We used capture-recapture data to assess population dynamics of endangered Lost River suckers (Deltistes luxatus) and shortnose suckers (Chasmistes brevirostris) in Upper Klamath Lake, Oregon. The Cormack-Jolly-Seber method was used to estimate apparent survival probabilities, and a temporal symmetry model was used to estimate annual seniority probabilities. Information theoretic modeling was used to assess variation in parameter estimates due to time, gender, and species. In addition, length data were used to detect multiple year-class failures and events of high recruitment into adult spawning populations. Survival of adult Lost River and shortnose suckers varied substantially across years. Relatively high annual mortality was observed for the lakeshore-spawning Lost River sucker subpopulation in 2002 and for the river spawning subpopulation in 2001. Shortnose suckers experienced high mortality in 2001 and 2004. This indicates that high mortality events are not only species specific, but also are specific to subpopulations for Lost River suckers. Seniority probability estimates and length composition data indicate that recruitment of new individuals into adult sucker populations has been sparse. The overall fitness of Upper Klamath Lake sucker populations are of concern given the low observed survival in some years and the paucity of recent recruitment. During most years, estimates of survival probabilities were lower than seniority probabilities, indicating net losses in adult sucker population abundances. The evidence for decline was more marked for shortnose suckers than for Lost River suckers. Our data indicated that sucker survival for both species, but especially shortnose suckers, was sometimes low in years without any observed fish kills. This indicates that high mortality can occur over a protracted period, resulting in poor annual survival, but will not necessarily be observed in association with a fish kill. A better understanding of the factors influencing adult survival and recruitment into spawning populations is needed. Monitoring these vital parameters will provide a quantitative means to evaluate population status and assess the effectiveness of conservation and recovery efforts.