Winter hardiness depends on the ability of plants to tolerate a wide spectrum of environmental stresses, which can be affected by climate change in complex ways. Here, empirical Partial Least Squares Regression (PLSS) models of winter survival (WS) of winter wheat (Triticum aestivum L.) and triticale (Triticosecale x Wittmack) were created using data from six years of field experiments at multiple locations throughout Poland. These included 553 winter wheat and 155 triticale accessions. Our aims were to: 1) predict WS on the basis of meteorological data; 2) identify the most critical weather events affecting WS of winter wheat and triticale under Polish conditions; and 3) predict WS for the simulated winters of 2040, 2060 and 2080 under climate change scenarios RCP2.6, RCP4.5, RCP6.0 and RCP8.5 for the experimental site with the lowest mean WS rate during the field experiments. The empirical models showed a high R² for winter wheat (0.751), and a low R² for winter triticale (0.160), because of the low winter damage to triticale observed during the experiments. The key climate factors affecting WS varied between species. Winter wheat was affected by winter severity, the number of freezing-thawing cycles, the thermal vegetation index and the freezing index in various winter months. Triticale was affected by late winter ice encasement and high numbers of freeze-thaw events. The predictions indicated that the WS of both winter wheat and triticale may decrease in the future, especially when more extreme climate change scenarios were considered. The main issue will be cold deacclimation connected with climate warming which will be more important for WS than the general increase in minimum winter temperatures. This finding indicates that deacclimation tolerance should be included in wheat and triticale breeding programs as a trait crucial for WS under future winters, at least in Central Europe.

冬季抗寒性取决于植物耐受范围广泛的环境压力的能力，这些压力可能以复杂的方式受到气候变化的影响。在这里，冬小麦 ( Triticum aestivum L.) 和小黑麦 ( Triticosecale ) 的冬季存活 (WS) 的经验偏最小二乘回归 (PLSS) 模型x Wittmack) 是使用来自波兰多个地点六年现场实验的数据创建的。其中包括 553 个冬小麦和 155 个小黑麦种质。我们的目标是：1) 根据气象数据预测 WS；2) 确定波兰条件下影响冬小麦和黑小麦 WS 的最关键天气事件；3) 预测2040年、2060年和2080年模拟冬季在气候变化情景RCP2.6、RCP4.5、RCP6.0和RCP8.5下，在田间试验中平均WS率最低的试验场的WS。经验模型显示冬小麦的R ²较高（0.751），而 R ²较低对于冬季小黑麦 (0.160)，因为在实验期间观察到对小黑麦的冬季损害较低。影响 WS 的关键气候因素因物种而异。冬小麦受冬季严重程度、冻融循环次数、热植被指数和冬季各月冻结指数的影响。黑小麦受到晚冬冰封和大量冻融事件的影响。预测表明，未来冬小麦和小黑麦的 WS 可能会下降，尤其是在考虑更极端的气候变化情景时。主要问题将是与气候变暖相关的寒冷去驯化，这对 WS 而言比冬季最低温度的普遍增加更为重要。