Unlike other North American prairie-nesting dabbling ducks, northern pintail (Anas acuta) populations have not increased since the early 1990s and remain well below the long-term average for traditional survey areas. Previously reported estimates of annual survival and recovery rates for pintails did not investigate any spatial or temporal factors to explain annual variation of these rates. We used band-recovery data from 1970 to 2003 to test the influence of temporal periods defined by differing harvest regulations and habitat conditions of breeding grounds with spatially delineated regions on survival and recovery rates of northern pintails in North America. We separated regions based on a multiresponse permutation procedure to identify banding blocks with dissimilar recovery distributions based on a cluster analysis. We categorized time by grouping years into temporal periods based on bag limits, season lengths, or overflight versus nonoverflight years. We used the Brownie approach in Program MARK to evaluate 46 a priori models estimating survival and recovery rates. The best approximating model indicated that survival varied with age, sex, and region with additive time and interactive time-by-age and time-by-region effects. Recovery rate was best represented by a fully interactive term comprised of age, sex, region, and year. There were no statistical differences among average annual survival point estimates between age and sex classes within each region, and our estimates were similar to previous unpublished studies. We found the eastern region had decreased survival and increased recovery rates compared to other regions. Trends in pintail survival suggest that variation in annual survival was not the cause of the initial decrease in the northern pintail population and is unlikely the dominant factor preventing the population from increasing. The influence of other population parameters, such as recruitment rate, should be investigated to further evaluate other causes for the population status of northern pintails. Use of the top-ranked model to estimate annual survival and recovery rates for northern pintails in North America, which indicated that annually varying estimates of survival rates were better supported by the data than grouping years into temporal classes (i.e., based on bag limits, season lengths, and overflight yr) can be used by managers and policy makers when considering annual harvest regulations and effects of conservation efforts. Managers should incorporate these estimates into future demographic studies of pintails as well as consider using the top-ranked model for future analyses of band-recovery data.