of occurrence of the maximum annual rainfalls in the four catchments. The modified curve number method has been utilized to obtain the changes expected in the future in flood magnitudes, considering the initial soil moisture contents estimated with the HBV model and the expected changes in annual maximum rainfalls. The methodology has been applied to the control period (1971–2004) to check the validity of the process. Then, the methodology has been applied to the future period (2011–2095), to quantify the changes expected in the future in flood magnitudes under climate change conditions. The combined use of the HBV continuous hydrological simulation with the modified curve number method improves the results provided by the HBV model. The proposed methodology allows a better characterization of the response of catchments in flood events. It also considers the expected variation in the antecedent moisture content in catchments in the future, driven by increasing temperatures and decreasing mean annual precipitations in the future. The results show that flood quantiles will increase in three of the four catchments considered.

气候预测指出，未来洪水的数量，频率和时间都会发生变化。但是，需要强有力的方法来量化气候变化将如何改变洪水事件中的集水区响应。连续的水文模拟通常可以平滑极端事件的幅度。本文提出了一种方法，可以改善对气候变化驱动的未来洪水变化的评估。使用根据政府间气候变化专门委员会（IPCC）的第五次评估报告（AR5）获得的EURO-CORDEX计划的气候变化预测。案例研究考虑了位于杜罗河流域的四个流域。对降水和温度预测进行了偏差校正，以减少控制期内（1971-2004年）观测值的误差。HBV连续水文模拟模型已被用来模拟四个流域最大年度降雨发生当天的土壤水分含量。考虑到用HBV模型估算的初始土壤含水量和年度最大降雨量的预期变化，改进的曲线数方法已被用于获得未来洪水幅度的预期变化。该方法已应用于控制期（1971-2004年），以检查过程的有效性。然后，该方法已应用于未来一段时间（2011年至2095年），以量化气候变化条件下未来洪水强度的预期变化。结合使用HBV连续水文模拟和改进的曲线数法，可以改善HBV模型提供的结果。所提出的方法可以更好地表征洪水事件中集水区的响应。它还考虑到未来由于温度升高和未来年均降水量减少而导致流域中先前含水量的预期变化。结果表明，在所考虑的四个流域中，有三个流域的洪水分位数将增加。