Each text file describes coupled climate and land use scenarios used to model species distributions under different greenhouse gas emissions trajectories and time periods. Three different greenhouse gas emissions trajectories (representative concentration pathways [RCPs]), were used as the basis for future scenarios, including the 2.6 (low emissions), 4.5 (medium-low emissions) and 8.5 (high emissions) W/m2 RCPs (IPCC 2013). Three GCMs were selected based on two criteria: 1) high performance in hindcasting climate variables for the study area (Sheffield et al. 2013), and 2) to encompass the full range of variability across future climate projections from all available GCMs (Knutti et al. 2013, Maloney et al. 2014). The GCMs selected included IPSL-CM5A (Institut Pierre-Simon Laplace), MRI-CGCM3 (Meteorological Research Institute), and NOR-ESM (Norwegian Climate Centre). The scenario-based climate data were downscaled for the study area through the use of weather generator techniques to provide monthly and annual air temperature and precipitation data for the years 2006 through 2099 (Schoof et al. 2007, Schoof et al. 2010). Three time periods were used in our scenarios: 2030-2039, 2060-2069, and 2090-2099. Climate variables were averaged across each time period. Logistic regression land cover models were developed for row-crops (corn, soybean and cotton), wheat, forest and grasslands within the study area (Stoebner and Lant 2014). Developed land and water classifications were not modeled and therefore assumed to not change under any scenario. Projections for each modeled land cover were made using the climate scenarios detailed above (Stoebner 2014). Each land cover projection produced a probability of occurrence of the land cover for each pixel in the study area, which were then calculated into percent within 12-digit hydrologic unit codes (HUC). All stream segments within a HUC were attributed the same land cover percentages.