Effective conservation of fish species of greatest conservation need (SGCN) requires an understanding of species-habitat relationships and distributional trends. Thus, modeling the distribution of fish species may serve as a potentially valuable tool for conservation planning. Our goals were to evaluate the status of fish SGCN in wadeable Iowa streams, test the effectiveness of existing species distribution models, and identify the relative influence and importance of habitat variables measured at multiple spatial scales on fish SGCN occurrences. Fish assemblage and habitat data were collected from 86 wadeable stream segments in the Mississippi River drainage of Iowa during 2009 and 2010. The frequency of occurrence of ten fish SGCN in stream segments where they were historically documented varied from 0.0% to 100.0% with a mean of 53.0% suggesting the status of Iowa fish SGCN is highly variable. The accuracy of existing species distribution models was evaluated with Cohen's kappa values and other model performance measures calculated by comparing field collected presence-absence data to model predicted presences and absences for twelve fish SGCN. Kappa values varied from 0.00 to 0.50 with a mean of 0.15, and indicated that only three models predicted species occurrence more accurately than would be expected by chance. Poor model performance likely reflects the difficulties associated with modeling the distribution of rare species and the inability of large-scale explanatory variables to explain variation in species occurrences. Thus, we developed occurrence models for seven fish SGCN using large-scale habitat variables (e.g., stream order, elevation, gradient), small-scale habitat variables (e.g., depth, velocity, coarse substrate), and habitat variables measured at multiple scales to identify the most influential spatial scale on species occurrences. On average, correct classification rates and Cohen's kappa values were greatest for multiple-scale models, intermediate for small-scale models, and lowest for large-scale models. However, large-scale models predicted the occurrences of two species with greater accuracy than small-scale models. Our results highlight the need for long-term monitoring efforts to better understand distributional trends and habitat associations of fish SGCN, and the necessity of understanding the factors that constrain the distribution of fishes across spatial scales to ensure that management decisions and actions occur at the appropriate scale.