Barrier islands provide important ecosystem services, including storm protection and erosion control to the mainland, habitat for fish and wildlife, and tourism (Barbier and others, 2011; Feagin and others, 2010). These islands tend to be dynamic due to their location along the estuarine-marine interface. Besides gradual changes caused by constant forces, such as currents and tides, barrier islands face numerous threats including hurricanes, accelerated sea-level rise, oil spills, and anthropogenic impacts (Pilkey and Cooper, 2014). These threats are likely to influence the future of barrier islands in the latter part of the 21st century, especially as climate-related threats to coastal areas are expected to increase in the future (Knutson and others, 2010; Hansen and others, 2016). As a result, natural resource managers are concerned with monitoring changes to these islands and modeling future states of these environments. Geomorphology regulates many abiotic factors that influence the performance of foundation plant species, including wave energy, salinity, inundation frequency, sea spray, Aeolian transport, and nutrient availability (Young and others, 2011). Researchers have established linkages between barrier island habitats and specific landscape position variables, such as distance from shoreline (Young and others, 2011) and elevation (Anderson and others, 2016; Foster and others, 2017; Halls and others, 2018; Young and others, 2011). Here, we built upon recent barrier island habitat model efforts by Foster and others (2017) and Halls and others (2018) to develop a machine learning-based habitat model for Dauphin Island, Alabama, USA. Our model incorporated elevation uncertainty for elevation-dependent habitat extraction and yields spatially explicit predictions of general barrier island habitats based on landscape position information, such as elevation, distance from shoreline, and relative topography. The habitats that were predicted in this model included: 1) Barrier flat; 2) Beach; 3) Dune; 4) Intertidal beach; 5) Intertidal flat; 6) Intertidal marsh; 7) Water-estuarine; 8) Water-fresh; 9) Water-marine; 10) Woody vegetation; and 11) Woody wetland. Models were developed for three tidal zones: 1) subtidal; 2) intertidal; and 3) supratidal/upland. Deterministic accuracy, fuzzy accuracy, and hindcasting were used for validation. This data release contains data used to develop and validate the machine learning-based habitat model including: 1) final contemporary habitat model results; 2) contemporary habitat model training data per tidal zone; 3) contemporary habitat model predictor variables per tidal zone; 4) contemporary habitat model validation data; 5) final hindcast habitat model results; 6) hindcast habitat predictor variables per tidal zone; and 7) hindcast habitat validation data. For more information, see Enwright and others (2019).