The only self-sustaining population of endangered whooping cranes (Grus americana) requires a network of conservation lands for wintering along the Texas Gulf Coast (USA), so that this increasing population can reach downlisting under the Endangered Species Act (1,000 birds). We identify locations providing the highest quality and most sustainable wintering habitat for these whooping cranes through 2100 by predicting future habitats under three projections of sea level rise (0.6, 1.0 and 2.0 m by 2100), while incorporating two scenarios of future urban development. Our method combines predictions of future habitat quality with current whooping crane density estimates to calculate the potential carrying capacity of whooping cranes for each 10 m pixel within this 17,725 km² area. We found whooping cranes used locations with salt marsh at twice the rate of places lacking marsh. Areas > 15 km from development or < 2 km from estuarine water had increased crane use. Predicted area of salt marsh habitat oscillated across time given different rates of sea level rise. One urbanization scenario predicted 3% and the other 1% of the area converting to development by 2100. We estimated the study area can support 4414 (95% CI: 4096-4789) whooping cranes currently, 4795 (95% CI: 4402-5269) with 0.6 m sea level rise, 3559 (95% CI: 3352-3791) with 1 m sea level rise, and 2480 (95% CI: 2375-2592) with 2 m sea level rise by 2100, under the more aggressive urban development scenario. By anticipating climate-induced habitat loss with species population expansion we provide the requisite spatial information for conservation planners to build a sustainable conservation estate for downlisting whooping cranes. By coupling wildlife biology with conservation planning and on-the-ground implementation, our work exemplifies a proactive approach to recover endangered species.