The Eastern Corn Belt Region (ECBR) is an important agricultural sector for the U.S. This study analyzes the climate extremes over the contemporary (1980–2018) and future (2036–2099) periods over the ECBR. We evaluated the performance of 32 downscaled models from the U.S. Global Change Research Program's Localized Constructed Analogs (LOCA) of the Coupled Model Intercomparison Project (CMIP5) to simulate extreme temperature and precipitation indices. The LOCA downscaled models were evaluated for the recent past against the National Aeronautics and Space Administration (NASA)-supported gridded observational dataset DAYMET. Results reveal key trends throughout the region that are consistent with previous studies, including significant increases in extreme minimum temperatures, reduction of cold nights, increase of warm nights, and decreases in diurnal temperature ranges. Much of the region demonstrates extreme warming trends in the coldest night of the year (more than 5 °C) and an increase in the heaviest precipitation events over 1980–2018. An optimal model ensemble (OME) was constructed using a Kling-Gupta Efficiency and Bhattacharyya coefficient evaluation to construct a comprehensive ranking procedure. Having outperformed a standard multi-model ensemble approach, the OME was used to evaluate the future changes of extreme climate indices under RCP4.5 and RCP8.5 scenarios. Though the OME showed consistently strong warming throughout the ECBR, variability among the optimal models and across watersheds is quite significant, especially for precipitation indices. Thus, constraining the uncertainty in future climate models, specifically as it relates to agriculture decisions that support climate-resilience, remains a challenge.

东部玉米带地区 (ECBR) 是美国的一个重要农业部门 本研究分析了 ECBR 在当代（1980-2018 年）和未来（2036-2099 年）期间的极端气候。我们评估了来自美国全球变化研究计划的耦合模型比对项目 (CMIP5) 的本地化构建类似物 (LOCA) 的 32 个缩小模型的性能，以模拟极端温度和降水指数。近期针对美国国家航空航天局 (NASA) 支持的网格观测数据集 DAYMET 对 LOCA 缩小模型进行了评估。结果揭示了整个地区与先前研究一致的主要趋势，包括极端最低气温显着增加、寒冷夜晚减少、温暖夜晚增加、并在昼夜温度范围内降低。该地区大部分地区在一年中最冷的夜晚（超过 5°C）表现出极端变暖趋势，并且 1980-2018 年期间最强烈的降水事件有所增加。使用 Kling-Gupta 效率和 Bhattacharyya 系数评估构建了一个最优模型集合 (OME)，以构建一个综合排名程序。OME 优于标准的多模式集合方法，用于评估 RCP4.5 和 RCP8.5 情景下极端气候指数的未来变化。尽管 OME 在整个 ECBR 中显示出持续强烈的变暖，但最佳模型之间和流域之间的变异性非常显着，尤其是降水指数。因此，限制未来气候模型的不确定性，