Rolling hill style topography is a common feature of agricultural land throughout the United States. Topographic complexity causes subfield variation in soil resources such as water and nutrients, leading to a mosaic of high- and low-productivity zones that can shift from one year to the next due to weather. Stabilizing yields across these productivity zones using agroecological methods may improve land use efficiency, prevent unnecessary cropland expansion, and reduce the environmental impact of these systems. Here, we hypothesized that cover crops may help to reduce soil water and nutrient losses and increase the stability of subsequent maize yields across time and space. We performed a field study to evaluate the effect of a cereal rye (Secale Cereale L.) cover crop on maize (Zea mays L.) yield at three landscape positions (summit, backslope, and toeslope) in Central KY in 2018–2019, and calibrated the DSSAT v4.7.0.001 computer simulation program to test our hypothesis across a thirty-year period. Our field trial showed pronounced variability in maize yield across different landscape positions, ranging from 6.3 Mg ha⁻¹ in the backslope, to 12.2 Mg ha⁻¹ in the toeslope. Model simulations were consistent with results from our field trial and indicated that low yields in the backslope were primarily due to water stress, with >10 % yield reductions in 17 out of 30 simulated years relative to simulations under irrigated conditions where water was not limiting. In contrast, the toeslope and summit positions experienced >10 % yield reductions due to water stress in only 6 of the 30 years. Growing a cereal rye cover crop before maize reduced the frequency of water stress and raised maize yields in the backslope by 6% (500 kg ha^-1) on average, and 24 % (1235 kg ha^-1) during dry years. The coefficient of variation across all weather conditions and landscape positions was reduced from 33 % to 26 % when maize followed a rye cover crop compared to fallow. The yield benefits of the cover crop were associated with decreased soil evaporation and runoff that increased water availability during anthesis and late maize reproductive phases. Crop model simulations allowed us to evaluate and parse out the fundamental drivers of the interaction between cover crops and complex topography under different weather scenarios. Overall, our study demonstrates the outsized potential of cover crops to increase and stabilize grain yields in rolling hill landscapes and emphasizes the value of cover crops as a tool for ecological intensification.

连绵起伏的丘陵式地形是全美国农业用地的共同特征。地形的复杂性导致土壤资源（如水和养分）的子领域变化，导致高生产力和低生产力地区的拼接，由于天气原因，该地区可能会从一年转移到下一年。使用农业生态学方法稳定这些生产力地区的单产可以提高土地利用效率，防止不必要的耕地扩张，并减少这些系统对环境的影响。在这里，我们假设覆盖作物可能有助于减少土壤水和养分流失，并提高随后的玉米产量在时间和空间上的稳定性。我们进行了田间研究，以评估谷物黑麦（Secale Cereale L.）覆盖作物对玉米（Zea mays）的影响L.）于2018-2019年在肯塔基州中部的三个景观位置（坡度，后坡和坡度）获得收益，并校准了DSSAT v4.7.0.001计算机模拟程序以检验我们在三十年期间的假设。我们的田间试验表明，不同景观位置的玉米产量存在明显差异，从后坡的6.3 Mg ha ^-1到12.2 Mg ha ^-1在坡道上。模型模拟与我们的田间试验的结果一致，表明后坡的低产量主要是由于水分胁迫，相对于灌溉条件下水不受限制的模拟，在30个模拟年中有17年产量降低了10％以上。相比之下，在30年中只有6年，由于水分胁迫，坡脚和山顶位置的收成减少了10％以上。在玉米前种植谷类黑麦覆盖作物可以减少水分胁迫的频率，并使后坡地的玉米平均产量提高6％（500 kg ha ^-1），平均提高24％（1235 kg ha ^-1））。与休耕相比，玉米跟随黑麦覆盖作物时，所有天气条件和景观位置的变异系数从33％降低至26％。覆盖作物的增产效益与减少的土壤蒸发和径流相关，从而增加了花期和玉米生育后期的水分利用率。作物模型模拟使我们能够评估和解析不同天气情况下地表作物与复杂地形之间相互作用的基本驱动力。总体而言，我们的研究表明，在连绵起伏的丘陵景观中，覆盖作物具有增加和稳定谷物产量的巨大潜力，并强调了覆盖作物作为生态集约化工具的价值。