Trap and haul programs are used to conserve fish populations by circumventing high mortality locations or events, and enhancing population abundance by reintroducing fish to historical habitats and mitigating for fish passage limitations. Spring run Chinook Salmon are transported in trucks upstream of barrier dams in Willamette River Tributaries as part of fish conservation efforts. Fish mortalities occurring during hauling minimizes the utility of the effort because natural origin fish are targeted for theses outplanting efforts. The objectives of this study were to develop models predicting hauling mortality and identify optimal hauling densities that minimize mortality risk and effort. We used an information-theoretic approach to evaluate multiple models predicting hauling mortality. Predictors identified varied between the two dams evaluated but were related to operations and annual or in-river conditions. The amount of time loading fish and the density of fish in tank trucks were positively associated with hauling mortality. Instream flows and thermal exposure were also identified as factors predicting with hauling mortality. We used the results of model selection to predict mortality risk and calculate daily hauling effort. Risk and effort were combined into a utility to identify optimal hauling densities for varying numbers of fish to haul and transport truck volume. Optimal hauling densities varied between dams reflecting whether loading time or hauling density was associated with hauling mortality. This analysis provides managers a way to integrate research, monitoring, and management to improve understanding of factors associated with hauling mortality and adjust optimal hauling densities using adaptive management.