Technical Abstract: Traditional approaches to understanding, predicting, and managing rangelands based on assumptions about homogeneity in space and stationarity in time of environmental drivers are expected to be insufficient in the future. Range scientists, ecologists, and global system scientists use approaches that emphasize three dimensions of problems differently: the role of humans, the importance of site-contingency, and the goal leading to specific questions. Here we argue that to overcome limitations and to effectively address future rangeland problems, scientists and land managers will need to successfully integrate these approaches in novel ways with developments in other disciplines. The objectives of this paper are: (1) to describe key challenges limiting effectiveness of current approaches, (2) to describe the background, historical development, and current status of different approaches, (3) to provide alternative ways to integrate these approaches, and (4) to discuss the implications of these integrated approaches to the future of range science when climate and human drivers are non-stationary. This integration will be critical for applying range science to the management of specific land units, contribute to and benefit from the development of general ecological principles, and the problems facing society at regional, continental, and global scales.