Long-term fisheries independent gill net surveys conducted in Texas estuaries from 1975 to 2006 were used to develop spatially explicit estuarine habitat use models for 3 coastal shark species: bull shark Carcharhinus leucas, blacktip shark C. limbatus, and bonnethead shark Sphyrna tiburo. Relationships between environmental predictors and shark distribution were investigated using boosted regression trees (BRT). Bull shark was the most abundant species (n = 5800), followed by blacktip (n =

more »

... ), and bonnethead sharks (n = 1793). Environmental conditions influenced distribution patterns of all species and relationships were nonlinear, multivariate, and interactive. Results showed very good model performance and suggested shark distribution is most closely linked to salinity, temperature, and proximity to tidal inlets. By interpolating the BRT models, maps of the probability of capture were produced using ordinary kriging. Results showed that the central region along the Texas coast contains the most important estuarine shark habitat. This area was characterized by warm temperatures, moderate salinities, and abundant tidal inlets. Bull sharks also extended into low salinity estuaries, while blacktip and bonnethead sharks were restricted to areas near tidal passes with moderate salinities. Juvenile sharks were frequently captured, suggesting the Texas coast may constitute important nursery areas for all 3 species. The development of these spatially explicit models allows for prioritization and conservation of areas in a region that has great potential for human disturbance and climate change impacts. These results provide new insight into the habitat requirements of coastal sharks in the northwestern Gulf of Mexico and practical information for managing this resource. KEY WORDS: Shark · Boosted regression trees · Essential fish habitat Resale or republication not permitted without written consent of the publisher Editorial responsibility: Hans Heinrich Janssen,