As our species moves deeper into an era in which we have an increasing influence over the climate and health of our planet, it is important to examine the likely effects of our activities as well as the tools we can use to adapt to coming changes. Occupying more land area than any other human activity and employing biological systems vulnerable to extreme heat, agriculture is chief amongst vital industries impacted by a changing climate. Previous work has focused on those impacts, finding potentially drastic effects for countries like the United States, the world's largest producer of maize and soybean, whose major production regions are fortuitously positioned near a climate optimum for those key crops. This dissertation examines various specific practices that could be deployed to build resilience and prevent the degradation of the U.S. agricultural system under potential 21st century climate regimes. Double cropping, crop rotation, cover cropping, and irrigation all have their place as potential adaptations. This work uses mechanistic and statistical models as well as newly available datasets and data processing methodologies to explore the expansion of suitability, the spatial variability, the yield effects, and the temporal trends in adoption of these practices respectively. Chapter 1 runs mechanistic phenological models for winter wheat and soybean under recent and future climate scenarios, finding that even small increases in expected temperature and growing season length can lead to large increases in double crop suitability. These changes in suitability have already been occurring over the last few decades and appear poised to accelerate along with our changing climate. While the increase the area suitable for this cropping practice is large, especially later in this century, the implied increase in agricultural production that accompanies it is substantially smaller than potential yield losses. Building on the first chapter but exploring inter-yearly crop rotation patterns versus intra-yearly patterns, Chapter 2 uses a large dataset of field-level yields to examine the yield penalties seen in continuous maize and soybean fields. Yield loss from continuous cropping found in the model was broadly consistent with findings from field trials. Additionally, the spatial breadth and temporal depth of the dataset enabled us to find that areas with large negative yield anomalies see worse yield penalties for continuous cropping, as do soybean crops grown in areas or years with low early season vapor pressure deficit and maize crops grown in areas or years with low early or late minimum temperatures. Chapter 3 examines another promising crop configuration with potential to serve as a climate adaptation; cover crops. In it, we build a cover crop classifier based on remotely sensed data and cross the classifier's output with already existing soil quality as well as maize and soybean yield maps. The raw classifier output shows that, as intended, cover crops are more likely to be found on poorer soils in the Midwest. Contrary to other sources, however, yield benefits for adopters of the practice are quite mild, even after a number of years following the practice. Combining this conclusion with the currently high cost of cover crop adoption, continued expansion of government funding for cover cropping appears necessary to propagate the practice. Chapter 4 uses methods built in Chapter 3, but with a different aim in mind -- mapping irrigation and its adoption in two key states in the western U.S. maize-soybean belt. Here we find that irrigation has indeed been on the increase over the last decade and a half in Nebraska, though no definitive trends were seen in Iowa. The increase in Nebraska does not appear to be driven by changes in the difference between irrigated and dryland yields, and irrigation adoption was more likely to be undertaken on higher quality land from 2003-2017 versus earlier in the practice's history.