To examine potential influence of tributaries on riparian habitat complexity along ~216 km of the Colorado River in Utah and ~300km of the Dolores River in Colorado and Utah, we first classified fluvial features and land cover of the bottomland on remotely sensed imagery. We then examined riparian and geomorphic patterns within the near channel zone with variably-sized spatial units. We used supervised image classification to create a 2-m resolution map of the primary land cover types within bottomlands of the Colorado and Dolores rivers, including two anthropogenic classes, four vegetation classes, bare ground, water and shadow. We selected these cover classes as major vegetation and land cover types that could be discerned from imagery. Our minimum mapping unit was 16m2. We were unable to map channel areas with flowing or standing water using supervised image classification, so we hand digitized channels based on a visual inspection of 2-m resolution imagery. We classified 6 channel classes based on their geomorphic characteristics and location within the river network (i.e., tributary vs. primary channel) or relation to the primary channel (e.g., split flow channels and secondary channels) and converted these to a 2-m resolution image (adapted from Moore et al 2012). We then combined land cover and channel classes to produce a single map representing both cover types along the Colorado and Dolores rivers. Our classification was based on 2-m resolution, multi-spectral (RGB NIR) aerial photographs for September 2013 and 2014 from the USDA National Agriculture Imagery Program (NAIP; http//www.fsa.usda.gov). We identified tributary junctions using the National Hydrography Dataset Plus Version 2 (NHDPlus V2) using the medium resolution (1:100,000 scale) National Hydrography Dataset (NHD) (http://nhd.usgs.gov/). To more accurately locate tributary junctions, we extracted flowlines corresponding to tributaries and converted each flowline to a point located at the terminus proximal to the channel centerline. We manually corrected tributary junction point locations with the NAIP images. We defined the near channel zone as within 20 meters of the edge of the Dolores low flow channel and within 100 meters of the edge of the Colorado low flow channel. These distances represented the average widths of the low flow channel for the two rivers. We assumed that habitat conditions closer to the channel would be more strongly influenced by fluvial processes and less strongly influenced by land management (e.g., farming, road development). We created spatial units for analysis within the near channel zone with Thiessen polygons - a polygon containing a point and defining an area closest to the point relative to all other systematically placed points (Fortin and Dale 2005). Beginning at the upstream study site boundary for each river, we placed regularly spaced points at three intervals: 10-, 25-, and 100-m to capture patterns for different sized spatial units around tributary junctions. For each point, we created a Thiessen polygon. Our use of Thiessen polygons as spatial units followed the example of other researchers (Alber and Piegay 2011). This data release includes shapefiles and associated metadata for: land and channel cover types along both rivers; tributary junction locations along both rivers; and the 10-, 25-, and 100-m Thiessen polygons along both rivers. Alber A., and Piégay H., 2011, Spatial disaggregation and aggregation procedures for characterizing fluvial features at the network-scale: application to the Rhône basin (France): Geomorphology, v. 125, p. 343-360. Fortin M.J., and Dale M.T., 2005, Spatial analysis: a guide for ecologists: Cambridge, Cambridge University Press, 365 p. Moore K., Jones K., Dambacher J., and Stein C., 2012, Aquatic inventories project methods for stream habitat surveys: Corvallis, OR, Conservation and recovery program, Oregon Department of Fish and Wildlife, 74 p.