We estimated age-specific fecundity and survival rates for peppered chub Macrhybopsis tetranema from the Canadian River, New Mexico and Texas. We used these estimates to construct a life history matrix model that assumed age-0 survival was related to river discharge. Model predictions agreed well with the observed abundance of peppered chub for the 6-year period from 1996 to 2001. Based on the Akaike information criterion, this model received greater support from observed catches of peppered chub than did two alternative null models (one null model assumed a static or fixed population and the other assumed a population with a constant growth rate over the 6-year study period). Elasticity analysis showed that the peppered chub population growth rate was most sensitive to changes in age-0 survival (elasticity = 0.48) and age-1 fecundity (elasticity = 0.44). We performed sensitivity simulations to determine the effect of parameter uncertainty on the observed elasticities. Based on 1,000 simulations, we found that the peppered chub population growth rate was most sensitive to age-0 survival and age-1 fecundity and was robust with respect to uncertainty in our estimates of these parameters. Our model accurately predicts changes in peppered chub abundance based on river discharge and provides a mechanistic explanation for previous anecdotal observations indicating that the reproductive success of peppered chub is related to river discharge.